2009 IEEE Nuclear Science Symposium and Medical Imaging Conference

Abstract Book

Contents
Nuclear Science Symposium .......................................................................................................... 5 NP: NSS Plenary .................................................................................................................... 5 N01: Photodetectors and Scintillation Detectors I ................................................................. 5 N02: Semiconductor Detectors I: Silicon Detectors and Applications .................................. 8 N03: Analog and Digital Circuits I ...................................................................................... 10 N04: Nuclear Measurements and Monitoring Techniques .................................................. 12 N05: New Detector Concepts and Instrumentation I ........................................................... 15 N06: Instrumentation for Homeland Security I ................................................................... 17 N07: Data Acquisition and Analysis Systems I ................................................................... 19 N08: Radiation Damage Effects I: Semiconductor Devices ................................................ 22 J01: Instrumentation for Medical and Biological Research I: Radionuclide Imaging......... 24 N09: Computing and Software for Experiments I: Simulation ........................................... 26 N10: Trigger and Front-End Systems I ................................................................................ 29 N11: Gaseous Detectors I: Development of Techniques ..................................................... 31 N12: High Energy Physics Instrumentation I ...................................................................... 33 J02: Instrumentation for Medical and Biological Research II: X-ray Imaging and Radiotherapy Applications .......................................................... 35 N13: Posters I ....................................................................................................................... 38 J03: Joint NSS/MIC 3 ........................................................................................................ 113 N14: Instrumentation for Homeland Security II ................................................................ 115 N15: Nuclear Physics Instrumentation I ............................................................................ 118 N16: Gaseous Detectors II: Varied Applications in Astrophysics and Particle Physics ... 120 N17: Computing and Software for Experiments II: New Computing Technologies......... 122 J04: Joint NSS/MIC 4 ........................................................................................................ 125 N18: Gamma-ray Imaging I: Compton Imaging ............................................................... 127 N19: Analog and Digital Circuits II................................................................................... 129 N20: Neutron Imaging and Detectors for Neutron Imaging .............................................. 131 N21: Trigger and Front-End Systems II ............................................................................ 134 N22: Semiconductor Detectors II: Silicon Devices ........................................................... 136 N23: Computing and Software for Experiments III: High Energy Physics Computing.... 138 N24: New Detector Concepts and Instrumentation II........................................................ 140 J05: NSS/MIC Joint Posters .............................................................................................. 143 N25: Posters II ................................................................................................................... 150 N26: Gamma-Ray Imaging II ............................................................................................ 229 N27: Analog and Digital Circuits III ................................................................................. 231 N28: Photodetectors and Scintillation Detectors II ........................................................... 233
2

N29: New Detector Concepts and Instrumentation III ...................................................... 236 N30: Accelerators and Beam Line Instrumentation........................................................... 238 N31: Semiconductor Detectors III: CZT Detectors ........................................................... 240 N32: Radiation Damage Effects II: Scintillators ............................................................... 243 N33: Computing and Software for Experiments IV: Software for Experimental Applications .................................................................. 244 N34: High Energy Physics Instrumentation II ................................................................... 247 N35: Astrophysics and Space Instrumentation I: Component development ..................... 249 N36: Nuclear Physics Instrumentation II ........................................................................... 251 N37: Astrophysics and Space Instrumentation II: Instruments ......................................... 253 N38: Synchrotron Radiation Instrumentation .................................................................... 255 N39: Computing and Software for Experiments V: Bio-medical Software ...................... 257 N40: Photodetectors and Scintillation Detectors III .......................................................... 259 N41: Semiconductor Detectors IV: CdTe and other Wide Band Gap Materials ............... 261 N42: Data Acquisition and Analysis Systems II................................................................ 263 N43: High Energy Physics Instrumentation III ................................................................. 266 N44: Gaseous Detectors III: GEM Applications in Particle Physics ................................. 268 N45: High Energy Physics Instrumentation IV ................................................................. 270 Medical Imaging Conference ...................................................................................................... 272 J01: Instrumentation for Medical and Biological Research I: Radionuclide Imaging ............................................................................................... 272 J02: Instrumentation for Medical and Biological Research II: X-ray Imaging and Radiotherapy Applications ........................................................ 272 J03: Joint NSS/MIC 3 ........................................................................................................ 272 J04: Joint NSS/MIC 4 ........................................................................................................ 272 M01: Plenary 1 ................................................................................................................... 272 J05: NSS/MIC Joint Posters .............................................................................................. 273 M02: Image Processing and Evaluation ............................................................................ 273 M03: Image Reconstruction 1 ............................................................................................ 275 M04: Quantitative Imaging Techniques ............................................................................ 277 M05: MIC Posters 1 ........................................................................................................... 280 M06: PET/SPECT instrumentation 1................................................................................. 327 M07: Image Reconstruction 2 ............................................................................................ 329 M08: X-ray imaging 1 ....................................................................................................... 332 M09: MIC Posters 2 ........................................................................................................... 335 M10: PET/SPECT instrumentation 2................................................................................. 383 M11: Plenary 2 / Multimodality Instrumentation and Techniques .................................... 385 M12: X-ray imaging 2 ....................................................................................................... 387
3

M13: MIC Posters 3 ........................................................................................................... 389 M14: Simulation and Modelling of Medical Imaging Systems ......................................... 436 M15: Animal Imaging Instrumentation and Techniques ................................................... 438 Special Focus Workshops ........................................................................................................... 442 High Performance Medical Imaging (HPMI) 2009 ................................................................ 442 HP1 Platforms and Architectures....................................................................................... 442 HP2 APIs and Applications ............................................................................................... 443 HP3 Physics and Algorithms ............................................................................................. 444 HPP Poster Session ............................................................................................................ 446 Nuclear Techniques Applied to Cultural Heritage ................................................................. 450 CH1 Nuclear Techniques Applied to Cultural Heritage I .................................................. 450 CH2 Nuclear Techniques Applied to Cultural Heritage II ................................................ 452 Contrast in Neutron Imaging .................................................................................................. 453 CN1 Contrast in Neutron Imaging 1 .................................................................................. 453 CN2 Contrast in Neutron Imaging 2 .................................................................................. 455 Author Index ............................................................................................................................... 457 Contributions from Collaborations ......................................................................................... 457 Contributions from Individuals ............................................................................................... 459

Nuclear Science Symposium

NP: NSS Plenary Monday, Oct. 26 NP-1: 08:30-12:00 International Ballroom Center

The U.S. Nuclear Renaissance and the Challenges It Presents

W. E. Burchill President, American Nuclear Society, La Grange Park, IL, USA Dr. Burchill will discuss the factors that are producing the renaissance of nuclear power in the United States, the current status of that renaissance, and the challenges that it presents. These challenges include re-establishing the United States nuclear infrastructure, addressing political issues, building public confidence, licensing the Yucca Mountain High Level Waste Repository, and closing the nuclear fuel cycle.
NP-2:

The Intelligence Advanced Research Projects Activity (IARPA) -- What It Is and Why You Should Care

L. J. Porter Office of the Director of National Intelligence, College Park, MD, USA Dr. Porter will give an overview of the newly created Intelligence Advanced Research Project Activity (IARPA). She will describe its mission, why it was established, what hard problems it is focused on solving, and how researchers can engage with IARPA to help solve those problems.
N01: Photodetectors and Scintillation Detectors I Monday, Oct. 26 N01-1: 13:30-15:30 International Ballroom North

A. Burger1, P. Battacharya1, M. Groza1, N. Cherepy2, S. Payne2, B. Sturm2, O. Drury2, E. van Loef3, R. Howrami3, W. Higgins3, K. Shah3, J. Ramey4, L. Boatner4 1 Fisk University, Nashville, TN 37208, USA 2 Lawrence Livermore National Laboratory, Livermore, CA 94550, USA 3 Radiation Monitoring Devices, Watertown, MA 02472, USA 4 Oak Ridge National Laboratory, Oak Ridge, TN 37830, USA This presentation will review recent developments in the emerging strontium iodide scintillator technology. The topics covered will include the handling of precursors, crystal growth and crystal characterization. Results related to purification, synthesis and doping, crystal uniformity and dopant distribution as well as the crystal performance for gamma spectroscopy will be presented and discussed. SrI2(Eu) yields >100,000 photons/MeV in the Eu2+ luminescence band (435 nm central wavelength), with a decay time of ~1.2 microsec, and exhibits excellent light yield proportionality (and hence intrinsic energy resolution) that is superior to that of Cedoped lanthanum bromide [1-5]. We have demonstrated energy resolution of 2.6% at 662 keV, and we observe evidence that a significant component of the resolution broadening derives from material inhomogeneity. For this reason, we are working on improving materials purification, handling and crystal growth methods. In addition, we are developing prototype SrI2(Eu)-based gamma ray spectrometers with optimized optics and pulse processing with the goal of realizing the intrinsic energy resolution limit of SrI2(Eu) for gamma ray spectroscopy, estimated at ~2% at 662 keV. References 1. N.J. Cherepy, G. Hull, A. Drobshoff, S.A. Payne, E. van Loef, C. Wilson, K. Shah, U.N. Roy, A. Burger, L.A. Boatner, W-S Choong, W.W. Moses Strontium and Barium Iodide High Light Yield Scintillators, Appl. Phys. Lett. vol. 92, p. 083508, (2008). 2. R. Hawrami, M. Groza, Y.Cui, A. Burger, M.D Aggarwal, N. Cherepy and S.A. Payne, SrI2, a Novel Scintillator Crystal for Nuclear Isotope Identifiers, Proc. SPIE, Vol. 7079, 70790 (2008).
Acknowledgements This work was supported by the Domestic Nuclear Detection Office in the Department of Homeland Security (A. Janos), and performed under the auspices of the U.S. DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. Oak Ridge National Laboratory is managed for the U.S. DOE by UT-Battelle under contract DE-AC05-00OR22725.

Recent Developments in Strontium Iodide Detectors

N01-2:

Luminescence Centers in Ca Co-Doped LSO:Ce Single Crystals

K. Yang, C. L. Melcher Scintillation Materials Research Center, University of Tennessee, Knoxville, TN, USA Recent reports show Ca co-doping significantly reduces the charged trap populations in cerium doped lutetium oxyorthosilicate (Lu2SiO5:Ce, LSO:Ce). Thus afterglow is suppressed and scintillation decay is accelerated. In the present study, we investigate the effect of Ca co-doping on the luminescence centers in LSO:Ce by analyses of photoluminescence decay time, excitation and emission spectra at low temperature as a function of Ca concentrations. It is well-known that Ce doped LSO has two types of Ce luminescence (Ce1 and Ce2) with two distinct sets of excitation and emission spectra and decay constants. In our investigations, emission and excitation spectra of both Ce1 and Ce2 are well resolved at low temperature. These spectra indicate Ca co-doping does not alter the energy level structure of either Ce1 or Ce2 luminescence centers in LSO but changes the relative emission intensities. Photoluminescence decay studies show Ca co-doping has minimal effect on the decay of Ce1 but decreases the decay time measured at Ce2 emission wavelength. Experimental data suggest Ca co-doping reduces the relative population of Ce2 to Ce1. Thus, the luminescence from Ce2 is suppressed, which contributes to the fast scintillation decay property of Ca co-doped LSO:Ce.
N01-3:

K. Roemer1, G. Pausch1, C.-M. Herbach1, Y. Kong1, R. Lentering1, C. Plettner1, J. Stein1, M. Moszyski2, L. Swiderski2, T. Szczniak2 1 ICx Technologies GmbH, Solingen, Germany 2 Soltan Institute for Nuclear Studies, wierk-Otwock, Poland Scintillator-based Compton cameras for remote localization and identification of radio nuclides require scatter detectors made of low-Z materials. The energy resolution of such detectors in a range dominated by Compton scattering is a crucial parameter. It has to be known for performance estimates, and it must be quantified and optimized for detector designs to be used in real systems, but it is hard to measure. The Compton Coincidence Technique (CCT) appears as the best method for reliable and direct measurements, but appropriate facilities are expensive. This paper suggests and investigates a modified CCT which provides less expensive means for qualifying of scatter detectors in a reasonable time frame. The assembly consists of a single HPGe detector, the scatter detector to be investigated, and one or more common gamma sources in close geometry. Pulse height and timing information from both detectors is gathered by multi-parameter data acquisition. Coincidences of both detectors are due to a plurality of Compton scattering angles and corresponding energy transfers. A thorough data analysis then allows extracting the detector resolution for multiple energies from data sets measured within hours. Results obtained for NaI, plastic, and CaF2 scatter detectors will be presented and discussed.
N01-4: Concentration Dependence of Nonproportionality of LaBr3(Ce), SrI2(Eu), and Other Scintillator Crystals

A Technique for Measuring the Energy Resolution of Low-Z Scintillators

S. Payne1, L. Ahle1, S. Sheets1, N. Cherepy1, W. Moses2, G. Bizarri2, W.-S. Choong2 1 LLNL, Livermore, CA, USA 2 LBNL, Berkeley, CA, USA

We have explored the nonproportionality of numerous crystals including: alkali halides [NaI(Tl), CsI doped with Tl and Na]; LaBr3(Ce) as a function of Ce-doping from 0.5 30%, and other crystals such SrI2(Eu), YAG(Ce), YAP(Ce), and LaCl3(Ce). We have found that we able to model the nonproportionality curves on the basis of three parameters: (dE/dx)ONS which accounts for the attraction between electrons and holes; EXC which is the maximal fraction of excitons that are created following the cascade; and (dE/dx)BIRKS which allows for exciton-exciton annihilation (and the concomitant fall in light yield at low electron energy). During the talk we will develop our model of nonproportionality to explain the electron light yield curves, and also show how Landau fluctuations can be interpreted to explain how the intrinsic resolution can be deduced. Interestingly, there is a small but reproducible change in the LaBr3(Ce) nonproportionality for different concentrations leading a prediction of slightly improved resolution at higher doping - due to diminished exciton creation and annihilation processes.
This work is supported by the National Nuclear Security Administration, Office of Defense Nuclear Nonproliferation, Office of Nonproliferation Research and Development (NA-22) of the U.S. Department of Energy and the Domestic Nuclear Detection Office of the Department of Homeland Security. Thanks to Eric Mattmann and Bruno Aleonard from St. Gobain for the loan of the LaBr3(Ce), BrilLanCe 380 crystals. We wish to thank Lynn Boatner (Oak Ridge National Lab), Arnold Burger (Fisk University) and Kanai Shah (RMD, Inc.) for providing crystals. This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC5207NA27344 and by Lawrence Berkeley National Laboratory under Contract No. DE-AC02-05CH11231.

N01-5:

Scintillation Properties of Cs2LiLaCl6

J. Glodo, E. V. D. van Loef, A. Churilov, W. M. Higgins, R. Hawrami, K. S. Shah Radiation Monitoring Devices, Inc., Watertown, MA, USA The elpasolite family is rich in compositions that are very promising for radiation detection. Some of these compositions contain Li ions. Since Li atoms capture neutrons, releasing at the same time charge particles as a byproduct of this process, Li based compositions may be used for thermal neutron detection. The first of the elpasolite crystals proposed for neutron detection was Cs2LiYCl6:Ce (CLYC). Recently, we have been investigating a similar composition: Ce doped Cs2LiLaCl6 (CLLC) for neutron and gamma detection systems. This material shows significantly higher light output than CLYC. While, typically CLYC provides about 20,000 photons/MeV, the light output of CLLC can reach 40,000 photons/MeV. High light output and good linearity allow CLLC to achieve excellent energy resolution of 3.4% FWHM at 662 keV. The thermal neutron detection is readily achieved with CLLC. The gamma equivalent energy for thermal neutron full energy peak is about 3 MeV, thus the light output per neutron can be as high as 120,000 photons. Similar to CLYC, CLLC also exhibits Core-to-Valence Luminescence that is only observed under gamma excitation. This allows pulse shape discrimination to be effectively implemented for this material in order to distinguish between gamma and neutron radiation. In this communication we report our recent results on this material, including its properties versus Ce concentration.
N01-6:

N. Cherepy1, J. Kuntz1, J. Roberts1, T. Tillotson1, S. Fisher1, R. Sanner1, W. Ralph1, R. Gaume2, O. Drury1, S. Payne1 1 Lawrence Livermore National Laboratory, Livermore, CA, USA 2 Stanford University, Stanford, CA, USA

Fabrication and Characterization of Transparent Ceramic Garnet Scintillators for Gamma Ray Spectroscopy

Transparent ceramics offer an alternative to single crystal scintillators for gamma ray spectroscopy. We have developed ceramics processing methods involving vacuum sintering and hot pressing for facile fabrication of phase-pure transparent ceramics. Gadolinium-based garnets evidence a large photopeak fraction, however, only a few chemical compositions offer adequate phase stability to be formed as large volume transparent optics. Additionally, materials purity is an important consideration in reducing undesirable afterglow. We have identified several Cerium-doped Gd-based garnet compositions that yield >40,000 Photons/MeV, acceptable phase purity and optical quality ( SCATT <0.1 cm-1). The luminescence spectra for Cerium-doped Gadolinium Gallium Aluminum Garnet, GGAG(Ce) and Gadolinium Yttrium Aluminum Garnet, GYGAG(Ce) are centered at 560 nm. In contrast, the peak for Gadolinium Yttrium Scandium Aluminum Garnet, GYSAG(Ce) is at ~590 nm, in a spectral range that is significantly offset from the best efficiency range for bialkali PMTs. While GGAG(Ce) has the highest Zeff, the garnet phase is not sufficiently stable for easy fabrication. For this reason, we explored addition of Yttrium, and we subsequently found that a limited compositional range of GYGAG(Ce) can be formed as pure phase garnet. We have measured energy resolution for gamma ray spectroscopy at 4-7% at 662 keV for 0.5-1.5 cm3 Gd-garnet ceramics. Pulsed x-ray characterization of afterglow and methods for afterglow reduction by improved synthesis and annealing treatments will be described. In addition, optical scatter criteria as a function of scintillator size will be discussed.
This work was supported in part by the Domestic Nuclear Detection Office in the Department of Homeland Security (A. Janos) and in part by the National Nuclear Security Administration, Office of Nonproliferation Research and Development (NA-22, D. Beach) of the U.S. DOE under Contract DE-AC03-76SF00098, and performed under the auspices of the U.S. DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. N01-7:

M. Moszynski1, T. Szczesniak1, M. Kapusta2, M. Szawlowski1, J. Iwanowska1, M. Gierlik1, A. Syntfeld-Kazuch1, L. Swiderski1, C. Melcher3, L. Eriksson4 1 Soltan Institute for Nuclear Studies, Otwock-Swierk, Poland 2 Photonis, Brive, France 3 University of Tennessee, Konxville, TN, USA 4 Siemens Medical Solutions, Konxville, TN, USA The discrepancy in the scintillation light output observed recently in measurements with different photomultipliers (PMT) prompted the study of the PMT devices used in evaluation of the light output. The photoelectron number measured by the classical method using the single-photoelectron response for PMT gain evaluation was compared to that determined using the pulse height resolution of the multi- photoelectron peak from the LED light pulses. Tests performed with a number of photomultipliers like Photonis XP2020Q, XP20D0, XP3060, XP5200, XP5301 and XP5302, and Hamamatsu R6231, R9420, R5320, R3998, and test of Hamamatsu R2059 at the Tennessee University showed for most of modern PMTs an excess of photoelectrons measured by evaluation of the single-photoelectron peak position over that calculated from the analysis of the pulse height resolution of the peaks from LED pulses. This results in discrepancy in light output evaluation and an erroneous 7

Characterization of Scintillators by Modern Photomultipliers - a New Source of Errors

light output. Particularly, the light output of BGO and LSO determined with R6231 and R2095, using the pulse height resolution method, is comparable to those measured with XP2020Q and S3590-18 pin photodiode. Several different tests are reported addressed to understand the origin of the excess of photoelectrons measured by the classical comparison to the single photoelectrons. Moreover, the implications on the characterization of new scintillators by modern PMTs are discussed in relation to the light output and the analysis of energy resolution.
N02: Semiconductor Detectors I: Silicon Detectors and Applications Monday, Oct. 26 N02-1: 13:30-15:30 International Ballroom Center

The LHCb Silicon Tracker Commissioning and First Data

A. Bchler Physik-Institut, Universitt Zrich, Zrich, Switzerland On behalf of the Silicon Tracker group The LHCb Silicon Tracker (ST) has been installed in the LHCb detector and was successfully put into operation in summer 2008. The ST plays a crucial role in reconstructing the trajectories of charged particles. It consists of two silicon micro-strip detectors, the Tracker Turicensis upstream of the LHCb dipole magnet and the Inner Tracker downstream of the magnet. Insights gained during the exciting commissioning phase, the problems encountered and their solutions are presented. The data collected during LHC injection tests in 2008 and 2009 allowed us to make a first measurement of the detector performance. The data is used to determine the mechanical alignment of the ST and to get a first estimate of detector resolutions and efficiencies. Furthermore, these tests have been extremely useful for tuning the operational parameters of the different parts of the Silicon Tracker.

A 1 cm2 Multi-Linear Silicon Drift Detector for 2D X-Ray Spectroscopic Imaging and Compton Scattering
N02-2:

A. Castoldi1, C. Guazzoni1, R. Hartmann2, L. Strueder3 1 Dip. Elettronica e Informazione, Politecnico di Milano and INFN, Milano, Italy 2 PNSensor GmbH and MPI Halbleiterlabor, Munich, Germany 3 Max Planck Institut fur Extraterrestriche Physik and Halbleiterlabor and Universitat Siegen, Munich, Germany

Multi-Linear Silicon Drift Detectors (ML-SDDs) are a recent evolution of silicon drift detectors. In these detectors the signal electrons generated by the interaction are confined within parallel drifting columns and transported towards point-like anodes by the electrostatic field. The multi-linear transport mechanism based on electrons drift accounts for a dramatic reduction in the number of channels required for true 2D position sensing (i.e. no. of channels equal to the square root of the pixels). Moreover it naturally allows the connection of the front-end electronics aside the detector chip thus simplifying the interconnection issue and it leads to readout times of only few microseconds. The deposited energy can be measured with spectroscopic resolution at (or very close to) room temperature thanks to point-like anodes of very low capacitance (100 fF) with on-chip JFETs. Therefore ML-SDDs open the way to time-resolved imaging and spectroscopy of X-rays and Compton electrons in addition to charged particle tracking. The excellent energy resolution and the relatively small Doppler broadening of silicon makes a ML-SDD detector an ideal candidate to reconstruct the original location of the gamma-ray with sub-millimeter position resolution. A new design of a 1 cm2 ML-SDD has been carried out and first detector prototypes were produced at the Halbleiterlabor of the Max Planck Institut in cooperation with PNSensor GmbH. The first part of the talk will be devoted to the discussion of the spectroscopic performance and of the timing properties of the detector chip. The second part will show applications to X-ray imaging and electrons tracking.
N02-3:

Pixel Sensors Based on High Density Microelectronic Technologies for the Layer 0 of the SuperB Silicon Vertex Tracker

V. Re Department of Industrial Engineering, University of Bergamo, Bergamo, Italy On behalf of the VIPIX Collaboration The paper reviews the development of monolithic and vertically integrated silicon pixel sensors in view of their application in the innermost layer of the vertex detector of the Super B-Factory. The first generation of Deep N-Well (DNW) MAPS with on-pixel data sparsification and time stamping is now available and was tested in a beam for the first time in September 2008. The analysis of the experimental results provides an assessment of the Deep N-Well MAPS potential with respect to the SuperB Layer0 specs in terms of detection efficiency, material budget and capability of handling high data rates. The paper also discusses the way R&D is progressing towards the design of a full size Layer0 detector system. The sensors could take advantage from technological advances in microelectronic industry, such as vertical integration, with potential benefit for various device features, 8

from the charge collection properties to the digital readout architecture. The design of the first prototype of a vertically integrated pixel sensor for the SuperB Layer0 is presented in the paper.
N02-4:

G.-F. Dalla Betta1,2, M. Boscardin3, L. Bosisio4,5, P. Gabos1,2, M. Koehler6, U. Parzefall6, C. Piemonte3, S. Ronchin3, L. A. M. Wiik6, A. Zoboli1,2, N. Zorzi3 1 Dipartimento di Ingegneria e Scienza dell'Informazione, Universita' di Trento, Trento, Italy 2 Gruppo Collegato di Trento, INFN, Sezione di Padova, Trento, Italy 3 Centro Materiali e Microsistemi, Fondazione Bruno Kessler (FBK-irst), Trento, Italy 4 Dipartimento di Fisica, Universita' di Trieste, Trieste, Italy 5 INFN, Sezione di Trieste, Trieste, Italy 6 Institute of Physics, University of Freiburg, Freiburg i.B., Germany We report on the characterization of 3D-DDTC (Double-sided Double-Type-Column) detectors fabricated at FBK-irst on p-type silicon substrates. Compared to the early prototypes made with this technological approach, these new detectors have optimized column depths, making the electric field configuration comparable to that of Full-3D detectors, so that a very good charge collection efficiency is expected, as confirmed by TCAD simulations. Results from the electrical characterization are very promising, with low depletion voltage (a few V) and very low leakage current (below 1pA/column). Functional tests with position resolved laser and beta source setups are under way, and a beam test has also been planned for the summer. Selected results from the characterization of these optimized 3D-DDTC detectors will be reported at the Conference.
This work has been partially supported by the Italian National Institute for Nuclear Physics (INFN), 5th Scientific Commission, under the project TREDI, and by the Autonomous Province of Trento. N02-5:

Characterization of 3D-DDTC Detectors on P-Type Substrates

S. Kleinfelder1, L. Jin2, S. Li1, X. Nguyen-Hu3 1 Electrical Engineering, University of California, Irvine, CA, USA 2 Physics, University of California, San Diego, CA, USA 3 Chemistry, University of California, San Diego, CA, USA

A High-Performance Charged-Particle CMOS Image Sensor with Per-Column Analog to Digital Conversion

A CMOS image sensor for charged-particle imaging with per-column analog signal processing and 10-bit ADCs has been designed and tested. A refined design including parallel data conversion enables up to 390 frames/s operation with sufficient sensitivity and noise performance for imaging using single-electron counting. Extensive circuit details plus measured performance will be presented, including modulation transfer-function (MTF) measurements, plus a discussion of imaging using single-electron counting mode. Measurements show that the charge to voltage conversion ratio is 0.024 mV per ionization electron at -10 C, determined using 55Fe X-ray tests. The dark current is 2.54 e-/ms, corresponding to 61 mV/s at 26 C, and 0.29 e-/ms or 7 mV/s at -19 C. The pixel and readout noise after CDS and pedestal removal is 19.4 e-, corresponding to 0.465 mV RMS at -29 C, and 22.1 e- or 0.53 mV RMS at 16 C. The average collected signal is 2,770 e-, integrating to 66.4 mV, per incident 200 keV electron, and 4,650 e- or 111.5 mV per 120 keV incident electron. The most probably response of a single electron event, when measuring the peak (central) pixel, is 338 e- or 8.1 mV, using 200 keV electron beam. The sensor has sufficiently low noise (S/N ratio of ~17.5) that it is possible to count single incident electrons with excellent efficiency. We have found that images acquired using a counting mode have advantages in contrast and sharpness over normal continuous image integration. However, imaging in single electron mode requires a low electron flux, short per-frame exposure times, and the acquisition of many hundreds or thousands of frames. Due to its highly-parallel digitization, the maximum frame rate of this chip in is 217 frames/s at 10 bits and with 8-bit A/D conversion (more than sufficient for electron counting) the frame rate reaches 390 frames/s.
N02-6:

A. M. Conway1, L. F. Voss1, C. E. Reinhardt1, R. T. Graff1, T. F. Wang1, R. J. Nikolic1, N. Deo2, C. L. Cheung2 1 Lawrence Livermore National Laboratory, Livermore, CA, USA 2 Chemistry, University of Nebraska, Lincoln, NE, USA Solid state thermal neutron detectors are desired to replace the current He-3 tube based technology, which suffer fieldability issues including high voltage requirements and sensitivity to microphonics. There have been several solid state thermal neutron detector concepts developed recently. This work reports our most recent experimental results of pillar structured silicon p+, intrinsic, n+ (PIN) diode array with an aspect ratio of 13:1 for thermal neutron detection. The detector fabrication process, electrical characteristics and neutron spectra for 2 m wide pillars, 4 m pitch, and height of 26 m are reported. A thermal

Si Based Pillar Structured Thermal Neutron Detectors

neutron efficiency of 20.1% was measured, which is the largest efficiency reported to date for a semiconductor based thermal neutron detector using B-10 as the converter material.
This work was partially supported by the Domestic Nuclear Detection Office in the Department of Homeland Security (Alan Janos) and this work was performed under the auspices of the U.S. DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNLABS-412870. N02-7:

Diamond, 3D-, and Planar-Silicon Pixel Detectors for Super Large Hadron Collider

F. Huegging, J.-W. Tsung, N. Wermes Physikalisches Institut, University of Bonn, Bonn, Germany Radiation-Hard detectors are in demand for the super Large Hadron Collider (sLHC), because the irradiation will be 10 times larger than LHC. Diamond and 3D-Silicon are promising radiation-hard sensors for the ATLAS pixel tracking detectors. In this research, their performances are compared to each other using the same pixel readout technique and the same characterization method. The sensors are bump-bonded to the ATLAS FE-I3 pixel readout chips. Their leakage current, noise, and threshold of signal are measured using the standard ATLAS pixel test setup. Charge collection performance can be assessed with radioactive source tests, using -ray from 90Sr. The number of generated electron and hole pairs is measured, and then the charge collection efficiency and signal to noise ratio are estimated. The sensors are also tested with the 100 GeV pion beam at CERN. To reconstruct the particle trajectory, a high resolution telescope is operated to determine the hit position on the pixel. Therefore, the position resolution, charge collection properties, and signal to noise ratios can be evaluated precisely. 5 types of sensors: sCVD Diamond, pCVD Diamond, 3D-silicon, n-on-p and n-on-n planar-Silicon sensors are characterized and are compared to each other.
N03: Analog and Digital Circuits I Monday, Oct. 26 N03-1: 13:30-15:30 Grand Ballroom 7

S. Callier1, F. Dulucq1, R. Fabbri2, C. de La Taille1, B. Lutz2, G. Martin-Chassard1, L. Raux1, W. Shen3 1 Laboratoire de l'Accelerateur Lineaire, Orsay, France 2 DESY, Hamburg, Germany 3 University of Heidelberg, Heidelberg, Germany

Silicon Photomultiplier Integrated Readout Chip (SPIROC) for the ILC: Measurements and Possible Further Development

The SPIROC chip is a dedicated very front-end electronics for an ILC technical prototype hadronic calorimeter with Silicon Photomultiplier (or MPPC) readout. This ASIC is due to equip a 2,000-channel demonstrator in 2009. The SPIROC chip is the successor of the ILC_SiPM ASIC presently used for the ILC AHCAL physics prototype incorporating additional features like autotriggering, pipelines, digitization as well as power pulsing. Realized in 0.35um SiGe technology it is designed in order to fulfill ILC final detector requirements of large dynamic range,low noise, low power consumption, high precision and large channel numbers. The SPIROC is a 36-channel chip. Each channel has bi-gain amplification,auto-triggering capability, a 16-bit depth analog memory array and a 12-bit Wilkinson ADC. It allows time and charge measurements at the same time with digitized data results. The digitization is controlled and read out by the digital part of the chip. After the submission in June 2007, extensive measurements have been carried out to characterize the chip. Results on linearity, noise, triggering, timing capability and the A/D interface etc. will be presented. The chip has been proven to be successful in calorimeter calibration as well as real physics experiments. Besides of the affirming measurement results, possible improvements have been proposed in order to make the chip even more versatile in dealing with a large variety of Silicon Photomultipliers. These new ideas will also be discussed.
N03-2:

A 12-Bit 35MS/s Pipelined ADC with a Dynamic Element Matching Correction for ILC / CALICE Integrated Readout

F. Rarbi LPSC - IN2P3, Grenoble, France On behalf of the CALICE collaboration A low power 12 bits analog to digital converter is a critical part of a fully integrated readout system for the next ILC ECAL. We present here a new design of 12-bit ADC up to 35-MS/s using a pipelined architecture in a CMOS 0.35 m process. The first frontend stage of 2.5 bits includes an efficient dynamic element matching scheme permitting to average its gain errors. The back-end converter is a set of seven 1.5 bit stages followed by a 3 bit full flash. The dynamic range covered is 2V. The analog part of the

10

converter can be quickly (in a couple of s) switched to a standby mode that reduces the DC power dissipation. The size of this converters layout including the output pads is 1.4mm*1.3mm, and the total power dissipation is only 45mW.
N03-3:

Radiation-Hard ASICS for SLHC Optical Data Transmission

K. K. Gan Physics, The Ohio State University, Columbus, OH, USA High-speed data transmission in a high radiation environment poses an immense challenge to detector design. We investigate the feasibility of using optical links for the silicon trackers of the ATLAS experiment at the planned upgrade of the Large Hadron Collider (LHC) at CERN (Geneva). The planned upgrade of the LHC with ten times higher collision rate will produce a similar increase in the radiation. One possibility for the optical transmission is to use VCSEL arrays operating at 850 nm to transmit optical signals while using PIN arrays to convert the optical signals into electrical signals. We have designed a high-speed driver for VCSELs and a receiver/decoder for the signal received at the PIN diode. In addition, we also designed a clock multiplier to produce a higher frequency clock to serialize the data for transmission. These chips were designed using a 130 nm CMOS process to enhance the radiation-hardness. We irradiated the chips with 24 GeV/c protons at CERN to the SLHC dosage of 70 Mrad, including a 50% safety factor. We observed no significant degradation except in the VCSEL driver. Post-irradiation analysis indicates that there is a significant threshold shift in the PMOS transistors fabricated in the thick oxide technology for the operation at 2.5 V to drive the VCSEL. We also studied the single event upset (SEU) rate of the receiver and clock multiplier. We will present the results of the study.
N03-4:

S. Martoiu1, G. Dellacasa1, S. Garbolino1, F. Marchetto1, G. Mazza1, A. Rivetti1, P. Jarron2, A. Ceccucci2, J. Kaplon2, P. Riedler2, M. Noy2, S. Tiuraniemi2, M. Fiorini3, A. Cotta Ramusino3, E. Martin Albarran4 1 INFN, Torino, Italy 2 CERN, Geneva, Switzerland 3 INFN, Ferrara, Italy 4 Universite' Catholique de Louvain, Louvain la Neuve, Belgium We describe a prototype of a silicon-pixel front-end ASIC, designed to meet the requirements of the Gigatracker beam spectrometer of the NA62 experiment at CERN SPS. The beam spectrometer consists of three hybrid silicon pixel detector stations which have to sustain high and non-uniform beam rate (up to 1.5 MHz/mm2 in the hot center and 0.8-1.0 GHz in total). In addition, it provides precise momentum, time and angular measurements on every single track of the secondary 75 GeV/c hadron beam with a timing precision of 150 ps (rms). The front-end solution described here integrates in the 300m x 300m pixel area a Constant-Fraction Discriminator (CFD) to compensate for the time-walk error, a low-power Time-to-Digital converter (TDC) and a 4-event deep mixed digital and analog buffer for local data de-randomisation. The chip contains a 15x7 pixel matrix which replicates the operating conditions of the final Gigatracker readout-chip, and can be bump-bonded to a sensor. The work presents the architecture, design and measured performance of the prototype chip.
N03-5:

A Pixel Front-End ASIC in 0.13 m CMOS for the NA62 Experiment with on Pixel 100ps Time-toDigital Conversion

G. Traversi1,2, L. Gaioni2,3, M. Manghisoni1,2, L. Ratti2,3, V. Re1,2 1 Department of Industrial Engineering, University of Bergamo, Dalmine (BG), Italy 2 INFN Pavia, Pavia, Italy 3 Department of Electronics, University of Pavia, Pavia, Italy

Charge Signal Processors in Sparse Readout CMOS MAPS and Hybrid Pixel Sensors for the SuperB Layer0

This work will discuss the design of analog circuits for processing the signals from deep n-well monolithic CMOS sensors and from high resistivity substrate pixel detectors. Several options for the Layer0 design of SuperB are being studied. Hybrid pixel detectors are nowadays a robust and mature technology for the innermost vertex detector layer, but they provide a relatively large material budget which can make them marginal for the foreseen application. CMOS MAPS technology has the potential for providing very thin detectors since the sensor and the readout electronics are integrated in the same substrate. Recently, a very promising approach based on the use of a vertically integrated CMOS technology has also been considered. Various solutions complying with different S/N ratio and detector capacitance constraints have been studied and implemented in a planar 130nm CMOS technology and in a 130nm CMOS technology with vertical integration capabilities. This paper intends to describe and compare the features of the different options by means of simulations and experimental results.

11

N03-6:

A. G. Antonacci1, J. L. Britton2, S. C. Bunch3, C. L. Britton4, M. N. Ericson4, B. J. Blalock5, R. Chun5, R. Greenwell5, D. S. McGregor6, L. Crow7 1 Texas Instruments, Inc., Dallas, TX, USA 2 Texas Instruments, Inc., Knoxville, TN, USA 3 Ametek, Inc., Oak Ridge, TN, USA 4 Oak Ridge National Laboratory, Oak Ridge, TN, USA 5 EECS, The University of Tennessee, Knoxville, TN, USA 6 Kansas State University, Manhattan, KS, USA 7 The Spallation Neutron Source, Oak Ridge, TN, USA

PATARA III: a 64-Channel Solid-State Neutron Detector Readout System with Integrated Analog and Digital Processing for the SNS

The Spallation Neutron Source (SNS) at the Oak Ridge National Laboratory (ORNL) in Oak Ridge, Tennessee, is the worlds most intense pulsed accelerator-based neutron source. This research facility will require a data acquisition system capable of acquiring data from a neutron detector array with spatial resolutions (pixel widths) on the order of 100um to 500um. The pulse rate of the neutron source requires that the read-out electronics are to obtain a response time of less than 10usec. The first prototype chip for such a system, named Patara, integrated 16 channels of a charge sensitive front-end amplifier and a semiGaussian shaper with baseline restore circuitry. Patara II, described in this paper, involved the addition of mixed signal circuitry and a digital read/write interface in order to operate at a system level. Patara II successfully integrated a 64-channel neutron sensor input that will synchronously or asynchronously output the binary encoded pixel event via Low Voltage Differential Signaling (LVDS) transmission. The final die size was 6.5mm by 6.7mm. This paper presents the design and measured performance characteristics of the completed chip.
N03-7:

J. R. Hoff1, T. Zimmerman1, R. Yarema1, J. Kapustinsky2, M. Brooks2 1 Fermilab, Batavia, IL, USA 2 Los Alamos National Laboratory, Los Alamos, NM, USA

FPHX: a New Silicon Strip Readout Chip for the Phenix Experiment at RHIC

The FPHX chip is a silicon strip readout chip developed at Fermilab for use in the end caps of the Phenix experiment at RHIC. Each front end consists of two stages, an integrator AC coupled to a shaper, followed by a 3-bit analog-to-digital converter. The backend is a novel architecture in three stages that permits dead-timeless operation. A slow controller provides an interface for all on-chip programmable functions. This chip has been fabricated in the 0.25um TSMC process. All of its functionality including the analog front-end, the digital back-end, and the slow controller has been verified experimentally.
N04: Nuclear Measurements and Monitoring Techniques Monday, Oct. 26 N04-1: 13:30-15:30 Grand Ballroom 8

M. Sweany1, A. Bernstein2, N. Bowden2, S. Dazeley2, S. Ouedraogo2, R. Svoboda1 1 Physics, University of California, Davis, Davis, CA, USA 2 Physics, Lawrence Livermore National Laboratory, Livermore, CA, USA Fission events from Special Nuclear Material (SNM) produce both neutrons and high energy gamma-rays, both of which can penetrate through shielding. The neutron multiplicity can be approximated by a Possonian distribution with an average of 2 to 3 neutrons per fission event. Time-correlated neutrons, therefore, are a strong indication of a fission event from SNM. After successfully demonstrating neutron detection with a 250-liter gadolinium-doped water Cherenkov detector, the Advanced Detectors Group (ADG) at LLNL has recently begun full operation of a 3800-liter detector. This new detector offers an efficient and cost-effective way to detect SNM with large solid angle coverage. The detector construction and calibration will be presented, along with early results and future plans.
N04-2:

Water Cherenkov Based Neutron Detection

A. Haefer1, G. A. Warren1, W. Bertozzi2, R. D. Hasty2, A. V. Klimenko2, S. E. Korbly2, R. J. Ledoux2, W. H. Park2 1 Pacific Northwest National Laboratory, Richland, WA, USA 2 Passport Systems, Inc, Billerica, MA, USA There is a growing interest in the use of nuclear resonance fluorescence (NRF) to address a wide variety of applications involving isotopic-specific detection and characterization. In evaluating the feasibility of NRF for a particular application, it is important to 12

Experimental Observation of the Background in Nuclear Resonance Fluorescence

determine both the strength of the NRF signal and of the background. While determining the strength of the signal is relatively straightforward, calculating the background is significantly more complicated due to the multi-step nature of the background. In this presentation, experimental observations of the background for nuclear resonance uorescence are reported. Measurements were conducted on four samples, graphite, aluminum, steel and lead, with a bremsstrahlung end-point energy of 5.3 MeV and a large high purity germanium detector at 110 degrees. It is observed that in the 1.5 to 3.0 MeV range the background scales with cube of the atomic number of the material. Above 3.0 MeV, simulations and measurements strongly diverge, suggesting either an unaccounted process in the simulations or unaddressed background in the measurements. This effort was funded by the U.S. Department of Energy Office of Defense Nuclear Nonproliferation, Office of Nonproliferation Research and Development (NA-22) and the U.S. Department of Homeland Security Domestic Nuclear Detection Office (DNDO).
N04-3:

S. McConchie1, P. Hausladen1, J. Mihalczo1, J. Jones2, S. Watson2, B. Blackburn3 1 Oak Ridge National Laboratory, Oak Ridge, TN, USA 2 Idaho National Laboratory, Idaho Falls, ID, USA 3 Raytheon Technical Services, Waltham, MA, USA

Pulsed Photonuclear Time Correlation Measurements of Shielded HEU Oxide Fuel Pins and DU Metal

Pulsed photonuclear time correlation measurements have been performed on HEU oxide fuel pins and DU metal with a 2-12 MeV Varitron electron accelerator at the Idaho National Laboratory Power Burst Facility. The HEU sample (46 wt%) consisted of 8 kg of U-235 and was surrounded by Borobond4 neutron absorber in an ES-3100 container. These measurements investigated the characteristic neutron response in various polyethylene, lead, and iron shielding configurations. The accelerator was operated at an endpoint photon energy of 10 MeV, a 3 microsecond pulse width, and a repetition rate of 125 Hz. An array of He-3 detectors was used to detect neutrons emitted from each sample. These neutrons were counted during and after the pulse with the Nuclear Material Identification System (NMIS) processor and used to measure the neutron differential die-way and coincidence distributions. Analysis of the data demonstrates (1) a failure mode of the differential die-way technique in which HEU shows no die-away behavior in the new ES-3100 shipping container and (2) the ability to correctly bound the multiplications of the HEU and DU samples. This paper presents results from the measurements.
N04-4:

W. J. Walsh1, S. D. Clarke1, S. A. Pozzi1, N. Cunningham2, S. Banerjee2, D. Umstadter2 1 Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA 2 Physics and Astronomy, University of Nebraska, Linclon, NE, USA

Application of a Laser-Wakefield Driven Monochromatic X-Ray Source to Nuclear Resonance Fluorescence

The Diocles laser facility at The University of Nebraska-Lincoln (UNL) is a 100-TW, 30-fs pulsed Ti:sapphire laser system that is routinely used to accelerate electron beams to high energy. These electrons may be used to scatter laser light to produce monochromatic X-rays with energy and spectral width that are directly related to those of the electron beam. This novel approach presents the possibility of producing a tunable, monochromatic X-ray source that has promising applications in nuclear material detection. By using a monochromatic light source at the appropriate energy it is possible to excite nuclei by resonance photon absorption. The excited nucleus then de-excites by emitting a photon of nearly identical energy. This process of excitation and de-excitation is known as nuclear resonance fluorescence (NRF). It is possible to analyze these emissions and determine the identity of an unknown material because every nucleus has distinct energy levels. This technique is of particular interest to the detection of special nuclear material (SNM) where isotopic characterization is paramount. In addition to photons emitted by resonance de-excitation, there will also be a significant amount of backscattered photons that will distort the NRF signal. The backscattered spectra must be characterized in order to design a detection system that maximizes the material identification probability. The results will include a comparison between the backscattered spectrum for U-238 and U-235. In addition, we will examine compare the performance of this tunable X-ray source with a conventional bremsstrahlung source. The results will be useful to evaluate the performance of each source for the detection of SNM.
N04-5:

B. Obryk1, P. Bilski1, M. Budzanowski1, M. Fuerstner2, M. Glaser2, P. Olko1 1 Institute of Nuclear Physics PAN (IFJ), Krakow, Poland 2 CERN (European Organization for Nuclear Research), Geneva, Switzerland

Behaviour of Various Types of LiF Detectors at High-Energy Mixed Fields

The thermoluminescent (TL) detectors are well developed technology in the field of passive radiation sensors; many different types of them are already widely used in the range of doses from a few micrograys up to several kilograys. LiF:Mg,Ti and LiF:Mg,Cu,P are well known standards of thermoluminescent dosimetry. While both of them are composed of the same base material, several of their properties are significantly different. A new method of thermoluminescent (TL) measurement of radiation doses ranging from micrograys up to a megagray has been 13

recently developed at IFJ. This method is based on a newly-discovered behavior of LiF:Mg,Cu,P detectors at high and ultra-high doses. While the shape of the glowcurve at doses ranging from microgray to a kilogray is practically identical, significant changes are observed at higher doses. After discovery of the unexpected behavior of LiF:Mg,Cu,P detectors at very high doses, a detailed study on the response of these sensors to specific types of radiation has been done. To enable an application of this method to high-energy mixed radiation fields, tests of six different LiF-based TL detectors have been carried out at the CERN-IRRAD6 facility that provides mixed high-energy particles. The response of standard LiF:Mg,Ti, 7LiF:Mg,Ti, 6LiF:Mg,Ti and high-sensitive LiF:Mg,Cu,P, 7LiF:Mg,Cu,P, 6LiF:Mg,Cu,P, was investigated. All detectors were produced at the Institute of Nuclear Physics (IFJ), Krakw. LiF detectors are sensitive to slow neutrons, their response to neutrons being enhanced by 6Li-enriched lithium or suppressed by using lithium consisting entirely of 7Li. Sensors were exposed over five months, the results are presented here.
This work was partly supported by a research project from the Polish Ministry of Science over the years 2008-2009 (No. N404156434). N04-6:

C. J. Stapels1, E. B. Johnson1, E. Chapman1, S. Mukhopadhyay1, F. L. Augustine2, M. R. Squillante1, J. F. Christian1 1 Radiaiton Monitoring Devices, Watertown, MA, USA 2 Augustine Engineering, Encinitas, CA, USA The ability to develop high-gain optical detectors in commercial CMOS creates the potential for complete integrated detector solutions. Circuit development in a CMOS platform can allow integration of pulse conditioning, processing and storage elements. This work describes the operation and evaluation of such a detector-on-a-chip. The chip consists of 300 signal-conditioned elements, a seven-channel level sorter and slow clock, and associated 16-bit counters. We evaluate the performance of this device in terms of dynamic range of energy, dynamic range of human equivalent dose, and fit for potential applications including radiation monitoring and personnel dosimetry.
N04-7:

Digital Detector on a Chip for Radiation Monitoring and Personnel Dosimetry

R. Coulon1, S. Normand1, G. Ban2, E. Barat3, T. Montagu3, T. Dautremer3, H.-P. Brau4, V. Dumarcher5, J.-L. Portier6, L. Barbot7, T. Domenech1, K. Boudergui1, A.-M. Frelin1, M. Michel1, V. Kondrasovs1, J.-M. Bourbotte1, P. Jousset1 1 CEA, LIST, Laboratoire Capteurs et Architectures Electroniques, Saclay, France 2 ENSICAEN, Caen, France 3 CEA, LIST, Laboratoir Processus Stochastiques et Spectres, Saclay, France 4 ICSM, Marcoule, Bagnols sur Ceze, France 5 AREVA NP, SET, Bollne, France 6 CEA, Marcoule, Bagnols sur Ceze, France 7 CEA, Cadarache, Saint-Paul-Lez-Durance, France On sodium fast reactors, power is usually measured by heat balance on water to vapor heat exchangers which is correlated with the ex-core neutron flux measurements. The ex-core neutron power measurement is done by fission chambers (five on the French SFR Phnix reactor) with several activity ranges settled at the bottom of the primary vessel to cover the whole neutron flux range of more than 11 decades. This measurement instantly estimates the neutron power but induces some shift problems. The clad failure detection is done by gamma measurement on argon cover gas sampling and neutron measurement on primary sodium sample. This work deals with the use of gamma spectrometry for fourth generation SFR power monitoring and clad failure detection. Usually gamma spectrometers could not manage on-line application but recent improvements in this research field may improve it. The Adonis digital signal processing prototype gives a time function response and is able to do spectrometry measurement with metrological grade under high gamma flux impinging the HPGe detector, with also high and fast activity transient. Previous works on Pressurized Water Reactors (PWR) show that gamma emitter concentration in primary coolant is also directly correlated with reactor neutron power. On SFR, the use of short decay period gamma emitters as the 20F and 23Ne tagging agent will allow a very fast response system without thermal hydraulic distortion effects. The study is divided in two parts. First part deals with a simulation study based on physical calculations and MCNP simulations to predict sensor signal as a function of neutron power. The second part is an experimental test under preparation at the French SFR Phnix. Measuring activation and fission products contained in primary sodium, the adaptive Adonis system set on primary coolant sample could be an innovating neutron power monitoring and clad failure detection system for SFR.

Sodium Fast Reactor Power Monitoring and Clad Failure Detection using Adonis System

14

N05: New Detector Concepts and Instrumentation I Monday, Oct. 26 N05-1: 16:00-18:00 International Ballroom North

R. Jalilian1,2, I. Childres1,2, M. Foxe3, G. Lopez2,4, I. Jovanovic3, Y. P. Chen1,2,4 1 Department of Physics, Purdue University, West lafayette, IN, USA 2 Birck Nanotechnology Center, Purdue University, West Lafayette, IN, USA 3 School of Nuclear Engineering, Purdue University, West Lafayette, IN, USA 4 School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA High-performance sensors for ionizing radiation (particularly -rays and neutrons) are important for many applications, for example, in nuclear security to detect special nuclear materials (SNM). We have explored the exceptional electronic properties of graphene (a single atomic layer of graphite) to develop graphene-based sensors to detect ionizing radiation. Unlike the established detector technologies based on charge collection in semiconductors, the graphene-based detector instead utilizes a sharp dependence of the electrical conductivity of graphene on a local electric field, which can be abruptly changed by charge carriers produced by ionizing radiation in an absorber material. This novel detection method offers a unique opportunity to deploy narrow-bandgap absorbers (such as InSb) without stringent temperature, purity, carrier mobility, or lifetime constraints. We have developed a prototype radiation sensor based on graphene field effect transistor (GFET) made of a graphene on an electrically gated absorber substrate (an undoped semiconductor). Using Monte Carlo simulations (MCNP), the interactions of both g-rays and fast neutrons with various absorber materials (such as Si, Ge, InSb) were modeled to calculate the position of interactions and the energy deposited to the Compton electrons by the -rays in the substrate. Also, transport of energetic Compton electrons in the absorber was subsequently modeled using CASINO. The calculated ionized charge distribution was used as an input into COMSOL Multiphysics to model the change in electric field that would modulate the conductivity of the graphene. We have performed some proof-of-concept experiments to study the transient electrical response of GFET actuated by pulsed laser irradiation. We have also studied the effect of charged particles (including ions and electron beams) irradiation on graphene, important for the long term robustness and reliability of the graphene based radiation detectors.
N05-2:

Graphene Based Sensors for Detecting Ionizing Radiation

T. Keri1, E. N. Cowie1, K. Foehl2, D. Glazier3, G. Hill1, M. Hoek1, R. Kaiser1, M. Murray1, G. Rosner1, B. Seitz1 1 Nuclear Physics Experiment Group, University of Glasgow, Glasgow, Scotland, United Kingdom 2 University of Giessen, Giessen, Germany 3 University of Edinburgh, Edinburgh, United Kingdom Proton-antiproton annihilation is a unique tool to address fundamental questions of the strong interaction and to explore the structure of the nucleon. The PANDA collaboration proposes to build a state-of-the-art universal detector system to study reactions of anti-protons impinging on a proton or nuclear target internal to the high energy storage ring HESR at the planned FAIR facility at GSI, Darmstadt, Germany. Superior particle identification of charged and neutral particles is mandatory to fulfill PANDA's physics aims. Detectors for particle identification comprise energy measurements in tracking detectors, precision Time-Of-Flight detectors, electromagnetic calorimeters, muon chambers and Cherenkov detectors based on the focusing principle or on the time-of-propagation principle. The central detector will feature a barrel DIRC covering the central region and a novel Disc DIRC providing particle identification capabilities in the forward region. Both DIRC systems will benefit from recent advances in detector technology aiming for a 3D DIRC design. In this presentation the technical design, the current status of the development and first results from the prototype test experiment during 28th Aug to 8th Sep for the Focusing Disc DIRC will be presented.
N05-3:

A Focusing Disc DIRC for PID for the PANDA Experiment at FAIR/GSI

Development of a Semi-Digital Hadronic Calorimeter Using GRPCs for Future Linear Collider Experiments

I. Laktineh IPNL-UCBL-IN2P3, Lyon, France On behalf of the CALICE collaboration A new concept of high granularity hadronic calorimeter using thin GRPCs as sensitive medium with embedded semi-digital readout electronics to be used in the future linear collider experiments is under development within the CALICE collaboration. Based on this concept, a small prototype was built and tested with success at CERN PS test beam in 2008. To validate completely this new concept a prototype of 1m3 is being conceived. Few GRPCs as large as 1m2 were built with a new design reducing the dead zones and improving on the gas distribution system. The GRPCs were tested with an electronics board of the same size. The board containing 144 of 64-channel ASICs was conceived and built for this purpose. It represents the largest ever built with the embedded electronics scheme. 15

N05-4:

S. A. Tupputi1, G. De Robertis2, G. Iaselli2,3, F. Loddo2, G. Pugliese2,3, Y. Shinde2 1 Universita' degli Studi di Bari, Bari, Italia 2 INFN Sezione di Bari, Bari, Italia 3 Politecnico di Bari, Bari, Italia

Development of Mini Strips Resistive Plate Chamber Prototype

Resistive Plate Chambers (RPC) provide excellent time resolution ( ns) and they are widely employed as triggering detectors in most of the high energy physics experiments. However their spatial resolution is of the order of cm. A prototype of a single gap RPC instrumented with read-out mini-strip has been implemented. A front-end board has been produced which enables to collect data from the RPC by reading out up to 64 strips connected to 4 chips, each strip with width of 0.8 mm. Chamber efficiency and charge distribution are studied by means of cosmic muons using a small vertical telescope. Different gas mixtures have been also tested. The results of the analysis following out from data collected are presented and the feasibility of a new generation of RPC with improved spatial resolution is discussed.
N05-5:

M. Christophersen1, B. F. Phlips1, H. F. F-W. Sadrozinski2, V. Fadeyev2 1 Code 7651, U.S. Naval Research Laboratory, Washington, DC, USA 2 Santa Cruz Institute for Particle Physics, University of California Santa Cruz, Sanat Cruz, USA Bump bonding is the standard approach for connecting pixel sensors with read-out ASICS. Current bump bonds are mostly solder or Indium based and lead to permanent bonds. This method is difficult to use in cases when either ASIC of sensor performance is variable, as is the case in extensive radiation testing or production QA analysis. We use photo-patternable silicone bumps in combination with a metal coating as flexible and reversible bump bonds. This process only requires a relatively low temperature, < 150 C, and leads to very homogeneous bump height because the silicone (Polydimethlysiloxane, or PDMS) is applied by spincoating. The technique is also applicable to brittle materials, where wire-bonding is problematic, such as gamma-ray detectors based on CdZnTe and CdTe. This paper describes the general concept and presents first results with ATLAS pixels sensors and electronics.
N05-6:

Photo-Patterned Silicone Bump Bonds for Sensors Interconnects

A. Manor1, C. Micou2, A. Osovizky3, F. Schulcz2, E. Marcus3, D. Ginzburg1, V. Pushkarsky1, R. Seif3, Y. Kadmon3, Y. Cohen3 1 Health Physics Instrumentation Department, Rotem Industries Ltd, Beer-Sheva, Israel 2 Health Physics Division, Mirion Technologies, Lamanon, France 3 Electronics & Control Laboratories, Nuclear Research Center - Negev, Beer-Sheva, Israel Radiation detector providing enhanced identification capabilities aimed for the use in homeland security market is presented. Identification ability is required in order to discriminate between Naturally Occurring Radioactive Material (NORM) and Medical isotopes from the Industrial source or Special Nuclear Material (SNM). The detector high resolution together with internal sophisticated algorithm enables decreasing the false alarms rate generated by innocent material as well as miss alarms caused by masking scenario is required. In order to approach these operational requirements, a new Spectroscopic Personal Radiation Detector (SPRD) was developed based on the advanced LaBr(Ce) scintillation crystal. Designated electronics and isotope identification algorithm were implemented. Measurements of sensitivity and isotope identification performance were performed. The device designed configuration as well as its functionality are presented and discussed. Electronic design. The high light output and the short scintillation decay time of the LaBr(Ce) set up a challenge for the accompanying electronic design. Specialized electronics was developed and implemented aiming to achieve low noise and low power consumption. Algorithm. In order to take advantage of the LaBr(Ce) extraordinary properties for high resolution spectroscopy a unique isotope identification algorithm was used, focusing on fast and reliable detection and identification along with masking and shielding scenarios. Special techniques were implemented to assure spectrum stability and analysis reliability over a wide temperature range. Identification over a wide temperature and measuring range required spectrum stabilization and dedicated electronics for pile-up rejection and base line restorer. The detector obtained performances in various operational modes and compare to conventional NaI(Tl) based detector emphasizes its optional contribution for the HLS needs
N05-7:

A Novel Personal Radiation Detection and Identification Device

T. Miyachi1, M. Fujii1, N. Hasebe1, O. Okudaira1, S. Takechi, A. Kurozumi, S. Morinaga, M. Kobayashi3 1 Advanced Research Institute for Science and Engineering, Waseda University, Tokyo, Japan 3 Planetary Exploration Research Center, Chiba Institute of Technology, Chiba, Japan

Study of the Characteristics of a Piezoelectric PZT Radiation Detector Using a Pulsed Xe Source

The detector characteristics of piezoelectric PZT was studied by directly irradiating high energetic Xe ions with a PZT detector. The detector comprised seven piezoelectric plates. The dimensions of each plate were 35 mm x 35 mm x 1 mm. Output signal 16

was observed by digital scopes as a signal form, which was analyzed offline. The radiation source was made by passing an extracted Xe beam at the energy of 400 Mev/n through a rotating slit. Consequently, the number of less than ~ 103 pps ions was available at the energy of 400 MeV/n. The energy injection rate was obtained over 100 s to 1 ms by adjusting the rotation speed. The energy of the incident ions was confined within the detector by placing another 1 mm thick plate in front of the detector. The output amplitude was herein regarded as an integral of the PZT signal form over the duration of the source observed by a scintillation counter. The amplitude increased with the thickness of the PZT element. Such an increasing behavior was understandable by considering that the output is closely related to the energy loss of Xe ions along the beam path. However, there exists a sizable discrepancy between the experimental results and calculated results based on the Bethe-Bloch formula.. The output amplitude per ion depended on the rotation speed. This is qualitatively consistent with phenomenological predictions that the amplitude depends on the rate of energy injection. On the other hand, it is not easy to explain why the output per ion depends on the rotation speed.
N06: Instrumentation for Homeland Security I Monday, Oct. 26 N06-1: 16:00-18:00 International Ballroom Center

S. R. Tornga1, M. C. Galassi1, A. S. Hoover1, M. Mocko1, D. Palmer1, L. J. Schultz1, M. S. Wallace1, B. Harris2, M. V. Hynes2, J. McElroy2, M. Toolin2, D. Wakeford3, R. Lanza4, D. Wehe5 1 Los Alamos National Laboratory, Los Alamos, NM, US 2 Raytheon Integrated Defense Systems, Tewksbury, MA, US 3 Bubble Technology Industries, Ontario, CA 4 Massachusets Institute of Technology, Cambridge, MA, US 5 University of Michigan, Ann Arbor, MI, US The Standoff Radiation Detection System (SORDS) Tri-modal Imager (TMI) is a mobile truck-based, hybrid gamma-ray imaging system able to quickly detect and identify, in position and energy, illicit radiation sources at standoff distances while minimizing the false alarm rate. The TMI consists of 35 NaI crystals each 5 x 5 x 2 arranged in a random mask array, followed by an array of 30 NaI position sensitive bars each 2.5 x 3 x 24. Reconstruction of gamma-ray sources is performed using a combination of two imaging modalities: coded aperture and Compton imaging. A third modality is offered by a shadow algorithm that provides directional information about sources that are not in the field of view of the imager. The active random array acts as both a coded aperture mask and scattering detector for Compton events. Coded aperture and Compton algorithms will be described as well as fusion of the two modalities, and image and energy refinement techniques. Both maximum likelihood and back-projection algorithms have been evaluated and will be compared. Extensive simulations using GEANT4 have been performed and validated against measured data and demonstrate the usefullness of such simulations. Results of image reconstruction and isotope identification algorithms at various speeds and distances will be presented. Signal-to-noise gains on energy spectra will also be shown by utilizing imaging information. The TMI is being developed for the SORDS program conducted by the Department of Homeland Securitys (DHS) Domestic Nuclear Detection Office (DNDO).
N06-2:

The SORDS Tri-Modal Imager: Imaging and Source Identification at Standoff Distances

L. J. Mitchell1, B. F. Phlips1, E. A. Wulf1, W. N. Johnson1, A. L. Hutcheson1, C. J. Lister2, K. D. Bynum1, B. E. Leas1, G. Guadagno1 1 Radiation Detection Section Code:7654, Naval Research Laboratory, Washington, DC, USA 2 Physics Division, Argonne National Laboratory, Argonne, IL, USA The Mobile Imaging and Spectroscopic Threat Identification (MISTI) system being developed to locate radiological threats in urban and rural environments is currently in the integration and test phase of development. MISTI utilizes the cost effective collection power of NaI for imaging and the sensitivity of high resolution HPGe for spectroscopy. MISTIs data acquisition system was developed with the latest commercially availed hardware that met MISTIs count rate and stability requirements. The performance of crucial software and hardware components will be discussed along with overall system performance. The results collected during the test phase (to be completed in the summer of 2009) will be compared to prior simulations and presented at the conference.

Mobil Imaging and Spectroscopic Treat Identification (MISTI): Field Trials

17

N06-3:

Performance of the Roadside Tracker Portalless Portal Monitor

K. P. Ziock, E. C. Bradley, A. Cheriyadat, M. F. Cunningham, L. Fabris, J. S. Goddard, D. E. Hornback, T. P. Karnowski, R. A. Kerekes Oak Ridge National Laboratory, Oak Ridge, TN, USA We have developed a proof-of-concept prototype, rapid-deployment, gamma-ray portal monitor that can uniquely link the radiation signatures and visible-light images of vehicles in the system field of view from the side of a multi-lane roadway. The instrument uses both visible-light and gamma-ray imaging to accomplish this. Vehicles entering the field of view of the visiblelight imaging system are identified and tracked by an automated target acquisition and tracking software engine. The vehicle locations provided by this code are reported to the gamma-ray imager which uses them to locate the vehicles in the gamma-ray images and to collect the vehicle-specific radiation signa-tures from those images. Using this technique the gamma-ray data is integrated over the entire residence time of the vehicle in the gamma-ray imager field of view. The complete instrument comprises ~ 1-m2 CsI(Na) detector area split between two units. The system has been designed to handle up to 5 lanes of traffic at highway speeds, with one unit located on either side of the roadway. Because no equipment is required in the roadway, the instrument can be deployed without impacting the flow of traffic. We will present details of the system design, including the visible-light and gamma-ray imaging systems as well as the results of an extensive test and evaluation campaign.
N06-4:

Boron Coated Straw Detectors as a Replacement for 3He

J. L. Lacy, A. Athanasiades, L. Sun, C. S. Martin, G. J. Vazquez-Flores Proportional Technologies, Inc, Houston, TX, U. S. A. The limited inventory and minute natural abundance of 3He gas on Earth necessitate the adoption of new technologies for the detection of neutrons. The exclusive source of 3He on Earth is the radioactive decay of tritium, which decays to 3He at a rate of 5.5% per year. The estimated current US tritium stock is 36 kg, giving an annual 3He production of just 1.98 kg. Despite the low 3He supply, and uncertain production rate in the future, this medium appears to be the only solution considered by government agencies, in a number of applications. The DHS and DOD plan to equip major ports of entry with large area monitors, in an effort to intercept the smuggling of nuclear materials. A modest deployment of 1500 such monitors will require about 10 kg of 3He, the whole US production of the isotope over a period of 5 years. Clearly, alternate neutron detection technologies must be considered. We propose a technology based on close-packed arrays of long aluminum or copper tubes (straws), 4 mm in diameter, coated on the inside with a thin layer of 10B-enriched boron carbide (10B4C). A close-packed array of straw detectors offers a stopping power for neutrons equivalent to that of 2.68 atm of 3He gas. In addition to the high abundance of boron on Earth and the very low cost of 10B, the boron-coated straw (BCS) detector offers distinct advantages over conventional 3Hebased detectors, including sensitivity to both neutrons and gammas rays (gamma rays are converted in the straw wall), low weight, safety for portable use (no pressurization), and low cost. Furthermore, in imaging applications, the BCS high level of segmentation supports high count rates and parallax-free position encoding, both difficult to achieve in conventional 3He pressure vessels. We review the use of the BCS detector in a variety of applications, pointing out its distinct advantages over conventional 3He tubes and present a novel new configuration of straws dispersed in a solid plastic moderator.
N06-5:

Accurate Modeling of the Terrestrial Gamma-Ray Background for Homeland Security Applications

G. A. Sandness, J. E. Schweppe, W. K. Hensley National Security Directorate, Pacific Northwest National Laboratory, Richland, WA, USA Pacific Northwest National Laboratory has developed computer models to simulate the use of radiation portal monitors to screen vehicles and cargo for the presence of illicit radioactive material. The gamma radiation emitted by the vehicles or cargo containers must often be measured in the presence of a relatively large gamma-ray background mainly due to the presence of potassium, uranium, and thorium (and progeny isotopes) in the soil and surrounding building materials. This large background is often a significant limit to the detection sensitivity for items of interest and must be modeled accurately for analyzing homeland security situations. Calculations of the expected gamma-ray emission from a disk of soil and asphalt were made using the Monte Carlo code MCNP and were compared to measurements made at a seaport with a high-purity germanium detector. Analysis revealed that the energy spectrum of the measured background could not be reproduced unless the model included gamma rays coming from the ground out to distances of at least 300 m. The contribution from beyond about 50 m was primarily due to gamma rays that scattered in the air before entering the detectors rather than passing directly from the ground to the detectors. These skyshine gamma rays contribute tens of percent to the total gamma-ray spectrum, primarily at energies below a few hundred keV. The techniques that were developed to efficiently calculate the contributions from a large soil disk and a large air volume in a Monte Carlo simulation are described and the implications of skyshine in portal monitoring applications are discussed. Issues include the effect of skyshine on the suppression of the gamma-ray background by the vehicles and cargo containers during scanning, the effect of the background suppression on the shape of measured energy spectra, and the dependence of these effects on the orientation of the detectors, particularly upward-looking versus side-looking detectors.

18

Prepared for the U. S. Department of Homeland Security, U.S. Customs and Border Protection and Domestic Nuclear Detection Office under U.S. Department of Energy Contract DE-AC05-76RL01830. N06-6:

Shielding of Skyshine Interference with Radiation Detection Systems

R. T. Kouzes, K. E. Conlin, W. J. Kernan, E. K. Mace, E. R. Siciliano, M. L. Woodring PNNL, Richland, WA, USA Skyshine, the radiation scattered from the air above a source, such as radiography or x-ray imaging sources, can produce interference with radiation detection systems even at very large distances. Skyshine has been studied for many years, but almost entirely within the context of nuclear power and radioactive waste storage. These earlier studies show that modeling of skyshine must take into account huge volumes of space in the models. Skyshine has a significant impact on border security applications of radiation detection equipment. This paper focuses on shielding approaches to the skyshine interference problem with radiation detection systems used for border screening.
N06-7:

S. E. Labov1, G. Clark1, T. Edmunds1, M. Gokhale1, L. Hiller1, G. Johannesson1, K. Nelson1, D. Slone1, R. Wheeler1, Y. Yao1, K. M. Chandy2, A. Liu2, R. McLean2, M. Wu2, J. G. Jernigan3, A. Ganem4 1 Lawrence Livermore National Laboratory, Livermore, CA, USA 2 California Institute of Technology, Pasadena, CA, USA 3 University of California, Berkeley, CA, USA 4 Zontrak Inc., San Ramon, CA, USA Radiation detection systems are currently deployed at ports of entry to scan incoming vehicles and cargo for radiological or nuclear threats. These systems can be effective for traffic coming through the port, but a different architecture is required to protect against threats that do not pass through these ports, are missed by the port systems, or are created within the country. We are developing a Distributed Nuclear Detection Array (DNDA), a broad-area radiation detection system designed to evaluate network architectures and data fusion algorithms for distributed radiation detectors. Our system is designed to work with a variety of network architectures, different distributed detector array configurations and a range of data fusion algorithms. Our measurements show many features that can have a large impact on distributed detector data fusion algorithms including nonstatistical background variations and non-uniform detector coverage and motion. The spatial, spectral and temporal variations in the radiation measurements are extremely complex and strongly dependent on the type of region, detector deployment and operation. These are the characteristics that make radiation detection with high sensitivity so challenging, and any distributed radiation detector analysis system must be designed to take this into account. We have therefore developed a technique that can simulate any number of detectors operating under conditions that include the variations found in the real measurements. We present an overview of the DNDA system along with the scenarios, network architectures and algorithms we are developing and evaluating. Results include performance comparisons with and without data fusion for a network of detectors operating serendipitously in vehicles and carried by personnel whos primary duty brings them to areas of interest but who are mostly engaged in non-radiation related activities.
This work was supported by the Domestic Nuclear Detection Office of the Department of Homeland Security and performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. N07: Data Acquisition and Analysis Systems I Monday, Oct. 26 N07-1: 16:00-18:00 Grand Ballroom 7

Data Fusion with Distributed Nuclear Detection Arrays

The Read-Out and Control System for the Dark Energy Survey

K. Honscheid Physics, Ohio State University, Columbus, Ohio, USA On behalf of the DES Collaboration We describe the data acquisition and control system of the Dark Energy Camera (DECam), which will be the primary instrument used in the Dark Energy Survey (DES). DECam is a 3 sq. deg. mosaic CCD camera mounted at the prime focus of the Blanco 4m telescope at the Cerro-Tololo International Observatory (CTIO). The DECam data acquisition system (SISPI) is implemented as a distributed multi-processor system. The core of the system software is written in Python with a few extensions coded in LabVIEW and C. We use PYRO, a powerful distributed object technology available free under the MIT license, to implement the Client-Server and Publish-Subscribe design patterns. In order to support remote operations and to allow our graphical user interfaces to run in standard web browsers we extended these communication protocols to web applications. In this paper we

19

discuss the SISPI architecture, new concepts used in the design of the infrastructure software and provide an overview of the remaining components of the DES read-out and control system.
N07-2:

A Time and Frequency Distribution System for Reactor Neutrino Experiment

G. Gong, X. Tao, Z. Feng, S. Chen Dept. of Engineering Physics, Tsinghua Univ. Beijing, Beijing, China The reactor neutrino experiment provides a clean way to measure the neutrino mixing angle 13 with high precision, this requires careful design for all aspects. This paper will present a time and frequency distribution system, which is used to broadcast a global highly precise frequency and time reference to the distributed underground experiment sites with baselines up to 2Km. A rubidium atomic oscillator provides a 10MHz frequency with an accuracy of 10-12 and stability of 10-11. A dedicated device converts the 10MHz reference frequency signal which is transmitted with optical fibers to the destinations, the signal is then reconstructed few Kilo-meters away with amplifiers. The reconstructed frequency is then processed with a precision clock conditioning circuit to programable frequency and fan-out to all electronics modules. The time reference is generated by a commercial UTC (Universal Coordinated Time) device and transmitted to different experiment sites via multimode optical fibers. The signal is then fan-out to all electronics modules and decoded there. Another optical fiber feeds back the time reference signal, the delays of cable length and distortion of temperature can be measured and monitored at the transmit side with a FPGA based TDC. Details of the design and performance will be described in this paper.
N07-3:

A. Kimura1, K. Furutaka1, S. Goko1, H. Harada1, M. Igashira2, T. Kamiyama3, T. Katabuchi2, T. Kin1, K. Kino3, F. Kitatani1, Y. Kiyanagi3, M. Koizumi1, M. Mizumoto2, S. Nakamura1, M. Ohta1, M. Oshima1, Y. Toh1 1 Research group for Innovative nuclear science, Japan Atomic Energy Agency, To-kai-mura, Ibaraki, Japan 2 Tokyo Institute of Technology, Tokyo, Japan 3 Hokkaido University, Sapporo, Hokkaido, Japan To measure neutron cross-section data of the minor actinides, a Ge detector-array, which consists of 30 Ge crystals and BGO anti-Compton shields, was installed at pulsed neutron beam line for the Time Of Flight (TOF) experiments. The authors developed a new DAQ and reported in NSS 2008. In TOF experiments, dead time is dependent on many factors, such as event rate, multiplicity of prompt gamma-rays, and data transfer rate. It is very difficult to obtain the dead time for each detector. To correct dead-time effect and measure high-precision neutron cross-section data, we have tested a dead-time correction method. In this method, random timing pulses are input to every pre-amplifiers of the Ge crystals via test-signal inputs. Both the random timing pulses and the other pulses originated from prompt gamma-rays are measured with our DAQ. At the same time, the inputs random pulses are counted with another system consisting of a fast timing amplifier, a CFD, and a counter. Because dead-time affects similarly both the random timing pulses and the measured prompt gamma-rays, we can calculate dead-time by comparing the number of the input random pulses measured using the counter with the area of the peak due to the pulses from the pulse generator measured with our DAQ.
This study was the result of "Study on nuclear data by using a high intensity pulsed neutron source for advanced nuclear system" entrusted to National University Corporation Hokkaido University by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT). N07-4:

A Dead-Time Correction Method for Multiple Gamma-Ray Detection

L. Toscano1, D. Falchieri2, F. Prino3, A. Rivetti3, S. Zannoli2 1 CERN, Geneve, Switzerland 2 INFN Sezione di Bologna and University of Bologna, Bologna, Italy 3 INFN Sezione di Torino, Torino, Italy

A Digital Filter with Common Mode Noise Rejection for ALICE Silicon Drift Detector

The Silicon Drift Detector (SDD) system of the ALICE experiment at CERN employs 10-bit A/D conversion on the front-end chip to measure the charge released in the sensors by an impinging particle. A 2-D data compression algorithm is then applied to the resulting data stream to reduce the data volume by a factor of 17. At high trigger rates (above 100 Hz) a few of the 260 modules have shown a common mode noise component that worsens the signal to noise figure typically by a factor of two. In order to get the best performance of the detector in term of both resolution and data reduction, common mode noise subtraction can be performed on-line before the data are compressed. For this purpose, a dedicated algorithm has been developed and implemented in a Xilinx Virtex-5 FXT evaluation board. The amount of common mode noise is measured event-by-event on a group of adjacent channels and then removed after baseline equalization. To implement an on-line algorithm a Field Programmable Gate Array (FPGA) with embedded Digital Signal Processing (DSP) was used. The functionality of this algorithm is confirmed by applying it to the raw data taken with ALICE SDD system.

20

N07-5:

X. Sun1, C. Tintori2, K. A. Lan1, M. J. Kibilko3, Y. Shao1 1 Imaging Physics, University of Texas MD Anderson Cancer Center, Houston, TX, United States 2 CAEN S.p.A., Viareggio, LU, Italy 3 SE Technical Sales, Inc, Windermere, FL, United States

Evaluation of a FADC with DPP for Applications of Gamma-Ray Detection in Nuclear Medicine Imaging

A FADC (flash ADC) with DPP (digital pulse processing) from CAEN Technologies (model V1724) has been evaluated for its applications in gamma ray imaging detector development. The detector consists of an 8x8 matrix of 1x1x3 cubic mm CsI crystals coupled to a Position-Sensitive Avalanche Photodiode. Since PS-APDs inherent high background noise and signal variations in both amplitude and time among different cathode channels due to its internal charge division, long integration signal shaping and peaking sensing are usually required after the charge-sensitive pre-amplifier, which demands substantial electronics and data acquisition. With the FADC and DPP, both signal shaping and amplitude measurement can be done for all signal channels with a single standalone module and digital signal processing to significantly simplify the electronics, and potentially reduce the cost and enhance the performance as well. Overall, the initial test has shown that the signal from each channel can be properly triggered and conditioned by FADC and be further processed with shaping and peak measurement through the DPP. A LabView based graphical user interface provides convenient parameter input and operation selection. Flood source crystal images show that the new approach can reach the same image quality as the analog one, although a common trigger mode will be really needed for gamma-ray imaging.
N07-6:

Bayesian Muon Tomography Using Fused Priors

G. Wang, J. Qi University of California, Davis, Davis, CA, USA Muon tomography is capable of detecting high-Z materials in vehicles or cargo containers. Bayesian image reconstruction method has been proposed to improve the detection performance of muon tomography by introducing a prior distribution on the scattering density images. In this paper we investigate using a fused prior which comprises of a spatial independent term and a spatially correlated term in the maximum a posteriori (MAP) image reconstruction. An mininorization-maximization algorithm is derived to find the MAP estimate iteratively. Receiver operating characteristic studies using Monte Carlo simulation show that MAP reconstruction using the fused prior can achieve higher detection performance than using independent or spatially correlated prior alone.
N07-7:

X. Wang1,2, Y. Chen3, Q. Yin3, S. Zeng1,2, Q. Xie1,2 1 Department of Biomedical Engineering, Huazhong University of Science and Technology, Wuhan,Hubei, China 2 Wuhan National Laboratory for Optoelectronics, Wuhan,Hubei, China 3 Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan,Hubei, China Higher count rate detectors are needed for high sensitivity positron emission tomography (PET) scanner. However, in past several decades, the maximum count rate of the detector is limited by the large pileup fraction in high count rate applications. Due to its analog electronics, most PET system reject the pileups without attempting to separate them. In this work, we investigate the use of digital signal processing (DSP) technologies for separating pileup events. We propose a DSP method to recover timing and energy information of the single events from a multi-events pileup event. For each recorded pulse, this DSP method discriminates multi-events pileup and single-event pulse by the number of leading edges, after that, each leading edge and the followed tail are associated to reconstruct new pulses in the pileup processing. With this method, a pileup pulse can be separated into two or more single event pulses. Consequently, the timing and energy of single events can be measured independently. Our preliminary results show that 154 event pulses can be extracted out of 120 record pulse waveforms, which were digitized by oscilloscope. These results show the potentials of digitally sampling scintillations techniques in pileup processing in PET. In addition, this DSP method can be implemented by a free running sampling ADC and a FPGA, which are feasible and commercially available. keywords:Positron Emission Tomography (PET), Digital Signal Processing (DSP), Pileup, Scintillation Pulse

Advantages of Digitally Sampling Scintillation Pulses in Pileup Processing in PET

21

N08: Radiation Damage Effects I: Semiconductor Devices Monday, Oct. 26 N08-1: 16:00-18:00 Grand Ballroom 8

Charge Collection Studies of Heavily Irradiated 3D Double-Sided Sensors

R. L. Bates Physics and Astronomy, The University of Glasgow, Glasgow, UK On behalf of the RD50 Collaboration Three-dimensional (3D) photodiode detectors offer advantages over standard planar photodiodes as more radiation hard radiation sensors. 3D detectors with the double-sided geometry have been fabricated as very short strip detectors with the same intercolumn spacing as proposed for the ATLAS pixel detector upgrade. The detectors have been irradiated to a fluence of 2 x 1016 cm2 1 MeV equivalent neutrons, which is twice the expected dose of the inner pixel layer of the ATLAS detector for superLHC operation. Charge collection studies have been performed with analogue readout with 25ns shaping time, as required for (super)LHC experiments. The response of the detectors to Sr-90 electrons is shown and compared with planar devices. The 3D detector is shown to have superior charge collection characteristics even at the highest fluences even when compared to planar devices operating at 1000V, which is in excess of that presently possible in the ATLAS experiment. The experimental results are compared to the simulation of charge transport in the devices.
RD50 collaboration for part funding this work N08-2:

Measurement of the Interstrip Resistance and Capacitance of Irradiated P-Type Silicon Strip Detectors

S. Lindgren, C. Betancourt, N. Dawson, G. Bredeson, M. Gerling, J. Wright, H. F. -W. Sadrozinski SCIPP, UC Santa Cruz, Santa Cruz, CA, USA We report on the measurement of interstrip parameters of p-type silicon strip sensors which we are developing in a large collaboration to be used in a future tracker for the LHC upgrade. We measure on test structures with about 1 cm long strips the interstrip resistance and the interstrip capacitance (at 1 MHz) pre-rad and after irradiation with 70 MeV protons to a fluence of 1.5*10^13 p/cm^2, corresponding to about 1 MRad, from prototyping runs with Hamamatsu Photonics and Micron Semiconductors. We report the values for a variety of isolation scenarios of p-stops, p-spray and a combination of both.
N08-3:

Radiation Damage Studies for the D0 Silicon Microstrip Tracker at the Tevatron

Z. Ye Fermi National Accelerator Laboratory, Batavia, Illinois, USA On behalf of the D0 Collaboration A Silicon Microstrip Tracker (SMT) has been operating at the D0 Run II experiment at the Fermilab Tevatron collider since 2001. It is one of the largest silicon detectors in the world and its information is heavily used for many D0 analyses. The SMT consists of four layers of double-sided and single-sided sensors organized in six barrels, and two sets of disks in the forward and backward regions. A new layer of single-sided sensors organized in eight barrels was installed at a very small radius in 2006. The sensor leakage current, full depletion voltage as well as signal to noise ratio are continuously monitored for radiation damage. During this monitoring process a bulk carrier-type reversal was observed in the innermost layer of the original detector. Predictions are also made for the detector performance towards the possible end of the Tevatron Run II. These predictions are based on the locations and types of the sensors, our knowledge and measurements of the phenomenology of radiation damage as well as the actual dose received in the Tevatron environment. We present the results of these aging studies and predictions in this talk.
N08-4:

C. Betancourt1, B. Colby1, G. Bredeson1, N. Dawson1, V. Fadeyev1, M. Gerling1, R. F. Hurley1, S. Lindgren1, P. Maddock1, H. F.-W. Sadrozisnki1, J. von Wilpert1, A. Affolder2, P. Allport2, G. Casse2, G. Kramberger3, I. Mandic3, M. Mikuz3, V. Cindro3, H. Brown2, I. Mcleod2 1 SCIPP, University of California Santa Cruz, Santa Cruz, Ca, USA 2 Department of Physics, University of Liverpool, Liverpool, United Kingdom 3 Jozef Stefan Institute, University of Ljubljana, Ljubljana, Slovena As part of the RD50 program to develop radiation-hard semiconductor detectors for the future high luminosity hadron collider upgrade, n-type and p-type magnetic Czochralski and Float Zone planar silicon devices were produced at Micron Semiconductor 22

Charge collection studies and annealing effects in heavily irradiated planar silicon strip sensors

Ltd. Planar silicon diode and microstrip devices were irradiated with protons, pions and neutrons up to fluences of 2.2*10^16 neq/cm2 in order to study the radiation hardness of these devices. These fluences are predicted to exceed the dose of the innermost layers of the LHC Upgrade detectors. In this study, the evolution of the depletion voltage, leakage current, and charge collection was studied as a function of fluence and annealing time. Charge collection measurements using Sr-90 beta sources were performed with both binary and analogue electronics. The depletion voltage determined by C-V and charge collection from a beta source is compared. These measurements demonstrate the viability of planar silicon sensors for both strip and pixel systems for the LHC upgrade.
N08-5:

M. J. Tadjer1, K. D. Hobart2, R. E. Stahlbush2, P. J. McMarr3, H. L. Hughes2, F. J. Kub2, S. K. Haney4 1 Department of Electrical and Computer Engineering, University of Maryland, College Park, College Park, MD, USA 2 Power Electronics Branch, Naval Research Laboratory, Washington, DC, USA 3 GSG, Inc., Crofton, MD, USA 4 Cree, Inc., Durham, NC, USA

Thermally Stimulated Current Separation of Hole and Acceptor Trap Density in 4H-SiC Epitaxial MOS Devices Using Co-60 Irradiation

The presence of traps near the SiO2 interface has been a detrimental factor for the mobility in Si and SiC devices [1-4]. By utilizing thermally stimulated current (TSC) measurements on Co-60 irradiated 4H-SiC MOS capacitors, we have separated the density of two traps previously shown to have the same activation energy, but different origin: hole traps and acceptor traps. The 5 m thick epitaxial p-type (Al-doped, NA=1x1016 cm-3, 50 nm NO-annealed thermal SiO2) 4H-SiC MOSCAP samples were provided by Cree, Inc. At temperatures lower than 160 K, the device entered the freeze-out regime. A total dose of 100 krad of Co-60 was delivered at a rate of 277 rad/s. The capacitors were biased at either 2 or 2 MV/cm during irradiation. The TSC method consisted of cooling the sample down to 20 K biased in accumulation, and then warming up to 250 K biased in depletion. TSC emission during warmup revealed peaks near 55 K and 80 K. The 55 K peak originated from traps spatially located between the gate and the SiO2/SiC interface [5]. The 80 K peak is thought to be a combination of emission from hole traps and Al acceptors in the SiC epitaxial layer. After Co-60 irradiation with an applied gate field of 2 MV/cm, the TSC spectra showed that the amount of filled hole traps during cooldown did not vary in irradiated devices, as opposed to non-irradiated ones. Therefore, TSC emission from an irradiated device occurred only from acceptor-related traps, whose occupation density did not depend on the accumulation bias. When the sample was irradiated with 2 MV/cm gate field, the 80 K peak for the capacitor split by about 5 K, resulting in two distinct peaks at 77 K and 82 K. This indicated a cooldown accumulation bias dependent shift in hole trap activation energy, and a bias independent emission from Al acceptors.
References: [1] D. M. Fleetwood, R. A. Reber, Jr., and P. S. Winokur, IEEE Trans. Nucl. Sci., vol. 38, no. 6, pp. 1066, 1991. [2] J. G. Simmons and G. W. Taylor, Solid-State Electronics, vol. 17, pp. 125 and 131, 1974. [3] H. Olafsson et al., Appl. Phys. Lett. 79 (24) 4034, 2001. [4] A. Agarwal, H. Fatima, S. Haney, S.-H. Ryu, IEEE Electr. Dev. Lett. vol. 28, no. 7, pp. 587, 2007. [5] M. J. Tadjer, R. E. Stahlbush, K. D. Hobart, E. A. Imhoff, F. J. Kub, S. K. Haney, Jour. Appl. Phys., in preparation. N08-6:

Polarization Effects in Radiation Damaged scCVD Diamond Detectors

S. Schuwalow DESY, Zeuthen, Germany On behalf of the FCAL Collaboration The current status of the research work of the International Linear Collider (ILC) Forward Calorimetry Collaboration (FCAL) is described. One of the planned Very Forward Detectors, the Beam Calorimeter, should be equipped with the sensors capable to withstand electromagnetic radiation doses up to several MGy annually. Charge collection properties of several candidate materials including GaAs, Diamond and Sapphire were investigated in details as a function of the absorbed dose. It was found that strong polarization occurs in the radiation damaged Diamond sensor due to the charge trapping. This effect significantly reduces charge collection efficiency of the detector. The corresponding model is developed to describe the observed phenomena. A method of routinely changing of the detector bias voltage polarity is introduced and experimentally proved in order to suppress the polarization.

23

J01: Instrumentation for Medical and Biological Research I: Radionuclide Imaging Tuesday, Oct. 27 J01-1: 08:00-10:00 International Ballroom North

D. Meier1, D. J. Wagenaar1, G. E. Maehlum1, B. M. Sundal1, B. E. Patt1, M. J. Hamamura2, W. W. Roeck2, S.-H. Ha2, O. Nalcioglu2, S. Chen3, J. Xu3, B. M. W. Tsui3 1 Gamma Medica - Ideas, Oslo/Sherbrooke/Northridge, Norway/Canada/USA 2 John Tu & Thomas Yuen Center for Functional Onco-Imaging, University of California, Irvine, California, United States 3 The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, Maryland, United States This article describes a stationary single photon emission computed tomograph (SPECT) camera which can be operated inside a state-of-the-art magnetic resonance imaging (MRI) system with a minimum bore diameter of 12 cm. The combined SPECT/MRI system allows one to acquire simultaneously the data from both modalities and co-register the images in space and time. The new SPECT camera is based on the solid-state semiconductor cadmium zinc telluride (CZT) and application specific integrated circuits (ASICs) two technologies which are MR-compatible and almost insensitive to magnetic fields. The results from our experiments show, firstly, that the MRI performance is minimally affected by the SPECT camera inside the bore and, secondly, that the SPECT camera works well inside the MRI system with negligible effect from the magnetic field. The SPECT camera has an energy resolution of 5.4 keV fullwidth-half-maximum at the 140-keV photo peak from 99m-Tc for both inside and outside of the MRI. We acquired data that were reconstructed to form SPECT images from capillary tubes. The line spread function was measured at 2.5 mm FWHM and two capillary tubes 4-mm apart can be easily resolved in the reconstructed image.
J01-2:

A Stationary SPECT Camera for Simultaneous SPECT/MRI

S. Marcatili1,2, M. G. Bisogni1,2, G. Llosa1, G. Magazzu2, F. Morsani2, A. Del Guerra1,2 1 Department of Physics, University of Pisa, Pisa, Italy 2 Sezione di Pisa, INFN, Pisa, Italy

Calibration and Performances of a Multichannel DAQ System for Silicon Photomultiplier (SiPM) Matrices in PET Applications.

The silicon photomultiplier (SiPM) [1] is nowadays a promising technology in various fields of physics involving radiation detection. In the last years, devices were mainly available in small area samples ranging from 1 mm2 active area to a 16 mm2 active area in a single pixel. However, imaging applications require detector with a big active area as well as the capacity to give information about the radiation hit position. At FBK-irst, 16-pixels (1 mm2 size) matrices of SiPM on a common substrate and negligible dead area, have been developed. However, the use of SiPM matrices makes it necessary to develop a multichannel read-out system capable to manage the SiPM matrices signals, especially in view of the high granularity as needed in a PET tomograph. At the University of Pisa we have developed an FPGA based DAQ system (SiPM0) for the read-out of two 4x4 pixels matrices in time coincidence for Positron Emission Tomography applications. Our goal is the construction of a small PET prototype composed by two detector heads based on LSO continuos crystals read by SIPM matrices [2]. We have already presented the DAQ architecture [3] and the first electric test on the board. Now, all the logic functions necessary for our application have been implemented, and the board has been fully characterized and calibrated. Here we present the characterization of the DAQ, and the measurements with the two detector heads in time coincidence.
[1] G. Llosa et al. Novel Silicon Photomultipliers for PET Applications. IEEE Trans. Nucl.Sci. vol. 55, no.3, pp877-881, June2008. [2] MoehrsS. etal. A detector Head design for small animal PET with silicon photomultipliers(SiPM). Phys. Med. Biol.vol. 51, pp. 1113-1127, 2006. [3] M. G. Bisogni etal. An FPGA based DAQ System for the Readout of SiPM Arrays in PET Applications.NSS MIC 2008, Nuclear Science Symposium Conference Record, N52-1. J01-3:

A. G. Weisenberger1, B. Kross1, J. E. Mckisson1, A. Stolin1, C. Zorn1, C. R. Howell2, A. S. Crowell2, C. D. Reid3, S. Majewski4, M. F. Smith5 1 Radiation Detector and Imaging Group, Thomas Jefferson National Accelerator Facility, Newport News, VA, USA 2 Department of Physics, Duke University, Durham, NC, USA 3 Department of Biology, Duke University, Durham, NC, USA 4 Department of Radiology, West Virginia University, Morgantown, WV, USA 5 Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland, Baltimore, MD, USA There are opportunities for the development of new tools to advance plant biology research through the use of radionuclides. Thomas Jefferson National Accelerator Facility, Duke University, West Virginia University and University of Maryland are 24

Positron Emission Tomography Detector Development for Plant Biology

collaborating on the development of radionuclide imaging technologies to facilitate plant biology research. Biological research into optimizing plant productivity under various environmental constraints, biofuel research and carbon sequestration research are areas that could potentially benefit from new imaging technologies. Using 11CO2 tracers, the investigators at Triangle University Nuclear Laboratory / Duke University Phytotron are currently researching the dynamical responses of plants to environmental changes forecasted from increasing greenhouse trace gases involved in global change. The biological research primary focus is to investigate the impact of elevated atmospheric CO2 and nutrients limitation on carbon and nitrogen dynamics in plants. We report here on preliminary results of 11CO2 plant imaging experiments using Jefferson Lab dual planar positron emission tomography detectors to image for 11CO2 tomography in live plants. New detector designs will be developed based on the preliminary studies reported here.
J01-4:

S. Seifert1, R. Vinke2, H. T. van Dam1, H. Loehner2, P. G. Dendooven2, F. J. Beekman1,3, D. R. Schaart1 1 Radiation, Radionuclides & Medical Imaging, Delft University of Technology, Deft, The Netherlands 2 Nuclear Physics Accelerator Institute (KVI ), University of Groningen, Groningen, The Netherlands 3 University Medical Centre Utrecht, Utrecht, The Netherlands Silicon Photomultipliers (SiPMs), are a relatively new class of photo sensors, with very favorable properties such as high gain, high quantum efficiency, fast response, and low excess noise. These make them ideal candidates for replacing conventional photomultiplier tubes (PMTs) in medical imaging applications such as single photon emission computed tomography (SPECT) or positron emission tomography (PET). In addition these devices are insensitive to magnetic fields, enabling integration of e.g. PET and MRI. Also, the combination of SiPMs and fast and bright scintillators, such as LaBr3:Ce, seems very promising for application in time-of-flight (TOF) PET. We therefore conducted a series of experiments with the goal of obtaining the best possible timing resolution with SiPM based scintillation detectors, focusing on the improvement of the measurement strategy and the instrumentation for precise timing. The detectors employed in our measurements consisted of two SiPMs (Hamamatsu MPPC-S10362-33-050C; 3x3 mm2 active area; 3600 cells), which were directly coupled to small LaBr3:Ce(5%) crystals (SaintGobain BrilLanCe 380; 3x3x5 mm3). The amplified signals from the SiPMs were recorded with synchronized, fast digitizers (Acqiris DC282) at 8 GS/s. The digitization of the signals allowed for offline determination of the times stamps for each detector. The setup was optimized in terms of electronic noise, detector bias voltage, and time pick off method. An excellent coincidence resolving time for 22Na 511 annihilation photons of 101.8 ps 0.6 ps FWHM was obtained at the optimized settings. These results compare well to the optimum intrinsic timing resolution predicted by a statistical model describing the single photo-electron timing jitter.
J01-5:

Ultra Precise Timing with SiPM-Based TOF PET Scintillation Detectors

K. A. Lan1, X. Sun1, Y. Shao1, P. A. Dokhale2, R. Farrell2, K. S. Shah2 1 Imaging Physics, UT MD anderson cancer center, Houston, TX, USA 2 Radiation Monitor Devices, Inc., Watertown, MA, USA

Design Studies of a Front-End Readout for Position Sensitive Avalanche Photo Diode

PSAPD has been recently used in medical imaging applications. It is still challenging to couple a CsI(Tl) array to PSAPD for detecting low energy gamma ray at room temperate due to relatively low signal-noise-ratio (SNR). Previous studies by other groups have shown suitable performance only at -20 to -60 C. Preliminary tests show the JFET operational amplifier (OA) is suitable for building PSAPD front end circuit. This presentation is focused on signal to noise analysis to provide design guidelines for optimizing the detector performance at room temperature. The crystal array consists of 8x8 array of 1x1x3 mm CsI(Tl) crystals. PSAPD has 8x8 mm sensitive area. Each of 4 position-sensitive outputs was connected to a preamplifier based on latest JFET OA. In the design, different OAs configurations at different working conditions were tested and optimized based on the measurement of SNR and flood crystal image. Our simulation and circuit analysis have shown that it is feasible to use a low-noise OA to develop a preamplifier as the front-end readout for a gamma camera to achieve high spatial image resolution. The 16 channel readout prototype circuit board was built and tested. Four trigger channel from anode signal also on the board. CAEN spectrum amplifiers and digitizer were used for signal processing. In the flood source image, all crystals at the central region of the array were clearly identified with ~3:1 peak-to-valley ratio, but overlapped at the edge due to reduced charge sharing sensitivity. This phenomenon had been reported previously. Further analysis will be reported. Main concerns for using OA are the heat generated the circuits that may fluctuate PSAPD gain. To solve this problem, a flex cable with 30 cm length and 1 mm pitch was connected between PSAPD and preamplifiers. Preliminary studies have shown that the same imaging quality can still be maintained with the use of flax cable.

25

J01-6:

Radiotracer Medical Imaging Technologies Applied to Environmental Remediation Systems

R. Boutchko, B. W. Reutter, T. F. Budinger, W. W. Moses, G. T. Gullberg Radiotracer Development & Imaging Technology, Lawrence Berkeley National Lab, Berkeley, CA We demonstrate an alternate application of high-sensitivity radiotracer imaging technologies developed for nuclear medical imaging. There are several available radioisotopes, namely 51Cr and 99mTc, that are particularly appealing for studying environmental remediation techniques, notably for modeling toxic element runoff through the soil near former nuclear facilities such as the Hanford reactors in Washington state (USA). 52Cr (stable) and 99Tc (2.3x106 year half-life) are the two most important reactor-generated contaminants that have entered the groundwater. There are numerous studies being conducted to model transport of these compounds in different types of soil such as sand, clay, dirt and gravel. The speed of transport through medium depends on chemical composition of the particles, chemical composition of water, physical properties of the medium (particle size, homogeneity, temperature) and on content of different types of microorganisms. Measurements are typically performed by flowing water containing 52Cr or 99Tc through sediment columns (tubes a few cm in diameter and ~1 meter long) and monitoring the in- and out- flows of contaminants. We will support the effort to develop different methods of reducing the contaminant transport speed by using SPECT imaging to study transport of the single-photon emitting surrogates 51Cr and 99mTc. By quantitatively imaging the contaminant distribution in the sediment column as a function of time, a much richer set of data can be obtained. We will compare contaminant transport in various media under different conditions including standard diffusion and flow at different speeds and volumes. Experimental work will be supplemented with theoretical analysis and modeling of transport processes, including studies of diffusion and chemical exchange using a standard multi-compartment model.
J01-7:

A. V. Stolin1, A. G. Weisenberger1, J. E. McKisson1, S. Majewski2 1 Detector and Imaging Group, Thomas Jefferson National Accelerator Facility, Newport News,VA, USA 2 Radiology, West Virginia University, Morgantown, WV, USA

Feasibility Study of Using Detection of Direct Positrons in Plant Imaging Research

While nuclear methods find application in plant biology there are very limited amount of instrumentation that is geared to the unique challenges that plant biology presents. Mostly researchers are utilizing devices created for clinical and preclinical imaging. The majority of the biologically significant elements involved in plant formation have positron emitting isotopes, making positron emission tomography a naturally fitting imaging modality. One particular caveat in using PET instrumentation in plant imaging is extremely small thickness of the leaf as an annihilation medium for emitted positrons. For positron energies around 1MeV ( as in case of C-11), about 1 mm of water equivalent material is required for positron conversion, so most of the positrons would not annihilate inside the leaf, therefore limiting both sensitivity and spatial resolution of the imaging system. In this work we made an attempt of determining whether direct imaging of positrons could be beneficial in plant biology studies. We compared performance of direct positron detector with a planar PET system. Detection efficiency ratio of ~ 650 in favor of direct positron detection system was measured. Direct comparison of spatial resolutions yielded comparable numbers for a setup, when a phantom was placed directly onto the entrance window of the positron detector. We also determined quick deterioration of spatial resolution as a function of source to detector distance. The last observation makes it necessary to maintain close proximity of the object to the detector, but this condition can be met in plant biology study.
N09: Computing and Software for Experiments I: Simulation Tuesday, Oct. 27 N09-1: 08:00-10:00 Grand Ballroom 2

MCNPX 2.7.X New Features Being Developed

G. W. McKinney, M. R. James D-5 Group, Los Alamos National Laboratory, Los Alamos, NM, USA MCNPX is the Monte Carlo N-Particle eXtended version of MCNP4C that has been under continuous development for more than a decade. In addition to all of the MCNP4C capabilities, MCNPX can transport 34 different particle types and heavy ions up to the teravolt energy range, offers a variety of modern high-energy physics models, contains an integrated transmutation and activation package, and much more. In this paper we describe recent developments of interest to the Monte Carlo transport community and provide results to demonstrate the use of these features. Since the release of MCNPX 2.5.0 in 2005, over 30 new features have been developed and are now part of the 2.6.0 release of MCNPX (April 2008). These new capabilities are divided into five broad categories: physics, source, tally, variance reduction, and other enhancements.

26

N09-2:

Geant4-Related R&D for New Particle Transport Methods

M. G. Pia INFN Genova, Italy, Genova, Italy On behalf of the Geant4-Nano5 Team Geant4 is nowadays a mature Monte Carlo system, which is widely used in a variety of experimental applications. Since the first release in 1998, new functionality has been added to the toolkit; nevertheless, the architectural design and fundamental concepts defining Geant4 application domain have remained substantially unchanged since their original conception. New experimental requirements have emerged in the recent years, which challenge general-purpose Monte Carlo transport codes like Geant4 in domains like nanodosimetry, nanotechnology-based detectors, radiation effects on components, nuclear power, plasma physics etc. A R&D project has been launched in 2009 to address fundamental methods in radiation transport simulation and revisit Geant4 kernel design to cope with these new experimental requirements. The project focuses on simulation at different scales in the same experimental environment: this set of problems requires new methods across the current boundaries of condensed-random-walk and discrete transport schemes. An exploration is also foreseen about exploiting and extending already existing Geant4 features to apply Monte Carlo and deterministic transport methods in the same simulation environment. An overview of this new R&D associated with Geant4 is presented, together with the first developments in progress.
The authors acknowledge S. Bertolucci, S. Giani, A. Montanari and A. Pfeiffer for valuable discussions. N09-3: Research in Geant4 Electromagnetic Physics Design, and Its Effects on Computational Performance and Quality Assurance

M. G. Pia1, P. Saracco1, M. Sudhakar1, A. Zoglauer2, M. Augelli3, E. Gargioni4, C. H. Kim5, L. Quintieri6, P. Queiroz7, D. de Souza Santos7, R. Schulte8, A. Wroe8, G. Weidenspointner9, B. Grosswendt10, M. Begalli11 1 INFN Genova, Italy, Genova, Italy 2 Space Sciences Laboratory, University of California at Berkeley, Berkeley, CA, USA 3 CNES, Toulouse, France 4 University Medical Center Hamburg-Eppendorf, Hamburg, Germany 5 Hanyang University, Seoul, Korea 6 INFN LNF, Frascati, Italy 7 IRD, Rio de Janeiro, Brazil 8 Medical Center, Loma Linda University, Loma Linda, CA, USA 9 MPE-MPI, Munich, Germany 10 PTB, Braunschweig, Germany 11 State University of Rio de Janeiro, Rio de Janeiro, Brazil

The Geant4 toolkit offers a rich variety of electromagnetic physics models; so far the evaluation of this Geant4 domain has been mostly focused on its physics functionality, while the features of its design and their impact on simulation accuracy, computational performance and facilities for verification and validation have not been the object of comparable attention yet, despite the critical role they play in many experimental applications. A project is in progress to study the application of new design concepts and software techniques in Geant4 electromagnetic physics, and to evaluate how they can improve on the current simulation capabilities. The application of a policy-based class design is investigated as a means to achieve the objective of granular decomposition of processes; this design technique offers various advantages in terms of flexibility of configuration and computational performance. The current Geant4 physics models have been re-implemented in terms of the new design as a pilot project. The main features of the new design and first results of performance improvement and testing simplification are presented; they are relevant to many Geant4 applications, where computational speed and the containment of resources invested in simulation production and quality assurance play a critical role.
N09-4:

Efficient Transport Simulations of Difficult Detection Problems Using ADVANTG

S. W. Mosher, T. M. Evans, T. M. Miller, J. C. Wagner Nuclear Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA In the transport community, it has been known for some time that Monte Carlo simulations of difficult shielding problems can be dramatically accelerated by employing a weight window game, where the weight targets are based on an approximate importance (adjoint) function. Such an importance function can be generated efficiently using a deterministic diffusion or transport code and a coarse model of the problem. For source-detector problems, a wealth of experience has shown this technique to be robust. It would be reasonable for one to suppose, then, that this methodology is practiced widely and routinely. In fact, this is not the case. Until very recently, no computational tools were widely available to automate the process of generating weight-window 27

parameters in this way. In this summary, we briefly describe the ADVANTG (Automated VariANce reducTion Generator) code, which has been developed at the Oak Ridge National Laboratory to implement the Consistent Adjoint Driven Importance Sampling (CADIS) methodology. In addition, we discuss the application of ADVANTG to a problem in which a cargo container holding a shielded neutron or gamma source is placed near a portal monitor. For these problems, ADVANTG obtains an O(104) speedup and, for a detailed gamma spectrum tally, an average O(102) speedup relative to implicit-capture-only simulations, including the deterministic calculation time. These and other results indicate that ADVANTG is an effective tool for modeling difficult shielding problems.
N09-5:

Validation of Fluorescence Transition Probability Calculations

M. G. Pia, P. Saracco, M. Sudhakar INFN Genova, Italy, Genova, Italy A systematic and quantitative validation of the K and L shell X-ray transition probabilities calculated according to different theoretical methods has been performed against experimental data. The values resulting from Hartree-Slater and Hartree-Fock calculation models are compared to an extensive collection of experimental data by means of rigorous statistical methods. This study is relevant to the selection and optimization of atomic data libraries based on these theoretical calculations, which are used by software systems dealing with X-ray fluorescence, namely Monte Carlo codes and PIXE analysis. Among them, the Evaluated Data Library (EADL) is used by Monte Carlo systems like EGSnrc, Geant4 and MCNP; it is based on Hartree-Slater calculations. The results of the validation support transition probabilities calculated according to the Hartree-Fock approach, which appear more accurate with respect to experimental measurements. The conclusions of this study suggest the need of a revision of the EADL data library: its update to reflect the state-of-the-art in this field of atomic physics would significantly contribute to the accuracy of the Monte Carlo codes based on it.
N09-6:

A Lewis-Theory for Landau/Vavilov Energy Straggling

A. K. Prinja, P. H. Smith Chemical and Nuclear Engineering, University of New Mexico, Albuquerque, NM, USA Recent numerical work has provided strong evidence for the existence of a Lewis-like theory underlying moment preserving approximations for energy straggling. Specifically, it has been noted that approximate transport models for energy straggling that are explicitly constructed to preserve a finite number of energy-loss moments of the analog differential cross section yield more accurate dose distributions and energy spectra than those that fail to capture these moments. Moreover, the accuracy can be made to approach analog with increasing numbers of true moments retained and very accurate results can be realized with surprisingly few moments. In this paper, we present a theoretical analysis that explains the high accuracy of numerical results observed for moment-preserving approximations to the Landau/Vavilov straggling model. We show explicitly that a finite number of moments of the energy spectrum or flux can be exactly preserved if the same number of moments of the corresponding energy-loss differential cross section are also exactly preserved. This result provides confidence in the application of moment-preserving straggling theories to charged particle penetration in thin targets as well as to energy straggling in condensed history Monte Carlo methods.
N09-7:

M. G. Pia1, T. Basaglia2, Z. W. Bell3, P. V. Dressendorfer4, A. Larkin4 1 INFN Genova, Italy, Genova, Italy 2 CERN, Geneva, Switzerland 3 ORNL, Oak Ridge, USA 4 IEEE, Piscataway, USA The Geant4 reference paper published in NIM A in 2003 has become the most cited paper in the whole Nuclear Science and Technology category of Thomson-Reuters Journal Citation Reports, the official reference for impact factor determination. Recently, Thomson-Reuter classified it among the current classic paper selections. This paper is currently the second most cited article ever among the publications authored by two major research institutes, CERN and INFN. This outstanding performance contrasts with the tradition of nuclear science and technology research, which is largely dominated by hardware rather than software - R&D, and with the relatively lower representation of software articles in scholarly literature concerning particle and nuclear physics, and related disciplines. The Geant4 reference paper published in TNS in 2006, though less cited than the previous one, has contributed significantly to the impact factor evaluation of the hosting journal, collecting by itself approximately 11% of the citations to this journal over the period 2006-2008. An overview of Geant4 presence (and absence) in scholarly literature is presented, and its implications are discussed. The patterns of this remarkable performance of Geant4 in the literature have been quantitatively examined; they highlight the

Geant4 in Scientific Literature

28

evolution of technological research in particle physics and related fields, and the impact of Monte Carlo simulation software in the experimental realm.
N10: Trigger and Front-End Systems I Tuesday, Oct. 27 N10-1: 08:00-10:00 Grand Ballroom 3

A New Paradigm Using GPUs for Fast Triggering and Pattern Matching at the CERN Experiment

G. Lamanna1, G. Collazuol1, G. Ruggiero1, M. Sozzi2 1 Scuola Normale Superiore & INFN, Pisa, Italy 2 University & INFN, Pisa, Italy We describe a pilot project for the use of GPUs (the standard computer video card processor) in an online triggering application at the CERN NA62 experiment, and the results of the first field tests together with a prototype data acquisition system. This approach is promising to achieve very high resolution in online reconstruction of physical observables used in trigger systems.
N10-2:

NA62

J. T. Anderson1, K. Byrum1, G. Drake1, A. Kreps1, F. Krennrich2, M. Schroedter2, A. W. Smith1 1 Argonne National Laboratory, Lemont, IL, USA 2 Iowa State University, Ames, IA, USA

Field Tests of a New High-Speed Pattern Recognition Trigger for Ground-Based Gamma-Ray Telescope Arrays

We have developed a three-stage, high-speed trigger for use in an array of imaging atmospheric Cherenkov telescopes (IACTs), which are used for the study of cosmic high-energy gamma-ray sources. This trigger has the ability to recognize patterns of Cherenkov light generated by gamma-ray initiated air showers in the atmosphere and correlate these patterns across multiple telescopes in the IACT array to form a stereoscopic real-time pattern recognition trigger. New hardware and firmware sampling at 400MHz with programmable coincidence recognition timing and programmable delay compensation over 500 pixel channels in an IACT camera has been produced which generates triggers from the coincidence of any three adjacent pixels within the camera. Reduction of the required coincidence time to 5ns and the concomitant reduction in dead time from the faster logic allow operation at lower discriminator thresholds relative to existing systems, enabling studies of lower energy gamma-ray events. A successful field test of the first two stages of this pattern recognition hardware has been performed on one of the IACTs of the VERITAS array located in Southern Arizona. We present the results of these tests and compare them to the performance of the existing VERITAS trigger system. A subsequent field test where hardware is connected to multiple VERITAS telescopes to exercise the stereoscopic features is in the planning stages.
N10-3:

An FPGA-Based Trigger for the Search of Mu -> E + Gamma Decay in the MEG Experiment

L. Galli Pisa University and INFN of Pisa, Pisa, Italy On behalf of the MEG Collaboration An FPGA-based trigger has been set up and is now operating in the MEG experiment at Paul Scherrer Institut, which aims at searching for the Lepton Flavour violating decay mu -> e + gamma with unprecedented sensitivity (10^(-13) for the branching ratio). An overview of the electronics is given with particular emphasis on the use of a Field Programmable Gate Array on board chip to implement the on-line event selection. The use of FPGAs makes the system adjustable to all the experimental requests and also permits to achieve optimum resolution on reconstructed observables while keeping latency at the order of 400 ns.
N10-4:

Feature-Extraction Algorithms for the PANDA Elctromagnetic Calorimeter

M. Kavatsyuk KVI/University of Groningen, Groningen, The Netherlands On behalf of the PANDA collaboration The PANDA collaboration at the future FAIR facility at Darmstadt, Germany, will employ antiproton annihilations to investigate yet undiscovered charm-mesons and glueballs with the aim to explain the origin of the hadronic mass spectrum. A multi-purpose detector for tracking, calorimetry and particle identification is presently being developed. A high discovery potential requires the experiment to run at high luminosities providing up to 2*107 interactions/s. In order to gain high flexibility for physics event selection, a triggerless data-acquisition system is employed, so that each sub-detector has to continuously provide data from 29

incoming physics events. Such conditions require new developments of the front-end electronics. The Electromagnetic Calorimeter (EMC) of PANDA comprises of the central target calorimeter, read out by large-area avalanche photodiodes and the forward endcap EMC, read out by vacuum phototriodes. Lead tungstate has been chosen as scintillator material. To achieve the required dynamic range of 10000, low-noise low-power preamplifiers were designed, namely the discrete component LNP [1] and the ASIC APFEL II [2]. With Sampling ADC readout energy and time information will be derived from the same signal. This approach complies with the triggerless data-acquisition system and avoids analogue delays. Dead time is eliminated as data will be processed on-line by fast FPGA. The feature-extraction algorithm should use only few FPGA resources, require a moderate sampling rate and be able to provide optimum energy resolution and a time resolution better than 1 ns. Further, the system should be able to operate at 500 kHz per channel incoming event rate. Feature-extraction algorithms fulfilling these requirements will be presented. The results of test-beam measurements with an EMC prototype will be discussed. [1] W. Erni, M. Steinacher, PANDA TPR, Univ. Basel, February 2005. [2] P. Wieczorek, H. Flemming, GSI report (2007) 30.
N10-5: Results from the ATLAS Barrel Level-1 Muon Trigger Timing Studies Using Combined Trigger and Offline Tracking

G. Salamanna Nikhef, Amsterdam, Nederland On behalf of the ATLAS Collaboration The ATLAS Level-1 Muon Barrel Trigger is one of the main elements of the first stage of event selection of the ATLAS experiment at the Large Hadron Collider. The challenge of the Level-1 system is a reduction of the event rate from a collision rate of 40 MHz by a factor 0(3), using simple algorithms that can be executed with a latency of the order of 1 microsecond. The input stage of the Level-1 Muon consists of an array of processors receiving the full granularity of data from a dedicated detector (Resistive Plate Chambers in the Barrel). Because of the different time-of-flights and cables and optical fiber lenghts, signals have to be adjusted in time in order to be correctly aligned before being processed. We present the analysis technics developed to allow for a study of time misalignements both among the RPC trigger sectors and globally with the LHC clock. These studies collect and integrate several informations, such as trigger times from the ATLAS Central Trigger Processor, both from the RPC and external triggers (e.g. the Inner Detector); and offline reconstructed muon tracks, that provide us with the ability to perform topology-based studies of the trigger timing by knowing the muon trajectory. These techniques have been tested using a large statistics sample of cosmic muon events in combined runs with the rest of the ATLAS detector; and are expected to provide a fast and detailed mapping of the status of the timing calibration with the early LHC beam data.
N10-6:

Operational Experience of the ATLAS High Level Trigger with SingleBeam and Cosmic Rays

I. Aracena SLAC, Stanford, CA, USA On behalf of the ATLAS Collaboration After giving an overview of the ATLAS trigger design and its innovative features, this paper focuses on the operational experience gained in running the trigger in the fast-changing environment of the detector commissioning. It will emphasize the commissioning of the High Level Trigger (HLT) system, including its monitoring and configuration. Preliminary results from initial LHC running in 2009 will be included if available. ATLAS is one of two general-purpose detectors at the LHC. Using fast reconstruction algorithms, the trigger system needs to efficiently reject a large rate of background events while keeping potentially interesting ones with high efficiency. After a first level trigger implemented in custom electronics, the trigger selection is made by software running on two processor farms (the High Level Trigger system), containing a total of around two thousand multi-core machines. To reduce the network data traffic and the processing time to manageable levels, the HLT uses seeded, step-wise event reconstruction, aiming at the earliest possible rejection of background events. The LHC startup and single-beam run periods in 2008 provided a "stress test" of the trigger system. Following this period, ATLAS continued to collect cosmic-ray events for detector alignment and calibration as well as for commissioning the trigger. These running periods allowed us to exercise the trigger system online, including its configuration and monitoring infrastructure, as well as reconstruction and selection algorithms. Several tracking, muon-finding, and calorimetry algorithms were commissioned under different running conditions. Frequent changes of the trigger configuration were required to cope with the parallel commissioning of the ATLAS subdetectors. The experience gained while running the trigger system online was very valuable to design and implement an optimal strategy for the collision data taking period of 2009.

30

N10-7:

The ATLAS Tau Trigger

S. Tsuno KEK, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan On behalf of the ATLAS T/DAQ Collaboration The ATLAS tau trigger has three levels: the first one (L1) is hardware based and uses FPGAs, while the second (L2) and third levels (EF -Event Filter-) are software based and use commodity computers (2 x Intel Harpertown quad-core 2.5 GHz), running scientific linux 4. In this contribution we discuss both the physics characteristics of tau leptons and the technical solutions to quick data access and fast algorithms. We show that L1 selects narrow jets in the calorimeter with an overall rejection against QCD jets of 300, whilst L2 and EF (referred together as High Level Trigger -HLT-) use all the detectors with full granularity and apply a typical rejection of 15 within the stringent timing requirements of the LHC. In the HLT there are two complementary approaches: specialized, fast algorithms are used at L2, while more refined and sophisticated algorithms,imported from the offline, are utilized in the EF.
N11: Gaseous Detectors I: Development of Techniques Tuesday, Oct. 27 N11-1: 08:00-10:00 Grand Ballroom 7

S. Duarte Pinto1,2, I. Brock2, E. David1, L. Ropelewski1, M. van Stenis1, H. Taureg1, R. de Oliveira1 1 CERN, Geneva, Switzerland 2 Physikalisches Institut der Universitt Bonn, Bonn, Germany

Spherical GEMs for Parallax-Free X-Ray Imaging

We developed a method to make GEM foils with a spherical geometry. Tests with this procedure and with the resulting spherical GEMs are presented. Together with a spherical drift electrode, a spherical conversion gap for x-rays can be formed. This would eliminate the parallax error in an x-ray diffraction setup, which limits the spacial resolution at wide diffraction angles. The method is inexpensive and flexible towards possible changes in the design. We show advanced plans to make a prototype of an entirely spherical triple-GEM detector, including a spherical readout structure. This detector will have a superior position resolution in both angular coordinates, also at wide diffraction angles, and a high rate capability. A completely spherical gaseous detector has never been made before.
N11-2: Study of New Materials for Bakelite Resistive Plate Chamber Operation in Avalanche and Streamer Modes

A. Sharma PH, CERN, Geneva, Switzerland The Resistive Plate Chambers (RPCs) are employed widely for various particle physics and neutrino-based experiments to detect high energy charged particles especially fast muons, due to their excellent time resolution, reasonable space resolution and high rate capability. High gain, simple electronics, robustness and low cost of these detectors have sparked off a lot of interest for cosmic-ray and low-rate accelerator experiments with operation both in streamer and avalanche mode. In high rate experiments such as those at LHC operation in avalanche mode is essential. In this work, new materials and coatings have been investigated for the construction of medium-sized RPCs. The bakelites with good electrical properties and coted with silicone have been used to build RPCs of various sizes (30cm x 30cm and 1m x 1m operated in streamer mode and 10cm x 10cm operated in avalanche mode) showing >92% efficiency and ~ 2 nsec time resolution for long-term operations. In order to study the suitability of this material for operation in high dose environment, these chambers are now going to be exposed to a high flux of photons at the Gamma Irradiation Facility at CERN. The goal is to monitor the performance during and after the irradiation dose to validate long term operation in avalanche and streamer modes. Results on validation of the new materials and subsequent chamber operation will be presented.
N11-3:

The Performance of GridPix Detectors

M. Fransen Nikhef, Amsterdam, Netherlands The performance of GridPix detectors Y. Bilevych, S. Bentvelsen, M. Chefdeville, A-P. Colijn, M. Fransen, H. van der Graaf, F. Hartjes, N. Hessey, P. de Jong, T. Ketel, E. Koffeman, W.C. Koppert, L.de Nooij, J. Timmermans, J.L. Visschers Nikhef, Amsterdam, The Netherlands V. Blanco Carballo, J. Schmitz Univ. of Twente, The Netherlands 31

N. de Groot, A. Konig, M. Rogers Radboud University, Nijmegen, The Netherlamds R. Kalter TU Delft, Delft, The Netherlands The GridPix detector is an assembly of a TimePix pixel chip, covered with a layer of silicon rich silicon nitride (SiNi) as protection against discharges, an integrated MPGD grid (InGrid), a drift gap and a cathode plane. Gossip is a GridPix detector with a drift gap of ~1mm. The concept is to replace the silicon in vertex detectors by ~1mm of gas. The performance of Gossip has been determined by simulations and by doing measurements in a testbeam at CERN. DICE is a GridPix detector with a drift gap of ~2cm and is used for nuclear research at the TU Delft. DICE also is used as reference detector for testing Gossip detectors in a testbeam. GridPix detectors containing 4 TimePix chips are being build as prototype TPC readout for ILC/CLIC studies (drift gap ~1m). In parallel a new type of optical data transmission (IfLink) and a power supply through optical fiber are being developed to reduce the amount of electrical cabling inside a (vertex) detector.
N11-4:

C. A. B. Oliveira1, J. J. F. C. A. Veloso1, A. L. Ferreira1, S. Biagi2, R. Veenhof3, J. M. F. dos Santos4, C. M. B. Monteiro4 1 Physics Department, University of Aveiro, Aveiro, Portugal 2 Physics Department, University of Liverpool, Liverpool, UK 3 SFT group, PH Department, CERN, Geneva, Switzerland 4 GIAN, Physics Department, University of Coimbra, Coimbra, Portugal Scintillation emitted during avalanche development in gaseous avalanche detectors is an useful information for triggering and/or tracking in gaseous detectors. Scintillation gases, like noble gases or CF4 present high values for electroluminescence yield, emitting mainly in the VUV region. The photons can be used for signal readout, using an appropriate photosensor. It was experimentally demonstrated that statistical fluctuations in the number of photons produced along the avalanche are not higher than those in the number of electrons produced. Recent experimental results show that micropattern detectors are good candidates for application in low-background and low rate experiments, as Dark Matter search, due to their high electroluminescence absolute yield which permit high gains. In this way, statistical fluctuations, electroluminescence yield, time properties and photon collection efficiency are important parameters to quantify, in order to understand how to use and take advantage of scintillation additional information. A simulation tool based on Magboltz/Garfield programs is under development to assess these parameters by following the excited states produced along the avalanche. Simulation results of the referred properties by applying the model to different micropattern gas detectors operating with different noble gases will be presented. Comparison between light and charge collection properties will be performed.
This work was supported by project CERN/FP/83487/2008 under the FEDER and FCT (Lisbon) programs. C. A. B. Oliveira is supported by FCT under Doctoral Grant SFRH/BD/36562/2007. N11-5:

Electroluminescence Assessment in Micropattern Gaseous Avalanche Detectors

T. Tamagawa1,2, A. Hayato1,2, T. Iwahashi1,2, S. Konami1,2, F. Asami1,2, H. Hamagaki3, Y. L. Yamaguchi3, K. Makishima4,1 1 Cosmic Radiation Laboratory, RIKEN, Wako, Saitama, Japan 2 Tokyo University of Science, Shinjuku, Tokyo, Japan 3 CNS, University of Tokyo, Wako, Saitama, Japan 4 University of Tokyo, Hongo, Tokyo, Japan We have produced gas electron multipliers (GEMs) using a laser etching technique in Japan since 2002. So far we have successfully developed fine pitch GEMs with a pitch of 50 micron and a hole diameter of 30 microns, thick-foil GEMs with a thickness of 100 microns, etc. In this paper, we show that the recent progress in producing the fine-pitch GEMs including the production process of the GEMs with various geometries and insulator materials, and show the gain characteristics, stability and uniformity of the GEMs. We also show the future satellite missions in astrophysics which out GEMs will be used for.
N11-6: Long-Term Study of the Operation of LHC Resistive Plate Chambers in an Optimized Closed-Loop Gas System

Progress of Fine-Pitch GEM Development in Japan Using a Laser Etching Technique

M. Capeans, I. Glushkov, R. Guida, F. Hahn, S. Haider Physics Department, CERN, Geneva, Switzerland The RPC systems installed as part of the large muon detectors of the LHC experiments are operated in a gas mixture containing 94-96% C2H2F4, 3-5% iC4H10 and 0.3% SF6. In this gas mixture, RPCs operated in a high radiation environment produce a large amount of impurities in the return gas that are potentially dangerous for the stable operation of the detectors, the materials in the detector and the gas system. We present the results of a series of systematic studies carried out in a dedicated set-up at the CERN Gamma Irradiation Facility in order to understand the impurities produced in the gas of highly irradiated RPCs, the properties of possible absorbers of those impurities and the long-term performance of RPCs operated in a optimized closed-loop gas system. The results of these tests will be used and implemented in the LHC and sLHC RPC systems. 32

N11-7:

G. C. Smith1, N. A. Schaknowski1, B. Yu1, D. A. Fischer2 1 Instrumentation Division, Brookhaven National Laboratory, Upton NY, USA 2 Materials Science and Engineering Laboratory, National Institute of Standards and Technology, Gaithersburg MD, USA A compact, efficient, soft X-ray detector has been designed and fabricated for use in synchrotron experiments that use x-ray absorption spectroscopy to study a wide range of materials properties. Fluorescence X-rays, in particular carbon K at 277eV, are converted in low pressure gas, and charge multiplication is derived from two gas electron multipliers, fabricated in-house from glass reinforced thermoset laminate, to enable single photon counting. The detector satisfies a number of demanding characteristics often required in synchrotron environments, such as compactness, UHV compatibility, energy resolving capability and long-term stability.
N12: High Energy Physics Instrumentation I Tuesday, Oct. 27 N12-1: 08:00-10:00 Grand Ballroom 8

A Low Pressure, Soft X-Ray Detector Using Gas Electron Multipliers

The Challenges and Possible Implementations of Super-LHC Detector Upgrades

S. Stapnes Univ. of Oslo, Oslo, Norway The LHC will soon begin operation and build up to nominal luminosity over the next few years. This will be followed by a programme of improvements leading eventually to a ten-fold increase above nominal, referred to as super-LHC (sLHC), providing a very challenging environment for detectors and electronics. Substantial upgrades to the general purpose experiments ATLAS and CMS will be necessary to deliver large data sets with good performance under these conditions, and also the more specialized experiments LHC-b and ALICE are preparing upgrades of their detectors. Radiation damages and ageing will also necessitate improvements of parts of the detectors. The physics goals and machine luminosity determine the performance requirements of the various detector parts. Much of the current detectors and all the magnets will work well at the sLHC, but several items including the entire inner trackers, some of the forward detectors, and substantial amounts of trigger and readout electronics, will need upgrading to cope with the high particle and background rates, and increased integrated radiation doses. This paper presents some of the main challenges involved in upgrading the LHC detectors and the main changes foreseen.
N12-2:

ATLAS Silicon Microstrip Tracker Operation

E. Berglund DPNC, University of Geneva, Geneva, Switzerland On behalf of the ATLAS SCT The ATLAS experiment at the CERN Large Hadron Collider (LHC) has started taking data last autumn with the inauguration of the LHC. The SemiConductor Tracker (SCT) is the key precision tracking device in ATLAS, made up from silicon micro-strip detectors processed in the planar p-in-n technology. The completed SCT has been installed inside the ATLAS experimental hall. Since then the detector was operated for many months under realistic conditions. Calibration data has been taken and analyzed to determine the noise performance of the system. In addition, extensive commissioning with cosmic ray events has been performed both with and without magnetic field. The cosmic muon data has been used to align the detector, to check the timing of the frontend electronics as well as to measure the hit efficiency of modules. The sensor behavior in magnetic field was studied by measurements of the Lorentz angle. For the initial running with unfocused LHC beam operation with undepleted sensors is foreseen. Efficiency and noise determination for various bias voltages was also performed. The current status of the SCT will be reviewed, including results from this year's latest data-taking periods, and from the detector alignment. We will report on the commissioning of the detector, including overviews on services, connectivity and observed problems. The SCT commissioning and running experience will then be used to extract valuable lessons for future silicon strip detector projects.
N12-3:

Results from the Commissioning of the ATLAS Pixel Detector with Cosmics Data.

C. Troncon INFN, INFN - Milano, Milano, Italy On behalf of the ATLAS Pixel Detector Collaboration The ATLAS Pixel Detector is the innermost detector of the ATLAS experiment at the Large Hadron Collider at CERN. It consists of silicon sensors equipped with approximately 80 M electronic channels and will allow to detect particle tracks and secondary vertices with very high precision. After connection of cooling and services and verification of their operation, the ATLAS Pixel Detector is now in the final stage of its commissioning phase. Prior to the first beams expected in Summer 2009, a 33

full characterization of the detector is performed. Calibrations of optical connections, verification of the analog performance and special DAQ runs for noise studies are ongoing. Combined operation with other subdetectors in ATLAS allows to qualify the detector with physics data from cosmic muons. The talk will show all aspects of detector operation, including the monitoring and safety system, the DAQ system and calibration procedures. The summary of calibration tests on the whole detector as well as analysis of physics runs with cosmics data will be presented.
N12-4:

Commissioning the CMS Pixel Detector with Cosmic Rays

M. E. Dinardo Physics, University of Colorado, Boulder, Colorado, USA On behalf of the CMS pixel collaboration The Compact Muon Solenoid (CMS) is one of two general purpose experiments at the Large Hadron Collider. The CMS experiment prides itself on an ambitious, all silicon based, tracking system. After almost 20 years of design and construction the CMS tracker detector has been installed and commissioned. The tracker detector consists of ten layers of silicon microstrip detectors and three layers of pixel detectors near the interaction point. The pixel detector consists of 66M pixels of 100x150m2 area, and is designed to exploit the shape of the actual charge sharing among adjacent pixels to gain hit resolution down to 12m. Results from cosmic ray studies with the CMS pixel detector will be presented, including the current status in calibration, alignment, data quality monitoring and track reconstruction.
N12-5:

M. Barbero1, D. Arutinov1, R. Beccherle2, G. Darbo2, S. Dube3, D. Elledge3, D. Fougeron4, M. Garcia-Sciveres3, D. Gnani3, V. Gromov5, T. Hemperek1, M. Karagounis1, R. Kluit5, A. Kruth1, A. Mekkaoui3, M. Menouni4, J. D. Schipper5, N. Wermes1 1 University of Bonn, Bonn, Germany 2 University of Genova, Genova, Italy 3 LBNL, Berkeley, USA 4 CPPM Aix-Marseille University, Marseille, France 5 NIKHEF, Amsterdam, The Netherlands The new ATLAS pixel chip FE-I4 is being developed for use in upgraded luminosity environments, in the framework of the Insertable B-Layer (IBL) project but also for the outer pixel layers of Super-LHC. FE-I4 is designed in a 130 nm technology and is based on an array of 80 by 336 pixels, each 50 by 250 m2 and consisting of analog and digital sections. The analog pixel section is designed for low power consumption and compatibility to several sensor candidates. The digital architecture is based on a 4 pixel unit called region, which allows for a power-efficient, low recording inefficiency design, and provides an elegant solution to the problem of timewalk. The chip periphery contains a control block, powering blocks, a data reformatting unit, an asynchronous storage FIFO, an 8b10b coder and a clock multiplier unit, which handles data transmission up to 160 Mb/s for the IBL. An increased power consumption in the inner layers of ATLAS translates into more material for cooling and power routing, which degrades the tracking and the b-tagging quality. Hence, the FE-I4 collaboration places severe constraints on the power consumption of all blocks.
N12-6:

FE-I4: the New ATLAS Pixel Chip for Upgraded LHC Luminosities

Longevity Studies in the CDF Silicon Detectors

S. Behari Dept. of Physics and Astronomy, The Johns Hopkins University, Baltimore, MD, USA On behalf of the CDF II Silicon group The silicon detectors at the CDF Experiment in Run II have been collecting data steadily in a hard radiation environment for the last 8 years. This has led to the presence of some aging effects, such as type-inversion of the substrate or increase of the intrinsic noise, that are carefully monitored in order to maintain the detector performance at the highest level and to predict the useful lifetime of the different sensors. In this presentation, a summary of the studies is given as well as the current conclusions and expectations for the future of the detectors.

34

J02: Instrumentation for Medical and Biological Research II: X-ray Imaging and Radiotherapy Applications Tuesday, Oct. 27 J02-1: 10:30-12:30 International Ballroom North

C. Ozkan1,2, A. Castoldi2,3, C. Guazzoni2,3, A. Bjeoumikhov4, R. Hartmann5,6, L. Strueder7,5,8, G. Royle9 1 Dip. Fisica, Universita' degli Studi di Milano, Milano, Italy 2 INFN sez. Milano, Milano, Italy 3 Dip. Elettronica e Informazione, Politecnico di Milano, Milano, Italy 4 IfG -Institute for Scientific Instruments GmbH, Berlin, Germany 5 Max Planck Institut HalbleiterLabor, Muenchen, Germany 6 PNSensor GmbH, Muenchen, Germany 7 Max Planck Institut fur Extraterrestriche Physik, Garching, Germany 8 FB Physik, Universitat Siegen, Siegen, Germany 9 Dept. of Medical Physics, University College London, London, United Kingdom

Performance Assessment of a High Resolution X-Ray Scatter Imaging System with Multi-Momentum Transfer Capability

X-ray Scatter Imaging is a technique based on the capability of coherently scattered x-rays to probe the structure of matter at molecular level which has the potential to produce images with higher contrast and specificity than conventional transmission imaging. We developed a novel X-ray scatter imaging system featuring improved spatial resolution and able to exploit photon energy information which are the main limitations of present systems. To this purpose we replaced the mechanical collimation system with a polycapillary parallel x-ray collimator. This approach leads to a dramatic improvement of position resolution, that can be pushed down to few micrometers, thus shifting the limiting factor on the spatial resolution from the size of mechanical collimator to the detector pixel size. Moreover we employed a 2D spectroscopic x-ray imaging device - the Controlled Drift Detector (CDD) - that allows use of monochromatic as well as polychromatic sources (e.g. conventional X-ray generator) hence easing the translation from the synchrotron environment to the laboratory. This also opens the way to the fascinating possibility of (simultaneous) acquisition of images at multiple values of the momentum transfer. In order to fully asses the performance of the upgraded setup for X-ray scatter imaging, a thorough experimental campaign was carried out under several defined conditions both with synchrotron radiation and in the laboratory with a conventional source. The obtained results, as well as the feasibility and limitations of the proposed approach will be fully discussed.
J02-2:

An Automated Portable Instrument for Rapid Screening for Zinc Deficiency Using Non-Destructive XRay Fluorescence

J. J. Kehayias, E. A. Gruber, C. E. Kehayias USDA Human Nutrition Research Center on Aging, Tufts University, Boston, MA, USA Analytical method: The purpose of this work was to develop and validate a portable instrument for the rapid, automated recognition of chronic zinc deficiency without the need for blood collection, so that the instrument can be used in field studies. The traditional method for screening for zinc deficiency is based on atomic absorption in blood samples. It requires significant sample preparation and only provides a snapshot of the zinc status as the concentration varies with time. To avoid these problems, we selected to test human hair instead, using X-Ray Fluorescence (XRF) analysis. Experimental set up: A 0.5mm thick, 25mm2 Si-PIN photodiode detector is used for the X-ray detection (Amptek Inc, Bedford, MA, USA). Energy resolution <240 eV is achieved with electronic cooling of the detector assembly at 218 K. The sample is irradiated with a silver-housed Cd-109 radioactive source through the vertical wall of the 1.0cc specimen vial. The detector is positioned at 90 , a geometry that allows for the simultaneous use of a second radioactive source. Medical applications: Chronic zinc deficiency is common in malnourished children because of poor nutrition and in the elderly because of poor absorption. It causes growth impairment and immune dysfunction leading to infections and especially pneumonia. The treatment is usually easy, but identifying this deficiency with a simple, widely-available field method has challenged IAEA scientists. Results: Sensitivity was estimated to about 5 micrograms for 10min of acquisition time, sufficient for our screening method that classifies 100 micrograms of zinc per g of hair as normal. Reproducibility for the same specimen was better than 2%. Compared to atomic absorption, we found that because of the lack of sample preparation the XRF method was more reproducible but not as sensitive. However, its lower detection limit is well below the level of zinc malnutrition and therefore qualifies for field use.
This work was supported by the US Department of Agriculture/ ARS 51000-061.

35

J02-3:

An Attenuation-Refraction-Scattering Computed Tomographic Experimental System with a Conventional X-Ray Tube: Systemic Optimization and Image Fusion

Z. Huang, L. Zhang, K. Kang Dept. of Eng. Phys., Tsinghua Univ., Beijing, China In this paper, a projection-based grating-based method under the incoherent condition different from Talbot-Lau interferometric method under the coherent condition is presented to realize attenuation-refraction-scattering computed tomography with a conventional X-ray tube. In theory, the systemic parameters, including the pitches of the gratings, the geometrical relationship and the energy of the X-rays, etc, were optimized by a Monte Carlo simulation according to the existing level of grating fabrication, the resolution of the detector and the focal spot size of the X-ray tube. Based on this theory, an attenuation-refractionscattering computed tomographic experimental system has been built at Tsinghua University in China, which can synchronously reconstruct the distributions of linear attenuation coefficient, the refractive index gradient and the second moment of the scattering angle of the tested sample in only one computed tomography scanning process. A fusion method of three kinds of images is presented to show integrated structures of the sample. Some elementary experimental results proved that our method obtains most comprehensive physical characteristics of the sample. A comparison of projection-based method and interferometric method is discussed theoretically finally.
This work was supported by a grant from the National Natural Science Foundation of China (No. 10875066) ([email protected] ). J02-4: Comparison of SOI Microdosimeter and Tissue Equivalent Proportional Counter Measurements at the CERF Facility

D. A. Prokopovich1, M. I. Reinhard1, G. C. Taylor2, A. Hands3, A. B. Rosenfeld4 1 Insitute of Materials Engineering, Australian Nuclear Science and Technology Organisation, Lucas Heights, Australia 2 National Physical Laboratory, Teddington, UK 3 Aerospace Division, QinetiQ, Farnborough, UK 4 Centre for Medical Radiation Physics, University of Wollongong, Wollongong, Australia

The CERN-EU High Energy Reference Field (CERF) facility is used in the calibration of neutron dosimeters for aviation, space and high energy physics applications. A comparison of the CERF facility's microdosimetric spectra obtained with a novel Silicon on Insulator (SOI) Microdosimeter and a HAWK Tissue Equivalent Proportional Counter (TEPC) are presented. Experimental measurements with a silicon PIN dosimeter diode are also described which indicate a small but significant contribution from charged particles to dosimetry measurements within the neutron dominated field.
J02-5:

H. Tanaka1, Y. Sakurai1, M. Suzuki1, S. Masunaga1, T. Mitsumoto2, S. Yajima2, H. Tsutsui2, T. Sato2, T. Asano3, G. Kashino1, Y. Kinashi1, Y. Liu1, K. Ono1, A. Maruhashi1 1 Department of Radiation Life Science and Radiation Medical Science, Kyoto University, Osaka, Japan 2 Sumitomo Heavy Industries, Tokyo, Japan 3 Stella Pharma Corporation, Osaka, Japan We have been developed an epithermal neutron source for boron neutron capture therapy(BNCT), consisting of a cyclotron accelerator that can provide a 1 mA, 30 MeV proton beam, a neutron production beryllium target and the moderator that can reduce the energy of fast neutrons to an effective energy range. In order to validate the simulations, we measured the depth distribution of the thermal neutron flux in water phantom located at the treatment position. The measured results were compared with the simulations using the MCNPX Monte Carlo code. The good agreement between the simulations and measurements was shown. The thermal neutron flux with the proton current of 430 A was 7.4 108(neutrons cm-2 s-1) at the depth of around 20 mm in the water phantom. This intensity corresponds to the neutron source of Kyoto University Research Reactor (KUR), at which 275 clinical trials of BNCT have been performed. We experimentally confirmed that our cyclotron based neutron source can use for clinical trials of BNCT for the first time in the world.

Measurement of the Thermal Neutron Distribution in a Water Phantom Using a Cyclotron Based Neutron Source for Boron Neutron Capture Therapy

36

J02-6:

D. M. Gonzalez-Castano1, F. Gomez1, L. Brualla2, J. Rosello2 1 Particle Physics, University of Santiago de Compostela, Santiago, Spain 2 Radiotherapy, ERESA, Valencia, Spain

Room Temperature Non-Polar Liquid Dosimetry for High Precision Radiotherapy

Nowadays, absolute radiotherapy dosimetry is performed with air filled ionization chambers which are reliable and whose output is easily converted to dose. Relative dosimetry is performed with radiographic or radiochromic film, with detector arrays or with a solid state detector attached to a motorized bench. The ideal characteristics of dosimeters are dose and spatial accuracy, linearity with dose, negligible dose rate nor energy dependence and readout convenience. Organic non polar liquids have been proposed as active medium to be employed in ionization-detection-based dosimeters. The Radiation Physics Research Group at Santiago de Compostela is developing several dosimetry systems based in 2,2,4 Tri-methylpentane and proved their operation in the clinical radiotherapy application. A linear matrix for verification of radiotherapy IMRT dose profiles with 1.4 mm3 cell (electric field 2 kV/mm) detector volume and 1.7 mm pitch has been used in the verification of treatments and commissioning of radiotherapy accelerators with a performance comparable to radiographic film devices in terms of space resolution. This device offers better relative reproducibility (standard deviation below 1%) than film and real time readout (100 projections per second). On the other hand the tissue-equivalent composition of the sensitive media offers a best matching of the clinical dosimetry. Additionally to prove the versatility of the device, we have produced a single ionization chamber based on this principle with a volume of 0.5 mm3, compatible with standard radiotherapy electrometers. This device has been tested in the dosimetry of small fields achieving similar results than a solid state semiconductor dosimeter. Tests of stability have shown that this device may be used also for absolute dosimetry since its stability showed a deviation of less than 2% after 30 kGy exposure along several weeks of irradiation. Currently a 2d matrix for radiosurgery dosimetry is under development.
Acknowledgments: This work was performed under the research contract PGIDIT06BTF20601PR from Xunta de Galicia J02-7:

A. L. Ziebell1, S. J. Dowdell1, M. I. Reinhard2, D. A. Prokopovich2, M. Petasecca1, M. L. Lerch1, B. J. Clasie3,4, A. J. Wroe5, R. W. Schulte5, V. Perevertaylo6, I. E. Anokhin7, A. B. Rosenfeld1 1 Centre for Medical Radiation Physics, University of Wollongong, Wollongong, NSW, Australia 2 Australian Nuclear Science and Technology Organisation, Menai, NSW, Australia 3 Massachusetts General Hospital, Boston, MA, USA 4 Harvard Medical School, Boston, MA, USA 5 Loma Linda University Medical Center, Loma Linda, CA, USA 6 SPO BIT, Kiev, Ukraine 7 Institute for Nuclear Research, Kiev, Ukraine In proton therapy, secondary neutrons are associated with an increased risk of second cancer induction. With the introduction of active beam delivery techniques, the secondary neutron dose delivered to patients has been reduced. However, neutrons produced via (p,n) reactions within the patient themselves cannot be avoided. As the radiation protection weight factor for neutrons can be as high as 25, it is important to have a quantitative understanding of the separate neutron and proton dose components delivered to a patient during proton beam therapy. A dosimetry technique to measure the separate dose components is presented. This method uses two planar Si pin diodes; one operated in a forward voltage mode to measure the non-ionising losses (NIEL) in Si, the other operated in standard spectroscopic mode to measure the ionising energy losses (IEL) in Si. These losses correspond to: Vf = Dn + Dp (1), and Q = (Dn + Dp) (2), where Dn and Dp are the respective neutron and proton absorbed tissue doses; Vf is the shift in forward voltage of detector1 due to NIEL; and Q is the charge deposited in detector2 as a result of IEL. In equation (1): >> as the ratio of neutron KERMA to proton KERMA is 10-100 for the particle energies expected. Callibration of the pin diode in Vf mode has been performed at the Harper Hospital fast neutron therapy facility, Detroit, and at the KEK Proton Therapy Facillity, Japan, for neutron and proton absorbed dose respectively. Results obtained at the clinical fast neutron beam, Harper Hospital, experimentally demonstrate the relative insensitivity of the Vf diode to protons in a mixed protonneutron field. Experimental results obtained using the dual detector system at the Francis H. Burr Proton Therapy Center, Massachusetts General Hosptial (MGH) will be presented along with GEANT4 simulation results. This work will contribute to the prediction of second cancers in proton therapy, irrespective of the technique used for beam delivery.

PIN Diodes for Measuring Out-of-Field Neutron Dose in Active Beam Proton Therapy

37

N13: Posters I Tuesday, Oct. 27 N13-1: 10:30-12:00 Grand Ballroom 4&5

TMRS MK III Engineering Analysis

K. A. Woloshun, G. G. Walthers-Ellis, R. A. Valicenti, J. A. O'Toole Los Alamos National Laboratory, Los Alamos, NM The Target-Moderator-Reflector System (TMRS) is the neutron source for nuclear physics experiments in the Lujan Neutron Scattering Center user facility. The TMRS is an assembly consisting of tungsten targets plus a network of moderators, reflectors and flight paths that deliver neutrons of specified energies and fluxes to surrounding experiments and detector stations. The second-generation TMRS (MK II), operational for 5 years, is scheduled for replacement in 2010. While most features of MK II will remain unchanged, all components of the new TMRS (MK III) have been analyzed for structural integrity and adequacy of cooling consistent with todays engineering standards. This analysis is presented in this paper, with emphasis on 2 features which deviate from the MK II design: 1) The tungsten components of the targets will be clad in tantalum to minimize erosion and consequent contamination of the water cooling system, and 2) A new H2-beryllium moderator/reflector system operating below 100 K will be incorporated.
N13-2:

TMRS MK III Engineering Design

J. A. O'toole, R. A. Valicenti, K. A. Woloshun AOT-MDE, Los Alamos National Laboratory, Los Alamos, NM, USA The Target-Moderator-Reflector System (TMRS) is the neutron source for nuclear physics experiments in the Lujan Neutron Scattering Center user facility. The TMRS is an assembly consisting of tungsten targets plus a network of moderators, reflectors and flight paths that deliver neutrons of specified energies and fluxes to surrounding experiments and detector stations. The second-generation TMRS (MK II), operational for 5 years, is scheduled for replacement in 2010. While many features of MK II will remain unchanged this paper describes the engineering design of important new features: 1) The tungsten components of the targets will be clad in tantalum to minimize erosion and consequent contamination of the water cooling system, and 2) A new liquid hydrogen moderator with a beryllium reflector/filter that doubles the neutron flux.
N13-3:

R. A. Valicenti1, T. Diaz2, A. T. Nelson3, J. A. O'toole1, D. F. Pruessmann4, K. A. Woloshun1 1 AOT-MDE, Los Alamos National Laboratory, Los Alamos, NM, USA 2 Diaz & Associates, Indian,Well, CA, USA 3 MST-8, Los Alamos National Laboratory, Los Alamos, NM, USA 4 Coronado Machine inc., Albuquerque, NM, USA The Target-Moderator-Reflector System (TMRS) is the neutron source for nuclear physics experiments in the Lujan Neutron Scattering Center user facility. The TMRS is an assembly consisting of tungsten targets plus a network of moderators, reflectors and flight paths that deliver neutrons of specified energies and fluxes to surrounding experiments and detector stations. The second-generation TMRS (MK II), operational for 5 years, is scheduled for replacement in 2010. While many components of the MK II will remain unchanged this paper describe in detail the fabrication and assembly of some of the more complex components of the MK III: 1) The tungsten components of the targets will be clad in tantalum to minimize erosion and consequent contamination of the water cooling system, and 2) A new liquid hydrogen moderator with a beryllium reflector/filter that doubles the neutron flux will be fabricated.
N13-4:

TMRS MK III Fabrication

D. J. Peake1, M. J. Boland1,2, G. S. LeBlanc2, G. J. O'Keefe1,3, R. P. Rassool1 1 School of Physics, The University of Melbourne, Parkville, Victoria, Australia 2 Accelerator Physics Group, The Australian Synchrotron, Clayton, Victoria, Australia 3 Center for PET, The Austin Hospital, Heidelberg, Victoria, Australia

Controlling Coupled-Bunch Instabilities at the Australian Synchrotron

In addressing the demands of advanced experiments on 3rd generation light sources, there is a quest to increase to stored beam current and reduce the overall beam size to enhance brilliance. However, these actions can be self-defeating as they can result in instabilities which may ultimately limit the photon flux available to the experimental users. Several techniques have been investigated at the Australian Synchrotron to passively and actively damp instabilities in the stored beam. During normal operations, our approach has been to use a vertical chromaticity of 11 to passively control coupled-bunch instabilities. More recently, we have completed the commissioning a "Bunch-By-Bunch" active feedback system designed around 38

the commercially available Libera Bunch-by-Bunch with hardware and software customised in-house. With all of the existing low-gap insertion devices fully engaged, it has been possible to maintain 200 mA of stored beam at nearzero vertical chromaticity, previously unachievable. Furthermore, the active feedback system also allows for a variety of beam diagnostic measurements to be taken including measuring instability growth rates and calculating beam parameters such as tune and chromaticity. Grow-damp analysis has been used to verify the performance of the system and quantify the potential improvements. This system is now vital for ensuring maximum possible photon flux for the beamline experimentalists.
N13-5:

D. J. Peake1, M. J. Boland1,2, G. S. LeBlanc1,2, G. J. O'Keefe1,3, R. P. Rassool1 1 School of Physics, The University of Melbourne, Parkville, Victoria, Australia 2 Accelerator Physics Group, The Australian Synchrotron, Clayton, Victoria, Australia 3 Center for PET, The Austin Hospital, Heidelberg, Victoria, Australia

Design, Construction and Diagnostic Methods of a Real-Time Fill Pattern Monitor at the Australian Synchrotron

Modern third generation light sources provide high-brilliance beams for experimental users, but to futher improve beam brilliance there is an increasing need to know about the temporal nature of the electrons. This article describes the design and construction of a device which allows real-time measurement of the electron density distribution within the storage ring at the Australian Synchrotron. Under the influence of the 500 MHz radiofrequency cavities, the bunch structure in the storage ring is 30-50ps wide gaussians seperated by 2 ns. Utilising the visible light portion of the synchrotron radiation spectrum, the system uses a 30GHz bandwidth MSM photodiode to measure the light emitted from the stored electrons. The signal is then amplified before being digitised by an 8GS/s Acqiris ADC CompactPCI card. An EPICS interface provides the processed signal to the control room which allows remote control of the parameters used to acquire the data. The real-time nature of the system linked with advanced control over the storage ring injection chain has led to the development of methods which enable repeated targeted injections, thereby maintaining an arbitrary fill pattern. This has been exploited in recent instability growth measurements to provide a known flat fill which eases theoretical calculations and allows for multiple measurements to be easily compared. In the future top-up mode user runs can be implemented using this technology. Fill pattern measurements are now being used as an essential tool in the control room for day-to-day operations and diagnostic methods.
N13-6:

Design and Development of Laser-RF Synchronization System for Thomson-Scattering X-ray Source at Tsinghua University

Q. Du, J. Li, C. Tang, W. Huang, Y. Du, L. Yan Dept. Engineering Physics, Tsinghua University, Beijing, China Femtosecond X-ray pulses could be generated by Thomson scattering between subpicosecond relativistic electron bunches and ultrashort infrared terra watt laser pulses. To ensure the stability and quality of the X-ray pulses, laser pulse must be synchronized to electron bunch generated by the photocathode RF gun. Here we demonstrate the design and implementation of the Laser-RF Synchronization system in Tsinghua University Thomson Scatter X-ray Source.
This work supported by National Natural Science Foundation of China under Grant No. 10735050 N13-7:

F. Garcia1, M. Kalliokoski1, E. Tuominen1, R. Rudolf Janik2, M. Pikna2, B. Sitar2, P. Strmen2, I. Szarka2 1 Detector Laboratory, Helsinki Institute of Physics and Department of Physcal Sciences, University of Helsinki, Helsinki, Finland 2 FMFI Bratislava, Comenius University, Bratislava, Slovakia

GEM-TPC Prototype for Beam Diagnostics of Super-FRS in NUSTAR Experiment of FAIR Facility

The FAIR [1] facility is an international accelerator centre for research with ion and antiproton beams. It is being built at Darmstadt, Germany as an extension to the current GSI research institute. One major part of the facility will be the Super-FRS [2] separator. The NUSTAR experiments will benefit from the Super-FRS, which will deliver an unprecedented range of radioactive ion beams (RIB). These experiments will use beams of different energies and characteristics in three different branches; the high-energy which utilizes the RIB at relativistic energies 300-1500 MeV/u as created in the production process, the low-energy branch aims to use beams in the range of 0-150 MeV/u whereas the ring branch will cool and store beams in the NESR ring. The main tasks for the Super-FRS beam diagnostics chambers will be for the set up and adjustment of the separator as well as to provide tracking and event-by-event particle identification. The Helsinki Institute of Physics and the Comenius University are in a joint R&D phase of a GEM-TPC detector which could satisfy the requirements of such diagnostics chambers in terms of 39

tracking efficiency, space resolution and count rate capability. The current status of the first prototype and simulations results will be shown.
N13-8:

C. I. Choi1, B. H. Kang1, Y. K. Kim1, I. W. Choi2, D. K. Ko2, J. M. Lee2, G. D. Kim3 1 Department of nuclear engineering, Hanyang University, Seoul, Korea 2 Center for Femto-Atto Science and Technology, and Advanced Photonics Research Institute, Gwangju Institute of Science Technology, Gwangju, Korea 3 Ion Beam Application Group, Korea Institute of Geoscience and Mineral Resources, Daejeon, Korea An ultraintense pulse-type laser having pulse width of subpicoseconds is one of the most interesting scientific topics and has been studied widely. If the ultraintense laser is focused on targets with intensity over 1018 W/cm2, protons having energies of a few to few tens MeV are generated by laser-plasma interaction called Target Normal Sheath Acceleration (TNSA) [1, 2]. Because the energy spectrum of accelerated protons assumes a shape like figure 1, proton radiography images which are obtained by using the ultraintense laser can show better contrast to thin objects than the mono-energy proton ones. Also, the ultraintense laser doesnt require large facilities and high cost in comparison with conventional accelerators. Because of these advantages, the ultraintense laser is expected to replace the conventional accelerators as energetic ion generation source. The experiments comparing the images of the ultraintense laser of Gwangju Institute of Science Technology (GIST) and those of the Tandem Van de Graaff accelerator of Korea Institute of Geoscience And Mineral resources (KIGAM) were performed. The used phantoms to obtain proton radiographic images were designed by using TRIM code, and then were fabricated with Mylar and polyethylene. The obtained images were evaluated by using factors like a spatial resolution, a contrast, and a Modulation Transfer Function (MTF). The spatial resolutions of images from the ultraintense laser were 9.3, 9.9, 10.9 for each phantom. Experiments using the Tandem accelerator in same condition and comparison of the obtained images will be performed.
Acknowledgment : This work was supported by the Ministry of Knowledge and Economy of Korea through the Ultrashort Quantum Beam Facility Program, and also supported by the Basic Atomic Energy Research Institute (BAERI), nuclear R&D program of MEST, Korea. N13-9:

A Study of Proton Radiography Through Comparison Between Ultraintense Laser and Tandem Accelerator

M. Iwasaki1, T. Tauchi2, T. Tsuboyama2 1 University of Tokyo, Bunkyo-ku, Tokyo, Japan 2 KEK, Tsukuba, Ibaraki, Japan

IP Beam Size Measurement During Collisions at Super-KEKB

We report studies of the direct IP beam size measurement during collisions at the future Super-KEKB asymmetric energy e+ecollider experiment by using e+e- pairs. A large number of e+e- pairs are produced during the collisions and deflected in a strong Coulomb potential made by an oncoming beam. Since the potential depends on the vertical and horizontal beam sizes, the deflected pairs are expected to carry the beam size information, especially in their angular distributions. This IP beam size measurement method has been developed for the future e+e- linear collider experiments, where the huge number of e+e- pairs and the strong Coulomb potential are expected because of their high energy beam collisions. We describe the idea of the application to the beam size measurement at Super-KEKB based on its superior high luminosity, and show the first results of the simulation studies.
N13-10:

H. Gong1, L. Hou1, M. Zeng1, B. Shao1, Y. Li2, J. Cai3 1 Dept. of Engeering Physics, Tsinghua University, Beijing, Beijing, China 2 National Synchrotron Radiation Laboratory, University of Science and Technology of China, Hefei, Anhui, China 3 Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai, Shanghai, China A Beam Loss Monitoring (BLM) system has been designed for Shanghai Synchrotron Radiation Facility (SSRF). It is helpful to find out the position of vacuum leakage, regulate the machine operation parameters, and study the beam lifetime. The beam loss mechanism is studied in the paper, followed by beam dynamics analysis which gives the location of lost particles along SSRF storage ring. A detailed beam trajectory tracking with the DBA structure of SSRF storage ring under different energy loss ratios also helps to predict the appropriate positions of beam loss monitoring. The beam loss mechanism points out that when particles hit the vacuum chamber wall a shower process occurs. Among this process, shower electrons carry more distinct position information of lost particles than photos and neutrons do. The PIN photodiode BLM detector originally designed at DESY is then employed in this situation for detecting the shower electrons, which is sensitive to shower electrons, but nearly has no response to photos and neutrons. The detectors are installed along the 432 meter ring, most of them are located at the appropriate positions given by calculation and few are mobile to provide more flexibility. A distributed data acquisition system based on embedded system and synchronous Ethernet technology is also designed to collect the signals from all the detectors and transfer to the data 40

Design and Application of Beam Loss Monitoring System for SSRF Storage Ring

storage and analyze workstation. The BLM System described in this paper has been built and contributed to the commissioning of the SSRF storage ring, and been qualified as a sensitive and steady machine diagnostic and radiation protection tool.
N13-11:

T. Z. Fullem1, A. M. Kovanen1,2, D. J. Gillich1,2, Y. Danon1 1 Dept. of Mechanical, Aerospace and Nuclear Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA 2 Dept. of Physics, United States Military Academy, West Point, NY, USA

Focused Ion Beam Production Using a Pyroelectric Crystal and a Resistive Glass Tube

We describe the use of a pyroelectric crystal and a resistive glass tube to generate a focused ion beam. A sharp tip was mounted on one surface of the crystal which ionized a low pressure deuterium gas and a resistive glass tube was mounted coaxially with the tip. The resultant ion beam produced an illuminated spot on a ZnS screen. It was found that the spot size increased as the distance between the screen and the end of the resistive glass tube increased.
N13-12:

P. S. Yoon1, D. P. Siddons1,2 1 National Synchrotron Light Source-II, Brookhaven National Laboratory, UPTON, NY, USA 2 National Synchrotron Light Source, Brookhaven National Laboratory, UPTON, NY, USA

Development of a Photodiode-Based X-Ray Beam-Position Monitor with High-Spatial Resolution for Use on NSLS-II Beamlines

We have developed a photodiode-based X-ray beam-position monitor with high-spatial resolution for use on the future project beamlines at NSLS-II. A ring array of 32 Si PIN photodiodes were fabricated as a photon sensor, and a low-noise HERMES4 ASIC die was integrated into the data-acquisition system. A series of precision measurements for electrical characterization of the multi-segmented Si-photodiode sensor and the ASIC die demonstrated that the inherent detector noise is sufficiently below tolerance levels. Following up with the efforts of modeling detector performance including geometrical optimization with a Gaussian beam, we built the first-version of prototype detector. As a result of the geometrical optimization via analytical modeling, the next versions of the ring diode are currently under development. To find an X-ray beam centroid using the ring photodiode, we devised a new method of fitting polar coordinates measured from a ring array of 32 segments. In this paper, we present the development of the new front-end monochromatic X-ray BPM and subsequent experimental measurements performed on an existing beamline at NSLS.
N13-13:

A. Ratti1, J.-F. Beche1, J. Byrd1, P. Denes1, L. Doolittle1, P. F. Manfredi1, H. Matis1, M. Monroy1, M. Placidi1, T. Stezelberger1, W. Turner1, H. Yaver1, E. Bravin2, A. Dress3, J. Stiller4, K. Chow1 1 Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 CERN, Geneva, Switzerland 3 Brookhaven National Laboratory, Upton, NY, USA 4 Physikalisches Institut, Ruprecht-Karls-Universitat Heidelberg, Heidelberg, Germany The LHC beam luminosity monitor is a gas ionization chamber that observes showers produced by very forward emitted neutrons at the IPs. The choice of a four quadrant detector could provide information on the beam crossing angle. The system must comply with extremely stringent requirements: resolving the 40 MHz bunch repetition rate of LHC and withstanding extremely high radiation doses (1 GGy/yr). The sensitive volume of the chamber is split into six 1mm gaps to reduce the drift electron clearing time below the 25ns bunch spacing. We chose a mixture of Argon with a few percent (nominally 6%) of Nitrogen. The amplitude of the signal is linearly proportional to the gas pressure and can be chosen up to 10 bar to obtain a desired pulse height. The systems performance was validated with beam in several facilities, such as the ALS, RHIC and the SPS. The 1.5 GeV electron beam from the ALS booster allowed single bunch testing at 1 Hz, with good control of the bunch intensity. In these tests, we were able to characterize the detectors position sensitivity, as well as its response to changes in gas mixture, pressure, or applied bias voltage, validating the models. Testing at an ALS x-ray beamline we could measure the speed of the system using a dedicated 38 MHz bunch pattern. We validated the operation of the device as a luminosity monitor during RHIC run 7. Using the standard RHIC luminosity measurement instrument, the Zero Degree Calorimeter, we placed the detector just between the first and second ZDC modules and found an excellent agreement (better than 99%) when comparing the event rates from the gas ionization chamber and the ZDC during dedicated vernier scans. Finally, the tests at the SPS confirmed the expected signals when measuring the MIPS in showers generated by a 350 GeV proton beam using different absorber thicknesses. This paper describes the final design of the devices as well as the systems modeling and experimental test results.
This work was supported by the Director, Office of Science, Office of High Energy Physics, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231

The Luminosity Monitoring System for the Large Hadron Collider (LHC)

41

N13-14:

P. Branchini1, A. Aloisio2, S. Loffredo1, V. Izzo1, R. Giordano2 1 INFN, Roma, Italy 2 Universita' Federico II, Napoli, Italy

A High-Precision Time-to-Digital Converter in a FPGA Device

The construction and design process of a highresolution time-interval measuring systems implemented in a SRAM-based FPGA device is discussed in this paper. The TDC use a delay line to interpolate the phase within the system clock cycle. A resolution better than 35 ps has been measured generating time intervals with high precision pulse system whose range spans from tens of ns up to 20 sec and whose resolution is better than 1 ps. The architecture implemented has a very small dead time (less than 2 ns) and it is multi-hit. In this paper we show the main characteristics of the board and the performances achieved in terms of stability, resolution integral and differential non linearity.
N13-15:

The GSI Event-Driven TDC with 4 Channels GET4

H. Deppe, H. Flemming Experimte Electronic, GSI Helmholtzzentrum fr Schwerionenforschung GmbH, Darmstadt, Hessen, Germany The GSI Event-driven TDC GET4 is the first prototype of a high resolution low power event driven TDC for the CBM-Time of Flight detector readout. The design specifications according to the CBM-ToF requirements are a very high time resolution of better than 25 ps and a double hit resolution of less than 5 ns. The TDC has to cope with an event rate of up to 50kHz per channel. The time core architecture is based on a Delay Lock Loop followed by a clock driven hit register. For the event handling derandomistion units adapted from FiFo's are implemented and the readout is controlled by a token ring structure. The serialized data transmission could cover different event types like data, error or synchronisation events. The ASIC was submitted in October 2008 and is presently under test at GSI Darmstadt.
N13-16:

An FPGA TDC for Time-of-Flight Applications

J. Wu Fermilab, Batavia, IL, USA An 18-channel time-of-flight (TOF) grade time-to-digit converter (TDC) has been implemented in a low cost FPGA device. The TDC has the following unique features. (1) The time recording structures of the TDC is based on the wave union TDC we developed in our previous work. A leading edge of the input hit launches a bit pattern, or wave union into the delay chain-register array structure which yields two usable measurements. The two measurements effectively sub-divide timing bins for each other especially the ultra-wide bins caused by the FPGA logic array block (LAB) structure and improves measurement precision both in terms of maximum bin width and RMS resolution. A coarser measurement on input signal trailing edge is also provided for time-over-threshold (TOT) applications. (2) The TDC supports advanced timing reference distribution schemes that are superior to conventional common start/stop schemes. The TDC has 16 regular measurement channels plus two channels for timing reference. The timing reference is established with multiple measurements rather than single shot common start/stop. An advanced scheme, the mean-timing approach even eliminates needs of high quality timing distribution media. (3) The ASIC-like encapsulation of the FPGA TDC significantly shorten the learning curve for potential users while maintain certain flexibility for various applications. Necessary digital post-processing functions including semi-continuous automatic calibration, data buffer, data link jam prevention logic etc. are integrated into the firmware to provide a turn-key solution for users.
N13-17:

WaveDREAM - a DRS4 Based 5 GS/s 12 Bit Digitizer with GBit Ethernet Readout

S. Ritt, R. Dinapoli, U. Hartmann TEM, Paul Scherrer Institute, Villigen, Switzerland A new waveform digitizing board has been designed at PSI, Switzerland. It is based on the patented DRS4 chip, which is capable to digitize eight inputs with 5 GS/s and 12 bits of resolution with only 350 mW of power. The small size of this board (50x100mm2) make it ideally suited to be mounted directly on a detector, while the read out is done through Gigabit Ethernet. A clock distribution system together with the PLL inside the DRS4 chip achieve a system-wide timing precision below 10 ps RMS. Application specific firmware allows to implement TDCs, CFDs, various ADCs and sophisticated triggers in software on this board. The high channel density, precise amplitude and time digitization together with the fast readout speed make this board suitable for a wide range of applications in various particle detectors, including gamma-ray astronomy, TOF systems and PET scanners.

42

N13-18:

F. Petulla'1,2, F. de Notaristefani1,2, V. Orsolini Cencelli1,2, E. D'Abramo1,2, A. Fabbri1,2, M. Marinelli3,4, G. Verona Rinati3,4 1 Deparment of Electronic Engineering, University of Rome Roma Tre, Rome, Italy 2 INFN Sez. Roma III, University of Rome Roma Tre, Rome, Italy 3 Deparment of Mechanical Engineering, University of Rome Tor Vergata, Rome, Italy 4 INFN Sez. Roma II, University of Rome Tor Vergata, Rome, Italy Monocrystalline diamond, grown with CVD techniques, is a very good base material for clinical dosimetry sensors. Diamond also shows good physical and electronic properties in term of tissue equivalence and very low dark currents. The sensor current readout is made by electrometers, these are expensive and bulky and, due to the need of shielded cables, can hardly be used for matrix sensor readout. In order to overcome these problems an ASIC, directly bonded to the sensor, could be the most suitable electronic readout. The ASIC shuld be capable of integrating currents smaller than the pA, with a serial intreface to minimize inteconnection cable volume, providing a compact and portable device. This paper describes the main functional ASIC blocks we designed. The ASIC has nine channels that will be able to operate in the clinical dosimetry current range, and one channel dedicated to dark currents integration. We present simulation results for every functional block and the nal chip layout.
N13-19:

Interleaved Dual Slope ADC for a Diamond Dosimeter ASIC

A Peak Detect and Hold Circuit Using Ramp Sampling Approach

J. R. Lin, H.-P. Chou Department of Engineering and System Science, National Tsing Hua University, Hsinchu, Taiwan The paper presents a peak detect and hold circuit for nuclear radiation pulse amplitude measurement. Two ramps with different slopes are used to sample the input voltage. The two sampled voltage levels are compared to determine whether the waveform is rising or falling and to detect the peak value. The comparator delays and offsets are also cancelled during the comparison. The circuit is designed using 0.18m 1.8V CMOS process. The error of the measured peak value is within 1mV measuring Gaussian pulses with amplitude 1 V and the peaking time from 1 to 12 s.
The work is under the auspices of National Science council, Taiwan. N13-20:

A Fast Single Slope ADC with Vernier Delay Line Techniques

W. F. Lin, H. P. Chou Department of Engineering and System Science, National Tsing Hua University, Hsinchu, Taiwan The paper presents a high precision fast integrating ADC with 0.18m CMOS technology. The ADC consists of a one-bit flash ADC, an amplitude-to-time converter, and a two-level vernier delay line unit for 8- bit timing measurement. The one-bit flash ADC divides the input range into half and can reduce the nonlinearity of the ramp signal. Three ramp-down generators are used for amplitude to time conversion. Thus, the timing errors associated with comparator delays are cancelled by pairs of timing marks generated at the same voltage level and a precise time interval can be obtained. The two-level vernier delay line unit has 16 delay stages in each level and the time resolution is about 50 ps and the conversion time is less than 0.1 s.
The work is under the auspices of National Science council, Taiwan. N13-21:

O. Gevin1, F. Lugiez1, E. Delagnes1, O. Limousin2, A. Meuris2 1 IRFU/SEDI, CEA, Saclay, France 2 IRFU/SAP, CEA, Saclay, France

IDEF-X SX0: a LOW POWER CMOS ASIC for the READOUT of CD(ZN)TE DETECTORS

Since few years, our group is developing a family of ASICs for space applications, named IDeF-X for Imaging Detector Frontend. IDeF-X SX0 is the very last member of the family. It has been designed in the standard AMS CMOS 0.35m process technology. IDeF-X SX0 is a 32 channel analog front-end with self-triggering capability optimized for the readout of 16x16 pixels CdTe or CdZnTe pixellated detectors to build a new low power micro Gamma camera. The architecture of the analog channel includes a Charge Sensitive Amplifier with a continuous reset system, a variable gain stage, a Pole-Zero cancellation stage, a variable shaping time second order low pass filter, a peak detector, and a discriminator. The energy thresholds can be individually settled in each channel thanks to an in channel 6 bit low power DAC. The channel has been optimized to reduce the power consumption which is now 600W per channel; it is five times lower than the power consumption of the previous IDeF-X chip. Moreover, the dynamic range of the ASIC can now be extended to more than 1MeV thanks to the in channel variable gain stage. When no detector is connected to the chip and no leakage current is programmed, the lowest ENC is achieved at a 10.7 s peak time: the noise is 35 electrons rms. In the paper we will detail the performances of the chip including the linearity and the

43

dispersion of characteristics between channels. Finally, the circuit will be evaluated by spectroscopy measurements performed with Cd(Zn)Te detectors.
N13-22:

A High-Speed 2nV/Hz1/2 16-Channel Current Amplifier IC for PET

J.-P. Rostaing, A. Peizerat, O. Billoint, G. Montemont, O. Monnet DCIS / DTBS, CEA, LETI, MINATEC, Grenoble, France A 16 channel front-end IC dedicated to small animal Positron Emission Tomography has been developed for crossstrip CdTe detectors. Each channel, designed to handle up to 20pF detector capacitance, includes both low voltage (2nV/Hz1/2) and low current (40fA/Hz1/2) noise, high bandwidth (50MHz at 20pF detector capacitance) current amplifier with a gain of 100 for a power consumption of 4.7mW. Both anode and cathode connection to the detector are possible thanks to an external control signal selecting the input common polarity. Input has been designed for maximum peak current of +/-1uA. Additional charge amplifier functionality is available via a single external control pin, extending functionality to X-ray or Gamma-ray detection, fast instrumentation. Circuit size is 2x8 mm2 in 0.35um 3.3V CMOS process. Each channel differential output is buffered by a transconductance amplifier with two balanced differential current outputs compliant with 50 Ohm terminations.
This work has been supported by French National Research Agency (ANR) through TecSan program (project Topase-Med n ANR-06-TECS009). The authors would like to thank O. Rossetto et J.P. Richer from LPSC, Grenoble, and all the people who have contributed to the realization of this ASIC. N13-23:

CLOSY: a Very Precise Clock Generation for Timing Measurements and Synchronization of the CBM ToF Wall

K. Koch Experiment Electronics Dept., GSI Helmholtzzentrum fuer Schwerionenforschung GmbH, Darmstadt, Germany In the concept of the CBM-Time of Flight (ToF) detector readout, two phase coupled high-performance frequencies with very low jitter are needed. One is directly used for time measurements with an event driven TDC (GET4 ASIC), the other for synchronous data transportation reasons. Due to that, a new electronics for precise clock generation and distribution has been designed. The main card (CBM-CLOCK-SYTEM) is based on a frequency synthesizer chip to create two phase coupled independent output frequencies with very low jitter ( < 5ps, distributed over a 20 m distance) and an additionally downstream fast CPLD that creates a synchronisation signal, which is needed as a periodically epoch marker for the system.
N13-24:

O. Rossetto1, J.-P. Rostaing2, J.-P. Richer1, O. Billoint2, J. Bouvier1, O. Monnet2, A. Peizerat2, G. Montmont2 1 Microelectronic, LPSC CNRS/IN2P3-Grenoble universite, Grenoble, France 2 DTBS, CEA-LETI, Grenoble, France In this paper, we present two circuits designed for pulse readout of a semiconductor PET system: a fast low noise low power front-end preamplifier/shaper, and the processing circuit performing time tagging, energy measurement and digital interfacing with the data acquisition system. Considerations on noise, speed and power consumption are discussed. The system level electronics architecture and its optimization according to the detector architecture is presented. Results of the circuits caracteristics are also presented.
N13-25:

INTEGRATED ELECTRONIC for a CdTe BASED PET SYSTEM

A. Abba1, A. Manenti1, A. Suardi1, S. Riboldi2, A. Geraci1 1 Dept. of Electronics, Politecnico di Milano, Milan, Italy 2 Dept. of Physics, Universita' degli Studi di Milano, Milan, Italy

High Performance Analog Front-End for Digital Spectroscopy

In any digital spectroscopy system the signal has to be properly conditioned before sampling by means of analog circuitry. The paper presents an analog fully-differential front-end, with digitally adjustable gain and offset, wide dynamic amplitude of input signals, low noise and linearity error less than 100 ppm. The analog section has been designed bearing in mind also multichannel applications and, consequently, taking into particular account power dissipation, physical size of hosting PCB, and allowing the system to operate in presence of electromagnetic noise, while at the same time minimizing its own radiated emissions.

44

N13-26:

M. J. Myjak1, D. Ma2, D. J. Robinson2, G. S. La Rue2, D. R. Hanlen1 1 Pacific Northwest National Laboratory, Richland, WA, USA 2 Washington State University, Pullman, WA, USA

Multi-Channel Data Acquisition System for Nuclear Pulse Processing

We are developing a compact, low-cost electronics platform for acquiring pulse-mode data from multiple radiation detectors. The system can process up to 64 input channels in parallel. We attempted to keep the design as simple as possible, both to reduce power consumption and allow the system to handle a large amount of parallel processing. A front-end ASIC contains a charge integrating amplifier, two comparators, and a switchable current driver for each channel. This circuit implements the first stage of a Wilkinson ADC, in which the integrated charge is converted to a digital pulse whose width is proportional to the charge amplitude. A back-end FPGA digitizes the pulse and controls the switchable current source. The FPGA also implements a novel technique to measure and compensate for leakage current in the sensor head. This paper presents the design of the multi-channel data acquisition system and discusses the performance of the electronics, including the noise, linearity, and count rate throughput.
N13-27:

Digital Readout Electronics for Microcalorimeters

H. Tan, J. Collins, W. Hennig, M. Walby, P. Grudberg, W. K. Warburton XIA LLC, Hayward, CA, USA Microcalorimeters are cryogenic radiation detectors that measure the energy of incident photons or particles by the increase of temperature in an absorber. They are capable of achieving ultra-high energy resolution, but only with small active volumes and pulse decay times in the order of milliseconds. Consequently, detection efficiency per detector is low. A practical detector thus requires a large array of microcalorimeters, and either time-domain or frequency domain multiplexing is used to minimize the number of leads exiting the cryostat to room temperature signal processing. Existing readout electronics generally stream the multiplexed data onto a computer hard drive for offline processing. XIA LLC is developing digital microcalorimeter readout electronics that offer real time signal processing. The electronics consist of two major parts: a set of daughter cards and a standardized core processor board. Each daughter card is designed to interface a specific type of microcalorimeter (single outputs, time multiplexed, frequency multiplexed, etc.) to the core processor. The combination of the standardized core plus a set of easily designed and modified daughter cards allows the design to not only meet present detector signal processing requirements but also ensure future expandability to as yet unspecified detector systems. We will first present the general architecture of the daughter cards and the core processor board. Then we will describe the core processor's interleaved multiplexing firmware that is capable of real time digital signal processing for generating either list mode event data or energy histograms. Analysis of a specific daughter card will clarify their design and function. Finally we will discuss the preliminary performance of our digital signal processing algorithms on real microcalorimeter pulses.
This research is supported by the U.S. Department of Energy under grant DE-FG02-07ER84760. N13-28:

The Readout Electronics of GEM Detector

Y. Zhao Div.of experimental physics center, Institute of High Energy Physics, Chinese Academy of Sciences, beijing, China The CSNS was under construction sine 2008. One of the spectrometers requires a neutron detector with spatial resolution of better than 0.2 mm in both X and Y directions. In order to meet the neutron detector requirement of CSNS, the GEM, a twodimensional position sensitive detector, was constructed. A prototype of the detector had been constructed at first, using cathode strip readout method to determine the avalanche position. A readout electronics system was also developed. The prototype of readout electronics system has been designed based on VME bus. The system makes up of three parts: Preamplifier Board, Control Logic Generator Board and Data Acquiring and Transferring Board. The Preamplifier Board is designed with charge sensitive amplifier. On the Data Acquiring and Transferring Board, we get the charge value by peak seeking, after AD conversion by the 40Msys 10 bit flash ADC. Both the peak seeking method and the interface with VME bus are achieved by the firmware in one FPGA. The electronics system has 2 work modes: online mode and calibration mode. Under the calibration mode, we got the charge measurement resolution about 0.4 fc. Under the online mode, we got the spatial resolution for the detector about 0.11 mm. The readout electronics system can work normally, and owned a pretty good performance in neutron discrimination
N13-29:

Development of a Counting Strip Detector Readout Chip for Precision Compton Polarimetry

M. Karagounis, G. Ahluwalia, M. Gronewald, M. Koch, H. Krueger, N. Wermes Physics Department, University of Bonn, Bonn, Germany A strip detector readout chip used in an individual photon counting detection system for precision Compton polarimetry has been chosen as a platform for the implementation, comparison and the insilicon test of several alternative architectural analog and 45

digital circuit concepts. An analytic result for the noise optimization has been derived based on the EKV model which gives the optimum width of the input transistor of the charge sensitive amplifier as a function of the inversion factor. Apart from the input transistor, the noise contribution from the biasing for a folded and a telescopic preamplifier topology has been also analyzed. The limitations of classic shaping circuits using MOS transistors operating in linear region as feedback elements are emphasized and alternative shaping topologies rest upon linearized transconductance amplifier and MOSFET-C filter are introduced. A potential hazard of synchronous counters reacting on very short clock pulses is described which becomes much less severe when the implementation is changed to asynchronous ripple counters. All digital circuitry has been implemented in Differential Current Logic to reduce crosstalk between the digital and the analog circuitry. The Trade-Offs which are related to the presented architectural options are extracted from measurements and simulations.
N13-30:

K. M. C. Koua1, J.-F. Pratte1, A.-A. I. Assane1, N. Viscogliosi1, C. Pepin2, R. Lecomte2, R. Fontaine1 1 Dept. of Electrical and Computer Engineering, Universite de Sherbrooke, Sherbrooke, QC, Canada 2 Dept. of Nuclear Medicine and Radiobiology, Universite de Sherbrooke, Sherbrooke, QC, Canada

Design and Performance of a 0.18-m CMOS Charge Sensitive Preamplifier for the LabPET II, a Novel 64-Channel APD-Based Detector for PET/CT

A novel 64-channel APD-based detector module has been developed for the next generation small animal PET/CT scanners achieving submillimetric spatial resolution. We report on the design and performance of a low-power, low-noise, 0.18-m CMOS Charge Sensitive Preamplifier (CSP) optimised for this new 64-channel APD-based detector. Timing resolution remains one of the most challenging issues with APD-based detectors, even with recent fast scintillators. Hence, fast, low noise performance is required from the CSP to achieve very good timing resolution. The effects of APD gain on Equivalent Noise Charge (ENC) and optimum shaping time was evaluated for a unipolar 3rd order semi-Gaussian filter. For an APD gain of 200 and a shaping time of 30 ns, an ENC of 450 e- rms was calculated using the EKV model, assuming realistic values of the APD capacitance and current noise spectral density. Also, given the high pixel density and large total number of channels, the power budget must be limited to 1 mW per CSP. The CSP consists of a modified telescopic cascode architecture with an NMOS input device. Electronic characterisation of the rise time, linearity, dynamic range, ENC, timing resolution and energy resolution, as well as the effects of APD gain on the performance are reported.
N13-31:

Noise Optimization of CMOS CSA in Weak and Moderate Inversion Regions

Y. Li, Z. Deng, Y. Liu Department of Engineering Physics, Tsinghua University, Beijing, China Noise optimization of CSA(charge sensitive amplifier) in scaled CMOS process has been well studied for the input transistor working in strong inversion region. However, it tends to work in weak or moderate inversion region in many applications requiring low power consumption and/or using deep sub-micron processes. 1/f noise coefficient and the gate capacitance are not constant and become more dependent to the bias conditions, which makes noise optimization more complicated. A CSA is designed and tested using 0.35 m CMOS technology. ENC at different bias currents are measured. At moderate inversion region, the maximum current does not yield the minimum ENC. As the bias current increases, the contribution of the series 1/f noise increases and the contribution of the series white noise is non-monotonic. A more accurate noise optimization method for weak and moderate inversion regions will be derivated, considering the bias dependence of the 1/f noise coefficient and gate capacitance.
N13-32:

A. Pullia1,2, F. Zocca2 1 Dept. of Physics, University of Milan, Milan, Italy 2 INFN, Milan, Italy

Fast Low-Impedance Output Stage for CMOS Charge Preamplifiers Able to Work at Cryogenic Temperatures

The quality of the output stage is key to the performance of low-noise CMOS preamplifiers of semiconductor detector signals, especially when the preamplifier signal is to be transmitted to a remote receiver. This work deals with the design and optimization of an output-stage circuit architecture particularly suited for low-noise integrated preamplifiers of X and gamma-ray detector signals. The aim of this development is to conjugate a few important features: low output impedance, ability to work at room and cryogenic temperatures, ability to drive a terminated coaxial cable, low power consumption, large voltage swing with a 100 ohm load. In particular we were interested to a large negative voltage swing. The standard solution, i.e. a source-follower stage realized with a p MOSFET, is not adequate because of the so-called body effect, which would severely limit the negative output voltage swing. The proposed circuit structure, inspired by the White follower, builds around a first n-MOSFET configured as source follower, a second n-MOSFET acting as driver for the load current and a negative-feedback loop which stabilizes the working current of the first MOSFET. As a result both a low output impedance and a large negative voltage swing are obtained. We realized the output stage protoype in a 5 V 0.8 m CMOS technology, and obtained a voltage range of -2.5 V with a negative 46

power supply of -3 V and with a 100 ohm load. Using the output stage in a custom JFET-CMOS charge-sensitive preamplifier for germanium detectors we obtained a large negative voltage swing of -2.5 V, a signal rise time of ~13 ns with a resistive load of 100 ohm and a detector capacitance of 16 pF, and a negligible additional noise. The design of a fully symmetric version, able to provide negative as well as positive rail-to-rail signals, will also be discussed.
N13-33:

A. Fabbri1,2, V. Orsolini Cencelli1,2, F. de Notaristefani1,2, E. D'Abramo1,2, F. Petulla'1,2, R. Pani3,2, G. Moschini4,2, F. Navarria5,2 1 Departmente of Electronic Engineering,, University of Roma Tre, Roma, Italy 2 INFN - Instituto Nazionale di Fisica Nucleare, Roma, Italy 3 Department of Experimental Medicine and Pathology, University of La Sapienza, Roma, Italy 4 Department of Physics, University of Bologna, Bologna, Italy 5 Department of Physics, Univeristy of Padova, Padova, Italy The gamma cameras built on a LaBr3 crystal and Hamamatsu H8500 that are being developed have a number of channel that can vary from 64 to 256 depending on the number of PSPMTs that are used and, due to the gain differences, channels have to be acquired and corrected individually. In this context, the readout electronics has to be able to acquire and possibly to make computations on a high number of channels at a rate of kiloevents. A possible solution is the use of an FPGA programmed to collect the data and, in parallel, make a preliminary analysis. The high number of pins available on a modern FPGA allows to acquire and to manage all the data coming from the ADCs, driving the control signals and acquiring the data from several device at the same time. The possibility, even on average grade FPGAs, of having clock rate in the 100 MHz range, makes it feasible to make preliminary energy discrimination in order not to overload the communication channel with the control PC that, due to the high number of data, can became the real bottleneck of system.
N13-34:

FPGA Based Readout Electronics for Multi Anode PSPMT

C. P. Lambropoulos1, E. G. Zervakis2, A. Nikologiannis2, G. Theodoratos2, D. S. Hatzistratis2 1 Sensors Electronics and Communications Lab and Aircraft Technology Dept., Technological Educational Institute of Chalkida, Psahna-Evia, Greece 2 Sensors Electronics and Communications Lab, Technological Educational Institute of Chalkida, Psahna-Evia, Greece The Photon 4dimensional Digital Information (P4DI) ASIC is a pixel readout integrated circuit to be used in hybrid imagers. It measures the charge induced on the detectors pixel electrodes from the impinging photons irrespective of the dose rate of the radiation field while the time tagging of each photon makes it appropriate for Compton scattering imaging applications. In pixel digitization and storage of the time and amplitude signal are performed. Only pixels with recorded hits are read out. It has been designed in UMC 0.18um CMOS technology and consists of 8x8 pixels matrix and control logic. The chip has been fabricated and bump bonded to a CdZnTe pixel detector. The ASIC architecture and measurement results will be presented.
N13-35:

The P4DI ASIC Architecture and Measurement Results

A. Wassatsch1, S. Herrmann2, R. H. Richter1, L. Andricek1 1 Halbleiterlabor, Max-Planck-Institut fr Physik, Munich, Germany 2 Halbleiterlabor, Max-Planck-Institut fr extraterestische Physik, Munich, Germany

A Clustering Engine for Data Rate Reduction in the Belle II Pixel Detector

The DEPFET technology has been chosen as the baseline for the vertex pixel detector of the Belle II experiment at KEK/Japan. This detector consist of 22 modules, each proposed to be equipped with 2x250x1k pixel. With the proposed average readout frequency of 10kHz and 30kHz peak rate a huge data stream of 100Gbit/s will be generated. To reduce this data stream to a more feasible value, we have to take advantage of the low occupancy of approx. 1-2% and the the hit distribution within the pixel array. For this purpose a pipe-lined clustering algorithm was developed, which can handle cluster sizes of 3x3 or 5x5 at full readout speed. These algorithm can handle data from non bricked sensors and also from bricked designs. It is able to reduce the data rate by up to 50%. The collection of the clustered data packages from the row wise parallel pipe-line are controlled by a specially developed output scheduler, which assures together with the corresponding FIFO structures the data-handling capability. To prove the concept and the performance of the algorithms and also to decouple the development of DHP (DataHandling-Processor) unit from the clustering unit, the clustering engine is in the first phase of the project implemented within a separated ASIC. In a second phase both units will be merged into a final data processing unit, which has to fit on the balcony of the thinned DEPFET detector module. In this paper we will present the overall structure of this DCE (DHP Clustering Engine) and also the details of the implemented clustering and output scheduling algorithms. Furthermore the impacts of the constraints on speed, power and area on the developed circuit and the chosen realization given by the experiment environment will be discussed in detail.

47

N13-36:

Current Mode Constant Fraction Discriminator for PET Using SiPM(MPPC)

W. Shen, H.-C. Schultz-Coulon Kirchhoff Institute for Physics, University of Heidelberg, Heidelberg, Germany Silicon photomultipliers (SiPMs) are a novel type of solid state photon detector,which have similar internal gain factors as Photomultiplier Tubes (PMT). Due to their low operation bias voltage, magnetic field immunity and small size, this new silicon photon detector can be used in a wide range of applications, such as Positron Emission Tomography (PET). The special features of SiPM such as large pixel number and high gain lead to a large current output in detection of 511keV photon in the PET application. Hence, a novel current-mode monolithic readout scheme can be implemented. As an important time pick-up unit, constant fraction discriminators are generally used in PET for coincidence measurement. Here, we report on a current mode constant fraction discriminator using a current mode non-delay-line method to generate the bipolar shape for the zero-crossing timing. One analog channel using this scheme has been designed and simulated in AMS 0.35m CMOS technology. Both the design and simulation results will be presented.
N13-37:

F. Zocca1, A. Pullia1,2, S. Riboldi1,2, C. Cattadori3,4, A. D'Andragora4 1 INFN-Milano, Milano, Italy 2 Dept. of Physics, University of Milano, Milano, Italy 3 INFN-Milano Bicocca, Milano, Italy 4 INFN-LNGS, L'Aquila, Italy

Setup of Cryogenic Front-End Electronic Systems for Germanium Detectors Read-Out

Front-end electronic devices for the read-out of ionizing radiation sensors must operate in many cases at cryogenic temperatures. Sometimes the front-end circuit is divided into a cold part operated at cryogenic temperature and a warm part operated at room temperature outside the cryostat. In other cases this is not possible owing to the physical constraints coming from the experimental setup, the detector system requirements or the apparatus geometry. In this latter cases the front-end circuitry has to operate in its entirety at cryogenic temperature. In this work we focus in particular on front-end read-out systems for High-Purity Germanium (HPGe) detectors, which are usually operated at liquid nitrogen (LN) temperature. We study the strong effects that the changed characteristics of the electronic active and passive devices have on the charge preamplifier design when operated in LN, while taking into account the particularly challenging requirements that the circuit has to meet: radio-purity, physical reliability under thermal cycling, good noise performance (0.1-0.2% resolutions), fast rise time (20 ns) needed for pulse shape analysis applications. The developed circuit consists of an external silicon JFET (Junction Field Effect Transistor), an external feedback network, and an ASIC (Application Specific Integrated Circuit) realized in a 5V 0.8m silicon CMOS technology. We discuss in particular the effects that changes of JFET and MOSFET transconductance have on noise and bandwidth performance of the circuit. We also discuss the effects that a changed performance of passive devices, such as high-value filtering capacitances, may have on the preamplifier response when operated in LN.
N13-38:

Low Noise 64-Channel ASIC for Si, GaAs and CdTe Strip Detectors

M. Kachel, P. Grybo, R. Szczygie Department of Measurement and Instrumentation, AGH University of Science and Technology, Cracow, Poland We present a 64-channel low noise ASIC called SXDR64 for low noise digital X-ray imaging systems. The ASIC is aimed to work with DC coupled CdTe and GaAs strip detectors as well as with Si strip detectors. The circuit can operate with input leakage currents in the range from -5 nA up to 10 nA and with the frequency of input pulses up to 1 MHz. The IC control and readout of the data are done via LVDS drivers/receivers, to ensure the possibility of circuit operation in continuous readout mode. Measured gain of the ASIC is 40 V/el and noise is ENC = 133 el. rms, with 1 cm long silicon strip detector with 100 m pitch.
N13-39:

W. Dabrowski1, F. Anghinolfi2, N. Dressnandt3, M. Dwuznik1, J. Kaplon2, D. La Marra4, M. Newcomer3, S. Pernecker4, K. Poltorak1, S. G. Sevilla4, K. Swientek1 1 Faculty of Physics and Applied Computer Science, AGH University of Science and Technology, Krakow, Poland 2 CERN, Geneva, Switzerland 3 Physics Department, University of Pennsylvania, Pennsylvania, USA 4 University of Geneva, Geneva, Switzerland

Design and Performance of the ABCN-25 Readout Chip for the ATLAS Inner Detector Upgrade

A primary challenge of tracking detectors being developed for the SLHC environment is a high occupancy, which affects directly granularity of the detectors and the number of electronic channels, to be about 5- to 10 times higher compared to the present Semiconductor Tracker (SCT) in the Atlas Experiment. As a result, power consumption in the readout chip is one of the most critical issues on top of usual requirements concerning noise and timing parameters, and radiation resistance. These requirements 48

have to be considered taking into account present and expected trends in development of industrial CMOS processes. We present the design and performance of the ABCN-25 readout chip implemented 0.25 m CMOS technology. The front-end design has been optimized for the short, 2.5 cm, silicon strips foreseen in the upgrade of the ATLAS Inner Detector. The core of the readout architecture includes binary front-end, two levels of data buffering, data compression and data serializing circuitry, and is similar to the architecture of the ABCD3T chip used in the present ATLAS SCT detector. In order to ensure required radiation hardness the hardening by layout technique has been used and SEU detection and correction circuitry have been added. The design includes on-chip power management circuitry comprising two types of shunt regulators and a serial regulator. This circuitry makes the ABCN-25 chip compatible with recent developments in the area of power distribution systems for the inner trackers in the S-LHC environment and in particular with serial powering of the detector modules. The chip has been fabricated in 0.25 m CMOS technology and full functionality has been obtained. The design and performance of the analog and digital circuits will be presented and discussed.
N13-41:

J. Kaplon1, A. Ceccucci1, P. Jarron1, A. Kluge1, M. Noy1, S. Tiuraniemi1, M. E. Martin Albarran2, F. Marchetto3, G. Mazza3, A. Rivetti3, S. Martoiu3, G. Dellacasa3 1 CERN, Geneva, Switzerland 2 Universit catholique de Louvain, Louvain-la-Neuve, Louvain-la-Neuve, Belgium 3 INFN, Torino, Italy We present the design and test results of a front-end prototype circuit developed in 130 nm CMOS technology for the readout of the Gigatracker pixel detector in NA62 experiment at CERN. The conditions of the experiment set very demanding requirements on the front end electronics in terms of speed, noise and power consumption. The main challenges for the front end amplifier are very high signal hit rate (dead time less than 100 ns, average signal rate 100 kHz) and 100 ps timing resolution combined with the level of affordable power consumption (<2W/cm2) and noise (<200e- ENC). The 5 ns peaking time prototype transimpedance amplifier has been submitted in IBM 130 nm process and expected back from the foundry in June. The predicted ENC levels for the nominal detector capacitance of 250 fF and maximum leakage current of the order of 20 nA are below 200 e-. The overall power consumed by the analogue and digital part of the pixel cell is in the order of 130 W. The optimization of the design as well as test results of the prototype front end chip are evaluated and discussed.
N13-42:

The 5ns Peaking Time Transimpedance Front End Amplifier for the Silicon Pixel Detector in the NA62 Experiment

C. Arnaboldi1, X. Liu2, G. Pessina1 1 INFN Milano Bicocca e Universita degli studi di Milano Bicocca, Milan, Italy 2 Department of Physics and Astronomy, University of California and Los Angeles, Los Angeles, CA, Usa We present the room operating preamplifier for the readout of the 988 large mass bolometers of the CUORE experiment. Each bolometer is based on a 750 g TeO2 crystal on which a Nuclear Transmutation Doped Ge thermistor, NTD Ge, is glued, in good thermal contact. To read out the detectors we have developed a differential voltage sensitive preamplifier based on a well selected and properly biased input J-FET and by a circuit solution based on operation amplifiers. The developed preamplifier has a low parallel noise 0.094 fA/Hz and a low voltage noise of 4.3 nV/Hz @ 1Hz and white 2.6 nV/Hz. For the DC readout of bolometric detectors, the presented preamplifier has also the circuit for the compensation of the detector bias and the offset of the input J-FET pair. A compensation circuit allows to minimize the voltage thermal drift. The common mode rejection ratio is digitally adjustable. DC coupling, low noise, small thermal drift and CMRR are stringent requirements of the CUORE experiment. We will describe in detail the circuit solutions and the selection criteria for devices used.
N13-43: Spectroscopic Performances of the GERDA Cryogenic Charge Sensitive Amplifier Based on JFETCMOS ASIC Coupled to Germanium Detectors

The Preamplifier for CUORE, an Array of Large Mass Bolometers

C. Cattadori1,2, A. D'Andragora2,3, A. di Vacri2, L. Pandola2, C. Ur4, A. Pullia5,6, S. Riboldi5,6, F. Zocca5,6 1 INFN - Milano Bicocca, Milano, Italy 2 INFN - LNGS, Assergi (AQ), Italy 3 Electrical and Information Engineering, Universita' degli Studi dell'Aquila, L'Aquila, Italy 4 INFN - Padova, Padova, Italy 5 Physics, Universita' degli Studi di Milano, Milano, Italy 6 INFN - Milano, Milano, Italy

In the GERDA (Germanium Detector Array) experiment, aiming to search for neutrinoless double beta decay, the germanium detectors organized in 3-fold strings, will be operated naked suspended in the center of a 70m3 cryostat filled with liquid Argon (LAr). In this application the use of cryogenic front-end electronics is mandatory. We tested two versions of Charge Sensitive Preamplifiers (CSA) circuits, namely a 1-channel (1-ch) and 3-channels (3-ch) based JFET-ASIC; the latter is produced in 5V 49

0.8m silicon CMOS technology. The 3-chs CSA are designed to serve the detectors of one GERDA detector string. While in the reference test the 1-ch circuit and the custom-encapsulated Ge detector (SUB) were operated both submerged in liquid nitrogen, in the naked detector test both the 1-ch circuit and the naked unsegmented Ge detector were both submerged in LAr. A resolution of 3.3 keV at 1332 keV 60Co line (2.7 keV at pulser) at 6 s shaping time has been obtained in the latter configuration to be compared to 2.2 keV (1.6 keV at pulser) obtained in the reference test. In the work, the discussion of the noise origin is presented, and some filtering techniques are implemented to improve the resolution. The 3-ch circuit, based on 3 JFETs connected to 3chs of the CMOS ASIC circuit, mounted on a Cuflon PCB has been tested (1-ch test) both coupled to the reference SUB detector and to two naked prototype detectors (two channels test) to address the cross-talk issue. The spectroscopic performances have been measured connecting the CSA output both to a spectroscopy amplifier and to a sampling ADC, through 12 m long cryogenic cables to simulate the real front-end to FADC connection in the GERDA environment. The spectroscopic performances obtained with the analog and digital pulse processing techniques are presented and discussed.
N13-44:

M. Ciobanu1, N. Herrmann2, K. D. Hildenbrand1, M. Kis1,3, A. Schuttauf1 1 GSI-Darmstadt, Darmstadt, Germany 2 Physikalisches Institut der Universitt Heidelberg, Heidelberg, Germany 3 Rudjer Boskovic Institut, Zagreb, Croatia

PADI-2,-3 and -4: The Second Iteration of the Fast Preamplifier Discriminator ASIC for Time-ofFlight Measurements at CBM

Within the European project JRA-12, we have designed a general-purpose PreAmplifier-DIscriminator (PADI) ASIC to be used as Front-End-Electronics (FEE) for the readout of timing Resistive-Plates Chambers (RPC) in the CBM experiment at FAIR. These fast timing detectors provide signal rise times tR <500 ps and generate a prompt charge in the range of 50 to 2000 fC [1]. Intrinsic time resolutions of tD <50 ps have been obtained with RPCs, which requires a preamplifier-discriminator stage with an intrinsic electronic resolution of tE <15 ps [2]. The first prototype developed was PADI-1, a three-channel ASIC in 0.18 m CMOS technology. From in beam tests of PADI-1 [3] we have understood that the minimization of the crosstalk between channels is of priority for further developments. In this paper we present the improved designs in which we have changed the biasing from voltage biasing to current biasing. The increase of crosstalk rejection ratio at the chip level is 20dB minimum. The PADI chip will be connected to a TDC chip which is currently under development (GET-4) [4]; it accepts LVDS signals as input. We have designed the two variants PADI-2 and PADI-3 which differ in the Time output levels: The former has +/- 100mV (fixed value, with the common mode DC voltage fixed at the optimum point of the LVDS receiver, e.g., 2/3*VDD), the latter +/-350mV (programmable by an external resistor) on 100 differential load. We have increased the number of channels to four and we have added the OR feature which allows to daisy-chain chips for trigger purposes. PADI-4 has been developed to be used with diamond detectors for timing. In this article, we describe these new prototypes PADI-2,-3 and -4 by pointing out their design characteristics; we present the first measurements concerning gain, linearity, bandwidth, noise, and time-over-threshold (ToT) behavior, as well as timing performance, crosstalk, common-mode-rejection and input impedance.
N13-45:

M. Ciobanu1, E. Berdermann1, N. Herrmann2, K. D. Hildenbrand1, M. Kis1,3, W. Konig1, M. Pomorski1, A. Schuttauf1 1 GSI-Darmstadt, Darmstadt, Germany 2 Physikalisches Institut der Universitt Heidelberg, Heidelberg, Germany 3 Rudjer Boskovic Institut, Zagreb, Croatia In this paper we describe the operation principles and the in-beam performance of Start Detector assemblies (SD) consisting of various Diamond Detectors (DDs) and FEEs. An SD is an important component of a Time of Flight system often used in particle physics experiments, allowing for a determination of the time-zero (T0) of the reaction. The diamond materials investigated here are produced by Chemical Vapor Deposition (CVD) of carbon containing radicals onto {100}-oriented silicon wafers (polycrystalline (PC) CVD diamond) or onto mono-crystalline {100}-oriented synthetic diamond substrates (single-crystal (SC) CVD diamond). In the framework of the European Joint Research Activity NoRHDia we optimized different sensors and FEE designs, particularly aiming at the development of relativistic proton SDs for the GSI experiments FOPI and HADES as well as for the upcoming FAIR experiment CBM. The challenging parameter in minimum-ionizing particles (MIP) DD applications is the high pair-production energy of diamond (r ~ 13 eV/e-h pair), which leads to rather small signal amplitudes. We present results from various SD assemblies obtained with relativistic heavy ions and protons. An excellent heavy-ion resolution (i < 30 ps) was obtained using the broadband preamplifier-discriminator plate FEE-1 with both, PC- and SC-devices. Due to the poor S/N ratio in the case of relativistic proton beams the use of that particular FEE plate was prevented. Best proton resolution of i = 100 ps was achieved with low-capacitance assemblies, where amplifier chips of low input capacitance were applied along with detector pcbs of minimized detector and parasitic capacitances. Two variants of single-channel amplifiers are discussed: a double gate MOS source follower and a microwave BJT in common-emitter configuration. For micro-patterned DD applications a fourchannel broadband PreAmplifier DIscriminator (PADI-4) ASIC was designed and realized in 0.18 m technology.

Diamond Start Detectors

50

N13-46:

GAS II: a Versatile Wire Chamber Readout ASIC

M. Newcomer, N. Dressnandt, N. Doshi Physics, University of Pennsylvania, Philadelphia, PA, USA GAS II is an 8 channel wire chamber readout ASIC fabricated in TSMCs um CMOS mixed mode analog and digital process. The design was motivated by the diverse needs of the wire chamber tracker and particle identification sub-systems of the GlueX detector at Jefferson Laboratory. It is suitable for both wire anode and cathode pad readout with selectable differential analog or fast comparator outputs capable of driving low impedance cable. The analog signal processing blocks consist of a preamplifier, shaper with tail cancellation and selectable comparator. The peaking time of ~12ns was chosen to allow sufficient signal collection while keeping GAS II suitable for tracking applications. A programmable gain selector allows the user to preserve wide dynamic range operation for pad or wire readout.. Special attention has been given to minimizing the noise contributions of the amplifier with out adding excessive power burden. Circuit details, layout and measurements will be presented.
N13-47:

G. De Geronimo1, E. Vernon1, K. Ackley1, J. Fried1, Z. He2, C. Herman2, F. Zhang2 1 Instrumentation, Brookhaven National Laboratory, Upton, NY, USA 2 Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA We report on the development of an application specific integrated circuit (ASIC) for 3D position-sensitive detectors. The ASIC is designed to operate with pixelated wide band gap sensors like Cadmium-Zinc-Telluride (CZT), Mercuric Iodide (HgI2), and Thallium Bromide (TlBr), and it measures, corresponding to an ionizing event, the energy and timing of signals from 128 anodes and two cathodes. Each channel provides low-noise charge amplification, high-order shaping with peaking time adjustable from 250 ns to 12 s, gain adjustable to 20 or 120 mV/fC (3.2 MeV or 530 keV in CZT), discrimination with 5-bit trimming, and positive and negative peak- and timing-detections. The readout can be full or sparse, based on a flag and single- or multi-cycle token. All channels, triggered only, or triggered with neighbors can be read out thus increasing rate capability of the system to more than 10 kcps. The ASIC dissipates 330 mW which corresponds to about 2.5 mW per channel.
N13-48:

ASIC for High-Rate 3D Position Sensitive Detectors

Y. Tazawa1, J. D. Cressler2, S. Diezb3, V. Emerson4, N. Dressnandt1, A. A. Grillo5, G. Mayers1, F. Martinez-McKinney5, I. Mandic6, S. Phillips2, S. Rescia4, H. F. W. Sadrozinski5, A. Seiden5, E. Spencer5, H. Spieler6, A. K. Sutton2, F. M. Newcomer1, M. Ullan3, M. Wilder5 1 Physics and Astronomy, University of Pennsylvania, Philadelphia, PA, USA 2 Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georga, USA 3 Centro Nacional de Microelectrnica, Barcelona, Spain 4 Instrumentation, Brookhaven National Laboratory, NY, USA 5 Santa Cruz Institute for Particle Physics, University of California Santa, Santa Cruz, CA, USA 6 Jozef Stefan Institute, Ljubljana, Slovenia Detector mounted electronics at the SLHC will be exposed to ten times the radiation levels of that at LHC. Due to activation and sheer size it will be nearly impossible to access the front end electronics in the inner parts of SLHC detectors. The total dose radiation requirement along with power and size constraints on the front end electronics points toward the use of full custom ASICS designed in deep sub-micron technologies. For maximum versatility a CMOS process with a well supported digital library accompanied by bipolar transistors for power efficient analog signal processing would be ideal. Measurements by our group on a small sample of vendor supplied devices has shown that IBM's BiCMOS 8WL process appears to meet these requirements. To qualify this process we submitted a test ASIC with 40 NPN transistors and 37 CMOS transistors to IBM in November. The parts were returned in March and we have begun a program of Gamma and Neutron Irradiation tests. This talk will primarily cover total dose and early results of dose rate studies employing ionizing radiation sources.
N13-49:

SiGBiT: a Dedicated SiGe and CMOS Device Test ASIC

B. Shi1, K. Shimazoe2, T. Fujiwara1, H. Takahashi2 1 Department of Nuclear Engineering and Management, The University of Tokyo, Tokyo, Japan 2 Department of Bioengineering, The University of Tokyo, Tokyo, Japan

Application of Time Over Threshold Method for Micro Strip Gas Counter

Micro Strip Gas Chamber (MSGC) is an attractive detector for X-ray and neutron detection for its high spatial resolution, high counting rate and high gas gain. However, utilizing popular time-over-threshold (ToT) method to readout MSGC will reduce its energy resolution due to the linearity and dynamic range problems. To overcome the drawback of conventional ToT technique, two new ToT methods have been adopted, the multi-level threshold ToT and the dynamic threshold ToT. We have implemented

51

the new ToT technique with discrete circuits and used it to readout MSGC detector with Fe-55 source. The experiment results validate that with improved linearity and dynamic range, ToT readout method is applicable for MSGC.
N13-50:

Advanced Waveform Digitization with Programmable Windowed Real-Time Trigger Capability

W. Huang, S.-H. W. Chiang, S. Kleinfelder Electrical Engineering, University of California, Irvine, CA, USA An advanced transient waveform digitizer (ATWD) design with continuous sampling capability and real-time programmable windowed trigger generation is presented. Designed in a 0.25 m process, the digitizer circuit contains a circular array of 128 sample and hold circuits for sample storage and attains 2 GHz sample speeds. Sample clock generation adopts a semisynchronous approach, combining a phase-locked loop (PLL) for high-speed clock generation and a high-speed fully-differential shift register for distributing clocks to all 128 sample circuits. This permits a low-frequency external reference clock to generate the high-speed sample clocks. A novel windowed trigger capability is included. Rather than merely set a single analog threshold on the input signal, the recorded signals are compared with programmable positive and negative thresholds, in parallel for all samples. Digital to analog converters set thresholds on a per-sample and per-threshold basis, allowing programmable corrections for offsets plus the ability to change thresholds on the fly. Windows that are 8 samples wide are interleaved across the 128 sample array, and all 128 windows can observe the input signals in parallel. A programmable logic array is used to set trigger conditions over the windows. The basic conditions can include high-threshold (over the programmable high threshold), lowthreshold, neither high nor low, and dont care (high, low or in-between). Over each window of 8 samples, up to 72 different patterns can searched for, allowing considerable flexibility. For example, its program may require 3 high samples followed by 3 low samples, or the opposite, for a valid trigger this would search for a bipolar signal of a certain strength and width. The ability to search for waveform profiles greatly increases trigger selectivity over that of a single threshold. After a trigger is produced, the analog signals are digitized to 10 bits by 128 parallel A/D converters.
N13-51:

C.-S. Hwang1, K.-H. Chen1, H.-W. Tsao2 1 Department of Electrical Engineering, National Yunlin University of Science and Technology, Douliu, Yunlin County, Taiwan 2 Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan In this paper, the new architecture of a timing generator using dual delay-locked loop (DLL) is proposed. With the aid of coarse and fine tuning mechanisms, the timing generator can provide sub-gate resolution with precise close-loop control and instantaneous switching capability. The circuit is implemented and simulated in TSMC 0.35 m 2P4M technology. The chip area occupies 1.36 mm2. It can interpolate the reference clock cycle with 80 divisions to obtain 45 ps resolution when running at 280 MHz. The DNL and INL are within -0.3~+0.6 and -0.8~+0.4 LSB, respectively.
N13-52:

Timing Generator Using Dual Delay-Locked Loop

CCD Base Line Subtraction Algorithms

I. V. Kotov, A. Kotov, J. Frank, P. O'Connor, V. Perevoztchikov, P. Takacs Instrumentation Div., Brookhaven National Laboratory, Upton, NY, USA High statistics astronomical surveys require higher accuracy of sensor calibration procedures. The first step in calibration procedures is the base line subtraction. The accuracy and robustness of different base line subtraction algorithms used for Charge Coupled Device (CCD) sensors are discussed. Also, specialized algorithm developed for analysis of CCD images contain sparse signals is presented. The algorithm statistical properties are discussed in section 3. The algorithm performance on $^{55} Fe$ data and comparison with bias exposures approach is presented in section 4. Details of the bias exposure analysis are presented in section 2.
Authors are greatfull to V.Radeka for support of this work.

52

N13-53:

C. Amoros1,2, M. Carty3, V. Cipolla4, F. Daly3, M. Fesquet3, A. Givaudan5, F. Gonzalez4, K. Lacombe1,2, H. Le Provost3, R. Pons1,2, D. Rambaud1,2, N. Remoue1,2, T. Tourrette3 1 CESR, Universite de Toulouse; UPS, F-31028 Toulouse cedex 4, France 2 UMR5187, CNRS, F-31028 Toulouse, France 3 IRFU, CEA, F-91191 Gif-sur-Yvette, France 4 CST, CNES, 31401 Toulouse cedex 9, France 5 APC, 75205 Paris cedex 15, France The ECLAIRs space telescope is the main French contribution on the payload of the SVOM Chinese-French mission, aiming at being the next gamma ray bursts multi-wavelength observatory. By detecting the very first lights of the gamma ray burst, the ECLAIRs telescope will be the initiator of the measurement process based on a complementary network of space and ground instruments. Since beginning 2006, this instrument has been developed within the SVOM international context. The whole architecture is based on a combination of 3 key solutions: the coded mask imaging, a compact ASIC-CdTe hybrid detector thermally controlled and an onboard powerful processor for gamma ray burst detection. Each of those elements has generated specific developments and optimized solutions in order to reach the expected performances. Those solutions and the status of the developments will be presented.
N13-54:

Development Status of ECLAIRs, a Gamma Ray Burst Observatory for the SVOM Mission

R. Sefri1, V. Tocut2, H. Lebbolo1, S. Bailey1, D. Martin1, P. Antilogus1, C. Delataille2, J. Jeglot2, M. Moniez2, F. Wicek2 1 LPNHE, IN2P3, Paris, France 2 LAL, IN2P3, Orsay, France The LSST camera will have more than 3000 video processing channels to readout its large and highly segmented focal plane, requiring a compact readout chain. The methods to perform Correlated Double Sampling signal processing on CCDs are Dual Slope Integrator (DSI) and Clamp and Sample (C&S). We have designed and implemented an ASIC to directly compare the strengths and weaknesses of these methods on a working device: the Analog Signal Processing asIC (ASPIC). Four channels of each method have been implemented on the same ASIC to perform direct comparisons and fine crosstalk measurements. Channel to channel crosstalk due to total electronics has to be no more than 0.05%, with a 0.01% goal at 500 kHz readout frequency. The other requirements on this readout chain are: - Readout noise < 5 e - Dynamic range 16 bits - Maximum power consumption of 25mW/channel - Working temperature 173 K - Differential outputs driving 1k // 50pF load The chosen technology is 5V compliant CMOS 0,35 by AMS. A second version, ASPIC 2, implements the LSST baseline design of 8 DSI channels with better noise performance and lower power consumption than ASPIC1. It was submitted at the end of 2008. This version is characterized by a 3 bit programmable gain (made by capacitive feedback) of the input amplifier, and a 3 bit programmable time constant integrator in order to match the CCD output conversion and the readout frequency. In order to reduce the power consumption, an idle mode has been implemented. ASPIC 2 is now under test. A CLAmp and Sample aSIC (CLASSIC) chip containing 8 channels was submitted in the beginning of April 2009 in order to perform a comparison between the improved versions of both of these methods. CLASSIC is also characterized by a programmable input amplifier gain and a programmable time constant output filter. Both ASPIC and CLASSIC will be tested with a custom readout board containing eight 18 bits ADC channels, 256 kwords memory and a USB interface.
N13-55:

ASPIC : LSST Camera Readout Chip

A. M. Ilyin1, I. A. Ilyina2 1 Physical Department, Kazakh National University, Almaty, Kazakhstan 2 TRIO, Moscow, Russia

High-Resolving Electrostatic Energy Analyzer for Space Measurements

There are many applications especially in space charged particles and plasma flows measurements, in which energies of electrons are a subject of interest. Therefore electrostatic energy analyzers, combining high-electron optical characteristics with a compact and simple design are needed. Well known analyzers with toroidal or spherical fields [1,2] used in the past investigations have a complicated shape of electrodes and need special fine electrical systems for fringe-field correction or additional lenses for operation. In many cases it is not acceptable for space applications. This paper presents a new type of high resolving instrument (the energy resolution nearly to 0.1 %), which is simple in design, can be made as a light and very compact device with cylindrical symmetry. The new analyzer has one more engineering advantage: it needs no any fringe-field correction systems. The instrument can be easily arranged for operation on the outer surface of the orbital station. References 1. Purcell E M, Phys.Rev. 54 (1938) 818 2. Sagara T, Boesten L, Nishida S and Okada K ,Rev.Sci.Instrum.71 (2000) 4201

53

N13-56:

First Detection of Extensive Air Shower with the EEE Experiment

C. Gustavino Laboratori Nazionali del Gran Sasso, INFN, Assergi (AQ), Italy On behalf of the the EEE Collaboration The Extreme Energy Events (EEE) project is devoted to search extremely high energy cosmic rays, by means of an array of telescopes distributed over the italian territory. The stations are based on the use of Multigap Resistive Plate Chambers (MRPC), that have excellent time resolution and good tracking capability. They are installed in selected secondary school institutes, where students are involved in the research work. The status of the experiment and the analysis of the first extensive air shower detected is presented.
N13-57:

E. C. Bellm1, J.-L. Chiu2, J.-S. Liang2, D. Perez-Becker1, S. E. Boggs1, M. Amman3, M. S. Bandstra1, H.-K. Chang2, Y.H. Chang4, M. A. Huang5, W.-C. Hung4, P. Jean6, C.-H. Lin7, Z.-K. Liu4, P. N. Luke3, C. B. Wunderer1, A. Zoglauer1 1 Space Sciences Laboratory, University of California, Berkeley, Berkeley, CA, USA 2 National Tsing Hua University, Hsinchu City, Taiwan 3 Lawrence Berkeley National Laboratory, Berkeley, CA, USA 4 National Central University, Taoyuan County, Taiwan 5 National United University, Miaoli County, Taiwan 6 Centre dEtude Spatiale des Rayonnements, Toulouse, France 7 National Space Organization, Hsinchu City, Taiwan The Nuclear Compton Telescope (NCT) is a balloon-borne gamma-ray telescope which uses cross-strip germanium detectors to study astrophysical sources of nuclear line emission. The compact design allows for wide-field imaging with excellent efficiency from 0.2 to 10 MeV. Moreover, the Compton imaging principle utilized by NCT provides polarimetric sensitivity from 0.2 to 1 MeV. We have conducted an extensive calibration campaign using radioactive sources prior to our flight in Spring 2009 from Ft. Sumner, New Mexico. I present results of our calibration of the effective area throughout NCT's field of view and compare with Monte Carlo simulations using a detailed mass model. Additionally, I assess NCT's polarimetric capabilities with observations of a partially-polarized beam.
The NCT project is supported in part by NASA under Grant NNG04WC38G for the NCT-US team and by the National Space Organization (NSPO) in Taiwan under Grant96-NSPO(B)-SP-FA04-01 for the NCT-Taiwan team. N13-58:

Efficiency and Polarimetric Calibrations of the Nuclear Compton Telescope

PoGOLite - a Balloon-Borne Soft Gamma-Ray Polarimeter

M. Pearce Dept. of Physics, The Royal Institute of Technology, KTH, Stockholm, Sweden On behalf of the PoGOLite Collaboration PoGOLite is a balloon-borne experiment which stands to open a new observational window on the Universe through measurements of the polarisation of soft gamma-rays from compact astrophysical objects, such as pulsars, and accreting black holes in the energy range 25 - 80 keV. The polarisation is reconstructed using the modulation of Compton scattering angles in the segmented sensitive volume of the instrument composed of plastic scintillators, and surrounded by a BGO anticoincidence shield. A pathfinder flight of PoGOLite from the Esrange facility in the north of Sweden is planned for August 2010. The status of the PoGOLite pathfinder will be reviewed, and science prospects discussed.
N13-59:

A. Chekhtman1,2 1 Space Science DIvision, HESE branch, Naval Research Laboratory, Washington, DC, USA 2 College of Science, CEOSR, George Mason University, Fairfax, VA, USA On behalf of the FERMI LAT collaboration

On-Orbit Calibration and Performance of the CsI Crystal Calorimeter of the FERMI Large Area Telescope

We will present procedures and results of on-orbit calibration of the CsI crystal calorimeter subsystem of the Large Area Telescope on board of Fermi Gamma-ray Space Telescope, launched by NASA on 11 June 2008. The calorimeter is built of 1536 CsI(Tl) crystals arranged as a 4x4 array of identical modules having 8 layers of 12 crystals each. Scintillation light is detected by PIN photodiodes at both crystal ends. This allows a measurement of the position along a crystal and thus reconstruction of a 3dimensional image of energy deposition in the calorimeter. This 3D image is used in background rejection and energy 54

reconstruction algorithms. During calibration procedures we measure pedestals, thresholds (for trigger, data readout and range selection discriminators), electronics linearity, light asymmetry curve and energy scale for each crystal. The peaks in the energy deposition spectrum of charged cosmic rays (from hydrogen to iron) were planned to be used to calibrate the energy scale, but due to non-linear dependence of the scintillation light yield of CsI(Tl) crystals on local ionization density, only the proton peak was used for absolute energy calibration during the first year after launch. To account for this effect we plan to apply corrections to the energy scale dependent on incident particle energy and type. These corrections were measured in beam tests and should be confirmed by a CsI(Tl) crystal scintillation model. On-orbit monitoring of all calibration parameters during the first year of data taking shows very good long-term stability.
N13-60:

Performance Evaluation of Low Complexity EDAC Systems for Application on-Board the Algerian Satellites

Y. Bentoutou Centre des Techniques Spatiales, Arzew, Oran, Algeria Error Detection And Correction (EDAC) systems onboard satellites aim to secure the transaction of data between the central processing unit of the onboard computer and its local memory. This paper is concerned with the performance evaluation of some onboard EDAC methods for application in the Alsat-1 On-Board Computer OBC 386. The OBC386 is an Intel 80C386EX based system that plays a dual role for Alsat-1, acting as the key component of the payload computer as well as the command and control computer for the microsatellite. A new EDAC system is described and implemented in Field Programmable Gate Array (FPGA) technology FPGA. The performance of the proposed EDAC method is measured and compared with two different EDAC devices, using the same FPGA technology.
N13-61:

S.-N. Ishikawa1, S. Watanabe1, H. Odaka1, S. Sugimoto1, S. Saito1, T. Fukuyama1, M. Kokubun1, T. Takahashi1, H. Tajima2, T. Tanaka2, Y. Terada3, S. Krucker4, S. Christe4, S. McBride4, L. Glesener4 1 High Energy Astrophysics Department, ISAS/JAXA, Sagamihara, Kanagawa, Japan 2 KIPAC, SLAC, Menlo Park, CA, USA 3 Department of Physics, Faculty of Science, Saitama University, Saitama, Saitama, Japan 4 Space Science Labolatory, University of Calofornia, Berkeley, Berkeley, CA, USA The Focusing Optics X-ray Solar Imager (FOXSI) is a NASA sounding rocket mission which will study particle acceleration and coronal heating on the Sun through unprecedented high resolution imaging in the hard X-ray energy band (5-15 keV). Energy release occurring in the quiet region of Sun may potentially play an important role in the coronal heating mechanism. With a combination of high-resolution focusing X-ray optics and fine-pitch imaging sensors, FOXSI will achieve superior sensitivity; two orders of magnitude better than that of the RHESSI satellite. We plan to use a Double-sided Si Strip Detector (DSSD) as the FOXSI focal plane detector, which fulfill the scientific requirements on the spatial resolution, energy resolution, and lower threshold energy. We have designed and fabricated a DSSD with a thickness of 0.5 mm, and a dimension of 9.69.6 mm containing 128 strips separated by a pitch of 75 m, which corresponds to 8 arcsec at the focal length of 2 m. The DSSD was successfully operated under a temperature of 20 C and a bias voltage of 350V, and obtained a spectrum from a single channel. The energy resolution was measured to be 1.6 keV (FWHM). Through this detector and an ASIC readout system, an energy resolution of ~1 keV (FWHM) and low energy threshold of ~5 keV are expected, sufficient for the FOXSI mission requirement.
N13-62:

Fine-Pitch Semiconductor Detector for FOXSI Mission

Effective Pixel Area Measurements on CZT Detectors in 3D

I. Kuvvetli, C. Budtz-Jrgensen Astrophysics, DTU Space National Space Institute, Technical University of Denmark, Copenhagen, Denmark The National Space Institute, Technical University of Denmark (DTU Space) coordinates the development of the Modular X- and Gamma-ray Sensor (MXGS) instrument of the ESA mission, the Atmosphere-Space Interaction Monitor (ASIM) in collaboration between University of Valencia, and University of Bergen. The main science goals of the ASIM mission is to measure Terrestrial Gamma Flashes (TGF) and Terrestrial Luminous Events (TLE) above thunderstorms at high altitudes in optical and X- and Gamma-ray bands. The MXGS detector plane is made of a number of 5-mm thick CdZnTe Detector Modules (DM) with a total sensitive area of 1024 cm2. At The DTU Space, we have characterized CZT detector samples from several suppliers and found that pixel-to-pixel uniformity measurements showed unexplained pixel-to-pixel count rate variations. Collimated monochromatic X-ray beams were used to scan a number of CZT pixel detectors. The main conclusion of these measurements is that the pixel-topixel count rate variations are correlated with variations in the effective pixel areas. In order to clarify this we performed a full 3D analysis employing the DOI method to determine the depth dependence of the pixel size variations as well. The investigated CZT pixel detector unit is 10 mm x 10 mm x 5 mm with 16 2.4 mm wide pixel electrodes at a 2.5 mm pitch. The detector was illuminated with a collimated 57Co radioactive source both through the pixel electrode side and through the planar electrode side. The full 3D analysis employing the DOI method revealed the effective pixel 55

boundaries as a function of depth. Figure 1 shows variation of pixel boundaries for the center pixels extracted from the double pixel events distribution as a function of depth. The effective pixel boundaries are clearly evolving as function of depth defining regular pixels close to the pixel electrodes and more deformed pixels at the common cathode electrode.
N13-63:

G. Santin1,2, D. J. Rodgers1, V. Fioretti3 1 TEC-EES, ESA - ESTEC, Noordwijk, Netherlands 2 RHEA Tech Ltd, Louvain-La-Neuve, Belgium 3 INAF/IASF Bologna, Bologna, Italy

Impact of Space Protons and Electrons Scattering Through the IXO Telescope Mirrors

A study was performed of the impact on the IXO focal plane detectors of electrons and low energy protons penetrating through the X-ray optics by a combination of reflection and scattering processes. Total (non-ionizing) dose and flux levels were obtained as a support to a detailed design of the mission configuration.
N13-64:

S. Herrmann1,2, R. Andritschke1,2, J. Elbs1,2, O. Haelker1,2, R. Hartmann1,3, N. Meidinger1,2, L. Strueder1,2, L. Tiedemann1 1 Max Planck Institute for Extraterrestrial Physics, Garching, Germany 2 MPI Halbleiterlabor, Munich, Germany 3 pnSensor GmbH, Munich, Germany The German X-ray observatory eROSITA is the prime instrument of the new Spectrum-RG mission. Launch of the Russian satellite is planned for the year 2011. The scientific goal of eROSITA is primarily an all sky survey with the detection and analysis of 100 thousand clusters of galaxies in order to study the large scale structures in the Universe and to test cosmological models. The therefore required large effective area is obtained by an array of seven identical and parallel aligned Wolter-I telescopes. In the focus of each mirror module, there is a large frame store pnCCD detector, providing a field of view of 60 arc minutes and with a sensitive energy band from 0.3 keV to 11 keV. This pnCCD is specifically designed to serve the scientific goals of eROSITA, resulting in a frame store device with a 28.8 x 28.8 mm2, 384 x 384 pixels imaging area. The conceptual design of the X-ray focal plane cameras is presented with the focus on the front end electronics. Central part of the camera is the pnCCD detector chip, which is developed and produced in our semiconductor laboratory, the MPI Halbleiterlabor. The detector assembly consists mainly of the pnCCD detector, three CAMEX readout ASICs, a six-layer ceramic PCB and a frontendproximate FR4 PCB. The CCD and the ASICs are mounted and wire bonded onto the ceramic, which is equipped in addition with RC-filters for noise optimization. Further active front end electronic components, such as the three analog output buffers and the CCD clock pulse generators, are implemented on a separate FR4 PCB, which is located in close distance (ca. 20cm) and connected by a flex lead to the detector PCB. Whereas the detector ceramics is cooled down to about -80C, the most heat generating components are displaced to the warm frontend-proximate PCB. It is thermally decoupled from the detector by a sufficient length of the flex lead and a minimum copper cross section. Experimental results reaching 133eV for the Mn-K line will be presented
N13-65:

eROSITA Focal Plane Instrumentation Design

R. M. Curado da Silva1, N. Auricchio2,3, E. Caroli2, A. Donati2, F. Schiavone2, J. B. Stephen2, S. Del Sordo4, A. M. F. Trindade1, V. Honkimki5 1 Departamento de Fsica, Universidade de Coimbra, Coimbra, Portugal 2 INAF-IASF-Bologna, Bologna, Italy 3 Dipartimento di Fisica, Universit di Ferrara, Ferrara, Italy 4 INAF/IASF-Palermo, Palermo, Italy 5 European Synchrotron Radiation Facility, Grenoble, France Polarimetry has been recognized as a very important observational parameter for high energy astrophysics (>100 keV), as confirmed recent results obtained from SPI/INTEGRAL data on the polarisation of the Crab pulsar. Therefore the capability of to perform accurate polarimetric measurement of high energy cosmic sources should be included in future space missions. This idea was borne several years ago with a view to implementation in a new generation of wide field telescope for high energy all sky survey and transient sources such as GRBs, but it has also become very appealing in the last three years in the development of a new telescope mission concept based on Laue focusing techniques for the next call for mission ideas of ESA Cosmic Vision. In order to optimize the design of a focal plane for this mission a hard X- and soft -ray polarimeter, a CdZnTe detector prototype has been tested at the European Synchrotron Radiation Facility under an ~100% polarized gamma-ray beam. A Monte Carlo simulation code based on the GEANT4 program has been developed, with which an extensive polarimetric study was performed in order to evaluate the response and performance of the detection plane to the type and level of linearly polarised radiation expected from different hard X and soft gamma ray astrophysical sources. The prototype detector tested is a 5 mm thick CdZnTe array with the anode segmented into 11x11 pixels of 2.5x2.5 mm2. The detector was irradiated by a monochromatic linearly 56

Polarimetry Study with a CdZnTe Focal Plane Detector

polarized beam with a spot diameter of about 0.5 mm at 250 keV. We will report on the obtained results in particular on the sensitivity of the detector to the level of the beam polarisation, the modulation of the Q factor as a function of the azimuthal angle between the detector plane axis and the polarisation plane and on the systematic effects produced by the beam incident angle with respect to the detector plane. These results will be compared with MC simulations based on the GEANT4 program.
N13-66:

K. T. Asai1, T. Takashima2, T. Koi3, T. Nagai1 1 Dept. of Earth and Planetary Sci., Tokyo Institute of Technology, Tokyo, Japan 2 ISAS, JAXA, Kanagawa, Japan 3 SLAC, California, U. A.

Data Calibration Using Geant4 for the Radiation Monitor Onboard the Akebono Satellite

Natural electrons and protons (keV-MeV) in the space contaminate the data reciprocally. In order to calibrate the energy ranges and to remove data contamination on the radiation monitor onboard the Akebono satellite, the detector is investigated using the Geant4 simulation toolkit. A semi-polar orbiting satellite, Akebono, launched in February 1989 is active now and has been observed the space environment at altitudes of several thousands km. The RDM detector onboard Akebono monitors energetic particles in the Radiation Belt. The data given by RDM are for electrons with energy of > 2.5, 0.95-2.5, 0.3-0.95 MeV, protons with energy of 30-38, 15-30, 6.4-15 MeV, and alpha particles with energy of 15-45 MeV. These energy ranges are however based on information of about 20 years ago so that the data seem to include some errors actuary. Therefore we need data calibration. The Geant4 simulation toolkit gives information of trajectories of incident and secondary energetic particles interacted in materials. We examine the RDM monitor in the simulation. It is found from the simulation that electrons show extremely complex trajectories caused by material interactions in the instrument. Some electrons are scattered in the shading material (Al) and the primary detector element (Si) before arriving the main detector elements of the instrument. Moreover, some one escape out of the instrument through the frame of the instrument after scattering and some one bounce between the elements of the instrument. Such scattering motion frequently occurs for electrons of energy around 1 MeV which is principal for the Earths Radiation Belt electrons. The results lead the corrected energy ranges as that E1: >4.2 MeV, E2: > 2.4 MeV, E3: > 1.8 MeV for electron, P1: 32-56 MeV, P2: > 18 MeV, P3: > 7 MeV for proton, and A: 18-56 MeV for alpha particle.
Acknowledgments: The Genat4 site, http://www.geant4.org/geant4/. N13-67:

Photon Detector Developments for the Next Generation Cherenkov Telescope Array AGIS

A. N. Otte SCIPP, University of California in Santa Cruz, Santa Cruz, CA, USA On behalf of the AGIS collaboration The Advanced Gamma-Ray Imaging System (AGIS) is a concept for the next generation observatory in ground based very high energy gamma-ray astronomy. Design goals are ten times better sensitivity, higher angular resolution, and a lower energy threshold than existing Cherenkov telescopes. Simulations show that a substantial improvement in angular resolution may be achieved if the pixel diameter is reduced to the order of 0.05 deg, i.e. two to three times smaller than the pixel diameter of current Cherenkov telescope cameras. At these dimensions, photon detectors with smaller physical dimensions can be attractive alternatives to the classical photomultiplier tube (PMT). Furthermore, the operation of an experiment with the size of AGIS requires photon detectors that are among other things more reliable, more durable, and possibly higher efficiency photon detectors. Alternative photon detectors we are considering for AGIS include both Geiger-mode APD (G-APD) and multi-anode photomultipliers (MAPMTs). Here we present results from laboratory testing of MAPMTs and G-APD along with results from the first incorporation of these devices into cameras on test bed Cherenkov telescopes.
N13-68: Basic Performance of the Polarimeter for Gamma-Ray Bursts Using MAPMTs and Segmented Scintillators

S. Gunji Faculty of Science, Yamagata University, Yamagata, Japan On behalf of the PHENEX Collaboration A gamma-ray burst is one of the most important topics in astrophysics. The radiation mechanism, though much effort has been made to clarify it, is still unknown. In such situation, new observations are required which have not yet been carried out. So we are developing a hard X-ray polarimeter for gamma-ray bursts. This polarimeter consists of several detectors called "unit counter". Each unit counter consists of segmented CsI(Tl) scintillators surrounding segmented plastic scintillators mounted on a multianode photomultiplier. The field of view is about 60 degrees (FWHM). As the incident hard X-ray enters into any plastic scintillator, the hard X-ray is scattered and the scattered hard X-ray is absorbed by any CsI(Tl) scintillator. By detecting the scattering and absorption positions, the two dimensional scattering angle can be determined. We can obtain the information on 57

the polarization of the incident hard X rays because the two dimensional scattering angle depends on the polarization direction. Constructing a prototype unit counter, we have carried out basic experiments using polarized beams. Since the detector response for the slant injection is very important, we investigated the modulation amplitude and the shape of counting rates when hard Xrays at 80 keV are injected at an angle of 30 degrees. As one of the results, we confirmed that the polarimeter can obtain the modulation factor of about 50% even for injected hard X-rays from 30 degrees off-axis. Moreover, we discovered that the modulation shape much depends on the threshold energies of the plastic scintillator and the CsI(Tl) scintillator. If the two energy thresholds are adjusted according to the incident direction, it is possible to keep the modulation curve symmetric even for slanted injection. We will present the results of the experiments for slanted injection and the computer simulation in use of EGS4 (Electron Gamma-ray Simulation Version.4) in detail.
N13-69:

J. M. Chirinos1,2, L. J. Otiniano Ormachea2, D. Allard3, I. Allekotte4, C. Alvarez5, H. Asorey4, H. Barros6, X. Bertou4, M. Castillo7, A. De Castro6, S. Flores8, J. Gonzalez9, M. Gomez Berisso4, J. Grajales7, C. Guada10, W. R. Guevara Day2, J. Ishitsuka8, J. A. Lopez11, O. Martinez7, A. Melfo10, E. Meza12, P. Miranda Loza13, C. Murrugarra6, L. A. Nunez10, G. Perez14, Y. Perez10, E. Ponce7, J. Quispe13, C. Quintero10, H. Rivera13, M. Rosales10, A. C. Rovero15, O. Saavedra16, H. Salazar7, J. C. Tello6, R. Ticona Peralda13, E. Varela7, A. Velarde13, L. Villasenor17, D. Wahl8, M. A. Zamalloa18 1 Michigan Technological University, Houghton, USA 2 Comision Nacional de Investigacion y Desarrollo Aeroespacial, CONIDA, Lima, Peru 3 APC, CNRS et Universite Paris 7, Paris, France 4 Centro Atomico Bariloche, Instituto Balseiro, Bariloche, Argentina 5 Universidad Autonoma de Chiapas, UNACH, Tuxtla Gutierrez, Mexico 6 Laboratorio de Fisica Nuclear, Universidad Simon Bolivar, Caracas, Venezuela 7 Facultad de Ciencias Fisico-Matematicas de la BUAP, Puebla, Mexico 8 Instituto Geofisico del Peru, IGP, Lima, Peru 9 Universidad de Granada, Granada, Spain 10 Universidad de Los Andes, ULA, Merida, Venezuela 11 Departamento de Fisica, Universidad Central de Venezuela, Facultad de Ciencias, Caracas, Venezuela 12 Universidad Nacional de Ingenieria, UNI, Lima, Peru 13 Instituto de Investigaciones Fisicas, UMSA, La Paz, Bolivia 14 Universidad Politecnica de Pachuca, Pachuca, Mexico 15 Instituto de Astronomia y Fisica del Espacio, Buenos Aires, Argentina 16 Dipartimento di Fisica Generale and INFN, Torino, Italy 17 Instituto de Fisica y Matematicas, Universidad de Michoacan, Morelia, Mexico 18 Universidad Nacional San Antonio Abad del Cusco, Cusco, Peru The LAGO (Large Aperture Gamma Ray Burst (GRB) Observatory)-Peru collaboration joined the international LAGO experiment last year. The project's goal is the detection of high energy GRBs. GRBs produce many high energy gamma rays that can reach our planet's atmosphere. When these gamma rays enter the atmosphere they generate showers of particles, which will produce an anomalously high flux of particles detected at ground level. LAGO uses water Cherenkov tanks suitable for detecting particles generated by GRBs. In order to achieve a good acceptance, we can either install a very large array of detectors or a smaller one at very high altitudes. LAGO is an experiment with sites mainly in developing countries, where we are taking advantage of their geography and that much of the hardware which has previously been developed and tested in the Pierre Auger Observatory (PAO). In this way, despite the limited local support for science, we can build high energy astrophysics experiments in Peru. LAGO already has detectors in Argentina, Bolivia, Mexico and Venezuela. The purpose of this paper is to report on the progress of LAGO-Peru and its expectations in the near future. We have constructed a low cost Cherenkov tank prototype with the acquisition electronics and photomultipliers of the engineering array of the PAO. This prototype is already installed in Lima at CONIDA. We are in the testing phase of the prototype, after which we will verify its operation at high altitudes at the Observatory of Huancayo at 3300m.a.s.l. and then finally install it in a very high elevation site (>4800m.a.s.l.) in the region of Junin (Lat/Lon: 12.04 S 75.32 W) where it will be working completely autonomously.
The LAGO project is very thankful to the Pierre Auger collaboration for the lending of the engineering equipment. J.Chirinos thanks the U.S. DOE for its support.

LAGO (Large Aperture GRB Observatory) in Peru

58

J.-L. Chiu1, Z.-K. Liu2, M. S. Bandstra3, D. Perez-Becker3, S. E. Boggs3, J.-S. Liang1, E. C. Bellm3, H.-K. Chang1, Y.-H. Chang2, M. Amman4, C. B. Wunderer3, A. Zoglauer3, P. N. Luke4, M. A. Huang5, C.-H. Lin6, W.-C. Hung2 1 Department of Physics, National Tsing-Hua University, Hsinchu City, Taiwan 2 Department of Physics, National Central University, Jhongli, Taiwan 3 Space Sciences Laboratory, UC Berkeley, Berkeley, USA 4 Lawrence Berkeley National Laboratory, UC Berkeley, Berkeley, USA 5 Department of Energy and Resources, National United University, Miaoli, Taiwan 6 National Space Organization, Hsinchu City, Taiwan The Nuclear Compton Telescope (NCT) is a balloon-borne soft gamma ray (0.2-10 MeV) telescope designed to study astrophysical sources of nuclear line emission and polarization. The heart of NCT is an array of 12 cross-strip germanium detectors, designed to provide 3D positions for each photon interaction with full 3D position resolution to < 2 mm3. Tracking individual interactions enables Compton imaging, effectively reduces background, and enables the measurement of polarization. The key to Compton imaging with NCTs detectors are: determine the energy deposited in the detector at each strips, and tracking the gamma-ray photon interaction within the detector. The 3D positions are provided by the orthogonal X and Y strips, and by determining the interaction depth using the charge collection time difference (CTD) between the anode and cathode. Our preliminary calibrations of the energy, the 3D position of interactions, and the imaging for the instrument have been completed. Here we will present the calibration techniques and results.
N13-71:

N13-70:

Energy, Depth Calibration, and Imaging Capability of Nuclear Compton Telescope

A 4D Monte Carlo Compton Scattering Code

C. Sun, Y. K. Wu Physics Department, Duke University, Durham, NC 27708, USA An x-ray or gamma-ray beam produced by Compton scattering of a laser beam with a relativistic electron beam has been used for nuclear physics research as well as radiography. To predict the spatial and energy characteristics of the Compton gamma-ray beam, which are critical for many application utilizing such beams, we developed a 4-dimensional (time and 3D space) Monte Carlo simulation code (MCCMPT) to model Compton scattering process. This code has been benchmarked against the measured energy spectrum of Compton gamma-ray beams produced at the High Intensity Gamma-ray (HIS) Source facility at Duke University.
N13-72:

A. Masoni1, M. Carpinelli2, G. Fenu3, A. Bosin4, D. Mura5, I. Porceddu6, G. Zanetti7 1 INFN Sezione di Cagliari, Cagliari, Italy 2 Department of Mathematics and Informatics, University of Sassari, Sassari, Italy 3 Department of Mathematics and Informatics, University of Cagliari, Cagliari, Italy 4 Department of Physics, University of Cagliari, Cagliari, Italy 5 Consorzio Cybersar, Cagliari, Italy 6 INAF, Cagliari, Italy 7 CRS4, Pula, Italy

CyberSar: a Lambda Grid Computing Infrastructure for Advanced Applications

CyberSar is a distributed computing infrastructure based on a core of sites interconnected by dark fibers. The whole system of 1500 processors is integrated as a grid infrastructure based on glite middleware and provides services to a variety of high end applications within the region spanning from astrophysics, biotechnology, climate modeling, geophysics and high energy physics. CyberSar provides computational resources to a wide range of local research groups inserted in outstanding international experiments. The network core based on dark fibers provides an optical network backbone that can support application driven dynamic creation of point to point multi-lambda optical circuits. This way it can dynamically allocate resources to applications via a general control plane orchestrating the system computational and network components. The CyberSar infrastructure is inserted in the Italian Grid Initiative (IGI) and trough IGI it contributes to the European Griod Infrastructure (EGEE) . It is also integrated in the South Italy Grid Initiative (GRISU) where five distributed grid infrastructures with over 8000 processors fully interoperate. This presentation report the infrastructure high level features and its impact for applications. This work makes use of results produced by the Cybersar Project managed by the CYBERSAR Consortium. The project is cofunded by the Italian Ministry of University and Research (MUR) within the Programma Operativo Nazionale 2000-2006 "Ricerca Scientifica, Sviluppo Tecnologico, Alta Formazione" per le Regioni Italiane dell' Obiettivo 1 (Campania, Calabria, Puglia, Basilicata, Sicilia, Sardegna) Asse II, Misura II.2 Societa' dell' Informazione, Azione a "Sistemi di calcolo e simulazione ad alte prestazioni. More information is available at http://www.cybersar.it. The project benefits also to the access the fiber network infrastructure deployed thanks to a dedicate investment of the Sardinian Autonomous Region.

59

N13-73:

The Architecture of BESIII Offline Database

Y. Chu Experiment Physics Center, Institute of High Energy Physics, Beijing, P.R.China This paper present the database for use in the current BES3 experiment at IHEP. The database can be used to manage history status data of detectors, electronics, slow control subsystem, and online subsystem. The database is also used to manage calibration data, DQM data and Monte Carlo data. These are the data necessary for the online DQM monitor, reconstruction, calibration and data analysis. The BES3 database also provide user with web services to make it possible to monitor the status of detectors, slow control subsystem, and online subsystem locally or remotely. The database engine is MySQL server. The database technology route and the database services for the distributed application environment are also presented in this paper.
N13-74:

A. Castoldi1, C. Guazzoni1, C. Ozkan2 1 Dip. Elettronica e Informazione, Politecnico di Milano and INFN, Milano, Italy 2 Dip. Fisica, Universita' degli Studi and INFN sez. Milano, Milano, Italy

A 3-D Simulation Tool for Design and Data Correction of X-Ray Scatter Imaging Experiments

We developed a complete three-dimensional simulation tool able to compute the interaction of the incoming x-ray beam with a sample having arbitrary x-ray cross-sections and to record the emitted radiation at a given scattering angle by the collimationdetection system. The philosophy was to develop a simple though efficient code with a high degree of flexibility in the choice of the experiment geometry and of the properties of the individual components to aid design optimization and data correction procedures of x-ray scatter experiments. Tests on representative case studies will be described where image contrast is estimated as several parameters (e.g. diffraction angle, sample properties, incoming beam, collimator) were systematically varied. The code was also validated by comparing the results of a real low-angle x-ray scatter system. A data correction procedure to minimize attenuation effects on the acquired X-ray diffraction images is also shown.
N13-75:

ATLASeditor3D, 3D Database Editor for the ATLAS Experiment

J. Molina-Perez, K. Pommes PH/ADO, CERN, Geneva, Switzerland The ATLAS Experiment is the biggest detector of the Large Hadron Collider, and contains thousands of tones of equipment. Existing software at CERN allowed the configuration of the ATLAS off-detector electronics, i.e. back end; although a description of the detector itself, i.e. front side, was missing. ATLASeditor3D, a graphical user interface in Java/Java3D which interacts with Oracle databases, was developed to implement a 3D view of the ATLAS detector. The detector is synthesized in nominal position starting from basic geometrical shapes, which represent the envelopes to be connected with the back end side. In addition, detector symmetries were exploited to create multiple objects at once. The geometry, position, and logical relationship of 3D objects created through this GUI are stored in a database, so that the detector layout can be easily loaded. ATLASeditor3D has granted the interconnectivity of the ATLAS back end and front end electronics, and due to the association of Java objects with Oracle tables, its functionality has been extended. By integrating survey data, this software tool was adapted to show displacements for the Tile Calorimeter sub-detectors. In order to ensure equipment traceability of all detector equipment for safety matters, it is currently used to identify equipment location within the ATLAS cavern. As a 3D-shape database editor, ATLASeditor3D can be used to synthesize any other detector of the LHC, and further development is focused on integrating a 3D view of the activation studies.
N13-76:

Parameterized Simulation of the CMS Calorimeter Using GFlash

D. Jang Physics, Carnegie Mellon University, Pittsburgh, PA, USA On behalf of the CMS Collaboration The simulation of the CMS detector operating at the CERN Large Hadron Collider (LHC) is based on the GEANT 4 software package. An alternative simulation of the CMS electromagnetic (ECAL) and hadronic (HCAL) central and forward calorimeters based on the GFlash parameterization package has been implemented into the CMS software structure. The GFlash program is a fast simulation of electromagnetic and hadronic showers using parameterizations of longitudinal and lateral shower profiles. We report on tuning the CMS calorimeter response to ECAL and HCAL test-beam data and compare the gain in CPU time to the standard GEANT 4 calorimeter simulation.

60

N13-77:

CVMFS-Based Easy-to-Install Linux to Distribute Large Computing Software to End Users

Y. Yao Lawrence Berkeley National Laboratory, Berkeley, CA, USA On behalf of the ATLAS Collaboration ATLASs tier-3 computers are managed locally in separate institutions. ATLAS software for tier-3 computers are primarily distributed via downloadable installation kits. Each kit has a size of ~8GB and there are frequent new releases and bugfix releases. The current distribution system has the following issues: the end user or site manager needs to frequently download or update the ATLAS software; keeping multiple copies of ATLAS releases on each tier-3 computer is a waste of disk space; some institutions put the ATLAS release on a network file system (e.g. NFS) and serve it to all tier-3 computers inside the institution which requires added maintenance of the server; nightly builds, which are very important in the process of software development, are not distributed to tier-3 computers due to their size and update frequency. The CernVM file System (CVMFS) can help to solve these problems. CVMFS is FUSE-based Linux file system that only obtains a file via HTTP when the file is accessed. By including CVMFS inside a Scientific Linux CERN (SLC), and making an automated installation CD, we can reduce the maintenance cost of tier-3 computers by eliminating the effort to update ATLAS software releases. Our procedure is as follows: 1. Determine a minimum but sufficient install of SLC, 2. Include CVMFS as a package, with some command line/Graphics Tools to manage the software releases, 3. Use tools like pungi to make installation CD/Repository of this SLC, automating the installation process with Anaconda and Kickstart. The site manager/end user can obtain the installation CD and automatic install it to any computer with minimum user interference. All packages that are available to SLC and all updates are accessible via the YUM (RPM) package manager. By making an Easy-To-Install Linux distribution, we can significantly reduce the maintenance cost of tier-3 computers in small institutions and increase the accessibility of ATLAS software.
Co-Author: Gordon Watts University of Washington, Seattle, WA, USA N13-78:

A New Specialized Data Format for Commissioning of the ATLAS Experiment and Physics Analysis

B. Radics University of Bonn, Bonn, Germany On behalf of the ATLAS Collaboration In the commissioning phase of the ATLAS experiment, low-level Event Summary Data (ESD) are analyzed to evaluate the performance of the individual subdetectors, the performance of the reconstruction and particle identification algorithms, and obtain calibration coefficients. In the GRID model of distributed analysis, these data must be transferred to Tier-1 and Tier-2 sites before they can be analyzed. However, the large size of ESD (~1 MByte/event) constrains the amount of data that can be distributed on the GRID and is available on disks. In order to overcome this constraint and make the data fully available, new data formats - collectively known as Derived Physics Data (DPD) - have been designed. Each DPD format contains a subset of the ESD data, tailored to specific needs of the subdetector and object reconstruction and identification performance groups. Filtering algorithms perform a selection based on physics contents and trigger response, further reducing the data volume. Thanks to these techniques, the total volume of DPD to be distributed on the GRID amounts to 20% of the initial ESD data. An evolution of the tools developed in this context serves to produce another set of DPD s that are specifically tailored for physics analysis. All selection criteria and other relevant information is stored inside these DPD s as meta-data and a connection to external databases is also established.
N13-79:

M. G. Pia1, M. Augelli2, M. Begalli3, L. Quintieri4, P. Saracco1, M. Sudhakar1, G. Weidenspointner5, A. Zoglauer6 1 INFN Genova, Italy, Genova, Italy 2 CNES, Toulouse, France 3 State University, Rio de Janeiro, Brazil 4 INFN LNF, Frascati, Italy 5 MPE-MPI, Munich, Germany 6 university of California, Berkeley, USA A variety of software implementations for each process of photon interaction with matter is available in the Geant4 toolkit; they correspond to nominally different modeling approaches. Despite their relevance to many experimental applications from calorimetry in high energy physics to bio-medical dosimetry and astronomy limited documentation is available in literature about their respective features and accuracy, apart from a study published in 2005 concerning a comparison to reference theoretical calculations available from the NIST. Moreover, the features of some of the implementations have evolved since the time of that publication. An extensive, quantitative study of Geant4 photon interaction models has been performed, addressing cross sections and final state generation. It involves the inter-comparison of the various Geant4 implementations, their physics accuracy against reference data and computational 61

Inter-Comparison and Validation of Geant4 Photon Interaction Models

performance. The different features of the Geant4 models have been estimated by means of rigorous statistical analysis methods. The results demonstrate various features relevant to experimental simulation applications, like the relative physics capabilities and computational speed of Geant4 models; they provide guidance for the optimal use of Geant4 physics models in experimental applications. Some features deserving attention in view of possible Geant4 improvement have been highlighted: duplication of the same physics functionality in different software implementations, discrepancies between implementation and specifications (in User Documentation), multiple and inconsistent definitions of fundamental atomic parameters affecting the simulation accuracy etc. An overview of the main results is presented and recommendations for the improvement of this Geant4 physics domain are discussed.
N13-80:

M. G. Pia1, M. Sudhakar1, G. Weidenspointner2, A. Zoglauer3 1 INFN Genova, Italy, Genova, Italy 2 MPE-MPI, Munich, Germany 3 university of California, Berkeley, USA

Prototype for Mutable Compton Scattering Simulation with Geant4

A Geant4-related R&D project has been recently launched to address fundamental methods in radiation transport simulation, which have been motivated by new experimental requirements. The project focuses on simulation at different scales in the same experimental environment: this set of problems requires new methods across the boundaries of condensed-random-walk and discrete transport schemes as currently treated in major Monte Carlo systems. The project elaborates a conceptual scheme for condensed and discrete simulation approaches to co-work in the same environment, and a software design capable of supporting it. This requirement implies the introduction of a new concept in the simulation mutable physics entities (process, model or other physics-aware object), whose state and behavior depend on the environment and may evolve as an effect of it. Such a new concept requires rethinking how Geant4 kernel handles the interaction between tracking and processes, and represents a design challenge in a Monte Carlo software system. A concept of mutability, at the present time applicable to discrete processes, has been designed and applied to Compton scattering simulation with Geant4. This process represents an ideal playground to study design solutions concerning the configuration and mutability of its physics functionality: in terms of physics features, various models are available to calculate its cross section and to determine the characteristics of the secondary particles it produces, also including effects like atomic relaxation. From an experimental perspective, the relevance of the simulation details varies with the photon energy and the environment where the process occurs, ideally requiring a process which is intrinsically capable of mutation in the course of the simulation. First results from a design prototype and application to a Compton telescope are presented.
N13-81:

S. Levinson1, B. Sarusi1, A. Osovizky2, V. Pushkarsky2, U. German1, E. Marcus1, Y. Cohen1, I. Belaish1 1 Electronics & Control Laboratories, Nuclear Research Center Negev, Beer-Sheva, Israel 2 Health Physics Instrumentation Department, Rotem Industries Ltd, Beer-Sheva, Israel "GammaGen" (GG) is a software computer program developed to generate synthetic pulse height spectra due to gamma rays in various detectors. A pulse height spectrum originated by a specific radio-nuclide is generated according to its activity, photo peak energy and yield. The detector efficiency, the resolution and peak to Compton dependence are also taken into consideration, as well as attenuation by absorbers placed between the radiation source and the detector. The resulting pulse height spectrum and its components are displayed on the computer screen. Zooming functions are available for easy and detailed view of the spectra obtained. The channel, energy and nuclide data are displayed according to the cursor position in the spectrum. The spectra can be displayed in several modes: as energy lines of the photo peaks, as Gaussian of each photo peak, or as sum of all Gaussians. The Compton contribution can also be included. The radio-nuclides and their activity can be selected at will. The spectra can be exported to formats required by commercial spectra analyzing programs and to Excel format. The software can be used as a tool to optimize the detector configuration when developing a radio-nuclides analysis system, by predicting the pulse height spectrum which will be obtained for a specific setup. Complicated overlapping peaks can be easily analyzed, and the effectiveness of the analysis system can be tested.

Synthetic Gamma-Ray Spectra for Homeland Security Radio-Nuclides Analysis

62

N13-82:

F. Garcia1, H. Andersson2, M. Roos2, K. Banzuzi2, J. Huovelin3 1 Detector Laboratory, Helsinki Institute of Physics and Department of Physcal Sciences, University of Helsinki, Helsinki, Finland 2 Oxford Instrument Analytical Oy, Helsinki, Finland 3 University of Helsinki Observatory, Helsinki, Finland BepiColombo is European Space Agencys (ESA) mission to Mercury[1]. It has been named after the Italian mathematician and engineer Giuseppe (Bepi) Colombo, known especially for his works on Mercury. The spacecraft will set off in 2013 on a journey lasting about six years. One of the scientific instruments onboard the mission will be SIXS, Solar Intensity X-ray and particle Spectrometer. It is developed in a Finnish consortium headed by the Observatory of the University of Helsinki. The Sensor Unit of SIXS consists of two parts: an X-ray detector and a particle detector for protons and electrons. The X-ray part consists of three GaAs sensors, one of which will always see the sun. The particle detector package consists of a core sensor (CsI scintillator with a photodiode) surrounded by five thin Si surface sensors. The particle sensors are covered by a collimator. Since SIXS measures the X-ray and particle fluxes coming from the Sun, its data are vital for the calibration of the MIXS (Mercury Imaging X-ray Spectrometer) instrument, which measures the X-ray fluorescence radiation coming from the surface of Mercury. For the simulations the Geometry description has been directly imported from CAD models[2] and the Sector Shielding Analysis Tool (SSAT)[3] has been used to obtain the dose in the components. The dose absorption map of the electronics components will be presented.
N13-83:

Electronic Components' Dose Mapping for the SIXS Sensor Unit of the BepiColombo Mission

H. Seo1, J. H. Park1, C. H. Kim1, S. H. Lee2, J. H. Lee3, C. S. Lee3, J. S. Lee4 1 Dept. of Nuclear Engineering, Hanyang University, Seoul, Korea 2 Dept. of Radiation Oncology, Asan Medical Center, Seoul, Korea 3 Dept. of Physics, Chung-Ang University, Seoul, Korea 4 Dept. of Nuclear Medicine and Interdisciplinary Program in Radiation Applied Life Science, Seoul National University, Seoul, Korea The Compton camera has a great potential to be a high-performance imaging device in nuclear medicine and industrial applications for its unique advantages including 3-D imaging capability without requiring multi-directional data acquisitions, high imaging sensitivity, multi-tracing capability, and wide field-of-view. The performance of a Compton camera is affected by the timing characteristics of the detectors and data acquisition electronics mainly because Compton imaging is based on coincidence detection. The timing characteristics, however, are usually neglected in Compton camera simulation, resulting in large errors in imaging sensitivity and less but still significant errors in imaging resolution. In the present study, the timing characteristics of the detectors and data acquisition electronics are explicitly modeled and added to the developing Compton imaging simulator (CIS) for more accurate simulation of a Compton camera, and the upgraded CIS was tested against the experimental results. The timing characteristics modeled in the present study include the detector time resolution, detector deadtime, coincidence window, and artificial system dead-time. Compared with the previous simulation results which did not includes the timing characteristics, the discrepancies of imaging sensitivities, between simulation and experiment, were reduced to an insignificant level. The imaging resolutions were also in a good agreement, i.e., < 1 mm FWHM of difference, after explicitly modeling the timing characteristics. Our results show that the new simulator produces much more accurate results both for imaging sensitivity and imaging resolution. This user-friendly GUI-based simulator will significantly reduce the time and efforts to simulate the Compton camera in the future.
This work was supported by the Nuclear R&D Program in Korea through BAERI and Basic Research Program. This work was also supported by Korean Ministry of Knowledge Economy (2008-P-EP-HM-E-06-0000)/Sunkwang Atomic Energy Safety Co., Ltd. N13-84:

Development of GUI-based Compton Imaging Simulator with Timing Characteristics of Compton Camera

Condition Data for the CMS Silicon Strip Tracker Reconstruction

M. De Mattia INFN and University of Padova, Padova, Italy On behalf of the CMS Collaboration The silicon strip tracker of the CMS experiment at LHC is the largest silicon tracker ever built for high energy physics experiments. With nearly 9.6 millions of analog read-out channels it poses significant challenges for the reconstruction, calibration and monitoring of the collected data. Beside the event-by-event data, recorded from the detector, the reconstruction is based on a set of non-event data that allow to handle the status of the devices in order to interpret the event data. These non-event data are referred to as condition data. The condition data are held in several databases designed to configure the read-out devices, the online high-level trigger system as well as the offline reconstruction and analysis applications. In this contribution we will 63

describe the condition data used for the silicon strip tracker data reconstruction, their optimizations connected to the timing and CPU constraints, the procedure to manipulate and transfer them from a database to another, assuring the consistency and synchronicity of the available information.
N13-85:

Use of Late-Binding Technology for Workload Management System in CMS

H. Pi University of California, San Diego, La Jolla, CA, USA Condor glidein-based workload management system (glideinWMS) has been developed and integrated with distributed physics analysis and Monte Carlo production at CMS. The late-binding between the jobs and computing element (CE), and validation of worknode (WN) environment help reduce the failure rate of user jobs. For CPU-consuming data production, opportunistic grid resources can be effectively explored via the extended computing resource pool of various heterogeneous sites. The VO policy is supported to simplify the configuration. GSI authentication and interfacing with gLExec allows a large user basis to be supported and seamlessly integrated with CMS computing infrastructure. The highly available and stable service for the general users of CMS virtual organization (VO) is built on glideinWMS, which benefits from continuous efforts of improving the scalability and streamlining the daily operation of the overall system. The enhanced monitoring allows administrators and users to track the system-level and job-level status respectively.
N13-86:

E. Vilucchi1, A. Andreazza2, D. Anzellotti3, D. Barberis4, S. Campana5, G. Carlino6, C. Ciocca7, L. dell'Agnello7, A. De Salvo3, A. Di Girolamo5, A. Doria6, M. L. Ferrer1, A. Italiano7, L. Luminari3, B. Martelli7, A. Martini1, L. Magnoni7, L. Merola8,6, E. Musto6, L. Perini9,2, M. Pistolese2, D. Rebatto2, S. Resconi2, L. Rinaldi7, D. Salomoni7, L. Vaccarossa2, R. Zappi7 1 Laboratori Nazionali di Frascati, Istituto nazionale di Fisica Nucleare, Frascati, Italy 2 Sezione di Milano, Istituto nazionale di Fisica Nucleare, Milano, Italy 3 Sezione di Roma1, Istituto nazionale di Fisica Nucleare, Roma, Italy 4 Sezione di Genova, Istituto nazionale di Fisica Nucleare, Genova, Italy 5 CERN, Geneva, Swizerland 6 Sezione di Napoli, Istituto nazionale di Fisica Nucleare, Napoli, Italy 7 CNAF, Istituto nazionale di Fisica Nucleare, Bologna, Italy 8 Universita' degli Studi di Napoli Federico II, Napoli, Italy 9 Universita' degli Studi di Milano, Milano, Italy With this work we present the activity and performance optimization of the Italian computing centers supporting the ATLAS experiment forming the so-called Italian Cloud. We describe the activities of the ATLAS Italian Tier2s Federation inside the ATLAS computing model and present some Italian original contributions. We describe StoRM, a new Storage Resource Manager developed by INFN, as a replacement of Castor at CNAF - the Italian Tier1 - and under test at the Tier2 centers. We will also show the failover solution for the ATLAS LFC, based on Oracle DataGuard, load-balancing DNS and LFC daemon reconfiguration, realized between CNAF and Tier2 in Roma. Finally we will describe the sharing of resources between Analysis and Production, recently implemented in the ATLAS Italian Cloud with the Job Priority mechanism.
N13-87:

Activities and Performance Optimization of the Italian Computing Centers Supporting the ATLAS Experiment

A. Fella1, E. Luppi2, L. Tomassetti2 1 INFN - CNAF, Bologna, Italy 2 Department of Physics, Universita' di Ferrara and INFN, Ferrara, Italy

Evaluation of AMGA as a Data-Handling Tool for a New HEP Experiment

High Energy Physics experiments need a huge amount of Monte Carlo Simulated events. Currently, Grid computing infrastructures are widely used for the distributed execution of this kind of CPU consuming jobs. To locate and retrieve files spread over the grid, an efficient tool to query and discover information about their content (metadata) is needed. AMGA is a Metadata Catalog Service, which is part of the gLite software stack, being developed for the EGEE project. In this work we wish to evaluate the capability of AMGA to fulfill the requirements of the SuperB experiment simulation production. For this purpose we setup a simple prototype tailored to meet SuperB characteristics. We estimate the overheads and address the performance issues when AMGA is used as a metadata service in our prototype.

64

N13-88:

Muon Reconstruction and Identification in CMS

D. Trocino Dip. Fisica Sperimentale, Torino University, INFN Torino, Torino, Italy On behalf of the CMS Collaboration Muonic final states will provide clean signatures for many physics processes at the LHC. One of the main goals of the Compact Muon Solenoid (CMS) design is thus to ensure efficient and accurate identification and reconstruction of muons. A sophisticated muon system is used for muon identification and stand-alone reconstruction and the inner silicon tracker exploits the high magnetic field to ensure a very precise transverse momentum resolution. The global reconstruction algorithms combine muons reconstructed in the dedicated spectrometer with tracks reconstructed in the inner detector. The CMS reconstruction software is well suited for both offline reconstruction and online event selection (HLT) and its performance has been studied in detail using Monte Carlo simulations. The muon reconstruction has also been employed successfully to reconstruct cosmic muons traversing the CMS detector. The design of the CMS muon identification and reconstruction is presented, as well as its performance on simulated and cosmic data.
N13-89:

A. Ribon1, J. Apostolakis1, G. Folger1, V. Grichine1, V. Ivanchenko1, M. Kosov1, V. Uzhinskiy1, D. Wright2 1 PH/SFT, CERN, Geneve, Switzerland 2 SLAC, Stanford, U.S.A. Transportation codes for medium- and high-energy applications need to simulate hadronic interactions in a wide energy range, typically from MeV to hundreds of GeV. Yet the range of applicability of detailed hadronic models typically spans more limited kinematic regions. Therefore, different models are combined in order to address the full spectrum of hadronic collisions. We present an overview of the transitions between hadronic models in Geant4, with the emphasis on configurations (Physics Lists) relevant to applications in LHC experiments. Unphysical discontinuities in calorimeter observables have been reported by ATLAS and CMS in the transition regions between hadronic models. We have investigated the microscopic origin of these effects, and present the first improvements in the energy dependence of key physical observables in Geant4 simulation.
N13-90:

Transition Between Hadronic Models in Geant4

R. C. Torres1, D. E. F. de Lima1, E. F. Simas Filho1,2, J. M. de Seixas1 1 Signal Processing Laboratory, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil 2 Simoes Filho Campus, Federal Institute for Education, Science and Technology of Bahia, Salvador, BH, Brazil Among LHC detectors, ATLAS aims at coping with such high event rate by designing a three-level online triggering system. The first level trigger output will be ~75 kHz. This level will mark the regions where relevant events were found. The second level will validate LVL1 decision by looking only at the approved data using full granularity. At the level two output, the event rate will be reduced to ~2 kHz. Finally, the third level will look at full event information and a rate of ~200 Hz events is expected to be approved, and stored in persistent media for further offline analysis. Many interesting events decay into electrons, which have to be identified from the huge background noise (jets). This work proposes a high-efficient LVL2 electron / jet discrimination system based on neural networks fed from preprocessed calorimeter information. The feature extraction part of the proposed system performs a ring structure of data description. A set of concentric rings centered at the highest energy cell is generated for each calorimeter layer, producing a total of 100 ring. Energy normalization is later applied to the rings, making the proposed system usable for a broad energy spectrum. For data compaction and signal decorrelation, a segmented preprocessing scheme is developed. For this, Principal Component Analysis (PCA) and Principal Component of Discrimination (PCD) are compared in terms of compaction rates and classification efficiency. Moreover, as signal decorrelation is performed either linearly (PCA) or nonlinearly (PCD) in the preprocessing phase, the classifier complexity can be reduced, which also favors signal preprocessing adoption in level two trigger strategy. For the hypothesis testing section, an artificial neural network was employed. Neural networks are fast to execute, and can provide non-linear cuts in a high dimensional space, providing better pattern separation for complex problems when compared to linear decision systems.
The authors would like to thank CAPES, CNPq, FAPERJ, FINEP (Brazil), and CERN and European Union for their nancial support. Also, we would like to thank the TDAQ collaboration of ATLAS for discussions concerning this work.

Neural Online Filtering Based on Preprocessed Calorimeter Data

65

N13-91:

D. S. Santos1, P. P. Queiroz1, M. C. Martins1, R. S. Silva2, M. Begalli2, M. G. Pia3 1 Instituto de Radioproteo e Dosimetria, Rio de Janeiro, Brazil 2 Physics Institute, State University of Rio de Janeiro, Rio de Janeiro, Brazil 3 INFN Genova, Genova, Italy

Anthropomorphic Phantoms and Geant4-Based Implementations for Dose Calculations

Various models of anthropomorphic phantoms have been developed in the past decades for usage in Monte Carlo simulation for radiation protection and other medical physics applications; they adopt two alternative approaches, either modelling the body components through analytical geometrical representations or through voxelized geometries. An original system has been developed in the context of the Geant4 simulation toolkit. It exploits Geant4 advanced capabilities to model geometrical components, and the object oriented technology to provide a variety of models of the human body usable in a simulation application. A design allows the creation of analytical or voxel phantoms of different sex and age. Phantoms based on established models are provided.
N13-92:

openEyA Tool for Medical Webcasting

M. Zennaro, L. Tenze, E. Canessa Science Dissemination Unit, The Abdus Salam International Centre for Theoretical Physics, Trieste, Italy We introduce to the medical community a free, linux-based openEyA platfor m to carr y out academic webcasting. It allows to record, archive and share traditional lectures/talks carried out using chalkboards as well as moder n presentation tools (PPT, PDF, animations, etc). The nal result includes audio and video recording in Flash for mat synchronized with high resolution zoomable images. This synchronization is done automatically in few minutes. OpenEyA is innovative because it does not require any dedicated human inter vention during recording and post-processing (no need for an operator or editor) and there are no special requirements for the speaker or lecturer (no need to press buttons during the lecture or to wear a microphone, etc).
N13-93:

Influence of Shielding Composition on Transmission Curves Determination for Diagnostic Radiology: a Monte Carlo Study Using the GEANT4 Code

G. Hoff, S. F. Firmino, W. P. de Souza Physics College, Pontifical Catholic University in Rio Grande do Sul, Proto Alegre, RS, Brazil The transmission curves published generally are defined for specific material composition, that represent a mean composition or the compositions defined by ICRU report 44 or by NIST recommendations. Based on that on this work was studied the influence of the composition of different material (barite concerte, barite mortar and steel) on determination of transmission curves applied on diagnostic radiology. To simulate the interaction of the photons to the matter was used the Geant4 tool kit (version 9.2.p01). The computational universe to determine the transmission of primary beam considered: an aluminum add filtration 3.0mm thicker; sensitive volume-source distance of 1.1m; a wall-surface distance to source of 1.0m been the wall 1.60m higher and 1.60m larger with different thickness according the shielding composition. Were simulated monoenergetic beams, from 60keV and 150keV, in steps of 10keV; and polienergetic spectra according the description of the "Catalogue of Diagnostic X-Ray Spectra and Other Data" for electrons accelerating tensions of 100kV, 120kV and 150kV. The results presented on this work shown mean variations on transmission data between -8% and 25%, depending of the shielding material composition and spectra energy studied. The data generated on this study shown that a spectra applied to diagnostic radiology can be significantly sensitive of shielding composition. Variations larger than 20% were calculated for monoenergetic and polienergetic beams. As the monoenergetic beam presented shown sensitivity to material shielding composition we suggest studies considering the influence on transmission of the different spectra characteristic of usual energies used to shielding thickness determination. That way it could be used to otimize the shielding thickness in diagnostic radiology without increase the risk due the radiation exposure for adjacent areas.
N13-94:

M. Beck1, M. Binkley2, J. Boote3, E. Boyd3, A. Brown3, B. Brown2, S. de Ledesma2, D. Engh2, E. Kissel4, T. Moore1, C. Sellers1, P. Sheldon2, M. Swany4, A. Tackett2, M. Zekauskas3, J. Zurawski3 1 Dept. of Electrical Engineering and Computer Science, University of Tennessee, Knoxville, TN, USA 2 Advanced Computing Center for Research and Education and Dept. of Physics and Astronomy, Vanderbilt University, Nashville, TN, USA 3 Internet2, Ann Arbor, MI, USA 4 Dept. of Computer and Information Sciences, University of Delaware, Newark, DE, USA The experience of the last few years has provided the global scientific community with a sobering object lesson in the magnitude of the challenge it faces in creating networks that can serve todays data intensive research collaborations. Despite the remarkable 66

Explicit Buffer Networking: a New Approach to High Throughput Data Transfer

build out of ESnet and allied research networks, substantial improvements in the rapid provisioning of lambdas and circuits, and years of intense work on performance optimizations for transport protocols such as TCP, we have now passed 2008 with the situation substantially unchanged. The E2E network challenge identified by the 2003 Roadmap - routinely achieving sustained transfer rates above 10Gbps - remains a distant aspiration. In this paper we discuss the results of experiments we performed to show that a new model for high-throughput data transfer that we call explicit buffer networking (EBN) offers a plausible strategy for solving this problem in a relatively cost effective and sustainable way. The goal of EBN is to design and develop a network that incorporates this new buffer paradigm. Its underlying hypothesis is reasonably straightforward: If we implement an explicit buffer service with memory/storage resources that are well provisioned and highly adapted for network access, then we can design protocols and algorithms that build on that service to implement overlay data services which routinely make use of long, sustained bursts to achieve order of magnitude improvements in network throughput. We believe that by using fast, plentiful buffer media to stage large aggregations of data, we can enable optimal utilization of dedicated network circuits by coordinating depot bursts at channel capacity for a few seconds (or few minutes) and then releasing the resource.
N13-95:

M. Begalli1, S. S. D. O. Fonseca-Rodrigues2, P. P. Queiroz2, D. D. S. Santos2 1 Physics Institute, State University of Rio de Janeiro (UERJ), Rio de Janeiro, Brazil 2 Institute for Radiation Protection and Dosimetry, Rio de Janeiro, Brazil

Simulation of an Ir-192 Brachytherapy Source Using Geant4 Code

In brachytherapy cancer treatments, methods for calculation of dose delivered to tumours and organs make extensive use of Monte Carlo calculations. The goal of the treatment planning is the maximization of dose deposition in the target volume, while minimizing the deposition in healthy tissues. The use of interstitial brachytherapy sources follows the recommendations of AAPM REPORT No. 51. The present work is based on Monte Carlo simulations with GEANT4 of the dose quantities recommended by that report. Two different packages of the latter code have been analysed, namely the Standard package and the Low Energy one. An Ir-192 Buchler HDR source was simulated and the geometry of dose deposition in media and primary spectra have been obtained. Results obtained with the Geant4 code have been compared with data obtained using Geant3, a previous version of that code.
N13-96:

The LHCb Muon Detector Alignment

S. Pozzi, S. Vecchi INFN and University of Ferrara, Ferrara, Italy The LHCb experiment, that will operate at the Large Hadron Collider at CERN, is dedicated to the physics of b-hadrons and CP violation. The muon detector provides a high transverse momentum muon trigger at the lowest level and muon identification for the high-level trigger. A correct alignment is fundamental for its purposes. It is composed by 1380 chambers arranged on five walls that can be opened to access the beam pipe. During the installation the chambers were aligned within a precision of 1-2 mm. Using the track reconstruction in the muon detector it is possible to monitor and improve the knowledge of the real position of the chambers. We present the results obtained with the Kalman Filter technique on MonteCarlo simulated data and on real cosmics as well.
N13-97:

N. Vega-Acevedo1, I. Lanas2, F. R. Rannou1 1 Departamento de Ingenieria Informatica, University of Santiago of Chile, (USACH), Chile 2 Departamento de Fisica, University of Santiago of Chile, (USACH), Chile We present a parallel model for GATE simulations in multiprocessor computers. The model is designed to accomodate the passage of time and coincidence processing, which impose limitations on how the parallelization can be carried out. The model consists of a coordinator process, a timer process and a number of worker processes. The coordinator is responsible for assigning events IDs which are unique across all processes. Since events are time-tagged, worker processes pool the timer for the current simulation time. They are also responsible for sending back the updated simulation time, event creation and simulation. The timer is only responsible for keeping the the current simulation time, which is global to all processes. Simulation experiments of a cylinder and a point-source with the OPET scanner were carried out in a 8 core computer. Percentage noise and image resolution are found to be similar for the sequential and parallel case. Experiments show an almost linear speed-up up to six worker processes. It appears that the speed-up is linear as long all worker processes can be hosted in separate cores without the need for processors competition. Output handling follows the same approach as the GATE cluster version, i.e. each worker outputs its own data file. However, the model could also be implemented with a centralized output handling, in which the coordinator is responsible for writing the results. Although the current implementation is process-based, the model is generic enough that can implemented with threads and shared-memory.

GATE Simulations on Multicomputer Architectures

67

N13-98:

Real-Time Multispeckle XPCS Data Reduction via a Field Programmable Gate Array and High Performance Computing

M. Sikorski, P. B. Fernandez, P. R. Jemian, T. J. Madden, S. Narayanan, A. R. Sandy, M. Sprung, B. J. Tieman, J. Weizeorick XSD, Argonne National Laboratory, Lemont, Illinois, USA We describe recent progress on a system that will perform real-time or post-acquisition autocorrelations of large volume (streaming) multi-speckle x-ray photon correlation spectroscopy (XPCS) data sets. The system consists of a fast (>= 30 Hz) direct detection CCD (~ 1 megapixel) interfaced to a frame grabber with a considerable on-board memory and a large field programmable gate array (FPGA). The frame grabber and host computer are connected to a computer cluster capable of reducing the data in the real time. The CCD camera streams data to the frame grabber which performs dark subtraction and thresholding and, eventually, mild compression before streaming data to the cluster where time autocorrelations are performed. The CCD camera and frame grabber are controlled via areaDetector software while a custom Matlab GUI application manages the entire process via EPICS channel access and web service access to a cluster-based service. Ongoing tests demonstrate that the system will be capable of time autocorrelating rapidly streaming XPCS data in real time. Our efforts will advance the general user program at 8-ID. In addition, the framework that we have developed will facilitate other efforts at rapid high volume data reduction and the adoption of even faster XPCS detectors on the near horizon.
N13-99:

Electronics for Monolithic Scintillator PET Detector Modules Based on Neural Network Position Estimators

W. Yonggang, D. Junwei Department of Modern Physics, University of Science and Technolgy of China, Hefei, Anhui, China We are currently developing a prototype of monolithic scintillator PET detector modules based on neural network position estimators towards issues for practical applications. The detector module comprises of a 25.5mm25.5mm10mm LYSO crystal coupled to Hamamatsu 64 channels multi-anode PMT H7546B. Comparing with classical pixelated detectors, possible drawbacks of neural network based detector modules are lower signal to noise ratio and more complicated signal readout and processing. The former needs high dynamic range of signal readout and high resolution digitization, while the latter resorts modern FPGA technology. After the optimization of signal readout geometry, the electronics including a number of digital nuclear signal processing techniques and neural network computing on-line by FPGA was achieved. The validation and performance tests show that the system was well designed and the detector prototype based on this electronics system can be a good platform for future physical experiments.
N13-100:

The Research of the Embedded Data Acquisition and SCADA System of the Gamma Camera Based on ARM

M. Ye, Y. Yang Experimental Physics Center, Institute of High Energy Physics , Academia Sinica, Beijing, CHINA Gamma camera has been widely used in the field of gamma ray detection since 1957. Embedded hardware and software technology have been used in the design and the implementation of the gamma ray data acquisition system. The gamma camera system was composed of the raw data acquisition, image reconstruction, image display, control, communication. It can complete the gamma ray resource detecting. At the first part of the paper the basic principle and the main techniques was introduced during the system design, including embedded system construction, wireless local network configuration and network program. And the system functions are also defined according to the project requirement. The design of the system was divided into two parts: hardware design and software design. The software realization is described in details, including the customization of the embedded Linux system, the Linux device driver migration, the realization of the MiniGUI interface and the sub-module implementation. Finally, the system is tested using simulate event data. The result shows that the whole data acquisition system can obtain the event data. After the event data reconstruction the system can not only realized the image display and process, but also provided the remote control function. The raw event data and image reconstructed will be transmitted to off-line analysis system in order to further data processing and analysis. Key Words: embedded Linux, data acquisition, device driver, gamma ray, MiniGUI, Wi-Fi, USB, wireless protocol,SCADA
N13-101:

Digital Configurable Processor for Acquisition and Elaboration of Data from Detector Arrays

A. Abba, A. Manenti, A. Suardi, C. Fiorini, A. Geraci Dept. of Electronics, Politecnico di Milano, Milan, Italy The paper presents a digital processor based on a configurable computing (FPGA) device for acquisition and programmable data processing in multi-detector systems. Main features of the processor are reconfigurability, linearity, low noise and processing data rate through hardware and firmware architecture solutions that also allow the system to operate in presence of 68

electromagnetic noise, while at the same time minimizing its own radiated emissions. We present the customization of the processor in the framework of the HICAM project, which deals with the development of a compact and high resolution Anger camera to be used in clinical and research environments.
N13-102:

Non-Linear Least-Mean-Squares Fitting in FPGA Devices for Digital Spectroscopy

A. Abba, A. Manenti, A. Suardi, A. Geraci, G. Ripamonti Department of Electronics, Politecnico di Milano, Milan, Italy In measurement tasks, such as the determination of the arrival time of events, severa1 methods based on weighted least-meansquares (LMS) routines have been demonstrated. Nevertheless, the high potential of digital least-mean-squares methods in highresolution nuclear spectroscopy is not fully exploited in practical due to its heavy calculation burden, especially in real-time applications and in spatial computing FPGA device implementation. The paper presents an implementation into a Xilinx Virtex-5 SX50-SG-1 FPGA device of the GaussNewton algorithm that is one of the most used methods to solve non-linear LMS problems.
N13-103:

A. Abba1, A. Manenti1, A. Suardi1, S. Riboldi2, A. Geraci1 1 Department of Electronics, Politecnico di Milano, Milan, Italy 2 Dept. of Physics, Universita' degli Studi di Milano, Milan, Italy

Adaptive Digital Trigger Architecture in FPGA

A fully digital trigger architecture for high-resolution spectroscopy is introduced. The proposed architecture is constituted by an ensemble of modules whose proper combination and initialization addresses at best sometimes conflicting requirements, as signal discrimination, multiple hit detection, finite charge-collection time, drifting baseline level, and timing precision. This can be achieved in extremely wide operative conditions, i.e. signal amplitude, event rate, noise features, disturbances. The architecture has been implemented and experimentally validated into a field programmable gate array (FPGA) device.
N13-104:

Studies of Data Transmission on Long Kapton Cables

V. A. Fadeyev, M. Norgren, P. Manning, J. Nielsen SCIPP, UCSC, Santa Cruz, CA, USA Tracking detectors for ATLAS and CMS experiments will have to be replaced after an LHC luminosity upgrade, due to radiation damage and more stringent performance requirements. The new sub-systems will be larger in scope of instrumented area, and therefore will feature more tightly integrated modules. A basic building block of both strip and pixel sub-systems of the ATLAS Upgrade is stave, a collection of modules with common mechanical support, thermal management system and readout. The data readout is planned to be done via thin Kapton cable which covers the full length of a stave, of up to 1.2 m, and supplies data I/O for each module. The large number of modules placed on the stave, and large track multiplicity at the upgraded detectors necessitate faster readout speed than in present ATLAS detector. We are investigating the performance of the data transmission along the cable at speeds of 160 and 320 Mbps, and how signal integrity depends on the parameters of the system.
N13-105: Scalable Multi-Channel Acquisition System for Radiation Imaging Applications and Nuclear Physics Instrumentation

A. I. Ivanov, V. Y. Pedash, V. A. Kolbasin Institute for scintillation materials NAS of Ukraine, Kharkov, Ukraine Scalable multi-channel acquisition system for radiation imaging applications and nuclear physics instrumentation was developed. It can acquire spectroscopic data and can accept and generate a number of signals with free user editable logic for synchronization. All parameters of acquisition system remain stable in high count rates operation. The core element of the system is 8 channel acquisition boards. Each acquisition board contains 8 free-running ADCs, field-programmed gate array, two CPUs and interfacing circuits. The sampling frequency of the ADCs can be set by user in the range from 5 MHz to 65MHz. It provides accurate pulse sampling for the majority of scintillation detectors. Two types of ADCs can be used in the board: 10- or 12-bit, depending on the required dynamic range and differential nonlinearity. FPGA is used for digital integration of the pulses, baseline restoration and global signals processing (master clock, start of acquisition, etc.). It is also functions as a controller of internal links. Programming of the FPGA can be carried out from post-processing servers via external links, which allows dynamic changing of the system functionality. CPUs perform pulse sorting and rejection. They can also execute user-defined algorithms for advanced pulse processing like calculation of flash coordinates. Designed acquisition system can be used for medical and security imaging and in nuclear physics instrumentation. The maximum number of input analog channels in current design is 256, and it can be increased in future. Acquisition system was successfully tested using conventional gamma camera head with 55 PMTs. 69

This work was sponsored by CRDF, project UAE2-1672-KK-02 N13-106:

Acquisition and Control Hardware and Software for an Advanced Enrichment Monitor

R. F. Parker, J. Goda, T. R. Hill, K. D. Ianakiev, T. Marks, C. E. Moss, H. Nguyen N-4, Los Alamos National Laboratory, Los Alamos, NM, USA Research into the design of an Advanced Enrichment Monitor (AEM), using an x-ray tube rather than a decaying isotope, to measure UF6 enrichment in gas centrifuge plants has led to the development and evolution of a data acquisition and control system consisting of hardware and software components. This paper provides a look into the hardware and software used for research and development of the AEM. The hardware used in the acquisition portion of the system acquires voltage and beam current readbacks from the high voltage supply, pressure, temperature from several points, and current readings from a flux-monitor diode. In addition, a NaI detector and MCA acquire spectra. The control portion uses inputs from the operator as well as data collected by the acquisition system to control the high-voltage supply. Analog and digital inputs and outputs are provided by a high-resolution off-the-shelf data acquisition module chosen based on its analog conversion characteristics, inputs and outputs, and ease of programming. The system software is derived from software used for enrichment measurements, modified to control and acquire data from the new hardware. The software acquires spectra at configurable intervals, calculates enrichment, and stores a wealth of information about the spectrum and other values acquired by the data acquisition module during the interval. The wealth of data allows realtime control feedback as well as analysis of longer-term trends in the data. Additional code allows for research and diagnostic operations such as automatic sweeping of the high voltage and beam current.
N13-107:

F. Ameli1, A. D'Amico2, A. Aloisio3,4, G. Giovanetti2, R. Giordano3,4, L. Capasso3,4, V. Izzo3,4 1 Sez. Roma1, Istituto Nazionale di Fisica Nucleare, Rome, Italy 2 Laboratori Nazionali del SUD, Istituto Nazionale di Fisica Nucleare, Catania, Italy 3 Universita di Napoli Federico II, Napoli, Italy 4 Sez. Napoli, Istituto Nazionale di Fisica Nucleare, Napoli, Italy Dense Wavelength Division Multiplexing (DWDM) networks include active and passive optical components and devices with their specific level of performance in term of data throughput, Bit Error Rate (BER), number of wavelengths, latency, power budget and distortion induced by non-linear optical effects. Therefore, in such enviroments the system architects have to deal with specific issues and requirements tightly binded to the optical transmission layer other than classical digital aspects, like serializer protocols, encoding schemes, error detection and correction codes. In this paper, we present and discuss the performance analysis of a complex DWDM network designed for the data acquisition system of the NEMO underwater neutrino telescope. The tests cover the qualification of the DWDM transponders, fibers, passive components (add-and-drop, optical muxes, circulators) and optical Raman amplifiers. Optical layer simulations are also compared to experimental results for different setups and fiber lengths.
N13-108:

Performance Analysis of a DWDM Optical Transmission System

H. C. Yousefzadeh1, R. Lecomte2, R. Fontaine1 1 Genie electrique et Informatique, Universite de Sherbrooke, Sherbrooke, Quebec, Canada 2 Faculte de Medecine Nuclaire, Universite de Sherbrooke, Fleurimont, Quebec, Canada

Wiener Filter-Based Crystal Identification Applied to Dual and Triple Layers Phoswich Detectors

Various types of fast and high luminosity scintillation crystals have recently been developed for use as detectors in high resolution Positron Emission Tomography (PET) scanners. Phoswich detector modules consisting of different crystal layers or slices coupled to a photodetector can now take advantage of these scintillators for higher pixelisation or depth of interaction (DOI) measurement using crystal identification (CI) techniques. Sophisticated digital signal processing methods based on adaptive filter theory have recently shown very promising results in separating crystals used in LabPET, a small animal PET scanner developed at Universit de Sherbrooke. This paper investigates the feasibility of discriminating up to three different crystals with decay time constants as close as 15 ns. A recently developed LabPET-like pulse generator was used to generate pulses from different phoswich assemblies. The Wiener filter-based CI algorithm was used to discriminate events. Results were compared to experimental data which show that various phoswich designs with two and three crystals can be separated with an error rate of 1.5%-6% for two layer phoswich detectors and 7% for a three layer detector. This study supports the opportunity to upgrade scanners with fast phoswich detectors having suitable crystal identification performance for DOI.

70

N13-109:

H. Kleines1, M. Ramm1, P. Wuestner1, P. Kaemmerling1, M. Mertens2, T. Stockmanns2, F. Huegging2, J. Ritman2 1 ZEL, Forschungszentrum Juelich GmbH, Juelich, Germany 2 IKP, Forschungszentrum Juelich GmbH, Juelich, Germany The experiment PANDA (Antiproton Annihilation at Darmstadt) is under development for the future accelerator facility FAIR in Darmstadt, Germany. In the target spectrometer of PANDA a micro-vertex detector (MVD) will be used as the central tracking detector for charged particles. The design of the MVD is based on silicon strip detectors at the outer layer and on silicon pixel detectors at the inner layers. The paper discusses the requirements and first concepts of the DAQ system for the MVD which has to be finished about 2013. It presents the design of the lab system that has been developed already now for the test and readout of different frontend chips for pixel detectors. The lab system consists of a flexible readout board coupled via the SIS1100 Gigabit link to PCI. A further improvement of the lab system employing a 2.5 Gbit/s optical link to PCIe is introduced.
N13-110:

Developments for the Readout of the PANDA Micro Vertex Detector

A Telescope Using CMS PSI46 Pixels and the CAPTAN for Acquisition and Control over Gigabit Ethernet

R. A. Rivera ESE/CD, Fermilab, Batavia, Illinois, USA The Electronic Systems Engineering department of the Computing Division at the Fermi National Accelerator Laboratory is assembling a pixel test beam telescope for high energy physics detector research and development. The telescope features CMS PSI46 readout chips and a data acquisition and control system known as the Compact And Programmable daTa Acquisition Node or CAPTAN. The CAPTAN is a flexible and powerful system that meets the readout and control demands of a variety of pixel and strip detectors for high energy physics applications. The CAPTAN functions in a gigabit Ethernet network, which facilitates the coordination of the multiple pixel planes of the telescope. Through the use of the CAPTAN hardware, a unified telescope system is attained encompassing both the CMS PSI46 pixel components and a device under test. This paper discusses results from the telescope project including mechanical design, alignment procedure, and attainable precision.
N13-111:

The Muon Conditions Data Management: Database Architecture and Software Infrastructure

M. Verducci University of Wuerzburg, Wuerzburg, Germany On behalf of the ATLAS Muon Collaboration The management of the Muon Conditions Database will be one of the most challenging applications for Muon System, both in terms of data volumes and rates, but also in terms of the variety of data stored and their analysis. The Muon conditions database is responsible for almost all of the 'non-event' data and detector quality flags storage needed for debugging of the detector operations and for performing the reconstruction and the analysis. In particular for the early data, the knowledge of the detector performance, the corrections in term of efficiency and calibration will be extremely important for the correct reconstruction of the events. In this work, an overview of the entire Muon conditions database architecture is given, in particular the different sources of the data and the storage model used, including the database technology associated. Particular emphasis is given to the Data Quality chain: the flow of the data, the analysis and the final results are described. In addiction, the description of the software interfaces used to access to the conditions data are reported, in particular, in the Offline Reconstruction framework ATHENA environment.
N13-112:

High Resolution Digital Spectroscopy Based on Multiple Interleaved ADCs

A. Abba, M. Binda, A. Manenti, A. Suardi, A. Geraci Department of Electronics, Dipartimento di Elettronica e Informazione - Politecnico di Milano University, Milano, Italy Interleaved sampling is a well known technique for achieving a sampling speed by the synergy of more than one ADC clocked at frequency lower than the sampling one. This allows significant reduction of power consumption and increasing of achievable resolution with respect to the use of a single ADC. Main problem of interleaved ADC structures is the definition of a precise and stable relationship among sampling combs of the different ADCs. While consolidated integrated solutions exist for interleaved ADCs mismatches, at off-chip system level the problem is still open. We propose an active mechanism of tracking and maintaining the correct relative position of the sampling combs of the ADCs by means of a control architecture implemented in a field programmable gate array (FPGA) device.

71

A. Valero1, V. Castillo1, A. Ferrer1, V. Gonzalez2, E. Higon1, J. Poveda1, B. Salvachua1, E. Sanchis2, C. Solans1, J. Torres2, J. Valls1 1 IFIC - (CSIC & Universidad de Valencia), Valencia, Spain 2 Ingenieria Electronica, Universidad de Valencia, Valencia, Spain TileCal is the hadronic calorimeter of the ATLAS experiment at the LHC collider at CERN. The Read-Out Drivers (ROD) are the core of the off-detector electronics. The main components of the RODs are the Digital Signal Processor (DSP) placed on the Processing Unit (PU) dautherboards. This paper describes the DSP code and its performance with calibration and real data. The code is divided into two different parts: the first part contains the core functionalities and the second one the reconstruction algorithms. The core acts as an operating system and it controls the configuration, the data reception, transmission, online monitoring and the synchronization between front-end data and the Trigger information. The reconstruction algorithms implemented on the DSP are the Optimal Filtering (OF), Muon Tagging (MTag) and Missing ET (MET) calculation. The OF algorithm reconstructs the deposited energy and the arrival time of the data on every calorimeter cell within a front-end module. This reconstructed energy is used by the MTag algorithm to tag low transverse momentum muons that may escape the ATLAS muon spectrometer Level 1 trigger whereas the MET algorithm computes the total transverse energy and the projection on X and Y for the entire module that will be used by the Level 2 trigger system. The DSP code performance is being validated with comparisons with offline reconstruction. The DSP performance is evaluated using calibration data from Charge Injection and Laser systems. In addition, cosmics rays data taking and reconstruction has been a part of the ATLAS detector commissioning and it has been also used for the DSP performance validation. The performance validation results are presented for each type of data including first beam data.
N13-114:

N13-113:

The ATLAS TileCal Read-Out Drivers Signal Reconstruction with Commissioning Data

Gigabit High Speed TCP Data Transmission for Positron Emission Tomography

B. Humbert, P. Bard, C. Fuchs, D. Brasse IPHC, Universite de Strasbourg, CNRS / IN2P3, Strasbourg, France AMISSA platform is a multimodality imaging system for small animal developed in our institute and combining microCT, microSPECT and microPET devices on the same gantry. The CT/SPECT components are already operational and used for biological experiments. The preclinical PET system is actually under development and is composed of four detection modules arranged around the animal. Each module consists in 768 LYSO:Ce crystals read at both ends by Micro-Channel Plates (MCP) Photo-Multiplier Tube (PMT) (Planacon XP85023/A1, Photonis Corp.). The particular geometry combined with an inner diameter of 61.2 mm lead to a detection efficiency greater than 15 % with a volumetric spatial resolution of 1 l. Six electronic cards are integrated to collect the data from one module of 2 x 768 channels. The data are then sent to a set of PCs. Monte Carlo simulations show a maximum estimated sustained transfer rate of 200 MiB/s (mebibytes per second : 220 bytes/s) from one module to PCs. The cheaper and more long lasting solution to transfer data at such rates must clearly involve standards such as Gigabit Ethernet. New FPGAs include processors which give the opportunity to use them in compact electronic card offering Ethernet communication. This paper focuses on our proposed solution to transfer data at high rate with a Gigabit Ethernet link. Between acquisition card and destination PC very simple and efficient software is developed to perform a socket to socket TCP/IP transmission. With a direct connection between an emitter board and a receiver PC, the 111.75 MiB/s rate is obtained with up to 23.8 % of CPU available. The proposed solution fulfils completely our present needs. It depends only on the Virtex-5 FXT choice and the availability of associated IPs. The entire knowledge at all level inherent in this solution, without neither Operating System nor TCP/IP stacks makes it immediately available for 10 Gigabits Ethernet.
N13-115:

The CMS Tracker Data Quality Monitoring Expert GUI

F. Palmonari INFN, Pisa, Italy On behalf of the CMS Collaboration The CMS Tracker data quality monitoring (DQM) is a demanding task due to the high granularity of the detector. The Tracker consists of 15148 strip and 1440 pixel modules corresponding to about 350,000 histograms defined and filled accessing information from different stages of data reconstruction to check the data quality. In order to manage such a large number of histograms, a tracker specific Graphical User Interface (GUI) was developed to simplify the navigation and to spot detector problems efficiently. The GUI is web-based and implemented with Ajax technology. A description of the framework and the specific features of the expert GUI developed for the CMS Tracker DQM system will be presented.

72

N13-116:

D. Falchieri1, F. Costa2, E. Gandolfi1, F. Prino3, M. Sitta3, L. Toscano2, S. Zannoli1 1 Department of Physics & INFN, Universita' di Bologna / INFN Bologna, Bologna, Italy 2 CERN, Geneve, Switzerland 3 INFN Torino, Torino, Italy

On-Line Data Reduction on the ALICE SDD Data Concentrator Card CARLOSrx

CARLOSrx is a FPGA based board used in the data acquisition chain for the Silicon Drift Detector (SDD), which is part of the Inner Tracking System of the ALICE experiment at CERN. Each CARLOSrx board is a 9U x 400 mm VME64x printed circuit board able to receive data coming from 12 SDD modules and send them to the standard data acquisition system (DAQ) after performing event building. During the ALICE shutdown phase of winter 2008/09, the firmware has been upgraded to allow a dramatic data-size reduction from 520 MBytes/s to 25 MBytes/s for all the 260 SDD detectors running at 1 kHz with almost empty events. In order to reach such a reduction, the end of row summary words containing information on the noise distribution has to be identified and removed and a new encoding algorithm has to be adopted. Because the data received from the front-end electronics is in variable length format, the goal has been accomplished using logarithmic combinational barrel shifters able to decode 12 input data streams at the same time and recode them using a fixed 32-bit format.
N13-117:

The ARGO-YBJ DAQ Monitoring System

S. Mastroianni Sezione di Napoli, INFN, Napoli, Italy On behalf of the ARGO-YBJ Collaboration The ARGO-YBJ detector has been completely installed at the Yangbajing Cosmic Ray Laboratory (Tibet, China), at 4300 m a.s.l., and is taking data with a trigger rate of about 4 kHz and a dead time of about 4%. The ARGO-YBJ Data Acquisition System is based on a two-layer readout architecture implementing an event-driven data collection. In order to sustain a high data throughput and have a good readout efficiency or keep the dead time as low as possible, the decoupling FIFOs have been housed on the data packet processing modules. In this paper, the hardware solution to monitor the DAQ system and to track in real-time anomalous data transfers is presented.
N13-118:

Data Quality Monitoring of the CMS Tracker

V. Adler IIHE, Vrije Universiteit Brussel, Brussels, Belgium On behalf of the CMS Tracker DQM group The Data Quality Monitoring (DQM) system of the CMS experiment has been developed within the CMS Software framework. It has been designed to be used during online data taking as well as during offline reconstruction. The goal of the online system is to monitor detector performance and identify problems very efficiently during data collection so that proper actions can be taken. On the other hand the reconstruction or calibration problems can be detected during offline processing using the same tools. The monitoring is performed with histograms. In the online system, the Tracker is monitored at full granularity (15148 Silicon Strip modules plus 1440 Pixel modules), but only on a subset of the data (~5Hz). On the other hand in the offline system the full statistics, calibration and final reconstruction results are available, but computing limitations restrict the monitoring to higher detector levels. Applications are in place to further process the information from these basic histograms by summarizing them in overview plots, by evaluating them with automated statistical tests, and by extracting their main qualities and filling them into trend plots, which monitor the behaviour of the detector over time. In particular we monitor readout errors, raw charge deposition information, as well as reconstructed hits and even tracks. Thus, we are able to detect, both, with fast turn-around (online), as well as high precision (offline), data corruption, mis-configuration and mis-calibration of the detector, as well as newly broken modules and dead or noisy strips and pixels. All histograms can be displayed in both, the central CMS DQM graphical user interface (GUI), as well as in Tracker expert GUI and in so-called Tracker Maps. We describe the CMS Data Quality Monitoring system of the Silicon Strip and Pixel Tracker of CMS, as well as first experience from the CMS cosmic ray data taking period in 2008.
N13-119:

Development of 500 MHz Multi-Channel Readout Electronics for Fast Radiation Detectors

W. Hennig, W. K. Warburton, S. Asztalos, D. Breus XIA, LLC, Hayward, CA, USA In this paper, we describe the development of readout electronics for fast radiation detectors that digitize signals at a rate of 500 MHz, process the digital data stream to measure pulse heights, bin the result in on-board MCA spectra, and optionally capture waveforms for pulse shape analysis. The electronics are targeted for applications requiring good energy resolution and precise 73

timing, for example life time measurements on exotic nuclei, timing measurements with fast scintillators such as LaBr3 or BaF2, or pulse shape analysis with liquid scintillators or phoswich detectors. Upgrading the existing XIA Pixie-4 spectrometer design with a 12-bit, 500 MHz analog to digital converter (ADC), we built a prototype of a 4-channel electronics module and evaluated its performance in terms of energy resolution, linearity of the ADC, timing resolution, and improvements in pulse shape analysis. Preliminary results are: energy resolutions of ~3% at 662 keV with a LaBr3 detector (matching existing electronics) and ~2.25 keV at 1.3 MeV with a germanium detector, and timing resolutions of ~20 ps with a pulser and ~53 ps with LaBr3. We will discuss issues limiting energy resolution and tradeoffs associated with the use of a 14-bit, 400 MHz ADC.
N13-120:

Readout Electronics for the FVTX Detector at PHENIX

S. A. Butsyk Physics and Astronomy, University of New Mexico, Albuquerque, NM, US On behalf of the PHENIX Collaboration The PHENIX experiment at RHIC at Brookhaven National Laboratory has been providing high quality physics data for over 8 years. The current PHENIX physics program will be significantly enhanced by addition of the Forward Silicon Vertex upgrade detector (FVTX) in the acceptance of existing muon arm detectors. The proposed tracker is planned to be put into operation in FY2011. Each arm of the FVTX detector consist of 4 discs of silicon strip sensors combined with FPHX readout chips, designed at FNAL. Full detector consists of over 1 Million active mini-strip channels with instantaneous bandwidth topping 3.4 Tb/s. The FPHX chip utilizes data push architecture with 2 serial output streams at 200 MHz. The readout electronics design require a split of DAQ into two separate pieces: 1) Read-Out Cards (ROC) located in the vicinity of the detector, 2) Front End Modules (FEM) located in the Counting House. ROC boards, based on the rad-tolerant ACTEL ProAsic3E FPGAs, combine the data from several chips, synchronizes data streams and send them to FEM over a Fiber Optics Link. The data are being buffered in the FEM (built on Xilinx Virtex-4 large-scale FPGAs) for 64 consecutive beam crossings and data from appropriate beam bucket are sent to a standard PHENIX DAQ interface upon Level-1 trigger request. I am going to present the current status of the readout electronics development and testing.
N13-121:

C. C. W. Robson1, C. Bohm1, K. Rehlich2, R. Kammering2 1 Department of Physics, Stockholms universitet, Stockholm, Sweden 2 Deutsches Elektronen-Synchrotron, Hamburg, Germany

Integrating Security into an Accelerator Control Systems Web Interface

Embedded devices for control and monitoring are becoming more powerful as technology evolves. It is common that these embedded devices are communicating through a local area network with other, sometimes more powerful, nodes for receiving and distributing commands and data gathering. This distributed system can then be connected to other systems over the Internet for data exchange, remote control and monitoring. It is essential in such complex environment that the users are authorized and authenticated as users of the system, and that these security credentials can be easily managed. But this is not enough since not all commands are suitable to be executed in every status of a complex system and by all users from all locations. The wrong command pattern or security policy could potentially cause great damage to the system being controlled. We are developing a system allowing remote monitoring and control of the European XFEL projects accelerators control system DOOCS (see doocs.desy.de). This remote system uses a service oriented architecture based on web services for communication between its nodes. The interface to the underlying classical accelerator control system will be implemented on a central server, whereby any remote access can be audited and controlled. To meet the mentioned requirements for security and safe control patterns, we found it necessary to add a security layer. We decided to implement the OpenSSO package from Sun in our distributed system for authentication and authorization, to make certain that only authorized users can execute commands according to the allowed policy, based on user credentials, location and system status. This paper presents our implementation of OpenSSO and our experiences from the implementation, including development cost, execution overhead and overall impact on the control system as well the impact on integration and communication with the underlying DOOCS layer.
N13-122:

A Modular High Speed Data Acquisition System for PHENIX TEC/TRD Front End Electronics

M. A. L. Leite Instituto de Fisica, Universidade de Sao Paulo, Sao Paulo, SP, Brazil On behalf of the PHENIX TEC/TRD group A modular data acquisition system was developed for the readout of a small subset of the PHENIX TEC/TRD detector front end electronics. In the full configuration, this PHENIX sub-system deploys more than 21 thousand channels from 48 multi-wire chambers. The signal from each wire is digitally sampled at 40 MHz, and groups of 128 channel waveforms of 80 samples each are serialized and sent to the back-end electronics using 1 Gbit/s optical links. Although the experiment full data acquisition 74

system has the flexibility to handle a subset of the full detector, its configuration and operation are quite complex, and can be only performed in systems located in the experimental facility. The necessity to use a mock-up of the front-end electronics under controlled conditions to perform the optimization of the detector readout chain and to provide an easy to use platform for hardware maintenance has dictated the need for a simple and high performance data acquisition, timing and control system. To accomplish this task, a back-end was designed using commodity devices and interfaces, such as Ethernet and High-speed USB. While it was necessary to redesign the slow control network to use a CAN-BUS protocol, a transition layer to the more common Ethernet network is provided by the hardware. An ARM9 micro-controller is responsible for the high level implementation ("bare-metal") of the information flow control, while the data transfer itself is handled by dedicated logic implemented using FPGAs and by a High Speed USB peripheral circuit. An HTTP server running inside the micro-controller provides easy access to all features of the system, and is further enhanced by a higher level framework on the computer side. A choice of high-speed FIFO memory or static RAM is used to buffer several events before the transfer to the computer. The micro-controller, memory and interfaces are built on daughter cards, allowing expansion and evaluation of newer devices or different protocols.
N13-123:

T. Konno1, M. Kuze1, M. Ishitsuka1, Y. Sakamoto2, Y. Nagasaka3, A. Sigemori4, A. Cabrera5 1 Tokyo Institute of Technology, Tokyo, Japan 2 Tohoku gakuin university, Sendai, Japan 3 Hiroshima institute of technology, Hiroshima, Japan 4 Niigata university, Niigata, Japan 5 CNRS/IN2P3 -APC Laboratory (Paris), Paris, France "Double Chooz" is a reactor-neutrino oscillation experiment in France to search for the last unknown neutrino mixing angle theta_13. The detector consists of 400 10-inch PMTs and other sub-detectors, and data-acquisition uses newly developed 500MHz FADCs and other distributed systems. We have developed an Online monitoring system for the whole experiment to monitor the status of data-acquisition, to perform fast data-quality monitoring using offline algorithms, and also to monitor slowcontrol variables such as power supply currents. In the DC monitoring system, common three data servers corresponding to a refresh rate of these data sources are used. According to the each data server manages a shared memory and mutex semaphores individually, the monitor server can control all data flow performances among a data source process and the GUIs (monitor viewers). That makes the data source process is able to work well independently of a performance of the monitor viewers. For GUIs, two kinds of schemes based on browser technologies and Java APIs are prepared, the former GUI adopts Ajax, Canvas and Google web toolkit, and the latter is built by using Java Swing. Those GUIs can show various histograms without the ROOT library and achieve enough performance because the GUIs receive only contents of histograms and neither receives pictures of the histograms nor analyzes there. Web browser techniques, especially Ajax and Java servlet, enable DC shift person access experiment information on web pages, because the servlet on a web server can communicate with the monitor server directly and derive the data from there, and the data are drawn by a client side browser itself. On the other hand, a GUI based on Java Swing provides a direct communication scheme to the data server and that achieves enough performance even if dealing with large size of data in one second. We have tested these monitoring systems on DC prototype DAQ and will show the results of the performance test.
N13-124:

Online Monitoring System for Double Chooz Experiment

J. Stelzer1, P. Bell2 1 DESY, Hamburg, Germany 2 Manchester University, Manchester, Great Britain

The Configuration System of the ATLAS Trigger

The ATLAS detector at CERN's LHC will be exposed to proton-proton collisions at a rate of 40 MHz. To reduce the data rate to a manageable level, only potentially interesting events are selected by a three-level trigger system. The first level is implemented in firmware, reducing the data output rate to about 100 kHz, with the second and third levels being software triggers with a final output rate of about 200 Hz. A system has been designed and implemented that enables the configuration of all three trigger levels from a centrally maintained relational database, for the purpose of both online data taking and offline trigger simulation. We present the current status of this trigger configuration system configuration system, covering the database design, client software and user interface tools, and putting emphasis on its multiple uses for data-taking, reconstruction and Monte-Carlo simulation.

75

N13-125:

L. M. D. Owsley1, J. J. McLaughlin1, L. Cazzanti1, S. Salaymeh2 1 Applied Physics Laboratory, University of Washington, Seattle, WA, 98105 2 Nonproliferation Technologies Section, Savannah River National Laboratory, Aiken, SC, 29808

A Novel Approach Using Speech Technology to Enhance Isotope ID and Classification

We are investigating whether the problem of automatic isotope identification based on gamma-ray spectroscopy can be informed by techniques and features which have proven valuable in the area of automatic speech processing. A cepstral feature representation is at the heart of the majority of modern speech and speaker recognition systems, and we believe the analogies to the spectroscopy application are strong. Channel-robustness techniques have been developed to minimize the effects of speech recording devices and environments, as well as to separate information about who is speaking from what is being spoken; we believe these may be useful in compensating for the effects of shielding and detector specifics on radiation spectra. Further, our understanding of the human ear's frequency-based sensitivity has lead to an approach to processing that is relevant for radiation detectors that have energy-dependent resolution. We have additionally studied an information-theoretic approach to spectraldifference measurement that is valuable in speaker identification but which we believe is robust to the types of variation seen among samples of a particular isotope collected in different conditions. The result is improved performance in the face of spectral differences which would normally require a high-complexity algorithm to model.
N13-126:

D. J. Lingenfelter1, J. A. Fessler1, C. D. Scott1, Z. He2 1 Dept. of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI, USA 2 Dept. of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA

Benefits of Position-Sensitive Detectors for Source Detection with Known Background

We address the question of whether or not the directional or imaging information offered by a position-sensitive Compton detector improves the detection accuracy when searching for a source of known shape amid a background of known intensity. We formulate the detection problem as a composite hypothesis testing problem and examine the behavior of the generalized likelihood ratio test (GLRT) in terms of the area under the receiver operating characteristic curve (AUC). Due to the analytical complexity of the GLRT in this case, we examine its asymptotic properties when the number of detected photons is large. We find that a detector of uniform sensitivity can more accurately detect a source when imaging information is used. In the case of a detector of nonuniform sensitivity, we provide simulation data to show how detector shape and source position affect detection performance.
This work was supported by the U.S. Department of Homeland Security - DNDO under Grant Award Number 2008-DN-077-ARI007-02 N13-127:

A Compact Compton Camera Using Scintillators for the Investigation of Nuclear Materials

W. Lee, T. Lee Radiologic Science, Korea University, Seoul, Korea Due to its availability to reconstruct high energy (up to several MeV) radiation distribution without using thick mechanical collimation, Compton camera has been used in many application such as astrophysics, industrial survey, homeland security and medical purpose. Whereas the performance of mechanical collimation decreases with radiation energy, Compton camera is effective for detecting the higher energy radiation. Even though scintillation detectors show large efficiency, high timing resolution and usability without cooling device, their relatively low energy resolution and bulky size of conventional photomultiplier tubes limit the scope of its applications to gamma ray imaging. To overcome the limitation, we constructed a Compton camera which combined cerium doped lanthanum chloride (LaCl3(Ce)) and position sensitive photomultiplier tube (PSPMT) with multi anodes connected with custom made circuits. The whole size of each detector is only about 5x5x10cm, and therefore, it can be carried in a hand to investigating nuclear materials in the field such as cargo or nuclear waste reservoir. Moreover, the high energy resolution of LaCl3(Ce) increases selectivity for various radiation energy sources. The custom made circuit, which can correct non-uniformity of the signals from the PSPMT, contributes to improve the compactness and robustness of the detector system. The design and performance of the detector system are presented.

76

N13-128:

C. Fiorini1,2, A. Marone1, P. Baraldi3, F. Cadini3, E. Zio3, F. Camera2,4 1 DEI, Politecnico di Milano, Milan, Italy 2 Sezione di Milano, INFN, Milan, Italy 3 Department of Energy, Politecnico di Milano, Milan, Italy 4 Dipartimento di Fisica, Universita' degli Studi di Milano, Milano, Italy

Employing Neural Networks to Determine the Position of Interaction of Medium-High Energy Gamma Rays

In this paper the Neural Networks (NNs) are used to analyze the gamma-ray interactions in a scintillation crystal coupled with Silicon Drift Detectors (SDDs) for the energy range of 100keV-1.1MeV. Among the interactions, the Compton effect is more relevant than the Photoelectric one and the Pair Production and as a result there is a great number of multiple interactions to be evaluated. In this framework, NNs show to be a valuable tool for satisfactorily modeling the detection system and to allow the reconstruction of the interaction event in the scintillator. Considering for instance a 1.1MeV gamma ray, the first point of interaction is located with our method with a precision of approximately 2mm for the coordinates X and Y, while the coordinate Z is determined with precision of about 2.9mm. The possibility to obtain the position of the second point of interaction is then examined throughout a specific case study. Finally, to test the efficiency and the performances of the NN created, other analysis methods are compared and experimental data are also processed.
N13-129:

M. P. Jones1, A. J. Boston1, H. C. Boston1, R. J. Cooper1, M. R. Dimmock1, A. N. Grint1, L. J. Harkness1, D. S. Judson1, J. McGrath1, P. J. Nolan1, J. R. Cresswell1, D. C. Oxley1, B. Pietras1, M. J. Joyce2, R. Mackin2, M. Aspinall2, A. Payton3, R. Van Silfhout3, G. Zhang3 1 Dept of Physics, University of Liverpool, Liverpool, UK 2 Department of Engineering, University of Lancaster, Lancaster, UK 3 School of Electronic and Electrical Engineering, University of Manchester, Manchester, UK The Distinguish collaboration is developing a technique capable of detecting and imaging hidden illicit substances such as explosives or narcotics in luggage and vehicles in transit [1]. To this end there is a requirement for a detection technique that is highly sensitive and highly specific. Pulsed Fast Neutron Analysis (PFNA) techniques [2] are used to stimulate the emission of characteristic gamma-rays, leading to the determination of the concentrations of the light elements (Oxygen 6.13MeV, Carbon 4.43MeV) that are normally used as primary components of explosive materials. This work is based on the Compton Camera principle [3] which aims to produce a 3D image of sources located in space. Compton Camera measurements are to be undertaken using two planar High purity Germanium (Ge) detectors. The characteristic fingerprint generated by the emitted gamma ray will be identified by the spectroscopic qualities of the detectors and the location of the emittance will be resolved using Compton kinematics. The Geant4 simulation toolkit is being used to optimise the detector setup and produce simulated Compton images for a direct comparison between simulated and experimental data in the energy range of 0.2 2.0MeV. The principles of the experiment, the Geant4 simulated Compton images, initial experimental results and the status of the project will be discussed. [1] M. Farahmand et al. Detection of explosive substances by tomographic inspection using neutron and gamma-ray spectroscopy, NIMB 261 (2007) 396-400 [2] Gozani, T., "Novel applications of fast neutron interrogation methods", Nucl. Instr. and Meth. A. 353 (1994) 635 [3] D. B. Everett, J. S. Fleming, R. W. Todd, and J. M. Nightingale, Gamma-radiation imaging system based on the Compton effect, Proc. ZEE, Vol. 124, pp. 995-1000, 1977.
N13-130:

The Development of a Compton Camera for the Imaging of Illicit Substances.

Statistical Methods for Chemical Compound Identification from Neutron-Induced Gamma-Ray

J. Noh1, R. C. Stevenson1, M. D. Hammig2 1 Galt LLC, Ann Arbor, Michigan, USA 2 Nuclear engineering and radiological science, University of Michigan, Ann Arbor, Michigan, USA In order to quickly, precisely, and non-destructively map the chemical compounds in the environment, one can employ a 14 MeV deuterium-tritium (D-T) neutron source for inducing gamma-ray emission via inelastic neutron scattering. If one is endeavoring to identify particular organic compounds - for instance, narcotics, conventional high explosives (HEs), or other contraband secreted from standard optical sensing methods - then one of the largest challenges is to separate the rare target of interest from the ubiquitous non-targets in the environment. In this paper, we investigate the performance of several statistical methods with which one can confidently distinguish compound chemicals of interest from other ordinary materials. As a simple approach, one can employ a standard correlation-based scheme to identify organic materials of interest from their gamma-ray spectra. For greater sophistication, one can exploit density mixture models such as the Gaussian mixture model (GMM) and the probabilistic principal component analysis (PPCA) mixture model in order to more precisely extract the unique spectral characteristics of 77

Spectra

compound materials of interest. With both simulated and empirical data, we quantify the success rates with which one can identify the HE compounds for even poorly sampled spectra - that is, for spectra with a limited number of counts and/or characterized by a poor energy-resolution detector - using various statistical techniques.
N13-131:

E. D'Abramo1,2, V. O. Cencelli1,2, F. de Notaristefani1,2, F. Petulla'1,3, A. Fabbri1,2, R. Pani3,4, G. Moschini5, F. L. Navarria6 1 Dep. of Electronic Engineering, University of Rome Roma Tre, Rome, Italy 2 INFN Sezione Roma III, Rome, Italy 3 University La Sapienza, Rome, Italy 4 INFN Sezione Roma I, Rome, Italy 5 INFN Sezione Bologna, Bologna, Italy 6 INFN Laboratori Nazionali di Legnaro, Legnaro, Italy In a preceding work a new compact gamma camera based on the Hamamatsu H8500 PSPMT and on a custom developed electronics was presented [1]. the core element of such electronics is a microcontroller that operates the transfer of the sampled data to a PC using a USB link in full speed modality; This transfer utilizes on one side a firmware that controls the operation performed by the microcontroller and on the other side a software that trough a graphical interface make the analysis of the acquired data possible. Both were optimized considering the nature of the data coming from the gamma camera and on the other the limitation imposed by the USB protocol. Although they were conceived for a particular gamma camera the requirements and optimizations realized still have a general interest for any position sensitive based gamma camera.
N13-132:

Optimized Software for a Compact Gamma Camera

B. Budden, G. L. Case, M. L. Cherry Dept. of Physics & Astronomy, Louisiana State University, Baton Rouge, LA, USA A gamma ray imager based on rotation modulation is capable of obtaining significantly better angular resolution than the fundamental geometric resolution given by the ratio of detector diameter to mask-detector separation. Compared to a standard coded aperture or Compton telescope, a rotational modulator (RM) can accomplish this with less need for fine position resolution and consequently less complexity. A prototype of this concept developed at LSU features high sensitivity and energy resolution, angular resolution of 15', and a simple readout system. The instrument consists of a single grid of 1.5 wide lead slats separated by 1.5 and spaced ~1.2 m above an array of 19 1.5 x 1 thick LaBr3:Ce detectors in a concentric circular layout. As the grid rotates, the transmission from a source is modulated on each detector between 0 and 100%. This count profile is cross-correlated with pre-calculated modulation profiles to produce an initial source image. Further processing of the image with a cleaning technique that incorporates information from the point-spread function can accurately resolve point sources. We present our reconstruction technique, cleaning algorithms, and imaging and spectral results for the RM prototype.
This work was supported by the Dept. of Energy National Nuclear Security Administration and NASA/Louisiana Board of Regents Space Grant Consortium. N13-133:

Imaging Results with a LaBr3-Based Rotational Modulator

B. T. Wells1, R. C. Stevenson1, J. Noh1, S. Ramadoss1, M. Joseph1, M. D. Hammig2 1 Galt Research LLC, Ann Arbor, MI, USA 2 Department of Nuclear Engineering and Rad. Sci., University of Michigan, Ann Arbor, MI, USA

Defeating IEDs, SNM and Contraband Secreting via Long Range Gamma-Ray Imaging of Neutron Interrogated Materials

The continued development of a depth sensitive gamma camera, designed to image neutron- interrogated materials in a large gamma background is reported. Bar detectors, comprised of inorganic scintillators, are used to measure the interaction locations of gamma-rays produced by the inelastic scattering of 14 MeV neutrons with the target material. Scanning a wide field of view and sensing target objects at long range presents a challenge because of the increased competition between counts returned from the targets of interest, which greatly diminish in number, and unwanted interactions that occur in the surrounding environment, which increase on a relative basis. Additional design challenges consist of dose exposure limitations to by-standers and operating personnel, near identical competing spectra to the targets spectra, high event rates and pulse pile-up issues, the physics limited aspects of detector materials, computational time limitations, and Doppler broadening of the interrogation neutron energy. In the following the preamplification electronic design and the downstream electronic chain are described for the partitioned detector. The angular resolution, which is limited by both energy and spatial-resolution limitations, is measured for imagers based on both NaI(Tl) and BaF2 scintillators, which exhibit complementary performance with regard to the precision with which they extract either the position-of-interaction or measure the deposited energy. The relative performance of the two tiers of the angular reconstruction algorithm - the simple back projection and the maximum likelihood (ML) based technique that is optimal but computationally expensive, is also presented. 78

N13-134:

Directional Isotope Identification Using 3-D Semiconductor Gamma-Ray-Imaging Spectrometers

C. L. Thrall, C. G. Wahl, Z. He Nuclear Engineering, Univeristy of Michigan, Ann Arbor, MI, USA Using three-dimensional-position-sensitive semiconductor detectors, our research group has demonstrated the ability to record multiple interaction locations and associated energy depositions due to incident gamma-rays in a pixellated CdZnTe (CZT) detector. These interaction locations and energies can be used to reconstruct a 4 image of radiation sources around the detector. We have previously demonstrated gamma-ray imaging techniques which provide an estimate of the incident spectrum in each direction. The performance of identification for two different simultaneous sources in different directions will be shown as a function of source intensity. Identification is unique for each direction and source.
N13-135: Implementation of Signal Decomposition for Compton Imaging Using High-Purity Germanium Detectors

J. Rohel, L. Mihailescu, K. Vetter, D. Chivers, J. Maltz, G. Gullberg Lawrence Berkeley National Laboratory, Berkeley, CA, United States A critical requirement for Compton-scatter based gamma-ray imaging is the ability to measure three-dimensional positions and energies of individual gamma-ray interactions with high efficiency and granularity. The latter refers to the ability to separate and localize interactions with high accuracy. So as to obtain this information, we combine Double-Sided Segmented (DSSD) HPGe Detectors with pulse-shape analysis. In order to improve the granularity of this system beyond the strip size, we have developed signal decomposition algorithms that allow us to separate and measure the positions even of closely located interactions within the same or adjacent strips. The signal decomposition algorithm is based on a comparison of measured signals and calculated signals, stored in a data base. Details of the electrical fields, the charge transport, and the electronic response is taken into account to properly parameterize the signals. Preliminary results indicate a significant improvement in the measurement of closely located interactions resulting in a substantial increase in sensitivity in HPGe detectors, particularly for lower gamma-ray energies. In this paper, we report on the approach and the results we have obtained for two HPGe DSSD based systems.
This material is based upon work supported by the U.S. Department of Homeland Security. N13-136:

Triggered Readout for CCD Imaging of Compton Electron Tracks

K. L. Macdonald, D. H. Chivers Nuclear Engineering, UC Berkeley, Berkeely, CA, USA Measuring position and energy information for Compton interactions between a gamma ray and a detector allows the position of the source to be extrapolated to within a back-projected cone. If trajectory information about the electron scattered by the first Compton interaction is incorporated, then the cone can be reduced to a cone segment. We use a charge coupled device to record both the first Compton interaction and the scattered electron track. However, timing information is necessary in order to correlate these data with data recorded by a second detector. Fabricating strip electrodes on the back of the charge coupled device will allow us to get a trigger signal whenever an interaction has taken place. We will use this trigger data to coordinate readout of the charge coupled device, and incorporate the necessary timing information.
N13-137:

B. C. Plimley1, A. Coffer1, B. Pourhamzeh2, A. Popkow2, T. Aucott2, D. Chivers1, K. Vetter1,3 1 Nuclear Engineering, UC Berkeley, Berkeley, CA, USA 2 Physics, UC Berkeley, Berkeley, CA, USA 3 Lawrence Berkeley National Laboratory, Berkeley, CA, USA Detection of the gamma ray signatures emitted by different radioisotopes is known to be an effective means of identifying nuclear materials. However, natural and man-made background radiation limits a detectors sensitivity, particularly when background fluctuations occur. Compton imaging has been demonstrated to improve detection sensitivity, as well as provide position information about the location and extent of a source, which can aid decision makers in applications such as homeland security. Multiple interactions by a gamma ray in a position-sensitive detector may be used to back-project the incident direction of the gamma ray. The symmetry of the measurement limits the precision to a cone in image space. This symmetry may only be broken if the direction of the scattered Compton electron is measured, in which case conservation of momentum narrows the backprojected cone to a segment of a cone, with an arc length determined by the uncertainty in electron trajectory. A silicon CCD detector is suitable for measuring electron tracks for several reasons. CCDs can have excellent spatial resolution (~10 m) and excellent energy resolution (~700 eV at 32 keV), and are more readily available than other proposed detectors. Si79

Electron Tracking Measurements for Advanced Compton Imaging

based CCD systems provide a good compromise between efficiency and resolution for Compton imaging. On one hand, the high granularity provides the initial vertex of the Compton-scattered electron, on the other hand, Si as a significantly higher efficiency to induce the Compton scattering process, e.g. compared to gases. In this paper, we report on measurements we have performed employing a high-resolution CCD system in combination with a 3D position sensitive Ge detector and a collimation system that allows us to fully determine the scattering kinematics. We will discuss the algorithms we have developed to extract the important initial direction of the Compton electron and preliminary results on the impact of electron tracking on Compton imaging.
This material is based upon work supported by the U.S. Department of Homeland Security under the Grant Award Number 2008-DN-077-ARI001-02. N13-138:

Development of Two-Dimensional Micro-pixel Gas Chamber Capable of Individual Line Readout for Neutron Measurement

K. Toh, H. Yamagishi, K. Sakasai, T. Nakamura, K. Soyama J-PARC Center, Japan Atomic Energy Agency, Tokai, Japan An improved micro-pixel detector element that showed high voltage tolerance was fabricated, and irradiation experiments using a Cf-252 neutron source were conducted on it. A gas-based neutron detection system capable of individual line readout, consisting of the developed detector element, gas chamber, amplifier-shaper-discriminator boards, and a fast data acquisition device, was constructed for the experiment. The micro-pixel gas chamber can identify signal-pulse peaks of neutrons and easily distinguish the neutron signals from background noises. The measured gas gain was approximately 280 at the supplied anode-cathode voltage of 670 V and total pressure of 0.5 MPa (0.45 MPa of 3He and 0.05 MPa of CF4).
N13-139:

H. T. Sipila1, H. J. Sipila1, J.-O. Lill2, S.-J. Heselius2 1 Turku PET Centre, Turku University Central Hospital, Turku, Finland 2 Turku PET Centre, Abo Akademi University, Turku, Finland

Determination of 68Ge-Trace Activities in 68Ga Eluates from a 68Ge/68Ga Generator Using X-Ray Spectrometry with a Gas Proportional Counter

Aim: The aim of this study is to introduce a new concept for measuring the remaining radioactivity in the 68Ga eluate or in the final radiopharmaceutical product. 68Ge decays to 68Ga through electron capture (EC, 100%). Using the 9.24 keV X-rays of 68Ga following the EC of 68Ge the remaining radioactivity in the 68Ga eluate can be measured with a gas proportional counter. Methods: The 68Ga eluate was analysed at 30 hours after elution of the 68Ge/68Ga generator (1.85 GBq 68Ge, Cyclotron Co. Ltd., Russia). Samples of 1 ul, 5 ul, 10 ul, 20 ul, 40 ul, 60 ul and 100 ul from the 68Ga eluate were applied on paper discs. The paper discs were placed one by one on the Be window of a gas proportional counter (Oxford Instr., Type 0460.3 Ar). X-ray spectra of each sample were acquired using a multi-channel analyser (Canberra Genie 2000 MCA). The peak areas were obtained by integrating the channel counts of the 9.24 keV X-ray peak. On the right side of the peak the same number of channels was integrated as background counts representing a continuous spectrum of 68Ga positrons. The background counts were subtracted from the peak counts. Reference radioactivities of the paper discs were measured using a gamma counter (1480 Wizard 3 in. Counter; EG&G Wallac). Reference X-ray spectra were also acquired using an IGP detector (Princeton Gamma-Tech, crystal area 25 mm2, thickness 5 mm). Results and Conclusion: X-ray spectrometry by means of a gas proportional counter is a sensitive technique for determining the amount of 68Ge in the 68Ga eluate from a 68Ge/68Ga generator or in the final radiopharmaceutical product. The method was linear in the activity interval measured. The limit of detection (LOD) utilizing X-ray spectrometry with the gas proportional counter was about 0.4 Bq. From the X-ray spectra it can also be noted that the gas proportional counter has a considerably higher sensitivity than the IGP detector for measuring trace amounts of 68Ge in 68Ga eluates.
N13-140:

Y. Wang1, J. Wang1, X. Zhu1, Y. Li1, J. Cheng1, N. Herrmann2, I. Deppner2, Y. Zhang2, P. Loizeau2, D. G. Diaz3 1 Engineering Physics, Tsinghua University, Beijing, China 2 Physikalisches, University Heidelberg, Heidelberg, Germany 3 Gesellschaft fur schwerionenforschung, Damstadt, Germany MRPC (Multigap Resistive Plate Chamber) has good time resolutin, high efficiency and other good performance such as low cost and high granularity. MRPC has been adopted to construct the time-of-flight (TOF) detector for the STAR experiment at RHIC. The MRPC TOF system can improve the particle identification capability in high moment area and will improve RHIC STAR in its K/Pi and K/p separation power to 1.7 and 3 GeV/c, respectively. Considering the demand of high particle rate experiments such as FAIR-CBM and HIRFL-CSR, two kind of high rate MRPC were developed. The high rate MRPCs are produced with resistive electrodes made of semiconductive glass which has bulk resistivity in the order of 1010 cm. The glass plate developed 80

Beam Test Results of Two Kinds of High Rate Multi-Gap Resistive Plate Chambers

in Tsinghua University is 0.7mm thick. Its mechanical dimension can reach 30cm*30cm. The gas gap of MRPC is 220 micron meters wide. The area of readout pads are 18 cm2. One MRPC consists of six gas gaps and the other consists of ten gas gaps. Beam test was performed in GSI from April 16-20. 2.5 GeV proton beam was used. By changing the distance from the detectors to the main beam, different particle rate from a couple of Hz/cm2 to 33 kHz/cm2 were obtained. The test results are as the followings: 1. For the ten-gap MRPC, when the rate increases from 1 to 33kHz/ cm2, its efficiency degrades from 98% to 94%, and its time resolution increases from 95ps to 115ps. This means this MRPC can afford high particle rate up to 33kHz/ cm2. 2. The performance of six-gap MRPC is not as good as the ten-gap MRPC. When the rate increases from 1 to 33kHz/ cm2, its efficiency degrades from 95% to 77%, and its time resolution increases from 110ps to 160ps. In this paper, We describe the layout of beam test and the experimental commission. The performances of two MRPCs are also presented.
This study was supported by the National Natural Science Foundation of China under fund No.10775082 and 10620210287. N13-141:

P. Legou1, M. P. Combet1, F. Nizery1, R. De Oliveira2 1 DSM\IRFU, CEA Saclay, Gif sur Yvette, FRANCE 2 PCB Workshop, CERN, geneve, Switzerland

A New Detector for Fragment Charges Particles from Proton to the Uranium

A new Time Projection Chamber is now under developpement for the future R3B experiement at GSI for the FAIR Program. The detector would be able to detect with a high efficiency charged fragments from the proton up to the uranium.This TPC will presents quite big dimensions, as matter of fact it will measure 8 m wide and 1.6 meter high. This detector will be capable to detect track events with a clear identification and a reconstruction of each track. We will use a gaseous detector called micromegas, and the detector will be used in TPC mode, in order to present the minimum amount of material along the beam to minimize parasitics interactions. We made three tests beams at GSI. We have tested in last april the very new resistive anode detector built at CERN in a high intensity beam. We have seen already very encouraging results concerning the space resolution and spark protection. In the poster we propose to present an overview of the whole detector, and the way the gaseous detector is read and integrated very close to the front-end electronics. we would present also a comparison between "classical" micromegas detector and the new resistive detector concerning the space resolution.
N13-142:

Y. Tian1,2, Y. Li1,2, Y. Li1,2, H. Qi3 1 Dept. of Engineering Physics, Tsinghua University, Beijing,100084, China 2 Key Laboratory of Particle & Radiation Imaging (Tsinghua University), Ministry of Education, Beijing,100084, China 3 Nuctech Company Limited, Beijing,100084, China Gas electron multiplier, which was first developed for high energy physics experiment, is also a promising candidate for imaging detector. In our lab, a GEM imaging detector prototype was successfully designed and constructed. During the systematic response measurement of this device, an unexpected saturation phenomenon was highlighted. Under moderate effective gain, the response of double GEM system is proportional to the X-ray intensity, as was expected. However, when the gain is increased by either increasing the voltage applied to each GEM foil or using triple GEM, apparent saturation effect is observed. At high pressure, such as 2atm or 3atm, there is a similar phenomenon. Intensive research on this effect has been carried out. Taking into account drift voltage, effective gas gain, protection resistor and area effect, we now significantly diminish the saturation of the detectors response and get some promising results. Key words: saturation, GEM, imaging detector
N13-143:

The Study of Saturation Effect of GEM Detector for X-ray Imaging

R. Fan1, F. Hou1, F. Yi1, Y. Chen1, Q. Ouyang1, T. Zhao2 1 Institute of High Energy Physics, Beijing, China 2 Department of Physics, University of Washington, Seattle, WA 98185, USA We report results of prototype micromegas chambers that have a resistive sheet glued to the readout PCB board using carbon loaded high resistivity adhesives. Two types of resistive sheets are studied. One is a thin plastic sheet extruded from a special acetal copolymer with nominal volume resistivity in the range of 10^9 to 10^11 .cm depending on the material compositions and the relative humidity. The thinnest sheet we tried is 0.30 mm. Also, special bakelite sheets with good surface quality developed for the BESIII RPCs and other applications are tried. In both cases, the stainless steel mesh is affixed on the surface of the resistive sheet and readout electrodes are outside of the gas volume. Our tests show that these resistive sheets can significantly extend the gas gain of standard micromegas without the resistive layer while providing good protection against sparks that dissipates all the charges stored on the mesh. The additional thickness of resistive layer above readout strips can help to spread micromegas signals laterally. This feature allows wider readout elements to be used and the total number of readout channels can be minimized. A readout scheme by using 2D readout strips etched on double sided PCBs is tested. 81

Mcromegas with Resistive Anode

N13-144:

A Novel Detector for 2D Ion Detection in Low-Pressure Gas and Its Applications

V. A. Bashkirov, R. F. Hurley, R. W. Schulte, J. M. Slater Loma Linda University, Loma Linda, USA A detector array for single-ion registration was developed and studied. The detector array was operated with various working gases, including propane, air, and water vapor at low gas pressure (a few mbar) in a limited Geiger mode, providing single ion sensitivity. The patterned detector structure, comprising sub-millimeter hole diameter, millimeter pitch, ground and 2D readout electrodes, were manufactured using standard PCB technology. A glass cathode was utilized to allow the limited discharge mode of operation. Essentially, the new detector is a combination of a hole-type micro-pattern detector and a resistive plate counter operating at reverse polarity in a low pressure gas. It was demonstrated that the unique combined properties of these well established individual detectors in the low pressure gas environment allows for single ion registration through ion-impact induced discharge confined inside an individual cell (hole) of the detector array. An ion time projection chamber, equipped with anode and drift region, was realized on the basis of this detector array. Ion drift time measurements and 2D ion detector readout provided, for the first time, a full 3D track structure reconstruction with ultimate (individual ions) resolution of ionization pattern. Possible applications of the new ion detector in radiation protection, hadron therapy, and mass spectrometry will be briefly discussed.
N13-145:

Performance of the BESIII Drift Chamber

M. Dong Institute of High Energy Physics, Beijing, China On behalf of the BESIII Collaboration The Beijing Spectrometer III (BESIII) is a high precision, general purpose detector for the Beijing electron-positron collider II (BEPCII), a high luminosity, multi-bunch collider running at -charm energy region. To meet the requirements of BESIII experiment, the center tracking detector is a low-mass cylindrical drift chamber with small-cell geometry, using helium-based gas and operating in a 1T magnetic field. After the construction and commissioning, the drift chamber works well. In recent running at the e+e-collision mode, the spatial resolution of 143m and the momentum resolution of about 0.82%@1.85GeV/c in center of mass system have been obtained, and the dE/dx is about 5.9%. These results show the drift chamber has reached its expected performance. We report the development and the performance of the drift chamber.
N13-146:

W-values for Heavy Ions in Gases

S. Sasaki, T. Sanami, K. Saito, K. Iijima, H. Tawara, T. Murakami Radiation Science Center, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan The average energy to create an ion pair, W, for heavy ions is known to show dependence on the ion energy. However, there are still few experimental data about W for heavy ions and W is one of indefinite parameters in the dose evaluation or in the presumption of absorbed energy for heavy ions. The systematic measurements are proposed. In such backgrounds, we aim to determine W for heavy ions as a function of energy and to provide data to application fields such as the heavy ion therapy and for explicating mechanisms of its Z and energy dependence. An ionization chamber used in the present study is equipped with an energy degrader to obtain particles with desired energy and a time-of-flight energy spectrometer (TOF) to measure the energy of incident ions. Several kinds of heavy-ion beams from HIMAC (Heavy Ion Medical Accelerator in Chiba) were used in the experiment. To measure number of charges per ion, we used two types of pulse operation modes of ionization chamber by changing electrode arrangement. One is a slow-pulse mode using a parallel-plate chamber and a fast current amplifier, and the other is a fast-pulse mode with electron collection using a gridded chamber for the measurements in rare gases. The collector in the ionization chamber was segmented into 10 pads in a row in the direction of beam, from which signals were taken independently. The pulse height distributions measured with the both modes show peaks corresponding to the number of incident ions. We have made measurements for He, C, N, O, Ne, and Ar ions in argon, air and tissue-equivalent gases, and determined the values of W as a function of ion energy. The energy dependence of W was observed in these gases for all ions except for He ions. Differential values of W (w-value) have been also determined using signals from a pad on the collector. The preset status of the study and the details of the recent results of W and w for heavy ions will be presented in the conference.

82

N13-147:

An Investigation of the Dependence of CMS Resistive Plate Chamber Operation on Environmental Parameters

L. Benussi Physics, INFN, Frascati, Italy On behalf of the CMS RPC Collaboration Resistive Plate Chambers (RPCs) are employed widely for various particle physics experiments to detect high energy charged particles especially fast muons, due to their excellent time resolution, reasonable space resolution and high rate capability. High gain, simple electronics, robustness and low cost of these detectors have sparked off a lot of interest for cosmic-ray and low- rate accelerator experiments with operation in streamer mode; while the high luminosity experiments ATLAS and CMS, at LHC employ them in avalanche mode. At the CMS experiment, the muon system consists of a large number of bakelite RPCs both in the barrel and forward regions of the experiment. In this work, the operation of Resistive Plate Chambers is characterized by using a neural network model. Dependences on temperature, humidity and other environmental parameters are investigated. After teaching the model with known data over a large period of time, the neural network is used to predict unseen data with good agreement.
N13-148:

D. Y. Jang1, B. H. Park1, S. Y. Kang1, B. H. Kang1, H. S. Kim2, J. H. Ha2, S. H. Park2, W. Lee2, G. S. Choi2, Y. K. Kim1 1 Nuclear Engineering, Hanyang University, Seoul, South Korea 2 Advanced Radiation Detection instrument & Sensor Lab, Korea Atomic Energy Research Institute, Daejeon, South Korea High-pressure xenon (HPXe) gas is a desirable gamma-ray detection medium due to its physical and nuclear properties. HPXe has a large detection efficiency for gamma ray energies on the order of hundreds of keV due to its large atomic number (Z = 54) when compared with Ar gas widely used for gamma-ray detection medium. And also, HPXe ionization chambers are considered as ideal detectors for the Radiation Monitoring System (RMS) in nuclear power plants, as this detectors response has been shown to be uniform over large temperature ranges up to 170&degC. By using the MCNPX code, the energy spectra were calculated with respect to the thickness of the outer shell and the dependence on the incident direction of the radiation sources in the nuclear power plant was simulated. A shielding mesh manufactured by using an etching method, which eliminates the charge contribution of the slow positive ions, was inserted between the collecting electrode and the potential electrode to increase its energy resolution. With the gas purification and injection system, Xe+7% 4He was purified and injected in the chamber and the performances of the chamber were observed. The leakage currents and saturation currents of the chamber were measured with an electrometer (Keithley 6517A), respectively. Linearity against low dose rates were also measured with 226Ra (0.906 mCi) source and the root-mean-square linearity was 0.994. In this study, high pressurized Xe ionization chambers were fabricated and evaluated to monitor environmental radiation level in the nuclear power plants. The chambers showed a low leakage current and good linear relationship against low dose rates. Thus, this chamber can be used for environmental radiation monitoring system. In a future work, energy spectra with the fabricated HPXe ionization chamber will be measured and compared with simulated energy spectra.
Acknowledgment : This work was supported by the Basic Atomic Energy Research Institute (BAERI), nuclear R&D program of MEST, Korea and partially supported by the Ministry of Knowledge Economy (2008-P-EP-HM-E-06-0000). N13-149:

A Cylindrical High-Pressure Xenon Chamber with Shielding Mesh Manufactured by Using an Etching Method

S. Sasaki1, K. Saito1, H. Tawara1, T. Doke2, K. Miuchi3, T. Komiyama4, H. Matsumoto4, Y. Uchihori5, K. Terasawa6 1 High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan 2 Waseda University, Shinjyuku, Tokyo, Japan 3 Kyoto University, Kyoto, Kyoto, Japan 4 Japan Aerospace Exploration Agency, Tsukuba, Ibaraki, Japan 5 National Institute of Radiological Science, Chiba, Chiba, Japan 6 Keio University, Yokohama, kanagawa, Japan

Present Status of Position-Sensitive Tissue Equivalent Proportional Chamber (PS-TEPC) for Space Dosimetry

Radiation dose for human body is evaluated using the dose equivalent H, defined as the product of the absorbed dose and the quality factor given as a function of the linear energy transfer (LET). In space, there exist many kinds of radiations such as galactic cosmic rays and geomagnetic trapped particles, where charged particles and neutrons are considered to be the main components contributing to radiation doses in space. LET of these radiations distributes from 0.2 to 600 keV/m. It is essential to measure LET to evaluate H in space. The Tissue Equivalent Proportional Counter (TEPC) has been used as a standard space dosimeter, which is a simple gas proportional counter made of tissue equivalent materials. In TEPC, since no position information is given, LET can not be measured directly and the lineal energy is used instead of LET. Obviously the lineal energy 83

does not represent LET accurately. The dose equivalent obtained using TEPC is reported to be more than twice that measured with the real LET spectrometer. We have started to develop a new dosimeter named Position Sensitive Tissue Equivalent Chamber (PS-TEPC), which is based on a time projection chamber (TPC) using a micro-pixel chamber (-PIC) as a 2D readout device. The -PIC consists of a double-sided printed circuit board. The anode and the cathode strip electrodes are orthogonally arranged with a pitch of 400m on the both side of 100m thick polyimide insulator. Pixel-like anode pillars of 50m diameter are formed at the center of each cathode opening through the insulator. Due to an electrical field between the anode and the cathode, a gas gain of 1000 is obtained without any additional gas multiplication devices. We have demonstrated successfully the feasibility of PS-TEPC from the results of 3D-tracking and the energy measurements obtained for heavy ion beams. We are presently tying to complete the prototype of PS-TEPC usable in space. The present status of PS-TEPC is described in detail.
N13-150:

A Study of 3He Detectors for Active Interrogation

E. H. Seabury, D. L. Chichester Idaho National Laboratory, Idaho Falls, ID, U.S.A. He proportional counters have long been used as neutron detectors for both passive and active detection of Special Nuclear Material (SNM). The optimal configuration of these detectors as far as gas pressure, amount of moderating material, and size are concerned is highly dependent on what neutron signatures are being used to detect and identify SNM. We present here a parametric study of the neutron capture response of 3He detectors, based on Monte Carlo simulations using the MCNPX radiation transport code. The neutron capture response of the detectors has been modeled as a function of time after an incident neutron pulse.
N13-151: A New Technique for Gaseous Gamma Ray Detectors: the Multigrid High Pressure Xenon Gas Proportional Scintillation Counter
3

F. I. G. M. Borges, S. J. C. do Carmo, J. C. R. Mariquito, A. M. F. Trindade, C. A. N. Conde Physics Department, University of Coimbra, Coimbra, Portugal A new technique is presented for high pressure gaseous gamma ray detectors, that leads to pulse amplitudes about one order of magnitude larger than ionization chamber based gamma ray detectors, and that does not rely on charge avalanche multiplication, like proportional counters do. The technique uses room temperature Xe based gaseous mixtures at pressures of about 5 to 20 atm (or higher) to detect ionizing radiation in a multigrid gas proportional scintillation counter (MGHP-GPSC) with a CsI photocathode and no optical windows. Experimental results are presented for a small prototype filled with pure Xe at 5 atm, concerning the dependence of the pulse amplitude on the voltages applied to the different grids and it is shown that its dependence has the behaviour expected for the proportional scintillation of Xe. The potential of this detector for large size, high energy resolution, ruggedized, room temperature gamma ray detection applications, like homeland security, is discussed and compared with that of competing detectors.
Patent pending N13-152:

L. M. N. Tavora1,2, A. M. F. Trindade2, P. N. B. Neves2, C. A. N. Conde2, J. A. S. Barata3 1 ESTG, Instituto Politecnico de Leiria, Leiria, Portugal 2 Departamento de Fisica, Universidade de Coimbra, Coimbra, Portugal 3 Departamento de Fisica, Universidade da Beira Interior, Covilha, Portugal

Measurement of the Drift Velocities of Positive Ions in Ar-CH4 Mixtures

We present the results for the experimental measurement of the drift velocities of the positive ions in Ar-CH4 mixtures. In the technique that we have develloped, the ions are produced inside the holes of a GEM covered with a CsI photocathode. The photons from a UV flash lamp release photoelectrons from the CsI film which then trigger an avalanche inside the GEM holes. Some of the ions produced during this avalanche are extracted and forced to drift, under a uniform electric field (E), towards a double grid where they induce a pulse. The concentration of CH4 in the mixture is varied as well as the total pressure of the gas. The results are presented in function of the drift reduced electric field (E/N). Depending on the pressure of the gas, the concentration of CH4 and the value of E/N, several ions are simultaneously present, as depicted in the ions' time arrival spectra. The nature of these ions is analyzed and their drift velocity values are discussed in the light of well established ion-mobility theories.
This work was supported by project PTDC/FIS/65569/2006.

84

N13-153:

E. Graziani1, A. Baroncelli1, D. Sacco1, P. Branchini1, F. Petrucci1, S. Di Luise1, M. Schioppa2, G. Morello2 1 INFN - sezione Roma3, Roma, Italy 2 dep. of Physics, UNICAL - INFN, Cosenza, Italy

Systematic Study of the Calibration of the Drift Tubes for Muon Tracking in the ATLAS Experiment at LHC and Possible Use of Fast Gas Mixtures for SLHC

The Muon Spectrometer of the ATLAS experiment consists of three large air core toroidal magnets, one barrel and two end-caps. The bending power is about 3 and 5 Tm respectively. Tracking in the range |eta|<2 is done with regular arrays of 3 cm diameter high pressure drift tubes, called MDT, arranged in three measurement stations. The aim of the experiment is a relative resolution of 10% at transverse momentum of 1 TeV/c. This implies a single point resolution which should not exceed 80 um per tube, averaged over the drift distance. The gas composition (Ar/CO2 93/7) and gain (10^4), chosen in order to limit aging effects, make the drift response function, r(t), slow and highly non linear, thus depending on temperature, pressure and local value of the magnetic field. The maximum drift time is typically around 700 ns. An auto-calibration method was developed to compute from the data themselves the r(t) function and to extract the corresponding resolution function. A test beam setup, with a small MDT chamber, a high resolution silicon strip telescope and an external, scintillator based, trigger, was organised at the Cern SPS H8 180 GeV/c muon extracted beam with the twofold goal of a) measuring the r(t) relation and the spatial resolution of the chamber with the standard ATLAS gas mixture, in order to compare them with the auto-calibration derived results; b) measuring the r(t) relation and the spatial resolution of the chamber with faster gas mixtures (for example Ar/CF4 93/7 and Ar/CF4 80/20) in view of a possible use at SLHC or elsewhere. Lower priority results may include c) detailed studies on the time slewing and temperature corrections for the standard ATLAS gas mixture. We describe the analysis of the high momentum muon data collected with this experimental setup and show the first, preliminary, results for items a) and b) and, possibly, c), and their comparison with the simulation.
N13-154:

E. Tashiro1, T. Ishizuka2, N. Ishihara3, G. Iwai3, H. Iwase3, K. Takahashi3, T. Ohama3, Y. Kato3, Y. Yamada3, T. Ishikawa4, T. Sumiyoshi4, S. Kitamura5, K. Tanaka6, R. Ito7 1 Graduate School of Engineering, Shizuoka University, Hamamatsu, Shizuoka, Japan 2 Faculty of Engineering, Shizuoka University, Hamamatsu, Shizuoka, Japan 3 High Energy Accelerator Research Organization (KEK), Tukuba, Ibaraki, Japan 4 Graduate School of Science and Engineering, Tokyo Metropolitan University, Hachioji, Tokyo, Japan 5 Nihon Institute of Medical Science, Irumagun, Saitama, Japan 6 Saitama Science Institute, Fukaya, Saitama, Japan 7 FutureScope, Shibuya, Tokyo, Japan A gaseous detector called DCBA (Drift Chamber Beta-ray Analyzer) is being developed at KEK in order to search for the events of neutrinoless double beta decay. DCBA is a momentum analyzer detecting charged particle tracks in a uniform magnetic field. It can individually identify the electron, the positron and the alpha particle. So, it is easy for us to distinguish beta ray events from background events. In DCBA experiment, the charged particle tracks of X-Y plane and X-Z plane are recorded. The track is reconstructed in three dimensions by the recorded data, and then its momentum and kinetic energy are obtained from a helical curvature and pitch angle. A beta ray track has been selected by the so-called eye-scan method from other high momentum events so far. Therefore a great deal of time and efforts are consumed to analyze data of tracks. To solve this problem, this study aims to automatically sort events by means of the Hough transformation focusing on the feature of DCBA. The Hough transformation is one of the methods used in general for detecting curves and lines in pictures. The original algorithm of Hough transformation was invented to recognize complex patterns in bubble chamber photographs. We have applied the Hough transformation for classifying the charged-particle-tracks into lines and curves. Experimentally, we sorted the track data by the results of line and circle detections. We will report the sorting methods, its results and classification accuracy.
N13-155:

Implementation of Automatically Sorting Events with the Hough Transformation for the DCBA Experiment

Another Method to Produce Laser Calibration Beams in Gaseous Detectors

A. N. Lebedev Brookhaven National Laboratory, Upton, NY, USA A new method utilizing production of UV laser beams by axicon optics provides very narrow laser beams, so-called Bessel beams, which significantly exceeds existing methods with focusing optics. The characteristics of laser beams and linear ionization created with different optical schemes are present. This method could be used for detector alignment and to produce calibration beams.

85

N13-156:

T. Fujiwara1, H. Takahashi1, B. Shi1, N. Iyomoto1, K. Fujita2 1 Department of Nuclear Engineering and Management, School of Engineering, The University of Tokyo, Bunkyo, Tokyo, JAPAN 2 Japan Atomic Energy Agency, Tokai, Ibaraki, JAPAN Gaseous radiation detectors are used in various applications, however, recent progress in the micro-pattern gaseous detectors can expand the application area of these relatively old radiation detectors. Normal metals such as Cr, Au are used in Conventional MSGCs, which prevent the use of optical image since the metal pattern absorbs the light. Indium Tin Oxide (ITO) is widely used transparent electrode material. If the MSGC is transparent, it is also possible to observe light from gas proportional scintillation counters . Thus, we have decided to develop a transparent MSGC by using ITO electrodes. Here, we introduce this new category of micropattern gas detectors. We have selected our multi-grid-type MSGC configuration in our ITO-MSGC since it enables the use of a variety of substrate glasses. In the multi-grid-type MSGC, grid electrodes are inserted between narrow anode strips and wide cathode strips. In order to test charge amplification properties of this plate, we have operated the plate in a PR gas (1 atm, Ar (70%)+CH4(30%)). The detector has been irradiated by X-rays from an Fe-55 source. A gas gain of ~2000 is successfully achieved at the anode voltage of 700 V. We have operated the ITO-MSGC in 1 atm CF4 gas, which is known as a good scintillation gas medium. The plate is operated in a clear chamber and a photomultiplier tube (Hamamatsu R928) is placed outside the chamber behind the plate. Optical photons are transmitted through the glass substrate plus the chamber wall. Observed optical signal taken with a PMT is very promising. We have successfully demonstrated the optical signal readout with ITO-MSGC. These charge and optical signals can effectively be used for position sensing. For example, if we obtain coarse position information from light and we obtain the detailed position from charge signal, we can identify incident position of radiation with very high precision and cheap cost. We are trying to develop such a readout system with APD.
N13-157:

A New Transparent MSGC for Both Charge and Optical Position Readout

J. Escada1, T. H. V. T. Dias1, P. J. B. M. Rachinhas1,2, F. P. Santos1, F. I. G. M. Borges1, C. A. N. Conde1, A. D. Stauffer3 1 Departamento de Fisica, Universidade de Coimbra, Coimbra, Portugal 2 Servico de Radioterapia, Hospitais da Universidade de Coimbra, Coimbra, Portugal 3 Physics and Astronomy Department, York University, Toronto, Canada Due to the interest of Xe doped with CF4 or CH4 in Time Projection Chambers for Xe-136 double beta decay experiments, we have carried out a Monte Carlo simulation study of the density-reduced secondary scintillation Y/N and corresponding fluctuations in these gaseous mixtures, as a function of the reduced electric field E/N for different CF4 or CH4 concentrations. While the addition of CF4 or CH4 to Xe may reduce electron diffusion and increase drift velocity, the secondary scintillation yield in the mixtures becomes lower and has larger fluctuations than for pure Xe, due to the vibrational excitation of the molecules by electron impact and to electron attachment, so that a compromise must be found to minimize energy resolution degradation.
N13-158:

A Study of the Scintillation Yield and Fluctuations in Xe Doped with CF4 or CH4: the Role of Electron Cooling and Attachment

F. Asami1,2, T. Tamagawa1,2, A. Hayato1,2, T. Iwahashi1,2, S. Konami1,2, H. Tawara3, K. Makishima1,4 1 Cosmic Radiation Laboratory, RIKEN, Wako, Saitama, Japan 2 Tokyo University of Science, Shinjuku, Tokyo, Japan 3 KEK, Tsukuba, Ibaraki, Japan 4 University of Tokyo, Hongo, Tokyo, Japan Gas Electron Multiplier(GEM) is a newly developed micro pattern gasdetector. The GEM has an impressive advantage to amplify electrons in two-dimension. We have developed various GEMs with a laser etching technique in Japan, and have applied them to photoelectric X-ray polarimeters for future space missions. Since gain uniformity is important for those applications, we measured gain and hole diameter distributions across the GEMs. The gain uniformity distributions of several GEMs were about 4%, and the hole diameter distributions were about 3%. Those results satisfied our requests. We will present the detail of the measurements and results at the conference.
N13-159:

Mapping Study of Hole Diameters and Gain of Japanese GEMs

A New Technique for Charge Dependent Corrections to the Time Response of the Drift Tubes of the ATLAS Experiment

S. Di Luise Roma University and INFN, Rome, Italy Monitored Drift Tubes (MDT), filled with a mixture of Ar:CO2 at 3 bar pressure, are used as tracking devices for the ATLAS muon spectrometer at the Large Hadron Collider. The MDT chamber tested is equipped with the final version of front-end electronics, which provides information about the collected charge through a Wilkinson ADC. We present an analysis of the 86

signal and timing characteristics, and of charge-dependent time response of the MDTs. A new method to improve the spatial resolution of MDT chambers using the ADC information to correct the time response was developed. The procedure is based solely on the data from the MDT chamber itself. The improvement with respect to previous methods is achieved thanks to an indirect derivation of the slope of the induced anode signal. This requires an accurate determination of the response function of the ADC. Response function which is obtained by injection pulse of known charge in the integrator circuit. The results are compared with a simulation of the tube response with the Garfield program.
N13-160:

C. D. Azevedo1, A. L. Gouvea1, A. L. Silva1, J. M. dos Santos2, J. F. Veloso1 1 I3N, Physics Department, University of Aveiro, Aveiro, Portugal 2 GIAN, Physics Department, University of Coimbra, Coimbra, Portugal

2D-Sensitive HpXe Gas Proportional Scintillation Counter

The operation of a detector working in Gas Proportional Scintillation Counter (GPSC) mode with high pressure xenon for imaging purposes will be presented. The GPSC uses a CsI- Micro-Hole and Strip Plate (MHSP) based photosensor for scintillation readout. To define the scintillation region, a pair of stainless steel meshes separated by a 2.8-mm gap is placed 3-mm above the photosensor. Voltage is applied between meshes to establish an electric field with a value above the xenon excitation threshold but below the ionisation threshold. The voltage difference between the radiation window and the first mesh establishes an electric field, set below the xenon excitation threshold, in the drift region, above the mesh. The resulting scintillation is generated when the electron cloud, produced in xenon following x- or -ray interactions, crosses the gap between the meshes. The interaction position of the x- or -photons in the xenon medium can be found by determining the centroid of the resulting photosensor illumination. The CsI-MHSP photosensor 2D capability is achieved by using two orthogonal resistive lines interconnecting the strips patterned on both surfaces. Experimental studies of the interaction position along the X- and Ycoordinate for different parameters of the detector as a function of xenon pressure will be presented.
Acknowledgements: This work was supported by project PTDC/FIS/65455/2006 through FEDER and FCT (Lisbon) programs. N13-161:

C. D. Azevedo1, A. L. Gouvea1, C. A. Santos1, J. M. dos Santos2, J. F. Veloso1 1 I3N, Physics Department, University of Aveiro, Aveiro, Portugal 2 GIAN, Physics Department, University of Coimbra, Coimbra, Portugal An experimental study of the photoelectron extraction efficiency from a CsI photocathode operating in high pressure xenon medium will be presented. A CsI reflective photocathode, deposited on a gold-coated copper substrate, is irradiated with photons from an Hg lamp (1855nm). The resulting photoelectrons are collected and measured on a stainless steel mesh placed 5 mm above the CsI photocathode. This mesh also serves to set the electric field above the CsI photocathode needed for photoelectron extraction. For the measurements, xenon pressures from 1 up to 6 bar will be considered and results as a function of extraction electric field will be obtained. The effect of the addition of small quantities of CF4 to the HpXe in photoelectron extraction will be also studied.
Acknowledgments: This work was supported by project PTDC/FIS/65455/2006 through FEDER and FCT (Lisbon) programs. N13-162:

CsI Photoelectron Extraction Efficiency in HpXe Medium

Improving the PHENIX Muon Trigger Using Resistive Plate Chambers

R. Towell Physics, Abilene Christian University, Abilene, TX, USA On behalf of the PHENIX Collaboration Many significant questions remain to be answered about the origin of the proton spin. A new fast Resistive Plate Chamber (RPC) based trigger system is being developed for the PHENIX muon spectrometer arms that will allow for the first time the measurement of the flavor structure of the quark polarization in the proton through the observations of W-bosons in polarized proton-proton collisions at the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory (BNL). The new PHENIX Muon Trigger will improve the efficiency by which the data acquisition system can identify potential W events by approximately two orders of magnitude. W-bosons can be detected through the appearance of a high-energy muon in one of the two existing muon spectrometers. The trigger upgrade is based on new front-end electronics for the muon tracking chambers and RPCs that will be installed in two stations of both muon arms. Components of the RPCs are being fabricated at many different locations around the world. After they are shipped to BNL, these components will be tested and then the RPC modules will be assembled at a newly completed facility. Once assembled, these modules will be extensively tested and jointed in half octant units for installation into the PHENIX spectrometer. Prototype chambers have been assembled, tested, and installed before the last RHIC run. The RPC upgrade will be described and results from the prototypes will be presented.

87

N13-163:

M. Ashurov1, M. Baydjanov1, E. Gasanov2, J. Ibragimov1, I. Nuritdinov2, U. Salikhbayev2 1 Scientific Center, State Scientific Industrial Enterprise, Tashkent, Uzbekistan 2 Radiation Physics of Optical Materials, Institute of Nuclear Physics, Ulugbek, Tashkent, Uzbekistan

Dose and Dose Rate Dependency of Emission in Quartz Fibers with High-OH Group Content

It is known that during utilization of quartz fibers with high-OH group content in radiation environment the unavoidable processes arises that bring to degradation of parameters of fibers and distortion of useful information. These processes attributed to induced absorption and luminescence. This work is devoted to study of light emission spectra arising in quartz fibers with high-OH group content under -irradiation influence of 60Co (1,25 MeV) source. We used the method of restoring of the true spectra taking into account the reabsorption along irradiated and transporting parts of the fiber. -induced emission spectrum consist of Cherenkov emission and luminescence bands at 450 and 650 nm ascribed to recombination of holes in E-center and non-bridging oxygen hole center (NBOHC) correspondingly. Cherenkov emission is caused by secondary electrons generated by -rays in the fiber. Increase rate of 450 nm band intensity is higher than that of 650 nm band with the growth of irradiation dose up to 10^8 Rad at dose rate 360 R/s. In addition, increase rate of absorption band at 610 nm is higher than increase rates of both luminescence bands. Absorption band at 610 nm is observed in high-OH fibers by many authors and interpreted as sum of absorption bands caused by two kinds of NBOHC. Disagreement of dose dependency of this band with dose dependence of luminescence bands can be caused by nonradiative relaxation of most part of NBOHC amount participating in luminescence at 650 nm band. Another possible suggestion that there exist absorption bands around 610 nm caused by centers other than NBOHC. The increase of mostly Cherenkov emission intensity and comparatively small increase of 450 and 650 nm luminescence bands intensities are observed during the change of dose rate from 6 to 360R/s. It is caused by increasing of the number of secondary electrons and luminescence centers.
N13-165:

Operational Experience with the CDF Run II Silicon Detector

M. J. Mathis Dept of Physics and Astronomy, Johns Hopkins University, Baltimore, MD, 21218 On behalf of the CDF II Silicon group The CDF Run II silicon detector is the largest operating silicon detector in High Energy Physics. Its 722,000 channels spread over 7 m2 of silicon micro-strip sensors allow precision tracking and vertexing. The CDF silicon detector played a critical role in the discovery of Bs mixing and is used extensively for the current Higgs Boson searches. Over the last 7 years, the detector efciency has remained stable at 95 % after the Run II commissioning period. The infrastructure (cooling, power supplies) problems dealt with are discussed.
N13-166:

Studies on Charge Collection of P-Type Silicon Detectors under Neutron Irradiation Expected for Super LHC

M. Minano, C. Garcia, C. Lacasta, R. Marco-Hernandez, S. Marti i Garcia, U. Soldevila Instituto de Fisica Corpuscular (IFIC), Valencia, Spain The existing technology (p-strip readout on a n-type substrate) used in the ATLAS Tracker is the limit for performances with 10 years of running at a LHC luminosity of 1034 cm-2s-1. The operation under an upgraded luminosity of 1035 cm-2s-1 (super LHC) implies to upgrade the semiconductor tracking systems of the LHC experiments. The expected dose for the inner detector tracker at the super-LHC is up to 1016 equivalent-1-MeV-neutron cm-2 after the envisaged five years of operation. Investigations have showed arguments in favour of implementing the n-strip readout on a p-type substrate. P-type microstrip detectors have been used for this work. In order to evaluate the radiation damage they have been irradiated with neutrons at several fluxes up to 1016 cm-2. Electrical and charge collection efficiency measurements have been carried out by means of a radioactive source setup as well as by an infrared laser illumination and the measurements normalised with a non-irradiated sensor as a reference. Two acquisition system have been used: 1. Electronic amplifier based system. The sensor signal was provided from the readout of all the strips shorted together and the charge was computed as the integral of the averaged signal pulse at the oscilloscope over several thousand triggers. 2. The ALIBAVA system. It is a compact and portable system which contains two front-end readout chips (Bettle chip) to acquire the detector signals. One of the advantages of the ALIBAVA system is that it uses LHC speed electronics. Another one is that it performs a pulse by pulse and strip by strip analysis.
N13-167:

A New Data Concentrator Card for the Hadron Calorimeter in CMS

E. S. Hazen, J. Rohlf, A. Heister, J. StJohn, S. Wu, P. Lawson Physics, Boston University, Boston, MA, USA A new data concentrator card has been designed for readout of the CMS hadron calorimeter. The data concentrators process digitized data from about 10k channels of photo-detectors mounted along with their digitizers on the CMS detector. Triggers at a 88

rate of 100kHz initiate transfer of data from trigger/readout cards to the data concentrators where event fragments are built and sent to the CMS high-level trigger. The data concentrators perform extensive error-checking and can actually repair various problems with the data stream to maintain pipeline synchronization. An upgraded design was completed and installed in 2009, providing enhanced processing bandwidth, integrated Ethernet-based monitoring, and seeded selective readout. This paper describes the design, installation, commissioning and performance of the upgraded data concentrator system.
N13-168: A Layer Correlation Technique for ATLAS Calorimetry Calibration at the 2004 ATLAS Combined Beam Test

K.-J. Grahn Physics Department, Royal Institute of Technology (KTH), Stockholm, Sweden On behalf of the The ATLAS Liquid Argon Calorimeter Group A method for calibrating the response of a segmented calorimeter to hadrons is developed. The ansatz is that information on longitudinal shower uctuations gained from a principal component analysis of the layer energy depositions can improve energy resolution by correcting for hadronic invisible energy and dead material losses: projections along the eigenvectors of the correlation matrix are used as input for the calibration. The technique is used to reconstruct the energy of pions impinging on the ATLAS calorimeters during the 2004 Barrel Combined Beam Test at the CERN H8 area. Simulated Monte Carlo events are used to derive corrections for invisible energy lost in nuclear reactions and in dead material in front and in between the calorimeters. For pion beams with energies between 20 and 180 GeV, the particle energy is reconstructed within 3% and the resolution is improved by about 20%.
N13-169:

Upgrade of the Cold Electronics of the ATLAS HEC Calorimeter for sLHC: Generic Studies on Radiation Hardness and Temperature Dependence.

H. G. Oberlack Max-Planck-Institut fuer Physik, Munich, Germany On behalf of the ATLAS Liquid Argon Calorimeter HECPAS Collaboration The signal amplification and summation electronics of the ATLAS Hadronic End-cap Calorimeter (HEC) is operated at the circumference of the HEC calorimeters inside the cryostats in liquid argon. The present electronics is designed to operate at irradiation levels expected for the LHC. For operation at the sLHC the irradiation levels are expected to be a factor 10 higher, therefore a new electronic system might be needed. The technological possibilities are investigated. From irradiation tests of the present HEC electronics it is known that it will operate up to a dose of 55 kGy of ionizing radiation and up to a neutron fluence of 3 * 10**14 n/cm**2, where it shows some degradation of performance. This matches well the requirements of up to 1.5 * 10**13 n/cm**2 for 10 years of LHC operation, including safety factors. For a subsequent sLHC running phase with 10 times higher expected irradiation levels, a more radiation hard HEC electronics will be needed. Therefore generic studies of different technologies have been carried out at the transistor level to understand the radiation hardness up to integrated neutron fluxes of ~2*10**16 n/cm**2 and the behaviour during operation at cryogenic temperatures. The S-parameter technique has been used to monitor the performance e.g. of gain and linearity during irradiation at room temperature. In addition, DC measurements before and after irradiation have been compared. Results of these tests and of accompanying noise tests are reported. In addition, results of S-parameter measurements will be reported operating the transistors in liquid nitrogen. Conclusions are drawn and the potential is assessed on the viability of using the tested technologies for carrying out the design of new HEC cold electronics for the sLHC.
N13-170:

F. Belli1, B. Esposito1, D. Marocco1, M. Riva1, A. Zimbal2, L. Giacomelli2, H. Schuhmacher2 1 Associazione Euratom-ENEA sulla Fusione, Frascati, Italy 2 Physikalisch-Technische Bundesanstalt, Braunschweig, Germany

Conceptual Design, Development and Preliminary Tests of a Compact Neutron Spectrometer for the JET Experiment

The Compact Neutron Spectrometer (CNS) is a Joint European Torus (JET) Enhancement Project, devoted to measure the neutron emission spectra in different plasmas scenarios. The results obtained by CNS during next JET campaigns will contribute to the development of neutron spectrometers suitable for application in the International Thermonuclear Experimental Reactor (ITER). The main components of the instrument are a liquid scintillation detector (BC-501A) for neutron (n) and gamma-ray (g) detection and a digital pulse shape discrimination (DPSD) acquisition system. The CNS will enable simultaneous recording and discrimination of n and g pulse height spectra up to total count rates of about 1 MHz. After unfolding the neutron spectra can be determined with energy resolution < 2% at 14 MeV and <4% at 2.5 MeV [1]. The work done in the last two years by ENEAFrascati and Physikalisch-Technische Bundesanstalt (PTB) respectively in the assembly of DPSD and scintillation detector, along with the preliminary tests and results performed on both the components and the assembled spectrometer are presented. 89

[1] A. Zimbal et al., Rev. Sci. Instrum, 75, 2004, 3553 N13-171:

Operation of the ATLAS End-Cap Calorimeters at sLHC Luminosities: an Experimental Study.

A. Savine Physics Department, University of Arizona, Tucson, AZ, USA On behalf of the The ATLAS Liquid Argon Calorimeter Group The expected increase of luminosity at sLHC by a factor of ten with respect to LHC luminosities has serious consequences for the signal reconstruction, radiation hardness requirements and operations of the ATLAS liquid argon calorimeters (EMEC, HEC, FCAL) in the endcap, respectively forward region. Small modules of each type of calorimeter have been built and exposed to a high intensity proton beam of 50 GeV at IHEP/Protvino. The beam is extracted via the bent crystal technique, offering the unique opportunity to cover intensities ranging from 10**6 pps up to 10**12 pps. This exceeds the deposited energy per time expected at sLHC by more than a factor of 100. In addition, the machine has been operated with the full RF bunch structure preserved and with 5-6 empty bunches between filled bunches. Thus the gap between two filled bunches was about 1 micro-s. This operation mode enables the unique opportunity to study the high flux response of the calorimeters not affected by pile-up from previous bunches and to reconstruct a clean signal of the response over the full drift time of electrons in the liquid argon gap. The correlation between beam intensity and the read-out signal has been studied. Also the dependence of the HV currents as well as calorimeter module temperature on the beam intensity has been measured.
N13-172:

The L1 Track Trigger and High Data Rate Acquisition System for the SLIM5 Beam Test

M. Villa Dept. Physics, Universita' di Bologna, Bologna, Italy On behalf of the Slim5 Collaboration In September 2008 the Slim5 collaboration submitted a low material budget silicon demonstrator to test with 12 GeV/c protons, at the PS-T9 test-beam at CERN. The beam test setup was composed of two different detectors (DUTs) placed inside a reference microstrip silicon tracker telescope. The first was a 4k-Pixel Matrix of Deep N Well MAPS, developed in a 130 Nm CMOS Technology, providing digital sparsified readout. The other one was a high resistivity double sided silicon detector, with short strips at 45 angle to the detector's edge, read out by FSSR2 chips. In total, the set-up contained 38 chips for 12k individual electronic channels. The data acquisition of the experiment was done by means of two high bandwidth, fully programmable 9U VME board (EDRO) that have been designed to stand a 12 Gbit/s input rate, 1.2 Gbit/s output rate and have the possibility to perform different types of trigger strategies on data. The most important one was the on-line track identification performed with the help of an Associative Memory board, which demonstrated the capability of the setup to trigger on identified tracks with a minimal latency (< 1 us). In the contribution, the beam test setup, the data acquisition and trigger systems are presented. The main remarkable trigger and data acquisition performances (2.5 MHz event evaluation rate, 40 kHz data acquisition rate, event selection on identified tracks) that lead to a successful data taking are discussed.
N13-173:

The Expected Performance of the ATLAS Inner Detector

J. Lee University of Sidney, Sidney, Australia On behalf of the ATLAS Collaboration The ATLAS inner detector will see of the order of 1000 charged particle tracks for every beam crossing at the design luminosity of the CERN Large Hadron Collider (LHC). This talk summarizes the design of the detector and outlines the reconstruction software. The expected performance for reconstructing single particles is presented, along with an indication of the vertexing capabilities. The effect of the detector material on electrons and photons is discussed along with methods for improving their reconstruction. The studies presented focus on the performance expected for the initial running at the start-up of the LHC. Results from the 2008 cosmic-ray running are also included.
R.Nickerson has submitted this abstract on behalf of the ATLAS Collaboration's Inner Detector Speakers Committe and will act as a contact in selection of a speaker if the abstract is accepted.

90

N13-174:

Y. Ikegami1, K. Hara2, T. Kohriki1, S. Terada1, Y. Unno1 1 Institue of Particle and Nuclear Studies, KEK, Tsukuba,Ibaraki, Japan 2 Institute of Pure and Applied Sciences, University of Tsukuba, Tsukuba,Ibaraki, Japan

Performance of Low-Mass and High Thermal Conductivity Hybrid for High Track Density Environment

Recent silicon microstrip detectors are designed to incorporate VLSI chips on the detector to readout the signals and communicate with the outside data acquisition system. The hybrid, the base for the VLSI chips and related circuit components, is required to be low mass to reduce multiple coulomb scattering effects for precision tracking. The proposed silicon strip detector at the super LHC experiment includes a short strip silicon detector system in order to cope with the high track density. The system has about 10,000,000 readout channels. In order to minimize the number of hybrids, a hybrid reads out 2 segments of strips with two rows of readout chips; one row in each edge. Thus, the density of readout chips is quite high, 20 chips on a hybrid in this application. We designed and fabricated a new hybrid, applying a technology of copper-polyimide flexible circuit board and carbon-carbon reinforce material which have been developed and adopted for the present LHC ATLAS silicon detector, to read out ABCN chips made with 0.25 m deep-submicron CMOS technology. The performance of the chip and the hybrid is reported. The characteristics of the chip are evaluated with the load connected to the input of the chips. The issues associated with the high density of chips are evaluated and discussed.
N13-175:

The Upgrade of the CMS Tracker for the Super-LHC

G. Steinbrueck Institute for Experimental Physics, University of Hamburg, Hamburg, Germany On behalf of the CMS Collaboration The CMS experiment at LHC is planning a major upgrade of its tracking system to adapt to an expected increase in luminosity of the accelerator. Present ideas suggest this will be achieved in two stages, with a major long shutdown after the experiments have reached their design lifetimes, about ten years after start-up. At this time the accelerator will be equipped to increase the luminosity by an order of magnitude to 1035 cm-2.s-1. The CMS Tracker will then have to cope with several hundred interactions per bunch crossing and fluxes of thousands of charged particles emerging from the 40MHz collisions. This will require major developments of detector technology and R&D has begun to address the expected challenges. Among the most important are the radiation tolerance of sensors and other components, the provision and distribution of power for both electronics and sensors, the removal of heat loads, which may significantly exceed those in the present detector, and the development of low power, but highly performing electronics in more advanced technologies. CMS has also identified another novel requirement, which is to provide tracker data to contribute to the first level trigger, which must maintain the 100kHz rate for compatibility with existing sub-detector systems while increasing the trigger decision latency by only a few s. The motivation for the upgrade and recent progress in several aspects of the R&D will be described, as well as the current status of designs of a new Tracker which could meet the SLHC requirements.
N13-176:

A Muon Detector Based on Extruded Scintillators, WLS Fibers and GM-APD Readout for a Super B

Factory

G. Cibinetto1, R. Calabrese1, W. Baldini2, L. Tomassetti2 1 University of Ferrara and INFN, Ferrara, Italy 2 INFN Ferrara, Ferrara, Italy The existing muon detectors for high energy physics (HEP) experiments are mainly made of gas chambers such as Resistive Plate Chambers (RPC), Limited Streamer Tubes (LST) or Multy Wire Proportional Chambers. With the increasing of the luminosity in the new accelerators, RPCs and LSTs are not capable to maintain good performances due to the consequent increasing of the particles rate. Moreover the aging process due to the charge accumulation can be a serious issues. The MWPC are rather expensive to cover the huge areas of the new HEP experiments. Extruded scintillators together with wavelength shifter (WLS) fibers can be the active core of a new kind of muon system for high energy physics experiments. The light collected by the WLS fibers is readout by Geiger-Mode Avalanche Photo-Diodes (GM-APD) which are cheaper with respect to the PhotoMultipliers and less sensitive to temperature and bias variations with respect to the APDs. GM-APDs can also work in a magnetic field and have an outstanding time resolution that is used to measure one of the coordinates of the muon system. This way the two coordinates of a detector plane are readout by the same detector element and the ambiguity in presence of more than one track is also solved. We will report our R&D results for different detector layouts and different GM-APDs (i.e. Hamamatsu MPPC and FBK-IRST SiPM): charge spectra and efficiency measurement will be discussed in association with the noise rate. The WLS fibers parameters (i.e. light yield and time response) are essential to have good detection efficiency and resolution: tests with Kuraray and Bicron fibers will be presented and an extended study of the time resolution will be shown along with its implication on the 91

detector performances. Setup optimization and experimental issues will be also discussed as well as the future plans and perspectives.
N13-177:

The CMS All Silicon Tracker Simulation

V. Cuplov Rice University, Houston, Texas, USA On behalf of the CMS Collaboration The CMS Tracker detector is the world's largest silicon detector with about 201 m2 of silicon strips detectors and 1 m2 of silicon pixel detectors. It contains 66 millions pixels and 10 million individual sensing strips. The quality of the physics analysis is highly correlated with the precision of the Tracker detector simulation which is written on top of the GEANT4 and the CMS object-oriented framework. The hit position resolution in the Tracker detector depends on the ability to correctly model the CMS tracker geometry, the signal digitization and Lorentz drift, the calibration and inefficiency. In order to ensure high performance in track and vertex reconstruction, an accurate knowledge of the material budget is therefore necessary since the passive materials, involved in the readout, cooling or power systems, will create unwanted effects during the particle detection, such as multiple scattering, electron bremsstrahlung and photon conversion. In this paper, we present the CMS Tracker simulation description and studies of its material budget.
N13-178:

CMS RPC Trigger and Detector Performance During the Cosmic Run at Four Tesla (CRAFT)

R. Trentadue Universit degli Studi di Bari, Bari, Italy On behalf of the CMS Collaboration / RPC group During the period from september to november 2008 CMS experiment at CERN has been busy in a long taking data, called Cosmic Run At Four Tesla (CRAFT), where several millions of cosmic muons have been collected. Among the most relevant goals to be achieved, the CMS detector calibration, synchronization and performance study are included. CMS RPC system have been careful monitored and studied in oder to enhance all the possible malfunctionalities and to evaluate the quality of its performace. Dedicated tools in order to manage the workflow of data analysis and algorithms to measure the detector and trigger efficiency have been implemented. Algorithm validation by means MC simulation has been performed as well. Moreover a thorough study concerning the noise and the possible effect on the trigger rate and the CMS physics has been successfully accomplished. Detailed results concerning all the issues listed before are reported.
N13-179:

T. Hemperek1, D. Arutinov1, M. Barbero1, R. Beccherle2, G. Darbo2, S. Dube3, D. Elledge3, D. Fougeron4, M. Garcia-Sciveres3, D. Gnani3, V. Gromov5, M. Karagounis1, R. Kluit5, A. Kruth1, A. Mekkaoui3, M. Menouni4, J. D. Schipper5, N. Wermes1 1 University of Bonn, Bonn, Germany 2 University of Genova, Genova, Italy 3 LBNL, Berkeley, USA 4 CPPM Aix-Marseille University, Marseille, France 5 NIKHEF, Amsterdam, The Netherlands With the high hit rate foreseen for the innermost layers at an upgraded LHC, the current ATLAS Front-End pixel chip FE-I3 would start being inefficient. The main source of inefficiency comes from the copying mechanism of the pixel hits from the pixel array to the end of column buffers. A new ATLAS pixel chip FE-I4 is being developed in a 130 nm technology for use both in the framework of the Insertable B-Layer (IBL) project and for the outer layers of Super-LHC. FE-I4 is 80 by 336 pixels wide and features a reduced pixel size of 50 by 250 m2. In the current design, a new digital architecture is introduced in which hit memories are distributed across the entire chip and the pixels organized in regions. Additional features include neighbor hit checking which allows a timewalk-less hit recording.
N13-180:

Digital Architecture of the New ATLAS Pixel Chip FE-I4

D. Pinci1, R. Antunes Nobrega1, V. Bocci1, E. Furfaro2, G. Martellotti1, G. Penso1,2 1 INFN Sezione di Roma, Roma, Italy 2 Sapienza Universita' di Roma, Roma, Italy

Performance of the MWPC of the First Station of the LHCb Muon System

The LHCb Muon System is composed of five detection stations (M1-M5) equipped with a total of 1368 Multi-Wire Proportional Chambers (MWPC) and 12 triple-GEM based detectors. The first station (M1) is placed upstream of experiment calorimeters and 92

thus is equipped with very ``light'' MWPC made by only two gas-gaps (instead of four) and with triple-GEM based detectors. The chambers have to provide a high detection efficiency, a fast response and a good space resolution. For the first time the performance of different M1 MWPC completely equipped with the final Front-End Electronics, have been tested with cosmic rays. The detection efficiency, the time resolution and the hit multiplicity were measured. Chambers with anode-wire-readout and with cathode-pad-readout were tested and the results are compared. The results obtained will be used to individuate the optimized working conditions of the apparatus and to make a more realistic detector description in the LHCb Monte Carlo simulation.
N13-181:

Y. Chang1, C.-H. Wang1, Y. B. Hsiung2, C.-Y. Wang3, H.-T. Hu3, M.-H. Cheng1 1 Dept. of Mechanical Engineering, National United University, Miaoli, Taiwan 2 Dept. of Physics, National Taiwan University, Taipei, Taiwan 3 Dept. of Civil Engineering, National Cheng Kung University, Tainan, Taiwan

The Stress Analysis on the Acrylic Vessels in Dayabay Experiment

Acrylic material has become an important structural material of neutrino detector in recent years. However, it is difficult to obtain reliable data for the mechanical properties of acrylic. Those properties are very important parameters in design and simulation of the acrylic components. We report here the mechanical property measurements of acrylic used in Dayabay experiment, and evaluate the design of the acrylic vessels of the antineutrino detector of Dayabay experiment. The static and dynamic stress distributions of the vessels and the vibration of the vessels during transportation would also be reported here.
N13-182:

First Results of the Engineering Prototype of the CALICE Tile Hadron Calorimeter

M. Reinecke DESY, Hamburg, Germany On behalf of the CALICE collaboration A new prototype of a tile hadron calorimeter (HCAL) for the International Linear Collider (ILC) detector is currently developed within the CALICE collaboration. The aim is to improve the energy resolution by measuring details of the shower development (particle flow). The prototype is based on scintillating tiles that are read out by novel Multi-Pixel Geiger-Mode Photo Diodes (MGPDs). The full prototype will contain about 2200 detector channels (one layer) and takes into account all engineering design aspects that are demanded by the intended operation at the ILC. For technical manageability reasons, the detector is subdivided into basic units (HCAL Base Unit: HBU) with typically 144 detector channels (scintillating tiles) and sizes of 36 x 36 cm each. In order to reduce the amount of non-absorbing material in the calorimeter, the thickness of the prototype is optimized to only 5.4mm. The presentation will focus on the first electrical and mechanical integration experiences of the first two realized HBUs. The operation with a newly developed data acquisition, based on a commercial FPGA board and two newly developed mezzanine modules, is described. Additionally, first results for an integrated light-calibration system (LCS) are shown, addressing the critical distribution of the fast trigger signals (pulse width of about 40ns) to all channels. The LCS is used for calibration and gain monitoring, addressing the temperature and bias dependencies of the MGPDs gains.
N13-183:

N. Amram1, D. S. Levin2, Y. Benhammou2, T. Dai2, E. Diehl2, E. Etzion1, J. Gregory2, J. Hindes2, C. Ferretti2, R. Thun2, C. Weaverdyck2, A. Wilson2, B. Zhou2 1 Raymond and Beverly Sackler School of Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel 2 Department of Physics, University of Michigan, Ann Arbor, Michigan, USA The ATLAS muon spectrometer contains two types of precision-measurement chambers and two types of muon trigger chambers. The Monitored Drift Tubes (MDTs) consist of more than 1200 chambers laid out both in the spectrometers endcaps and barrel, containing a total of 350,000 drift tubes. The tubes are three cm diameter, one to six meter length, and filled with a three bar mixture of 93/7 Ar/CO2. The composition of the gas affects the performance of the MDT chambers dramatically, and as such must be carefully monitored. The Michigan-Tel Aviv gas monitoring system is a mini MDT chamber located in the ATLAS gas facility at CERN. The system monitors simultaneously the input feed and exhaust gas to the complete MDT system, sensitive to changes of about 1 part per mil. The system has been working continuously for over 2 years and provides accurate drift-time measurements for the gas, allowing good calibration of the radius-time (RT) relation of the drift tubes. The ambient humidity of the ATLAS cavern has affected the stability of the MDT gas, causing drift-time variations that roughly follow changes in the cavern humidity. The injection of 1000 parts per million of water, which is meant to improve the lifetime of the Noryl endplugs used in the MDTs, is seen to increase the maximum drift time by 71 ns, consistent with expectations from Garfield simulations of drift tube performance. In addition, we show that the measurement of maximum drift time can be used to infer the amount of water in the MDT gas. We will present a detailed study of the effects both humidity and water injection on the gas performance.

The Effects of Water Injection on MDT Gas Performance

93

N13-184:

Commissioning of CMS with Beam and Cosmic Ray: Strategy and Performance Results

P. Paolucci INFN of Napoli, Napoli, ITALY On behalf of the CMS collaboration board The status of the CMS experiment is described. After a brief review of the detector design and construction of the Compact Muon Solenoid (CMS) experiment, the installation of its subsystems and the general commissioning strategy is presented. Particular emphasis is made on the performance measured with cosmic muons prior and after the collision data-taking and on the observation of the first beam events. Finally the talk will focus on the readiness of CMS for the first physics with coliding beams.
N13-185:

The CMS Tracker Calibration, Methods and Experience with Cosmic Ray Data

G. Kaussen Universitaet Hamburg, Institut fuer Experimentalphysik, Hamburg, Germany On behalf of the CMS Collaboration The CMS Silicon Strip Tracker (SST), comprising 9.6 million readout channels from 15 148 modules covering an area of 198m2, needs to be precisely calibrated in order to correctly interpret and reconstruct the events recorded from the detector, ensuring that the SST performance fully meets the physics research program of the CMS experiment. Calibration constants may be derived within several workflows, from promptly reconstructed events with particles as well as from commissioning events gathered just before the acquisition of physics runs. These calibration procedures have been exercised since summer 2008, when the CMS detector has been commissioned using cosmic muons with and without magnetic field. During these data taking the performance of the SST have been carefully studied: the noise of the detector, together with its correlations with the strip length and the temperature, the data integrity, the S/N ratio, the hit reconstruction efficiency, the correlation with the trigger patterns have been all investigated with time and for different conditions, at the full detector granularity. In this paper we describe the reconstruction strategies, the calibration procedures and the detector performance results from the latest CMS operation.
N13-186:

Front End Electronics for Compact Silicon-Tungsten Calorimeter FOCAL

A. Y. Sukhanov Physics, Brookhaven National Laboratory, Upton, NY, USA On behalf of the PHENIX Forward Upgrade Group The compact silicon-tungsten sandwich sampling calorimeter FOCAL is one of a major planned upgrades of PHENIX detector at RHIC. The device is built from bricks, each brick is composed of between 8 and 14 super-towers Each supertower is a silicontungsten sandwich composed of silicon layers divided into 16 1.5x1.5 cm2 pads/layer, and 4mm thick tungsten plates . The structure is read out in 3 longitudinal segments, EM0, EM1 and EM2. In addition EM0 is augmented with 4 x-y planes of 0.5 mm strips read out individually for the reconstruction of overlapping showers. In our work we present 1) simulation, design and test results of signal conditioning and digitizing electronics for pad sensors; 2) design and test results of digital readout electronics for strip sensors 3) generation of trigger primitives based on pad sensors; 4) results of the test beam run with a full-size prototype. The strips are digitized using SVX4 ASIC. For digitization of pad sensors we use 14bit ADCs, located 10m from hybrid preams. The calorimetric towers are formed by adding signals from 7 pads at preamp inputs. To reduce system noise we drive signals from preamps differentially. The differential shaper on receiving end is used to filter out high frequency noise as well as reflections due to mismatched impedances. The cut-off frequency is chosen low enough to enable for several samplings of welldefined signal shape, doing this we triple the signal-to-noise ratio. For trigger generation we transfer all signals from each segment to a single FPGA which makes fully pipelined calculations of trigger primitives related to the segment. The primitives from each segment are sent to a central FPGA, it generates decisions such as energy and coordinates of showers and jets, em/hadron identification, total energy. All electronics is remotely configurable using ethernet JTAG microcontrollers, the same controllers are used for test data acquisition and for online monitoring.
N13-187:

A Method for Designing Cable Equalizers

M. E. Johnson, M. J. Utes Particle Physics Div, Ferm National Accelerator Laboratory, Batavia, IL, USA A method for designing cable equalizers is described. Equalizers compensate for the frequency-dependent attenuation characteristics of cables by providing amplification that just compensates for these attenuation characteristics. This method uses an approximation for the poles and zeros in an amplifier transfer function. It then utilizes a Monte Carlo technique to randomly select the poles and zeros in the chosen frequency band, converts these poles and zeros to component values and then minimizes the RMS of the difference between the inverse of the cable attenuation curve and the actual circuit transfer function. This method can be used for a wide range of circuits. In a slightly modified form it can be used to generate a model of a cable for circuit 94

simulation programs such as SPICE. We present an example circuit using three capacitors and three resistors in a 3 pole and 3 zero configuration which can fit many types of cables over a wide frequency range. The Monte Carlo minimization for component values converges reliably to stable values.
N13-188:

The EUDET High Resolution Beam Telescope - the Final Digital Readout

E. P. Corrin DPNC, University of Geneva, Geneva, Switzerland On behalf of the EUDET Collaboration A high resolution ( < 3 m) beam telescope based on monolithic active pixel sensors (MAPS) has been built within the EUDET collaboration. EUDET is a coordinated detector R&D programme for a future linear collider providing test beam infrastructure to detector R&D groups. The telescope consists of six sensor planes with a pixel pitch of either 10 or 30 m for the demonstrator, or 18.4 m for the final telescope. These are located on two arms, between which a device under test may be positioned. A general purpose cooling and positioning infrastructure is available, along with a custom-made trigger logic unit, a flexible data acquisition system based on dedicated VME readout boards, a platform-independent, lightweight DAQ framework, and a data analysis tool based on the standard ILC software framework. Since the first installation of a demonstrator telescope in 2007, the DAQ system has been continuously improved and adapted to new sensor types, and has been used by a total of about ten groups over the summers of 2008 and 2009 as a reference system for tests at DESY and the high energy hadron test beam facility at CERN. In 2008 the sensors were upgraded to the high-resolution Mimosa18 chips, providing the user with the option of enhanced resolution, at the expense of readout speed. In parallel with the 2009 test beam campaign, the final sensors will be tested, allowing the commissioning of the final telescope soon afterwards. After an overview of the pixel telescope, its data acquisition system and its performance, the commissioning of the final telescope will be presented. The main feature is the Mimosa26 sensor chip that includes on-sensor data sparsification and digital readout, allowing greatly enhanced readout speed, while maintaining good resolution.
This work is supported by the Commission of the European Communities under the 6th Framework Programme "Structuring the European Research Area", contract number RII3-CT-2006-026126. N13-189:

Performance of the ATLAS Hadronic Tile Calorimeter at the LHC Startup

C. Solans Sanchez IFIC (Universitat de Valencia - CSIC), Valencia, Spain On behalf of the ATLAS Collaboration The Tile hadronic calorimeter is the central hadronic barrel calorimeter of the ATLAS experiment at CERN. Its performance has a direct impact on physics for the LHC start up. Good pulse timing is required for good off-detector energy reconstruction, where the digital samples are calibrated to the electromagnetic scale. We present the current status on detector timing with latest results from beam data, initial electronic noise characterization with cosmic data and the status of the energy intercalibration to the electromagnetic scale.
N13-190:

Tile AHCAL Test Beam Analysis: Positron and Hadron Studies

R. Fabbri FLC/Calice, DESY, Hamburg, Germany On behalf of the CALICE collaboration The CALICE collaboration has constructed a hadronic sandwich calorimeter prototype with 7608 scintillating plates, individually read out by multi-pixel silicon photomultipliers (SiPMs). For the first time ever the read out is performed using SiPMs on a large scale. Results of test beam operations with muon, positron and hadron beams at CERN are presented here, validating the feasibility of the novel SiPM technology. The calibration procedure, at a cell-by-cell basis, is presented. Hadron shower development is investigated in terms of longitudinal and lateral profiles, and compared with Monte Carlo simulations. Exploiting the unprecedent high hadron calorimeter granularity also allows to measure the profiles with respect to the point of the first hadronic interaction. Results of the application of the "particle flow approach" in shower energy reconstruction are also presented, using real data.

95

N13-191:

The Upgrade R&D of the CMS HCAL HE and HF for Super-LHC

U. Akgun Physics, University of Iowa, Iowa City, Iowa, USA On behalf of the CMS Collaboration The luminosity of Large Hadron Collider is planned to increase each year, reaching to 1035 cm-2s-1 in 2020. As the integrated luminosity of the LHC increases CMS hadronic calorimeter will need some detector upgrades. For CMS Hadronic Endcap (HE) calorimeter we propose to substitute the scintillator tiles in the original design with quartz plates. To enhance the light production we treated the quartz plates with p-Terphenyl, and constructed a 20 layers calorimeter prototype. The CMS Hadron Forward (HF) calorimeters currently use quartz fibers, with a plastic clad, as the active component. Here we discuss the possible solution scenarios for radiation damage of fibers and abnormal events produced on the PMT window. We report from the R&D studies on PMT and quartz fiber replacement candidates for the HF calorimeter.
N13-192:

Clusterless Data Analysis for Position Sensitive Detector Characterization

T. Maenpaa, M. J. Kortelainen, T. Lampen Helsinki Institute of Physics, Helsinki, Finland The luminosity upgrade of the CERN Large Hadron Collider (LHC) is one of the driving forces of development of new radiation hard detectors. As part of this research, performance of severely irradiated detectors need to be measured. The final characterisation of these detectors is carried out in a particle beam whose tracks are measured with reference detectors. Signals of the detectors are compared with positions of reference tracks, which allows the measurements of eg. resolution, efficiency and predictive value in addition to signal and noise levels. The devices tested in research are often hand-made prototypes. To speed up the process of sensor development, it should be possible to measure the characteristics in a standardized way even if there are issues in manufacturing that do not typically appear in well-estabilished mass production. The test beam data analysis traditionally comprises pedestal and noise level calculation, cluster construction, hit position assignment and comparison to reference data. This approach corresponds to the actual usage of position-sensitive detectors, and works well provided that the detectors perform as expected. In case a detector performs in a different way, the result of this kind of analysis can be seriously biased and can provide very little useful information. We present a robust and simple approach for detector characterization. In this approach cluster construction is omitted to avoid the related bias. The clusterless analysis complements the traditional analysis and provides a standardized way to compare poorly performing detectors. The clusterless testbeam data analysis is described and compared to traditional methods. Scenarios where the clusterless analysis is beneficial are discussed.
N13-193:

Comparison of Simulated and Measured Charge Transfer Inefficiencies in a CCD with High-Speed Column Parallel Readout

A. Sopczak Physics, Lancaster University, Lancaster, UK On behalf of the LCFI Collaboration Charge Coupled Devices (CCDs) have been successfully used in several high energy physics experiments over the past two decades. Their high spatial resolution and thin sensitive layers make them an excellent tool for studying short-lived particles. The Linear Collider Flavour Identification (LCFI) Collaboration has developing Column-Parallel CCDs for the vertex detector of a future Linear Collider which can be read out many times faster than standard CCDs. The most recent studies are of devices designed to reduce both the CCDs intergate capacitance and the clock voltages necessary to drive it. A comparative study of simulated and measured Charge Transfer Inefficiency values for a range of readout frequencies and operating temperatures is presented.
N13-194:

Construction of a Large Scale Prototype for a SiW Electromagnetic Calorimeter for the ILC EUDET Module

Calice The CALICE collaboration is preparing large scale prototypes for highly granular calorimeters for detectors to be operated at the International Linear Collider, ILC. During the years 2009/10 a prototype of a SiW electromagnetic calorimeter will be assembled which in terms of dimensions and layout meets already most of the requirements given by the ILC Physics Program and hence 96

the detector design. In particular the Very Front End electronics will have to fit within alveolar layers of less than 1 cm in height. In this contribution the design of this protoype is outlined and the steps towards the realisation will be presented. At the time of the conference it is expected that results on thermal dissipation and mechanical stress the module will suffer from can be reported. These results will be obtained with a first version of the EUDET module which is constructed during spring 2009.
N13-195:

The Status of the Precursor Prototype for the PANDA-TPC and the PANDA GEM-Trackers

B. Voss Dept. Detector Laboratory, GSI Helmholtzzentrum fuer Schwerionenforschung GmbH, Darmstadt, Germany On behalf of the JointGEM Collaboration In 2007 the construction and building of a medium volume (300mm x 650mm) Time Projection Chamber (TPC) based on Gaseous Electron Multipliers (GEM) as amplification stages has been started. The detector is foreseen to be integrated in several experiments e.g. at CB-ELSA/Bonn as well as at FOPI/GSI-Darmstadt. It serves as a precursor prototype for an even larger TPC (840mm x 1500mm) to be incorporated in PANDA/FAIR as a central tracking device which will be built in the frame-work of a joint venture project of twelve institutions in Europe. This detector will cover the angels between -45 and 22. The forward region below 22 will be measured by three to four planar-GEM trackers of large area (up to 1500mm) installed 810mm to 1900mm downstream the interaction point inside the target spectrometer. The general design of these detector systems as well the current status of their assembly and testing will be presented.
N13-196:

M. T. Koch1, P. Fischer2, S. Furletov1, H. Krueger1, C. Kreidl2, I. Peric2, N. Wermes1 1 Dept. for Physics, University of Bonn, Bonn, Germany 2 Dept. for Computer Engineering, University of Heidelberg, Heidelberg, Germany

Development of a Fast Readout System for DEPFET Pixel Sensors

DEPFET pixel sensors are based on a fully depleted high resistivity silicon substrate, and offer first stage low noise in-pixel amplification by integrating a field effect transistor directly into every pixel. Signal charges created by impinging particles are collected in a potential minimum below the FET, thereby modulating the transistor current. DEPFET sensors are an integrating device, readout is non-destructive, and they feature a special clearing process to remove the collected charge. The DEPFET pixel sensor is currently being developed for a pixel detector for the upgrade of the Belle experiment (Belle II) at the High Energy Accelerator Research Organization (KEK) in Japan. The physics goals require radiation hard large area detectors with low material budget, low power consumption and fast readout. This talk concentrates on the development and results of a prototype readout system which addresses specifically the readout speed of the DEPFET sensors. This prototype readout system consists of the DEPFET sensor matrix itself, steering ASICs (SWITCHER3), and a newly developed readout ASIC (DCD2). The DCD2 chip digitizes the DEPFET signal current employing algorithmic current-based ADCs with eight bit resolution in every input channel. The DCD2 chip is successfully operated at its maximum design data rate. A line rate of 12.5MHz corresponding to a sampling time of 80ns can be achieved with reasonable noise figures. Furthermore, a newly developed Virtex4 FPGA-based control and data acquisition system allows operation at a burst data rate of 14.4Gbit/s per 144 DCD2 input channels.Detailed measurements of the DCD2 readout chip and performance with a connected DEPFET sensor will be presented.
N13-197:

B. Bilki1, J. Butler2, G. Drake3, E. Hazen2, J. Hoff4, G. Mavromanolakis4,5, E. May3, E. Norbeck1, J. Repond3, D. Underwood3, L. Xia3, Q. Zhang3,6 1 Department of Physics & Astronomy, University of Iowa, Iowa City, IA, USA 2 Physics Department, Boston University, Boston, MA, USA 3 HEP Division, Argonne National Laboratory, Argonne, IL, USA 4 Fermi National Accelerator Laboratory, Batavia, IL, USA 5 Cavendish Laboratory, University of Cambridge, Cambridge, UK 6 Institute of High Energy Physics, Beijing, PRC We present the concept of a Digital Hadron Calorimeter (DHCAL) for use in a detector optimized for the application of Particle Flow Algorithms to the measurement of jet energies. Resistive Plate Chambers (RPCs) with 1 x 1 cm^2 readout pads are used as active elements. The front-end electronic readout is integrated on the pad-boards of the chambers and applies a single threshold (1-bit) to the signal charges, hence the designation of digital readout. We report on detailed measurements with a small scale prototype in the Fermilab test beam using muons, positrons, pions, and protons and in the laboratory using cosmic rays. The results validate the concept and serve as basis for the design of a large prototype calorimeter. 97

Tests of a Digital Hadron Calorimeter

We also report on the status of the ongoing construction of the larger prototype calorimeter, of the size of a cubic meter. We expect to test this prototype calorimeter at Fermilab test beam within a year.
N13-198:

E. Castro1, A. J. Bell2 1 CMS, DESY, Zeuthen, Germany 2 BCM BRM, CERN, Geneva, Switzerland

Data Readout Concept and Test Results for the BCM1F Adn BSC Sub-Detectors of CMS

At CMS, several Beam Conditions Monitors will provide, with different time resolution, permanent radiation monitoring within the CMS experiment, preventing the ITS operation in adverse beam conditions. The two sub-detectors with the best time resolution are the BSC (Beam Scintillator Counters) and BCM1F (Fast Beam Condition Monitoring) . The BSC is made of 36 PVT scintillation plastic tiles that cover ~ 1.2m2 area at 10.86m from the IP and that are mounted on the two Hadron Forward calorimeters whilst the BCM1F is composed by two carriers of 8 radiation-hard single crystalline diamonds placed at 1.8 from the IP inside the CMS pixel detector. Both subsystems cover areas of different expected radiation load, however they share two main features: they have to supply better bunch-by-bunch resolution and to be MIP sensitive. This suggests a similar data acquisition and data processing systems. The current data acquisition is a VME based system that includes: Time-to-Digital Converter (TDC) to record hit rates and precise timing information; Analog-to-Digital Converters (ADC) that sample the signal and supplies pulse height and shape; and scalers to supply information about the hit rates. During the successful initiatory run of the LHC in September 2008, these BCM sub-detectors recorded successfully recorded data, allowing first studies of their performance. In 2009 several improvements are done and realistic tests in the laboratory where performed. A summary of the performance studies of the whole system is given and first data after re-startup of LHC may be shown.
Acknowledgements to CMS BCM BRM group N13-199:

Studies of Scintillation Plates with Waveshifter Fiber and SiPM Read Out

J. Marchant, B. Baumbaugh, M. Kirzeder, S. Mathews, E. Fidler, R. Ruchti, E. Shearer, M. Vigneault, T. Williamson Physics Department, University of Notre Dame, Notre Dame, IN, USA We have been studying the performance of SiPM devices and scintillation plates read out with waveshifter fiber and SiPM. Applications are for the detection of ionizing radiation, for example in particle and nuclear physics experiments. An example is to use SiPM devices to transduce waveshifted scintillation signals from scintillation tiles with embedded waveshifter fiber containing Y11 and newer, fast shifters such as the DSB2 family of fluorescent dyes. Results are presented on performance of the SiPM using pulsed LEDs and tile/fiber detectors excited by radioactive sources.
N13-200:

The Sea Bed Power and Data Transmission Network for the KM3NeT Neutrino Telescope

R. Papaleo Laboratori Nazionali del Sud, Istituto Nazionale di Fisica Nucleare - INFN, Catania, Italy On behalf of the KM3NeT Consortium The KM3NeT EU-funded consortium, pursuing a cubic kilometre scale neutrino telescope in the Mediterranean Sea, is developing technical solutions for the construction of this challenging project, to be realized several kilometres below the sea level. The power supply system, the data transmission and distribution network will be presented with focus on: main electrooptical cable, deep sea power conversion, subsea data and power distribution network, connection systems issue.
N13-201:

J. Va'vra1, D. Leith1, B. Ratcliff1, K. Nishimura2, L. Ruckman2, G. Varner2 1 Stanford University, SLAC, Menlo Park, USA 2 Department of Physics & Astronomy, University of Hawaii, Honolulu, USA

Progress in Development of the TOF Detector with a New ASIC-Based Waveform Digitizing Electronics

Abstract - We are developing a novel Time-of-Flight detector for a possible use at the Super-B factory. Using two identical 64pixel Photonis Microchannel Plate Photomultipliers and the Ortec electronics, to provide start and stop signals, each having a 1 cm-long quartz Cherenkov radiator, we have already achieved a timing resolution of _Single_detector ~14 ps in the Fermilab beam; with a laser, we obtained ~15-20 ps resolution for equivalent charge. The novel feature of this contribution are new laserbased test results with a new waveform digitizing electronics, based on the TARGET ASIC chip, which allows a waveform 98

digitization with sampling every 400 ps. The TOF prototype is also going to be tested in a large cosmic ray telescope providing muon tracks with ~1mrad angular resolution and energy cutoff of ~1.5 GeV. The paper will present description of detector, electronics and the first results.
N13-202:

J. Hasi1, S. Watts1, R. Thompson1, S. Kolya1, C. Da Via1, C. J. Kenney2, S. I. Parker3, E. Westbrook4 1 University of Manchester, Manchester, UK 2 SLAC National Accelerator Laboratory, Menlo Park, CA, USA 3 University of Hawaii, Honolulu, HI, USA 4 Molecular Biology Consortium, Chicago, IL, USA Proper thermal and mechanical design of large scientific instruments such as collider detectors and telescopes is crucial to their success. Although computer simulations allow a systems parameter space to easily explored, construction of prototype physical models is an important stage in the design flow. A silicon chip, with the same thermal and mechanical properties as the ATLAS pixel front-end read-out circuit has been fabricated. It allows thermal loads to be generated in both the analog and digital sections independently. Having the same mechanical dimensions and wire-pad locations enables realistic mockups of the final system with inexpensive and easily operated components. These devices are being used to design the assemblies for the ATLAS forward physics experiment at the Large Hadron Collider.
N13-203:

Thermal-Mechanical Mimic of the ATLAS Read-Out Chip

The ATLAS Planar Pixel Sensor R&D Project

D. Muenstermann Physik E 4, TU Dortmund, Dortmund, Germany On behalf of the ATLAS Planar Pixel R&D collaboration According to the updated schedule of LHC, the luminosity delivered to the ATLAS detector might exceed the design luminosity of 1034 cm-2 s-1 already in 2013 and reach 3 1034 cm-2 s-1 around 2016. A further 'phase 2' upgrade could take the luminosity to 1035 cm-2 s-1 around 2020. The inner tracker of ATLAS was not designed for such high fluences and will suffer from both, radiation damage and high occupancies. To cope with the accelerator upgrades, a two-step approach is foreseen: In 2013, a new innermost pixel layer (Insertable B-Layer IBL) with a radius of only 3.7 cm from the beam will be installed. In 2017, a full replacement of the inner tracker is planned. For both enterprises, planar pixel sensors might be used if it can be shown that they will survive the harsher radiation environment and can be produced at affordable cost to instrument a larger area. To investigate the behaviour of pixel sensors using the proven planar technology, an ATLAS R&D project was established comprising 16 institutes and more than 70 scientists. Main areas of research are the performance of planar sensors at fluences of up to 2 1016 neq cm-2 and the exploration of possibilities for cost reduction. The presentation will give an overview of the agenda of the R&D project and present first results in the fields of charge collection efficiency, test beam activities, new sensor designs, low-threshold studies and cost reduction activities.
N13-204:

Performance of the CMS Silicon Strip Tracker Local Reconstruction

N. A. Cripps Dept. of Physics, Imperial College London, London, United Kingdom On behalf of the CMS Collaboration The CMS silicon micro-strip tracker is unprecedented in terms of its scale and complexity, comprising 16 thousand front-end modules and 9.6 million analogue read-out channels. The reconstruction of hit positions and tracks with high efficiency is extremely challenging, due to the high particle multiplicity of the LHC environment, the scale of the read-out system and the enormous amounts of raw data produced.A demanding constraint is the limited time budget available to make a software-based high-level trigger decision using tracks, which is achieved using a novel, regional approach to reconstruct hits. Additionally, the complexity of the read-out system requires procedures to detect and handle faults and error conditions in order to maintain data integrity and minimize the effect on track reconstruction. We report on the methods developed to meet the challenges of local reconstruction and the tools developed for error detection and handling.

99

N13-205:

The Design, Construction and Testing of a 2.1m Long Straw Detector Prototype Operated in Vacuum for the NA62 Experiment

H. Danielsson PH, CERN, Geneva, Switzerland On behalf of the NA62 Collaboration We have designed, built and tested a full-length prototype (2.1m) for the straw tracker in NA62 (rare Kaon decay). To minimize the multiple scattering of the outgoing pion, it is proposed to operate the magnetic spectrometer in vacuum, upstream of the decay region. This means that the decay and spectrometer regions are not separated and share a common vacuum volume. In this paper we present some of the considerations, which led to the present design in terms of detector layout, material choice and mechanical design. Very tight requirements on straw straightness, wire positions and leak tightness put great demands upon design and assembly. In order to verify the design, a 64-straw prototype was built. The straws have a diameter of 10 mm and are filled with a gas mixture of 84% CO2, 10% CF4 and 6% Isobutane. The overall mechanics as well as the detailed design of individual components are presented. In particular, the measured characteristics of the straws tubes are shown. Furthermore, the principles of dedicated assembly and quality control tooling, e.g. straw straightness, wire tension and leak rate are discussed. Finally, the results from the experimental validation including high voltage performance and test beam studies of the front-end electronics are presented.
N13-206:

M. Koutalonis1, E. Cook2, J. Griffiths2, J. Horrocks1, C. Gent3, S. Pani1,4, L. George3, S. Hardwick3, R. Speller2 1 Department of Clinical Physics, Barts and the London NHS trust, London, UK, London, United Kingdom 2 Department of Medical Physics and Bioengineering, University College London, London, United Kingdom 3 Home Office Scientific Development Branch, Sandridge, United Kingdom 4 Department of Physics, University of Surrey, Guildford, United Kingdom

Designing an in-Field Diffraction System for Illicit Drug Detection Using X-Ray Diffraction

One way of smuggling drugs into a country is via the postal and courier services. Automated systems are necessary to scan incoming parcels and make quick decisions on whether they contain drugs or not. Few false positive and negative results are an important requirement for the end users of such a system, as neither parcels containing drugs should be lost nor parcels without drugs should stop the workflow. According to previous studies, x-ray diffraction has demonstrated the potential to meet this requirement, as it has shown high ability in identifying drugs, compared to other methods. This is mainly due to the crystalline pattern of the drugs and their unique diffraction signature. The same technique has also been applied in explosives and calculi identification in the past with great success. In this study, a simulation model was developed performing energy dispersive x-ray diffraction on the powder diffraction profiles of several materials that could be found in a common parcel. A database containing several thousands of such materials has been collected. In this way, the profile that would be obtained by an x-ray diffraction system can be extracted, taking into account several geometric parameters, as well as various detector energy resolutions. The aim of this study was to test several possible in-field systems for drug identification and decide on the optimum that will be developed in the lab. To this direction, several geometries (including distances, collimation, scattering angles etc.), x-ray spectrum energies and detector energy resolutions (HPGe, CZT, Si and NaI) were tested. A variety of parcel sizes and compositions were designed and simulated and the results were analyzed using Multivariate Analysis (MVA). Results showed that several geometries and detectors can lead to a system with high sensitivity and specificity. The next step of this study is the development of these systems in the lab.
N13-207:

Isotope Identification in the GammaTracker Handheld Radioisotope Identifier

C. E. Seifert, M. T. Batdorf, W. K. Hensley, D. S. Barnett, B. J. Burghard, L. J. Kirihara, S. J. Morris, M. J. Myjak Pacific Northwest National Laboratory, Richland, WA, USA Isotope identification is a critical function of handheld spectroscopic radiation detectors. At a minimum, the ability to distinguish naturally occurring radioactive materials and medical radiopharmaceuticals and from potential radiological and nuclear threats is essential to widespread deployment of handheld radiation detection systems. The GammaTracker handheld radioisotope identifier uses a peak-based isotope identification method implemented on an embedded computing platform within the device. Peak-based isotope identification has an advantage over template-based methods in that a comprehensive library of measured spectral templates is not required. In fact, a peak-based method can identify a radionuclide not previously measured, as long as the corresponding gamma-ray energies are tabulated. In this work, we present the spectral smoothing, peak search, and multiplet decomposition methods used for isotope identification and describe their implementation on the GammaTracker platform. Performance of the identification algorithms is presented using measured data from gamma-ray sources and compared to the performance obtained using standard commercial packages on a PC platform.

100

N13-208:

Low Count Spectral Anomaly Detection Algorithm for Search Applications

B. D. Milbrath, D. M. Pfund, K. D. Jarman Pacific Northwest National Laboratory, Richland, WA, USA A spectroscopic radiological detector system operating in a drive-around search mode will encounter several difficulties. Not only must it deal with nuisance alarms (such as medical isotopes) and low-count statistics, but it will also observe a changing background with systematic variations that far exceed Poisson scatter about a constant mean. We developed an algorithm specifically to address these issues. In particular the algorithm operates on vectors of spectral comparison ratios to maximize the detection of certain threats and to minimize the sensitivity to the variable benign background. As part of the development of the Standoff Radiation Imaging System (SORIS) being built by General Electric as part of DNDOs Standoff Radiation Detection System Advanced Technology Demonstration program, we have modified our earlier algorithm work substantially to perform better in urban search environments based on search data collected in Seattle. Modifications to be reported on include optimization of signal integration time, rejection of medical isotopes, and lowering the energy range of interest.
N13-209:

Materials Identification by X-Ray and Photoneutron Transmission

Y. Yang, B. Wu, Y. Li, T. Li Department of Engineering Physics, Tsinghua University, Beiing, P.R.China In this paper, we present the method research of materials identification based on X-ray and photoneutron transmission. X-rays are produced by 7MeV LINAC. Photoneutrons are emitted as the byproduct of X-rays when beryllium convertor is used to convert X-rays to neutrons. The different attenuation coefficients of X-rays and neutrons in materials are used to form a factor, named V, to discriminate materials. The main problems of MIXPT (materials identification by X-ray and photoneutron transmission) come from (1) the interference of X-ray pulses to the detection of photoneutrons and (2) varied V value of materials originated by continuous spectra of X-rays and photoneutrons when penetrating materials; To the first problem, a thermal neutron detector is designed to detect the fast penetrating photoneutrons. With the aid of neutron moderation substances surrounding BF3 detector, fast photoneutrons are thermalized and timely delayed thus could be detected free of X-ray pulses interference and with high detection efficiency. A time window of 950s width after each X-ray pulse is carefully selected according to both simulation and experimental research to measure the penetrating photoneutrons. To the second problem, the result of simulation with MCNP5 shows that each material has a definite curve of V versus mass per unit area and could be discriminated from other materials effectively. The experimental results of polyethylene, aluminium, iron, copper, lead and explosive simulant also confirm these curves; Keywords: X-ray transmission, photoneutron transmission, materials identification
N13-210:

Explosives Detection Using Dual Energy X-Ray Imaging and Photoneutron Analysis

Y. Yang, T. Li Department of Engineering Physics, Tsinghua University, Beijing, P.R.China To find the explosives concealed in cargos, X-ray penetration information and neutron induced information are both helpful. The combination of them would improve the capability of explosives detection in complex situation like metal shielded or other organic materials disturbed. In this paper, we present the research of combining dual energy X-ray imaging and photoneutron induced -ray analysis. The former technique provides the materials identification capability based on effective atomic number that indicates organic, light metal, inorganic and heavy metal areas. The area containing organic material will be suspected further that will undergo neutron analysis. High energy X-ray induced photoneutron analysis provides the 2-dimensional elemental distribution of hydrogen, nitrogen and iron. Area wealthy of nitrogen and hydrogen element may be suspected as to be explosive. Several different shielding situations are researched in the experiments to evaluate the explosive detection capability, including explosive standing alone, shielded by 2cm iron plate, surrounded by sugar and so on. 9MeV and 6MeV X-ray pulses are produced interlaced with 5s pulsed width and 3000rad/min@1m to form dual energy X-ray image. Neutrons are produced by X-ray with 20kg heavy water converter. The experiments results show that areas of organic materials could be first notified by dual energy imaging effectively, then the 2-dimensional elements information of nitrogen and hydrogen are used to confirm the existence of explosive. Key words: dual energy X-ray imaging, photoneutron analysis, explosives detection
N13-211:

Feasibility Study: Low-Cost Dual Energy CT for Security Inspection

Y. Liu, J. Cheng, Z. Chen, Y. Xing Department of Engineering Physics, Tsinghua University, Key Laboratory of Particle & Radiation Imaging (Tsinghua University), Ministry of Education, Beijing, China Dual energy CT (DECT) has become a hot topic among security inspection recently for its high detection precision and robust material identification ability. However, the high cost of the system become a big limitation for its wide usage. In this paper earlier, we proposed a new dual energy CT reconstruction method with reduced data (DECT-RD) requiring much fewer data to 101

reduce the cost of detectors. To fully investigate the feasibility of its application in security inspection, we are developing such an experimental imaging system (DECT-RDS). Practical experiments are done by using only 20% detector bins instead of completesample bins in each projection. Results give an relative error less than 5% between complete sampling and undersampling. The results demonstrated that detector bins can be greatly reduced in dual energy CT imaging, hence much lower the system cost. We believe this type of system configuration will drive DERT into wide practical applications.
This work is supported by the National Natural Science Foundation of China under the project No. 60871084, No. 60772051 and No. 10575059, and the Program for New Century Excellent Talents in University No. NCET-05-0060. N13-212:

P. A. Marleau1, N. Bowden2, S. Mrowka1, J. Steele1 1 Radiation and Nuclear Detection Systems, Sandia National Laboratories, Livermore, CA, USA 2 Lawrence Livermore National Laboratories, Livermore, CA, USA Organic scintillators are widely used for fast neutron (>~100 keV) detection, however these materials are rarely used for spectroscopy purposes for two reasons. First, the majority of neutrons leave the detector before depositing all of their energy. Second, neutrons typically undergo multiple scatters and the scintillators non-linear response results in a non-deterministic relationship between the measured and actual energy deposited. By segmenting the detector, multiple scatters may be resolved and the non-linearity can be corrected. We will present a technique that addresses the issue of full energy deposition and is based on the fact that neutrons deposit an average of half of their energy per scatter. The average fraction of energy deposited can then be estimated as (1 - (1/2)^n) where n is the number of scatters. Using segmentation to select for multiple scatters, events in which a high fraction of the total energy is deposited can be identified.
N13-213:

Selection of Fast Neutron Full Energy Deposition Using Segmentation

E. A. Wulf1, B. F. Phlips1, A. Hutcheson2, W. N. Johnson1, E. I. Novikova1 1 Naval Research Laboratory, Washington, DC, USA 2 National Research Council/Naval Research Laboratory, Washington, DC Large Area Scintillation Array (LASA) is a coded mask instrument housed in a standard 6.1 m (20') refrigerated shipping container making it easily transportable by the worldwide shipping industry and able to operate in most maritime and land environments. LASA consists of a 13x6 array of NaI detectors that are 15 cm in diameter and 15 cm deep with average resolution of 4.3% at 2614 keV. The coded aperture is a random 23x11 array chosen by simulation to minimize artifacts and to be 50% open. Lead blocks 5 cm thick are used to make up the mask elements. The mask and detectors are mounted on opposite walls of the shipping container for maximum separation and the instrument has a pixel image resolution of 5 deg. The fully coded eldof-view of LASA is ~50x30 deg.
N13-214:

Large Area Scintillation Array (LASA)

K. L. Hertz1, D. L. Chichester2, C. E. Holland3, P. J. Resnick4, P. R. Schwoebel5, B. B. Johnson5, B. Reichenbach5, I. Solano5 1 Rad/Nuc Detection Materials & Analysis, Sandia National Laboratories/CA, Livermore, CA, USA 2 Idaho National Laboratory, Idaho Falls, ID, USA 3 Microsystems Development, SRI International, Menlo Park, CA, USA 4 MEMS Core Technologies, Sandia National Laboratories/NM, Albuquerque, NM, USA 5 Physics and Astronomy, University of New Mexico, Albuquerque, NM, USA The use of an Electrostatic Field Desorption (EFD) ion source would constitute a significant advancement in the design and operation of neutron generators. The results would directly benefit the use of neutron generators for active interrogation in the search for Special Nuclear Material (SNM), particularly in man-portable scenarios. The novel EFD approach uses high electrostatic fields to produce pure atomic deuterium ions from a conductive surface, rather than ions produced from a deuterium plasma. This new concept has the potential to surpass current state of the art sealed neutron tube designs in many key performance areas including lifetime, reliability, efficiency, and neutron yield. The neutron generator yield in this design scales with the ion current which scales in proportion to the area of the conductive surface. To maximize ion production and minimize applied voltage, the ion source is comprised of a microfabricated array of field tips based on the geometry of field emission arrays. Electrostatic modeling and simulations have aided the design of the arrays. To date, several iterations of tip arrays have been fabricated and tested. The current tip arrays, relevant modeling, and test results will be discussed.
This work was supported by the U.S. Department of Energy through the National Nuclear Security Administrations Office of Non-proliferation Research and Development (NA-22).

An Electrostatic Field Desorption Ion Source for Active Neutron Interrogation

102

N13-215:

Development of the Neutron Interrogation System for Underwater Threat Detection and Identification

I. Novikov, A. Barzilov Physics and Astronomy, Western Kentucky University, Bowling Green, KY, USA Wartime and terrorist activities, training and munitions testing, dumping and accidents have generated significant munitions contamination in the coastal and inland waters in the United States and abroad. Although current methods provide information about the existence of the anomaly (for instance, metal objects) in the sea bottom, they fail to identify the nature of the found objects. Field experience indicates that often in excess of 90% of objects excavated during the course of munitions clean up are found to be non-hazardous items (false alarm). The technology to detect and identify waterborne or underwater threats is also vital for protection of critical infrastructures (ports, dams, locks, refineries, and LNG/LPG). We propose a compact neutron interrogation system, which will be used to confirm possible threats by determining the chemical composition of the suspicious underwater object. The preliminary results of computer simulation of the system performance (neutron dynamics in underwater environment and gamma ray signatures of various threats) will be discussed.
This work was supported by the U.S. Department of Homeland Security and the National Institute for Hometown Security (contract No. 04-07WKU). N13-216:

W. Bi1, Z. Chen1, L. Zhang1, Y. Xing1, Y. Wang2 1 Department of Engineering Physics, Tsinghua University, Beijing, China 2 Nuctech Co., Ltd., Beijing, China

Real-Time Visualize the 3D Reconstruction Procedure Using CUDA

In some CT systems for online control or inspection, people need to do some operations referring to the median reconstruction results during the CT scanning. It requires a very high computing ability to realize an online reconstruction and immediate volume rendering. In this paper, we present a complete Real-Time Visualization method (RTV) for Cone-beam CT with latest GPU Tesla C1060. Different from the former solutions, our method can visualize the intermediate results of the reconstruction procedure during scanning each projection. We can see clearly that how the filtered projection data group forms a 3d volumetric reconstruction, and how broken projections and bunch of error pixels affect the final result. The final volume rendering result is obtained as soon as the scanning is completed. This method can also be a good solution candidate for 4D CT visualization.
N13-217:

C. B. Wunderer1, M. Galloway2, A. Zoglauer1, M. Amman3, P. N. Luke3, J. S. Lee3, L. Mihailescu3, K. Vetter2 1 Space Sciences Laboratory, UC Berkeley, Berkeley, CA, USA 2 University of California, Berkeley, Berkeley, CA, USA 3 Lawrence Berkeley National Laboratory, Berkeley, CA, USA The High Efficiency Multimode Imager (HEMI) is an instrument currently under development for the purpose of detecting, locating, and spectroscopically characterizing gamma-ray emission sources at long-range standoff distances. 1cm3 CZT coplanar grid detectors in a compact, low-mass mounting are arranged in two planes, with the front plane only partially populated with detectors. This partially populated plane acts as (active) coded aperture, with the back plane recording the shadow pattern, enabling imaging from tens of keV. Simultaneously, the array of 2 detector planes acts as Compton detector, providing capabilities up to a few MeV. We use Monte Carlo simulations and Compton and Coded Aperture event and image analysis algorithms developed in the context of nuclear astrophysics to predict capabilities of a full-scale HEMI. We validate these predictions through detailed comparisons of modeled and measured spectra from a single detector element and a small test array.
N13-218:

Performance Predictions for the High Efficiency Multimode Imager

A. W. Hunt1,2, H. A. Seipel2,1, S. J. Thompson2,1, E. S. Cardenas2,1, E. T. E. Reedy2,1, B. H. Failor3, M. T. Kinlaw4 1 Idaho Accelerator Center, Idaho State University, Pocatello, Idaho, USA 2 Department of Physics, Idaho State University, Pocatello, Idaho, USA 3 ATG-Pulse Sciences, L-3 Communications, San Leandro, California, USA 4 National and Homeland Security, Idaho National Laboratory, Idaho Falls, Idaho, USA

The Comparison and Combination of Delayed and Prompt Neutron Signatures for the Detection of Fissionable Materials

Over the last several years, there has been a substantial research and development effort into active inspection technologies that can nondestructively detect fissionable materials. These active inspection technologies use a probing radiation source to stimulate nuclear reactions and then monitor for secondary emissions that are a signature of fissionable materials. While a single signature can be utilized for detection, the combination of multiple fissionable material signatures will increase the detection sensitivity, 103

decrease the dose required for detection or decrease the inspection time. In addition, the use of multiple signatures increases the probability of defeating potential shielding. In this presentation, delayed neutron and high-energy prompt neutron signatures from photofission are compared and combined. The fission reaction in these experiments were induced using bremsstrahlung beam produced by a variable energy pulsed radio frequency linac with endpoint energies varied between 7 and 19 MeV. The fissionable material targets consisted of aqueous solutions containing ~100 g of Th-232 or U-238 for determining the minimal detectable mass of unshielded material. In addition, ~2 kg metallic Th-232 and U-238 targets were used to investigate the effects of high-Z and hydrogenous shielding. Combining both delayed and high energy prompt neutrons, defines a phase space unique to fissionable materials.
N13-219:

Performance Evaluation of Mobile Radiation Detection System Against Nuclear Terrorism

S. W. Kwak, H.-S. Yoo, S. S. Jang, J.-S. Kim, W.-K. Yoon Dept. of Nuclear Security and Physical Protection, Korea Institue of Nuclear Non-proliferation and Control, Daejeon, South Korea Recently the threats relating to nuclear and radioactive materials have become a matter of increased international concern. Detection of illicit transport and trafficking of nuclear and radioactive materials is of vital importance for prevention of nuclear terrorism, since failure in detection might lead to catastrophic results. A mobile radiation detection system plays an important role in preventing and detecting the potential dangers posed by illicit transport and trafficking of such dangerous materials because it can monitor the suspicious vehicle at place beyond terrorists expectation which circumvents the portal monitoring system which have already been deployed at seaports, airports, and key traffic checkpoints. The mobile radiation detection system using one NaI, two plastic scintillation, and two He-3 detectors has been developed. This paper describes the developed mobile radiation detection system and experimental results for its performance assessment.
N13-220:

L. Swiderski1, M. Moszynski1, D. Wolski1, T. Szczesniak1, T. Batsch1, J. Iwanowska1, J. Szabelski1, G. Pausch2, J. Stein2, C. Plettner2, P. Schotanus3, C. Hurlbut4 1 Soltan Institute for Nuclear Studies, Otwock, Poland 2 ICx Radiation, GmbH, Solingen, Germany 3 SCIONIX Holland B.V., Bunnik, The Netherlands 4 Eljen Technology, Sweetwater, TX, USA

Comparison of Neutron Detection Efficiency Using a He-3 Counter and a Boron-10 Loaded Liquid Scintillator EJ309B5

Performance of a He-3 counter and a Boron-10 loaded liquid scintillator EJ309B5 has been studied in terms of neutron detection efficiency. The measurements were performed in a mixed field of neutron and gamma radiation from a strong (~106 n/s/4) 238 Pu-Be and a weak (~104 n/s/4) 252Cf source. The response of both detectors to background and high intensity gamma radiation (137Cs + 241Am + 60Co) has been measured to establish neutron background count rate and -rays cut-off point, respectively. The measurements covered the tests of neutrons count rate, neutron-over-background yield and false alarm rates. The performed tests of both detectors are followed by a discussion of sensitivity to neutron radiation, taking into account the usage of possible shielding materials.
N13-221:

K. Karafasoulis1, C. Lambropoulos2, D. Loukas3, C. Potiriadis1, K. Zachariadou3 1 Greek Atomic Energy Commision, Athens, Greece 2 Technological Educational Institute of Chalkida (TEI of Chalkida), Chalkida, Greece 3 Institute of Nuclear Physics, National Center for Scientific Research Demokritos, Athens, Greece

Preliminary Simulation Studies of a Multilayer Cd(Zn)Te Detector (COCAE) Used for Security Purposes

We present preliminary simulation studies regarding the development of a multilayer Cd(Zn)Te detector (COCAE). The challenges for the COCAE are to make spectroscopic measurements with efficiency equivalent to that of NaI and energy resolution close to that of HPGe devices but without using cryogenic systems, to find the direction and the distance of radioactive sources and to work at a wide range of absorbed dose rates. Extensive Monte Carlo simulation studies are performed with the Geant4 simulation toolkit and the software library MEGALIB and results concerning the energy and angular resolution of the detector and the efficiency of different event reconstruction algorithms are presented.

104

N13-222:

Detection of Concealed Special Nuclear Materials Using Nuclear Resonance Fluorescence Technique

H. Yang, D. K. Wehe Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA Smuggling of special nuclear materials has become a contemporary concern globally. The ability to detect and identify clandestine nuclear material is under heavy investigation. In this work, we have studied detection methods based on NRF technique, using both HPGe detectors and lanthanum halide scintillation detectors. We used a Van de Graff electron accelerator as the photon source. The detection method involves exposing materials to a bremsstrahlung x-ray beam and detecting the resonantly-scattered photons. Customized DAQ system based on FPGA technology was developed to be used together with fast lanthanum halide scintillator in high rate scenarios. Experimental results with various samples, including depleted uranium, will be presented. This work shows that the energy resolution of lanthanum halide scintillation detectors is sufficient to detect strong resonance from certain isotopes. HPGe detectors give the best signal-to-noise ratio because of its superior energy resolution.
We would like to thank Dr. David Bartels at the Radiation Laboratory of the University of Notre Dame for his help with the LINAC. N13-223: Metal Artifact Reduction in Dual Energy CT by Sinogram Segmentation Based on Active Contours Model and Sinogram TV Inpainting

H. Xue Department of Engineering Physics,Tsinghua University, Beijing, China Metal artifact is a major problem in computerized tomography (CT). The main approaches for metal artifact reduction can be classified into two groups: sinogram interpolation and iterative reconstruction. More complex sinogram inpainting methods based on PDEs are also presented. In dual energy CT, it is a more serious problem that metal artifacts not only degrade the image quality but also cause great inaccuracy in reconstructed effective atomic number, even when the metal objects are very small in size. Thus brings difficulty for material recognition. In this article, a metal artifact reduction method (MAR) is proposed. Firstly, dual energy decomposition is applied to compute the photoelectric coefficient projection and Compton coefficient projection from the initial dual-energy attenuation measurements. We treat the decomposition coefficients respectively. As the reconstructed Compton coefficient is always relatively accurate, metal artifact reduction is not necessary. A sinogram segmentation based on active contours model, which is self-adaptive, is implemented to detect the metal projection region (MPR) in photoelectric coefficient projection, after which total variation (TV) inpainting is adopted to fill the data gaps in MPR. Finally the decomposition coefficients are computed by conventional filtered back projection (FBP) and effective atomic number and electric density are calculated. Experiments are taken to demonstrate the effectiveness of this method. The experiment data is obtained from a scan of luggage containing a tin of water. The metal bars and screws lead to severe streak artifacts and the reconstructed effective atomic number of water deviates from the true value, with an error of 12.8%. After MAR the error is decreased to only 2.3% and most of the streaks are removed. Our method greatly improves the image quality and the accuracy of the reconstructed value.
N13-224: Concept Study of a Two-Plane Compton Camera Designed for Location and Nuclide Identification of Remote Radiation Sources

C.-M. Herbach1, A. Gueorguiev2, Y. Kong1, R. Lentering1, G. Pausch1, C. Plettner1, J. Stein1 1 ICx Techologies GmbH, Solingen, Germany 2 ICx Radiation Inc., Oak Ridge, Tennessee, USA

The Compton camera approach has been well established in photon measurements providing distinguished energy spectra and source-position information. This contribution describes the concept of a two-plane Compton camera designed to meet the requirements of Homeland Security applications. Searching for remote sources calls for large absolute detection efficiencies and, consequently, for large-scale devices. On the other hand, nuclide identification requires low detection thresholds and a proper resolution in the total energy spectrum. At last, the position resolution which is particularly biased by the granularity of the Compton camera device determines the camera performance for source location and background suppression. The results of comprehensive Monte-Carlo simulation calculations are discussed in detail, demonstrating the features of Compton cameras equipped with scintillation detectors as C9H10, CaF2, YAlO3, NaI, and LaBr3, respectively. Detection efficiencies, angular resolutions, and nuclide identification capabilities are governed by a variety of parameters. The contribution presents a method on how to optimize all these parameters as a function of the incident photon energy within the range of 100-1200 keV. The applied simulation program has been designed to model the photon physics well in accordance to the G4LECS module of Geant4. However, our implementation has been focused to rather simple geometrical setups, thus providing an exceptional program performance fast enough to proceed despite of the very large number of necessary calculations.

105

N13-225:

E. S. Cardenas1,2, E. T. E. Reedy1,2, H. A. Seipel1,2, B. H. Failor3, A. W. Hunt2,1 1 Physics, Idaho State University, Pocatello, ID, USA 2 Idaho Accelerator Center, Pocatello, ID, USA 3 ATG Pulse Sciences, L-3 Communications, San Leandro, CA, USA

Identifying Fissionable Materials Through the Detection of Delayed -Rays Using High-Efficiency Bi4Ge3O12 Detectors

An array of high-efficiency Bi4Ge3O12 detectors were utilized in the detection of delayed -rays to identify fissionable from nonfissionable materials. Photoemission spectra for 238U, 232Th, Pb and Be as well 238U and 232Th in aqueous solution were analyzed. Photonuclear reactions were induced using a pulsed bremsstrahlung beam at endpoint energies from 7 to 22 MeV. Shielding effects were also analyzed. By using energy and time cuts, Bi4Ge3O12 detectors proved to be a viable detection method.
N13-226:

An Integrated Approach for Multipurpose Fast Deployment Environmental Radiation Monitoring

E. Vax1, B. Sarusi1, M. Sheinfeld1, S. Levinson1, I. Brandys1, E. Marcus1, A. Osovizky2, Y. Kadmon1, Y. Cohen1 1 Nuclear Research Center Negev, Beer Sheva, Israel 2 Rotem Industries Ltd., Mishor Yamin, Arava, Israel An integrated approach for fast deployment of Environmental Radiation Monitoring System (ERMS), for first response teams, is presented. Following Sep. 11th there was a growing HLS need for monitoring system designed for pre and post radiological event. A network of stationary and portable radiation monitoring stations enables coping with various threats, with the need for regular environmental radiation monitoring and upgrade the data quality and quantity for risk assessment. A complete real-time system for collecting radiation and environmental measurements utilizing portable and fixed monitoring stations through DSL and cellular IP networks was developed. The rugged stations were designed to comply with demanding requirements arising from concept of operation for post event scenario, such as: long shelf life, high availability, fast deployment, friendly operation and ease of decontamination. These stations are stored ready for immediate use. After deployment each unit periodically transmits data containing: time stamp, GPS position, radiation dose and rate, temperature, local wind speed and direction and various status parameters. The data from all stations flows to a Data Base (DB) in the main server ('Control Center') and from there to remote clients via LAN or any other IP network to online view on graphs and geographic maps. The default system -detectors were selected for enabling a wide measuring range, from low background level up to high radiation fields, with dedicated stabilization algorithm. In addition to its use for HLS needs, pre-allocated stations can regularly monitor environmental radiation background in strategic locations, such as harbors and nuclear reactors surroundings. The new system provides an integrated and comprehensive solution for various scenarios and needs. The continuous data fusion of position, radiation, wind and temperature enables denser temporal and spatial sampling for radiation safety and post event risk analysis.
N13-227:

System

Analysis and Classification of Liquids Using Ultra-Low Field MR Relaxation Imaging

P. L. Volegov, C. Carr, C. J. Hanson, R. H. Kraus, A. N. Matlashov, J. C. Mosher, T. Owens, H. Sandin, L. J. Schultz, A. V. Urbaitis, V. S. Zotev, M. A. Espy Los Alamos National Laboratory, Los Alamos, NM 87544, USA A magnetic resonance imaging technique has been developed for the inspection of liquids, with potential application for security and quality control of food, chemicals, or medicines. The technique is designed to measure 2D/3D relaxation maps (longitudinal and transversal). The key feature of the technique is that the longitudinal relaxation time is measured at relatively high field (~30 millitesla) and the transversal relaxation time is measured at ultra-low field (~ 30 microtesla). It is shown that such combination provides robust parameter space for classification of liquids. The method is capable of classification of multiple samples, in random configurations. Details of the technique, pulse sequence, data analysis and sample classification algorithms are presented.

106

N13-229:

A. S. Hoover1, M. C. Galassi1, R. M. Kippen1, M. Mocko1, D. M. Palmer1, L. J. Schultz1, S. R. Tornga1, M. S. Wallace1, M. V. Hynes2, M. J. Toolin2, B. Harris2, J. E. McElroy2, D. Wakeford3, R. C. Lanza4, B. K. P. Horn4, D. K. Wehe5 1 Los Alamos National Laboratory, Los Alamos, NM, USA 2 Raytheon Integraed Defense Systems, Tewksbury, MA, USA 3 Bubble Technology Industries, Chalk River, ON, Canada 4 Massachussetts Institute of Technology, Cambridge, MA, USA 5 University of Michigan, Ann Arbor, MI, USA

Simulation and Modeling of the Tri-Modal Imager (TMI) for the Stand-off Radiation Detection System (SORDS) Using GEANT4

The Stand-Off Radiation Detection System (SORDS) is a mobile truck-based platform performing detection, imaging, and spectroscopic identification of gamma-ray sources. A Tri-Modal Imaging (TMI) approach combines active-mask coded aperture imaging, Compton imaging, and shadow imaging techniques. Monte Carlo simulation and modeling using the GEANT4 toolkit has been a critical component of the development effort. We report on the SORDS simulation effort and present comparisons to recently acquired real data. Particular focus is given to the simulation of accurate terrestrial background data based on a complete model of soil composition, radioisotope emissions, and subsequent self-absorption and scattering of photons. We also compare the spectral and imaging response of simulated and real data. This instrument is being developed for the Stand-Off Radiation Detection System (SORDS) program being conducted by the Domestic Nuclear Detection Office (DNDO) of the Department of Homeland Security (DHS).
N13-230:

M. Ghelman1, E. Vax1, A. Osovizky2, E. Marcus1, D. Ginzburg2, E. Gonen1, R. Seif1, U. Wengrowicz1, Y. Kadmon1, Y. Cohen1 1 NRCN, Beer-Sheva, israel 2 Rotem Industries Ltd, Beer-Sheva, Israel This paper describes and compares novel methods for improving detection performance of Radiation Portal Monitors (RPM). The pattern of the RPM detector reading, caused by measuring a moving radiation source versus the background provides an opportunity to employ methods based on Matched filter (MF) algorithm and the Generalized Likelihood Ratio Test (GLRT) method. The advantage of the proposed methods upon the traditional is presented as well. Conducted experiment using MF and GLRT showed approximant SNR improvement of 10 fold. To validate this concept, a detection configuration was established according to measuring time limit and Minimal Detectible Activity (MDA) defined by ANSI 42-35 regulation. A mathematical software simulation was performed, followed by a scale down configuration experiment in order to confirm the benefits of the suggested algorithms. The results of the experimental setup were correlated with a synthetic signal and analyzed by the MF algorithm and the GLRT method. An additional study was carried out examining the influence of a vehicles length L on the uncertainty of the location of radiation point source hidden in it. This analysis is required in order to evaluate the necessity of location/time estimation of the passing source, using the GLRT. The study indicates that adding location parameter decreases the estimations accuracy for small vehicles. The implementation of the MF in RPM detection algorithm improves its performance s. The method provides either: a) higher detection reliability level, b) the ability to detect lower activity level or c) the need for fewer detectors to achieve the same MDA obtained without the matched filter. The proposed method is almost costless for implementation and requires minor hardware modifications. A further research si required in order to optimize the available advantages of the MF and the GLRT in cases of spread source and source which is not located in the center of moving vehicle.
N13-231:

Improving the Detecting Performances of Radiation Portal Monitors Using Matched Filter Algorithm and Generalized Likelihood Ratio Test

Calibration and Simulation of a Coded Aperture Neutron Imaging System

J. Brennan, E. Brubaker, P. Marleau, J. Steele Sandia National Laboratories, CA, Livermore, CA, USA Active coded aperture neutron imaging detectors have the potential to be a powerful tool for the detection of special nuclear material at long range or under heavy shielding, using the signature of fast neutrons from spontaneous fission. We are building a 64-channel prototype system using 32 liquid scintillator cells, measuring 20" x 2.5" x 2.5" each, in a reconfigurable arrangement. A cross-calibration of the observed detector data with the output of Monte Carlo simulation can both improve the sensitivity of the detector to fast neutron sources and increase the simulation accuracy, allowing the study of next-generation detector designs. Here we describe the tools and procedures developed to calibrate and simulate the detector response, including energy scale and resolution, time of flight measurement, and gamma-neutron separation using pulse shape discrimination. We present results, obtained using the tuned simulation, concerning possible future development of this technology for future homeland security applications. 107

This work was supported by Laboratory Directed Research and Development (LDRD) at Sandia National Laboratories. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energys National Nuclear Security Administration under Contract DE-AC04-94AL85000 N13-232: Comparison of Passive Measurements on Well-Described Mixed-Oxide Fuel Pins for Nuclear Safeguards Applications

J. L. Dolan1, M. Flaska1, S. A. Pozzi1, D. L. Chichester2 1 Dept. of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, U.S. 2 Idaho National Laboratory, Idaho Falls, ID, U.S.

Passive measurement of special nuclear material is a promising technique and is used in areas such as nuclear nonproliferation, safeguards, materials controls and accountability, and nuclear security programs. In this work, mixed-oxide fuel pins of wellknown composition located at the Idaho Nation Laboratory will be measured with an innovative method for the accurate detection and characterization of special nuclear material developed at the University of Michigan. Monte Carlo simulation results are used to arrive at anticipated results through the use of the MCNP-PoliMi [1] code.
References: 1. S.A. Pozzi, E. Padovani, and M. Marseguerra, MCNP-PoliMi: a Monte Carlo code for correlation measurements, Nucl. Instrum. Meth. A, 513/3, 550558, 2003. N13-233: Directional Stand-off Detection of Fast Neutrons and Gammas Using Angular Scattering Distributions

P. E. Vanier1, I. Dioszegi1, C. Salwen1, L. Forman2 1 Nonproliferation and National Security, Brookhaven National Laboratory, Upton, NY, USA 2 Ion Focus Technology, Miller Place, NY, USA

Detection of radiation sources at long stand-off distances requires large area, low cost sensors that can maximize the subtended solid angle. Plastic scintillator detectors are sensitive to both gammas and fast neutrons and can be manufactured in large areas for relatively low cost. Two planes of scintillators coupled to high-bandwidth coincidence-timing electronics can be used to measure both the direction and the energy spectrum of the fast neutrons, and distinguish them from gamma rays by time of flight. We have previously demonstrated a method of data analysis in which the angle of scatter of a neutron is estimated using the kinematic equations with measured values for the energy deposited in the first plane and the time of flight between the planes. The resulting images are limited in precision by the statistical uncertainty of total amplitude of the small number of optical photons collected per event. We now discuss a different method of analysis, based entirely on timing, in which it is not necessary to make an amplitude measurement. Instead, we make use of the scattering angle probability distribution, which peaks at 45 degrees for neutrons scattering in hydrogenous material. An analogous method can be used to extract directional information from Compton scattered gamma rays, which have an angular distribution that becomes forward directed at relativistic electron energies, above 2 MeV. Even though plastic scintillators do not have sufficient energy resolution to work effectively as a Compton imager, the method allows us to determine the direction to a point source of gammas. The detector system therefore provides two independent estimates of the directions to a gamma source and a neutron source that may be, but are not necessarily, co-located. We have demonstrated these methods with measurements of a source at distances over 200 meters.
N13-234:

M. Agelou1, F. Carrel1, D. Dore2, E. Dupont2, M. Gmar1, F. Laine1, B. Poumarede1 1 CEA LIST, Saclay, France 2 CEA IRFU, Saclay, France

Detecting Special Nuclear Materials Inside Cargo Containers Using Photofission

In the last years, terrorism threats have become a major preoccupation in the developed countries. Among all the forms of these threats, the possibility for a terrorist group to import a "dirty" bomb has to be taken into account. The need of new and improved means for the detection of the smuggling of nuclear material in huge shipping containers becomes important. In the framework of the fight against these threats, CEA-LIST initiated an R&D program of three years called DEMIP (Detection de Matieres nucleaires par Interrogation Photonique) whose goal is to conceive a non-destructive detection device of nuclear matter in cargo container using high energy photons to induce reactions of photofission on actinides. The measurement of delayed neutrons and delayed gammas emitted by the fission products and coming out of the container is a clear signature of the presence of actinides. This project is divided into three parts. First, fundamental nuclear parameters of photofission are measured (cross section, time characteristics and abundance of delayed particles). The second part is the simulation of the expected sensitivity of the method and the design of the device. The third part will be the realization and the testing of this device in real conditions.

108

N13-235:

Measurements of Continuous-in-Energy Neutron Sources Using the BC-523A Capture-Gated Liquid Scintilator

M. Flaska, S. A. Pozzi NERS, University of Michigan, Ann Arbor, MI, USA Passive measurement of special nuclear material (SNM) is a promising technique and is used in areas such as nuclear nonproliferation, safeguards, materials controls and accountability, and homeland security. The identification of SNM with capture-gated organic liquid scintillators using fast digital systems is a promising method and is based on the neutron spectroscopic information that is available from the detectors. The capture-gated detectors are based on standard organic scintillators that are loaded with material with high neutron absorption cross section (typically 10B, 6Li, and/or natGd) and have been studied extensively in the past [1, 2]. The capture-gated neutron-spectroscopy principle is based on the fact that a single neutron creates two pulses that are related in time (typically several hundreds of nanoseconds apart). The primary scattering neutron pulse from the scintillator is then accepted only if a subsequent capture neutron pulse is detected. Also, the amplitude of the scattering can be used to estimate the incident neutron energy. In this work, we focus on the BC-523A detector, a boronloaded (4.41 wt% of 10B) liquid scintillator. In the paper, new experimental results obtained with the BC-523A detector are presented. The detector is used in conjunction with typical neutron sources such as 238Pu-Be, 239Pu-Be, 241Am-Be, 252Cf, and Am-Li and the measurement results are compared to Monte Carlo simulations. The main objective of this work is to accurately characterize the detectors response for a broad spectrum of neutron energies. This response will be then used to design an advanced neutron-spectroscopy system based on the BC-523A detector.
N13-236:

A Field Portable Fast Neutron Imager for SNM Detection

N. Mascarenhas, J. Brennan, C. Greenberg, P. Marleau, S. Mrowka Homeland Security, Sandia National Laboratories, California, Livermore,CA, USA We have developed a field portable fast neutron imager for the detection of special nuclear material (SNM). The detector is sensitive to fast fission neutrons from 0.3-8MeV. The effective area has been increased > 5X from our previous imager and the stand off distance is > 75m. The camera is compact, rugged and has improved background rejection. It can be transported in a 6m long trailer to enable field deployment. We present latest results obtained with the portable neutron scatter camera.
N13-237:

S. Mukhopadhyay1, C. J. Stapels1, E. B. Johnson1, E. Chapman1, P. Linsay1, T. Prettyman2, J. F. Christian1 1 Research, Radiation Monitoring Devices, Inc., Watertown, MA, USA 2 Planetary Science Institut, Tucson, AZ, USA Efficient, compact, low-cost radiation detectors, which can be deployed in large numbers, are needed for detecting illegally trafficked nuclear materials. We have designed a neutron detector based on a solid-state optical sensor, a CMOS solid-state photomultiplier (SSPM), which enables the development of compact spectrometers that can efficiently detect neutrons. SSPMbased spectrometers have high sensitivity and high specificity for neutron detection, which provides a robust alternative to the existing 3He detectors. In this work, we characterized the performance of CMOS SSPM for neutron detectors using a Cs2LiLaBr6 (CLLB) scintillator and compared the performance to a helium tube and a PMT-based instrument.
N13-238:

Improved Solid-State Neutron Detection Devices

Cryogenic Microcalorimeter Detectors for Ultra-High-Resolution Alpha-Particle Spectrometry

M. K. Bacrania Safeguards Science and Technology, Los Alamos National Laboratory, Los Alamos, NM, USA On behalf of the LANL/NIST/Star Cryoelectronics Microcalorimeter Collaboration Alpha spectrometry is well suited for analyzing low activity samples of radioactive material, such as those encountered in nuclear forensics, environmental sampling, and analytical chemistry. However, typical mixtures of isotopes in such samples exhibit spectra with energy peaks too similar to resolve with conventional silicon detectors. We have demonstrated microcalorimeter detectors with resolution as good as 1.06-keV FWHM for 5.3-MeV alpha particles, approximately a factor of eight better resolution than silicon detectors. With such high resolution, expensive and time-consuming separation of isotopes in a sample may not be required. For example, a microcalorimeter measurement of a mixture of 241-Am and 238-Pu allows for clear separation and identification of energy peaks in the 5-MeV range; such separation is unobtainable with silicon detectors. These microcalorimeter detectors use a superconducting transition-edge sensor, operating at < 100 mK, to measure the temperature change in an absorber from energy deposited by an interacting alpha particle. We have developed a four-detector system that allows for simultaneous measurement of four individual radioactive samples. Signal readout is accomplished by a commercially available superconducting quantum interference device (SQUID) amplifier system. In this talk, we will present an

109

overview of and recent measurement results from our microcalorimeter system, and discuss developments in detector pixel design, signal processing, and alpha spectrum analysis.
N13-239:

Charged Particle Energy Loss Radiography for Homeland Security Applications

K. N. Borozdin, C. Morris, A. M. Fraser, J. A. Green, F. G. Mariam, L. J. Schultz, L. Cuellar, N. W. Hengartner, A. Saunders, P. L. Walstrom Los Alamos National Laboratory, Los Alamos, NM, U.S.A. We discuss an innovative low-dose approach for detecting shielded SNM based on measuring the energy loss of energetic charged particles penetrating an object. For mono-energetic particles, energy loss depends mainly on the amount of material (or density multiplied by the material thickness) along the tracks, though it is also a function of atomic number Z. The probability density function for energy loss is described by a narrowly peaked Landau-Vavilov distribution, therefore each single particle provides a good measurement of the amount of material it penetrated. A few particles are enough to measure material parameters with high precision, providing for high-quality radiography with extremely low dose. Even a single-projection energy-loss-only radiography would be qualitatively superior to existing technologies based on attenuation of high-energy photons (e.g., X-ray radiography , CAARS). We can however further improve the performance by using multiple beam projections for 3d tomographic imaging and/or by incorporating signals from the multiple Coulomb scattering and nuclear attenuation of the beam. Detection geometry can be modified as required for a particular application. Out method is ideal for moving objects, allowing to scan them at the speed of their natural movement with mobile or stationary system. In future applications, the scanning can be made both at chokepoints or using mobile platforms, such as boats, planes or vehicles.
N13-240:

Fast Detection of 3D Planes by a Single Slice Detector Helical CT

W. Bi, Z. Chen, L. Zhang, Y. Xing Department of Engineering Physics, Tsinghua University, Beijing, China In order to accelerate the imaging process and lower the overall system cost, many practical CT systems use single slice detectors and big-pitch helical scanning, especially in the next generation Baggage Scanning systems. Geometry shape detection represents a very important function in these systems. In this paper, we present an online 3D-plane detection solution for these kinds of systems. First, we improve traditional Hough Line detection method and realized the 2.5D Hough Transform to detect the 3D plane. Second, an implementation scheme of the 2.5D Hough Transform and 3D-plane detection using CUDA is presented. We achieve performance 20 times faster than the requirement for practical applications. At Last, we test the method on a series of real baggage scanning results. Our method works effectively in practical applications.
N13-241:

M. Petasecca1, M. L. F. Lerch1, C. J. Jackson2, A. F. Gektin3, A. B. Rosenfeld1 1 Centre of Medical Radiation Physics, University of Wollongong, Wollongong, Australia 2 SensL, Cork, Ireland 3 AMCRYS-H, Kharkov, Ukraine

Characterization of an anti-Compton spectrometer based on a CsI(Tl) scintillator and silicon photomultipliers

Since 2001 considerable attention has been directed toward the development of reliable radiation portal monitor (RPM) systems to interdict illicit radioactive material at border control points. In most situations primary screening with RPMs is based simply on the measurement of the radiation count rate. The RPMs do not have the capability to distinguish naturally occurring radioactive material (NORM) or legal shipments of radioactive material, such as medical isotopes, from suspicious nuclear material (SNM). For this reason, following a primary radiation alarm, a secondary screen is required using handheld portable instrumentation with radioisotope identification capabilities to detect the cause of the initial radiation alarm. In this work we suggest an innovative approach and design of the radiation detector module based on the use of two annular CsI(Tl) concentric scintillators coupled to an array 4x4 of 3x3 mm^2 Silicon Photomultipliers (SiPM). This combination allows for the design of very compact and low power detection assemblies.
N13-242:

Addressing Different Active Neutron Interrogation Signatures from Fissionable Material

D. L. Chichester, E. H. Seabury Idaho National Laboratory, Idaho Falls, ID, USA In a continuing effort to examine portable methods for implementing active neutron interrogation for detecting shielded fissionable material, simulations and experiments have been conducted to investigate the utility of analyzing multiple timecorrelated signatures. Time correlation here refers to the existence of unique characteristics of the fission interrogation signature related to the start and end of an irradiation as well as characteristics present in between individual pulses of an irradiating source. 110

Traditional measurement approaches in this area have typically worked to detect die-away neutrons after the end of each pulse, neutrons in between pulses related to the decay of neutron emitting fission products, or neutrons or gamma rays related to the decay of neutron emitting fission products after the end of an irradiation exposure. Recognizing that the problem of detection is a problem of low count rates, we have been exploring methods to integrate these signatures with other rarely used signatures to improve detection capabilities for these measurements. In this paper we will discuss these approaches together with observations of some of the strengths and weaknesses of using these different signatures.
N13-243: Measurement of Fast Neutron/Gamma-Ray Cross-Correlation Functions with Cf-252 and Pu-Be Neutron Sources

M. Flaska1, A. Enqvist2, S. A. Pozzi1 1 Department of Nuclear Engineering & Radiological Sciences, University of Michigan, Ann Arbor, MI, USA 2 Department of Nuclear Engineering, Chalmers University of Technology, Goteborg, Sweden

Several techniques used in the field of nuclear nonproliferation rely on the measurement of neutron multiplicity distributions to assess and characterize fissile materials in various forms. These techniques are based on the thermalization of neutrons from fission in polyethylene moderators and subsequent detection using He-3 counters [1]. Signatures that rely on coincidence measurements of neutrons and gamma rays from fission have been shown to be useful in the detection and characterization of nuclear materials [2]. These correlations are of interest because they contain information that can be used to accurately identify and characterize fissile isotopes. Indeed, since for a given material and geometry the time distributions of the correlated events are very characteristic, they can be used as distinctive signatures of different material-geometry configurations. The neutrons and gamma rays from the radioactive sources have characteristic times of flight to the detectors. These arrival times depend on the type of particle (neutron or photon), and, in the case of neutrons, on their initial energy. Consequently, a characteristic crosscorrelation distribution is obtained for a given source and geometry. The goal of this work is to further explore the feasibility of using the cross-correlation functions to accurately identify radioactive materials such as special nuclear material (uranium, plutonium) and typical continuous-in-energy neutron sources (Cf-252, Pu-Be, Am-Be, Am-Li, etc.). Specifically, crosscorrelations for several source-shielding configurations will be measured and compared for Cf-252 (spontaneous-fission source) and Pu-Be (alpha-neutron) sources to assess potential differences in measured coincidences.
N13-244:

B. A. Maestas1, A. Poitrasson-Rivire1, P. R. Stanfield1, S. D. Clarke1, M. Flaska1, S. A. Pozzi1, A. Guergueiev2, G. Pausch2, C.M. Herbach2, J. Stein2 1 Department of Nuclear Engineering & Radiological Sciences, University of Michigan, Ann Arbor, MI, USA 2 ICx Radiation Inc., Oak Ridge, TN, USA Compton cameras have been used for astronomical and medical imaging purposes since the 1970s. Interest in their potential for the detection and localization of special nuclear material and radioactive sources has led to increased investigation. In this work, a large area (approximately 1 m2) two-dimensional imaging Compton camera consisting of two planar arrays of gamma-ray scintillation detectors was simulated for several common scintillation materials: NaI, LaBr3, LaCl, YAP, and CaF2. The incident gamma-ray energy ranged from 186 to 1001 keV. The camera was optimized by varying the following parameters: 1) plane thickness; 2) voxel size; 3) distance between planes; and 4) distance between pixels in a given plane. In the paper, the MCNPPoliMi simulation results is discussed in detail. In addition, optimized detection systems based on various scintillation materials is compared and ideal voxel size for each system is chosen and discussed.
N13-245:

Monte Carlo Investigation of a High-Sensitivity Two-Plane Compton Camera for Long-Range Detection of SNM

Soft Cosmic Ray Tomography for Detection of Explosives

N. W. Hengartner, L. Cuellar, K. Borozdin, A. Green, L. Schultz Los Alamos National Laboratory, Los Alamos, NM, USA Charged particle tomography has been gaining ground due to its unique probing characteristics, and the ability of cosmic ray muons to deeply penetrate matter, allowing sensitivity to atomic number, radiation length and material density via detection of the probing particles before and after target traversal. But not all particles completely traverse the target volume. This paper explores the usefulness and information content in the stopped particles, referred to as soft particles. We show, in simulations, that the stopped particles are sensitive to low Z material and therefore provides complementary information to the scattering data.

111

N13-246:

M. Hohlmann1, K. Gnanvo1, L. Grasso1, J. B. Locke1, A. Quintero-Segovia1, D. Mitra2 1 Physics and Space Sciences, Florida Institute of Technology, Melbourne, FL, USA 2 Computer Science, Florida Institute of Technology, Melbourne, FL, USA

Design and Construction of a First Prototype Muon Tomography System with GEM Detectors for the Detection of Nuclear Contraband

Current radiation portal monitors at sea ports and international borders that employ standard radiation detection techniques are not very sensitive to nuclear contraband (HEU, Pu) that is well shielded to absorb emanating radiation. Muon Tomography (MT) based on the measurement of multiple scattering of atmospheric cosmic ray muons traversing cargo or vehicles that contain highZ material is a promising passive interrogation technique for solving this problem. We report on the design and construction of a small first prototype MT station that uses compact Micro Pattern Gas Detectors. Specifically, the prototype MT station employs 10 tracking stations based on 33cm x 33cm low-mass triple-GEM detectors with 2D readout. The detectors are arranged into tracking superlayers at the top, bottom, and two sides of the probed volume. Due to the excellent spatial resolution of GEMs it is sufficient to use a gap of only a few cm between tracking stations. Together with the compact size of the GEM detectors this allows the GEM MT station to be an order of magnitude more compact than MT stations using traditional drift tubes. GEANT4 simulations demonstrate that such a compact GEM system is expected to achieve angular resolutions of a couple of mrad, i.e. similar to that of a larger drift tube system, while providing better acceptance for a given size of detector area. We present details of the production and assembly of the GEM-based tracking stations in collaboration with CERN and the RD51 collaboration as well as the design of the corresponding front-end electronics and readout system with ~15k channels. Implications of the results for the design and construction of a planned second MT prototype with large-area GEM detectors (1m x 1m) are discussed.
N13-247:

M. Pivovaroff1, S. Labov1, K. Nelson1, Y. Yao1, D. Cohen2, I. Shokair2, A. Dubrawski3, J. Ostlund3, S. Ray3 1 Lawrence Livermore National Laboratory, Livermore, CA, USA 2 Sandia National Laboratory, Livermore, CA, USA 3 Carnegie Mellon University, Pittsburgh, PA, USA Significant effort is underway throughout industry, academia and the federal laboratory system to develop systems capable of detecting the illicit importation of a nuclear weapon, special nuclear materials (SNM) or a radiological device. To increase system sensitivity, most of the focus has been devoted to improving detector performance (e.g., increasing detection efficiency or improving spectral resolution). Another approach, though, is to better utilize the existing hardware systems by folding contextual information (e.g., manifest information) into the analysis of the detector data. We are currently performing research in this area to exploit this untapped potential. Funded by the Department of Homeland Security (DHS), we are developing a contextually aware nuclear evaluation system (CANES) based on machine learning (ML) algorithms that are trained through repeated exposure to sample incidents and are subsequently able to return an assessment (classification) of a new incident. We present here the overall system design including data inputs, feature vector extraction, and system operation. We will show how the system is optimized to ingest domain expertise directly from nuclear analysts to provide both decision support assistance and complete analysis. In this presentation, we pay particular attention to the techniques used to derive features from both radiation detectors and contextual information. By tailoring the feature extraction procedures, we can assemble meaningful training tables used for tuning the ML algorithms. We also discuss our recent efforts to incorporate the CANES methodology into a larger analysis framework developed as part of the DHS Secure Freight Initiative.
This work was supported by the Domestic Nuclear Detection Office of the Department of Homeland Security and performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. N13-248:

Contextually Aware Nuclear Evaluation System

Detailed Photofission Physics Library for Monte-Carlo Radiation Transport Codes

J. M. Verbeke, D. M. Wright Global Security, Lawrence Livermore National Laboratory, Livermore, CA, USA Detection of nuclear weapons and special nuclear materials (SNM, certain types of uranium and plutonium) is crucial to thwarting nuclear proliferation and terrorism and to securing weapons and materials worldwide. Congress has funded a portfolio of detection R&D and acquisition programs, and has mandated inspection at foreign ports of all U.S.-bound cargo containers. Nuclear weapons contain SNM, which produce unique or suspect signatures that can be detected. For instance, since they are very dense, they can be seen by x-rays in the absence of heavy shielding. One class of SNM signatures detectable passively is the emission of neutrons and gamma-rays due to both natural radioactivity of actinides, and cosmic-ray induced fission. Passive detection systems rely on a variety of such signatures ranging from spectroscopic information from the decay of actinides, all the way to the detailed time structure of the fission chains in the multiplying SNM. These signatures can be enhanced by inducing fissions in the SNM using a neutron or photon beam in an active interrogation detection scheme. National labs and private corporations are presently developing detection technologies based on prompt and delayed fission 112

gamma-rays and neutrons. Current physics modeling capabilities of established Monte-Carlo radiation transport codes are unfortunately limited and often do not offer the detailed fission physiscs modeling required by both passive and active detection systems. The fission physics library developed at Lawrence Livermore National Lab aims at filling that need of the simulation community by offering detailed data-based fission physics. The data-driven library models spontaneous, neutron-induced fission as well as photofission. Since mcnpx v27b, the library can be turned on transparently via a programmatic interface.
N13-249:

Modeling and Measurements for Mitigating Interference from Skyshine

W. J. Kernan, E. K. Mace, E. R. Siciliano, K. E. Conlin, E. L. Flumerfelt, R. T. Kouzes, M. L. Woodring Pacific Northwest National Laboratory, Richland, WA, U.S.A. Skyshine, the radiation scattered in the air above a high-activity gamma-ray source, can produce interference with radiation portal monitor (RPM) systems at very large distances up to even many hundred meters. At Pacific Northwest National Laboratory (PNNL), a measurement and modeling campaign has been conducted to understand and develop strategies to mitigate the interference from skyshine in RPMs. Various approaches to shielding of skyshine radiation include walls, plates, collimators, louvers and even buildings, or combinations thereof, have been considered. This paper will concentrate on two topics: measurements and modeling with Monte Carlo transport calculations to characterize skyshine from an iridium-192 source and testing of a prototype louver system, designed and fabricated at PNNL, as a shielding approach to limit the impact of skyshine interference in RPM systems.
Prepared for the U. S. Department of Homeland Security U.S. Customs and Border Protection and Domestic Nuclear Detection Office under U.S. Department of Energy Contract DE-AC05-76RL01830 J03: Joint NSS/MIC 3 Tuesday, Oct. 27 J03-1: 13:30-15:30 International Ballroom North

H. T. van Dam1, S. Seifert1, R. Vinke2, H. L&oumlhner2, P. Dendooven2, F. J. Beekman1, D. R. Schaart1 1 Dept. TNW, Delft University of Technology, Delft, The Netherlands 2 KVI - University of Groningen, Groningen, The Netherlands Silicon Photomultiplier (SiPM) photosensors, also referred to as Multi Pixel Photon Counters (MPPC), are of high interest in many scientific environments, such as high energy physics, astronomy, bio-molecular imaging and medical imaging, as they are fast, have high gain and low noise, and are compact and potentially cost-effective. However, effects such as after-pulsing, crosstalk and saturation may have an important impact on the detector performance, especially when working with pulsed lasers or fast scintillators as in TOF PET detectors. Since the saturation effect depends on the number of incident photons per unit time, the detector shows a non-linear response with respect to photon intensity. In detector design it is desirable to understand and quantify this response in order to choose the proper detector components, to perform non-linear energy spectrum corrections, and to predict performance. A model of the SiPM response might furthermore be used as a tool for determining difficult-to-measure SiPM parameters. For these purposes a model of the expectation value of the SiPM response is presented. This model accounts for the total number and the temporal distribution of the absorbed photons, cross-talk, after-pulsing and recharging of the microcells. The model is demonstrated for various scintillators and SiPM parameter values. The response of a Hamamatsu MPPC S10362-33-050C is shown for LaBr3:Ce, LSO:Ce, BGO, NaI(Tl) and CsI(Tl) in response to 140 and 511 keV gamma quanta. It is shown that, depending on the scintillator and the fraction of absorbed photons, the response can decrease by up to a factor of 2 due to saturation. Also the influence of microcell recharge time, cross-talk, the probability of after-pulsing and the average release time of trapped charge carriers is shown.
J03-2:

Silicon Photomultiplier Response Model

Simulation Study of Muon Scattering for Tomography Reconstruction

D. Mitra, S. White, R. Hoch, M. Hohlmann, K. Gnanvo CS&PSS, Florida Institute of Technology, Melbourne, FL, USA Muon Tomography (MT), based on the measurement of multiple scattering of atmospheric cosmic ray muons traversing cargo or vehicles, is a promising technique for Homeland security application of nuclear contraband detection. Many groups around the world are studying and developing MT systems. For the purpose of reconstruction the standard ART (Algebraic Reconstruction Technique, where the probed volume is dvoxelilzed) based algorithms are adapted from medical imaging. A special feature of the underlying scattering event in MT is the existence of a point of closest approach (poca) between the incoming and outgoing vectors of a muon track, which represents the scattering region quite well. A geometrical algorithm (POCA-algorithm) that detects poca points and their corresponding angle of scattering is surprisingly efficient in identifying most of the metallic targets 113

in a probe volume of a MT station. Expectation Maximization algorithms based on maximum likelihood estimation are very useful in estimating scattering parameter of each voxel, but have slow convergence and typically computes the parameters for all voxels even when many of them are not of interest. In this work we are studying the muon scattering properties for possible target materials by using poca points generated by the tracks. We have used GEANT4 simulation software, enhanced with the CRY package as the source for cosmic ray generated muons, for this purpose. Empirical studies of muon scattering from a target, with a MT-station like geometry, is essential in developing tomography systems and reconstruction algorithms. One of our objectives is to develop a non-ART fast algorithm for the reconstruction purpose for MT systems. Our results presented here will also be useful in making EM algorithms more efficient for MT. We expect our work will make a broader impact on some transmission tomography applications including those used in medical areas.
Acknowledgment: US Department of Homeland Security, and US National Science Foundation J03-3:

A. Bousselham1, H. H. Barrett1, K. Shah2 1 Center for Gamma-Ray Imaging and Dept. of Radiology, University of Arizona, Tucson, AZ, USA 2 Radiation Monitoring Devices, Inc., Watertown, MA, USA

Photoelectron Anticorrelations and Sub-Poisson Statistics in Scintillation Detectors

In this work we propose a new experimental approach to studying statistics of the scintillation light from scintillation detectors. Two photodetectors are used for detecting the light output from the two faces of a thin scintillator crystal. From the correlation of the signals from the two photodetectors, the Fano factor Fopt of the scintillation light and the energy resolution of the scintillator detector are determined [1]. Data on several inorganic crystalline scintillators (CsI:Na, LaBr3, Cs2LiLaBr6:Ce, SrI2 and CdWO4) are presented, and in some crystals it is observed that the correlations are negative and Fopt < 1 (sub-Poisson statistics).
Keywords: Fano factor, Photoelectron anticorrelation, sub-Poisson statistics. J03-4:

J. S. Maltz1, J. Hartmann1, A. Dubouloz1, A. Paidi1, B. Gangadharan1, G. Hoerauf2, A. R. Bani-Hashemi1 1 Oncology Care Systems Group, Siemens Medical Solutions, USA, Inc., Concord, CA, USA 2 Siemens AG Healthcare, Forchheim, Germany

Thick Monolithic Pixelated Scintillator Array for Megavoltage Imaging in Radiation Therapy

We describe the fabrication and evaluation of a thick pixelated scintillator for megavoltage (MV) imaging composed of a ceramic containing over 99.9% gadolium oxysulfide. This sintered material offers a 59% increase in density over the Lanex Fast B (LFB) phosphor screens most commonly employed in MV imaging. The sintered pixelated array (SPA) is fabricated from a single slab of ceramic. This obviates the need to assemble over a million separate crystals in order to cover a 40.96 cm x 40.96 cm detector area. As a consequence, the design is amenable to fabrication using methods of mass production. Method and Materials: A 1.8 mm-thick 274 x 256 pixel SPA with 0.4 mm pixel pitch is attached to the light-sensitive surface of an amorphous silicon flat panel detector (Perkin Elmer XRD1640AN). Image quality is characterized using 1 MU exposures of the 6MV beam of a Siemens Primus Linac. A QC-3V phantom (Standard Imaging) is employed to calculate the modulation transfer function (MTF) and contrast-to-noise ratio (CNR). The detective quantum efficiency (DQE) is computed. A LFB screen is then evaluated under identical conditions for comparative purposes. Cone beam CT (CBCT) imaging is performed with four arrays tiled side-by-side on the detector surface. Results: The half-maximum value of MTF occurs at 0.32 and 0.34 lp/mm for the SPA and LFB, respectively. The DQE(0.1 lp/mm) of SPA is 5.8%, versus 1.0% for LFB. The SPA offers a 235% improvement in CNR over LFB. Previously undetectable low-contrast phantom inserts are clearly visible in SPA MV-CBCT images. Conclusions: The SPA appears to offer a practical and cost-effective means of attaining major improvements in MV image quality. The measured MTF and DQE values underestimate the achievable performance, since the SPA and detector photodiode arrays were imperfectly aligned during these evaluations. A DQE(0) value closer to the maximum attainable 7.8% is expected.
J03-5:

M. L. Purschke1, B. Ravindranath2, S. S. Southekal2 1 Physics Department, Brookhaven National Lab, Upton, NY, USA 2 Department of Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA

Massively Parallel Image Reconstruction for the BNL Breast Scanner PET Tomograph Using CUDA

A new trend is evolving in high-performance computing which makes use of commodity graphics processors (GPUs) for massively parallel computing tasks. The development of GPUs, driven by high-end computer gaming, can easily be used for CPU-intensive tasks. Off-the-shelf systems providing multiple TeraFlops of processing power are available at commodity price levels today. The BNL Breast Scanner is a development derived from the RatCAP tomograph and re-uses elements of the detector design, the readout technology, and the image reconstruction method. While the images from the much smaller RatCAP can be reconstructed 114

in a few minutes, the data from the breast scanner in its full configuration will present a significant computing challenge. We have investigated the "Compute Unified Device Architecture" (CUDA) framework, which makes use of NVIDIA GPUs in either a CUDA-enabled graphics card, or in a dedicated GPU board. Even low-cost graphics cards provide a number of individual "Multiprocessors", which in turn provide a high number of processing cores, such as 32, 64, or 128 each. The CUDA software libraries and development kits are available for free. In initial tests with a parallel processing approach using CUDA on a commodity GPU card, we were able to reduce the processing time for images of a breast scanner prototype from about 16 hours to 8 hours. We will give a brief overview of the CUDA framework, and present the current benchmarks for the image reconstruction with the CUDA system, and compare them to the same process on conventional CPUs.
J03-6:

Y.-C. Shih1, F. W. Sun1, L. R. MacDonald2, B. P. Otis1, R. S. Miyaoka2, W. McDougald2, T. K. Lewellen2 1 Dept. of Electrical Engineering, Univerisity of Washington, Seattle, WA, USA 2 Dept. of Radiology, Univerisity of Washington, Seattle, WA, USA

An 8x8 Row-Column Summing Readout Electronics for Preclinical Positron Emission Tomography Scanners

This work presents row/column summing readout electronics for an 8x8 silicon photomultiplier array. The summation circuit greatly reduces the number of electronic channels, which is desirable for pursuing higher resolution positron emission tomography scanners. By using a degenerated common source topology in the summation circuit, more fan-in is possible and therefore a greater reduction in the number of electronic channels can be achieved. In addition, the cross-talk between neighboring pixels is greatly reduced. Variable gain amplifiers are used as preamplifiers for the spatial channels to provide gain calibration. The timing signal is retrieved from a common anode, which allows the use of a single fast-sampling analog to digital converter (ADC) for the timing channel and slower, lower power ADCs for the 64 spatial channels. Preliminary results of one column summation of the 8x8 readout electronics exhibited an energy resolution of 13.75% and 12.14% with and without multiplexing, respectively. The measured timing resolution is 2.54ns and 1.7ns with and without multiplexing, respectively. This circuit design is being investigated to support the depth of interaction micro crystal element (dMiCE) detector being developed at the University of Washington. Successful implementation will lead to the development of an efficient application specific integrated circuit (ASIC) based upon this architecture.
This work was supported in part by the U.S. NIH under Grants EB0002117 and EB001563, DOE DE-FG02-05ER15709 and by Zecotek Photonics Inc. J03-7:

Effect of Pixel Dimensions and Thickness on Energy Resolution and Sensitivity of CZT Detectors Used in Nuclear Medicine Applications

M. E. Myronakis, D. G. Darambara Joint Department of Physics, Institute of Cancer Research & Royal Marsden NHS Foundation Trust, London, UK Cadmium zinc telluride (CZT) is considered a promising material in detector manufacturing for gamma-ray imaging. The potential of using CZT detectors with pixellated anodes to fulfill the requirements of better spatial and energy resolution as well as higher sensitivity has not been yet fully investigated. Pixellated anodes are advantaged by the small pixel effect but are susceptible to charge sharing and photon scattering within the detector especially in smaller pixel dimensions. In the current work, we combine the finite element method (FEM) and Monte Carlo simulations to model charge transport and photon interactions in the detector. The effect of pixel size on energy resolution and sensitivity was investigated for pixel pitch size from 0.4 up to 1.6 mm and for 5 and 10 mm thickness under 140 and 511 keV photon irradiation.
N14: Instrumentation for Homeland Security II Tuesday, Oct. 27 N14-1: 13:30-15:30 Grand Ballroom 2

First Light for the GammaTracker Handheld Radioisotope Identifier

C. E. Seifert, M. J. Myjak, D. S. Barnett, M. T. Batdorf, B. J. Burghard, W. K. Hensley, L. J. Kirihara, S. J. Morris Pacific Northwest National Laboratory, Richland, WA, USA GammaTracker is a high-resolution, high-efficiency handheld radioisotope identifier for search, survey, and characterization of radioactive materials. The system incorporates eighteen pixelated 15x15x10-mm3 CdZnTe detectors into a single unit, together with associated high voltage, ASIC readout, single-board computer, and graphical display. The unit can operate on two lithiumion batteries for nearly 6 hours. The embedded firmware performs automated energy calibration, isotope identification, and

115

simple source localization using Compton imaging techniques. This paper presents the first measured spectra from the handheld instrument and comments on the accuracy of the isotope identification and directionality algorithms.
N14-2:

M. Amman1, P. N. Luke1, J. S. Lee1, L. Mihailescu1, C. B. Wunderer2, A. Zoglauer2, K. Vetter1,3, M. Galloway3, H. Chen4, P. Marthandam4, S. Awadalla4, S. Taherion4, G. Bindley4 1 Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 University of California, Space Sciences Laboratory, Berkeley, CA, USA 3 University of California, Berkeley, CA, USA 4 Redlen Technologies, Sidney, British Columbia, Canada The High Efficiency Multimode Imager (HEMI) is an instrument currently under development for the purpose of detecting, locating, and spectroscopically characterizing gamma-ray emission sources at long-range standoff distances. The instrument design consists of multiple planes of relatively large-volume, good-energy-resolution detector modules configured for combined coded aperture and Compton scatter (multimode) gamma-ray imaging. The basic building block of HEMI is a detector module consisting of a CdZnTe coplanar-grid detector packaged along with its front-end and pulse-shaping electronics. We have developed a module that can be used to form close-packed arrays, has low dead mass, and is easily adjusted for optimum spectroscopic performance. In this paper, we provide an overview of HEMI and then detail the design, production, and testing of the detector modules.
N14-3:

Detector Module Development for the High Efficiency Multimode Imager

ULIS: An Unattended Luggage Inspection System

C. J. Groiselle, J.-L. Dumont, J.-S. Lacroix, M.-J. Lopez-Jimenez, F. Moutrousteguy, F. Thebault, I. Lefesvre, E. Poirrier, P. Paul, M. Mangeard, C. Dardennes, B. Vernet, K. Soudani, P. Le Tourneur DPAS/DASN, EADS - SODERN, Limeil Brevannes, France The need for detection of illicit materials has been increasing since 9/11/2001 and developments associated to Homeland Security are evolving within the evolution of requirements and threats. We are developing a fully integrated and self-sufficient device for radionuclide, explosive, and chemical detection fitting in a 30kg suitcase including a neutron generator, its VHV supply, detectors, a data acquisition system and all the electronics. ULIS, as it was named for Unattended Luggage Inspection System, is then placed close to the object to be inspected. The acquired information is sent to a remote laptop that performs the required analysis allowing locating and identifying an eventual threat within the parcel. This system was mainly designed to equip bomb squads but could also be part of Coast Guards or military table of equipments (TOE).
N14-4:

Variant Designs and Characteristics of Improved Microstructured Solid-State Neutron Detectors

S. L. Bellinger, W. J. McNeil, D. S. McGregor Mechanical and Nuclear Engineering Department, Kansas State University, Manhattan, KS, USA Microstructured semiconductor neutron detectors have superior efficiency performance over thin-film coated planar semiconductor detectors. The microstructured detectors have patterns deeply etched into the semiconductor substrates subsequently backfilled with neutron reactive materials. The detectors operate as pn junction diodes. Two variations of the diodes have been fabricated, which either have a rectifying pn junction selectively formed around the etched microstructures or have pn junctions conformally diffused inside the microstructures. The devices with the pn junctions formed within the perforations have lower leakage currents and better signal formation than the devices with selective pn junctions around the etched microstructure patterns. Further, pulse height spectra from conformally diffused detectors have the main features predicted by theoretical models, whereas pulse height spectra from the selectively diffused detectors generally do not show these prominent features. The improved performance of the conformal device is attributed to stronger and more uniform electric fields in the detector active region. Also, the system noise, which is directly related to leakage current, has been dramatically reduced as a result of the conformal diffusion fabrication technique. Sinusoidal patterned devices with 100 m, 150 m, and 200 m deep perforations, backfilled with 6LiF, have been fabricated and intrinsic detection efficiency for 0.0253 eV neutrons have been found for each, calibrated against thin-film planar semiconductor devices and a 3He proportional counter.
N14-5:

Event Reconstruction for Pixelated CdZnTe Detectors

R. McLean, F. Harrison, W. R. Cook, H. Miyasaka, S. Kaye California Institute of Technology, Pasadena,CA, USA We have built and tested a series of hybrid detectors that use 5-7 mm thick CdZnTe material for good high-energy sensitivity. The ASIC is arrayed into 32x32 pixels which are directly coupled to the detector electrode at a pitch of 604 microns. In this paper

116

we review some of the challenges to achieving high-resolution and review various reconstruction techniques suitable for the ASIC architecture.
N14-6:

M. Mayorov1, R. Arlt1, J. Blackadar2, D. Blumenthal2, K. Frame3, E. Mark4, M. Milovidov5, R. York6 1 Department of Safeguards, International Atomic Energy Agency, Vienna, Austria 2 Domestic Nuclear Detection Office, US Department of Homeland Security, Washington, USA 3 Los Alamos National Laboratory, Los Alamos, USA 4 National Nuclear Security Administration, US Department of Energy, Washington, USA 5 Nuclear Physics Researches Center, St. Petersburg, Russia 6 Nuclear Science and Technology, Oak Ridge National Laboratory, Oak Ridge, USA

Semi-Empirical Approach for Performance Evaluation of Radionuclide Identifiers

Under the Coordinated Research Project "Development and implementation of instruments and methods for detection of unauthorized acts involving nuclear and other radioactive material" the IAEA, in cooperation with U.S. National laboratories, established a testing campaign to assess the performance of radiation detection equipment for nuclear security applications. The test targets instruments featuring radionuclide identification: spectrometric portal monitors, radionuclide identifiers, spectrometric portable radiation scanners, etc. The test takes a semi-empirical approach, utilizing instruments replay and reevaluation capabilities, also known as injection studies. The test framework envisions a four step approach to assess reliability of the instruments identification algorithm and its hardware/software tolerance to statistical fluctuation, masking and shielding scenarios, and operational extremes. The steps are: (1) Hardware assessment. The value of distortion of the energy scale and response functions due to external influencing factors (count rate, temperature, EMI) will be assessed and quantified experimentally; (2) Spectrum generation. Series of spectra samples will be computed by downgrading statistics, distorting the spectra, and combining various radionuclide signatures. Well defined experimental spectra will be used for computing the sample one for the injection; (3) Replication. The identification algorithm will be assessed through replication of computed spectra, simulating those, taken under the field conditions; (4) Validation. A limited number of the experimental tests, aiming identification of radionuclide(s) under specified conditions, will be carried out to ensure conformity with the results of PC-based evaluation. The paper describes details of the methodology, approach, requirements for replication software, and interpretation of the identification results.
N14-7:

Tracking of Weak Radioactive Sources in Crowded Venues

R. B. Vilim nuclear engineering, argonne national laboratory, argonne il, usa Background Monitoring for the clandestine transport of nuclear and radiological materials is one element among a collection of administrative and surveillance tools being developed as terrorism countermeasures for homeland security. In one scenario, the terrorist attempts to disperse radioactive materials at a large public gathering, or special event, such as the inauguration. The capability to detect and track radioactive materials as they approach or move through such a venue would provide useful information for lawenforcement personnel. One important practical consideration is the cost of such a system as measured in terms of hardware and the number and technical sophistication of support personnel needed to deploy and operate it. With the availability of inexpensive off-the-shelf hardware and software for building integrated networked sensor systems, this cost is no longer prohibitive. Systems costing less than a hundred thousand dollars and deployed and run by a single trained individual are possible. Key Technical Challenges The main focus of this paper is on the technical solutions developed by the authors in designing and building such a system. The paper will describe how we are able to determine position in near real time. Second we show how this can be done in the presence of large and time-varying numbers of people. Crowds act to not only reduce signal strength through shielding, effectively lowering the signal-to-noise ratio (SNR), but also to change the relative counts rates among multiple detectors thereby confounding estimation of position. Third, since cost dictates large detector spacing which drives down SNR, methods are developed to achieve maximally informative position estimates in the face of weak signals.

117

N15: Nuclear Physics Instrumentation I Tuesday, Oct. 27 N15-1: 13:30-15:30 Grand Ballroom 3

C. Woody1, B. Azmoun1, C.-Y. Chi2, Z. Citron3, M. Durham3, T. Hemmick3, A. Iordanova4, J. Kamin3, A. Milov5, M. Naglis5, R. Pisani1, M. Proissl3, I. Ravinovich5, S. Rolnick4, T. Sakaguchi1, D. Sharma5, S. Stoll1, I. Tserruya5 1 Physics Department, Brookhaven National Lab, Upton, NY, USA 2 Nevis Laboratory, Columbia University, Irvington, NY, USA 3 Physics Department, Stony Brook University, Stony Brook, NY, USA 4 Physics Department, University of California Riverside, Riverside, NY, USA 5 Department of Particle Physics, Weizmann Instistut of Science, Rehovot, Israel

Initial Performance of the PHENIX Hadron Blind Detector at RHIC

The PHENIX Hadron Blind Detector (HBD) was successfully operated during the high energy polarized proton run at RHIC during the spring of this year. The detector was operated for several months under actual beam conditions, which provided an opportunity to study and optimize many of its performance parameters. Results will be presented on the operation of the GEM detectors and their CsI photocathodes, determination and calibration of the gain using scintillation light produced by charged particles in CF4, the ability to identify single and double electrons using the signal from Cherenkov light, and the level of sensitivity of the detector to charged hadrons. A description will also be given of the construction of the detector and its installation into PHENIX, including the methods used for producing the CsI photocathode GEMs, the CF4 gas system and its monitoring components, and the performance of the electronics and readout system. Plans for operating the detector in the heavy ion run at RHIC that is planned for next year will also be discussed.
N15-2:

A GEM-Based TPC Prototype for PANDA

M. Vandenbroucke Physik Departement, E18, Technische Universistt Mnchen, Garching, Germany On behalf of the GEM-TPC collaboration A GEM-Based TPC is a very promising option for the central tracker of the PANDA experiment. PANDA is a spectrometer which will detect charged and neutral particles emitted over the full solid angle from 2x107 proton-antiproton annihilations per second to perform high-precision studies in the field of non perturbative QCD. The central tracker of the spectrometer has to provide good momentum resolutions and at the same time have a very small material budget in order to minimize secondary interactions or photon conversion. A TPC ideally fulfills these requirements, and, in addition, provides particle identification by measuring the specific energy loss. The GEM technology, used for gas amplification, provides an intrinsic suppression of ion backflow usually done by an ion gate for wire chamber, inappropriate here because of the continuous nature of the beam. A small GEM-TPC prototype was built and characterized with cosmic muons. An average resolution of 200 m had been achieved using the PASA/ALTRO readout electronics. In order to perform more detailed investigations of the performance of the detector in various conditions, a tracking telescope was set up at the electron beam at ELSA, Bonn, Germany. The characterization of the external tracking telescope has begun in December 2008 and the first electron tracks was observed in the TPC.
N15-3:

L. Benussi1, A. Braghieri2, G. Boca2,3, S. Costanza2,3, P. Genova2,3, P. Gianotti1, L. Lavezzi2,3, V. Lucherini1, P. Montagna2,3, D. Orecchini1, D. Pierluigi1, J. Ritman4, M. Roeder4, A. Rotondi2,3, A. Russo1, P. Wintz4 1 INFN Laboratori Nazionali di Frascati, Frascati, Italy 2 INFN Sezione di Pavia, Pavia, Italy 3 Dipartimento di Fisica Nucleare e Teorica, University of Pavia, Pavia, Italy 4 IKP Forschungszentrum Jlich, Jlich, Germany PANDA [1] is a new experiment that will be installed at HESR, the new antiproton storage ring under construction as a part of the new FAIR facility at Darmstadt, Germany. This experiment will investigate QCD in the charmonium mass regime, and other aspects of particle and nuclear physics. It will be a fixed target detector with a central spectrometer, surrounding the interaction point, and a forward one for detecting particles emitted in the forward region. The central tracker is one of the essential part of the PANDA detector, providing information about primary and secondary decays vertices, momenta and types of charged particles emitted after antiproton-proton annihilation. In this talk, the design of the Straw Tube Tracker (STT) together with experimental results of the R phase will be presented.
[1] http://www-panda.gsi.de/auto/_home.htm

The Straw Tube Tracker of the PANDA Experiment

118

N15-4:

Design and Optimization of the CBM Time of Flight Wall

D. Gonzalez-Diaz CBM-HADES, GSI, Darmstadt, Hessen, Germany On behalf of the CBM collaboration Within the new FAIR-SIS100 complex based at GSI-Darmstadt, the CBM (Compressed Baryonic Matter) experiment will explore strange, charm and di-lepton production in heavy ion collisions up to AGS energies. The quest for unprecendented statistics in both bulk and rare observables needed for exploring critical parameters of the phase-diagram of nuclear matter requires of a spectrometer with excellent tracking and trigger abilites, together with particle identification over a large momentum-range. The CBM-TOF wall has been designed for the later purpose: the main part of its 150 square meters will be a carpet of the multi-strip tRPC technology pioneered at GSI at an anticipated 80 ps resolution. Timing RPC signals, with risetimes as short as 250 ps constitute a challenge for any stripped design, resembling in many aspects the PCB-routing in digital systems, but with a design that must combine harmoniously a dense packing at an acceptable cross-talk. An extense survey of the simulation and modeling of the RPC response (avalanche formation, induction, transmission, termination and amplification) and the different experimental means to address them will be summarized in the talk and presented in a systematic way. Few words will be said on the research of low resistivity materials, a topic of central importance for the inner CBM-TOF region due to the high particle fluxes present there. Being covered with ordinary pads, their electrical design is nevertheless of much reduced complexity as compared with the outer stripped area.
N15-5:

Performance of the Fine Grained Detector Built for T2K Neutrino Experiment

J. Zalipska Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada On behalf of the T2K/FGD Collaboration The Fine Grained Detector (FGD) is a finely segmented scintillation detector which combines, for the first time on a large scale, well established scintillation tracker techniques with a novel photo-sensor technology: Multi-Pixel Photon Counters (MPPC). The FGD is composed of bars, 0.96 cm square and ~1.8 m long, which are glued together to create a ~2 m x ~2 m surface of X and Y oriented tracking planes. Scintillation light is transported through wavelength shifting fibers to the MPPCs, where signal readout is by waveform digitalization. The FGD is part of the tracking system of the near detector (ND280) for the Tokai-to-Kamioka (T2K) long-baseline neutrino oscillation experiment. It acts as a target for neutrino interactions, detecting short-ranged charged particle tracks, and matching to longer tracks which are measured in other parts of the ND280 detector. It has a good efficiency for detecting Michel electrons from stopped muon or pion decays. This presentation is aimed at summarizing the basic performance of the FGD, based on beam test measurements which have been performed at the TRIUMF laboratory in Canada, prior to shipment to its final destination at the J-PRARC accelerator complex in Japan.
N15-6:

The Photon Veto System for the NA62 Rare Kaon Decay Experiment

P. Valente INFN Roma, Roma, Italy On behalf of the NA62 Collaboration The NA62 experiment at CERN SPS, aimed to measure ~80 events of very rare decay K+->+ anti-(BR~8.510-10), poses stringent requirements on PID capabilities to reject the overwhelming +0 (63%) and K2(21%) backgrounds. A 0 detection inefficiency of few units of 10-8 is needed to complement kinematical rejection. As a main veto detector, the NA48 liquid Kr calorimeter will be used. To have full acceptance, a set of veto counters should be placed along the vacuum decay tank, to catch large-angle 's (<50 mrad) with an inefficiency <10-4 in a wide energy range (50 MeV-35 GeV). Good energy resolution (~10% at 1 GeV) for threshold definition, good time resolution (~1 ns) to be used at the trigger level, sensitivity to mip, for calibration with muons of the beam halo are needed. A moderate segmentation in the azimuthal angle is desirable, for reducing the counting rate and providing information on the direction. We performed an intense R&D on 3 solutions: "spaghetti" calorimeter, lead/scintillator sandwich calorimeter, and original re-use of the existing barrel of the OPAL lead-glass e.m. calorimeter. Studies have been performed at the Frascati BTF beam and all 3 meet the efficiency requirements. The final choice, uses a peculiar radial arrangement of lead glasses in rings:incoming 's hit the crystal laterally on the short side. Multiple staggered layers of rings should be arranged in a station:with 5 layers (of 32 to 48 crystals depending on radius), at least 3 are fully crossed by particles at <50 mrad. A small 20-blocks prototype, with final geometry and assembly solutions (supports, cabling, feedthroughs), has been tested in the NA48/NA62 vacuum tube in Oct. 2008 with and K beams.Studies on the front-end electronics are under way, for designing a readout covering the 3 orders of magnitude of the signal, contributing to the trigger, and integrated in the general TDAQ, while keeping low cost and simplicity. The first full veto station is now in costruction.

119

N15-7:

Design and Performance of the Compact W-Si Forward Calorimeters for the PHENIX Experiment at

RHIC
E. P. Kistenev PH, Brookhaven National Laboratory, UPton, NY 11973, USA On behalf of the PHENIX Forward Upgrade Collaboration The Relativistic Heavy Ion Collider (BNL) collides heavy nuclei to create a medium at unprecedented density and temperature. Jets of hadrons from quarks and gluons experiencing initial state hard scattering provide a probe of transmission through the QGP that exists early in the collision. Characteristic event structures associated with probes may be efficiently selected with calorimeters, which can provide triggers on single high-pT particles, "jets" of particles, or large transverse energy, along with precision measurements of the structures. The importance of calorimeters in studies of ultrarelativistic heavy-ion collisions was first recognized by W.Willis The key requirements are good electromagnetic resolution for electron and photon identification and measurements, and excellent granularity to handle extreme occupancies common to this kind of interactions. Electromagnetic calorimetry is also the most important single tool for photon/ 0 measurements and as such drives the detector dimensions and cost. In this talk we present a new fully developed and beam tested concept of the W-Si sampling calorimeter able to handle extreme multiplicities typical of heavy-ion collisions in proximity to collision vertex. The calorimeter is designed following industrial approaches to detector construction and ensuring the reliability. Novel features of this design are concepts of silicon micromodules, use of microconnectors for alignment purposes and passive signal summation to form readout towers. A prototype calorimeter was built in collaboration between BNL and a number of University groups from USA, Russia, Korea, Finland and Czech Republic. Results of the performance measurements made exposing this prototype to particle beams in H2 beam line at CERN will be presented.
N16: Gaseous Detectors II: Varied Applications in Astrophysics and Particle Physics Tuesday, Oct. 27 N16-1: 13:30-15:30 Grand Ballroom 6

Systematic Study of the Calibration and Resolution of Drift Tubes for Muon Tracking in the ATLAS Experiment at the LHC

S. Di Luise Physics, Roma University INFN, Rome, Italy On behalf of the ATLAS Muon Collaboration The Muon Spectrometer of the ATLAS experiment consists of three large air core toroidal magnets, one barrel and two end-caps. The bending power is about 3 and 5 Tm respectively. Tracking in the range of pseudorapidity [-2:2] is done with regular arrays of 3 cm diameter high pressure drift called MDT, arranged in three measurement stations. The aim of the experiment is a relative resolution of 10% at transverse momentum of 1 TeV/c. At this value, the resolution is limited by the detectors alignment and the single point resolution. The second should not exceed 80 microns per tube, averaged over the drift distance. The gas composition ( 93%Ar-7% CO2) and gain (20000), chosen in order to limit the aging, makes the drift response function, r(t), highly non linear, thus it depends on temperature, pressure and local value of the magnetic field. An auto-calibration method has been developed to calculate from the data themselves the r(t) function and to extract the resolution function. This method has been extensively tested on a beam of high momentum muons to study systematic effects and the statistics needed to guarantee that the calibration error is indeed much smaller than the single point resolution. We describe the analysis of these data, and present the results on the convergence of the auto-calibration as a function of the track angular range, the number of tracks, the track fit quality requirements, the correlation between angle and position measurement, for both three and four tube layers.
N16-2:

DMTPC: a TPC with Optical Readout for Directional Detection of Dark Matter

G. Sciolla Physics Department, MIT, Cambridge (MA), USA On behalf of the DMTPC collaboration We developed a TPC with optical readout with the goal of detecting the sense and direction of the elastic recoils generated by Dark Matter interactions. The detector, filled with CF4 gas at low pressure, is equipped with a mesh-based amplification region that allows for a 2D imaging of the recoils in a CCD camera. The third coordinate of the recoil is provided by PMTs. The sense of the direction is determined by measuring the energy loss along the recoil track. The performance of this detector has been measured using alpha particles, low-energy neutrons, and x-rays. In particular, head-tail discrimination has been demonstrated for nuclear recoils down to ~100 keV. A 10-liter DMTPC detector is scheduled to start underground operation in summer 2009. A 1m3 detector will then be built. Such a detector will be able to increase the present sensitivity to DM spin-dependent interaction by almost two orders of magnitude. 120

N16-3:

The Luminosity Monitor of the ATLAS Experiment

M. Villa Dept. Physics, Universita' di Bologna, Bologna, Italy On behalf of the ATLAS Luminosity and Forward Physics working group The LUCID (LUminosity Cherenkov Integrating Detector) apparatus is composed by two symmetric arms deployed at about 17 m from the ATLAS interaction point. The purpose of this detector, installed in summer 2008, is to monitor the luminosity delivered by the LHC machine to the ATLAS experiment. Each arm of the LUCID detector is based on an aluminum vessel containing 20 Cherenkov tubes, 15 mm diameter and 1500 mm length, filled with C4F10 radiator gas at about 1.1 bar. The Cherenkov light generated by charged particles above the threshold is collected by photomultiplier tubes (PMT) directly placed at the tubes end. The intrinsically fast response of the detector and its readout electronics makes it ideal to follow the number of interactions per LHC bunch crossing therefore providing also an interaction trigger to the ATLAS experiment. The electronic boards performing the on-the-fly evaluation of the LHC luminosity with different algorithms will be discussed. The detector took already some data from the first beam interactions produced by LHC. In this talk, we will report the relevant details of the detector design and its associated monitor and electronics.
N16-4:

NEXT: Neutrinoless Double Beta Decay Experiment with a Gaseous Xenon TPC

M. Ball Instituto di Fisica Corpuscolar (IFIC), Valencia, Valencia, Spain On behalf of the NEXT The goal of the next generation of neutrino-less double beta experiments is to explore fully the degenerate hierarchy, which is corresponding to an effective neutrino mass up to 50 meV. This neutrino mass range will be only accessible for experiments that can simultaneously achieve a large fiducial mass and negligible backgrounds. A high-pressure, gaseous Xenon TPC has good prospects to combine many desirable features to achieve this goal. Our collaboration is planing a Xenon detector of a mass of about 100 kg. Such an object could explore the degenerated hierarchy and provide a deep understanding of the experimental techniques to suppress backgrounds required for larger detectors. The current status of our R&D towards a Neutrino Experiment with a Xenon TPC (NEXT) will be presented.
N16-5:

Streamlined Calibration of the ATLAS Muon Spectrometer Precision Chambers

D. S. Levin Physics, University of Michigan, Ann Arbor, MI, USA On behalf of the ATLAS Muon Collaboration The ATLAS Muon Spectrometer, a six story structure embedded in an air-core magnetic field, is comprised of nearly 1200 optically Monitored Chambers (MDTs) containing 354,000 aluminum drift tubes. These chambers are configured in barrel and end-cap regions. The operating gas, 92.9% + Ar 7% CO2 + 0.1 % H2O at 3 bar is pumped through the spectrometer at 100,000 l/hr. Tracking gas variability is critical to the spectrometer calibration. The design momentum resolution required for the LHC physics reach (dp/p = 10% at 1 TeV) demands that MDT calibrations be determined daily. These calibrations (RT functions) enable the track impact parameters to be extracted from the measured drift times. They are optimally determined at chamber level granularity. This CPU and data intensive task is relegated to dedicated CPU farms in Tier-2 centers which receive a special calibration data stream and are intended to operate continuously with a one day turnaround. For initial LHC running and for backup purposes, we have established an alternate, less resource demanding procedure: A Universal RT calibration, derived from an inline gas system monitor chamber, and compensated for localized temperatures in the spectrometer, is applied to each spectrometer chamber. The gas monitor, in all operational respects identical to the MDTs, continuously records cosmic-ray induced drift-time spectra. With separate gas partitions for spectrometer gas inflow and outflow, diurnal changes to gas properties, such as drift velocity, are faithfully tracked with better than 0.1% precision. We report results on spectrometer performance in cosmic ray commissioning runs using the Universal RT calibration.
N16-6:

Glass RPC Detector R&D for a Mega Neutrino Detector

S. Bheesette Dept. of High Energy Physics, Tata Institute of Fundamental Research, Mumbai, Maharashtra, INDIA On behalf of the INO Collaboration India-based Neutrino Observatory (INO) collaboration has proposed a 50 kton magnetised iron calorimeter (ICAL), whose primary goals are to precisely determine oscillation parameters of the atmospheric neutrinos, to study matter effects on the oscillations and finally to use it as a long baseline detector for the neutrino beams. The main requirements of the ICAL detector are large target mass, modularity, good tracking and energy resolutions. Good directionality (translating to a time resolution of 121

about 1nS) and charge identification of the detecting particles are the other essential capabilities of this detector, which is proposed to compliment the potentials of other existing and proposed detectors. ICAL will comprise of about 150 layers of low carbon 56 mm thick iron plates. Lateral dimensions of this cubical geometry detector are 48m x 16m x 14.5m. The iron absorber will be magnetised to a strong field of about 1.5Tesla. Sandwiched between these layers are glass Resistive Plate Chambers (RPCs), which are used as the active detector elements. RPCs are rugged and low-cost gas detectors and provide excellent spatial and temporal resolutions. The ICAL detector will cover an area of about 100,000m2 and will use about 30,000 RPCs of dimensions 2m x 2m. An aggressive R&D program to develop and characterise RPCs operating in the avalanche mode was undertaken. We have fabricated a large number of glass RPC prototypes of 1m x 1m in size and have studied in detail, their performance and long term stability. A detector stack of 12 RPCs, readout by an indigenously built electronics and data acquisition system is currently tracking about 150,000 cosmic ray muons per day for about two years. We will review our various activities towards development of glass RPCs. We will present results of their characterisation studies and discuss our plans to develop 2m x 2m sized RPCs. Some interesting results obtained by the detector stack will also be highlighted.
N16-7:

Neutron Calibration of the 3He Proportional Counters of the Sudbury Neutrino Observatory

K. Boudjemline Department of Physics, Carleton University, Ottawa, Ontario, Canada On behalf of the SNO Collaboration The Sudbury Neutrino Observatory (SNO) used an array of 3He proportional counters (Neutral Current Detectors, NCDs) to measure the rate of neutral-current interactions in heavy water and precisely determined the total active 8B solar neutrino flux. The neutron detection efficiency and the response of the NCD array have been determined with a variety of neutron calibration sources. Neutron point sources (252Cf and 241AmBe) were frequently deployed throughout the detector volume to measure the temporal stability and the detector gain of the NCD array. The NC neutron detection efficiency was studied by using an isotropic source of neutrons produced by mixing 24Na in the form of activated NaCl, into the heavy water. Neutrons were produced by deuterium photodisintegration induced by the 2.75 MeV gammas. The largest uncertainties on the neutron detection efficiency were associated with the knowledge of the 24Na source strength and the ability to determine the uniformity of its mixing in the heavy water. The inferred NC neutron capture efficiency for the NCD array was 0.211+/- 0.007 in good agreement with the 0.210 +/-0.003 given by a Monte Carlo simulation verified against point-source data.
N17: Computing and Software for Experiments II: New Computing Technologies Tuesday, Oct. 27 N17-1: 13:30-15:30 Grand Ballroom 7

Simulations of Distributed Systems in a Computing Centre

E. Ronchieri, D. Dongiovanni, L. Dell'Agnello, T. Ferrari INFN CNAF, Bologna, 40126, Italy A computing centre of High Energy Physics has to deal with the difficult dimensioning and configuration of several services accessed by a variety of users from different experiments. Often both the identification and parameter optimization of factors involved in the service deployment are a painful issue. For example is a new configuration able to support a certain number of requests per minute? Sometimes these activities are not well-planned and well-verified by computing centre administrators causing problems during the usage of such services and forcing administrators to apply emergency solutions. In addition, it can happen that computing centres do not have hardware machines to set up a testbed environment to be adopted in order to validate each service configuration according to experiments requirements. In this paper, we define a procedure of services modeling in order to develop a flexible simulator of possible distributed services available in a computing center. After having identified a service to model and a referencing metric to evaluate its performance, we select a set of variables possibly contributing to model the service and analyze their relevance to the service model. In particular it is possible to identify which variables do not affect the measure, eventually reaching a subset of variables from the initial set which mainly characterize the model. A second phase focuses on analyzing the impact of variable changes on the referencing metric. We apply it to the GridFTP service and IBM GPFS in order to model its performances over a SAN using simulation results of a storage testbed setup at Tier1-CNAF. Unfortunately that storage testbed is not always usable for dimensioning and configuration issues, therefore we believe that the development of such simulator can help. We were able to model a given service and to integrate it in a simulator prototype.

122

N17-2:

Hadoop Distributed File System for the Grid

H. Pi University of California, San Diego, La Jolla, CA, USA On behalf of the Open Science Grid Data distribution, storage and access are essential to the CPU- and data-intensive high energy computing. A newly emerged storage system, the Hadoop Distributed File System (HDFS), is deployed and tested within the Open Science Grid middleware stack. Various efforts have been taken to improve the performance of CPU utilization and data accessing throughput. The scalability test shows sustained high inter-datanode data transfer can be achieved with the cluster fully loaded with user analysis jobs. The WAN transfer to hadoop supported by Bestman and tuned gridFTP shows very high scalability and robustness of the system. The hadoop client can be easily deployed at interactive machines and laptop to support remote data access. The ability to automatically replicate precious data is especially important for Tier-2 and Tier-3 sites within the LHC computing grid to store user level data reliably. The simplicity of operations of HDFS significantly reduces the cost of ownership of Petabyte scale storage over alternative solutions.
N17-3:

Using the "Java Platform, Enterprise Edition" for Data Handling, Transfer and Storage.

S. Patton ASC/CRD, Lawrence Berkeley National Laboratory, Berkeley, CA, USA On behalf of the Daya Bay Reactor Neutrino Collaboration

With a wide availability of open source implementations of the Java Platform, Enterprise Edition, (Java EE) has become a good candidate for deploying applications that handle many standard housekeeping tasks such as data handling, transfer and storage. The Daya Bay Reactor Neutrino Collaboration is a joint China/USA experiment designed to measure Theta_13, has decide to use this approach to deploy its solutions to these tasks and this presentation will review cost, benefits and results of this decision. The Java EE platform is a collection of interface specifications that cover the most common tasks needed to provide a robust enterprise application. These standards cover such areas as persistency, web presentation and services, email and inter-application messaging. The immediate benefit that can gained from writing software to these interfaces is that not only can one tap in to the vast store of experience that went into creating them, but there are now plenty of implementations of them so not only is it possible to avoid re-inventing the wheel, but it is not even necessary to build it but simply to download it and be up and running in a short time. The complete Java EE platform can be a bit overwhelming to begin with, therefore this presentation will demonstrate the power of Java EE by highlighting how it copes with Daya Bay's data handling, transfer and storage. This data handling covers the retrieval of data from the Online/DAQ system, preparing it for offline analysis, including passing it to on-site reconstruction, transferring it from the experimental site to the major data centers of the experiment in both China and the USA, and managing it in those warehouses once it has arrived.
N17-4:

Virtualization for Physics Analysis in ATLAS

Y. Yao Lawrence Berkeley National Laboratory, Berkeley, CA, USA On behalf of the ATLAS Collaboration The software systems for data analysis in large-scale HEP experiments (e.g. ATLAS) are very complex and difficult for nondevelopers to learn and use. With modern computers equipped with large physical memory and multiple CPUs (cores), one can run ATLAS software inside a virtual machine (VM) on a users laptop without much performance penalty. Free virtualization software (hypervisors) are readily available for most operating systems and are easy to install and use for non-computer professionals. Another benefit of running ATLAS software inside a VM is that it eliminates the effort to port it to other platforms. We choose the CernVM as the virtual appliance to deploy ATLAS software. CernVM provides a powerful yet easily extendable web-interface to manage virtual machines. One extension is to enable users to connect to a X11 server running inside the VM. This enables users to run complex X11 applications (e.g. ATLAS Event Display, Eclipse) regardless of their OS. It can also eliminate the developers effort to support analysis software for multiple OS. The size and update frequency of ATLAS software makes it very hard to run ATLAS software on a personal computer. CernVM builds-in the CernVM Files System, which only download the needed files. It also allows us to distribute the nightly builds, which is not easy to do with conventional method. Many physics analysis software (e.g. ROOT, JAS, PAW) or GRID tools 123

(gLite) can also be deployed with CVMFS, which can turn any laptop/desktop into a physics analysis oriented workstation. Furthermore, CernVM is easily portable to IaaS clouds (e.g. Amazon EC2), which enables physicists to setup low-cost online analysis workstation. We integrate with CernVM the Virtual Machine Logbook to help users save work states, manage multiple projects and share their work. In this work we aim to utilize virtualization technology to increase the usability of ATLAS software and the efficiency in the development and data analysis process.
Co-Author: Asoka De Silva , TRIUMF, Vancouver, BC, Canada N17-5:

Ganga - a Single, User Friendly Grid Job Toolkit Across Multiple Disciplines

B. H. Samset Department of Physics, University of Oslo, Oslo, Norway On behalf of the ganga team The use of computing grids is ever increasing in many areas of science. To avoid exposing new users to the compexities of grid technology, flexible but intuitive user tools must be developed. We present the ganga toolkit, a universal job definition, submission and management toolkit for grid computing. With version 5 released in 2008, ganga is a mature, stable and widelyused tool with long-term support from the high energy physics community. It is in production use by two major experiments at the LHC, and also has users across medical physics, fusion technology and several other areas. The main strengths of ganga are its easily extendible modular structure, allowing plugins to be written combining any grid solution and experiment software, and its user-friendly interfaces, allowing fast access to any tecnology with an existing plugin. These include traditional batch and grid backends such as Condor, LSF, PBS, gLite/EDG and ARC. A point-to-point job execution via ssh on remote machines is now also supported, and tools for automated job submission are provided. Ganga is python based, open source and well documented, allowing any interested science collaboration to try it out with a minimum of effort.
N17-6:

Software Quality Engineering by Parts

T. M. Evans Nuclear Science and Technology, Oak Ridge National Laboratory, Oak Ridge, TN, USA We discuss a unit-testing based software practice that can greatly improve the production of scientific software. Defects in parallel scientific simulation codes can be very difficult to find, and they can invalidate years of published results. Unit-testing is a means of software verification that attempts to remove bugs as they are introduced in the code. We have used this technique in the Denovo discrete ordinates transport code at Oak Ridge National Laboratory since its inception in 1997. Using this practice, we have not had any bug reports since Denovo entered production usage in January 1998.
N17-7:

Optimizations in Python-based Analysis

W. T. L. P. Lavrijsen LBNL, Berkeley, CA, USA Analysis codes that are written in python do so mostly to take advantage of the ease of writing in python (such as interactivity, not having to declare variables, the garbage collector, etc.) which allows for a quick turn-around when testing ideas. Although the language offers them, analysis codes are typically not using many dynamic features, especially if external libraries written in a static language, such as C++, are used extensively. That is, when looking at the code actually being executed at run-time, it is possible for a tool to determine the types and functions used, and with that information ``freeze out" a compiled or even parallelized version of the analysis program. The current generation of tools, one of which based on PyPy will be presented here, does however still require guiding structure from the analysis author. This paper will show how to construct and write python code with an eye for optimization and possible parallelization by automated tools.

124

J04: Joint NSS/MIC 4 Tuesday, Oct. 27 J04-1: 16:00-18:00 International Ballroom North

The Digital Silicon Photomultiplier - A Novel Sensor for the Detection of Scintillation Light

C. Degenhardt, G. Prescher, T. Frach, R. de Gruyter, A. Schmitz, R. Ballizany Philips Corporate Technologies, Aachen, Germany Silicon Photomultipliers (SiPMs), arrays of avalanche photodiodes operated in Geiger-mode, are attractive alternatives to Photomultiplier Tubes for reasons of ruggedness, compactness or insensitivity to magnetic fields. Other advantages of solid state detectors are their low operating voltage, low power consumption and large scale fabrication possibilities. On the other hand, current solid state detectors are limited when it comes to the detection of very low light fluxes or precise timing measurements and their gain is very sensitive to temperature variations. The Digital Silicon Photomultiplier (dSiPM) presented here overcomes those problems by early digitization of the Geiger-cell output and integrated electronics on chip. We developed a Digital SiPM of 3.8x3.3mm2 in size containing 8188 individual cells. Each detected photon is converted into a digital signal as early as possible in each of the Geiger-mode cells of the sensor. In addition, the complete trigger logic and the time-to-digital converter are integrated into the sensor. To show the performance of the sensor, scintillation crystals of different sizes and materials using different reflector materials were coupled to the sensor and irradiated by gamma radiation of 662keV, 511keV and 49keV. The energy resolution at 511keV using LYSO scintillators was determined to be 10.8%, being comparable to results obtained with PMTs. The timing resolution of 190ps FWHM for 4x4x5 mm3 LYSO crystals constitute the best timing resolution ever obtained with this scintillator material. In addition, the influence of temperature on photon detection efficiency and timing characteristics of the sensor will be presented. The results show that the Digital Silicon Photomultiplier is a promising novel detector for the detection of low light fluxes as encountered in scintillation detector readout, especially in cases where a good timing resolution is mandatory.
J04-2:

V. V. Nagarkar1, V. Gaysinskiy1, O. Ovechkina1, S. Miller1, B. Singh1, L. Guo2, T. Irving2 1 RMD, Inc., Watertown, MA, USA 2 Illinois Institute of Technology, Chicago, IL, USA We report on a novel approach to produce oxygen-doped zinc telluride (ZnTe:O), a prominent and remarkable group II-IV semiconductor scintillator, in the columnar structured or polycrystalline forms needed to fulfill the needs of many highly demanding X-ray and -ray imaging applications. Having one of the highest conversion efficiencies among known scintillators (well above 100,000 photons/MeV), emission around 680 nm (which is ideally suited for CCD sensors), high density of 6.4 g/cm3, fast decay time of ~1 s with no afterglow, and orders of magnitude higher radiation resistance compared to commonly used scintillators, ZnTe:O has numerous applications in X-ray imaging, nuclear medicine (particularly SPECT), room temperature radioisotope identification, and homeland security, to name a few. Additionally, ZnTe:O offers distinct advantages for synchrotron-based high resolution imaging due to the absence of atomic absorption edges in the low energy range, which otherwise reduce resolution due to secondary X-ray formations. Our approach to growing ZnTe:O uses a specialized vapor deposition technique that allows large-area structured scintillator fabrication in a time- and cost-efficient manner. This method makes use of commercially available ZnTe material that we ball mill in an O2 (air) atmosphere and then anneal in a forming gas to obtain ZnTe:O. We then vapor deposit the resulting material onto a suitable substrate. We have identified favorable deposition conditions that have already allowed us to grow up to 1 mm thick material in columnar form or polycrystalline non-columnar layers, with stoichiometry appropriate for producing efficient scintillation. Here we will report on our fabrication of ZnTe:O films and characterization of their optical, scintillation and imaging properties.
J04-3:

Bright Semiconductor Scintillator for High Resolution X-Ray Imaging

Z. Wang1,2, Z. Huang1,2, L. Zhang1,2, Z. Chen1,2, K. Kang1,2 1 Engineering Physics, Tsinghua Univercity, Beijing, China 2 Key Laboratory of Particle & Radiation Imaging (Tsinghua University), Ministry of Education, Beijing, China X-ray phase contrast imaging (PCI) is a hot research field in the last decade. Many novel methods such as diffraction enhanced imaging, grating-based imaging et al come to the fore. As a result, the demand on Monte Carlo simulation in PCI emerges. The Monte Carlo simulation can simplify the phase contrast system modeling and is useful for system optimization and new method validation. Geant4 is a popular Monte Carlo toolkit that has been applied successfully in various research fields. Geant4 embeds general low energy x-ray physics which is suitable for x-ray phase contrast imaging research. However, currently Geant4 does not provide the capability for x-ray phase contrast physics. In this paper, a Monte Carlo tool based Geant4 that implements the xray refraction effect is proposed. It adds the capability of phase contrast imaging to Geant4. The tool is validated by a simple 125

Implement X-Ray Refraction Effect in Geant4 for Phase Contrast Imaging

sphere simulation. The result of the validation is in perfectly accordance with the theoretical value. The tool is also applied to model the grating-based imaging method to present the potential applications in x-ray phase contrast research.
The work was supported by a grant from the National Natural Science Foundation of China (No. 10875066). J04-4:

A. Manor1, A. Osovizky1, E. Dolev2, E. Marcus2, D. Ginzburg1, V. Pushkarsky1, Y. Kadmon2, Y. Cohen2 1 Health Physics Instrumentation Department, Rotem Industries Ltd, Beer-Sheva, Israel 2 Electronics & Control Laboratories, Nuclear Research Center - Negev, Beer-Sheva, Israel

Compensation of Scintillation Sensor Gain Variation During Temperature Transient Conditions Using Signal Processing Techniques

This work presents a theoretical analysis along its practical application aiming to improve stabilization of a spectrum in a transient temperature conditions both at slow temperature variations and extreme temperature transients. The gain stabilization of the detector spectrum has a crucial importance in hand held isotope identification device. The temperature dependence appears as spectrum gain shift that may result in a faulty identification result. Existing gain compensation methods operates very well under steady state conditions. However, the time needed to achieve stabilization of the spectra gain after a temperature transient, proved to be long. A step by step method for constructing a comprehensive scintillation detector temperature gain compensation scheme was developed and tested. An application example was used to evaluate the process on detectors of different configuration. The scheme can be easily applied to solve the real life problem of scintillation sensor gain variation due to temperature transient. A set of experiments has been conducted and discrete time methods have been employed to analyze the data and devise a digital filter to supplement an existing compensation scheme for the spectrum gain stabilization. A transfer function was obtained using discrete time analysis methods namely system identification and transform to model the transient response. Once a model was obtained it was transformed into a difference equation: This equation is in fact a digital filter that estimates the actual radiation sensor gain from the temperature sensor measurements. The data provided by the transfer function precisely estimates the gain of the radiation sensor. Therefore this approach can provide a practical method for a gain compensation schema that can be employed to stabilize the sensor gain. The improved gain estimation allows the employment of identification algorithms otherwise not possible or not reliable without the use of radioactive marker source.
J04-5:

F. Nishikido1, Y. Yazaki2, H. Osada2, N. Inadama1, T. Inaniwa1, S. Satoh1, K. Shibuya3, E. Yoshida1, T. Yamaya1, H. Murayama1 1 National Institute of Radiological Sciences, Chiba, Japan 2 Chiba University, Chiba, Japan 3 Tokyo University, Tokyo, Japan We are investigating the influences of heavy ion irradiation on the detectors for in-beam OpenPET by using the Heavy Ion Medical Accelerator in Chiba (HIMAC) at the National institute of Radiological Sciences (NIRS). OpenPET which consists of two detector rings separated axially is suitable for in-beam PET in heavy ion therapy due to the open space between the separated detector rings. All of the annihilation gamma rays from irradiated volumes enter obliquely into the OpenPET detectors. Therefore, detectors which can measure depth of interaction (DOI) are required to reduce parallax error. Previously, we have succeeded in crystal identification using the DOI detector consisting of four-layer crystal blocks and a 256-channel positionsensitive photomultiplier tube by optimization of the reflector arrangement. We plan to adopt DOI detectors using this method as the OpenPET detectors; however, secondary particles entering into this type detector have a possibility of degradation of detector performances. In the experiment described here, the energy and intensity of the 12C beam were 290MeV/u and 108 -109 particle per second (pps). Carbon ions entered into a water phantom and the influences of irradiation, especially fragment particles, on DOI detectors were evaluated. In this presentation, we report the results of this experiment.
J04-6:

Influence of Secondary Particles to Detectors for in-Beam OpenPET in Heavy Ion Therapy

M. Zhuravleva1, C. L. Melcher1,2, L. Eriksson1,2,3,4,5 1 Scintillation Materials Research Center, University of Tennessee, Knoxville, TN, USA 2 Materials Science and Engineering Department, University of Tennessee, Knoxville, TN, USA 3 Siemens Medical Solutions Molecular Imaging, Knoxville, TN, USA 4 Department of Physics, University of Stockholm, Stockholm, Sweden 5 Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden

Photoluminescence Characterization of Scintillators for Phoswich Detectors

Phoswich detectors made of two or more layers of scintillator crystals with different decay constants can provide information about depth-of-interaction (DOI) of gamma rays where identification of the crystal layer is determined via pulse shape discrimination techniques. Advantages of the concept have been demonstrated in the high resolution PET system HRRT using LSO/GSOor LSO/LYSO crystals [1]. Previous work [2] has shown that for certain pairs of scintillators, the light emitted by one crystal layer can be absorbed (photo-excitation) and re-emitted (photoluminescence) by the second layer. A scintillators rise time 126

and decay time under photo-excitation are often different from gamma-excitation due to the absence of electron and hole trapping during energy transfer from crystal host to luminescent center after irradiation [3]. Therefore, selecting the most suitable combination of scintillators may depends on their photoluminescence as well as their scintillation properties. In this paper, we report experiments on the photoluminescence properties of some promising phoswich materials such as LSO:Ce, LuAG:Pr, GSO:Ce, LPS:Ce, YAG:Ce, LuYAP:Ce, with an emphasis on investigating excitation/emission spectra, rise times and decay times and to clarify the interactions between those if applicable. Photoluminescence decay profiles were recorded using the timecorrelated single photon counting technique with a spectrofluorometer (Fluorolog 3, Horiba Jobin Yvon). The excitation source was a light emitting diode with a pulse width of 1 ns. Fundamental understanding of luminescence processes resulting from crystal-to-crystal interactions is very useful to optimize pulse shape discrimination in various detector configurations. Accurate quantitative data on complex photoluminescence and scintillation kinetics provide practical information for construction of real phoswich detectors.
References: [1] H.W.A.M. de Jong, F.H.P. van Velden, R.W. Kloet, F.L. Buijs, R. Boellaard, and A.A. Lammertsma, Phys. Med. Biol. 52 (2007) 1505 [2] L. Eriksson, C.L. Melcher, M. Eriksson, et al., IEEE Trans. Nucl. Sci. 56 (2009) 182 [3] M. Nikl, A. Vedda, V.V. Laguta, Radiation Measurements 42 (2007) 509 J04-7:

L. F. Carramate1, C. A. Santos1, C. A. Oliveira1, A. L. Silva1, M. Peres1, A. M. da Silva2, J. M. dos Santos3, J. F. Veloso1 1 I3N, Physics Department, University of Aveiro, Aveiro, Portugal 2 DETI/IEETA, University of Aveiro, Aveiro, Portugal 3 GIAN, Physics Department, University of Coimbra, Coimbra, Portugal A small field of view energy resolved Computed Tomography (CT) system based on a MicroPatterned Gaseous Detector (MPGD) is proposed for small animal imaging and envisaging mammography. MPGDs present excellent properties for single photon detection, 2D capability, absence of electronic noise (by setting a threshold), high counting rate capability, etc. The present system, based on the CT third generation, uses an x-ray detector filled with pure xenon using a MicroHole and Strip Plate for electron amplification. The MHSP is a hybrid microstructure with two charge multiplication stages, allowing charge multiplication above 10^4. A very simple implementation for 2D-imaging was build using two thin orthogonal resistive lines; between the strips on the top side and between the anode strips on the bottom side, allowing obtains the actual position. Taking advantage from detector energy resolution capability, Energy Weighting Techniques (EWT) will be applied for contrast enhancement. First multi-slice energy resolved x-ray tomographic images produced by the system will be evaluated and presented together with a discussion of the image quality and future prospects.
Acknowledgements: This work was supported by project CERN/FP/83487/2008 through FEDER and FCT (Lisbon) programs. N18: Gamma-ray Imaging I: Compton Imaging Tuesday, Oct. 27 N18-1: 16:00-18:00 Grand Ballroom 2

Small Field of View Quantum Computed Tomography System Based on a MPGD

A. Zoglauer1, S. E. Boggs1, R. M. Kippen2 1 Space Sciences Laboratory, University of California at Berkeley, Berkeley, CA, USA 2 Los Alamos National Laboratory, Los Alamos, NM, USA

Image Reconstruction for Compton Telescopes Based on a Partially-Binned List-Mode Approach

We present the latest developments of a partially-binned list-mode imaging approach for Compton telescopes, especially the Nuclear Compton Telescope (NCT). Its response representation combines the best aspects of binned mode (a complete description of all aspects of the response, even those not easily accessible by an analytical description) and list-mode (low memory requirements). This is achieved by only binning certain aspects of the response (especially those describing the shapes of the Compton cones in the data space) and calculating the rest on-the-fly. Combined with a standard MLEM iterative image reconstruction algorithm the given response representation results in higher source to noise ratios compared to a conventional list-mode approach (since there are no more events outside the response) while keeping the angular resolution.
N18-2:

L. Mihailescu1, J. Rohel1, K. Vetter1,2, J. Maltz1, D. Chivers2, M. Amman1, G. Gullberg1 1 Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 Nuclear Engineering Department, UC Berkeley, Berkeley, ca, USA The purpose of the present work is imaging of gamma-ray sources distributed from near- to intermediate-fields in three dimensions. For increased imaging performance and processing speed, a CAD-generated scene of objects is integrated into the 127

Computer-Aided Design Integrated Gamma-Ray Imaging

gamma-ray image reconstruction process. The imported CAD model is used to restrain the gamma-ray image field of view to the interior of physical objects. Subsequently, a dynamic pixelization algorithm is used to create imaging voxels inside CAD objects. The imaging method and the associated imaging algorithms are described. Results of test demonstration measurements performed in the laboratory are presented.
N18-3:

Application of the Energy-Imaging Integrated Deconvolution Algorithm for Source Characterization

W. Wang, C. G. Wahl, J. M. Jaworski, Z. He Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA The energy-imaging integrated deconvolution algorithm is capable of deconvolving the source image at any specific energy, as well as the incident spectrum in any direction. This spectrum image reconstruction method takes place in integrated spatial and energy space using the maximum likelihood expectation maximization (MLEM) algorithm. By including Compton scattering events in the system model and using a sensitivity image to correct for non-uniform efficiency, the MLEM solution estimates the true incident gamma-ray spectrum. The full-energy peak areas from unshielded sources hence show the true branching ratios. We can then identify the sources placed around the detector based on the peak energy and the relative peak areas for each direction. After identifying the isotope, the source activity can be estimated. From the reconstructed image and the difference between expected and measured peak-area ratios, the presence of shielding and the type and thickness of the shielding can be determined.
N18-4: Point-Source Detection Using Energy and Imaging Information from 3D-Position-Sensitive Semiconductor Detectors

C. G. Wahl, Z. He Dept. of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, United States Three-dimensional-position-sensitive semiconductor detectors provide both the energy and the locations of each gamma-ray interaction in the detector. From these data, Compton imaging methods can reconstruct the source distribution around the detector as a function of energy. Automated detection methods have been developed to use this combined energy and imaging information to detect the presence of point sources of radiation in an unknown background environment, as well as to identify the isotope, localize its position, and estimate the source strength. Performance will be compared with other standard detection methods as a function of false-alarm probability using experimental data collected on a 20 mm x 20 mm x 15 mm CdZnTe detector.
N18-5:

J. S. Maltz1, L. Mihailescu1, G. T. Gullberg1, D. H. Chivers1,2, J. Rohel1,2, K. Vetter1,2 1 Lawrence Berkeley National Lab, Berkeley, CA, USA 2 University of California at Berkeley, Berkeley, CA, USA

Joint Spatio-spectral Compton Camera Reconstruction Algorithm for Arbitrary Detector Geometries

We present a reconstruction algorithm for Compton imager (CI) data that is able to simultaneously determine source location, activity, and energy spectrum while accounting for the non-uniform geometry and composition of the detector system. By performing joint spatio-spectral reconstruction, it is possible to estimate incoming photon energy even for Compton continuum (CC) events where some of the photon energy escapes the detector. Methods: In classical formulations of CI system response, the sequence probability for each event is described in terms of products of probability density functions (pdfs) describing interaction probabilities. We formulate pdfs that represent the joint probability of interaction and detection, rather than simply interaction. In doing so, we are able to remove the bias present in the classical formulation in cases where pdfs are normalized over interactions that cannot be detected owing to non-isotropic response. The algorithm makes use of attenuated ray tracing to calculate detection sensitivities on-the-fly. We evaluate the method using Monte Carlo simulation applied to the geometry of a CI with a highly anisotropic response that is composed of 4 parallel planar semiconductor detectors: 2 large-volume Si(Li) and 2 large-volume HPGe detectors. This imaging system is available in our laboratory and is used for demonstration experiments. Results: The accuracy with which the algorithm determines the correct initial photon energy for CC events is increased approximately 30% relative to algorithms based on interaction pdfs only. Conclusion: The geometry-aware algorithm models elements of the pdfs describing interaction sequences that are often more important in determining sequence probability than classical considerations such as the Klein-Nishina differential cross section. Although geometric modeling increases computational burden, the most costly operation involved, that of ray tracing, is highly amenable to hardware acceleration.
This material is based upon work supported by the U.S. Department of Homeland Security.

128

N18-6:

The Power of Gamma-Ray Imaging to Detect Special Nuclear Material

D. Wright, L. Hiller, K. Nelson, K. Kazkaz Physics, Lawrence Livermore National Laboratory, Livermore, CA, USA Gamma-ray imaging has the potential for making a large impact in the detection of special nuclear material (SNM) in a host of application scenarios from non-proliferation to counterterrorism. There is also a very wide range of imaging technology modalities that have dramatically different trade offs in key performance characteristics such as efficiency, resolution, field-ofview, etc. It is not obvious which modalities are best suited for a particle scenario, or even if a specific technology implementation is capable of achieving a useful level of performance in a real-world situation. The gamma-ray spectra from sources of interest tend to be complex, weak, and similar to environmental background while technology tests in the laboratory often resort to test sources that are far easier to detect than the real SNM, but are also much easier to obtain. We have developed a quantitative evaluation framework to provide a capability to compare specific examples of imaging technologies and measure their absolute performance for realistic application scenarios. Using a combination of simulations and measurements, we have applied this framework to an urban search scenario in which we compare both imaging and non-imaging detectors, as well as evaluate the impact of advanced algorithms on the final performance.
N18-7:

H. Kubo1, K. Hattori1, N. Higashi1, C. Ida1, S. Iwaki1, S. Kabuki1, S. Kurosawa1, K. Miuchi1, K. Nakamura1, H. Nishimura1, J. D. Parker1, T. Sawano1, A. Takada2, M. Takahashi1, T. Tanimori1, K. Taniue1, K. Ueno1 1 Dept. of Physics, Graduate School of Science, Kyoto University, Kyoto, Japan 2 Scientific Balloon Laboratory, ISAS/JAXA, Kanagawa, Japan

Compton Gamma-Ray Camera Using an Electron-Tracking Gaseous TPC and a Scintillation Camera

We have developed a Compton gamma-ray camera which measures the three dimensional track of a Compton-recoil electron and thereby constrains the incident direction of each photon to an arc of the Compton event circle. The camera consists of two parts; One is a gaseous time projection chamber (-TPC) with a size of 30x30x30 cm^3 with a micro-pixel-chamber (-PIC) readout, which measures the three dimensional track and energy of the Compton-recoil electron. The other is a position-sensitive scintillation camera surrounding the -TPC, which consists of arrays of pixellated GSO(Ce) scintillators coupled to multi-anodes PMTs and measures the direction and energy of the scattered gamma ray. By reconstructing the incident gamma ray from these measurements, we have successfully obtained a gamma-ray image in the 0.3-1 MeV range. The angular resolutions of an angular resolution measure (ARM) and a scatter plane deviation (SPD) are 11 and 110 degrees (FWHM) for single photon with 662 keV, respectively. With this camera, we are also developing an imaging method for a gamma ray with energy of several dozen MeV by measuring tracks and energies of an electron-positron pair produced in the -TPC.
N19: Analog and Digital Circuits II Tuesday, Oct. 27 N19-1: 16:00-18:00 Grand Ballroom 3

S. Conforti1, C. de La Taille1, G. Martin-Chassard1, F. Dulucq1, W. Wei2 1 Laboratoire de l Accelerateur lineaire, Orsay, France 2 IHEP, Bejing, China

PARISROC, a Photomultiplier Array Readout Chip

PARISROC is a complete read out chip, in AMS SiGe 0.35m technology [1],for photomultipliers array. It allows triggerless acquisition for next generation neutrino experiments and it belongs to an R&D program funded by French national agency for research (ANR) and called PMm2:Innovative electronics for photodetectors array used in High Energy Physics and Astroparticles [2] (ref.ANR-06-BLAN-0186). The ASIC integrates 16 independent channels with variable gain and provides charge and time measurement by a 12-bit ADC and a 24-bits Counter.
N19-2:

BASIC: an 8-Channel Front-End ASIC for Silicon Photomultiplier Detectors

C. Marzocca DEE- Politecnico di Bari, BARI, ITALY On behalf of the INFN-DASiPM2 collaboration A current-mode analog front-end suitable for Silicon Photo-Multiplier (SiPM) detectors has been designed in a standard CMOS 0.35m technology to fulfill the severe constraints imposed by this kind of detectors in terms of dynamic range and speed of operation, without excessive power consumption. Timing accuracy better than 60ps has been obtained by means of a fast current discriminator. The circuit is intended for medical imaging applications but programmability of detector bias, gain and threshold current enhances the flexibility of the architecture, making it suitable for different kind of SiPM detectors and applications. An 8129

channel ASIC (BASIC), suitable for the read-out of SiPM arrays has been manufactured following this approach. Here we report the measurement results obtained by coupling a single channel to a single SiPM detector excited by a pulsed blue LED.
N19-3:

G. De Geronimo1, G. Carini1, W. Chen1, J. Fried1, J. Keister1, Z. Li1, D. A. Pinelli1, P. Rehak1, D. P. Siddons1, E. Vernon1, J. A. Gaskin2, B. D. Ramsey2 1 Instrumentation, Brookhaven National Laboratory, Upton, NY, USA 2 Marshall Space Flight Center, Huntsville, AL, USA We present an application specific integrated circuit (ASIC) for high resolution x-ray spectrometers (XRS). The ASIC was designed to read out signals from pixelated silicon drift detectors (SDDs). There is no integrated field effect transistor (FET) in the pixel, rather, the readout is done by wire-bonding the anodes to the inputs of the ASIC. The ASIC dissipates 32 mW and provides 16 channels of low-noise charge amplification, high-order shaping with baseline stabilization, discrimination, and peak detection with analog memory. The readout is sparse and based on a custom low-power tristatable low-voltage differential signaling (LPT-LVDS). A unit of 64 SDD pixels, read out by four ASICs, covers an area of 12.8 cm2, and it dissipates less than 20 mW/cm2. The 64-pixel units can be tiled to cover an arbitrarily large detection area. Our preliminary measurements show a FWHM of 153 eV at the 5.9 keV peak of a 55Fe source and of about 90 eV on a test pulse line at 282 eV.
N19-4:

ASIC for SDD-Based X-Ray Spectrometers

G. De Vita1,2, S. Herrman1,2, T. Lauf1,2, M. Porro1,2, J. Treis1,2, A. Wassatsch2,3, L. Bombelli4,5, C. Fiorini4,5 1 Max Planck Institut fuer extraterrestrische Physik, Munich, Germany 2 MPI HLL, Munich, Germany 3 Max Planck Institut fuer Physik, Munich, Germany 4 Dipartimento di Elettronica e Informazione, Politecnico di Milano, Milan, Italy 5 INFN, Milan, Italy

64-Channel ASTEROID ASIC: Experimental Performance and Measurements with Macro Pixel Arrays for X-Ray Astronomy

We present the latest experimental results on 64-channel ASTEROID ASIC, designed to readout DEPFET pixel matrices for astronomy. The sensor has been designed for the X-ray Astronomy application BepiColombo. This is a satellite based mission that requires a detector system with high speed readout, high energy resolution and radiation hardness properties. In order to cope with the high resolution requirement, ASTEROID implements a trapezoidal weighting function that is the time-limited optimum filter for white series noise. The performance of the 64 channel ASTEROID ASIC, implemented in the 0.35m 3.3V AMS CMOS technology, coupled with a 64x64 Macro Pixel array prototype with pixel area of 500x500m2 have been tested for the first time. A spectroscopic resolution of 128.5eV FWHM with the reconstruction of all split events has been measured on the Mnk of a 55Fe source operating ASTEROID with a total readout time of 4.2s. The ASIC comprises a digital section that generates the timing signals to operate the 64 channels in parallel. This digital section is based on SEU-immune dual port memory cells. Radiation hardness measurements on the digital part both with a 10MeV proton flux and an Americium alfa particles source have been performed.
N19-5:

T. Fusayasu1, S. Koshimuta2, H. Hamagaki3, Y. Tanaka1, M. Inuzuka4 1 Nagasaki Institute of Applied Science, Nagasaki, Japan 2 SciEnergy Co., Ltd., Kanagawa, Japan 3 University of Tokyo, Saitama, Japan 4 National Research Institute for Cultural Properties, Tokyo, Japan

A Frontend LSI for Large-Area Gamma-Ray Imaging Detectors with Gas Electron Multipliers

A prototype frontend LSI has beed designed and fabricated for GEM-based gamma-ray imaging. Our aim is to develop a compact, low-cost and real-time gamma-ray imaging instrument for non-destructive inspection of heavy materials like metal pipelines. For reduction of number of channels and fast readout, we adopted X- and Y-strip readout method. Obtained charges are converted to digital by 10-bit 10Msps pipelined ADCs on the chip. In order that the ghost images are reduced, which are the drawback of the X- and Y-strip method, simple time-to-digital converters with resolution of 10ns are designed on the same chip. Design with 9 channels on about 5mm x 5mm die size for the TSMC 0.25m process was submitted for fabrication and will be obtained and tested by the conference.
This study is supported by the program JST Development of Systems and Technology for Advanced Measurement and Analysis. The LSI design development has been helped by the software support of VLSI Design and Education Center (VDEC), the University of Tokyo.

130

N19-6:

A Pixel Readout Chip Designed in 90nm CMOS Process for High Count Rate Imaging Systems

R. Szczygiel, P. Grybos, P. Maj Department of Measurement and Instrumentation, AGH University of Science and Technology, Krakow, Poland We report on the design of an ASIC called PX90, dedicated for readout of hybrid pixels detectors in imaging applications. The PX90 was designed in a 90 nm technology. It contains a matrix of 40 x 32 pixels with the size of 100m x 100m. Each pixel uses fully differential readout scheme to limit the front-end sensitivity to the noise generated by the digital part. There are two discriminators followed a 16-bit counter each in every pixel. The chip can operate in the continuous readout mode and in readout mode separate from exposure. In the paper we also present the preliminary tests results of the PX90 regarding its functionality, power consumption and effective discriminator threshold spread.
N19-7:

R. Ballabriga1, M. Campbell1, E. H. M. Heijne1, J. Jakubek2, X. Llopart1, S. Pospisil2, L. Tlustos1, Z. Vykydal2 1 PH Department, CERN, Geneva, Switzerland 2 Institute of Experimental and Applied Physics, Czech Technical University, Prague, Czech Republic The Medipix3 chip is a 256x256 channel hybrid pixel detector readout chip working in Single Photon Counting Mode. It implements a new pixel architecture that mitigates the effects of charge sharing in segmented sensors allowing improved energy resolution. Summing circuits are placed at each pixel corner adding the signals of the 4 nearest neighbours. The outputs of the summing circuits are connected to discriminators and those are in turn connected to arbitration logic. The arbitration logic is used to allocate a given hit to the pixel corner with the highest local charge sum. Optionally charge summing can be switched off and each pixel acts independently like a conventional single photon counting detector. The pixel cell occupies an area of 55mx55m. It contains an analog front-end and digital processing circuitry. The analog front end consists of a preamplifier, a shaper and two threshold discriminators. The first discriminator is used to define the lower threshold and as input to the arbitration logic when charge summing is enabled. The second discriminator can be used to define a second threshold. Clusters of 4 pixels can be grouped in a single detection unit allowing up to 8 thresholds for energy binning. The digital circuitry contains control logic, arbitration modules for hit allocation, circuitry for storage of the pixel configuration data and two registers that can be configured as two 1-bit, 4-bit or 12 bit counters or as a single 24 bit counter. In 24-bit counting mode only one discrimination level is used and DAQ(write) and readout(read) are sequential. With lower counter depths it is possible to have 2 discrimination levels per pixel in sequential read/write or a single threshold per pixel in continuous read/write. The chip was designed and manufactured in an 8-metal 0.13m CMOS technology. It contains ~115M transistors. Preliminary measurements show an Equivalent Noise Charge of ~60e- rms (Single Pixel Mode) and ~120 rms (Charge Summing Mode).
This project has been undertaken in the framework of the Medipix3 Collaboration. http://www.cern.ch/MEDIPIX N20: Neutron Imaging and Detectors for Neutron Imaging Tuesday, Oct. 27 N20-1: 16:00-18:00 Grand Ballroom 6

Characterization of the Medipix3 Pixel Readout Chip

Development of a Neutron Imaging Detector Based on the -PIC Micro-Pixel Gaseous Chamber

J. D. Parker, K. Hattori, C. Ida, S. Iwaki, S. Kabuki, H. Kubo, S. Kurosawa, K. Miuchi, H. Nishimura, M. Takahashi, T. Tanimori, K. Ueno Dept. of Physics, Cosmic Ray Group, Kyoto University, Kyoto, Japan A new detector employing the micro-pixel gaseous chamber (-PIC) is currently being developed as a thermal neutron imaging detector for applications in small angle neutron scattering (SANS), neutron radiography, and radioactive material detection. Neutron detection is achieved through the 3He(n,p)3H absorption reaction by adding 3He to the usual Argon-Ethane gas mixture. The -PIC, with a pixel pitch of 400 microns, is coupled with an FPGA-based data acquisition system with a 100 MHz internal clock. This combined system has excellent spatial (< 1mm) and time (10 ns) resolutions and has been shown to be capable of handling counting rates up to ~5 MHz. These qualities make it well suited to SANS measurements at pulsed neutron sources such as the J-PARC Neutron Source. Here, we report the performance of a new dedicated -PIC based neutron imaging detector system along with the results of an upcoming test experiment at a small neutron source currently under development at Kyoto University. This new detector features a 10 x 10 x 5 cm3 active volume and operates at gas pressures up to 2 atm. With this design, we expect a neutron detection efficiency of ~30% for a gas mixture containing 30% 3He at a total pressure of 2 atm. Additionally, an improved FPGA encoder program will allow the simultaneous measurement of the track length and energy deposition, resulting in a finer position resolution and strong rejection of the gamma-ray background.

131

N20-2: Position Sensitive Detection of Fast Neutrons with High Spatial Resolution Using the Timepix Pixel Detector

J. Jakubek1, J. Uher2 1 Institute of Experimental and Applied Physics of the Czech Technical University, Prague, Czech Republic 2 CSIRO Minerals, Lucas Heights, NSW, Australia

Thanks to accessibility of DT neutron generators (14 MeV neutrons), fast neutrons are increasingly used in many fields. Fast neutrons are conventionally detected by scintillators with relatively large volume and low spatial resolution. In this contribution we present a novel detection technique based on tracking of protons recoiled by fast neutrons. The tracking is performed by the silicon pixelated detector Timepix (300 um thick silicon sensor, 256 x 256 square pixels with 55 um pitch) covered by a hydrogen rich convertor (polyethylene up to 2 mm thick). The performance of the TimePix device for detection of highly ionizing particles such as protons was already published. The technique utilizes the charge sharing effect and it is based on proper analysis of individual recorded tracks (clusters). The range of protons recoiled by 14 MeV neutron is greater than the pixel size allowing to determine not only energy and position but also the impact angle (precision is better than 2 degrees for 5 MeV protons). Having all information about the recoiled proton and knowing the original energy and direction of the neutron it is possible to reconstruct the exact position of the neutron-proton collision in the converter. The final spatial resolution of the neutron position determination reaches the subpixel level (about 20 um). Experimental results and some Monte-Carlo simulation will be presented.
N20-3:

One Meter Square High Rate Neutron Imaging Panel Based on Boron Straws

J. L. Lacy, L. Sun, C. S. Martin, A. Athanasiades, T. Lyons Proportional Technologies, Inc, Houston, TX, U. S. A. In order to fully realize the enhanced potential of powerful neutron scattering technique, higher neutron count rate capability must be reached in large size detectors. We have developed a neutron detector which has a large sensitive area (1 m2), offers 3D spatial resolution, high sensitivity and high count rate capability, and it is economical and practical to produce. The detector technology is based on 10B thin film conversion of neutrons in long straw-like gas detectors, as a novel and superior replacement of 3He detectors. Twenty two detector modules, each consisting of 50 straws (1 m long, 4 mm in diameter, natural boron carbide coated), have been constructed and mounted into a robust housing to form a square meter panel detector. Another major accomplishment achieved was the development of an enriched boron carbide coating process. One enriched detector module has been fabricated. Readout electronics including preamplifiers, signal conditioning modules and digitization modules have been designed and tested to meet the required count rate in a 1 m2 detector. The panel detector has been tested successfully by imaging objects of different materials and different shapes. The enriched module was integrated into the panel detector and the enhanced sensitivity from 10B-enriched straw material was quantitatively evaluated. The sensitivity increased by a factor of 4 in the enriched module, compared to that in natural boron module as predicted by modeling studies. The neutron imaging detector has achieved required performance for SNS applications and substantially increases performance compared with 3He while utilizing a cheap and inexhaustible conversion medium.
This project was supported by DOE Grants Nos. DE-FG02-05ER84251 and DE-FG02-08ER84997. N20-4:

V. V. Nagarkar1, V. Gaysinskiy1, Z. Bell2, M. Bliss3, S. Miller1, K. J. Riley1 1 Radiation Monitoring Devices, Inc., Watertown, MA, USA 2 Oak Ridge National Laboratory, Oak Ridge, TN, USA 3 Pacific Northwest National Laboratory, Richland, WA, USA

A Neutron Imaging Detector from Bundled Lithium Silicate Glass Fibers

Scientific applications of neutrons ranging from protein crystallography to non-destructive inspections require sensitive neutron detectors having a high spatial resolution that are able to discriminate against unwanted gamma ray events. To address this problem, 2 and 5 mm thick prototype detectors were fabricated as a bundled array of 120 m diameter cerium activated lithium silicate glass fibers cast in a low-index epoxy resin that helps improve light transmission. Performance of this detector was characterized by measurements of detection efficiency and pulse height spectra using 22Na, 133Ba and 137Cs isotopic gamma-ray sources and a moderated 10 Ci Am/Li neutron source that were compared with parameters determined from a commercially available bulk glass specimen. Imaging experiments were also performed in a polychromatic neutron beam using an electronmultiplying CCD. 1.5 MeV of electron equivalent light output during neutron capture events was measured from pulse height spectra of the bulk glass which is comparable to commercial glasses. This high light output is useful for discriminating against gamma ray events depositing less than 1.5 MeV in the scintillator. Light output for neutron events in the 2 mm thick detector was about 50% lower than the bulk sample but gamma rays were similarly affected. The 5 mm thick sample exhibited poorer light output with some 132

asymmetry in the neutron capture peak that is associated with light losses in the fiber bundle. The neutron detection efficiency of the 2 and 5 mm thick detectors was 68 and 98%, respectively. Both detectors successfully imaged patterns fashioned from cadmium neutron absorbers and indicate a spatial resolution of approximately 250 m. These data demonstrate the feasibility of this approach and further studies are underway to improve overall light output and spatial resolution.
N20-5:

S. D. Hunter1, G. A. de Nolfo1, S. Son1, J. T. Link1, N. A. Guardala2 1 NASA, Goddard Space Flight Center, Greenbelt, MD, USA 2 NSWC, Carderock, West Bethesda, MD, USA We describe the operation and performance of the Neutron Imaging Camera (NIC) being developed by NASA/GSFC and NSWC/Carderock. The NIC is based on the Three-dimensional Track Imager (3-DTI) technology developed at GSFC for gamma-ray astrophysics applications. The 3-DTI, a large volume time-projection chamber, provides accurate, ~0.4 mm resolution, 3-D tracking of charged particles. The incident direction of fast neutrons, En > 0.5 MeV, are reconstructed from the momenta and energies of the proton and triton fragments resulting from 3He(n,p)3H interactions in the 3-DTI volume. The angular resolution of the NIC from accelerator tests is presented.
This work is funded by DTRA. N20-6:

Performance of the Neutron Imaging Camera

A Liquid Scintillator Fast Neutron Double-Scatter Imager

A. L. Hutcheson, B. F. Phlips U.S. Naval Research Laboratory, Washington, DC, USA A liquid scintillator fast neutron imager based on double-scatter neutron detection is in development for the purpose of identifying and locating special nuclear materials. A prototype instrument is being tested consisting of two planes of four 1 m 5 cm 5 cm liquid scintillator detectors each viewed at either end by a photomultiplier tube. The instrument is capable of reconstructing energy and location of a source of fast neutrons via time-offlight and neutron double-scatter techniques. Pulse shape discrimination is utilized to separate neutron signals from gamma-ray background. Initial results and analysis for the instrument prototype will be presented.
N20-7:

N. O. Boyce1, B. R. Kowash1, D. Wehe2 1 Dept. of Engineering Physics, Air Force Institute of Technology, Dayton, OH, USA 2 Dept. of Nuclear Engineering and Radiological Sciences, The University of Michigan, Ann Arbor, MI, USA A neutron Rotating Modulation Collimator (RMC) is desirable because it provides a secondary indicator of a source, especially on heavily shielded devices since neutrons are difficult to shield. RMCs are a class of mechanically collimated, imaging systems that produce a modulated signal, similar to the illustration in Figure 1, that is directly related to the location of radioactive sources in the field of view [1,3]. The primary strength of such systems is that they provide an imaging capability using a single nonposition sensitive detector. Previous work has focused on mask and detector designs for modulating gamma ray sources (high Z materials), but it is apparent that the same modulation principles can be applied to neutron imaging by substituting the masks and detector with materials that favor neutron scattering and/or absorption (low Z materials). This research demonstrates that an RMC can be used to image thermal and epithermal neutron source distributions by employing many of the principles that are currently used in gamma ray imaging. Prior to executing the experiment, a series of Monte Carlo simulations were run to calibrate intuition about the correlation between the gamma ray imaging capabilities of the RMC and its neutron imaging potential. The results of this experiment demonstrate that the RMC method of imaging is suitable for mono-energetic neutrons below 24 keV and that the system performance is comparable to that for gamma ray imaging with the RMC (< 1 spatial resolution). Furthermore, the results demonstrate that an organic liquid scintillator can be used with the RMC, which provides a basis for future work with fast and fission spectrum neutron sources. Finally, these low energy results are combined with the Monte Carlo results of high energy mono-energetic and fission spectrum neutrons to discuss the application of these systems in more realistic imaging scenarios.
References 1. Kowash, B.R., Wehe, D.W, Fessler, J.A. "A Rotating Modulation Imager for Locating Mid-Range Point Sources". NIM - A. pp. 477-483. 21 April 2009. 2. Kowash, B.R., A Rotating Modulation Imager for the Orphan Source Search Problem, Ph.D. Dissertation, University of Michigan. September 2008. 3. Hurford, G.J., Schmahl, E.J., Schwartz, R.A., Conway, A.J., Aschwandan, M.J., Csillaghy, A., Dennis, B.R., Johns-Krull, C., Krucker, S., Lin, R.P., McTiernan, J., Metcalf, T.R., Sato, J., Smith, D.M. The RHESSI Imaging Concept, Kluwer Academic Publishers. 2002.

Thermal Neutron Imaging with a Rotationally Modulated Collimator

133

N21: Trigger and Front-End Systems II Wednesday, Oct. 28 N21-1: 08:00-10:00 Grand Ballroom 1

SiGe Front-End Prototype for the Upgraded ATLAS LAr Calorimeter

S. Rescia Instrumentation Div., Brookhaven National Laboratory, Upton, NY, USA On behalf of the The ATLAS Liquid Argon Calorimeter Group We have designed and fabricated a very low noise preamplifier and shaper to replace the existing ATLAS Liquid Argon (LAr) readout for use at Large Hadron Collider upgrade (sLHC). IBMs 8WL 130nm SiGe process was chosen for its radiation tolerance, low noise bipolar NPN devices, wide voltage range and potential for use in other LHC detector subsystems. The required dynamic range is accomplished with a single stage, low noise, wide dynamic range preamplifier followed by a dual range shaper. Preliminary measurement of the fabricated circuits indicates their performance is consistent with the design specifications.
N21-2:

The TDC Based Integrated Trigger System of the NA62 Experiment at CERN

R. Fantechi INFN - Sezione di Pisa, Pisa, Italy On behalf of the NA62 Collaboration The TDC based integrated trigger and data acquisition system of the NA62 experiment at CERN will be presented. The system architecture, the trigger algorithm and its implementation in commercial high performance FPGAs will be described. The results of test and characterization of the custom components as well as those of extensive field tests performed on a system prototype handling up to 512 input channels will be discussed.
N21-3:

S. Amerio1, A. Annovi2, M. Basile3, M. Bettini4, M. Bucciantonio3, P. Catastini5, J. Cenni3, F. Crescioli3, M. Dell'Orso3, P. Giannetti6, E. Giuliani3, D. Lucchesi1, M. Nicoletto4, M. Piendibene3, N. Rafanelli3, G. Volpi3 1 University of Padova & INFN, Padova, Italy 2 INFN Laboratori Nazionali di Frascati, Frascati, Italy 3 University of Pisa & INFN, Pisa, Italy 4 INFN Padova, Padova, Italy 5 University of Siena & INFN, Siena, Italy 6 INFN Pisa, Pisa, Italy

The GigaFitter: a Next Generation Track Fitter to Enhance Online Tracking Performances at CDF

The Silicon-Vertex-Trigger (SVT) is a processor developed at CDF experiment to perform online fast and precise track reconstruction. SVT is made of two pipelined processors: the Associative Memory finds low precision tracks looking for coincidences between hits from the tracking detectors and track candidates (roads) stored in memory; the Track Fitter refines the track quality whith high precision fits. The GigaFitter is a next generation track fitter, developed to reduce the degradation of the SVT efficiency due to the increasing instantaneous luminosity. It reduces the track parameter reconstruction to a few clock cycles and can perform many fits in parallel, thus allowing high resolution tracking at very high rate. The core of the GigaFitter is implemented in a modern Xilinx Virtex-5 FPGA chip, rich of powerful DSP arrays. With respect to the current Track Fitter, the GigaFitter is faster and provided of much more memory to store a greater number of roads and constant sets to be used in the fit; this results in an increased SVT efficiency as more track candidates can be reconstructed. The GigaFitter has been installed in parasitic mode at CDF and has been tested against the current Track Fitter in terms of speed and track reconstruction efficiency. We will describe the GigaFitter architecture, the parasitic installation at CDF and the performances with respect to the current system.
N21-4:

Fast Control and Timing Distribution based on FPGA-Embedded Serial Transceivers

A. Aloisio, R. Giordano, V. Izzo Universitdi Napoli Federico II and INFN, Napoli, Italy The Fast Control and Timing distribution System (FCTS) is a key component of the electronics of every High Energy Physics experiment. It must distribute the clock to front-end boards with minimum jitter in order to reduce the uncertainty of sampling points and it must transfer data with a fixed latency. In fact, data transferred include trigger signals (accept/reject and qualifiers) and fast control commands, whose timing must be preserved. So far, implementations based on Application Specific Integrated Circuits have enabled designers to address all the described issues. The drawback of such an approach is the higher design effort 134

and the lack of flexibility of the system, which requires a re-design of the ASICs and of the boards in case of an upgrade. Latest Field Programmable Gate Arrays (FPGAs) include high-speed Serializers-Deserializers (SerDes). In this work, we present an architecture fully based on FPGA-embedded SerDes. Data and clock are recovered with a fixed latency even after a power cycle or a loss lock, in such a way to realize a virtual ribbon cable between the two ends of the link. We implemented our architecture with a Xilinx GTP Transceiver embedded in Virtex 5 FPGAs. Data is transferred at 2.4 Gb/s and the distributed clock runs at 60 MHz. Our link supports both the 8b10b encoding, a wide-spread standard, and the scrambling method adopted by the GigaBit Transceiver project under development at CERN. We present our test results in terms of jitter performance on the recovered clock. We performed measurements both in the time and in the frequency domain (i.e. phase noise spectrum measurements). We also present and discuss a version of our architecture including an external jitter cleaner to reduce the phase noise on the recovered clock.
N21-5:

The Calorimeter ReadOut Card (CROC) a System of LHCb Experiment

O. Duarte CNRS / IN2P3 Laboratoire de l'accelerateur Lineaire, Orsay, France On behalf of the LHCb collaboration LHCb, one of the four particle physics experiments at the Large Hadron Collider at CERN will perform studies of CP-symmetry violation and rare decays of B-hadrons. The main purpose of the calorimeter system is the selection and identification of hadrons, electrons and photons, and the measurement of their energies and directions. The calorimeter front-end crate is a standard 9U VME. In each crate, 16 front-end boards and a single CROC board are connected. The frontend boards perform the acquisition of the detector channels. The CROC is the board which gathers the front-end data of a calorimeter crate and sends them through optical fibres to the LHCb Data Acquisition. The CROC also provides the clock, trigger signal, broadcast and slow control commands for all the boards in a crate. The CROC receives the event bits incoming from the FEBs serialized at 7x40MHz, the data flow reaching up to 13.5 Gbit/s per crate. It supports 5 FPGA, called Fe-PGAs (4 copies on the board) and Spy-PGA. The formers receive the data from 4 FEBs each. Then, they perform the header detection of the incoming bits, add the header, the control word and finally send them to the GOL optical emitters. The latter PGA is the conductor of the board and performs the actions which are not specific to a data channel. The clock, trigger and channel B signals are provided to the CROC by a TTCrq mezzanine board. The Experiment Control System (ECS) is done by a SPECS (Serial Protocol for the Experiment Control System) mezzanine. The CROC is located in a region where the radiation dose will reach 200 rad per year. This requested to perform irradiation tests in proton and heavy ion beams. All components of the CROC are radiation tolerant. The FPGA are ACTEL flash technology components and cope the radiation level expected. A mitigation technique (triple voting) has been implemented in the firmware of the FPGAs to prevent the effects of single event upsets.
N21-6:

All Data-Out Readout Electronics for the ATLAS LAr Calorimeter at Super-LHC

H. Ma Physics Department, Brookhaven National Laboratory, Upton, NY, USA On behalf of the ATLAS Liquid Argon Calorimeter Group The ATLAS Liquid Argon (LAr) calorimeters have a total of 182,486 signals to be digitized and processed at every triggered event, to provide time and energy in each detector element. The current readout electronics is not designed to sustain the ten times higher radiation levels expected at super-LHC in the years ~ 2017, and will be replaced by new electronics with a completely different readout scheme. The future on-detector electronics will send out all data continuously at each bunch crossing (~ 100 Gbps for each board) through multiple high speed radiation resistant optical links, to the off-detector processing units, which will not only process the data in real-time but will also implement trigger algorithms. This talk will review the various components necessary to develop such a complex system and will describe the R&D activities and architectural studies undertaken by the ATLAS LAr Calorimeter group. The talk will describe the on-going design of the mixed-signal front-end ASICs, the SOS based optical-link, the high-speed off-detector FPGA-based processing units and the power supply distribution scheme.
N21-7:

An on-Chip Fast Readout Sparsification for a 256-Pixel 3D Device

A. Gabrielli Physics department University of Bologna, INFN, Bologna, Italy On behalf of the VIPIX Collaboration A prototype of an ASIC built up of a fast readout architecture that interfaces with a matrix of 256 pixel sensor, via Chartered 130 nm CMOS technology, was recently submitted. The readout logic shares one layer of double-layer design that will be stacked at the end of the fabrication process. The two layers form the matrix of 256 pixels that, along with parallel sparsification readout logic, build a 3-dimensional stacked device. Square groups of 16 pixels form a macro-pixel (MP). Each MP can be latched via 135

single pixels, 40 by 40 um each, and an 8-bit time-stamp is associated with the frozen condition. This fast architecture could overcome the well-known readout speed limit of big matrices. As the output port can only accept one-hit information at a time, an internal queuing system has been provided to stand local high hit-rates. The ASIC can be connected to an actual full-custom matrix of 3D sensors or to a digital matrix emulator composed of std-cells (flip-flops), for testing facilities. For both operating modes a slow-control phase is required to load the chip configuration via mask bits to select which MPs are to be read and which are not, for example in case they are too noisy or burned up. Previous versions of similar ASICs were designed and properly tested. The work is aimed at improving the design of a pixel detector with an on-chip fast sparsification system, for particle tracking, to match the requirements of vertex tracking system of future high-energy physics experiments. In fact, the implemented readout architecture can be data driven extending the flexibility of the system to applications in first level triggers on tracks in vertex detectors. Simulations and tests on previous prototypes designed and fabricated via other foundries, still using the same 130 nm technology node, prove that the readout system can easily cope with an average hit-rate up to 100 MHz/cm2 if a master clock of 40 MHz is used, while maintaining an overall efficiency over 99%.
N22: Semiconductor Detectors II: Silicon Devices Wednesday, Oct. 28 N22-1: 08:00-10:00 Grand Ballroom 2

J. Ninkovic1, L. Andricek1, G. Liemann1, G. Lutz2, H. G. Moser1, R. H. Richter1 1 Semiconductor Laboratory, Max Plancl Institute for Physics, Munich, Germany 2 PN Sensor GmbH, Munich, Germany

The First Measurements on an Avalanche Diode Array with Bulk Integrated Quench Resistors for Single Photon Detection

A Silicon Photomultiplier (SiPM) is an avalanche photodetector that is entering many application areas as a replacement of conventional photomultiplier tubes (PMTs). Its Geiger mode operation requires high ohmic polysilicon as quench resistor that becomes an obstacle for light and is one of the most cost and yield driving technological issues. We have proposed a new detector concept which has the quench resistor integrated into the silicon bulk avoiding polysilicon resistors. Extensive simulation results showed the feasibility of the concept. The quenching mechanism has been demonstrated in a proof of principle production performed in house. The first prototype fabrication (second production run) on silicon on isolator substrates has been done and allows testing of the device performance. The results from the first measurements will be evaluated in comparison with the simulations. Based on these results the inherent advantages and drawbacks compared to standard SiPMs will be discussed.
N22-2:

Y. Arai1, T. Miyoshi1, Y. Unno1, T. Tsuboyama1, S. Terada1, Y. Ikegami1, T. Kohriki1, K. Tauchi1, Y. Ikemoto1, R. Ichimiya1, Y. Fujita1, H. Ikeda2, K. Hara3, H. Miyake3, M. Kochiyama3, T. Sega3, K. Hanagaki4 1 Institute of Particle and Nuclear Studies, KEK, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, JAPAN 2 Institute of Space and Astronautical Science, JAXA, Sagamihara, Kanagawa, JAPAN 3 Institute of Pure and Applied Sciences, Univ. of Tsukuba, Tsukuba, Ibaraki, JAPAN 4 Department of Physics, Osaka Univ., 1-1 Machikaneyama, Toyonaka, Osaka, JAPAN We have been developing a monolithic pixel detector with 0.2 m Silicon-on-Insulator (SOI) CMOS technology. It has both a thick, high-resistivity sensor layer and thin LSI circuit layer on a single chip. This structure is ideal for a monolithic pixel detector, but it also reveals interference between the sensor and the electronics. To circumvent this, we modified the process in which an additional p well (buried p-well) is created. In addition, we are also trying vertical integration of two SOI chips by using a -bump technique. This increases integration density and separates the sensor and the electronics part.
N22-3:

New Techniques in SOI Pixel Detector

Sensor/ROIC Integration Using Oxide Bonding

Z. Ye Fermi National Accelerator Laboratory, Batavia, IL, USA On behalf of the Fermilab Pixel R@D Group We explore the Ziptronix Direct Bond Interconnect (DBI) technology for the integration of pixel sensors and readout integrated circuits (ROICs) for high energy physics. The technology utilizes an oxide bond to form a robust mechanical connection between layers which serves to assist with the formation of metallic interlayer connections. The technology can provide low mass interconnects with pitch as small as 3 microns. We report on both bench and test beam testing results of sample sensors bonded to ROICs and thinned to 100 microns.

136

N22-4:

Backside Passivation of CMOS Sensor Study for Low Energy Charge Particle Tracking

R. Cluzel, R. Barbier CNRS/IN2P3 Institut de Physique Nucleaire de Lyon - Universite Lyon 1, Lyon, France Monolithic Active Pixel Sensor is promising for silicon detectors in order to track charged particle: it consists in a sensitive and not biased low doped epitaxial layer on a highly doped substrate with a high granularity, and a readout signal processing integrated in the same chip. Some applications require backside detection of low energy electrons below 10 keV. In this specific application the substrate is totally removed and a passivation process is performed. The passivation process of the epilayer has to replace the highly p-doped substrate in his mirroring effect of the charge carriers to avoid the surface recombinations and to maximise Charge Collection Efficiency (CCE). At low energy operations the passivation has to be as shallow as possible with the highest doping and activated boron concentration. The passivation techniques consist in an ultra shallow layer achieved by boron implantation and laser annealing. Secondary Ion Mass Spectroscopy (SIMS) analysis shows that the P++ doped layer can be divided in two area: the first 25 nm exhibits an increasing doping concentration where the induced electric field Ep1 makes drift the generated electrons toward the recombining surface. As these electrons cannot be collected by the diodes, this area is called dead zone. The second following area (from 25 nm up to 70 nm) shows a decreasing doping concentration, where Ep2 tends to push back the electrons inside the epilayer where the generated electrons can diffuse towards the collection diode. To characterize the sensitivity of the backside thinned CMOS an experiment has been set up to estimate the electron multiplication gain and the diode CCE. The curve of gain vs High Voltage exhibits a good linearity above 4 kV which shows that the electrons lost in dead zone are constant. The talk will further develops the analysis in terms of CCE and gain, thanks to Monte Carlo simulations coupled with Silvaco TCAD software.
N22-5: Thin, Fully Depleted Monolithic Active Pixel Sensor with Binary Readout based on 3D Integration of Heterogeneous CMOS Layers

W. Dulinski1, G. Bertolone1, C. Colledani1, Y. Degerli2, A. Dorokhov1, C. Hu1, F. Morel1, L. Ratti3, V. Re4, X. Wei1, M. Winter1 1 IPHC/IN2P3/CNRS, Strasbourg, France 2 CEA/IRFU/SEDI, Gif-sur-Yvette, France 3 University of Pavia, Pavia, Italy 4 University of Bergamo, Bergamo, Italy

On the way towards fast, radiation tolerant and ultra thin CMOS sensors, we propose new generation of devices based on recent progress and commercial availability of vertical integration of several CMOS wafers (3D Electronics). In this process, each wafer is thinned down to less than 10 microns end equipped with through-silicon vias (TSV) allowing for electrical interconnection between wafers at very small pitch (few microns) and with minimum material budget. The proposed prototype device is a 245x245 pixel array with a pitch of 20 microns, providing active area of 5x5 mm2. In the first silicon layer charge sensing diode and first stage buffer amplifier (source follower) are integrated, using 0.6 micron CMOS process on high resistivity epitaxial wafers. Pioneering application of such substrate (fully depleted at less than 5 V) for particle tracking sensors results in higher signal seen at the seed pixel, reduces strongly cluster multiplicity and in particular, due to fast charge collection, improves radiation hardness with respect to non-depleted CMOS MAPS by at least order of magnitude. Outputs of buffer voltage amplifiers are vertically coupled (through poly-poly capacitor) to the following stage of processing electronics (charge integration, time continuous shaping and signal discrimination), placed in the second silicon layer (0.13 micron CMOS). Expected equivalent noise charge (ENC) is less than 15 electrons, for a pulse peaking time of about 500 ns and total power dissipation of ~5microW/pixel. In connection with fully depleted, 14 micron thick epitaxial substrate as a charge sensing layer, this brings very comfortable signal-to-noise ratio of more than 40 for detection of minimum ionizing particles. The third silicon layer (also 0.13 micron CMOS) is used for implementation of digital (binary) readout, with a fast, data driven, self-triggering data flow.
N22-6:

J. Baudot1, G. Bertolone1, A. Brogna1, G. Claus1, C. Colledani1, Y. Degerli2, R. De Masi1, A. Dorokhov1, G. Doziere1, W. Dulinski1, M. Gelin2, M. Goffe1, A. Himmi1, C. Hu1, K. Jaaskelainen1, M. Koziel1, F. Morel1, F. Orsini1, M. Specht1, I. Valin1, G. Voutsinas1, M. Winter1 1 IPHC, IN2P3/CNRS - Universite de Strasbourg, Strasbourg, France 2 SEDI/DSM, IRFU - CEA Saclay, Gif Sur Yvette, France

First Test Results of MIMOSA-26, a Fast CMOS Sensor With Integrated Zero Suppression and Digitized Output

Numerous MIMOSA pixel sensors developed in Strasbourg have demonstrated attractive features for the detection of charged particles in high energy physics: a thin sensitive volume which allows for a low material budget, a high granularity resulting in spatial resolutions from about 1 to 3 micrometers depending on the pixel pitch and tolerance to radiations in the range of 10 kGy and 10^12 neq/cm2 equivalent fluence. These results are, so far, mainly obtained with analog readout, which limits the output 137

rate to about 1000 frames/second for a full-size sensor. The new MIMOSA 26 sensor improves mainly the readout speed and the data throughput. It incorporates pixel output discrimination for binary readout and a zero suppression micro-circuit at the sensor periphery to stream only fired pixel out. The MIMOSA 26 chip is 2.2 cm2 wide and composed of about 0.7 millions of pixels with a 18.4 micrometers pitch. The readout frequency reaches close to 10,000 frames/second and the single point resolution approaches 3.5 micrometers. The sensor is back from foundry since march 2009 and is being characterized in laboratory MIMOSA 26 will equip the final version of the EUDET beam telescope and first beam tests will occur in summer 2009 at CERN-SPS. This type of sensor prefigures the architecture of MAPS for coming vertex detectors (STAR, CBM and ILC experiments). To address integration issues, a prototype ladder equipped, on both sides, with a row of 6 MIMOSA 26-like sensors is under study, aiming for a total material budget of about below 0.2 % or less. This talk will first review the main design features of the MIMOSA 26 sensor, regarding the applications it is aimed for. We will also remind the test results of previous circuits which provide its building blocks. Then, the first results of the in-beam characterization of MIMOSA-26 will be presented. Finally, an overview of its extension achieving faster readout (a few tens of micro-seconds or below), will be provided.
N22-7:

M. Battaglia1,2, D. Bisello3,4, D. Contarato2, P. Denes2, P. Giubilato2,3,4, R. M. Yee2, S. Mattiazzo3,4, D. Pantano3,4, N. Pozzobon3,4, C. Q. Vu2 1 Dept. of Physics, University of California at Berkeley, Berkeley, CA, USA 2 Physics Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 3 Dept. of Physics, University of Padova, Padova, Italy 4 INFN Sezione di Padova, Padova, Italy Monolithic pixel sensors for charged particle detection and imaging applications have been designed and fabricated using commercially available, deep-submicron Silicon-On-Insulator (SOI) processes, which insulate a thin layer of integrated CMOS electronics from a high-resistivity substrate by means of a buried oxide. The substrate is contacted from the electronics layer through vias etched in the buried oxide, allowing pixel implanting and reverse biasing. This paper presents the design and test results of an advanced prototype manufactured in a 0.20 micrometer SOI process, implementing both analog and binary pixel architectures and optimized for fast readout up to 50 MHz clock frequency. The binary pixels feature a novel architecture with inpixel Correlated Double-Sampling (CDS) and a clocked comparator, and multiple outputs for high frame rate. Tests of analog and digital pixels, performed with infrared lasers, radioactive sources and a high momentum electron beam will be presented.
N23: Computing and Software for Experiments III: High Energy Physics Computing Wednesday, Oct. 28 N23-1: 08:00-10:00 Grand Ballroom 3

Fast Analog and Binary Monolithic Pixels in Deep-Submicron SOI Technology

A. Fella1, E. Luppi2, L. Tomassetti2 1 INFN - CNAF, Boogna, Italy 2 Department of Physics, Universita' di Ferrara and INFN, Ferrara, Italy

Study of a Distributed Computing Model for the SuperB Experiment

SuperB is an international project aiming at the construction of a very high luminosity asymmetric e+e- flavour factory. SuperB is focused on new physics and flavour physics studies. Since the first steps of detector design, the load in computing issues in terms of Monte Carlo alghorithm execution requires the exploitation of resources in a full distributed model. The experiment choice in such a theme is to build an as possible Grid integrated system identifying the consolidated and more important, the upcoming solutions to accomplish the typical HEP experiment requirements. The in progress design and setup of SuperB computing model is presented including the complete discussion of Grid services selection.
N23-2:

Commissioning of the ATLAS Reconstruction Software with First Data

W. Lampl University of Arizona, Tucson, AZ, USA On behalf of the ATLAS Collaboration Looking towards first LHC collisions, the ATLAS detector is being commissioned using all types of physics data available: cosmic rays and events produced during a few days of LHC single beam operations. In addition to putting in place the trigger and data acquisition chains, commissioning of the full software chain is a main goal. This is interesting not only to ensure that the reconstruction, and monitoring chains are ready to deal with LHC physics data, but also to understand the detector performance in view of achieving the physics requirements. The recorded data have allowed us to study the ATLAS detector in terms of 138

efficiencies, resolutions, channel integrity, alignment and calibrations. They have also allowed us to test and optimize the subsystems reconstruction as well as some combined algorithms, such as combined tracking tools and different muon identification algorithms. The status of the integration of the complete software chain will be presented as well as the data analysis results.
N23-3:

V. Rapsevicius1, A. Meyer2, I. Segoni2 1 Physics, University of Florida, Gainesville, FL, USA 2 CMS, CERN, Geneva, Switzerland

Data Certification Workflow and Tools in CMS Data Quality Monitoring

The data quality monitoring and certification system of the CMS experiment at the Large Hadron Collider is presented. The tools, workflows and procedures for the assessment of the data quality range from prompt online visual inspection of the data during the data taking to the final certification of the data for physics analysis. Both automatic data quality and certification algorithms as well as manual logging and bookkeeping of relevant information are organized through the Run Registry which consists of a database with GUI front end. The Run Registry is used to track the monitoring and certification procedure for each run and to publish the results to the collaboration. Quality and certification information is entered sequentially by online and offline shift persons; in weekly sign-off meetings the correctness of the information is confirmed by subsystem experts; finally, the Run Registry is used to make the certification decisions available for physics analysis. The key workflow components are described and the main results and lessons learnt during the CMS data quality assessment of commissioning data are discussed.
N23-4:

Alignment of the ATLAS Inner Detector Tracking System

G. Cortiana Max-Planck-Institut fuer Physik, Munich, Germany On behalf of the ATLAS Collaboration The Large Hadron Collider (LHC) at CERN is the world largest particle accelerator. It will collide two proton beams at an unprecedented center of mass energy of 14 TeV. ATLAS is equipped with a charge particle tracking system built on two technologies: silicon and drift tube based detectors, composing the ATLAS Inner Detector (ID). The alignment of the tracking system poses a challenge as one should solve a linear equation with almost 36000 degrees of freedom. The required precision for the alignment of the most sensitive coordinates of the silicon sensors is just a few microns. This limit comes from the requirement that the misalignment should not worsen the resolution of the track parameter measurements by more than 20%. So far the proposed alignment algorithms are exercised on several applications. We will present the outline of the alignment approach and results using real data from cosmic rays and large scale computing simulation of physics samples mimicking the ATLAS operation during real data taking. The full alignment chain is tested using that stream and alignment constants are produced and validated within 24 hours. Cosmic ray data serves to produce an early alignment of the real ATLAS Inner Detector even before the LHC start up. The impact of the alignment on physics measurements will be discussed.
N23-5:

First Alignment of the Complete CMS Silicon Tracker

M. Weber I. Physikalisches Institut B, RWTH Aachen University, 52056 Aachen, Germany On behalf of the CMS Collaboration We present the first results of the full CMS Silicon Tracker alignment based on several millions reconstructed tracks from the cosmic data taken during the commisioning runs with the detector in its final position. The all-silicon design of the CMS Tracker poses new challenges in aligning a complex system with 15148 silicon strip and 1440 silicon pixel modules. For optimal trackparameter resolution, the position and orientation of its modules need to be determined with a precision of several micrometers. For the modules well illuminated by cosmic ray particles, the ultimate precision has been achieved with data from the silicon modules traversed in-situ by charged muons used in combination with survey measurements. The achieved resolution in all five track parameters is controlled with data-driven validation of the track parameter measurements near the interaction region, and tested against prediction with detailed detector simulation. Outlook for expected tracking performance with the first collisions is given.

139

N23-6:

The Challenge of Mass Storage System for the Experiments at LHC: the Solution Developed at INFNCNAF

L. dell'Agnello, A. Cavalli, E. Ronchieri, B. Martelli, P. Ricci, V. Sapunenko, S. dalPra, A. Prosperini, D. Gregori CNAF, INFN, Bologna, Italy The mass storage challenge for the experiments at LHC remains a critical issue for most of the sites even when LHC is almost ready to start and the real data to flow to the Tier-1s. In fact, the requirements for the Mass Storage Systems (including transparent access to the tapes) are quite high, amounting to several PB of data that should be available for near-line access at any time. Besides the solutions already widely developed and adopted by the High Energy Physics community (e.g. CASTOR and dCache), INFN-CNAF has developed a Mass Storage System based on industrial standards (such as GPFS and TSM by IBM) able to deliver with high reliability and to satisfy the required performances. This system has been achieved by the integration of StoRM as SRM layer, GPFS as disk file system and TSM for data movement to tape; it is already in production at the INFN Tier1 since several years. In particular, since May 2008 also the so-called Storage Class T1D1 (requiring the files to be also copied to tape) is in production, while we have recently proved the feasibility of a pure Hierarchical Storage Manager system (implementing the so-called T1D0 Storage Class). In this paper we report the technical details about this system as well the results of the stress tests to verify the scalability and performance limits of the most critical part (the T1D0 storage class).
N23-7:

Statistical Software in ROOT

L. Moneta, D. Gonzalez Maline PH/SFT, CERN, Geneva, Switzerland ROOT, a data analysis framework, provides advanced mathematical and statistical methods needed by the LHC experiments for analyzing their data. Statistical tools for regression analysis, for density estimations, for establishing signal s ignificances and confidence levels and for classification using multi-variate analysis techniques are currently present in ROOT. A recent effort has been put in re-organizing the classes providing these tools in a new set of coherent and modular statistical libraries. A coherent design of the libraries allows extensions and easy integration of new statistical methods developed by the physics or statistical community. Emphasis is put on the quality and the performance of the statistical methods but also on the easiness of use in order to allow the comparison of similar methods without difficulty. These new developments are performed in close collaboration with the LHC experiments with the final goal to provide common standard implementations of statistical methods required for the analysis of the LHC data. We present an overview of these statistical methods and their recent developments introduced in the latest ROOT software release. We will show as well examples and some results of the validation and performance tests of these new statistical classes.
N24: New Detector Concepts and Instrumentation II Wednesday, Oct. 28 N24-1: 08:00-10:00 Grand Ballroom 7

Low-Noise Low-Mass Front-End Electronics for Low-Background Experiments Using Germanium Detectors

P. N. Luke, C. S. Tindall, K. Vetter, Y.-D. Chan Lawrence Berkeley National Laboratory, Berkeley, CA, USA A low-noise resistive-feedback front-end electronics assembly has been developed for use with p-type point contact (PPC) Ge detectors in neutrinoless double-beta decay and other low background experiments. The front end was designed to have a low mass and potentially low radioactivity. It is fabricated on a fused silica substrate, and consists of a low-noise FET, a feedback resistor formed using amorphous-Ge thin film, and feedback capacitor utilizing the stray capacitance between circuit traces. The fused silica substrate provides the appropriate thermal impedance to allow the FET to operate at the optimal temperature from self-heating when one side of the substrate is held at liquid nitrogen temperature. With no added capacitance at the input, a noise level of 90 eV FWHM at ~20 microsecond peaking time was achieved. With a 2 pF capacitor added to the input to simulate the expected capacitance when connected to a PPC detector, a minimum noise level of 99 eV FWHM was obtained.

140

N24-2:

T. Kin1, K. Furutaka1, S. Goko1, H. Harada1, M. Igashira2, T. Kamiyama3, T. Katabuchi2, A. Kimura1, K. Kino3, F. Kitatani1, Y. Kiyanagi3, M. Koizumi1, M. Mizumoto2, S. Nakamura1, M. Ohta1, M. Oshima1, Y. Toh1 1 Japan Atomic Energy Agency, Tokai-mura, Naka-gun, Ibaraki-ken, 319-1195, Japan 2 Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo 152-8550, Japan 3 Hokkaido University, N13, W8, Kita-ku, Sapporo 060-8628, Japan For development of advanced reactors such as fast breeder reactors or accelerator driven systems, nuclear data of minor actinides (MAs) and long-lived fission products (LLFPs) are important for calculation of burning rates of fuels and transmutation rates to reduce the radio-toxicity of wastes. However, these data are insufficient at present in quantity and quality. To overcome/improve such a situation, we developed a 4 germanium spectrometer to measure neutron capture cross sections of MAs and LLFPs by time-of-flight measurements using pulsed neutron beams. The spectrometer has been extended from the one developed for experiments at Kyoto University. The spectrometer has been installed on a neutron beam line No.4 of the materials and life science experimental facility (MLF) in J-PARC. The distance between neutron source and the sample position is 21.5 meters. The spectrometer consists of two cluster and eight coaxial germanium detectors. Each germanium detector is surrounded by BGO Compton suppressors. Collimators made of lead are located between the sample position and each germanium detector for shielding of BGO Compton suppressors and reduction of cross-talk events. All holes of the collimators are filled with conically-shaped 6LiH powder to prevent neutrons scattered by a sample and helium gas in beam guide from hitting the germanium crystals. This spectrometer can be surrounded with a shield. At the conference, we will report details of the arrangement and characteristics of the spectrometer, such as detection efficiency, energy resolution and signal-to-background ratio, using results of measurement of standard sources. We will also show example of TOF spectra to report time resolution.
This study was the result of Study on nuclear data by using a high intensity pulsed neutron source for advanced nuclear system entrusted to National University Corporation Hokkaido University by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT). N24-3:

Development of a 4 Germanium Spectrometer for Nuclear Data Measurements at J-PARC

Segmented P-Type Point Contact Germanium Detector

M. Amman, P. N. Luke, Y. D. Chan, K. T. Lesko Lawrence Berkeley National Laboratory, Berkeley, CA, USA Large-volume point-contact detectors made from high-purity p-type Ge have recently emerged as promising technologies for many scientific and technical applications. This detector configuration offers the advantages of high efficiency, excellent energy resolution, low energy threshold, and the ability to distinguish single site from multi-site interaction events. We have recently produced a segmented p-type point contact (SPPC) detector that further provides the ability to deduce the location of the energy deposition sites. This added information naturally enhances the ability to discriminate between single site and multi-site events. In this paper we provide the details on the SPPC detector and its fabrication. Measurements obtained with the detector demonstrating its performance and ability to spatially locate events are also presented.
N24-4:

A. Gabrielli1, D. Demarchi2, E. G. Villani3 1 Physics Department, INFN & University of Bologna, Bologna, Italy 2 Chilab Laboratory, Politecnico of Turin, Turin, Italy 3 Science & Technology Facility Council, Rutherford Appleton Laboratory, Didcot, UK The stimulated ignition of latchup effects caused by external radiation has so far proven to be a hidden hazard. Here this effect is described as a novel approach to detect particles by means of a solid-state device susceptible to latchup effects. In addition, the device can also be used as a circuit for reading sensors' signals, leaving the capability of sensing to external sensors. The paper first describes the state-of-the-art of the project and its development over the latest years, then the present and future studies are proposed. An elementary cell composed of two MOS transistors connected in a thyristor topology is shown. The study started using traditional bipolar transistors since the latchup effect is originated as a parasitic circuit composed of such devices. Then, an equivalent circuit built up of MOS transistors is exploited, resulting an even more promising and challenging configuration than that obtained via bipolar transistors. As the MOS transistors are widely used at present in microelectronics devices and sensors, this latchup-based cell is proposed as a novel structure for future applications in particle detection, amplification of sensors' signals and radiation monitoring. This study indicates that this very simple circuit can operate like the more complicated data acquisition chains used today in modern pixel detectors. In principle, an integrated device designed via modern CMOS technologies may work as particle detector and as a readout circuit for general sensors' signals, provided the increased performance. The circuit has a noise spread of the threshold smaller than 1 mV, power consumption due to leakage currents of 141

Exploiting a Latchup Circuit via Commercial CMOS Technologies

the order of 1 W, estimated charge sensitivity of the order of 1 pC and very good repeatability. Possible applications range from heavy-ion discriminator to beam monitor provided they deposit an over-threshold charge in the cell.
N24-5:

T. Lauf1, F. Aschauer1, S. Herrmann1, M. Hilchenbach2, M. Krumrey3, P. Lechner4, G. Lutz4, P. Majewski4, M. Porro1, R. H. Richter5, F. Scholze3, L. Strueder1, J. Treis1, G. de Vita1 1 MPI Halbleiterlabor, Max-Planck-Institut fuer Extraterrestrische Physik, Munich, Germany 2 Max-Planck-Institut fuer Sonnensystemforschung, Katlenburg-Lindau, Germany 3 Physikalisch-Technische Bundesanstalt, Berlin, Germany 4 MPI Halbleiterlabor, PNSensor GmbH, Munich, Germany 5 MPI Halbleiterlabor, Max-Planck-Institut fuer Physik, Munich, Germany DePFET macropixels are the combination of a silicon drift detector (SDD) and the integrated readout device DePFET. The use of the SDD allows arbitrary pixel sizes with lateral lengths from 100m to 1cm, while the central DePFET adds the ability for charge storage, low noise readout on demand, fast column-parallel readout and random accessible pixels. Macropixels are thus suited for X-ray optics where the spatial resolution is in the area of square millimetres but high energy resolution is required. Also readout time and power dissipation are reduced because of a lower pixel count per detector area. Using sidewards depletion to create a fully depleted bulk, radiation can enter the detector through the homogeneous entrace window on the backside resulting in a high peak to background ratio. In a DePFET detector each single pixel has its own individual first-stage amplifier. To study the homogenity of the gain over the matrix as well as its linearity over the energy range a prototype was investigated at energies from 100 eV to 10 keV the synchrotron facility BESSY. Also the charge splitting between single pixels, the homogenity of charge collection over the pixel area and the quantum efficiency of the detector were examined.
N24-6:

Performance and Spectroscopic Behaviour of DePFET Macropixels

Power Studies for the CMS Pixel Tracker

M. A. Turqueti, R. Rivera, A. Prosser, S. Kwan, A. Todri Fermilab, Batavia, Illinois, USA The Electronic Systems Engineering Department of the Computing Division at the Fermi National Accelerator Laboratory is carrying out R&D investigations for the upgrade of the power distribution system of the Compact Muon Solenoid (CMS) Pixel Tracker at the Large Hadron Collider. Among the goals of this effort is that of analyzing the feasibility of alternative powering for the forward tracker, including DC to DC voltage conversion techniques using commercially available and custom switching regulators and switched capacitor circuits. To reduce the material budget in the tracker, the use of signaling over the power circuits is also being investigated. Tests of these approaches are performed using the PSI46 pixel readout chip currently in use at the CMS tracker. The tests will provide comparative results between the different kinds of regulators including parameter variation versus switching frequency. Performance measures of the detector electronics will include pixel noise and threshold dispersion results. Issues related to electromagnetic susceptibility to switching noise will be studied and presented.
N24-7:

P. P. Jarron1, E. E. Auffray1, S. S. Brunner1, H. H. Hillemanns1, A. A. Kluge1, P. P. Lecoq1, M. M. Morel1, T. T. Meyer1, F. F. Powolony1, M. C. S. C. Williams2, M. M. Despeisse3 1 PH, CERN, Geneva, Switzerland 2 University of Bologna, Bologna, Italy 3 IMT, EPFL, Neuchatel, Switzerland Time of flight (TOF) technique for PET is very demanding in timing performance, ideally less than 100ps FWHM precision. We present a time based differential technique to readout SiPM having less than 10ps rms electronic jitter. The novel readout is a fast front end circuit based on a first stage differential current mode amplifier with 20 ohm input resistance. The amplifier inputs are connected differentially to the SiPM anode and cathode ports. DC current of the amplifier input branches are offset to provide a discrimination threshold. The second stage of the current mode front end circuit is a fast differential amplifierdiscriminator circuit performing a time-over-threshold signal processing. The leading edge of the output signal provides the time information, the trailing edge the energy information. SPICE simulation results of the precise 3x3 mm SiPM model and the front end electronics design in 0.25um CMOS technology are presented and compared to experimental results obtained with a 3x3x20mm LSO scintillator Crystal readout with a SiPM. Time coincidence precision and energy spectra are also presented and interpreted with the SPICE simulation

Time Based Readout of Silicon Photomultiplier (SiPM) for Time of Flight PET Tomography

142

J05: NSS/MIC Joint Posters Wednesday, Oct. 28 J05-1: 10:30-12:00 Palm 3, 4 & 5

O. J. Roberts1, D. G. Jenkins1, P. Joshi1, O. Dorvaux2, M. Rousseau2 1 Physics, University of York, York, UK 2 Institut Pluridisciplinaire Hubert CURIEN (IPHC), Strasbourg, France

Nuclear Spectroscopy with a Novel LaBr3:Ce Scintillator and SiPM Detector

LaBr3:Ce crystals are high resolution scintillators that hold significant potential within the field of gamma-ray spectroscopy. The self-activity due to the 138La isotope is present in the scintillator at all times. and has been observed and used to gain a more comprehensive understanding of the inner dynamics of the detector. Effects due to neutron activation in these crystals was investigated through study of gamma-ray and neutron pulse shapes, as well as determination of its activated spectrum from a 241 Am/9Be source. It was determined that (n,) discrimination is not possible, and that the majority of gamma emissions in the spectra after neutron activation were due to excited states of Lanthanum and Bromine. Tests with a novel SiPM detector were also investigated. As gain fluctuates greatly with temperature, temperature stabilization is required in order to allow energy resolution and other information from the scintillator to be maintained. Measurements with regard to these two parameters were thus measured and their results are discussed.
J05-2:

A New Micro-Pattern Gas Detectors Used for Medical Imaging

N. Kahlaoui MDTN, National Center for Nuclear Sciences and Technology, Tunis, Tunisia On behalf of the RD51 Collaboration RD51 is an international collaboration involving 54 Universities and Research Laboratories from 20 countries in Europe, America, Asia and Africa. This collaboration aims at facilitating the development of advanced gas-avalanche detector technologies for applications in basic and applied research. One of these detectors was the Micromegas (Micro Mesh Gaseous Structure). Micromegas detector can be used in several areas such as medical imaging material science, and security inspection. Our goal is to develop a new medical imaging instrument based on Micromegas as a detector. A primary investigation was about determining performances of our detector like energy and spatial resolution. The first results showed an energy resolution at about 17% (Fe 55). Simulator work was also performed for this detector. Simulations were conducted on the relation between the pressure gas and the electron collected. Several gas mixtures are studied to determine its performances for the particles transport. Garfield and Maxwell codes are used like particle physics simulation programs. Actually detector optimisation was started in aim to study effect of different type and pressure gas in the particles collection and the effect of several X ray and gamma ray sources.
J05-3:

Determination of Incident X-Ray Spectra Using the Medipix3 Detector

P. Takoukam Talla, P. Bartl, J. Durst, T. Michel, G. Anton ECAP, University of Erlangen Nuremberg, Erlangen, Germany The hybrid semiconductor detector Medipix2 has been used in a broad range of applications. The detector can be used as spectrometer, since it delivers information about the number and the energy of photons impiging on it. This detector was used to determine high intensity incident X-ray spectra from an X-ray tube . In order to correct the charge sharing we applied deconvolution methods to the measured spectrum in order to get the incident spectrum. In order to minimize the influence of charge sharing the Medipix3 detector was developed. With its Charge Summing Mode, charge sharing can be corrected. In the "Charge Summing" the charges collected in each 2 x 2 pixel neighbourhood are summed up and attributed to only one pixel whose counter is incremented. Based on simulations the aim of this paper is to investigate how good we would be able to determine incident X-ray spectra using the Medipix3 detector: can the measured spectrum be directly interpreted as the real spectrum? Is deconvolution still necessary in order to determine the incoming spectrum? If yes, should we then use the same algorithm as in the case of the Medipix2 detector or a modified version?

143

J05-4:

Y. Yin1,2, S. Komarov1, H. Wu1,2, Q. Li3, A. Garson III3, D. Paul3, G. Simburger3, H. Krawczynski3, Y.-C. Tai1 1 Dep. of Radiology, Washington University in St. Louis, St. Louis, MO, US 2 School of Nuclear Science and Technology, Lanzhou University, Lanzhou, China 3 Dep. of Physics, Washington University in St. Louis, St. Louis, MO, US CdZnTe detectors with 350 m pixel pitches have been developed in order to evaluate the charge sharing of 350 m pixels. We fabricated a special 9-pixel-pattern on a CZT substrate with additional traces of conductor connecting the eight edge pixels to 8 readout pads of 1.41.4mm2 for easier probing. We set up a 2-D translation-stage that holds a collimator and a Gamma ray source to move the collimated beam to different location on the CZT surface. The data acquisition systems include (a) system with 10channels of Amptek A250 preamplifiers and a fast 500 MHz FADC for each channel, which gives the detail information of signal shape as well as the neighbor pixels, (b) a digital readout system based on three units NCI ASICs from BNL (32 channels each) for 64 anodes and 1 cathode signals, which gives distribution of triggered events with specific beam location and the spatial resolution performance of CZT detectors. Up-to-date, the CZT substrate has been contacted and both data acquisition systems have been developed and tested. Experiments will be conducted in the summer to measure (1) energy, timing and spatial resolution of the detectors, (2) DOI resolution based on either time difference between the anode and cathode signal or the ration of the signal amplitudes, and (3) charging sharing among 350 m anodes.
J05-5:

Characterization of Highly Pixellated CZT Detectors for Sub-Millimeter PET Imaging

Effects of Impurities and Secondary Phases on the Performance of CdZnTe Radiation Detectors

S. K. Swain Center for Materials Research, Washington State University, Pullman, WA, USA CdZnTe (10 % Zn) crystals for nuclear radiation detectors were grown using a vertical Bridgman growth technique. Analysis of the effects of impurities and secondary phases on the detector properties will be shown. Results will be presented showing the deteriorating effects of unintentional impurities on the carrier transport in CdZnTe. Attempts to establish correlation between total impurities with dopants and the mu-tau will be discussed. Discussion will be presented on the effects of % volume of secondary phases on the CdZnTe detector performance. Mu-tau values as high as ~ 1E-2 Cm2/V have been obtained from ingots with relatively lower % volume of Te inclusions. However no correlation was observed between the % volume of secondary phases and mu-tau indicating there are other parameters affecting the mu-tau value. Results documenting the simultaneous avoidance of unintentional impurities and secondary phases will be presented. Our attempts to minimize secondary phases will also be discussed.
The authors thank the US Department of Energy, NA-22, Contracts DE-FG52-06/27497/A000 and DE-FG52-08NA28769 for their financial support of this research. J05-6: From Imaging to Dosimetry: GEANT4-Based Study on the Application of Medipix to Neutron Personnel Dosimetry and Experimental Application

M. A. R. Othman1, D. G. Marinaro2, S. Guatelli1, M. Petasecca1, M. L. F. Lerch1, D. Cutajar1, M. Reinhard3, D. Prokopovich3, J. Jakubek4, S. Pospisil4, A. B. Rosenfeld1 1 Center for Medical Radiation Physics, University of Wollongong, Northfield Ave 2522, Wollongong, N.S.W., Australia 2 Human Protection and Performance Division, Defence Science and Technology Organisation, 506 Lorimer St, Fishermans Bend 3207, VIC, Australia 3 Australian Nuclear Science and Technology Organisation, Lucas Heights, N.S.W., Australia 4 Institute of Experimental and Applied Physics, Czech Technical University in Prague,12800 Prague 2, Prague, Czech Republic

The application of Si detectors with polyethylene convertor for fast neutron personal dosimetry is well known. An ideal dosimeter should be simply based on counting recoil protons only while the energy is independent for dose equivalent measurements. Adjustment of the energy response can be achieved by varying the thickness of the converter, however this is limited for a single detector. Recent developments of Si pixelated detectors with multichannel readout chip coupled with a detector brought flexibility in design of neutron personal dosimeters. The novel approach consists of adding a converter of polyethylene on top of the sensitive volume of the pixelated Si detector device that is configured into segments with varying converter thickness. It allows simple flexibility in adjustment of the weight of the converter with particular thickness by changing the area of the segment (i.e. number of readout pixels under this segment).Intensive Monte Carlo simulations with GEANT 4 of the response of Medipix2 covered by structured polyethylene converter allows the optimization of the number of segments and their polyethylene thickness to make the dosimeter energy independent for 0.3-15MeV neutrons in terms of counts/mSv and will be presented along with the response of the Medipix2 detector with simulated structured convertor on Cf-252 and Pu-Be sources.

144

J05-7:

S. P. Lansley1,2, G. T. Betzel1, F. Baluti1,3, L. Reinisch4, J. Meyer1 1 Department of Physics & Astronomy, University of Canterbury, Christchurch, New Zealand 2 MacDiarmid Institute for Advanced Materials & Nanotechnology, Christchurch, New Zealand 3 Oncology Service, Christchurch Hospital, Christchurch, New Zealand 4 Department of Physical & Earth Sciences, Jacksonville State University, Jacksonville, Alabama, USA A range of commercially-available synthetic diamond fabricated using chemical vapour deposition (CVD) was used for the fabrication of x-ray detectors. These detectors are intended for use in radiotherapy dosimetry (beam calibration and profiling, insitu dose measurements etc.), and hence they were appropriately packaged and tested in a clinical environment, using clinical apparatus and following clinical procedures. They could also be used as detectors of other forms of radiation and in other radiation detection environments, such as high-energy physics experiments. Response dynamics, e.g. rise and fall times, sensitivity, stability, and linearity of dose and dose-rate dependence, were observed as a function of the operating voltage, within the limitations of the Farmer dosemeter used. The effect of the contact material is also being investigated through the use of different contact materials, including the use of novel pyrolysed photoresist films instead of metallic contacts. The combination of linear dose-rate dependence of the photocurrent, negligible dark current levels (sub-pA, compared to nA-level photocurrents), low priming doses (few Gy) and high specific sensitivities (of up to 460 nCGy-1mm-3, compared to reported values of 50-140 nCGy-1mm-3 for a commercial natural diamond-based x-ray detector) demonstrates the potential of these devices as simple-touse, small-size, tissue-equivalent, sensitive x-ray dosimeters.
S P Lansley would like to acknowledge the Foundation for Research, Science and Technology (FRST), New Zealand, for the receipt of a NZ Science and Technology Post-doctoral Fellowship (UOCX0702), and the MacDiarmid Institute for Advanced Materials and Nanotechnology for research funding. J05-8:

Prototype X-Ray Detectors Based on CVD Diamond

A Technique to Locate Stored Iron in the Liver Using Attenuation Correction for Neutron Stimulated Emission Computed Tomography

G. A. Agasthya, A. J. Kapadia Dept of Biomedical Engineering, Duke University, Durham,NC, USA Abstract: Neutron stimulated emission computed tomography (NSECT) is a quantitative spectroscopic technique to detect element concentrations in the body. In previous work, we have demonstrated the ability to detect non-uniform distributions of iron overload in liver (hemochromatosis) with a sensitivity of approximately 5mg/g. The diagnosis of hemochromatosis is performed by detecting characteristic gamma photons emitted by iron nuclei after they undergo inelastic scatter with incident neutrons. The efficiency of detection of the gamma photons is a combination of the attenuation of neutrons passing through the body and the attenuation of gamma photons before reaching the detectors. With non-uniform iron distributions, therefore, the resulting total attenuation depends on the position of the iron store within the body with respect to the neutron beam and the gamma detectors. We are developing an attenuation correction technique which takes into consideration the position of the ironstore in the liver to compute a correction factor based on a combination of neutron and gamma attenuation. In this work we present results from a Monte-Carlo simulation study exploring the effect of the location of the iron-store within the liver. The NSECT scanning geometry used for data collection was simulated in GEANT4. A lesion of iron was placed at different locations within the liver and scanned to obtain an estimate of the detected signal. An estimate of the unattenuated signal was obtained and used to determine the total attenuation in the liver tissue. The attenuation profile was obtained for each position of the lesion and compared against a theoretical value. The results were found to be in agreement with each other, indicating that a theoretically calculated attenuation profile can be accurately used to perform attenuation-correction in NSECT.
Reference: Kapadia AJ and Floyd CE, "An attenuation correction technique to correct for neutron and gamma attenuation in the reconstructed image of a neutron stimulated emission computed tomography (NSECT) system," Proceedings of SPIE Medical Imaging 2005, vol. 5745, pp. 737-743. Willams,M.B,Narayanan D et al, "Analysis of position-dependent Compton scatter in scintimammography with mild compression",Nuclear Science, IEEE Transactions on, Volume 50, Issue 5, Oct. 2003 Page(s): 1643 - 1649. Kapadia AJ, Harrawood BP, Tourassi GD, GEANT4 simulation of NSECT for detection of iron overload in the liver, Proceedings of the 2008 SPIE Symposium on Medical Imaging, vol. 6913, pp. 691309. J05-9:

H. T. van Dam1, W. Drozdowski2, S. Seifert1, P. Dorenbos1, D. R. Schaart1 1 Dept. TNW, Delft University of Technology, Delft, The Netherlands 2 Institute of Physics, Nicolaus Copernicus University, Torun, Poland The recently discovered scintillator LaBr3:Ce3+ is of much interest for application in new radiomolecular imaging instruments, such as SPECT and time-of-flight (TOF) PET. Nowadays, research on new detectors is often supported by Monte Carlo (optical) simulations. Such optical simulations can be difficult, since scintillation photons undergo many interactions before being detected 145

Optical Scattering and Absorption in LaBr3:Ce3+

by the light sensor. To obtain accurate results each interaction has to be modeled with high accuracy, requiring detailed physical models and precise knowledge of the necessary input parameters. For LaBr3:Ce3+, these parameters, notably the optical bulk scattering and absorption properties, are not yet accurately known. The hygroscopicity of LaBr3:Ce3+ is a complicating factor. In optical measurements, even the slightest surface degradation due to moisture affects the results. In the present work we determine the absolute optical bulk transmission properties of LaBr3:Ce3+ as function of wavelength and sample thickness, using a method that circumvents the problems associated with hygroscopic samples. The absolute transmission of a polished, rectangular LaBr3:Ce3+ crystal inside a moisture tight sample holder equipped with quartz optical windows is measured. A dedicated data analysis program, which has as input all known optical parameters of the windows, the dimensions of the crystal and the measured absolute transmission values, then calculates the wavelength dependent combination of refractive index and optical attenuation length. As a validation the parameters obtained from 29 x 39 x 71 mm3 and 20 x 25 x 100 mm3 crystals are fitted with an absorption model. One of the fit parameters can then be compared with previously measured excitation peaks. As a second validation the absorption model is also used to fit measured emission spectra. The fit result is compared with the parameters obtained from the transmission measurements.
J05-10:

V. Linhart1, V. Borshchov2, D. Burdette3, E. Chesi3, V. Cindro4, N. H. Clinthorne5, E. Cochran3, B. Grosicar4, K. Honscheid3, H. Kagan3, C. Lacasta1, O. Listratenko2, G. Llosa1, M. Mikuz4, M. Protsenko2, V. Stankova1, A. Studen4, I. Tymchuk2, P. Weilhammer3, D. Zontar4 1 IFIC/CSIC-UVEG, Valencia, Spain 2 SE SRTIIE, Kharkov, Ukraine 3 Ohio State University, Columbus, OH, USA 4 Joef Stefan Institute, Ljubljana, Slovenia 5 University of Michigan, Ann Arbor, MI, USA

Development and Test of TAB Bonded Silicon Pad Detectors and Microcables for the Construction of Silicon Probes for Imaging Devices

The work described in this paper is focused on the development and construction of a small stack of self-dependent silicon detector modules which can find application in a number of devices where a silicon probe can enhance the performance as is the case in Compton Cameras (1), Small Animal PET (2) and PET magnifying probes (3) among others. To this end, the TAB (Tape automated bonding) technology has been intensively studied as an alternative to rigid bonding of silicon pad detectors, allowing for a large encapsulation factor without degradation of the probe performance. The multichannel chip cables are built by using aluminum-polyimide foiled dielectrics with trace pitch of 50 um and below. We have assembled together a module with the sensor TABed to the the readout ASICs (4). Tests based on spectroscopy of different radiation sources and the internal calibration circuitry of the ASICs allow to demonstrate that the use of this microcable technology can be used to increase the encapsulation factor without any degradation of the overall system performance. (1) Llosa, G et al. Last results of a first Compton probe demonstrator. IEEE Trans. Nucl. Sci. 55, 936-941, 2008. (2) Park, S.-J. et al. A prototype of very high-resolution small animal PET scanner using silicon pad detectors. Nuclear Inst. and Methods in Physics Research, A, 570 (3), p.543-555, Jan 2007. (3) The Collaborative Project "MADEIRA" (www.madeira-project.eu), cofunded by the European Commission through EURATOM Seventh Framework Programme (Grant Agreement FP7-212100). (4) VATAGP3, a Gamma Medica-Ideas designed application specific integrated circuit (ASIC).
J05-11:

C. M. Pepin1, M. Bergeron1, S. Shimizu2, N. Viscogliosi3, R. Fontaine3, R. Lecomte1 1 Departement de medecine nucleaire et radiobiologie, Universite de Sherbrooke, Sherbrooke, Qc, Canada 2 Hitachi Chemical Co., Ltd.,, Hitachinaka, Ibaraki, Canada 3 Departement de genie electrique et informatique, Universite de Sherbrooke, Sherbrooke, Qc, Canada

Fast High Lutetium Content Scintillators as Phoswich Candidates for Depth-of-Interaction (DOI) PET Detectors

The recent introduction of several new fast, high light yield, versions of generic cerium-doped lutetium silicates (LSO, LGSO, LYSO, LFS,...) has widened the list of potential candidates for phoswich detectors dedicated to medical imaging applications. Furthermore, digital signal processing methods based on adaptive filter theory now provide powerful approaches for crystal identification, even with scintillators having small decay time differences () . The LabPET digital signal processing technology was used to investigate the crystal identification performance of phoswich scintillator pairs with decay times in the range of 32 to 77 ns. The scintillator initial photon emission rate I0 was found to directly impact the identification error rate, particularly for fast phoswich crystal pairs with small . Even with a difference as small as 3 ns, an identification error rate of only 7% was achieved for discriminating the highest I0 phoswich pairs currently available . Overall, error rates less than 10% were measured for phoswich pairs with > 5 ns and less than 1% for > 20 ns, which is amply sufficient for typical PET imaging operation. 146

J. Imrek1, G. Hegyesi1, G. Kalinka1, J. Molnar1, D. Novak1, I. Valastyan1, L. Balkay2, M. Emri2, G. Opposits2, S. A. Kis2, L. Tron2 1 Electronic Deptartment, Institute of Nuclear Research of the Hungarian Academy of Sciences, Debrecen, Hungary 2 Pet Center, University Medical School of Debrecen, Debrecen, Hungary We report on a distributed online coincidence detection implementation that we have recently added to the miniPET-II small animal PET scanner. The implementation uses standard Ethernet and IP multicasting techniques, therefore no architectural changes were necessary to the existing system. For 2D reconstruction the implementation scales with the number of detectors in the system, so it is suitable for larger ring configurations, too. The online coincidence detection makes the system suitable for real time imaging applications.
J05-13:

J05-12:

Distributed Online Coincidence Detection Using IP Multicast for the miniPET-II Detector

S. Kabuki1, K. Ueno1, S. Kurosawa1, S. Iwaki1, H. Kubo1, K. Miuchi1, Y. Fujii2, D. Kim3, J. Kim4, R. Kohara5, O. Miyazaki5, T. Sakae6, T. Shirahata5, T. Takayanagi2, T. Terunuma6, Y. Tsukahara7, E. Yamamoto5, K. Yasuoka6, T. Tanimori1 1 Department of Physics, Gradulate School of Science, kyoto university, Kyoto, Japan 2 Energy and Environmental systems Laboratory, Hitachi, Ltd, Ibaraki, Japan 3 Department of physics, Seoul National University, Seoul, Korea 4 Department of Biomedical Engineering, National Cancer Center, Gyeong-do, Korea 5 Application Development Office, Hitachi Medical Corporation, Chiba, Japan 6 Proton Medical Research Center, University of Tsukuba, Ibaraki, Japan 7 Department of Medical System Engineering, Faculty of Engineering, Chiba University, Chiba, Japan Radiation therapy with proton and heavy-ion beams has been better established lately and the patient throughput is increasing. Although the therapy beam is controlled with high accuracy, it is difficult to know the location of distal dose falloff in the body. If real-time monitoring of the location is realized, the treatment quality would be improved. We have developed an electrontracking Compton Camera (ETCC) for real-time monitoring on the proton therapy. Our ETCC has a wide energy dynamic range (200-1300 keV) and a wide field of view. Therefore, ETCC has a potential as a quality assurance tool for proton therapy. In this presentation, we report the results of simulation and experiment with a 140 MeV proton beam.
J05-14:

Study on the Use of Electron-Tracking Compton Gamma-Ray Camera to Monitor the Therapeutic Proton Dose Distribution in Real Time

S. Cool1, S. Miller1, C. Brecher2, H. Lingertat2, V. Sarin3, V. V. Nagarkar1 1 RMD, Inc., Watertown, MA, USA 2 ALEM Associates, Boston, MA, USA 3 Boston University, Brookline, MA, USA We have recently fabricated tellurium-doped zinc selenide (ZnSe:Te) in a robust optical ceramic form, in the first synthesis of this remarkable material from a precursor powder, using powder consolidation techniques. We utilized two techniques in particular hot isostatic pressing and uniaxial hot pressing as well as cold pressing, in order to achieve the desired physical and optical properties, and we are continuing our efforts to refine processing methodologies and parameters to improve this materials physical characteristics and performance. ZnSe:Te promises major advances in radiation imaging and spectroscopy applications with its potential of well over 100,000 photons/MeV brightness, fast emission (~3 s and ~40 s), negligible afterglow (5.2 gm/cm3), light emission in the visible range (610-650 nm), and extraordinary radiation hardness (106 Gy). This paper will describe our fabrication methods and report our performance characterization results.
This project was partially funded by Department of Defense/DTRA Grant No. HDTRA 1-07-P-0242. J05-15: Investigation of Lu1.8Gd0.2SiO5:Ce (LGSO) Scintillators with APD Readout for Medical Imaging Applications

Fabrication of ZnSe:Te by Hot Pressing Techniques

S. Shimizu1, C. M. Pepin2, R. Lecomte2 1 Hitachi Chemical, Tsukuba, Japan 2 Sherbrooke Molecular Imaging Centre, Universite de Sherbrooke, Sherbrooke, QC, Canada

New high Lutetium content Lu1.8Gd0.2SiO5:Ce (LGSO) scintillators with promising physical properties for PET, but also for SPECT and X-ray counting CT applications, have recently been developed. The new crystals have high density (7.3 g/cc), high light output (>90% of NaI(Tl)), suitable emission wavelength for APD readout (420, 480 nm), and fast decay time that can be adjusted from 34 to 45 ns by varying Ce-doping concentration. LGSO photoelectron yields measured with a high performance 147

UV-enhanced APD readout were 12380 570, 14250 570 and 17680 760 pe/MeV with 0.025, 0.15 and 0.75 mol% Ce, respectively. Energy spectra of 68Ge (511 keV, PET), 99mTc (140 keV, SPECT) and 241Am (60 keV, CT) sources were obtained with the crystals coupled to the APD. Timing resolution was measured in reference to a fast PMT-plastic detector. FWHM energy resolutions at 511, 140 and 60 keV reached 10.5%, 17.6% and 32.2%, respectively. Timing resolutions of 1.3-1.5 ns were achieved. It is concluded that LGSO-APD detectors can be considered as very promising candidates for multimodality medical imaging applications.
J05-16:

L. J. Harkness1, A. J. Boston1, H. C. Boston1, J. R. Cresswell1, A. N. Grint1, M. Jones1, D. S. Judson1, P. J. Nolan1, D. C. Oxley1, D. P. Scraggs1, I. H. Lazarus2, J. Simpson2 1 Department of Physics, University of Liverpool, Liverpool, UK 2 STFC Daresbury Laboratory, Warrington, Cheshire, UK The development of the ProSPECTus prototype semiconductor Compton camera for medical imaging applications is underway. This project seeks to vastly improve the sensitivity of conventional Single Photon Emission Computed Tomography (SPECT) systems through electronic collimation. The key benefits of such a system are shorter data acquisition periods and the potential to reduce the amount of dose received by the patient. Geant4 simulations have been used to determine the optimum system geometry for the commonly used medical radioisotope, 99mTc. The system will now be built to the optimised specifications of 1cm thick Si(Li) and 1cm CZT and its imaging performance assessed. High statistics experimental Compton camera data is being acquired from the existing SmartPET small animal imaging demonstrator to assess the performance of the latest generation digital electronics and image reconstruction methods. Results of the tests will be analysed and discussed with regards to the consistency between simulated and experimentally acquired data. The status of the optimised ProSPECTus imaging system will also be outlined.
J05-17:

Development of the ProSPECTus Semiconductor Compton Camera for Medical Imaging

Characterization of Poly-Si TFTs for the Development of Large Area Active Pixel X-Ray Imagers

M. Koniczek, L. E. Antonuk, Y. El-Mohri, Q. Zhao Dept. of Radiation Oncology, University of Michigan, Ann Arbor, MI, USA Active matrix, flat-panel imagers (AMFPIs) are widely used in medical x-ray imaging for many applications. At low exposures, high spatial frequencies, or for small pixels, additive noise (associated with readout of the pixel signal) decreases the signal-tonoise ratio significantly, imposing limits on system performance. One promising approach for overcoming such limits is the incorporation of amplification circuits, based on low-temperature polycrystalline silicon (poly-Si) thin-film transistors (TFTs), in the pixels. In this study, measurements of the transfer and output characteristics obtained from poly-Si TFTs having several geometries, and obtained from various locations on different wafers and fabrication runs, are presented. From these measurements, parameters suitable for the RPI poly-Si device model were extracted and used for circuit simulations. From additional analysis, current noise densities were obtained as part of the evaluation of the quality and uniformity of devices across wafers and fabrication runs. The study concludes with a comparison of simulations and measurements of source follower amplification stages similar to those used in existing active pixel imager prototypes. These studies are being conducted in the spirit of guiding the design of poly-Sibased, active pixel imagers with a goal of optimizing their detective quantum efficiency. Research supported by NIH grant R01EB000558.
J05-18:

M.-S. Yun1, Y.-B. Kim2, M.-W. Kim2, S.-H. Jung2, G.-W. Jang3, S.-H. Nam4 1 Biomedical Engineering, Inje University, Gimhae, Gyungnam, South Korea 2 Medical Imaging Science, Inje University, Gimhae, Gyungnam, South Korea 3 Korea Institute for accreditation Medical Image, KIMI, Seocho, Seoul, South Korea 4 Biomedical engineering & Medical Imaging Research Center, Inje University, Gimhae, Gyungnam, South Korea

The Optical Characteristics of the Sample Based Liquid Crystal for Using Radiation Detector.

The digital radiation detectors are used clinically by diagnostic apparatus. However the digital radiation detector has some problems like high operating voltage, light blurring, low conversion efficiency and low fill factor. Thus we propose a new radiation detector that the photoconductor layer and liquid crystal layer are coupled in sandwich structure. X-ray absorption in the photoconductor layer controls the state of the liquid crystal via creation of charge carrier and the light modulation of liquid crystal make image formation. The advantages of the new radiation detector are that high resolution image is acquired and the signal amplification is possible by external visible light source. In this study, we study the optical properties and electrical properties of the new radiation detector to irradiate X-ray. The HgI2 was used by photoconductor material, and the aluminum is used by reflective layer. The thickness of HgI2 is about 200 and the operating voltage of the liquid crystal is 1.5~5V. The electrical property of HgI2 was measured, and the transmission efficiency of liquid crystal was measured by modulation

148

potential. Key Words : radiation detector;liquid crystal; photoconductor;

This research was financially supported by the Ministry of Knowledge Economy(MKE) and Korea Industrial Technology Foundation (KOTEF) through the Human Resource Training Project for Strategic Technology. J05-19:

Simulation and Measurements of the Internal Electric Field of a CZT Detector under High X-Ray Flux for Medical Imaging

O. Alirol, F. Glasser, E. Gros dAillon, J. Tabary CEA LETI, GRENOBLE, FRANCE Simulation of room-temperature radiation detector for high X-ray flux application leads to investigate the created space charge. Existing models are based on transport, trapping and subsequent detrapping of charge. Problem is that parameters of these models are difficult to measure and simulation is very sensitive to any variation of these parameters especially for the space-time solution. Measuring the electric field provides important information on the distribution and the dynamics of the electric phenomena for a better understanding of the detector behaviour under high X-ray flux. Observations show non uniform electric field and a polarization under X-ray irradiation depending on the applied voltage. Experimentals measurements and simulations based on charge transport models with trapping effects have also a good correlation.
J05-20:

Diffraction Enhanced Imaging with Pulsed Terahertz Radiation

Y. Wang, Z. Zhao, Z. Chen, L. Zhang, K. Kang Department of Engineering Physics, Tsinghua University, Beijing, China Terahertz radiation can penetrate through most non-conducting materials. Imaging with terahertz waves for these materials could provide an appropriate contrast compared with infrared, visible light and X-ray. Edge diffraction in terahertz pulsed imaging (TPI) enables the image contrast to be further enhanced, which exhibits considerable application potential to recognize shape features in the sample under test, especially for those weak absorption materials. In this report, we present a diffraction enhanced imaging method using pulsed terahertz radiation and demonstrate its capability of improving the image quality, without altering the conventional TPI system.
J05-21:

Y. Wang1, G. Baldoni1, J. Glodo1, E. V. Loef1, W. H. Rhodes2, C. Brecher2, K. S. Shah1 1 Materials Research, Radiation Monitoring Devices, Inc., Watertown, MA 02472, USA 2 ALEM Associates, Watertown, MA 02472, USA

Transparent LuAG:Ce and LuAG:Pr Ceramic Scintillator for PET

Rare-earth doped LuAG is a promissing scintillator for medical imaging applications such as Positron Emission Tomography (PET) because of its high gamma ray stopping power, high light output and fast response. Ce3+ and Pr3+ activated LuAGs show fast decay and are efficient high-response scintillators. Compared to single crystal growth method, ceramic processing route offers flexibility in dimension control and the relative low processing temperature. Since the crystal structure of LuAG is cubic, the isotropic optical properties allow fabrication fully transparent polycrystalline optical ceramics (TOC) and enable us to obtain high-performance, inexpensive LuAG for PET application. In this study, optically transparent polycrystalline LuAG:Ce and LuAG:Pr ceramic were prepared by sintering-HIP approach. The properties and microstructures of the ceramic were controlled by varying the processing parameters during sintering. The highly-dense, transparent LuAG ceramics were obtained by hotisostatic-pressing (HIP). Pores have been effectively removed during the HIPing process and the density reaches 99.97% after HIPing. Highly transparent polycrystalline LuAG:Ce and LuAG:Pr ceramics are obtained. Optical and scintillation properties of LuAG:Ce and LuAG:Pr ceramics are examined in this study and the results are compared to those of BGO. High light outputs as well as high energy resolution (8% (LuAG:Ce) and 5.5% (LuAG:Pr ) (FWHM) (137Cs source)) were obtained. The Ce and Prdoped LuAG ceramics shows fast scintillation decays of 43 and 21 ns. In summary, phase pure transparent LuAG:Ce and LuAG:Pr ceramic were obtained with its scintillation properties rivaling those of single crystalline LuAG, which makes LuAG ceramic a promising candidate for medical imaging applications such as PET.
J05-22:

An Engineering Validation of the Semi-Classical Trigger Approach for PET Coincidence Finding

J. Wu Fermilab, Batavia, IL, USA The classical trigger system finds coincidence in time domain by manipulating pulses with accurate arrival time and width. All cables carrying such pulses from front end modules to coincidence module must have well controlled propagation delay and all discriminators generating hit pulses are to be adjusted precisely. Contemporary full digital approaches need only precise timing at 149

the TDC in the front end module and eliminate requirement of accurate timed cables and modules for data merging. However, when the detector hit rate is fluctuated high, some hits may be discarded in the front end without checking for coincidence, which causes not only lower efficiency but also possible inappropriate emphasis or de-emphasis in final image. In the semi-classical trigger PET coincidence approach described in this document, the coarse times of all hits are transmitted to the coincidence module for coincidence check, as in classical trigger system. On the other hand, the communication channels are digital serial links so that no accurately timed cables and merging modules are needed. The advantages of the full digital system, i.e., robustness and confining precise time at the TDC in the front end are preserved. An engineering validation design and test has been done and is discussed in this document.
The author would wish to thank W. Moses, S. Choong and C. Vu of Lawrence Berkeley National Laboratory for very helpful discussions that motivated this work. J05-23:

Y. Kim1, J. Joung2, K. Park1, J. Lee3, K. Lee1, Y. Ahn3, K. Yoon3, Y. Choi4 1 Korea University, Seoul, Korea 2 Siemens Medical Solutions, Chicago, USA 3 Osteosys, Seoul, Korea 4 Samsung Medical Center, Seoul, Korea

A Fast Data Acquisition System with a Global Reset Scheme for Multi Channel Solid-State Detectors

Investigation has been conducted to design a fast data acquisition system for a multi channel solid-state detector. Previously we developed a DAQ (data acquisition) system for a detector that consists of an array of 67 SDDs. However, the local reset scheme was not adequate for multi-channel systems and it prevented the DAQ system from running fast. In addition, since the DAQ system using a wired-OR method connecting all channels into a single comparator, there were reset current drop and interference problems among channels while the sensors were resetting. In this study, we have developed a global reset scheme that using a gated-OR circuit to generate global reset signal based on a comparator circuit of each SDD. This scheme does not interfere among channels and provides enough current to operate all channels. The prototype detector used in this study consists of an array of 67 hexagon-shaped SDDs that is coupled to a single slab of a scintillator (10x10 cm2, 5mm thick). The size of the SDD is 16.4 mm in diameter. Our study showed that the DAQ system works well for the 67 channel detector and it can be operated as fast as 1 usec reset interval.
N25: Posters II Wednesday, Oct. 28 N25-1: 10:30-12:00 Palm 3, 4 & 5

J. He1, G. OKeefe2, T. Ackerly3, S. Gong2, M. Geso4 1 Department of Biomedical Engineering, Kunming University of Science & Technology, Kunming, China 2 Centre for PET, Austin Hospital, Melbourne, Australia 3 Radiotherapy Centre, Alfred Hospital, Melbourne, Australia 4 School Of Medical Sciences, RMIT University, Melbourne, Australia

Respiratory Motion Correction Utilizing Geometric Sensitivity in 3D PET: A Simulation Study

Artifacts caused by respiratory motion are an issue of concern in clinical PET imaging, they can be a significant factor that limits the PET image quality. To improve image quality, correction techniques utilizing external monitoring track systems or datadriven methods, were developed to compensate for subject motion, thereby improving lesion correction. In this paper, we present a new correction method that utilizes the geometric sensitivity correction (GSC) of a 3D mode PET scanner system operating in list event acquisition mode. The frame event rate from a given organ will depend on the axial location of the organ due to the change of the geometric sensitivity along the axial direction in a scanner. As a result the translation of respiratory motion frames to the reference can be estimated from frame event rate changes which are determined to suitable temporal resolution from the list-mode data stream. GSC method has several advantages over the existing methods, it only uses LOR events and is noninvasive, no additional hardware device systems are required and there is no additional patient preparation required. Using GATE (GEANT4 Application Tomographic Emission) and NCAT (NURBs-Based Cardiac Torso) phantom with an Allegro PET configuration, simulations of respiratory motion demonstrate that GSC can be used for respiratory motion correction.

150

N25-2:

A. Castoldi1, C. Guazzoni1, C. Ozkan2 1 Dip. Elettronica e Informazione, Politecnico di Milano and INFN sez. Milano, Milano, Italy 2 Dip. Fisica, Universita' degli Studi and INFN sez. Milano, Milano, Italy

Polycapillary Parallel Collimators for X-Ray Imaging: Experimental Characterization of the Energy and Angular Response Function

The full exploitation of polycapillary collimators to x-ray imaging systems, e.g. based on x-ray scattering or on x-ray fuorescence, is strictly related to the energy dependence of the critical angle of the incoming photons. This means the output beam is strongly dependent on the properties of the incoming beam therefore polycapillary optics need a design tailored to the specific experiment and require a careful performance qualification. We carried out an experimental study of a polycapilary parallel collimator at the synchrotron and with a conventional x-ray generator in order to obtain a complete qualification of the energy-angular response function. The impact of all relevant effects like energy and angular response, uniformity, penetration, point-spread function were studied. The obtained response matrix allows precise modelling of photon transport in the range of interest and full exploitation of the micrometer spatial resolution achievable with polycapillary technology when applied to x-ray scatter imaging. This experimental study can be taken as a reference to ease translation of such imaging systems from synchrotron to conventional x-ray sources.
N25-3:

Simulation of a Medical Linac with Evaluation of Dose Profiles Behind an Electron Applicator

I. Muenster, B. Kreisler, J. Durst, T. Michel, G. Anton ECAP, University of Erlangen, Erlangen, Germany Cancer therapy with electrons requires a well defined electron beam. Especially a flat dose profile in the tissue is important for the treatment planning. The normal, healthy tissue should be harmed as little as possible. Therefore the understanding of the physical properties behind the linear accelerator is of outstanding importance. A simulation provides an elegant access to the track of the electrons and gamma rays towards the patient. Our simulation was performed with the Monte Carlo simulation tool Geant4. The setup of a real apperatus was implemented with multiple regions for the evaluation of doses and spectral analysis. The results of the simulation provide the possiblity to test new designs before prototyping.
N25-4:

Test of the E/E Silicon Microdosimeter at the CATANA Facility

A. Fazzi, S. Agosteo, A. Pola, V. M. Introini, V. Varoli Dept. of Energy - CeSNEF, Politecnico di Milano, Milano, Italy A monolithic silicon E/E microdosimeter has been developed in the last years. The proper design of the detectors, the custom electronics for front-end and acquisition, and the data correction for the tissue- equivalence make the developed system a unique tool for the characterization of radiation fields. Recently, test runs have been performed at the CATANA facility (INFN-LNS, Catania, Italy) on the therapeutic 62 MeV proton beam line for ocular tumors treatments. Preliminary results are reported and discussed. The comparison of the experimental microdosimetric spectra with the ones measured with mini-TEPCs is fairly satisfactory.
N25-5:

A. T. Abdul Rahman1, D. A. Bradley1, A. Nisbet2,1, R. Thomas3,2 1 Centre for Nuclear and Radiation Physics, Department of Physics, University of Surrey, Guildford, GU2 7XH, Surrey, United Kingdom 2 Departments of Medical Physics, The Royal Surrey County Hospital (RSCH) NHS Trust, Edgerton Road, Guildford, GU2 7XX, Surrey, United Kingdom 3 National Physical Laboratory, Hampton Road, Middlesex, TW110LW, Teddington, United Kingdom Interest focuses on the use of a novel thermoluminescence (TL) system in radiotherapy dosimetry. Investigation has been made of commercially available germanium- (Ge) doped silica optical fibers, with characterization including inherent uncertainties, reproducibility and fading effects. Primary interest is in the radiological properties of the dosimeter, investigations including linearity and energy dependence. Investigation was also made of photoelectron dose enhancement resulting from use of iodinated contrast media as might be applied in radiation synovectomy and tumor treatments. For irradiation, use was made of electron beams of energy 9 - 20 MeV and photon beams of energy 6 - 15 MV at fixed dose rate, produced by a Varian LINAC 2100 CD accelerator, and a 90 kVp x-ray spectrum produced by a PANTAK Superficial X-Ray machine, both facilities being located at the Royal Surrey County Hospital. The readout system used is a Solaro TL Reader (Vinten TLD, Reading, UK). The results indicate this new dosimetric system to offer strong potential, both for quality control measurements as well as in in-vivo dosimetry. The optical fibers demonstrated reusability ( 0.5%) and low residual signal and fading. At fixed dose-rate, the Ge-doped optical fibers 151

The Investigation of Ge-Doped Silica Optical Fibres for Radiation Therapy Micro Dosimetry System

were found to produce a flat response of better than 4% (1 S.D) and showed good linearity of response up to a dose of 50 Gy for both electron and photon beams. The TL yield enhancement due to photoelectron production for fibers in an iodinated contrast medium was observed to be ~ 67% above the photon dose component measured in the absence of the contrast medium at the depth of measurement within a phantom. The loss of TL response due to fading, for electron and photon beams at fixed energy and constant dose was found to be (21.0 0.5) % and (24.0 0.5) % respectively over a period of 15 days post irradiation. Taken together with the small size and the various features make these commercially produced optical fibers a very promising TL material for use as a dosimetric system in radiotherapy.
The authors would like to thank University of Surrey for providing the facilities, Royal Surrey County Hospital and its Department of Medical Physics for their help in performing the irradiations and the University of Technology MARA Malaysia (UiTM) and The Ministry of Higher Education of Malaysia (MOHE) for supporting a research studentship. We also would like to thank all colleagues in the Dosimetry Group, University of Surrey for their support and suggestion. N25-6:

I.-C. Cho1, C.-H. Chen2, H. Niu3, C.-H. Hsu1 1 Department of Biological Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan, Republic of China 2 Department of Physics, National Tsing Hua University, Hsinchu, Taiwan, Republic of China 3 Nuclear Science and Technology Development Center, National Tsing Hua University, Hsinchu, Taiwan, Republic of China

A Novel Cell Irradiation System Using Backscattering Technique

The ion beam cell irradiation system is an important tool in radiobiological investigations. Generally, a vertical irradiating design with low particle flux is suitable for the cell irradiation system. In this work, a backscattering cell irradiation system was setup with the NEC 9SDH-2 pelletron accelerator at National Tsing Hua University. A schematic view of this system is shown in figure 1. Incident particles produced from the accelerator bombarded the self supported Au foil, then scattered to exit window. In order to minimize both the energy loss and straggling, a 100 nm thick Si3N4 film was mounted on the top of the irradiation chamber as the exit window. The window size, namely the irradiating area, is 1 1 mm2. The preliminary test was performed with 4.0 MeV proton beam, and a 100 nm thick Au foil tilted 45 with respect to incident direction. A Si detector was mounted on the Si3N4 window to monitor the outgoing protons. The scattered proton energy is at 3.947 MeV with an FWHM of 44 keV, as shown in figure 2. The particle flux can be varied by adjusting incident particle current and thickness of the Au foil. In this test, the particle flux was about 2.74/mm2-sec. The uniformity of the outgoing protons was checked by a CR-39 track detector. Figure 3 shows a picture of a selected area in the etched CR-39 track detector, indicating that the impinging protons were uniformly distributed in the irradiating area. The preliminary results manifested the backscattering technique appropriate for cell irradiation.
The support of this work was given by the National Science Council under Grant No. NSC96-2221-E-007-065-MY3. N25-7:

C. Guardiola1, C. Fleta1, M. Lozano1, G. Pellegrini1, F. Teixidor2, C. Vinas2 1 Instituto de Microelectronica de Barcelona (IMB-CNM, CSIC), Bellaterra, Barcelona, Spain 2 Instituto de Ciencia de Materiales de Barcelona (ICMAB, CSIC), Bellaterra, Barcelona, Spain

Silicon Neutron Detectors for Individual Dosimetry in Radiotherapy LINAC Rooms

A new type of thermal neutron detector for use in radiotherapy linacs is being developed. The device is optimized for environments with both high neutron and high gamma fluences. It is based in particle detection and not in activation, and therefore allows for individual dose measurement. The device is based in the conventional PIN silicon diodes covered with a converter layer structure but takes advantage of two new developments: first, new boron-based converter materials that are capable to provide good sensitivity to neutrons, and second, ultrathin silicon detectors that allow obtaining a high reject factor to the gamma background.
N25-8:

A Dedicated Processor for Monte Carlo Computation in Radiotherapy

V. Fanti, R. Marzeddu, C. Pili, P. Randaccio, S. Siddhanta, J. Spiga, A. Szostak Department of Physics, University and INFN of Cagliari, Cagliari, Italy A high speed Monte Carlo simulator for radiotherapy is being developed at INFN, Cagliari. During radiotherapy treatment planning, while performing Monte Carlo simulations of the radiation dose delivered to the human body, the Compton interaction of a photon with an electron plays an important role. Monte Carlo simulations of the radiation dose delivered to the human body give precise results over empirical methods but necessitate of a huge computing time. A fast, pipelined and cost effective design for real time simulation of the Compton interaction and dose calculation has been implemented on FPGA based hardware, running at more than 100 MHz, making it feasible to perform high speed Monte Carlo simulations for practical purposes and allowing the real time building of dose distribution maps. 152

N25-9:

C. Broennimann1, M. Schneebeli1, A. Fauler2, M. Fiederle2, B. Henrich3 1 DECTRIS AG, 5400, Baden, Switzerland 2 Albert-Ludwigs-Univ. Freiburg, Freiburg, Germany 3 Paul Scherrer Institut, 5232 Villigen PSI, Switzerland

A LARGE CdTe HYBRID PIXEL DETECTOR BASED on the TECHNOLOGY USED for the PILATUS DETECTORS

The technology developed for the hybrid pixel X-ray detectors of the Pilatus detector family1) was used to build a large 83.8 x 33.5 mm2 CdTe module operating at room temperature. The module consists of two 41.9 x 33.5 mm2 CdTe sensors provided by the Freiburger Materialforschungszentrum (FMF)2) The sensors are 1 mm thick and have been processed from wafers from Acrorad3). The ASICs (Application Specific Integrated Circuit) and the readout electronics is the same as the one used for the Pilatus detectors. These assemblies, equipped with silicon sensors, are very well characterized and calibrated4). 8 ASICs are bump-bonded5) onto one CdTe sensor. The CdTe sensors are pixelated with 240 x 195 pixels of a size of 172 m2. Between the two sensors there is a seven pixel wide gap. To our knowledge, this is the largest CdTe assembly ever built. The advantage of hybrid pixel X-ray detectors is in the direct conversion of X-rays and single photon counting electronics implemented in the ASIC. The sensor material determines the sensitive energy range of the detector. The usage of CdTe sensors together with the named ASICs opens up a new field of higher energy applications. The module has been characterized with monoenergetic X-rays. The behavior of the CdTe module is very similar to silicon modules. The linearity of the analog pulse height versus the incident X-ray energy has been studied using the analog out signal available for each pixel. We performed threshold scans for various energies. The module can be calibrated down to energies of around 8 keV. We will show results of a fully calibrated module as well as stability measurements of the sensor material. 1) Ch. Broennimann et al., J. Synch. Rad. 13 (2006) 120-130 2) M. Fiederle et al., IEEE Transactions on Nuclear Sience 51 (2004) 1864-1868 3) http://www.acrorad.co.jp 4) P. Kraft et al., J. Synch. Rad. 16 (2009) 368375 5) Ch. Broennimann et al., Nucl. Inst. and Meth. A 565 (2006) 303-308
N25-10:

L. A. P. Santos1, C. M. S. Magalhaes2, M. C. Sobrinho2, D. N. Souza2, A. M. Figueiredo2 1 CRCN - NE, CNEN, Recife, Brazil 2 Dept. Fisica, Universidade Federal de Sergipe, Aracaju, Brazil

An Innovative Instrumentation System for Measuring the Computed Tomography Dose Index

Dose surveys in computed tomography (CT) have indicated the need for assessment of patient dose and improvement in quality assurance procedures. The pencil ionization chamber is commonly used as ionizing radiation dosimeter to measure the CT dose index (CTDI the physical quantity established by the International Commission on Radiation Units & Measurements ICRU) [13]. As an alternative to the pencil ion chamber a commercially available SMT (Surface-Mount Technology) phototransistor was evaluated for dosimetry in computed tomography X ray beam. In search of a better dosimeter, three detection systems were built: the former with a single device; the second one with two devices one soldered back to other; and the third one was built soldering four devices making right angle one each other. The angular response was evaluated free-in-air for these three detection system. The dosimeters were irradiated in 45 increments for a full 360 rotation setting the CT scanner at a tube voltage to be 120 kV and the tube current to be 200 mA. Each detection system was under an X ray beam of 10 mm width at the center of rotation. The results showed that the detection system with four devices has a higher output signal, however, it has an appreciable directional dependence. The detector with one phototransistor displayed that the relative sensitivity was approximately flat. Such a result probably occurs because the X ray tube performs a 360 revolution around the gantry center, and consequently the detector is uniformly irradiated. The detection system with two phototransistors showed a response nearly twice of the single detection system, as was expected, but the angular dependence was not significantly observed. This analysis supported to identify the best geometry of an innovative instrumentation system for measuring the CTDI.
[1] International Electrotechnical Commission IEC 61674: Medical equipment Dosimeters with ionization chamber and/or semiconductor detectors as used in X-ray diagnostic imaging, Geneve, 1997. [2] International Commission on Radiation Units and Measurements ICRU 74: Patient dosimetry for X rays used in medical imaging, Oxford, 2005. [3] European guidelines on quality criteria for computed tomography (EUR 16262). Luxemburg, 1999. N25-11:

Scaling and Design of a 16 M-Pixel Charged-Particle Image Sensor

S. Kleinfelder, S.-H. W. Chiang Electrical Engineering, University of California, Irvine, CA, USA The design and scaling of a large-scale 2 by 2 cm CMOS image sensor with 10-bit digital readout for charged-particle imaging, EM7, is presented. The sensor contains ~50 million transistors spanning its 16 million pixels, and includes over 4,100 parallel analog-processing and A/D conversion circuits, over 4,100 double-buffered readout registers, and 12 parallel 10-bit readout busses for high data throughput. With sixty-four times as many pixels and nine times as many parallel readout circuits, a nave 153

scaling by replicating the basic circuit blocks of EM5 to build EM7 would have resulted in a catastrophic failure. The new clock distribution design in EM7 minimizes the clock delay by dividing the chip into multiple parallel sections driven locally by a treelike clock structure. By this technique, simulations showed that the readout shift-register clock delay is reduced from 4.7 ns to 0.14 ns, and the row shift-register clock delay is reduced from 1.7 ns to 0.12 ns. With local buffering, the ADC Gray code counter distribution delay is reduced from 35 ns to 0.9 ns. With the improved architecture and speed, the simulations show that EM7 can sustain an image rate of 75 frames/s, for a continuous data throughput of over 10 Gb/s. The large chip dimensions and the increased power consumption also require more robust power distribution. The IR drop is shown to be proportional to the resistance of the supply line, the amount of current drawn by each sink, and the square of the number of current sinks. The pixel matrix IR drop is reduced from 20 mV to 8 mV. Similarly the pixel current source IR drop was reduced from 80.7 mV to 2.58 mV. The pixel source follower worst-case bandwidth is increased from 6.92 MHz to 14.4 MHz. The opamp ground IR drop is reduced from 236 mV to 20.2 mV, and VDD IR drop from 90.7 mV to 14.9 mV. The opamp gain variation is reduced from 525% to 28%. The worst-case opamp bandwidth is increased from 0.87 MHz to 764 MHz.
N25-12:

R. M. N. Doesburg1, M. N. Clyne2, D. A. van Leeuwen3, N. J. Cook4, P. H. Butler1,5, A. P. Butler1,5,6 1 Dept. of Physics and Astronomy, University of Canterbury, Christchurch, New Zealand 2 ILR Ltd, Christchurch, New Zealand 3 Dept. of Electrical and Computer Engineering, University of Canterbury, Christchurch, New Zealand 4 Dept. of Medical Physics, Canterbury District Health Board, Christchurch, New Zealand 5 CERN, Geneva, Switzerland 6 Dept. of Radiology, Christchurch School of Medicine, University of Otago, Christchurch, New Zealand The MARS readout is a gigabit ethernet camera for MARS-CT scanners using an array of Medipix photon counting detectors. Six Medipix detectors can be read out at a frame rate of up to 100 Hz. A field programmable gate array (FPGA) based controller, with hybrid software and hardware, provides power and control to all the Medipix devices simultaneously. It implements an application programming interface (API) for workstations on a gigabit ethernet network. The network and transport layers are the widely known standard UDP/IP. The camera is mountable on the MARS-CT gantry.
N25-13:

Fast Ethernet Readout for Medipix Arrays with MARS-CT

N. Kawachi1, N. Suzui1, S. Ishii1, S. Ito1, N. S. Ishioka1, K. Kikuchi2, T. Tsukamoto3, T. Kusakawa3, S. Fujimaki1 1 Plant Positron Imaging Group, Quantum Beam Science Directrate, Japan Atomic Energy Agency, Takasaki, Gunma, JAPAN 2 Molecular Genetics and Physiology Research Team, National Institute of Vegetable and Tea Science, National Agriculture and Food Research Organization, Tsu, Mie, Japan 3 Chiba Prefectural Agriculture and Forestry Research Center, Chiba, Japan In recent years, radionuclide-based imaging technologies have been providing researchers with exciting new opportunities to study biology. On the other hand, there are a few case of success in the field of plant science, because most of the radionuclidebased methods for studying plants are invasive and require a statistical process with a large number of test plants. The positronemitting tracer imaging system (PETIS), which has a planar PET scanner, is one of the most powerful techniques used for conducting plant researches in order to study the distribution and translocation of water, sugar, nutrients, and environmental pollutants. In the sink-source relationship in the plant body, source abilities of a leaf imaged using a compartmental model analysis with PETIS data. In this case, to clarify the mechanism of the growth and development of the agricultural produces, we performed imaging experiments of sugar translocation to sink organ of fruits. A near leaf of the target fruits (Eggplant & Tomato) inhaled carbon-11 labeled carbon dioxide (100 MBq), and the translocation of carbon-11 labeled photoassimilate into fruits was imaged by PETIS for two hours. As a result, serial images of graduate increasing 11C activity and its ununiformly distribution in the fruit were acquired successfully. And also velocities of photoassimilate translocation and changes in the contributing rate with time of translocation from the leaf were estimated by analysis of PETIS data. This experimental method will be useful in not only investigating plant physiology, such as the mechanism of fruit growth, but also in solving certain environmental and food problems.
N25-14:

Molecular Imaging for Plant Physiology - Imaging of Carbon Translocation to Sink Organs with Positron Emitting Tracer Imaging System (PETIS)

Performance Evaluation of Gd2SiO5(Ce) Detector Applications for Blood Radio-Activity Measurement

J. G. Kim, K. M. Kim, J. S. K. Kim, S. K. Woo, J. A. Park, C. W. Choi, S. M. Lim, G. J. Cheon Molecular Imaging Research Center, Korea Instiute of Radiological and Medical Sciences, Seoul, Republic of Korea A GSO(Ce) scintillation detector has been designed, tested, and developed for blood radio-activity measurements with high sensitivity, compactness and less background. Four of GSO(Ce) crystals optically coupled to PMTs(15mm diameter) are used for increasing sensitivity of the detector. The size of 1.0 mol% cerium doped GSO crystal (17mm x 17mm x 20 mm with front-end 154

polished) is determined by GATE & MCNPX simulation codes. The Small size module type PMT (Hamamatsu H7826) is also used for compactness. The performance of GSO detector is evaluated with Energy resolution and sensitivity. An F-18 solution with activity ranged from 0.005 to 50 Ci/ml is tested for detectable sensitivity in Blood-activity measurement. In the last, Linearity of detector response and pulse height shift over broad count rate range is discussed.
N25-15:

Parameters Research of Gadolinium or Boron Coated MCP Thermal Neutron Collimator

Y. Yang, N. Lu Department of Engineering Physics, Tsinghua University, Beijing, P.R.China In this paper, we present the parameter research of coated MCP as thermal neutron collimator. B2O3 and Gd2O3 are selected as coating materials to absorb thermal neutrons for their large cross-sections. Before coating, the effect of doping and coating is compared by simulation. Results show that coating can provide better collimation effect than doping. As the rocking curve of coated MCP looks like sawtooth, we define the first maximum point in rocking curve as the parameter responding to the collimation effect of MCP. The parameters of coating materials, coating thickness, pore diameter, pitch, thickness of MCP are varied to achieve better collimation effect. Contour plot of transmission probability is used to evaluate the performance of MCP collimation. With a market available MCP of 13.5m pore diameter, 16.25m pitch and 510m pore length, 1m of Gd2O3 coating may implement the collimation result of 0.02 (transmission probability of 25.3 neutron that should be shielded). Two coated MCP collimators and one BF3 neutron counter are used to setup an experiment to measure the rocking curve. Keywords: neutron collimation, MCP, coating, doping
N25-16:

Micro Capillary Technology for Fast Neutron Detection and Imaging

M. A. Grohman, M. S. Derzon, R. F. Renzi International R&D Dept., Sandia National Labs, Albuquerque, NM, USA Advancements in micro-fabrication are enabling the manufacture of high pressure micro-capillaries for fast neutron detection and imaging. Each individual capillary acts as a unique micro-sensor, and the net result is the pixilation of a bulk gaseous neutron scatter detector. The current research uses natural helium-4 gas at very high pressure to populate numerous micro-scattering chambers. Micro-electrodes are used to measure the energy transferred to an ionized helium molecule in each scatter chamber. From coincident scattering events, existing algorithms can be used to reconstruct the energy and cone direction of a fast neutron event. We will present the proof-of-principle results from the first prototype system being constructed in calendar year 2009.
N25-17:

D. Vavrik1, J. Jakubek2 1 ITAM CAS, v.v.i., Institute of Theoretical and Applied Mechanics, Prague, Czech Republic 2 IEAP CTU in Prague, Institute of Experimental and Applied Physics, Prague, Czech Republic

Combination of Neutronography and Radiography for Characterising of Light Material Distribution in Dense Material Matrix

Neutronography and radiography are complementary methods from point of view of the visibility and contrast of the material imaged. Substances containing hydrogen are very well visible even behind denser material as Si or Fe. One of the interesting tasks is observation of the water solution containing salting materials which penetrate into stone, plaster or silicate matrix, see image for such sample. Water from this solution evaporates and salt is deposited in pores. Neutronography is very good tool for observation of the penetration and deposition kinematics but related matrix structure is difficult to observe. Complementary Xray radiography allows finding the connection between solution behavior kinematic and matrix structure. Methodology used can be applied also for engineering problem like hydrogen distribution in the steel, resin damages in metal-glass epoxy composites etc.
This work has been carried out in frame of the CERN Medipix Collaboration and has been supported in part by the Research Grant Collaboration of Czech Republic with CERN No. 1P04LA211, by the Fundamental Research Center Project LC06041 and the Research Program 6840770040 of the Ministry of Education, Youth and Sports of the Czech Republic. N25-18:

T. Kamiyama1, N. Miyamoto2, S. Tomioka1, T. Kozaki1 1 Division of Quantum Science and Engineering, Graduate School of Engineering, Hokkaido University, Sapporo, Japan 2 Institute for Genetic Medicine, Graduate School of Medicine, Hokkaido University, Sapporo, Japan Neutron resonance absorption spectroscopy (N-RAS) with a pulsed neutron source can distinguish the dynamics of individual nuclides. It analyzes the dynamics of nuclides from the Doppler broadening of the resonance peaks due to their thermal motion. From the line shape analysis of the peak, we can deduce a parameter that is called effective temperature, which implies the local 155

Quantitative Epithermal Neutron Tomography with Compact Pulsed Neutron Source

temperature on the nuclide. In actual measurements, the obtained spectrum is considered to include internal information of the sample object, which can be reconstructed as the nuclide and temperature tomograms using computed tomography (CT) technique. We have been studying such tomography as N-RAS/CT to obtain the information of nuclide or temperature distributions. We constructed the instrument for the compact neutron source on the Hokkaido University 45MeV electron liniac. The flight path length from the moderator to a sample is 13.3m, and the absorbed neutron energy is determined from the neutron flight time on the path. The CT stage has two motions, one is sample rotation and the other is neutron slit movement. The neutron slit can select the appropriate width form 0 to 3mm. All stage motion can control remotely. The CT experiment was carried by irradiating neutrons through 1 mm width slit. The slit was moved as 1mm step for 10 times in a certain sample rotational angle. The series of obtained spectra were fitted with the resonance absorption peak equation, then we obtained the mean temperature and projected nuclide densities, respectively. Finally, we obtained the quantitative temperature tomogram and nuclide density tomogram with the CT reconstruction about them. In the presentation, we will discuss the quantitative spectrum analysis and CT reproducibility. Present study is the result of research of analysis technique for a MOX fuel pellet by neutron resonance absorption entrusted to Hokkaido University by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT).
N25-19:

T. Nakamura1, R. Yasuda2, M. Katagiri1, K. Toh1, K. Sakasai1, A. Birumachi3, M. Ebine3, K. Soyama1 1 J-PARC Center, Japan Atomic Energy Agency, Tokai, Japan 2 Quantum Beam Directorate, Japan Atomic Energy Agency, Tokai, Japan 3 Mechanical Engineering and Electronics Section, Japan Atomic Energy Agency, Tokai, Japan

High-Spatial-Resolution Neuron Image Detector Based on Wavelength-Shifting Fiber Read Out for Time of Flight Measurements

A wavelength-shifting-fiber based neutron image detector with a spatial resolution less than 200 m was developed. The detector had a capability of clear imaging with a temporal resolution of several s. The detector equipped with a ZnS/6LiF scintillator where the scintillation light was read out with individual fibers based on a photon counting method. The fiber optic tapers (FOTs) were implemented in between the scintillator and the fibers to magnify a neutron image. The prototype detector that equipped with the FOTs exhibited the spatial resolution, which was improved by an order better compared that with the original detector. This kind of detector would be a quite unique and powerful tool for energy-selective neutron radiography at a pulsed source.
N25-20:

Exploring Neutron Scatter Camera Performance Using MCNP-PoliMi

C. Greenberg, J. Brennan, P. Marleau, N. Mascarenhas, S. Mrowka Homeland Security, Sandia National Laboratories, California, Livermore,CA, USA The neutron scatter camera is a fast neutron imager for detecting SNM. We developed a Monte Carlo model for the neutron scatter camera. Using the Monte Carlo N-Particle (MCNP) modification MCNP-PoliMi, we explored camera design options by varying cell geometry and optimized detection efficiency. We used PoliMi to investigate concepts for camera improvements and to analyze some threat scenarios. The Monte Carlo model has been used to improve spatial and energy calibration which previously relied on discrete sampling. The simulations have been benchmarked and have reasonable agreement with data.
N25-21:

J. Uher1, J. Jakubek2 1 CSIRO Minerals, Lucas Heights, NSW, Australia 2 IEAP-CTU, Prague, Czech Republic

Monte-Carlo Simulation of Fast Neutron Detection with Timepix

The imaging semiconductor pixel detector Timepix allows measurement of charge collected in each of its 256x256 pixels (55 m pitch). This feature makes it suitable for applications where energy and position of incoming radiation needs to be measured. We are developing a novel fast neutron spectroscopy and radiography technique based on tracking of protons recoiled by fast neutrons. The proposed setup consists of a polyethylene (PE) bead placed above the Timepix detector surface and irradiated by fast neutrons. Protons recoiled by neutrons from the PE bead hit the detector and generate a track. Analysis of the track gives energy, position and direction of flight of the proton. This information is used to reconstruct energy of the incoming neutron. The neutron source position can be calculated as well. If both the source position and spectrum are known, the PE bead can be replaced by an unknown hydrogen rich sample and 3D tomographic reconstruction of the sample can be done. A Monte-Carlo simulation of the fast neutron detection process was prepared. The code was based on combination of ROOT and SRIM/TRIM packages. The proton recoil and transport through the geometry was simulated. Energy deposited in the detector for each event was calculated. The code contains also a model for charge sharing between pixels to correctly simulate the detected proton tracks. Results of simulations compared with experiment and details of the neutron source spectrum/position and object 3D reconstruction will be presented.

156

N25-22:

J. M. Ryan1, C. Bancroft1, P. F. Bloser1, U. Bravar1, D. Fourguette2, C. Frost1, J. S. Legere1, L. Larocque2, G. Ritter2, J. Wood1, R. S. Woolf1 1 Space Science Center, University of New Hampshire, Durham, NH, USA 2 Michigan Aerospace Corporation, Ann Arbor, MI, USA We describe the development, fabrication and testing of the Neutron Spectroscope (NSPECT) prototype detector capable of realtime neutron source location and identification. Real-time detection and identification will fill an important gap in the current technology for locating Special Nuclear Materials (SNM). These materials, specifically uranium and transuranics, emit neutrons via spontaneous or induced fission. Unlike the other forms of radiation produced by SNM (e.g. gamma rays), copious and penetrating neutron emission is unique to fissionable material. NSPECT is highly suitable for locating SNM by detecting these fission neutrons and constructing images of their source. The instrument will be able to locate a neutron point source and report the fission spectrum. Neutron detection is of particular interest for SNM identification for security and proliferation deterrence, as well as for nuclear waste detection and monitoring. NSPECT is an instrument with imaging and energy measurement capabilities that is sensitive to neutrons in the 1-20 MeV range. The detection principle is based upon multiple elastic neutron-proton scatters in an organic scintillator. The instrument utilizes two detector panel layers. By measuring the recoil proton and scattered neutron energies, the direction and energy spectrum of the incident neutrons can be determined and discrete sources identified. Event reconstruction gives NSPECT the capability to provide an image of the source of interest and location. The design of NSPECT is a low power, low mass, rugged instrument, suitable for field deployment operations to locate and identify SNM. Its modular design allows the user to combine multiple units for increased sensitivity.
We acknowledge support from DTRA N25-23:

Development of the Real-Time Portable Neutron Spectroscope (NSPECT) for Detection and Identification of Special Nuclear Materials

A Neutron Imaging Facility Based on a RFQ Accelerator

Z. Guo, Y. Lu, Y. Zou, S. Peng, K. Zhu State Key Laboratory of Nuclear Physics and Technology, Peking University, Beijing 100871, China The neutron source based on radio frequency quadrupole (RFQ) accelerator or RFQ/DTL combination has been developed in recent years[1-2], which can be used in many fields. A neutron imaging facility based on a RFQ accelerator is being constructed at Peking University, which consists of a RFQ based fast neutron source, a moderator-collimator-shielding assembly and a thermal neutron imaging system. Fast neutrons are generated by deuteron-beryllium reaction. The deuteron beam is extracted from a 2.45 GHz ECR ion source and accelerated in RFQ to 2 MeV. The RFQ is working at 201.5 MHz in pulsed mode. The designed peak deuteron current is 40 mA with a duty factor of 10%. The expected fast neutron yield is higher than 3E12 n/s. The advantage to choose the frequency of 201.5 MHz comparing with 425 MHz is that the RF transmitter can use tetrode amplifier instead of the klystron, so the system can be more compact. The main moderation material is polyethylene. Due to the higher ray yield of deuteron-beryllium reaction there is a layer of lead around the central moderator to shield the ray. Outside the lead a layer of polyethylene is used to attenuate the fast neutrons, and a layer of polyethylene containing boron is used to absorb thermal neutrons. The thermal neutron imaging system consists of neutron scintillator, mirror, lens and CCD camera. There are two imaging positions with the radiograph size 10 cm X 10 cm and 20 cm X 20 cm, respectively. The additional shielding for CCD is needed. This facility will be used for thermal neutron radiography of small samples and neutron radiography technology development. References [1] Hamm, R. W., Proceedings of SPIE 4142, 2000, 39. [2] Baxter, D., Cameron, J., Derenchuk, V., et al. Nucl. Instr. and Meth. B241, 2005, 209.
N25-24:

Phase Contrast Neutron Imaging at a Medium Intensity Neutron Source

K. K. Mishra, A. I. Hawari Nuclear Engineering, North Carolina State University, Raleigh, NC, USA Neutron Imaging is a technique for the non-destructive testing (NDT) of materials. It is characterized by its sensitivity to materials composed of low atomic number (Z) elements such as hydrogen. Enhancement of the spatial resolution and the contrast of the obtained images are primary objectives that are continuously being pursued in the development of this technique. An approach of improving the contrast at the material edges of a sample is non-interferometric phase contrast neutron imaging, which has been demonstrated at the start of this decade. Since then it has been performed at multiple facilities located at high flux neutron sources. However, due to its physical requirements (e.g., the use of pinhole apertures), the applicability of this technique remains primarily limited to such high flux facilities. Consequently, to expand its utilization in NDT applications, its performance at low and medium intensity neutron sources, that are available around the globe, should be explored. In this work, the feasibility of performing phase contrast neutron imaging at the North Carolina State University PULSTAR reactor was studied. The 1-MW 157

PULSTAR represents a prototypical medium flux neutron source. It is demonstrated that phase contrast imaging can be successfully implemented if careful attention is given to tailoring the neutron beam with the objective of maximizing the thermal neutron flux (and content) while minimizing the gamma-ray and fast neutron noise. Results are presented of imaging exercises demonstrating the phase contrast concept using a Fuji BAS-NDG image plate detector.
N25-25:

Z. Ziao1, K. K. Mishra1, A. I. Hawari1, H. Z. Bilheux2, P. R. Bingham2, K. W. Tobin2 1 Nuclear Engineering, North Carolina State University, Raleigh, NC, USA 2 Oak Ridge National Laboratory, Oak Ridge, TN, USA

Investigation of Coded Source Neutron Imaging at the North Carolina State University PULSTAR Reactor

A neutron imaging facility is located on beam-tube #5 of the 1-MWth PULSTAR reactor at the North Carolina State University. An investigation has been initiated to explore the application of coded imaging techniques at the facility. Coded imaging uses a mosaic of pinholes to encode an aperture, thus generating an encoded image of the object at the detector. To reconstruct the image data received by the detector, corresponding decoding patterns are used. The optimized design of coded masks is critical for the performance of this technique and will depend on the characteristics of the imaging beam. In this work, Monte Carlo (MCNP) simulations were utilized to explore the needed modifications to the PULSTAR thermal neutron beam to support coded imaging techniques. In addition, an assessment of a potential coded mask design has been performed. The simulations indicated that a 12 inch single crystal sapphire filter is suited for such an application at the PULSTAR beam in terms of maximizing flux with good neutron-to-gamma ratio. A gadolinium aperture thickness of 500 m was used to construct the mask using a 38 34 URA pattern. Experiments using such a URA design are currently underway.
N25-26:

A Linear Position Sensitive Neutron Detector Module Design for the ORNL Spallation Neutron

Source
L. L. Funk Neutron Facilities Development Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States On behalf of the Detector Group and Data Acquisition Systems Group A linear position sensitive neutron detector design has been developed for use in several instruments of the Spallation Neutron Source at Oak Ridge National Laboratory. The detectors use cylindrical detector tubes containing He3 biased to operate as a proportional counter. Neutron detections occur through the n + He3 p + H3 reaction. The position of an event along the tube is obtained by comparing signal strengths at each end of the tube. Short detector tube to preamplifier connections, a current sensitive preamplifier input stage, and early digitization of signals are used to achieve a position resolution and linearity of position detections superior to previous designs. The basic principles of operation will be presented along with some typical performance data. There will be an overview of the electronics used to detect and digitize the detector signals. Methods the detectors use to communicate with data acquisition systems will be outlined.
N25-27:

E. Takada1, A. Inoue1, H. Okada2, S. Naka2, J. Kawarabayashi3 1 Toyama National College of Technology, Toyama, Japan 2 University of Toyama, Toyama, Japan 3 Nagoya University, Nagoya, Japan

Application of Hetero Junction Organic Photodiode to Radiation Measurement

Organic optical photodiodes (OPDs) have obtained considerable interest in the fields of photonics devices, as they have the potential for large area and flexible photo detector. So as to study their applicability to radiation measurement, OPDs with classical hetero junction structure were fabricated and applied to X-ray measurements. Device structure was ITO (150nm)/ NPD (100nm) / td-PTC (50nm, 65nm, 80nm) / Al (100nm) with sensing area of 2mm x 2mm. The device showed rectifying property under forward and reverse voltage. Under irradiation of white light with intensity of 100mW/cm2, current density was 104 times larger than the dark current density. X-ray irradiation experiments have been carried out. It has been shown that X-rays could increase the current density, although there are some problems such as increasing dark current with irradiation. At acceleration voltage of 200kV, the current density increased to about 0.16nA/cm2 from about 0.035nA/cm2 at dark condition. Xray shielding experiments with Al and X-ray intensity profile measurements have been carried out and the both results have shown that the OPDs worked as radiation detector. An inorganic scintillator on the OPD could enhance the X-ray induced current, which showed the possibility of combined system of OPD and usual scintillating material as new radiation detector. While further study is necessary to improve the unstable property of dark current, we can conclude that OPD can be a new radiation detector.

158

N25-28:

C. Fuentes1,2, B. Allongue2, S. Buso3, G. Blanchot2, F. Faccio2, S. Michelis2,4, S. Orlandi2, G. Spiazzi5 1 UTFSM, Valparaiso, Chile 2 CERN, Geneva, Switzerland 3 Dept. of Technical Management of Industrial Systems, Padova University, Padova, Italy 4 EPFL, Lausanne, Switzerland 5 Dept. of Information Engineering, Padova University, Padova, Italy The upgrade of the LHC experiments at CERN brings new challenges in the design and integration of the detectors. Among them and to achieve the required channel density, the power distribution to the front-end electronics of the trackers has to be implemented in a more efficient and compact manner. This paper proposes a new powering scheme for the trackers based on the use of custom on-board DC-DC converters. The converters are designed to tolerate high levels of radiation and intense magnetic fields present in the experiments.
Cristian Fuentes gratefully acknowledges the support provided by MECESUP-Chile under grant FSM0601 Stefano Michelis has been supported by a Marie Curie Early Stage Research Training of the European Community's Sixth Framework Programme under contract number MEST-CT-2005-020216 - Elacco. N25-29:

Power Distribution with Custom DC-DC Converters for SLHC Trackers

M. Subramanian1, B. Phlips2, F. Kub2 1 Physics and Astronomy, George Mason University, Fairfax, VA, 22030 2 Space Science Division, Naval Research Laboratory, Washington, DC, 20375 We have characterized several Silicon On Insulator (SOI) devices for use as neutron detectors. SOI neutron detectors offer the advantage of a rugged structure with large area while incorporating a thin Si layer coated with a converter material. The performance of these SOI detectors depends on the thickness of the Si layer, the dimensions of the strips, on the convertor material and its thickness. This paper describes the design of the SOI wafer, the capacitances of the devices on the SOI wafer and the initial testing of the devices as neutron detectors. SOI wafers of two different Si thicknesses, with varying dimensions for the strip pitch were characterized. The IV and CV properties of the different devices were studied. The devices are then coated with the converter material and are used to detect neutrons.
N25-30:

Characteristics of a Silicon on Insulator Neutron Detector

The Characterization of CsI-Based Reduced-Hydroscopicity Nanocomposite Scintillators

J. M. Cook, B. L. Bennett, M. W. Blair, L. O. Brown, R. D. Gilbertson, A. Li, E. A. McKigney, R. E. Muenchausen, R. E. Del Sesto, N. A. Smith, S. C. Tornga, D. A. Wrobleski Los Alamos National Laboratory, Los Alamos, NM, USA The need for inexpensive, easily-formable gamma-ray detectors may be satisfied by nanocomposite scintillators. Nanocomposite scintillators based on CsI present the possibility of detectors that are resistant to damage from moisture and that are suitable for mating with photodiodes. The optical and physical properties of CsI-based nanocomposite scintillators will be compared to those of bulk powder and single crystals. The suitability of the nanocomposite material for gamma-ray detection will be presented, including the radiation response, linearity, and light yield. The expected cost and characteristics of field-placeable detectors will be evaluated.
N25-31:

A Low Noise Pixel Architecture for Scientific CMOS Monolithic Active Pixel Sensors

R. Coath, J. Crooks, A. Godbeer, Z. Zhang, M. Stanitzki, M. Tyndel, R. Turchetta, M. Wilson CMOS Sensor Design Group, STFC - Rutherford Appleton Laboratory, Oxford, Oxfordshire, UK This paper presents the design and characterisation of FORTIS (4T Test Image Sensor), which is a low noise, CMOS monolithic active pixel sensor for scientific applications. The pixels present in FORTIS are based around the 4-transistor (4T) pixel architecture, which is already widely used in the commercial imaging community. The sensor design contains thirteen different variants of the 4T pixel architecture to investigate the effects of changing its core parameters. The variants include differences in the size of the pixel, the diode and the in-pixel source follower, and differences in the capacitance of the floating diffusion node (the input node of the in-pixel source follower). Processing variations have also been studied, which include varying the resistivity of the substrate and investigating the existence or absence of a special deep P-well layer. By varying these parameters, the 4T pixel architecture can be optimised for scientific applications where detection of small amounts of charge is required.

159

N25-32:

Development of Lead Chalcogenide Nanocrystalline (NC) Semiconductor Detectors

G. Kim, M. D. Hammig Nuclear Eng. and Radiological Sci., University of Michigan, Ann Arbor, MI, USA Nanocrystalline (NC) semiconductor materials show unique properties which arise due to the strong quantum confinement effect. The direct generation of multiple carriers upon the impingement of visible light has previously motivated the research, for applications such as solar cells. If the multi-exciton states have highly uniform multiplicities across macroscopic NC samples, then one can also potentially quench the statistical counting noise associated with charge-carrier creation in the bulk material. Based on the favorable features of lead chalcogenide materials for their application to the detection of ionizing radiation, PbSe and PbTe NC particles of different size and shape were synthesized by changing the reaction conditions, and their physical and opto-electric properties were investigated. By sandwiching the closely packed NC assembly, assembled using its selfagglomerating nature, between various metal contacts, its performance as an NC diode was tested. Furthermore, the viability of using lead chalcogenide NC materials as a basis for the detection of ionizing radiation is assessed and confronting empirical challenges are discussed.
N25-33:

D. S. Levin1, P. S. Friedman2, R. W. Ball1, J. W. Chapman1, C. Weaverdyck1, B. Zhou1, E. Etzion3 1 Randall Lab, University of Michigan, Ann Arbor,MI, USA 2 Integrated Sensors,LLC, Toledo,OH, USA 3 School of Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel

A Plasma Panel Sensor Detector for SLHC-ATLAS Muon Spectrometer Upgrade

Muon spectrometer performance at high and super-high luminosity at the LHC ( 2 1034 cm-2 sec-1 after the 2014 shutdown and 1035 cm-2 sec-1 after the 2017 shutdown) and at future colliders demands high resolution tracking detectors with very fast time response and excellent temporal and spatial resolution. Towards these goals, we are investigating a new radiation detector technology based on Plasma Display Panels (PDP), the underlying engine of panel plasma television displays. The design and production of PDPs is supported by an extensive and experienced industrial base with four decades of development. Emerging from this television technology is the Plasma Panel Sensor (PPS), a novel variant of the micropattern radiation detector. The PPS at its core, is a dense array of micro-Geiger plasma discharge cells operating in a non-ageing, hermetically sealed gas mixture. This is an inherently digital device with the potential for very fast response times of order ns per pixel, very fine grain (pixel pitch can be less than 100 m) and low cost. We report on the PPS development program, including the design, construction and staging of a PPS Test Cell.
N25-34:

Innovative, Reworkable, and Robust Packaging of Semiconductor Detectors

B. Cardoso, M. Wrosch, A. Soriano Creative Electron, INC, SAN MARCOS, CA, USA In this paper Creative Electron, Inc. (CEI) presents an innovative, reworkable, and robust packaging technology to address the production of semiconductor based radiation detector systems for applications related to homeland security, defense, astrophysics, and medical imaging. This packaging technology is both reworkable and robust based on CEIs Transient Liquid Phase Sintering (TLPS) bonding material and CEIs metallization process of semiconductor detectors. The production of semiconductor radiation detection systems suffers from the relatively low yield of crystals and associated electronics compounded with the limited availability of some of the system components. Furthermore, the before mentioned applications require robust attachment techniques that must withstand mechanical stress and must have high bump-bond yield. The TLPS technology was initially developed for semiconductor die attachment in radar applications. CEIs experience in customizing the formulation of TLPS has led to complete characterization of assemblies using Cadmium Zinc Telluride (CZT) detectors. Tests show that tje pixel pads on the CZT and on the board have not been damaged after multiple rework processes. Thus, great cost reduction can be achieved because the same CZT detector can be used multiple times without the need of replacing the pixel pads on the CZT or on the board. The TLPS bonding material in tandem with our CZT metallization technology results in a packaging technology with the following characteristics: - Capability to reuse CZT crystals and readout electronics multiple times - Strong mechanical properties shear force > 2,500psi - High inter-pixel impedance capacitance < 1pF and resistance > 100G - Low bonding electrical resistance bump resistivity < 20.cm - Low assembly temperature reflow < 120C.

160

N25-35:

F. Garcia1, G. Pellegrini2, M. Lozano2, J. Balbuena2, R. Orava1, M. Ullan2 1 Detector Laboratory, Helsinki Institute of Physics and Department of Physcal Sciences, University of Helsinki, Helsinki, Finland 2 Centro Nacional de Microelectronica, Barcelona, Spain An ultra thin silicon detector called U3DTHIN[1] has been designed and built for neutral particle analyzers (NPA) and thermal neutron detection. The main purpose of this detector is to provide a state-of-the-art solution for detector system of NPAs for the ITER experimental reactor and to be used in combination with a Boron conversion layer for the detection of thermal neutrons. Currently the NPAs are using very thin scintillator - photomultiplier tube [2,3], and their main drawbacks are poor energy resolution, intrinsic scintillation nonlinearity, relative low count rate capability and finally poor signal-to-background discrimination power for the low energy channels. The proposed U3DTHIN detector is based on very thin sensitive substrate which will provide nearly 100% detection efficiency for ions and at the same time very low sensitivity for the neutron and gamma radiation. To achieve a very fast charge collection of the carriers generated by the ions detection a 3D electrode structure [4] has been introduced in the sensitive volume of the detector. One of the most innovative feature of these detectors has been the optimal combination of the thin entrance window and the sensitive substrate thickness, to accommodate very large energy dynamic range of the detected ions. An entrance window with a thickness of tens of nanometres together with a sensitive substrate thickness varying from less than 5 m, to detect the lowest energetic ions to 20 m for the height ones has been selected after simulations with GEANT4. To increase the signal to background ratio the detector will operate in spectroscopy regime allowing to perform pulse-height analysis. The technology used to fabricate this 3D ultra thin detectors developed at Centro Nacional de Microelectronica in Barcelona and test structure measurements will presented.
N25-36:

U3DTHIN - Ultra Thin 3D Silicon Detector for Plasma Diagnostics at ITER Tokamak

K. S. Song1, S. Jeon1, B. Kim1, D. H. Lee1, Y. G. Hwang1, H. Cho2, S. H. Nam3, Y. Huh1 1 Advanced Medical Device Research Center, Korea Electrotechnology Research Institute, Ansan, Gyeonggi-do, Korea 2 Department of Radiological Science, Younsei University, Wonju, Gangwondo, Korea 3 Department of Biomedical Engineering, Inje University, Kimhae City, Kyungnam, Korea X-ray imaging has been playing a very important role in diagnostics. The large area flat panel detector for digital radiology, called an active matrix, consists of a two-dimensional array of TFTs, which are fabricated on the a-Si thin film layer. In practice it has been the investment in the development of flat-panel displays, for use in laptop computers and now desktop PC and largely flat panel television. Normally, there are two type methods on the large sized flat panel display technology, so called LCD (liquid crystal display) and PDP (plasma panel display). LCD is based on the thin film transistors (TFTs) to realize each pixilated display image on the amorphous silicon (a-Si) thin film layer. And PDP flat panel display uses the rib structure. TFTs have been applied for X-ray digital image sensor to diagnosis. However, PDP is not considered this application. In this work, we evaluate the possibility of medical application of PDP on digital X-ray image sensor. The fabrication process of flat panel X-ray sensor based on PDP structure uses the customized process on commercial PDP fabrication process. Each pixel of X-ray image is realized by rib structure used in PDP to pixellate display image. We evaluate the first ionized electron signal on the charged gas and the secondary amplified electron signal on MgO layer. These X-ray signals are very small but comparable that of the noise. We should consider the specifically optimized mechanism to readout these electron signals on PDP structure.
This work was supported in part by a Grant-in-Aid for Strategy Technology Development Program from the Ministry of Knowledge Economy, Republic of Korea. N25-37: Monolithic Active Pixel Sensors on-Pixel Sparsification Architecture with Pulse Amplitude Information Storing.

Gas Typed X-Ray Image Sensor Using Plasma Display Panel (PDP) Structure

E. Spiriti1, J. Mlynarczyk2 1 Sezione di RomaTre, INFN, Rome, Italy 2 Department of Electronics, AGH, Krakow, Poland

Monolithic Active Pixel Sensors (MAPS) are becoming a realistic alternative to hybrid pixels in particles physics application. The proposed new colliders like the ILC (International Linear Collider) and SuperB factory are considering them for the vertex detectors due the low material budget and high space resolution possible with those sensors. The large number of needed pixels, of the order of hundreds of millions, produces a huge data throughput that, due to the low level occupancy for this application, could be reduced by two orders of magnitude if we implement an on pixel sparsification procedure to readout only hitted pixels. The paper presents a new approach for on-pixel sparsification using only N-MOS transistors to implement the on pixel signal processing (the use of P-MOS transistors would create N-type competing N-well that would drain part of the signal charge) and a Vertically Integrated version (CMOS version). The architecture implements: correlated double sampling, signal discrimination and analog information storing on each pixel. The results of a test chip, submitted as a multiproject run in the STMicroelectronics 161

130nm technology, will be described ( chip characterization with radioactive sources and on a test beam, measurements of threshold mismatch, temporal and fixed pattern noise, digital to analog cross talk) and the new Vertically Integrated version, submitted in the Chartered/Tezzaron 130nm technology in the 3DIC(3D Integration Consortium) framework will be described.
N25-38:

S. Duval1, T. Oger1, E. Morteau1, H. Carduner1, P. Leray1, J.-S. Stutzmann1, J.-P. Cussonneau1, J. Lamblin1, D. Thers1, A. Breskin2 1 UMR 6457, Subatech, Nantes, France 2 Departement of Physics, The Weizmann Institute of Sciences, Rehovot, Israel

Scintillation Detection with a Gaseous Photomultiplier for Compton Imaging with Liquid Xenon

We will describe and provide recent advances of a liquid-xenon time-projection chamber, designed to validate a new medical imaging concept named 3 imaging. The advantages of detection gamma-induced scintillation light with a gaseous photomultiplier (GPM) operating in cryogenic mode are discussed; first results are presented.
N25-39:

The Barrel DIRC of the PANDA Experiment at FAIR

J. Schwiening GSI Helmholtzzentrum fr Schwerionenforschung GmbH, Darmstadt, Germany On behalf of the PANDA-DIRC Group Cooled antiproton beams of unprecedented intensities in the momentum range of 1.5-15 GeV/c will be used at the PANDA experiment at FAIR to perform high precision experiments in the charmed quark sector. The charged particle identification in the barrel region needs a thin detector operating in a strong magnetic field. Both requirements can be met by a Cherenkov detector using the DIRC principle. Combining the time of arrival of the photons with their spatial image determines not only the particles velocity, but also the wavelength of the photons. Therefore, dispersion correction at the lower and upper detection threshold is possible. We present details of the design of a prototype for the Barrel DIRC detector of the PANDA experiment at FAIR and its performance in a proton test beam.
Work supported by EU FP6 grant, contract number 515873, DIRAC secondary-Beams. N25-40:

X-Ray Microscopy with 100nm Resolution for Tomographic Applications Using a MicroCT SEM Attachment

P. Bruyndonckx, A. Sasov, B. Pauwels, X. Liu SkyScan, Kontich, Belgium A microCT attachment for any scanning electron microscope (SEM) was developed to take advantage of the highly focused x-ray spot sizes achievable in these instruments. At very high spatial frequencies (small details), image contrast is also governed by the absorption and phase contrast transfer functions. This can result in contrast of small object structures being severely reduced at spatial frequencies that otherwise wouldnt suffer yet from the smoothing effect of the finite x-ray spot size. Both analytical and experimental studies were made to determine which instrumentation and operational parameters are most important for increasing the spatial frequency where loss of image contrast occurs towards higher frequency. After adjusting the experimental parameters of the MicroCT attachment accordingly, structures down to nearly 100 nm could be visualized in a star resolution pattern. To improve the resolution even further, the X-ray spot size needs to be reduced. Results of Monte Carlo simulation studies to determine the best combination of target material, target thickness, support material and electron beam potential to minimize the radial distribution of the x-rays leaving the target (x-ray focus spot) will be presented. The developed microCT attachment opens new possibilities for non-destructive 3D imaging in wide range of applications using any conventional SEM, eventually at sub-100 nm resolutions.
N25-41:

R. B. Borade1, E. D. Bourret-Courchesne2, M. J. Weber1, S. E. Derenzo1 1 Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA

Syntheses and X-Ray Excited Luminescence Properties of Europium Doped Ba3P2O8, Ba3P4O13, Ba2P2O7 and BaP2O6

Luminescence properties of Ba3P2O8:Eu2+ has been investigated by many authors (1-7) but their scintillation properties have not been reported so far. In this work Eu2+ doped and undoped Ba3P2O8, Ba3P4O13, Ba2P2O7 and BaP2O6 compounds have been synthesized as microcrystalline powders by solid-state reaction under reducing atmosphere utilizing Eu2O3 as the dopant. Barium carbonate and the diammonium hydrogen phosphate were used as starting materials. The physical, optical, and scintillation 162

properties of these materials were evaluated by powder x-ray diffraction, photoluminescence, pulsed x-ray, and x-ray excited luminescence measurements. All of these Eu activated compounds show broad emission band with the peak center at about 420 nm. Of the four compounds, Ba3P2O8: Eu2+ is the most luminous with an estimated luminosity of 21,000 ph/MeV with two decay components at 270 ns @ 22% and 620 ns@78%. The Ba3P2O8 has about 2 times the scintillation light yield of BGO microcrystalline powder. A complete report of the measured photoluminescence and scintillation properties of each europiumdoped barium phosphate compound will be given.
References [1] W. J. Schipper, J. Hamelink and G. Blasse, Phys. Stat. Sol. A 1994, 141, 231. [2] T. Miyata, Y. Mochizuki, T. Minami, Thin Solid Films, 2006, 496, 174. [3] Y. Gao and C.S. Chi, Chin. J. Chem., 1996, 14, 399. [4] H. Liang, Y. Tao, Q. Zeng, H. He, S. Wang, X. Hou, W. Wang, Q.Su, Mat. Res. Bull. 2003, 38, 797. [5] S. H. M. Poort, J. W. H. van Krevel, R. Stomphorst, A. P. Vink, and G. Blasse, J. Solid State Chem. 1996, 122, 432. [6] I. Tale, P. Kulis, V. Kronghauz, J. of Luminescence, 1979, 20, 343. [7] C. C. Lago, J. Electrochem Soc. 1968, 1271; C. C. Lago, J. Electrochem Soc. 1970, 117, 1189. This work was supported by the U.S. Department of Homeland Security and carried out at the Lawrence Berkeley National Laboratory under U.S. Department of Energy Contract No. AC02-05CH11231 N25-42:

D. A. Bennett1, J. N. Ullom1, W. B. Doriese1, J. A. Beall1, G. C. Hilton1, R. D. Horansky1, K. D. Irwin1, V. Kotsubo1, L. R. Vale1, M. K. Bacrania2, A. S. Hoover2, N. J. Hoteling2, P. J. Karpius2, M. W. Rabin2, C. R. Rudy2, D. T. Vo2 1 Quantum Electrical Metrology Division, National Institute of Standards and Technology, Boulder, CO, USA 2 Los Alamos National Laboratory, Los Alamos, NM, USA

A 256 Pixel Array of Superconducting Gamma-Ray Microcalorimeters with Improved Readout Circuitry

Superconducting microcalorimeters are a promising technology for the nondestructive assay of nuclear materials by gamma-ray spectroscopy. We have recently performed measurements with several tens of gamma-ray microcalorimeters that demonstrate remarkable ability to distinguish between emission from 235U and 226Ra and to clarify complex Pu spectra. However, the defining challenge for gamma-ray microcalorimeter technology is to increase the size of pixellated arrays in order to combine unmatched spectral resolution (as good as 22 eV at 100 keV) with array efficiencies and count rates comparable to semiconductor detectors. We present an analytic model for microcalorimeters that includes the thermal properties of the attached absorber and the large inductance in the TES circuit bias. We show how the model can be used along with detailed pixel characterization to maximize the number of sensors multiplexed into a single readout channel and to minimize the response time of individual sensors. We also show improved multiplexer switching speeds. Finally, we present preliminary results from a 256 pixel array with 5 square cm active area.
N25-43:

FRAMEWORK and STRATEGIES for QUALITY ASSURANCE and RISK ASSESSMENT in the KM3NeT NEUTRINO TELESCOPE PROJECT

C. Sollima INFN - Universita' di Pisa, Pisa, Italy On behalf of the KM3NeT Consortium KM3NeT is an undersea neutrino detector currently being designed. The submitted paper will describe the framework and the strategy for the introduction of the quality management system (QMS) and risk assessment (RA) into the KM3NeT project. QMS and RA are set up as an integrated system for the improvement and optimisation of components of the KM3NeT neutrino detector as well as its production and operation.
N25-44: The Thermoluminescence Response of K2GdF5 Crystals Doped with Dy3+ Trivalent Ions to X, Gamma and Neutron Radiation Fields

E. C. Silva1, N. N. M. Khaidukov2, E. Vilela3, L. O. D. Faria1 1 Servico de Materiais e Combustvel Nuclear, Centro de Desenvolvimento da Tecnologia Nuclear - CDTN, Belo Horizonte, Brazil 2 Institute of General and Inorganic Chemistry, Moscow, Russia 3 Div. de Dosimetria, Centro Regional de Cincias Nucleares do Centro Oeste - CRCN-CO, Recife, Brazil

Recently, K2YF5 crystals singly doped with rare earth ions (RE), e.g. Ce3+, Tb3+, Dy3+ or Tm3+, have been shown to be attractive TL materials for detecting and discriminating different types of radiation. The investigation of such materials containing high concentrations of optically active RE ions is a promising direction for the developing of novel TL phosphors by taking into account that trivalent RE ions can efficiently capture electrons and/or holes and can be simultaneously recombination and luminescent centers. In this context, recently, we have demonstrated that K2YF5 crystals doped with 10.0 at. % Tb3+ and doped with 1.0 at. % Dy3+ have high TL sensitivity to photon radiation fields (Rad. Prot. Dosim. 112, pg 435, 2004 and Rad. Measurem., v. 42, pg. 311-315, 2007) Encouraged by these results and taking into account that neutron dosimetry is widely performed for personnel working with neutron sources in a wide range of situations such as medicine, industry, research and 163

agriculture, we now propose an investigation of the TL response of a new material, i.e., the K2GdF5:Dy3+ crystals, to ionizing radiation fields. The crystals doped with 5.0 at.% of Dy3+ ions have been discovered to be very sensitive to photon and thermal neutron radiation fields. After deconvolution of the gamma TL output, the data lead us to conclude that the higher temperature peak (234.2 0C), which is present in the doped sample and absent in the undoped one, is probably originated by the addition of Dy3+ ions and is responsible for the higher TL response of the doped sample. The X-rays sensibility for effective energy of 52.5 keV was found to be about 15 times higher than Cs-137 energy. Concerning the TL response for neutron fields, the samples showed a very high response for thermal neutrons, comparable to that of TLD-600. These results indicate that K2GdF5 crystals are very promising TL materials to neutron-gamma dosimetry.
The authors acknowledge the financial support from Brazilian government agencies CNPq and FAPEMIG. N25-45:

A Digital Coincidence Measurement System Using FPGA Techniques

S. C. Hsieh, H. P. Chou Institute of Nuclear Engineering and Science, National Tsing Hua University, Hsinchu, Taiwan The present work is to use field programmable gate arrays to develop a digital coincidence measurement system. A digital approach is also employed for pulse height measurements by converting the pulse height to pulse width and measuring with a digital counter. Calibration has been performed with a pulse generator as well as gamma sources using and comparisons was also made with conventional analog systems. Adjustment of hardware settings encountered in analog system is replaced by changing the software inputs easily. The proposed method provides a simple and flexible system and can be incorporated easily with a personal digital assistant as a portable spectrometer for isotope assaying in field applications.
The work is under the auspices of National Science Council, Taiwan. N25-46:

Computation of Neutron Multiplicity Statistics Using Deterministic Transport

J. K. Mattingly Sandia National Laboratories, Albuquerque, New Mexico, USA Nuclear nonproliferation efforts are supported by measurements that are capable of rapidly characterizing special nuclear materials. Neutron multiplicity counting is frequently used to characterize the bulk properties of fissile materials, including neutron source strength, multiplication, and lifetime. Different classes of model have been used to estimate these and other properties from the measured neutron multiplicity counting distribution. We have implemented a technique that employs deterministic neutron transport modeling. It is substantially faster than methods based on Monte Carlo models, and it does not suffer from the restrictions inherent to point models.
N25-47:

Studies on the Effect of Heat Treatment on Reusability of the CaSO4:Dy Teflon Disc TL Dosimeters

G. Chourasiya, A. K. Bakshi, S. Chatterjee, S. Kumar Radiological Physics & Advisory Division, Bhabha Atomic Research Centre, Mumbai, Maharastra, India Thermoluminescent dosimeter based on CaSO4:Dy TL phosphor-embedded Teflon discs are being used in India. It offers a number of advantages like reusability over a number of cycles by simply annealing it to a temperature of 230C for 4 hours after the TL readout. This erases the residual TL and restores the TLD disc to the background level. This characteristic of TLDs makes it most economical of all the dosimeters in use. However, during the repeated reuse of CaSO4:Dy Teflon disc over a certain number of cycles, it develops some coloration, bulging, and deformation, which adversely affects its dosimetric performance. It was therefore, considered worthwhile to explore the feasibility of salvaging the abandoned dosimetric discs and extend their life, by subjecting them to a high temperature treatment and to study changes in their dosimetric characteristics so as to explore the possibility of their reusability. In the present study the dysfunctional discs were cleaned with 6N NaOH followed by rinsing with distilled water. Subsequently, discs were cleaned with acetone and dried. These TL discs were sandwiched between two glass plates and were subjected to heat treatment at 400C for one hour in an air circulating oven. Optical density of the discs of experimental and standard TLD cards was measured with a densitometer. It is seen that the optical density decreases by about 37% after the thermal treatment. Further it has been observed that there is no change in the glow curve structure after 400 C thermal treatment and there is a marked improvement in the TL sensitivity which increases by a factor of two. Hence it is concluded that the heat treatment restores the TL sensitivity and the discs can be put back into use. This study helps to extend the life of dosimeters for much longer period of use and thereby saving enormous cost without compromising on the quality of dosimetric results in terms of sensitivity, accuracy and precision.

164

N25-48:

G. Su1,2,3, Z. Zeng1,2,3, L. Liu1,2,3, J. Cheng1,2,3 1 Department of Engineering Physics, Tsinghua University, Beijing, China 2 2. Key Laboratory of Particle & Radiation Imaging (Tsinghua University), Ministry of Education, Beijing, China 3 3. Key Laboratory of High Energy Radiation Imaging Fundamental Science for National Defense, Beijing, China Recently,CZT detectors are being developed by many researchers to measure the radiation dose and to identify radionuclides. This article explores the in-situ spectrometer made of CZT which is to be used in seawater.Monte Carlo simulation is done to calculate the energy spectra of the in-situ CZT spectrometer dipped in seawater for identification of artificial radionuclides. A variety of photons from main natural and artificial radionuclides were simulated using MCNP codes for CZT detector and for NaI detector as comparison. The CZT detector presented outstanding energy resolution in energy range below 1000keV.The results of various photons from main natural radionuclides were summed together according to their respective amounts and then a background spectrum was obtained.This article analyzes the effective volume of seawater and reveals that the sphere of seawater centering the detector, with radius of about 30cm, contributes mostly to the measurement. The detection efficiency defined as full energy peak counts per particle entering the detector is 2.23% for CZT, comparing to 22.7% for NaI detector. This paper provides an analysis of minimum detectable activity concentration (MDAC) of artificial radionuclides. The MDAC result is 0.42 Bq/L for CZT when time is 100 hours, and 0.14 Bq/L for NaI when time is 1 hour. This work would be helpful to develop in-situ spectrometer used in seawater.
N25-49:

Monte Carlo Simulation of In-situ CZT Spectrometer in Seawater

H. Iwai1, T. Iguchi1, H. Tomita1, J. Kawarabayashi1, M. Isobe2, C. Konno3 1 Quantum Engineering, Nagoya University, Nagoya, Aichi, Japan 2 National Institute for Fusion Science, Toki, Gifu, Japan 3 Japan Atomic Energy Agency, Naka, Ibaraki, Japan

Development of 2.5MeV Neutron Spectrometer for Helical DD Plasma Experiment

The ion energy distribution in magnetic confinement fusion (MCF) plasma depends on the heating methods applied, such as neutral beam injection and RF heating. With heating of deuterium plasma, neutrons with an energy of 2.5 MeV are emitted as accompanying products of DD fusion reaction. The detailed studies on the neutron spectra can serve as a valuable diagnostic tool for the reacting deuterium ions. For the measurement of neutron spectra in helical MCF deuterium plasma, we are developing a high-energy-resolution neutron spectrometer based on coincident detection of the scattered neutron and recoiled proton from a plastic scintillator as the incident neutron target (called a radiator). The neutron energy is derived from sum of the deposit energy in the radiator, the recoiled proton energy measured by a Si SBD and the scattered neutron energy measured by a time-of-flight technique. We have made a prototype of the DD neutron spectrometer and experimentally evaluated the detector response by using the accelerator DD neutron source To achieve the high energy resolution for DD neutron, we propose to extract the events that have low energy deposition inside the radiator from all coincident events in the spectrometer. We will show an improvement in the energy resolution through this data extraction method and discuss an applicability to the helical MCF DD plasma experiment.
N25-50:

Development of Multiple Activation Counter Using Liquid Light-guide for Pulsed Neutron Dosimetry

J. Kawarabayashi, Y. Sato, M. Hayashi, H. Tomita, S. Maeda, H. Tsuji, T. Iguchi Quantum Engineering, Nagoya University, Nagoya, JAPAN In next-generation nuclear facilities such as international thermonuclear experimental reactor (ITER) and spallation neutron source (SNS or J-PARC), high intense pulsed neutrons will be generated and the neutron dosimetry in such facilities is important for safe and efficient operation. Conventional detectors are not suitable for dosimetry of pulsed neutron sources because pulse pile up in these detectors leads to considerable degradation of reliability by the neutron counting loss. Alternatively, neutron activation analysis (NAA) is applied for the dosimetry of neutron burst. On conventional activation counters, a plastic scintillator or a proportional counter is used as the detector. However, for solid-state detectors, radiation damage caused by intense neutron burst is not negligible. In order to apply the activation counters to the intense neutron burst, we have proposed a novel activation counter using a liquid light-guide (LLG) and a suitable group of activation materials. In this paper, we evaluate the counting efficiencies of several activation foils which are necessary to calculate the neutron dose from the counter responses and discuss the efficiency trend for the maximum energy of the beta-particle. The experimental detector responses were successfully obtained with the fast neutrons and 14MeV D-T fusion neutrons. From these results, the detection efficiency increases exponentially at beta decay energy of 1.5 to 3.5MeV. This trend is explained by the fact that the detection efficiency of light in LLG increases when photons are emitted in opposite direction, and the angle of emitted Cherenkov photons and incident particle are getting right angle when the energy of the particle increases. From this result, the detector counts can be converted to the beta activity with other activation material such as Cr, Co and Li.

165

N25-51:

C.-H. Baek1,2, S.-J. Lee1,2, J. Y. Hwang1,2, Y. Choi3, Y. H. Chung1,2 1 Department of Radiological Science, College of Health Science, Yonsei university, Wonju,220-710, South Korea 2 Institute of Health Science, Yonsei University, Wonju,220-710, South Korea 3 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul,135-710, South Korea Pinhole imaging is receiving notice as a method to enable gamma camera for nuclear survey. In our previous work, the pinhole gamma camera with depth of interaction (DOI) based on three layers of monolithic crystals and maximum-likelihood positionestimation (MLPE) algorithm was designed for the environmental monitoring and the methodology for 3D event positioning in the detector was established by using Monte Carlo simulation. The aim of this study was to evaluate the performance of our detector module experimentally. The proposed detector module was developed by stacking three layers of monolithic CsI(Tl) crystals, each of which has a dimension of 50.0 x 50.0 x 2.0 mm3. The bottom surface of the third layer was directly coupled to an 8 x 8 channel position-sensitive photomultiplier tube (PSPMT, Hamamatsu H8500C). The PSPMT was read out using a resistive charge divider, which multiplexes 64 anodes into 8(X) + 8(Y) channels. Gaussian-based MLPE methods have been implemented by using experimentally measured detector response functions (DRFs). The results demonstrated that the detector module could identify the position of the gamma ray interaction inside the crystal with a resolution of 2.0 mm in all three directions. In this paper, our new detector proved to be a reliable design to characterize the event position in all three dimensions with high and uniform spatial resolution and the large-angle pinhole camera could be a useful tool in the environmental monitoring.
N25-52:

Development of a Large-Angle Pinhole Gamma Camera Using Depth-of-Interaction Detector for Nuclear Survey

TeO2 Thin Film Based Real-Time Gamma Dosimeters for High Doses

S. L. Sharma, T. K. Maity Physics and Meteorology, Indian Institute of Technology, Kharagpur, Kharagpur-721 302, INDIA In order to develop real-time gamma dosimeters for high doses, we have studied in detail the gamma radiation induced changes in the structural, optical and electrical properties for the thin films of tellurium dioxide (TeO2) of thicknesses in the range 3001500 nm. These thin films were prepared by thermal evaporation in vacuum. The XRD patterns show that the as deposited and gamma radiation exposed thin films of all thicknesses are amorphous in nature. The FESEM images show that (a) the as deposited thin films of all thicknesses are non-uniform, (b) the exposed thin films of all thicknesses become more and more uniform as the gamma radiation dose is increased and (c) formation of large size clusters takes place in the exposed thin films at higher doses. Optical measurements for the as deposited and exposed thin films show that the optical band gap decreases with the increase of gamma radiation dose up to a certain value. The current versus voltage plots for the as deposited and exposed thin films show that the current increases near linearly with the gamma dose over a wide range of doses at different applied voltages. Further, the range of doses is higher for the thicker films. The near linear increase of the current with the gamma radiation dose can partly be attributed to the decrease of the optical band gap with the gamma radiation dose. Employing the dose dependence of the current, we have designed and fabricated real-time gamma dosimeters based on TeO2 thin films of thicknesses 300, 450, 600, 900, 1200 and 1500 nm and these have been found to possess sensitivity in the range 6-38 mA/cm2/Gy.
N25-53:

K. K. Mitev1, D. S. Pressyanov1, V. N. Zhivkova2 1 Dep. of Atomic Physics, Sofia University "St. Kl. Ohridski", Sofia, Bulgaria 2 Dep. of Radiochemistry, PP1, "Kozloduy" NPP, Kozloduy, Bulgaria

New Sensitive Technique for Measurement of Krypton-85 Based on Absorption in Polycarbonates and Liquid Scintillation Counting

The objective of this work is to present the liquid scintillation counting (LSC) as a new technique to be used within the polycarbonate method for measurements of Kr-85 activity concentrations in air or water. Currently, the polycarbonate method employs the high absorption ability of some polycarbonates to noble gases for sampling Kr-85 from the environment and then the absorbed activity is measured by gross beta counting or gamma spectrometry. This work explores the applicability of LSC for measurement of Kr-85 beta particles emitted from the exposed polycarbonates. Experiments with exposure of polycarbonates to Kr-85 in water and subsequent LSC counting are presented. The results from these experiments demonstrate good linear correlation between the measured net LS counting rate and the activity concentration of Kr-85 during the exposure. The usage of the proposed technique leads to approximately 10 times better sensitivity for Kr-85 compared to gross beta counting or gammaspectrometry of the exposed polycarbonates. A particularly advantageous feature that has been identified in the present work is that the polycarbonate material is transparent for the LS cocktail light, thus adding polycarbonate in the LS cocktail do not result in significant adverse effect on the light output. The results from the conducted experiments imply that quantitative measurements of Kr-85 activity concentrations in air or water by LS counting of polycarbonates are feasible. The technique is applicable for monitoring of Kr-85 concentrations in air or water in nuclear power plants (including at-reactor pools), spent fuel 166

storage facilities and reprocessing plants. It allows simultaneous monitoring of Kr-85 concentrations in a large number of points in these facilities.
N25-54:

Performance Optimization of Cosmic-Ray Charged Particle Detector Using Wavelength Shifter Fiber

S. Y. Kang, D. Y. Jang, C. H. Lee, S. M. Kang, B. H. Kang, Y. K. Kim Nuclear Engineering, Hanyang University, Seoul, South Korea Since the plastic detector is easy to fabricate and has good property, it has been used as various forms of cosmic-ray detectors. Major factors to determine the detectors efficiency are size of plastic scintilltor, scintillation transfer material, and the photo acceptor equipment. Size of plastic scintillator and photo acceptor equipment (Photo multiflier tube, Photodiode, etc.) can be decided simply in the design process but scintillation transfer material has to be selected circumspectly. Light guide using Polymethylmethacrylate (PMMA) has very high transmittance, but it is large size and is very expensive. While a wavelength shifter fiber can transfer scintillation by several quantities, but the transfer efficiency is not so good. Therefore, detector using wavelength shifter fiber requires a proper mounting method for maximum efficiency. In this study, three detectors of 200 70 10 mm3 sizes were fabricated. And the wavelength shifter fibers were mounted by three different ways and the performance of charged particle detection was evaluated.
Acknowledgments : This work was supported by the Basic Atomic Energy Research Institute (BAERI), nuclear R&D program of MEST, Korea, and also supported by the Ministry of Knowledge Economy (2008-P-EP-HM-E-06-0000). N25-55: Sequential Probability Ratio Test Using Scaled Time-Intervals for Environmental Radiation Monitoring

P. Luo, T. A. DeVol Environ. Eng. and Earth Sciences, Clemson University, Anderson, SC, USA Sequential probability ratio test (SPRT) of scaled time-interval data (time to record N radiation pulses) was evaluated against commonly used single-interval test (SIT) and SPRT with a fixed counting time on experimental and simulated data. Experimental data were acquired with a DGF-4C (XIA, Inc) system in list mode. Simulated time-interval data were obtained using Monte Carlo techniques to perform a random radiation sampling of the Poisson distribution. The three methods were compared in terms of the detection probabilities and the average time to make a decision. For both experimental and simulated data, SPRT with scaled time-intervals provided a similar detection probability as other two tests, but is able to make a quicker decision with fewer pulses at relatively higher radiation levels. The scaled time-intervals SPRT has a provision for varying the sampling time depending on the radiation level, which may further shorten the time needed for radiation monitoring. Parameter adjustments to scaled time-intervals SPRT method for increased detection probability for the same sampling time are presented.
This research is funded in part by a DOE, Office of Biological and Environmental Research Grant No. DE-FG02-07ER64411, entitled "Radionuclide Sensors for Subsurface Water Monitoring" N25-56:

Isotopic Abundance Analysis Using MGA on Spectra Generated by Monte Carlo Simulation

H. Yang, W. Russ, R. Venkataraman, A. Bosko Canberra Industries, Meriden, CT, USA Monte Carlo simulation has been widely used in designing and benchmarking proposed nuclear measurement systems. By carefully modeling of detector response and counting geometry, accurate gamma ray spectra can be generated using Monte Carlo method given specific source information. This work demonstrates this approach as a useful tool in design of waste assay systems using gamma spectroscopy technique. MGA is a gamma ray analysis program for determining the isotopic abundances of plutonium and other actinides in a wide variety of samples. Using advanced fitting technique, MGA is able to analyze the Pu abundance based on the more complicated but more intense region (94 keV 104 keV). Spectral Forge is a full-featured spectroscopic analysis utility developed at Canberra. It can be used to acquire data, perform nuclide identification, predict and evaluate system performance and to generate archive and reference system performance information. Spectral Forge can calculate peak and scattering response functions for a specific detector in a specific counting geometry, based on MCNP simulation results. Spectrum can then be generated given the detector energy and shape calibration and source information. In this work, we accurately modeled a Canberra REGe 1519 HPGe detector inside a slab counter. Spectra were generated via Spectral Forge and then analyzed in MGA. MGA reported isotopic abundance was then compared with input values.

167

N25-57:

E. C. Miller1, S. D. Clarke1, S. A. Pozzi1, J. K. Mattingly2 1 Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA 2 Sandia National Laboratories, Albuquerque, NM, USA Efforts to prevent the proliferation of special nuclear materials rely in part on methods capable of characterizing those materials. One method that can be used to quickly characterize an unknown fissile sample is neutron multiplicity analysis. This method characterizes fissile material by detecting neutrons correlated by fission chain-reactions, whose temporal and number distribution are distinct from spontaneous neutron emissions. This is a well established assay in the field of nuclear materials, control and accountability. The objective of our current work is to benchmark the ability of the MCNP-PoliMi code to simulate the neutron multiplicity distributions measured from a heavily moderated fissile system using 3He proportional counters. To perform this benchmark, simulations of a polyethylene reflected plutonium sphere are being performed using MCNP-PoliMi. The subject of the experiment was a 4.483-kg plutonium sphere that was covered with various thicknesses of a polyethylene reflector/moderator. The simulated data is post processed to include the effects of dead time of the 3He tubes and their associated electronics. The post processor then applies a shifting window at regular, non-overlapping intervals to accumulate the number of detection events versus window width. The Feynman Y values are calculated for each window width. MCNP-PoliMi is able to provide a close estimate of the Feynman-Y in this case, within 5% of the experimental value. Further simulations are being performed to simulate this system in the more heavily moderated cases.
N25-58:

Plutonium Sphere Multiplicity Simulations with MCNP-PoliMi

N. Hoteling1, M. K. Bacrania1, A. S. Hoover1, P. J. Karpius1, D. T. Vo1, D. A. Bennet2, J. N. Ullom2, W. B. Doriese2, K. D. Irwin2, M. W. Rabin1 1 Los Alamos National Lab, Los Alamos, NM, USA 2 National Institute of Standards and Technology, Boulder, CO, USA

New Measurements and Lessons Learned with High-Resolution Microcalorimeter Detector Arrays

Cryogenic microcalorimeter detectors have emerged as a new and innovative technology for X-ray and gamma-ray detection. Operated at temperatures below 100 mK, these detectors have achieved single-pixel energy resolution down to 22 eV at 103 keV, a factor of 20 better than commercial high-purity germanium detectors. With these powerful new tools, one is able to completely resolve peaks separated by less that 1 keV. For instance, our group has recently demonstrated for the first time a complete separation of 235U and 226Ra gamma rays at 185.7 and 186.2 keV, respectively. In this presentation, we will highlight this and other recent measurements, including spectra from a calibrated Pu source, where the separation of isotopic signatures is important to a quantitative analysis, and minor actinide-mixed oxide material where a complex arrangement of peaks makes analysis with conventional detectors extremely difficult. In addition, we will address practical issues in the data analysis such as energy calibration and automation.
N25-59:

Assessing the Lower Bound on Uncertainty in Ultra-High Resolution Microcalorimeter Isotopic Ratio Measurements

P. J. Karpius On behalf of the LANL-NIST Microcalorimeter Collaboration The unprecedented energy resolution of cryogenic microcalorimeter detectors relative to conventional high-purity germanium (HPGe) detectors gives promise for significant advancement in quantitative analysis of gamma-ray spectra. With ever increasing detection efficiency over an expanding dynamic energy range we begin to address the issue of the lower bound on the uncertainty of isotopic ratio measurements in the infinite-count limit. A primary aspect of this is the determination of the form of the function describing the relative efficiency for detection over the range of energy containing those gamma-ray peaks of interest. In addition to generally well-known branching ratios, the determination of relative efficiency depends ultimately on extracted gamma-ray peak areas, which in turn depends on proper background subtraction and peak fitting performance. It is often the case that data taken with HPGe detectors yield peaks required for such analyses that overlap adjacent peaks. The order-of-magnitude improvement in energy resolution of microcalorimeter detectors over HPGe systems will mitigate, and in many cases virtually eradicate, the error associated with peak overlap in the determination of relative efficiency and ultimately isotopic ratios. We examine the potential magnitude of this improvement based on current data from the LANL-NIST multi-pixel array microcalorimeter system and attempt to quantify it in the extension to the far greater statistics that will be achievable in future generations of arrays.

168

M. W. Cooper1, T. W. Bowyer1, D. A. Haas1, J. Forrester1, L. Lidey1, J. I. McIntyre1, R. Payne1, R. Thompson1, R. Williams1, B. Hosticka2, A. Faanhof3, K. Elmgren4, A. Ringbom4, M. Aldener5, N. Wilson6, P. Saey6 1 Pacific Northwest National Laboratory, Richland, WA, USA 2 University of Virginia, Charlottesville, VA, USA 3 NECSA, Pelindaba, South Africa 4 FOI, Stockholm, Sweden 5 Gammadata, Uppsala, Sweden 6 CTBTO, Vienna, Austria Radionuclide monitoring is a technique used by the International Monitoring System (IMS) to detect and provide evidence of potential nuclear explosions. Due to the complexities inherent in the analysis it is extremely important to know the back ground terms around the world. As part of that goal, the EU Project has been initiated to determine background levels of radioxenon at locations that do not currently have noble gas sampling stations. The EU Project has deployed a SAUNA noble gas detection system to various places around the world, and the data collected at Mafikeng, South Africa and Cape Point, South Africa is presented in this paper. Atmospheric transport modeling has been performed for each data set that has detected a radioxenon event.
N25-61:

N25-60:

Radioxenon Measurements in South Africa

Upgrade of Mini-DDL Applied in the Radiation Field Measurement

X. Han, M. Zeng, B. Shao, T. Xue Department of Engineering Physics, Tsinghua University, Beijing, China Mini-DDL is a specific device designed for radiation field real-time monitoring, which has already been applied in NSRL, SSRF and BEPC. Although these applications have proved this device to be functional and helpful, the original version still has some disadvantages. Based on brief review of the previous Mini-DDL, this paper introduce technologies which are applied to upgrade the Mini-DDL, including the OLED display, the GSM module ,as well as the FATFs module. The new version of Mini-DDL will offer more flexible data transfer interfaces, larger and various storage capacity, and so on.
N25-62:

Stability of X-Ray Tube-Based Transmission Source for UF6 Gas Enrichment Monitoring Technology

K. D. Ianakiev, J. M. Goda, T. R. Hill, C. W. McCluskey, C. E. Moss, H. Nguyen, R. F. Parker, M. T. Swinhoe N Division, Los Alamos National Laboratory, Los Alamos, NM, USA Measuring the 235U enrichment of UF6 gas in-line is one of the pressing technical issues for international safeguards at gas centrifuge enrichment plants (GCEPs). The classical method based on of the transmission measurement of UF6 gas density combined with gamma spectroscopy of the 186-keV line of 235U is still the only one proven technology for non invasive measurement of UF6 gas density through a few millimeter thick aluminum pipe. In order to achieve reasonable sensitivity, the 22 keV Ag X-ray line of decaying isotope Cd-109 was the only choice for a transmission source for low UF6 gas density. The major problem was the high level of maintenance necessary for periodic recalibration and replacement of the decaying isotope source. We proposed to replace the decaying isotope source with an X-ray tube and a notch filter. The initial results from a laboratory prototype have shown sensitivity similar to that of a Cd-109 source and great design flexibility for the energy and intensity of the transmission source, but at the price of higher complexity to handle the inherent instability of X-ray generating devices. In this work we discuss the stability issues affecting accuracy of X-ray based transmission sources and different measures to reduce their effect (improving stability of the HV power supply, radiation flux monitoring, etc). Increasing the voltage difference between the K-edge of the notch filter material and the bremsstrahlung cutoff voltage reduces sensitivity to high voltage variations and flux monitor tracking error. However, operating at higher voltages and higher power increases complexity and reduces the reliability of the X-ray tube and HV power supply. The overall design of the transmission source will be discussed as a complex tradeoff between operational characteristics of the X-ray generator and parameters of the notch filter.
N25-63:

Development of a Model of an X-Ray Tube Transmission Source

J. M. Goda, K. D. Ianakiev, C. E. Moss N Division, Los Alamos National Laboratory, Los Alamos, NM, USA An enrichment monitor for gaseous UF6 in a pipe or container requires a measurement of the 186-keV gamma ray from 235U and a measurement of the attenuation of a transmission source by the UF6. In support of the development of an x-ray tube based source for transmission measurements of UF6 gas, we have developed a spreadsheet-based model of the spectra involved. Starting with the spectrum produced by an x-ray tube we apply the linear attenuation coefficients for various notch filters, the aluminum pipe, and UF6 gas. This model allows calculation of the transmitted spectrum based on the type of filter, the thickness of the filter, the x-ray tube high voltage, the Al pipe thickness, and 169

the UF6 gas pressure. The sensitivity of the magnitude of the transmission peak produced by the notch filter to any of these variables can be explored quickly and easily to narrow the choices for experimental measurements. To validate the spreadsheet based model, comparisons have been made to various experimental data.
N25-64: A Hybrid Pulse Shape Discrimination Technique with Enhanced Performance at Neutron Energies below 500 keV

S. D. Ambers, M. Flaska, S. A. Pozzi Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, Michigan, USA Neutron and gamma-ray pulse shape discrimination (PSD) using organic scintillation detectors is a widely adopted technique in fields such as nuclear nonproliferation, international safeguards, nuclear material control and accountability, and national security. In contrast to thermal neutron detectors such as He-3 tubes, organic scintillators are able to detect high-energy neutrons and do not need to use moderating material. At the same time, these detectors are sensitive to gamma rays, which makes them very suitable for measurements in mixed neutron/gamma-ray fields. A large number of papers were published on PSD performance of liquid scintillators. In this paper, we present a new PSD approach that is based on detailed knowledge of the detector response to a given radiation. Specifically, an average detector response is obtained for several energies of interest, which is later used as a reference for particle identification. The main objective of this work is to show that there is potential for improvement of current PSD performance at neutron energies below 500 keV with commercially available measurement equipment. The average-pulses PSD will be utilized at energies below 500 keV while a traditional charge integration PSD will be used at higher energies. This hybrid PSD approach will provide a higher level of accuracy at low energies.
N25-65:

C. Boiano1, F. Cappuzzello2, M. Cavallaro2, A. Cunsolo2, A. Foti3, P. Guazzoni1, S. Moser1, S. E. A. Orrigo2, F. Riccio1, M. R. D. Rodriguesb2, L. Zetta1 1 Fisica, Universit di Milano and INFN, Milano, Italy 2 Fisica, Universit di Catania and INFN-LNS, Catania, Italy 3 Fisica, Universit di Catania and INFN, Catania, Italy

Digital Signal Processing Applied to the Position Start Detector of the MAGNEX Spectrometer

The capability of the digital pulse shape technique to acquire data from the focal plane detector of MAGNEX spectrometer in order to measure the position coordinates using digital signal processing was already demonstrated in one of our previous work. This contribution presents the preliminary results obtained illuminating the Position sensitive Start Detector (PSD) with an alphasource. The multiplexed analog signals produced by the preamplifier boards based on GASSIPLEX chips are sampled with a sampling analog to digital converter, at 50 MHz sampling frequency and 14-bit resolution. The obtained histograms are then compared with those obtained at the same conditions with the traditional approach used for the detector. They are fully comparable to those obtained with the commercial VME-multiplexed readout boards, leaving the additional advantage to apply possible optimum algorithms for data processing without any further hardware change.
N25-66:

F. Amorini1, A. Anzalone1, C. Boiano2, G. Cardella3, A. Castoldi4, S. Cavallaro1, E. De Filippo3, E. Geraci3, L. Grassi3, C. Guazzoni4, P. Guazzoni2, E. La Guidara3, I. Lombardo1, S. Moser2, A. Pagano3, S. Pirrone3, G. Politi3, F. Porto1, F. Riccio2, F. Rizzo1, P. Russotto1, L. Zetta2 1 Dipartimento di Fisica e Astronomia, Laboratori Nazionali del Sud and Universita' di Catania, Catania, Italy 2 Dipartimento di Fisica, Universita' degli Studi di Milano and INFN, Milano, Italy 3 Dipartimento di Fisica e Astronomia, Istituto Nazionale di Fisica Nucleare and Universita' di Catania, Catania, Italy 4 Dip. Elettronica e Informazione, Politecnico di Milano and INFN, Milano, Italy In intermediate nuclear physics experiments, the CHIMERA multidetector represents the most powerful apparatus with its capability of detecting 94% of the total solid angle and measuring energy and identifying mass and charge of the reaction products. Several identification techniques can be used for extracting mass and charge information on the reaction products, depending on the fact that they punch through, or not, the silicon detectors. We are fixing our attention to the products stopping in the silicon detector that can be identified in mass by measuring Energy and Time of Flight - (E, ToF) technique - and in charge by measuring the Energy and the Rise Time of the signal - (E, RT) technique. With this last technique, one of the points of largest physical interest is the minimum energy of the reaction products that allows charge identification. To this purpose we have done some tests, using the ion beam of the Superconducting Cyclotron of Laboratori Nazionali del Sud in Catania. We have compared the results obtained considering only the reaction products that stop in silicon using, respectively, the (E, ToF) and (E, RT) techniques, using signals from silicon detectors mounted in front and in reverse mode. The aim of this contribution is two-fold. On one side we will discuss the experimental results obtained in reverse mounting in a true beam experiment and we will critically analyze the advantages and drawbacks of reverse mounting w.r.t. front mounting. On the other side we will correlate more deeply the measured detector properties in terms of physical properties (including noise of the detector + preamplifier 170

CHIMERA Silicon Detectors in Reverse Mode: Preliminary Tests

system) and biasing conditions to the measured results, in order to understand the physical reason of the different values of the energy thresholds in reaction product identification for front and reverse mounting.
N25-67: A New Detector for Track Reconstruction and Identification of Charged Fragments particles from the Proton to Uranium

P. Legou1, M. P. Combet1, F. Nizery1, R. de Oliveira2 1 DSM\IRFU, CEA Saclay, Gif sur Yvette, FRANCE 2 PCB Workshop, CERN, geneve, Switzerland

We are working on a Time Projection Chamber to be placed at GSI downstream after the future GLAD magnet in the R3B hall experience. The detector would be able to detect with a high efficiency charged fragments from the proton up to the uranium. This detector will be capable to detect track events with a clear identification and a reconstruction of each track. To fit with these requirements, we made the choice of a gaseous detector called micromegas used in TPC mode. We made three tests beams at GSI. We have tested in a hign intensity beam, during last april the very new resistive anode detector. We have seen already very encouraging results concerning the space resolution and spark protection. In order o read the detector we developped a very fast new front-end electronics card with a spark protection.
N25-68:

T. Bressani1,2, M. Agnello3,2, E. Botta1,2, M. Bruschi4, S. Bufalino1,2, M. De Napoli5,6, A. Feliciello2, A. Fontana7, B. Giacobbe4, L. Lavezzi8,9, G. Raciti5,6, E. Rapisarda5,6, A. Rotondi7,9, C. Sbarra4,10, C. Sfienti5,6, A. Zoccoli4,10 1 Dipartimento di Fisica Sperimentale, Universita' di Torino, Torino, Italy 2 Sezione di Torino, INFN, Torino, Italy 3 Politecnico di Torino, Dipartimento di Fisica, Torino, Italy 4 Sezione di Bologna, INFN, Bologna, Italy 5 Dipartimento di Fisica, Universita' di Catania, Catania, Italy 6 Sezione di Catania, INFN, Catania, Italy 7 Sezione di Pavia, INFN, Pavia, Italy 8 Dipartimento di Chimica Generale, Universita' di Pavia, Pavia, Italy 9 Dipartimento di Fisica Nucleare e Teorica, Universita' di Pavia, Pavia, Italy 10 Dipartimento di Fisica, Universita' di Bologna, Bologna, Italy The HPGe detectors are a milestone in experimental nuclear physics. However they are seldom operated in magnetic fields. Thus their behaviour in these conditions is not well known and to our knowledge only two studies are present in the literature [1,2]. In this work we have studied the performances of a commercial coaxial n-type HPGe detector manufactured by ORTEC in the very high magnetic field of the SOLE superconducting solenoid at the LNS-INFN and we have analyzed the effects of the field on the gamma-ray spectra from three different radioactive sources (141Am, 137Cs and 60Co). The results are new with respect to the previous studies for two aspects: first, the intensity of the magnetic field used, 2.5 T is the highest in which a HPGe detectors has been so far tested; second, the crystal axis is coaxial with the field axis. In addition, we explored the effects on the spectra obtained by varying the angle between the field and the crystal axes. The most important phenomenon expected in HPGe detectros operated in high magnetic fields is a deflection of the charge carriers trajectories due to the bending Lorentz force. This effect increases the signal rise time and modifies the shape of the full energy peaks in the spectra and can give, with particular trajectories of the carriers, a loss of charge with a reduction of the efficiency. A significant worsening of the energy resolution was found, but with a moderate loss of the collected charge and then of the efficiency: this is mainly due to a change in the rise time distribution of the pulse and confirms the previously published results. The most relevant features of the peak shapes, described by bigaussian fuctions, have been parametrized in term of field intensity and gamma energy: this allows to correct the spectra measured in magnetic field and to recover the energy resolution almost completely. [1] A. Sanchez Lorent et al. NIM A573 (2007) 410-417; [2] K. Szymanska et al. NIM A592 (2008) 486-492.
This work has been granted by Progetti di Ricerca di Interesse Nazionale 2005 of the Italian Goverment (Hypergamma Program)

On the Behaviour of HPGe Detectors Immersed in Magnetic Fields up to 2.5 T

171

N25-69:

Investigation of the Properties of Large Volume LaBr3:Ce Scintillators with High Energy Gamma

F. Camera1, N. Blasi2 1 Physics, University of Milano and INFN sect. of Milano, Milano, Italy 2 InFN sez. of Milano, Milano, Italy The properties of cylindrical 1 x 1 inch, 3 x 3 inches and 3.5 x 8 inches LaBr3:Ce scintillators coupled to different Photonis and Hamamatsu tubes have been investigated with gamma-rays with energy up to 15 MeV. The used photomultipliers were XP2060, R5300, XP5301, R6233, R6233-100, XP5700, XP3540 and R10233-100. The tests have been done using the voltage dividers provided with the phototubes and, in some cases, using a special designed voltage divider which is able to provide one preamplified signal from one of the firsts dinodes and one signal directly from the anode. The detectors have been analyzed in the energy range 100 keV - 15 MeV in terms of their energy resolution and linearity for different values of the supply voltage.
N25-70:

Rays

F. Camera1, N. Blasi2 1 Physics, University of Milano and INFN sect. of Milano, Milano, Italy 2 INFN sect. of Milano, Milano, Italy

Position Sensitivity of Large Volume LaBr3:Ce Detectors

The position sensitivity of two large volume LaBr3:Ce crystals has been measured using collimated beams of 140 keV and 662 keV gamma rays. The crystals have been coupled to a PSPMT and/or a shielded PMT. The used detectors have sizes of 1 x 1 inch and 3 x 3 inches. The results indicate that in the 3 x 3 inches crystal it should be possible to trace the interaction point of a 662 keV gamma ray within 1-2 cm while this does not seem possible for a 140 keV gamma ray. GEANT+SCIDRA simulations confirm the results and show that, as expected, the position sensitivity improves as the gamma ray energy increases. New tests with a 3.5 x 8 inches large crystal are in progress.
N25-71:

F. C. L. Crespi1, V. Vandone1, S. Brambilla2, F. Camera1, B. Million2, O. Wieland2, D. Bortolato3, F. Recchia3 1 University of Milan / INFN Milan, Milan, Italy 2 INFN sez. of Milan, Milan, Italy 3 University of Padova / INFN Padova, Padova, Italy

Characterization of a HPGe Segmented Detector Position Response Using Pulse Shape Comparison Techniques

In next-generation arrays for gamma spectroscopy with exotic beams like AGATA or GRETA the path of the gamma ray inside the active volume of segmented HPGe detectors will be fully reconstructed. The gamma ray tracking will allow to correct for the Doppler broadening effect and to reject the background events which do not deposit their full energy inside the array or do not originate from the target position. Pulse Shape Analysis techniques are used to localize the gamma interaction points with a 5 mm spatial resolution. This in turn requires the HPGe detector position response to be known. The use of pulse shape comparison techniques (Pulse Shape Comparison based Scan, PSCS) opened the possibility to characterize the full 3D position response of a HPGe segmented detector experimentally, in a reasonable amount of time. In this work the application of a variant of the PSCS technique on an asymmetric AGATA segmented HPGe detector is presented. This variant of the PSCS method requires no collimators, thus reducing even more the time duration of the measurements. Specific Geant4 simulations of an AGATA triple cluster with a 60Co source placed in different positions have been performed. The technique relies on the fact that the combination of the number of counts in each detector voxel is in principle unique, provided that a sufficient number of measurements with a source illuminating the detector from different directions (3 at least) is available. In this way the calculated number of counts per voxel can be compared with the number of events with a specific line-shape acquired in the 3 measurements and it will be therefore possible to associate each line-shape with a position, thus measuring the detector position response. In order to test this technique experimental pulse shapes have been acquired in a measurement performed at LNL-INFN with a 60Co source and an AGATA cluster composed of 3 asymmetric segmented HPGe detectors.
N25-72:

J. P. Martin1, C. Mercier1, N. Starinsky1, P. Garrett2 1 Physics, R.J.A. Levesque Laboratory, University of Montreal, Montreal, Quebec, Canada 2 Dept. of Physics, University of Guelph, Guelph, Ontario, Canada

A GHz 8-Channel Trigger/DAQ Module with Real-Time Digital Neutron-Gamma Discrimination.

Analog trigger systems featuring neutron-gamma discrimination in radiation detectors have been in usage for decades. They are all based indirectly on the difference in response of the detectors as a function of the ionization density of the incoming radiation. Various analog methods has been developed to enhance features from the detector signals that are characteristic of the two types of responses. Typically, for liquid scintillators, the resulting light output follows an average behaviour consisting of a sum of 172

exponentials of different time constants with amplitude coefficients ratios dependent upon the nature of the radiation. The analog methods process the detector signals in order to enhance certain electrical differences, such as peaking time or zero crossover in the time domain, or amplitudes ratios at different time intervals. For a large detector arrays, the analog approach is less appealing due to its complexity and delicate adjustment procedures. For the DESCANT auxiliary neutron detector array under development at the TRIUMF-ISACII-TIGRESS facility in Vancouver, Canada, a custom digital module has been designed to achieve this neutron-gamma trigger functionality. The module is based on 1 GS/Sec digitizers coupled to an FPGA based neutron-gamma discrimination algorithm. The digital algorithm evaluates in real time the amplitude coefficients associated with the various time constants associated with the light output decay, and makes the neutron-gamma identification accordingly. The trigger decision algorithm is pipelined and has latency less than 1 microsecond. The modules have a VME 6U 4HP form factor and hold 4 channels. They can be used either to generate neutron triggers on a fast LVDS serial link compatible with the TIGRESS trigger/DAQ architecture, or perform the same task in a standard VME environment. Of course, the module can also be used as basic multi-channel ADC's to simply collect signal samples at GS/Sec rates.
N25-73:

A. A. Naqvi1, F. Z. Khiari1, M. I. Al-Jarallah1, A. H. Isab2, K. U. Rehman1, M. Raashid1, A. U. Islam1 1 Physics, King Fahd University of Petrol, Dhahran, Saudi Arabia 2 Chemistry, King Fahd University of Petrol, Dhahran, Saudi Arabia The Prompt Gamma Neutron Activation Analysis Technique (PGNAA) has proven to be an efficient technique to determine concentration of C, O, H and N elements in low Z elements bulk samples for the detection of contraband. The elemental ratio C/O in low Z samples is used to distinguish between contraband and innocuous materials. The elements C N, and O are detected in neutron inelastic scattering (NIS) reactions using 6.0-14 MeV neutrons while Thermal Neutrons are used to detect H through Thermal Neutron Capture gamma rays. Therefore, an efficient detection of H, C, N and O in PGNAA setup requires a moderator geometry that produces a maximum fast/slow neutron flux ratio. A polyethlene moderator can be used in conjunction with 14 MeV neutron source to produce such mixed flux. Previously polyethlene moderators have been used successfully at King Fahd University to produce thermal neutrons flux using 2.8 MeV neutrons from D(d,n) reactions . In this study Monte Carlo simulations were carried out to optimize the design of a high density Polyethylene moderator to produce the desired mixed flux of fast and thermal neutrons at the sample location using a 14 MeV neutron source. In order to optimize the fast/slow neutron yield, channels with varying diameters and inclination angles were drilled through the moderator thereby providing paths of variable attenuation to the 14 MeV neutrons. The effect of the number, size, and inclination angles of the channels on the fast/slow yield ratio was studied. The results of these studies will be discussed.
Acknowledgement: This study is part of Research Project No.IN090033 funded by King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia N25-74:

Optimization of Fast/Slow Neutron Yield for a PGNAA Moderator

Further Characterization of a Low-Background, Internal-Source Proportional Counter

A. Seifert, C. E. Aalseth, A. R. Day, E. W. Hoppe, T. W. Hossbach, B. J. Hyronimus, M. E. Keillor, K. E. Litke, E. E. Mintzer, J. L. Orrell, G. A. Warren Radiation Detection and Nuclear Sciences, Pacific Northwest National Laboratory, Richland, WA, USA Pacific Northwest National Laboratory (PNNL) researchers have recently employed ultra-high-purity electroformed copper and the techniques of low-background high-purity germanium spectrometer work for the design and construction of an internalsource gas proportional counter. Such counters can be used to effectively measure many radioisotopes, including low-activity measurements of tritium, 14C, radon, and 37Ar, and can also probe both bulk and surface radiopurity for the copper construction material. Prior work has described the design, construction, and initial testing, including energy resolution, stability, linearity, and performance as a function of pressure, for the PNNL design. The work presented here will describe performance enhancements realized over the last year due to design and construction technique improvements, detail further characterization results for detector efficiency from testing with radioactive gases, and assess background performance, including details on pulse shape discrimination and anti-cosmic rejection.
N25-75:

Thermal Neutron Depth Profiling

J. Vacik, V. Hnatowicz, J. Cervena, D. Fink Nuclear Physics Institute, Academy of Sciences of the Czech Republic, Husinec - Rez, Czech Republic Principle and instrumentation of Thermal Neutron Depth Profiling (TNDP), a specific nuclear analytical technique, is briefly described. The technique is based on analysis of residual energies of charged particles from the exoergic nuclear reactions, such as (n,p) and (n,alpha), with thermal neutrons, and enables to determine concentration profiles of the relevant elements in the nearsurface regions of solids. TNDP in the Nuclear Physics Institute (NPI) at Rez is based on a multi-detector multi-target 173

spectrometer utilizing one- and two-dimensional data acquisition techniques. Specific pulse-shape discrimination circuits are employed in order to suppress parasitic background in the energy spectra. The TNDP at NPI Rez has proven to be a versatile investigative tool, which has been utilized in various studies of the oriented and interdisciplinary research, e.g., in the study of the ion exchange mechanism of the nonlinear optical materials (such as LiNbO3), study of the interaction (diffusion) of dopants in polymers and ISOL targets, technologically important metallic alloys, carbon derivates and composites synthesized from carbon, silicon, nitrogen and boron. The motivation for these studies deals with the fundamental questions of the interaction and diffusion of elements in various materials, materials research in transmutation technologies, refining of metallic surfaces, biomedical application, etc. Long-term experience has shown that TNDP is a useful technique, complementary to Rutherford Backscattering, which has several unique characteristics: it is nondestructive, isotopic sensitive and posses an irreplaceable ability to determine light elements in various structured substrates up to tens of micrometer depths.
This study has been supported by the Ministry of Industry and Trade (under the project No. 1H-PK2/05) and by the Grant Agency of the Czech Republic under the project GACR 202/05/0410. N25-76:

A. Chabre1, A. Lyoussi2, M. Giot3 1 DEN, CEA, Saclay, FRANCE 2 DEN, CEA, Cadarache, FRANCE 3 SCK-CEN, Louvain University, BELGIUM

Advancements in Nuclear Instrumentation, Measurement Methods and Their Applications

The aim of this oveview talk is to give a summary of highlights of the ANIMMA conference helds in Marseille France (7-10 June 2009) . This event bring together scientific, academic, and industrial communities interested in, or actively involved in research and developments related to nuclear instrumentation and measurement methods. The program is focused on all instrumentation, but emphasizes the latest developments in all measurements stages: nuclear radiation detection and in-pile measurements, modeling, electronics, signal acquisition and analysis, interpretation and associated training/education activities.
N25-77: Experimental Orthogonal Functions and Principal Oscillation Patterns of an Unstable Event in a BWR Nuclear Power Reactor

G. Verdu1, D. Ginestar2 1 Dept. Ingenieria Quimica y Nuclear, Universidad Politecnica de Valencia, Valencia, Spain 2 Dept. Matematica Aplicada, Universidad Politecnica de Valencia, Valencia, Spain

Unstable events have been observed in BWR nuclear power reactors, indentified as oscillations present in the neutronic power. There is a great interest on developing fast and accurate techniques for detecting, classifying and suppressing these oscillations based on signal analysis techniques, using the signals provided by the LPRM detectors installed in the reactor core. A technique based on the singular system analysis or principal component analysis of the neutronic signals allows to decompose the oscillating neutronic signals in the in-phase contributions and the out-of-phase contributions. In this work, we show the principal component analysis allows to define experimental orthogonal functions (EOFs), which are spatial functions that can be used for the classification and decomposition of the neutronic signals. The EOFs are computed using only the information provided by the neutronic signals. Another important concern of the time series techniques is the characterization of the stability of a BWR is to determine the linear stability parameters of the core, that is, its decay ratio (DR) and the fundamental frequency (FF) of the oscillation. We propose to compute the principal oscillation patterns (POPs) of the core using the whole set of signals and to obtain the DR and the FF of the core from the dominant oscillation. We will investigate two possibilities of computing the POPs of the core. A first possibility is based on the computation of the Hilbert transform of the neutronic signals, and fitting a vector AR(1) model to the extended set of signals composed of the neutronic signals and their Hilbert transforms. A second method to compute the POPs is based on a vector AR(2) model for the neutronic signals. These methodologies will be tested and compared using the set of signals provided in the Ringhlas 1 stability benchmark.
N25-78:

Characterizing the Ultra-High Energy Resolution TES Array Gamma-Spectrometer Response Function for Isotopic Analysis

O. B. Drury, M. Velazquez, J. G. Dreyer, S. Friedrich Advanced Detector Group, Lawrence Livermore National Laboratory, Livermore, CA, USA We are developing ultra-high resolution gamma-ray detectors based on arrays of superconducting transition edge sensors (TES) for nuclear safeguards and non-proliferation applications. The spectrometer utilizes an adiabatic demagnetization refrigerator to operate the array at temperatures around 0.1 K without liquid cryogens. These detectors have achieved an energy resolution from 50-100 eV FWHM at 100 keV, an order of magnitude improvement over conventional high-purity germanium detectors. Better energy resolution can improve the speed and accuracy of the non-destructive isotope analysis of nuclear materials, which can 174

reduce the burden on more time consuming methods such as mass spectrometry. We have modeled different detector configurations to understand their influence on the response function and improve the detection limits for weak gamma emissions from the fissile Pu and U isotopes between ~40 keV and ~200 keV. Our work guides an optimization of the spectrometer geometry to reduce Compton scattering and increase the peak-to-background ratio. We compare calculations with the observed response functions, and demonstrate a peak-to-background ratio of several 100 at low energy. Emphasis is placed on the isotopic analysis of fissile material in mixed-isotope spent fuel samples whose emission lines are typically masked by intense Compton background.
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344. N25-79:

Silicon Photomultiplier Fabricated from Spin-on Dopants, Rapid Thermal Processing, and Indium Tin Oxide

P. J. Barton, D. K. Wehe Nuclear Engineering, University of Michigan, Ann Arbor, MI, USA A silicon single photon avalanche photodiode (SPAD), fabricated using spin-on dopants (SODs) and rapid thermal processing (RTP) is investigated. Creation of an ultra-shallow photodiode junction is desirable for the detection of low wavelength emissions common to many scintillators. Spin-on dopants are explored as an alternative to the more traditional ion implantation. Doping profiles and junction depths are measured via spreading resistance profilometry. Sheet resistivities for both boron and phosphorus SODs of 20-500 ohms/square are achieved with temperatures below 1000C and times on the order of seconds to minutes. We also present results for RF-sputtered indium tin oxide (ITO) readout lines and quench resistors. Transparent conductors and resistors on the SPAD surface increase detection efficiency versus opaque polysilicon resistors and aluminum readout lines. Resistances and optical transmission of ITO films are measured for various annealing times and temperatures. A 0.7 mohm-cm ITO film with optical transmission of 85% at 400 nm is achieved after a 30 minute hotplate anneal at 500C. Patterning of ITO has been successfully achieved down to several microns using a lift off process incorporating a photoresist inhibition layer. A 5:1 dilute HCl wet etch is shown to produce similar results with an etch rate of 200 nm/min. Implications of these results for large area silicon photomultipliers are discussed.
N25-80: Impact of Ce Concentration and Co-Doping on the Scintillation Properties of Lu2xGd2(1-x)SiO5:Ce (LGSO, X=0.2) Single Crystals

Y. Kurata1, H. Yamamoto2, T. Usui1, S. Shimizu1, N. Shimura1, H. Ishibashi1 1 Inorganic Materials R&D Dept., Hitachi Chemical Co., Ltd. Yamazaki Works(Katsuta), Hitachinaka, Ibaraki, Japan 2 Tokyo University of Technology, Hachioji, Tokyo, Japan

We developed Lu2xGd2(1-x)SiO5 :Ce (LGSO, x=0.2) single crystal as gamma-ray detector for medical imaging applications. Scintillation properties of LGSO were improved by optimized growth conditions and advanced post-growth annealing. The emission spectrum showed two peaks around 420 and 500 nm. In the previous study, the scintillation properties especially emission intensity around 500 nm of the LGSO with Ce 2.0% were improved by the optimized conditions including some dopants and the LGSO was shown to be useful for APD detection. In this study, we have investigated the scintillation properties of LGSO with different Ce concentration (higher than Ce 2.0%) and with different Ca concentration as a co-dopant to enhance the emission intensity around 500 nm for APD detection. Emission intensity around 500 nm under X-ray excitation of LGSO with Ce 4.0% and Ca 0.04% or with Ca 0.1% and Ce 2.0% has been enhanced significantly. It has been shown that the increase of Ce and Ca further enhanced the emission intensity around 500 nm and found to be the most useful for APD detection by optimization of other scintillation properties.
N25-81:

Cerium Bromide - Methanol Adduct, CeBr3(CH3OH)4: A Novel Lanthanide Halide Complex as Inorganic Scintillator

E. V. Van Loef, J. Glodo, W. M. Higgins, K. S. Shah Radiation Monitoring Devices, Inc., Watertown, MA, USA A novel lanthanide halide complex with formula CeBr3(CH3OH)4 was grown from anhydrous methanol solution into large single crystals. CeBr3(CH3OH)4 shows a reasonable light yield, an emission that is well-matched to bialkali photomultiplier tubes, and fast scintillation decay. The radioluminescence spectrum of CeBr3(CH3OH)4 exhibits a broad band peaking in the near-UV due to Ce3+ 5d 4f emission. The emission is shifted to shorter wavelengths when compared to the emission of single crystal CeBr3. We estimate the light yield of CeBr3(CH3OH)4 to be about 5,000 ph/MeV when compared to a BGO crystal. The energy resolution is 15% at 662 keV. Scintillation decay time spectra under gamma-ray excitation show a fast 33 ns decay component.

175

This work was supported by the U.S. Department of Energy under contract No. DE-FG02-08ER85104. N25-82:

L. C. Stonehill1, J. S. Salacka1, I. J. Owens1, M. W. Rabin1, R. Shirey1, O. H. W. Siegmund2, A. S. Tremsin2, J. V. Vallerga2 1 Los Alamos National Laboratory, Los Alamos, NM, USA 2 Space Sciences Laboratory, UC Berkeley, Berkeley, CA, USA Novel microchannel plate (MCP) imaging detectors using cross strip (XS) anodes have been developed recently and demonstrated to be capable of position resolution better than 10 microns, nanosecond timing accuracy, and event rates of greater than 300 kHz. These detectors use charge division and centroiding of MCP charge signals directed onto two orthogonal layers of sensing anode strips to encode event position, event time, and signal amplitude. In this paper, we describe measurements designed to understand the maximum possible event rates that such detectors are capable of handling. For these measurements, the signals from each strip (32 on each axis for the anode discussed here) are sent to a preamplifier ASIC that outputs shaped unipolar pulses with ~40 ns rise time and ~200 ns fall time, which are then digitized at 62.5 mega-samples per second and written to disk. Offline analysis is performed on long data records (up to 25 milliseconds) containing many individual events at event rates up to hundreds of MHz with different spatial distributions over the detector area. We present analysis algorithms and results on our measured ability to resolve events spatially and temporally as the event rate is increased. We also briefly discuss future plans to develop compact, fieldable systems based on this XS anode technology for a variety of applications, from astronomy to remote sensing.
N25-83:

Cross-Strip Anodes for High-Rate Single-Photon Imaging

K. Sakasai1, Y. Iwamoto2, T. Nakamura1, K. Toh1, K. Takakura3, C. Konno3 1 J-PARC center, Japan Atomic Energy Agency, Tokai-mura, Ibaraki 319-1195, Japan 2 Nuclear Scinece and Engineering Directorate, Japan Atomic Energy Agency, Tokai-mura, Ibaraki 319-1195, Japan 3 Fusion Research and Development Directorate, Japan Atomic Energy Agency, Tokai-mura, Ibaraki 319-1195, Japan

Storage Characteristics of KCl:Eu2+ Phosphors with Radiators by Irradiation of Fast Neutrons

Storage phosphors such as KCl:Eu2+ with low gamma-ray sensitivity were applied to detection of fast neutrons using polyethylene (PE) sheets as a proton generator that were set in front of the phosphors. The maximum PSL yields were obtained at a PE thickness of 3 mm when the neutron energy was 14.8 MeV. These experimental results well agreed with those by Monte Carlo simulation using PHITS code. Estimated PSL yields by gamma-ray associated with fast neutron fields were negligible. This combination use of such phosphors and PE sheets will be usable for fast neutron monitoring or imaging in a high gammaray fields.
N25-84:
1

K. Toh , K. Sakasai1, T. Nakamura1, K. Soyama1, S. Nagata2, B. Tsuchiya2, T. Shikama2 1 J-PARC Center, Japan Atomic Energy Agency, Tokai, Japan 2 Institute for Materials Research, Tohoku University, Sendai, Japan

Proton-Induced Fluorescence and Long Lasting Emission of Sr4Al14O25:Eu2+,Dy3+

Fluorescence and long lasting emission properties of Sr4Al14O25:Eu2+, Dy3+ were examined under proton irradiation by varying incident energy. The fluorescence spectra have three strong peaks at 400, 480, and 570 nm. The first two peaks are attributed to the luminescence of Eu2+ and exhibit the long lasting property. The peak at 570 nm is attributed to the luminescence of Dy3+; it exhibits no luminescence after the irradiation has been stopped. The fluorescence intensity ratio of Eu2+/Dy3+ exhibits incident energy dependence, and its value increases with an increase in the proton energy. Radiation damages show activator dependence, and Eu is more sensitive to incident particles than Dy is. Long-lasting emission is observed after the cessation of irradiation, and the decay time of the long-lasting emission is different for different irradiation fluence.
N25-85:

Beta/Gamma Coincidence Measurements of Radioxenon Using a Triple-Layer Phoswich Detector

A. Farsoni, D. Hamby Nuclear Engineering, Oregon State University, Corvallis, USA Beta/gamma coincidence measurement of radioactive xenon has been shown to be the best measurement solution when there is an ultra-low activity of radioxenon. Majority of detection systems for radioxenon monitoring uses separate detectors to detect beta/gamma coincident events. To simplify the coincidence detection setup, a triple-layer phoswich detector has been designed and characterized. The phoswich detector uses a thin plastic, CaF2 and NaI(Tl) scintillators to detect beta/conversion electrons, X-ray and gamma-rays respectively. Coincident events are discriminated from single events using digital pulse shape analysis. In this paper, we present our recent measurements with radioactive lab sources and 133Xe.

176

N25-86:

Characterization of CMOS Position Sensitive Solid-State Photomultipliers

M. McClish, P. Dokhale, J. Christian, C. Stapels, E. Johnson, R. Robertson, K. S. Shah Radiation Monitoring Devices, Inc., Watertown, MA, USA We have designed position sensitive solid-state photomultipliers (PS-SSPM) using a complementary metal-oxide-semiconductor (CMOS) process. While only needing four signal output channels to readout, the device provides spatial information on the micro-pixel level. Three variations of the PS-SSPM design were characterized for their energy and coincidence timing resolution, spatial resolution, and scintillator array imaging. Each PS-SSPM is 1.5 x 1.5 mm2, however, each device has different micropixel geometries and different micro-pixel electrical readout for event position sensing. The FWHM energy resolution at 511 keV was measured using a 1 x 1 x 20 mm3 LYSO crystal. The resolution varied, however, one PS-SSPM design achieved 11.6%. The LYSO scintillator coincidence timing resolution also varied between designs with results ranging from 2.1 to 1.0 nsec. Spatial resolution studies were conducted using a focused (~ 15 m beam spot diameter) pulsed 635 nm diode laser. For each PS-SSPM, its X and Y FWHM spatial resolution was measured. Lastly, we demonstrate the PS-SSPM imaging capabilities using a LYSO scintillator array with 500 x 500 m2 pixels uniformly irradiated by 22Na.
N25-87:

P. Anfre1, B. Hautefeuille1, G. Burato1, O. Tillement2, S. Dewonck3 1 AXINT, 13 avenue Albert einstein, 69003 Lyon, France 2 Laboratoire de Physico-Chimie des Matriaux Luminescents, CNRS, UMR 5620, F-69622 Villeurbanne, France 3 ANDRA, Laboratoire de recherche souterrain de Meuse/Haute-Marne RD 960, 55290 BURE, France We have developed a new range of miniaturized setups for the detection of ionizing radiation (alpha, beta, gamma, X). The device consists in a plurality of high accessibility scintillation probes. Each probe is made of a scintillating crystal and a photodetector. This basic setup can be adapted for the detection of different kinds of radiations by changing the composition or the dimensions of the scintillating crystals. The use of many probes (and thus, of many crystals) can also allow to detect simultaneously different kinds of radiations with different type of probes. Finally, it is possible to insert one or several optical fibers between the photodetector and the crystal. This particular setup allows setting photodetectors and electronics up to 20 meters out of the detection area. Two applications are using currently this type of delocalized probes. The first one studies the diffusion of beta emitter tracer (36Cl, Emax 710 KeV) in geological clay layers, with probes inserted in 20 m long and 20 mm wide drillings in the clay. The second one uses the compactness of the probes in a medical application: a probe can be inserted in the operating channel of an endoscope, then the mainly structural information delivered by the endoscope camera is completed by the functional radiotracer information delivered by the gamma probe.
N25-88:

Scintillating Setup for High Accessibility Detection of Low Activity Ionizing Radiations

B. Hautefeuille1, P. Anfre1, G. Burato1, O. Tillement2, S. Dewonck3 1 69100, AXINT 13 boulevard Einstein, VILLEURBANNE, FRANCE 2 69622, CNRS, UMR5620, Laboratoire de Physico-Chimie des Matriaux Luminescents, VILLEURBANNE, FRANCE 3 55290, ANDRA, Laboratoire de recherche souterrain de Meuse/Haute-Marne RD 960, BURE, FRANCE

Development of a New Beta Emitter Detector: a Mini-Invasive Setup for Radiotracer Study in Geological Clay Layer

We have developed a new beta emitter detector dedicated to radiotracer diffusion studies in geological clay layer. Study of radiotracer diffusion in clay layers is a quiet important issue for the design of future nuclear waste storage. The ANDRA (French agency for nuclear waste) leads studies for that purpose at the Bure underground laboratory (Meuse/Haute Marne, France). The DRN experiment aims measuring long distance radiotracer diffusion, simulating a storage leakage. Experiment will be running during 5 years, and diffusion distance is estimated to range between 0,5 and 1 meter. The global detection system, including detectors and electronics, will be installed in an underground gallery, and the probes will be inserted in 20 m long, 20 mm wide drillings in the clay. Each probe is made of a scintillating crystal coupled to 7 optical fibers. The outputs of these fibers are detected in coincidence by 3 photomultipliers: a light pulse emitted by the crystal will be validated if photons are detected by each photomultiplier simultaneously. The use of optical fibers allows a maximum compactness of the probes, as well as a very high sensitivity. It allows also a human access to the photodetectors and electronics, which will be needed for recalibration and control during the 5 year long study. First calibration test of this system will be discussed, and new developments for industrial radiotracer applications will be introduced.

177

N25-89:

A. O. Li1,2, B. L. Bennett3, M. K. Bacrania1, J. M. Cook1, M. P. Croce1, D. C. Gautier4, E. A. McKigney1, R. E. Muenchausen3 1 N-1, Los Alamos National Laboratory, Los Alamos, NM, USA 2 Nuclear Engineering, North Carolina State University, Raleigh, NC, USA 3 MST-8, Los Alamos National Laboratory, Los Alamos, NM, USA 4 P-24, Los Alamos National Laboratory, Los Alamos, NM, USA A pulsed x-ray streak imaging system (Hamamatsu C4334 streak scope) has been set up for measuring scintillator light emission. In this setup, a scintillating sample is excited by a pulsed x-ray source, and the resulting intensity distribution is measured as a function of time, wavelength and temperature. This information will aid in distinguishing between different luminescence mechanisms, for example, between the energy transfer from self-trapped excitons and defect luminescence. In this work, streak measurements of x-ray induced luminescence will be presented for common scintillators such as NaI(Tl), CsI(Tl), and LaBr3, and the results will be discussed in terms of known luminescence mechanisms.
LA-UR 09-02922 N25-90:

Streak Measurements of Luminescence Mechanisms of Scintillators

J. Iwanowska1, M. Moszynski1, L. Swiderski1, T. Szczesniak1, A. Syntfeld-Kazuch1, P. Sibczynski1, N. Galunov2, N. Karavaeva2 1 Soltan Institute for Nuclear Studies, Otwock-Swierk, Poland 2 Institute for Scintillation Materials of National Academy of Sciences of Ukraine, Kharkov, Ukraine

Composite Scintillators as Detectors for Fast Neutrons and Gamma-Radiation Detection in the Border Monitoring

Composite crystals are solid, non-toxic and non-flammable scintillators designed for spectroscopy of fast neutrons and gamma radiation. They are prepared by adding grain size fractions of p-terphenyl and stilbene to optically transparent silicon glue with high efficiency of fast neutron detection [5]. On the other hand, they are characterized by large opacity, which results in low light output. We present n/ discrimination properties of composite scintillators based on p-terphenyl and stilbene. The results obtained with a stilbene single crystal and a liquid scintillator (BC 501A) are shown as a comparison. The pulse shape discrimination (PSD) has been performed by means of a zero-crossing (ZC) method. The tested scintillators were coupled to a Photonis XP 5500 photomultiplier. We have used a 239Pu-Be source of fast neutrons and gamma radiation as well as a 137Cs source of gamma radiation for light yield measurements. First conclusion of this work is that composite scintillators can be an attractive proposal for the border monitoring.
N25-91:

Scintillation Properties of Eu - and Tl, Eu - Doped NaI Crystals

N. V. Shiran, A. V. Gektin, Y. A. Boyarintseva, S. Vasyukov Institute for Scintillation Materials, Kharkov, Ukraine New market trend related to ultra efficient scintillator need for advanced spectroscopy portals can afford scintillation materials with microsecond decay times. This claim has inspired a renewed interest to Eu2+ doped crystals and resulted in the renaissance of alkaline earth iodide scintillators (Eu-doped CaI2, SrI2, BaI2). At the same time conventional alkali halide scintillators could be potentially upgraded by Eu doping also. NaI crystals with Tl and/or Eu doping were the subject for the scintillator properties modification. Eu2+ doped NaI crystals were studied as a useful scintillation material as well as Tl, Eu - codoped NaI scintillators. Single crystals with the dopant concentration in the range of 10-4 - 0.1 % were grown to establish a correspondence between the luminescence and scintillation properties of Tl - doped and Tl, Eu - codoped NaI. Absorption, photo- and radioluminescence spectra, decay kinetics as well as scintillation parameters were studied. Optical absorption spectrum of NaI: Eu consists of two broad structured bands in 255-270 and 347-360 nm ranges with a crystalline field splitting 10 Dq 10 752 cm-1, which depends on the aging and thermal treatments given to the samples. A narrow band with maximum at 440 nm, caused by the parity allowed electric dipole transition from the 4f7 (8S7/2) ground state to the excited states of the 4f6 5d configuration of Eu2+, is observed in photo- and radioluminescence spectra. Decay kinetics of photoluminescence is close to 970 ns. It was shown that in decaying of NaI: Tl, Eu radioluminescence two components 250 and 1000 ns, typical for Tl+ Eu2+ ions accordingly, are present; and their contributions depend on the co-activators content. Reasonable decay time of NaI: Eu radio luminescence point out the possibility to use this material as scintillator as well.

178

N25-92:

L. Swiderski1, M. Moszynski1, W. Czarnacki1, T. Szczesniak1, R. Marcinkowski1, A. Syntfeld-Kazuch1, G. Pausch2, K. Roemer2 1 Soltan Institute for Nuclear Studies, Otwock, Poland 2 ICx Radiation, GmbH, Solingen, Germany A method based on Compton coincidence technique was applied to study the energy resolution of Compton electrons for various scintillators in a wide energy range. The experimental setup comprised of a High Purity Germanium (HPGe) detector and a tested NaI:Tl scintillator coupled to a PMT. The detectors were set in a face-to-face geometry and the source was placed between them. Thus gating on events backscattered in the tested scintillator and detected in HPGe allowed us to measure the energy resolution of Compton scattered electrons at energy corresponding to Compton edge for a given source. This study confirms the importance of the electron scattering (-rays) as the main component of the intrinsic resolution in scintillators. The results are particularly important for a discussion of the light yield non-proportionality contribution to the energy resolution of scintillators.
N25-94:

Energy Resolution of Compton Electrons in Scintillators

Rectangularly Shaped LAAPDs for PANDA

A. Wilms Hadronphysics 1, GSI Darmstadt, Darmstadt, Hessen, Germany On behalf of the PANDA EMC group The electromagnetic calorimeter (EMC) of the PANDA detector as an important main part of the target spectrometer will consist of PbWO4 crystals with significantly improved light yield. The whole calorimeter will operate at a temperature of approximately T = -25C, which will be challenging in terms of radiation hardness long-term management, not only in case of the used scintillation crystals, but also the radiation hardness of the used photo sensors has to be guaranteed at this envisaged temperature. The whole target spectrometer will be mounted inside a solenoid providing a magnetic field of nearly B 2 T, which precludes the usage of conventional photomultipliers in the region of this detector part of PANDA. APDs were chosen as readout devices for the PANDA-EMC barrel part. As a result of a change of the primary individual crystal geometries, the shape of the chosen avalanche photodiodes changes from quadratic ((10x10) mm2) to rectangular ( (7x14) mm2). Each crystal of the EMC barrel part will be read out via two of these large area APDs (LAAPDs) which results in a total amount of nearly 44,000 APDs for the whole EMC (including the endcaps). The internal structure of those newly developed rectangular LAAPDs has been adopted from the structure of the quadratic devices, whereby the main optical and temperature dependent properties of those two different APD types are nearly identical. One main difference caused by the diode geometry is the increased bulk dark current Idb of the rectangular LAAPDs, which has a main impact on the signal to noise performance of the crystal-APD-preamplifier system. The influence of this parameter change as well as the results of the measurements of the other main APD parameters at different temperatures, which will be part of the PANDA LAAPD mass screening procedure, will be presented and a comparison between the two different tested types of shape will be drawn.
N25-95:

Y. Zhang1, Y. Zhang2, H. Xia1, Z. Xu2, X. Wang2, Y. Wang3, Y. Heng3, T. Zhao4 1 School of Information Science and Engineering, Ningbo University, Ningbo, Zhejing Province, China 2 Department of Modern Physics, University of Science and Technology, Hefei, Anhui Province, China 3 Institute of High Energy Physics, Beijing, China 4 Department of Physics, University of Washington, Seattle, WA 98195, USA Combined electromagnetic and hadron calorimeter based on a total shower absorption principle using scintillating crystals has been proposed recently for the future ILC. Searching for a suitable calorimeter media that is dense and inexpensive is critical for the success of this proposal. Around the year 2000, heavy rare earth metal gadolinium oxide based scintillating glasses doped with Ce3+ were proposed and studied. This type of glasses is relatively inexpensive and the density can reach a level exceeding 5 grams per cubic centimeter. Unlike lead or bismuth oxides commonly used to increase the density of glasses, gadolinium oxide does not quench the luminescence of Ce3+ in the glass matrix. In order to study the glass forming ability and scintillation properties, samples with mole fractions of gadolinium oxide in the range from 10% to 42% and different cerium-doping levels are fabricated. Optical properties, scintillation light yields and decay time of these samples will be reported.
N25-96:

Development of Gd2O3 Based Dense Scintillating Glass

Growth and Scintillation Properties of Nd Doped LiCaAlF6 Scintilltor

T. Yanagida, Y. Yokota, S. Maeo, A. Yoshikawa, N. Kawaguchi, S. Ishizu, K. Fukuda Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Japan In this study, we develop 0.5% Nd3+ activated LiCaAlF6 single crystalline scintillator by micro-PD method. After cut and polishing processes, optical properties, including both transmittance and emission, were evaluated. Then, X-ray excited radio luminescence spectrum and 241Am alpha-ray excited pulse height spectrum was investigated. 179

N25-97:

T. Yanagida1, Y. Fujimoto1, Y. Yokota1, S. Maeo1, A. Yoshikawa1, M. Miyamoto2, H. Sekiwa2 1 Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Japan 2 Mistubishi Gas Chemical Company, Inc., Tokyo, Japan

Comparative Study of Ga, In, and Mg Doped ZnO Thin-Film Scintillator with Geiger Mode APD

In the present study, in order to develop efficient alpha-ray detector for nuclear plants, we grow pure and Ga, In, and Mg doped ZnO crystals by the Liquid Phase Epitaxy (LPE) method. Optical properties including transmittance and alpha-ray excited luminescence spectra were investigated. Then, using PMT R7600, the light yield and decay time excited by alpha-rays were evaluated, and a dependence of the dopant concentration was studied. Finally, coupling with Multi PiXel Photon Counter (MPPC, one kinds of Geiger mode APD), also light yield and decay time was compared with PMT results, to investigate total properties of exciton luminescence scintillator coupled with Geiger mode APDs as a next generation radiation detector.
N25-98:

K. Kamada1, K. Tsutsumi1, T. Yanagida2, T. Endo1, Y. Usuki1, A. Yoshikawa2 1 Materials Research Laboratory, Furukawa Co., Ltd., Tsukuba, Japan 2 IMRAM, Tohoku University, Sendai, Japan

Large Size Single Crystal Growth of Pr:Lu3Al5O12 and Their Uniformity of Scintillation Properties.

Pr-doped Lu3Al5O12 (Pr:LuAG) single crystals were grown by Czochralski method. The crystals were seeded-grown in the [100] direction. Dimensions up to 100mm in length and 92mm in diameter were achieved without cracking. Using 5 x 5 x 1 mm size sample coupled with photomultiplier (Hamamatsu R7600), energy resolution of 4.6% at 662 keV was achieved. Uniformity of light yield and energy resolution was evaluated in whole crystal.
N25-99:

A New Neutron Scintillator Based on 6Li Metal Sheet and ZnS(Ag) Phosphor Layer

R. Engels, G. Kemmerling, H. Noeldgen, J. Schelten Zentralinstitut fuer Elektronik, Forschungszentrum Juelich GmbH, Juelich, Germany A new neutron scintillator based on a 0.1mm thick 6Li metal sheet close to a thin phosphor LJ 440 was investigated. The phosphorous film consists of an optically transparent of 0.25mm thin polyester film, which on one side is deposited by Ag activated ZnS, with a layer thickness of 3.25mg cm -2. The function and property of this new scintillator type will be described in this paper. From neutron measurements the mean light emission, the emission distribution, and the neutron detection efficiency are deduced. The pulse height channel number is calibrated to 3H energy deposited in the phosphorous layer and to the number of photons hitting the photo cathode of a standard photomultiplier (PM) tube. We measured the gamma sensitivity with a 1.1MeV and 1.3MeV radiation from a radioactive 60Co source and its value is very low, if not zero. Excellent detection stability is verified by comparing continuously measured runs of one hour. For a direct comparison, measurements are also performed with the standard 6Li glass neutron scintillator.
N25-100:

D. H. Lee1, Y. Huh1, S. Jeon1, B. Kim1, Y. G. Hwang1, K. S. Song1, Y. K. Kim2 1 Advanced Medical Device Research Center, Korea Electrotechnology research institute, Ansan, Kyonggi-do,, Korea 2 Department of nuclear engineering, Hanyang University, Seoul, Korea

On the Optimal Design of Flat-Panel Image Detector Based CsI(Tl) for C-Arm Fluoroscopy System

For effective and precise diagnostic, radiology, the digital fluoroscopy devices that provide visualized moving anatomical structure image have been required and developed continually. Therefore we have developed fluoroscopy system which combined fluoroscopy providing realtime image and CT providing three-dimensional image together. The aim of this study was to find optimal condition of X-ray detection system which will be built into this C-arm fluoroscopy system. Before this we found good condition of X-ray generator system for guarantee to acquire high quality image with low dose environment. Our method to construct X-ray detector for digital radiography was indirect approach. This indirect method has high X-ray absorption efficiency but the spatial resolution encounters limitation because of dispersion of light generated in scintillators. Therefore convertors design is very important for good DR images because the scintillator geometry as thickness affects dispersion of light. In the present study CsI(Tl) scintillators with various thickness range were prepared and we found optimal design of scintillators by MCNPX simulation and experimental result. The characteristics of CsI(Tl) with various thickness were determined by experiments with X-ray tube and various device. In addition, energy absorption efficiency which related to X-ray detection and light transmission efficiency which related to light collection were calculated using MCNPX and DETECT97 simulation code. In order to obtain best energy resolution with given scintillators, collection of light emitted is very crucial. Therefore the techniques to maximize the light collection efficiency were investigated and tested experimentally. For this experiment, a number of wrapping material as reflector used to optimize the light collection from scintillators were tested for their reflectivity and their response in term of improving the energy resolution in CsI(Tl) scintillators.

180

This study was supported by a grant from the cooperative R&D Program funded by the Korea Research Council Industrial Science and Technology, Republic of Korea. N25-101:

Evaluation of a 1024 Anodes Micro-Channel Plate PMT for Preclinical PET Imaging

D. Brasse, J. Wurtz, S. Salvador, M. Imhoff, B. Humbert IPHC, Universite de Strasbourg, CNRS / IN2P3, Strasbourg, France A new generation of flat panel photomultiplier tube (PMT) based on micro-channel plates (MCP) offers characteristics suitable for PET imaging. We are developing a preclinical PET system composed of four detection modules arranged around the animal. Each module consists in 768 LYSO:Ce crystals red at both ends by MCP PMT. The particular geometry combined with an inner diameter of 61.2 mm lead to a high detection efficiency with a volumetric spatial resolution of 1 l. The purpose of this work is to characterize the 1024 anodes MCP PMT (Planacon XP85023/A1, Photonis Corp.) used in our preclinical PET system. The Planacon has very compact size (58 x 58 x 13.7 mm3) with an active area of 53 x 53 mm2. The anode array consists in a 32 x 32 matrix in which each individual anode is 1.4 x 1.4 mm2 with a pitch of 1.6 mm. Due to the lack of backside connectors and in order to individually readout the anode current, a dedicated connection board has been developed. The dark current, the gain and the timing resolution of one anode are measured as well as the cross talk, the gain uniformity and the intrinsic spatial resolution on the entire active field of view. With a timing resolution below 100 ps, an intrinsic spatial resolution of 400 m, a low dark current and a relatively high gain, the Planacon XP85023/A1 offers a promising photodetector for PET imaging.
N25-102: A Time Efficient Optical Model for GATE Simulation of a LYSO Scintillation Matrix Used in PET Applications

D. A. Baptista Bonifacio1,2, N. Belcari2,3, S. Moehrs2, M. Moralles1, V. Rosso2,3, S. Vecchio2,3, A. Del Guerra2,3 1 Research Reactor Center (CRPq), Energy and Nuclear Research Institute (IPEN), Sao Paulo, SP, Brazil 2 Department of Physics E. Fermi, University of Pisa, Pisa, PI, Italy 3 National Institute of Nuclear Physics (INFN), Pisa, PI, Italy

Monte Carlo simulation is a very useful tool for the assessment of the performance of medical imaging devices in emission tomography and, as a consequence, for the optimization of acquisition protocols and the development of image reconstruction algorithms and correction techniques. In this work, we performed the Monte Carlo simulation of a LYSO scintillation matrix seen by a position sensitive photomultiplier, as used in the DoPET (Dosimetry with a Positron Emission Tomograph) tomograph. The GATE (Geant4 Application for Emission Tomography) simulation code for tomographic applications, based on the Geant4 toolkit, was used for that task. The simulation includes a time efficient optical model for light transport and collection. The objective is the assessment of DoPET performance and its agreement with GATE results including all the relevant physical aspects for PET applications. A very good agreement was found between experimental and simulated data with the inclusion of the optical model.
Acknowledgements: CNPq (Brazilian National Research Council) funded this study. N25-103:

A. Syntfeld-Kauch1, P. Sibczyski1, M. Moszyski1, A. V. Gektin2, W. Czarnacki1, M. Grodzicka1, J. Iwanowska1, M. Szawowski1, T. Szczniak1, L. widerski1 1 the Andrzej Soltan Institute for Nuclear Studies (IPJ), Otwock, Poland 2 Institute for Single Crystals, Kharkov, Ukraine The presented study on CsI(Na) alkali halide crystals originated from our new big interest in the study of non-proportionality and energy resolution of crystals exibiting two components in the scintillation pulse. The extensive study of CsI(Tl) showed that the non-proportional response of the CsI(Tl) crystal to -rays and the energy resolution can be improved when a long signal formation time (several s) is set in the amplifier. For the tested CsI(Na) (10x10 mm and 1"x1") the non-proportional response and intrinsic energy resolution also improve at longer shaping times. The measured energy resolution of the bigger crystal is as good as 5.8% at 662 keV for the shaping time equal to 12 s. Since CsI Tl-doped specimens exhibit a light pulse shape that consists of three exponential components with the longest decay time up to several microseconds, the light pulse shape of the CsI Na-doped sample was also inspected to know its composition and relative contributions of observed light components. A fast Tektronix DPO7254 digital oscilloscope was used to measure the average pulse shapes of CsI(Na) irradiated by rays from 22Na (511 and 1274.5 keV) and 54Mn (834.8 keV) sources. The light pulse shape of CsI(Na) can be describe by three decay times: ~470 ns (fast), ~1.9 s (slow) and ~8 s (tail) with the relative intensities 41%, 28% and 31%, respectively. The good value of energy resolution (5.8% at 662 keV) measured for 1"x1" CsI(Na) at 12 s shaping time in the amplifier makes this crystal a candidate for hand-held radioisotope identification instruments. On the other hand, such aspects as a long-term stability, efficiency as a function of -ray energy, linearity of the detector, peak shift as a function of countrate, and others are also important with respect to the radioisotope identification. 181

Performance of CsI(Na) Scintillators in Gamma-Ray Spectrometry

T. Szczesniak1, L. Swiderski1, M. Moszynski1, A. Syntfeld-Kazuch1, D. Wolski1, G. Pausch2, J. Stein2, F. Kniest3, M. R. Kusner4, P. Schotanus5, C. Hurlbut6 1 Soltan Institute for Nuclear Studies, Otwock-Swierk, Poland 2 ICx Radiation GmbH, Solingen, Germany 3 Saint-Gobain Crystals, Holland Office, Soest, The Netherlands 4 Saint-Gobain Crystals, Newbury, USA 5 SCIONIX Holland, Bunnik, The Netherlands 6 Eljen Technology, Sweetwater, USA Liquid scintillators loaded with Boron-10 or Lithium-6 are capable to detect gammarays, fast neutrons and also thermal neutrons. One of the popular method applied in order to distinguish events originating from different particles is Pulse Shape Discrimination (PSD). The previously presented study of Boron-10 loaded liquid scintillators showed different discrimination performance using PSD method in scintillators such as BC523A, BC523A2, EJ339A2 and EJ309B5. It triggered a further study of light pulse shapes in these scintillators originating from events related to gamma-rays and fast and slow neutrons. The light pulse shapes were recorded using single photon method and included in the 2-dimensional n/gamma discrimination data acquisition as the third parameter. Next, the recorded light pulse was gated using energy and PSD information to extract pulses characteristic for only one kind of particles. Finally, the study of the light pulse shapes was performed with multi-exponential fits to experimental data and calculation of decay time constants and intensities of components. The results were compared with the well known liquid scintillator BC501A, not sensitive to thermal neutrons.
N25-106:

N25-104:

Light Pulse Shapes in Liquid Scintillators Originating from Gamma-Rays and Neutrons

J. S. Neal1, D. M. DeVito1, J. J. Henry1, B. L. Armstrong1, X. Yang2, N. C. Giles3, J. Y. Howe1, L. A. Boatner1 1 Oak Ridge National Laboratory, Oak Ridge, TN, USA 2 Department of Physics, West Virginia University, Morgantown, WV, USA 3 Department of Engineering Physics, Air Force Institute of Technology, Dayton, OH, USA

Investigation of Spark Plasma Sintering Techniques for Fabricating ZnO-Based Polycrystalline Ceramic Scintillators

ZnO-based scintillators are well-suited for use as the associated particle detector in a deuterium-tritium (D-T) neutron generator. ZnO, ZnO:Ga, ZnO:In, ZnO:In,Li, and ZnO:Er,Li have demonstrated fast (sub-nanosecond) decay times with relatively low light yields. Unfortunately, detectors using powders are difficult to assemble, and the light yield from powders is less than satisfactory. Single crystal ZnO of sufficient size possessing the necessary scintillation properties are not available. New applications for D-T neutron generators require better timing resolution and higher count rates than are currently available with associated particle detectors using YAP:Ce as the scintillator. ZnO-based polycrystalline ceramic scintillators offer the advantages of high light yield, ease of fabrication, low cost, and robust mechanical properties. In our previous work, two ceramic ZnO:Ga bodies both formed using spark plasma sintering (SPS) methods maintained good near-band-edge (NBE) to visible emission ratios after sintering and demonstrated improved luminescence after H2 treatments. These new studies seek to improve the optical transparency of the ceramic bodies while optimizing the scintillation characteristics since the application of these ceramic scintillators requires a transmission geometry for the scintillation light. Precursor powders used in these studies included ZnO mixed with varying amounts of SiO2 and ZnO:Ga powders. Ongoing studies of lutetia (Lu2O3) and lutetium oxy-orthosilicate (LSO) ceramic scintillators suggest the beneficial properties of SiO2 when used as a sintering agent. Specifically, improved transparency and associated light transmission have been demonstrated when SiO2 was used in pre-sintered green bodies. Photoluminescence measurements demonstrated the detrimental effects (in terms of fast, NBE emission) of air treatments and the beneficial effects of hydrogen treatments (increased NBE and reduced visible emissions).
N25-107:

G. Ban1, K. Bodek2, T. Lefort1, O. Naviliat-Cuncic1, C. Plonka3, G. Rogel1 1 LPC CAEN:ENSICAEN CNRS:IN2P3, Caen, France 2 Jagellonian University, Cracow, Poland 3 Institut Laue Langevin, Grenoble, France Ultra cold neutrons (UCN) produced at reactors or spallation sources are used by precision experiment to probe the standard model. As neutrons their unusual properties like wall reflections, small penetration depth raise several problems for detecting them. They can be detected with traditional techniques such as Helium 3 counters but with the new powerful UCNs sources build around the world and a foreseen increase factor of ten to a hundred in UCN intensity , Helium 3 counters can no longer be used. To detect high UCN flux we test several lithium doped Glass scintillators, UCN efficiency, gamma sensitivity, escape peak, count rate has been quantified, and another option will be gas scintillation.

Glass Scintillators for Ultra Cold Neutrons

182

N25-108:

P. Muralidharan1, S. J. Lee1, B. K. Cha2, J. Y. Kim2, D. K. Kim1, G. Cho2 1 Materials Science and Engineering, KAIST(Korea Advanced Institute of Science and Technology), Daejeon, South Korea 2 Nuclear and Quantum Engineering, KAIST(Korea Advanced Institute of Science and Technology), Daejeon, South Korea

Hydrothermal Synthesis and Characterization of Nano Gd2O3(Eu) Scintillator for High Resolution X-Ray Imaging Application

In the last decade, digital X-ray imaging systems have been replacing analog X-ray imaging systems with conventional X-ray film-screen for radiography applications. Between the direct- and indirect-detection methods for X-ray imaging, indirectdetection type consisted of an X-ray converter (or a scintillator film) and 2D imaging devices are more widely used in medical diagnoses and industrial fields due to its high stability and high performance. Scintillation materials such as terbium doped gadolinium oxysulfide (Gd2O2S:Tb, Gdox) and europium- doped gadolinium oxide(Gd2O3:Eu) are commonly used because of their high luminescence efficiency and emission wavelength(visible light region) well matching to silicon sensors such as a-Si:H panel detector, CCD and CMOS type imaging devices. In this work, nano Gd2O3(Eu) scintillator powders was synthesized by a hydrothermal process. Thus obtained sample was further calcined at different temperature, time and concentration in the electric furnace. The synthesized samples with nano- like rods morphology were observed as compared with granular-type commercial powder. Gd2O3(Eu) scintillator films with different thicknesses were manufactured by mixing and thermal hardening of powder, dispersion agent and other organic additives by screen printing method. In order to investigate the characterization such as the crystal structures and microstructure of nano Gd2O3(Eu) scintillator, X-ray diffraction(XRD) and scanning electron microscopy(SEM) measurement were implemented. The scintillation properties such as photoluminescence and X-ray imaging performance in terms of the light response to X-ray exposure dose, signal-to-noise-ratio (SNR) and spatial resolution were evaluated and analyzed by combining the screen films with CCD or CMOS imaging detectors under radiographic systems conditions.
N25-109:

H. Ohgaki1, T. Kii1, K. Masuda1, H. Harada2, F. Kitatani2, T. Hayakawa3, T. Shizuma3, N. Kikuzawa4, R. Hajima4, N. Nishimori4, H. Toyokawa5 1 Insititute of Advanced Energy, Kyoto University, Uji, Kyoto, Japan 2 Nuclear Science and Engineering Directorate, Japan Atomic Energy Agency, Tokai, Ibaraki, Japan 3 Advance Photon Research Center, Japan Atomic Energy Agency, Kizugawa, Kyoto, Japan 4 Advance Photon Research Center, Japan Atomic Energy Agency, Tokai, Ibaraki, Japan 5 Research Institute of Instrumentation Frontier, National Institute of Advanced Industrial Science and Technology, Tsukuba, Ibaraki, Japan Nuclear Resonance Fluorescence (NRF) is powerful tool for investigation not only of the nuclear physics, but also of isotope detection for the homeland security. The required performance of the detector used in the NRF experiments are high energy resolution, high full energy efficiency, and high counting rate. Therefore most of the NRF experiments have been carried out by using high purity germanium (HP-Ge) detectors with large volumes. The HP-Ge detectors show the best resolution, about 2keV at 5 MeV, however, the counting rate is typically below 10 kHz. Moreover, a large volume HP-Ge is very expensive. We can use scintillators which have superior detection efficiency than HP-Ge and by which cost effective system could be constructed. However, scintillators have poor energy resolutions. For instance a NaI(Tl) scinttilator shows about 6% energy resolution (FWHM) at gamma-ray energy of 662 keV. Recently, a new scintillator material, LaBr3(Ce), has been developed and shows excellent energy resolution. Therefore, it is worth to test the performance of the LaBr3(Ce) scintillation detector in the NRF experiments, especially with the energy resolution at several MeV energy gamma-ray. The LaBr3(Ce) scintillation detector used in this test experiment has size of 3-inch x 3-inch. The preliminal experiment shows the energy resolution of 2.8%(FWHM) at 1.332 MeV from 60Co at room temparature. A NRF experiment has been performed in the AIST LCS beamline[4]. The target nuclei are 208Pb and 12C and the gamma-ray energies of 4.44, 7.33, and 15.1 MeV. We will compare the energy resolutions of LaBr3(Ce) with those of HP-Ge and of NaI(Tl). The energy dependence of the energy resolution will be discussed as well.
This work was supported by the Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan, #19360431 and #21360467.

Performance of the LaBr3(Ce) Scintillator for Nuclear Resonance Fluorescence Experiment

183

N25-110:

N. Kawaguchi1,2, T. Yanagida1, Y. Yokota1, K. Kamada1, K. Fukuda1,2, T. Suyama1, A. Yoshikawa1,3 1 IMRAM, Tohoku University, Sendai, Miyagi, Japan 2 Tokuyama corp., Sendai, Miyagi, Japan 3 NICHe, Tohoku University, Sendai, Miyagi, Japan

Study of Crystal Growth and Scintillation Properties as a Neutron Detector of 2-Inch Diameter Eu Doped LiCaAlF6 Single Crystal

We are working on development the neutron detector based on fluoride single crystal scintillator. For efficient materials screening, we use the micro-pulling down method which allows us to grow high quality single crystal with high speed. Lithium with a large neutron absorption cross section is suitable for the neutron detection. Eu doped LiCaAlF6(Eu:LiCAF) contains Lithium, and its gamma rays sensitivity is the lowest one in fluoride crystals. In our past report, the neutron irradiated evaluation of Eu:LiCAF had been done for first time. As a results, it is found that the light yield of Eu:LiCAF is 150 % of that of Li-glass scintillator, so it is a promising scintillator for neutron detection. However, up to now, we only have made the 2mm diameter Eu:LiCAF single crystal had not been done, and we have never made more big size Eu:LiCAF single crystal. In this report, 2inch diameter Eu:LiCAF single crystal was grown by using Czochralski method. Our Eu 2% :LiCAF crystal contained transparent phase and opaque phase, and transparent area of Eu2%:LiCAF crystal was cut and polished to 497 sample peaces. We measured the light yield of 497 samples from alpha ray (241Am) exited pulse height spectra of these samples. As a results, it is thought that the light yield of transparent phase is roughly uniform, considering about 10% measurement error.
N25-111: Probability Distribution and Noise Factor of Solid State Photomultiplier Signals with Cross-Talk and Afterpulsing

S. Vinogradov, T. Vinogradova, V. Shubin, D. Shushakov, K. Sitarsky Amplification Technologies, New York, NY, USA Operating principles of Solid State Photomultipliers (SSPM) are based on Geiger mode avalanche breakdown limited by strong negative feedback. This operating mode provides both high gain and ultra-low excess noise of internal amplification resulting in ability to detect single photons. On the other hand high gain is accompanied with cross-talk and afterpulsing processes changing probability distribution function of output signals and arising excess noise. There are many mentions regarding its influence on signal distribution and noise, and we suppose that some analytical expressions would be useful. We will present our model of probability distribution of output signals in presence of cross-talk and afterpulsing (namely compound Poisson distribution). Model results allow us to calculate excess noise factor of cross-talk and afterpulsing as well as to find simple way to measure its probabilities.
N25-112: Improvement of Energy Resolution via Correction on Non-Uniform Light Collection in Large Scintillation Detectors

H. Yang, R. Venkataraman, N. Menaa Canberra Industries, Meriden, CT, USA Because of their low cost, high efficiency, moderate energy resolution and robustness, large scintillation detectors are quite popular in field applications. In addition to the relatively large band gap between the valence band and the conduction band (thus requiring a higher energy to produce each signal carrier), non-uniform light collection efficiency throughout the crystal is also a reason for degradation of energy resolution in large scintillation detectors. At Canberra Industries, accurate MCNP models of NaI scintillation detectors (the MD series stabilized NaI detectors) have been developed using detailed knowledge of the detector construction and materials. This has been done as part of the ISOCS (In Situ Object Counting System) characterization process for each detector type. These models were then verified using measurement results. In this work, these models were input into the computer code DETECT2000 for calculation of light collection efficiency distribution throughout the whole crystal. Monte Carlo simulation was then performed for a specific counting geometry, at interested energies. Energy deposition together with interaction positions was recorded in list mode. Combining the knowledge of light collection efficiency at these interaction positions, response function of the scintillation detector at interested energies was defined, for a specific counting geometry. The response function was used to deconvolve the measured spectrum to improve energy resolution.

184

N25-113:

Y. Furuya1, N. Kawaguchi1,2, N. Abe1, Y. Yokota1, T. Yanagida1, M. Nikl3, A. Yoshikawa1,4 1 IMRAM Tohoku univ, Sendai, Japan 2 Tokuyama Corporation, Sendai, Japan 3 Institute of Physics, ASCR, Prague, Czech 4 NICHe, Sendai, Japan 5d-4f Nd3+ luminescence (parity and spin allowed transition) in VUV region is well known to have fast decay time, because of relationship between wavelength and emission lifetime. However, most of the crystal hosts absorb VUV luminescence, i.e. the emission can't be observed. Therefore, we have investigated fluoride single crystal which shows high transparency from IR to VUV. NdF3 single crystal was reported in the past; however, optical properties havent been discussed in VUV region. In this study, NdF3 single crystals were grown by the micro-pulling-down (-PD) method, and optical properties and -ray responses were investigated.
N25-114:

Crystal Growth and Scintillation Properties of NdF3 Single Crystal

H. Tanaka1,2, N. Kawaguchi1,3, N. Abe1, Y. Yokota1, T. Yanagida1, M. Nikl4, A. Yoshikawa1,5, Y. Kawazoe2 1 IMRAM,Tohoku University, Sendai, Miyagi, Japan 2 IMR, Tohoku University, Sendai, Miyagi, Japan 3 Tokuyama corporation, Sendai, Miyagi, Japan 4 Institute of Physics AS CR, Prague, Czech 5 NICHe, Sendai, Miyagi, Japan Fluoride single crystal has high transparency between UV and IR region. Therefore, Fluoride single crystal is expected to be used in the various applications, such as laser material, scintillator and so on. Recent years, due to the requirement of efficient nuclear cycle, detection of pollution from plutonium became important. In this study, based on this background, we grow 1, 5, 10, 20, 30, and 40% Nd doped CaF2 single crystal as an alpha-ray scintillator. Optical properties and radiation responses were investigated for different Nd concentration.
N25-115:

Crystal Growth and Scintillation Properties of Nd Doped CaF2 Single Crystal

S. Stange1, E. I. Esch1, A. J. Couture1, R. E. Del Sesto1, R. D. Gilbertson1, T. M. McCleskey1, E. A. McKigney1, R. E. Muenchausen1, R. Reifarth2 1 Los Alamos National Laboratory, Los Alamos, New Mexico, USA 2 Gesellschaft fr Schwerionenforschung, Darmstadt, Germany The neutron capture cross-sections of a number of short-lived isotopes are needed for applications in nuclear astrophysics, the nuclear fuel cycle, and interpretation of data from nuclear weapons tests. However, existing neutron capture cross-section experiments, such as DANCE and nTOF, are limited to measuring the cross-sections of isotopes with half-lives greater than a few hundred days. The use of cerium fluoride (CeF3) could allow these cross-sections to be measured. The slow component of cerium fluoride has a decay time of 27 ns, compared with the 630 ns decay time of barium fluoride (BaF2), currently the standard detector for neutron capture measurements. However, cerium fluoride detectors are not available in large sizes or quantities. Los Alamos National Laboratory is currently conducting a program to develop nanocomposite scintillators, which consist of nanoparticles of a scintillating material dispersed in a matrix material. Nanocomposite scintillators may enable the use of scintillator materials such as cerium fluoride without requiring the growth of large crystals. We have successfully fabricated cerium fluoride nanoparticles with sizes < 10 nm and have characterized their optical, structural, and radiation detection properties. We have tested nanocomposite scintillators in a neutron beam, and have observed neutron capture resonances not seen by a barium fluoride detector. We are working to develop a non-hydrogenous liquid matrix for the nanoparticles. Current research results will be presented. LA-UR-09-02921
N25-116:

Development of Nanocomposite Scintillators for Neutron Capture Measurements

S. K. Lee1, B. H. Kang1, W. G. Lee1, J. K. Kim1, Y. K. Kim1, N. Z. Galunov2, G. D. Kim3 1 Nuclear Engineering, Hanyang University, Seoul, South Korea 2 Institute for Scintillation Materials of National Academy of Science of Ukraine, Kharkiv, Ukraine 3 Korea Institute of Geoscience and Mineral Resources, Daejeon, south Korea

Scintillation Properties of Large Area Composite Stilbene Crystal for Neutron Detection

Stilbene is one of representative scintillator, which has been used in fast neutron detection. In experiments with non-stationary neutron sources, it is often necessary to measure the neutron-yield evolution with a high temporal resolution against the background of accompanying gamma rays. Organic scintillator such as stilbene crystals characterized by good temporal resolution, and the ability of pulse-shape discrimination of neutron and gamma ray signals, are the most suitable devices for this 185

problem. Stilbene single crystal has a good spectrometry characteristics and light yield in comparison with plastic and liquid scintillators, but they are more expensive and their diameters were not larger than 60-80 mm. Crystalline grains of stilbene have been preliminarily obtained by mechanical grinding of stilbene single crystals, and a large size composite stilbene scintillator was manufactured with crystal grains into organosilicone base. The major aim of this work is to compare detector responses of the large area composite stilbene scintillator of diameter 200 mm x 20 mm size with reference stilbene single crystal of diameter 25 mm x 20 mm size when they are used for the detectors of fast neutrons. To study a degree of heterogeneity of light yield, we have measured the response of the large area scintillator at different points of surface. The measurements were carried out by using BURLE 8850 and 8575 photomultiplier tubes. We measured and investigated response properties of alpha-particles , gamma-rays and neutrons. The neutron source used for this study is 252Cf and mono-energy fast neutrons by using the accelerator in Korea Institute of Geoscience and Mineral Resources (KIGAM). The measured results are compared to the response of a small size (diameter 25 mm x 20 mm) stilbene single crystal.
Acknowledgment : This work was supported by the nuclear R&D program of MEST, Korea and also supported by the Ministry of Knowledge Economy (2008-P-EP-HM-E-06-0000). N25-117:

Cross Talk Study to the Single Photon Response of a Flat Panel PMT for the RICH Upgrade at

C. Arnaboldi1, M. Artuso2, M. Calvi1, E. Fanchini1, C. Matteuzzi1, D. L. Perego1, G. Pessina1, J. Wang2 1 Istituto Nazionale di Fisica Nucleare di Milano Bicocca and Universit degli Studi di Milano Bicocca - Dipartimento di Fisica, P.za della Scienza 3, 20126, Milano, Italy 2 Physics Department, 201 Physics Building, Syracuse University, Syracuse, NY 13244-1130, Syracuse, New York The Ring Imaging CHerenkov, RICH, detector at LHCb is now readout by Hybrid Photon Detectors. In view of its upgrade a possible option is the adoption of the flat panel Photon Multipliers Tubes, PMT. An important issue for the good determination of the rings produced in the sensitive media is a negligible level of cross talk. We have experimentally studied the cross talk from the 64x64 pixels of the H9500 PMT from Hamamatsu. Results have shown that at the single photon signal level, as expected at LHCb, the statistics applied to the small number of electrons generated at the first dynode of the PMT chain leads to a cross talk mechanism that must be interpreted in term of the percentage of the number of induced signals rather than on the amplitude of the induced signals. The threshold to suppress cross talk must be increased to a significant fraction of the single photon signal for the worst case. The number of electrons generated at the first dynode is proportional to the biasing voltage. Measurements have shown that the suppressing threshold can be lowered acting on this latter parameter.
N25-118:

LHCb

H. Chagani1, R. Dolenec1, S. Korpar1,2, P. Krizan1,3, R. Pestotnik1, A. Stanovnik1,4, R. Verheyden1 1 Experimental Particle Physics Department, Jozef Stefan Institute, Ljubljana, Slovenia 2 Department of Chemistry and Chemical Engineering, University of Maribor, Maribor, Slovenia 3 Department of Mathematics and Physics, University of Ljubljana, Ljubljana, Slovenia 4 Department of Electrical Engineering, University of Ljubljana, Ljubljana, Slovenia The use of Silicon Photomultipliers (SiPMs) as photon detectors in Positron Emission Tomography (PET) modules offers significant advantages over conventional light sensors, including application in a magnetic field, better resolution and easier operation. Two PET modules have been constructed by coupling 4 x 4 arrays of LYSO scintillation crystals of area 4 x 4 mm2 and length 20 mm to SiPMs. Two types of SiPM have been tested: the Hamamatsu S10931-100P and Photonique PCB-PET07 of active surface areas 3 x 3 mm2 and 2.1 x 2.1 mm2 respectively. The energy, time and spatial resolutions of the arrays are presented in view of arrangement into a larger module. Results from both modules are also compared with light collection simulations performed in GEANT4.
N25-119:

Tests of Silicon Photomultiplier PET Modules

K. Saito1, S. Sasaki1, T. Sanami1, H. Tawara1, E. Shibamura2 1 Radiation Science Center, High Energy Accelerator Research Organization, Ibaraki, Japan 2 Saitama Prefectural University, Saitama, Japan Scintillation photons from helium of more than atmospheric pressure, where most of exited helium atoms form excited helium dimers, lie in the VUV region (60-100 nm) and have long decay times (10 s). We have been testing the exchange of luminescence origin from helium to xenon by adding a small amount of xenon to helium. In this study, scintillation properties of helium mixed with xenon, such as luminescence spectra, time profile and scintillation yield are measured. Since ionization yields can be measured simultaneously with scintillation in rare gases and those is important to understand energy loss processes of radiation, those in He/Xe are measured simultaneously with scintillation yields. Scintillation mechanism in He/Xe is clarified on 186

Scintillation Mechanism in Helium Mixed with Xenon

the basis of these experimental results. The peak wavelength in the luminescence spectra in the He/Xe mixture is 173 nm. The shape of the luminescence spectra in He/Xe is almost the same as that in pure xenon. The rise time of scintillation in He/Xe is faster than 4 ns, and the decay time of that is shorter than the decay time of an excited helium dimer. The scintillation yields increase with increasing partial pressure of xenon. However, scintillation yields under the electric field of an enough strength to prevent the recombination of ion pairs completely, are saturated above 0.06 MPa of partial xenon pressure. When the partial pressure of xenon was 0.066 MPa for a total pressure of 1.0 MPa with the electric field, the scintillation yield was comparable to that in 0.1 MPa pure xenon. Increasing the rate of scintillation yields depends not only on the xenon pressure but also on the helium pressure. Ionization yields scarcely depend on xenon pressure. A mechanism of scintillation in He/Xe will be discussed in detail on the basis of these experimental results in the presentation.
N25-120: Measurement of Light Yield Proportionality in Scintillator Samples Through a Coincidence Technique

J. W. Cates, J. P. Hayward, K. Meeks Nuclear Engineering, University of Tennessee, Knoxville, TN, USA Characterizing light yield proportionality in scintillators is one of a few important considerations in understanding their energy resolution. Thus, several recent efforts have been made to efficiently quantify relative light yield as a function of energy deposited in new scintillator samples. A Compton spectrometer was set up at the University of Tennessee to augment the test facilities of the Scintillation Materials Research Center (SMRC) at the university. It is a coincidence setup between a scintillator sample and a HPGe detector. The setup is read out by a compact, multi-channel acquisition system Light yield proportionality was analyzed through post-processing of collected data. The system was benchmarked with NaI and BGO crystals by comparing our measurements with results in literature. Data from SMRC test samples is also shown.
N25-121: Functional Possibilities of Inorganic-Organic Hybrid Scintillator; Pr: LuAG Scintillator Covered with Plastic Scintillator.

A. Yoshikawa1,2, T. Yanagida1, Y. Yokota1, K. Kamada3, Y. Usuki3, M. Nikl4 1 IMRAM, Tohoku University, Sendai, Japan 2 New Industry Creation Hatchery Center (NICHe), Tohoku University, Sendai, JAPAN 3 Furukawa Corporation, Tsukuba, Japan 4 Institute of Physics, the Academy of Sciences of the Czech Republic, Prague, Czech Republic

Recent years, Pr: LuAG has attracted much attention due to its fast response as well as its highest energy resolution among heavy ion containing scintillators. However, it has emission peak around 310nm and therefore, glass window of normal PMT absorb the emission. Moreover, due to its short wavelength emission, semiconductor photo-detectors such as photodiode or avalanche photodiode (APD) are also not suitable for Pr: LuAG. In both cases the scintillation spectrum is not in full conformity with the spectral sensitivity of the photo detectors. In this study, we have investigate the effect of covering of Pr:LuAG with plastic scintillator aiming to examine possibilities of shifting of luminescence towards the region of higher spectral sensitivity of the photodetectors. Plastic scintillators, which have emission peak around 400-450nm with fast decay time, were selected as the candidate for the effective wavelength shifter of Pr: LuAG. 30 % increase of light yield is observed on the sample with 6planes covered with BC-499. It is concluded that the functional possibilities of inorganic-organic hybrid scintillator; Pr: LuAG scintillator covered with practice scintillator, got positive results.
N25-122:

G. A. Bizarri1, W.-S. Choong1, W. W. Moses1, L. Ahle2, N. Cherepi2, S. A. Payne2, S. Sheets2 1 Life Science, LBNL, Berkeley, USA 2 LLNL, Livermore, USA

Non-Proportionality in Alkali Halide Scintillators: Relationship to Energy Resolution Degradation

We recently proposed a model t describe the origin of non-proportional response of scintillators. This approach was developed around two separate steps, a semi-analytic treatment of the energy carrier density formed during the track creation (Generalized Oscillator Strength model), and a set of kinetic equations to model the energy transfer from the track to the emitting center. We now extend this model to quantitatively predict how the non-proportionality degrades the energy resolution. The fundamental assumption is that statistical fluctuations in the ionization density along the track (Landau fluctuations), coupled with nonproportionality (dependence of the luminous efficiency on ionization density) is the dominant cause of energy resolution degradation. This implementation is an important step bridging the electron response of a material to its energy resolution and allows a direct quantification of the microscopic mechanisms responsible of the energy resolution degradation. In this presentation, each step of the model will be detailed, discussed, and applied to alkali halide scintillators. Special attention will be given to NaI:Tl, CsI:Tl and CsI:Na scintillators. The similarity between the different channels responsible for energy loss, such as non-radiative Auger de-excitation, will be shown. We will emphasize the influence of impurities/traps on the proportionality 187

response. The role of each energy loss channel in the degradation of the energy resolution will also be discussed. The model allows us to move a step further toward understanding the non-proportionality and a quantitative prediction of the energy resolution versus energy.
This work is supported by the National Nuclear Security Administration, Office of Defense Nuclear Nonproliferation, Office of Nonproliferation Research and Development (NA-22) of the U.S Department of Energy and carried out at Lawrence Berkeley National Laboratory under Contract NO. AC02-05CH11231. N25-123:

A. Dominjon1, E. Chabanat1, P. Depasse1, R. Barbier1, J. Baudot2, W. Dulinski2, A. Dorokhov2 1 IPNL / University of Lyon & CNRS-IN2P3, Lyon, France 2 IPHC / University of Strasbourg & CNRS-IN2P3, Strasbourg, France

LUSIPHER : a New Large-Scale Ultra-Fast SIngle PHoto-Electron Recorder

We present a new Hybrid Photon Detector (HPD) called EBCMOS - for Electron Bombarded CMOS - sensitive to single photon with a micrometer range resolution and kHz frame rate. The CMOS which is a Monolithic Active Pixel Sensor (MAPS) is backthinned and passivated to be able to detect low energy photoelectrons. An EBCMOS demonstrator, EBMI5, was produced in 2007. This first prototype is sensitive to single photon with a spatial resolution of 27 m (FWHM) and a rate of 30 frames per second. This kind of photodetector is already a good candidate for single molecule detection in fluorescence microscopy but it was not dedicated to a fast temporal tracking. To overcome the limitations of the EBMI5, a new chip was fabricated in 2008 with a 0.25 m CMOS technology. This new sensor is composed of 400x800 pixels with a pitch of 10 m and microcircuits optimization to reach low noise (ENC = 8 electrons) at ambient temperature. To achieve a rate of 1000 frames per second, 8 parallel analog outputs have been implemented with a readout clock of 40 MHz. In order to optimize the quantum efficiency in a wide visible spectrum from green to red, a photocathode of type S25 was processed by PHOTONIS. Moreover, the gap between photocathode and sensor was reduced by a factor two compared with EBMI5 i.e. 600 m for improving the spatial resolution. With these new parameters, the photoelectrons detection threshold is measured around 3 keV. From measurements on LUSIPHER chip before tube integration, we have characterized the gain linearity versus high-voltage and the charge collection efficiency of the optimized diodes and the passivation process. We will report on the complete characterization of the new prototype of LUSIPHER: global gain, charge collection efficiency, spatial and energetic resolutions, dark count versus temperature and high-voltage. All these results will be discussed to demonstrate the potential of this detector concept for nano bio-photonic applications
N25-124:

Tertiary Scintillation Gas Proportional Scintillation Counter (TS-GPSC): First Experimental Results

L. P. M. M. Carita, C. A. N. Conde, F. P. Santos Physics Departament, University of Coimbra, Coimbra, Portugal We present a new concept for gaseous detectors: the Tertiary Scintillation Gas Proportional Scintillation Counter (TS-GPSC). The electric field induced secondary gas scintillation is first detected by a CsI-coated GEM-like structure and thus the photoelectrons are released and transferred through the GEM holes, with no charge multiplication, to another region where tertiary scintillation is produced and detected by a planar CsI-coated photocathode. Since there is no avalanche charge multiplication in the detector, the gain will be quite stable and so an improvement of the energy resolution is expected. The first prototype of the tertiary scintillation detector was tested in a xenon atmosphere and for hard X-rays and experimental results are presented.
N25-125: Empirical Estimation of Band Gaps, and Ce3+ and Eu2+ 4f-5d Absorption Energies for New Radiation Detector Materials

S. E. Derenzo, G. A. Bizarri Life Sciences, Lawrence Berkeley National Laboratory, Berkeley, CA, U.S.A The number of possible crystalline materials is vast and few have been explored as radiation detector materials. In this work we present an empirical approach for estimating band gaps, and Ce3+ and Eu2+ absorption energies of compounds using only their chemical formulas. The result would be useful in identifying candidate semiconductor detector materials (band gap between 1.5 and 2.5 eV) and activated scintillator detector materials (low band gap that is greater than the activator absorption energy). For each type of energy gap we assign a parameter to each anion and cation and vary those parameters for a best least squares fit between A-C and the measured values, where A and C are the average of the anion and cation parameters, respectively, in each compound. Four fits were performed: (1) 114 measured band gaps between 1 and 3 eV, (2) 190 measured band gaps between 3 eV and 8 eV, (3) 220 measured Ce3+ 4f-5d absorption energies, and (4) 180 measured Eu2+ absorption energies. The rms deviations between the best fit estimates and the measured values, and the correlation coefficients r for the fits were as follows: (1) rms = 0.34 eV, r = 0.82; (2) rms = 0.68 eV, r = 0.89; (3) rms = 0.30 eV, r = 0.89; (4) rms = 0.21 eV, r = 0.87. To test the 188

reliability of the estimations, a randomly chosen 10% of the measured values was excluded, the ionic parameters were fitted to the remaining 90%, those parameters were applied to the deleted set, and the rms deviations of the deleted set was calculated. When this process was repeated 70 times, the rms deviations of the deleted sets had the following average values: (1) rms = 0.44 eV, (2) rms = 0.83 eV, (3) rms = 0.34 eV, (4) rms = 0.24 eV. Note that these deviations are comparable to the differences in published values for many compounds.
This work was supported by the U.S. Department of Homeland Security and carried out at the Lawrence Berkeley National Laboratory under U.S. Department of Energy Contract No. DE-AC02-05CH11231. N25-126:

R. Pani1, M. N. Cinti1, R. Pellegrini1, P. Bennati2, S. Ridolfi1, V. Orsolini Cencelli3, F. de Notaristefani3, D. Sacco4, F. Navarria5, N. Lanconelli5, G. Moschini6, P. Boccaccio6, R. Scafe'7, G. De Vincentis1 1 Dept. of Exp. Medicine, Sapienza University of Rome and INFN, Rome, Italy 2 EDEMOM PhD shool, Roma Tre University of Rome and INFN, Rome, Italy 3 Dept. of Electronic Engineering, Roma Tre University of Rome and INFN, Rome, Italy 4 ISPESL and INFN, Monteporzio Catone, Rome, Italy 5 Dept. of Physics, University of Bologne and INFN, Bologne, Italy 6 Legnaro National Laboratory, INFN LNL, Padua, Italy 7 Casaccia, ENEA and INFN, Rome, Italy In this work we studied the performance of the Hamamatsu H8500 new series MA-PMT (H8500C-100 MOD) with reduced dynode stages (eight) and super bialkali photocathode (38% Q.E.). This tube represents the latest technological advance to improve energy and spatial resolution of gamma imagers. It is particularly suitable for applications involving new high light yield scintillators. To evaluate image performance improvement of this tube, we compared the energy and spatial resolution response with the analogous ones from a standard H8500 MA-PMT coupled to a LaBr3:Ce scintillation crystal with the same identity. All measurements were performed with a 64 channel electronic readout. An improvement of about 15% in spatial resolution has been found, as expected from the increased quantum efficiency. On the contrary no noticeable improvement in energy resolution was measured.
N25-127: Properties of Gd2O3:Eu3+ Nanoparticles and SiO2/Gd2O3:Eu3+Core/Shell Nanoparticles for Scintillation Detectors

Evaluation of Hamamatsu H8500 New Series MAPMTs for Readout of High-Resolution LaBr3:Ce Scintillation Crystal

T.-K. Tseng, J.-H. Choi, P. H. Holloway Dept of Materials Science and Engineering, University of Florida, Gainesville FL 32611-6400, USA Luminescent monoclinic Gd2O3 nanoparticles 10-15 nm in diameter and doped with Eu3+ were synthesized using a hot solution technique. For a reaction time of 1.5 hr at 320 C, photoluminescence (PL) from the Gd2O3:Eu3+ nanoparticles was from the 5D0-7F2 transition of Eu3+ at 609 nm. At a reaction time of 2.5 hr, the PL emission peak transformed from a singlet to a doublet, with maxima at 607 and 622 nm. Photoluminescence excitation (PLE) data showed that emission from Eu3+ results from excitation through the oxygen europium charge transfer band (CTB) between 240 and 280 nm. The quantum yield (QY) of nanoparticles with a singlet peak was 55 %, whereas for a doublet peak, it was only 15 %. The origin of the shape change of the PL peak decrease in quantum yield will be discussed. In addition to uncapped Gd2O3:Eu3+, nanoparticles with a SiO2 core and a Gd2O3 shell, where either the core or the shell was doped with Eu3+, were synthesized with a sol-gel process. Transmission electron microscopy (TEM) data showed that an ~15 nm Gd2O3 shell was successfully coated onto ~220 nm mono-dispersed SiO2 nanocores. Concentrations of nominally 5at% Eu3+ ions in Gd2O3 exhibited photoluminescent (PL) emission from the SiO2/Gd2O3 nanoparticles after being calcined at 8000C for 2 h. The SiO2 remained amorphous after calcining, while the Gd2O3 crystallized to a cubic structure. The PL emission spectra were still dominated by the 5D0-7F2 transitions of Eu3+ at 607 and 622 nm. Photoluminescence excitation (PLE) data showed that emission from Eu3+ could result from direct excitation, but was dominated by the oxygen to europium CTB, similar to the uncapped Gd2O3:Eu3+ nanoparticles. The QY from excitation via the CTB were 4-6% for SiO2:Eu3+/Gd2O3 and 17-20% for SiO2/Gd2O3:Eu3+ core/shell nanoparticles. Finally, the above core/shell nanoparticles were passivated with a shell of SiO2 to create SiO2/Gd2O3:Eu3+/SiO2 nanoparticles, but the QY was lower than for unpassivated nanoparticles.

189

N25-128:

Rare-Earth Activated Lutetium Aluminum Garnet Powders Synthesized via a Citrate-Nitrate

P. A. Cutler1, M. Zhuravleva1, D. Carey2, P. Szupryczynski2, M. Spurrier1, C. Melcher1 1 Scintillation Materials Research Center, University of Tennessee, Knoxville, TN, USA 2 Siemens Medical Imaging, Knoxville, TN, USA Semi-transparent, polycrystalline ceramics of cerium-doped lutetium aluminum garnet (LuAG) have been successfully demonstrated using preformed powders synthesized from a citrate-nitrate combustion process. [1] We have synthesized praseodymium-doped LuAG (Lu3Al5O12:Pr) and terbium- and cerium-codoped LuAG (Lu3Al5O12:Tb,Ce) powders using a combustion technique mixing metal nitrates and citric acid as a fuel. X-ray diffraction analysis confirmed a single phase garnet structure was formed in each powder. Luminescence data taken from these powders are compared to single crystals with the same nominal compositions. Fluorescence measurements of LuAG:Tb,Ce single crystal and powder show split emission of the 5d-4f transition of Ce3+ at 508nm and 543nm. Fluorescence decay time for LuAG:Tb,Ce was measured at 58ns for the single crystal and 62ns for the powder. Fluorescence measurements of LuAG:Pr powder closely match the spectra for the single crystal. Both samples exhibit 5d-4f and 4f-4f transitions characteristic of Pr3+ energy transitions. The powders will subsequently be pressed into transparent ceramics and luminescence data will be measured and reported.
[1] N. Cherepy, et al., Cerium-doped single crystal and transparent ceramic lutetium aluminum garnet scintillatorsNucl. Instr. and Meth. A 579 (2007) 3841 N25-129:

Method

Empirical Estimation of the Gamma-Ray Full-Energy Detection Efficiency of Cylinders and Slabs of Thickness 2.5 cm and Greater from 0.14 to 2 MeV

S. E. Derenzo, W.-S. Choong Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, U.S.A. We present an empirical formula for estimating for the full-energy gamma-ray detection efficiency of slabs and cylinders larger than 2.5 cm from 140 keV to 2 MeV. It can be applied to any material whose chemical formula and density are known, from CaF2 to PbWO4. Monte-Carlo calculations can be used for this purpose, but they require more computer time and resources than the formulas presented here. To provide control points for the empirical fitting procedure, we used Geant4 (version 4.9.0.p02 with the Penelope low energy electromagnetic code) to track the interactions of 10 million gamma rays for each point and determine the fraction that are fully absorbed in the cylinder or slab. The control points included cylinders and slabs of seven different thicknesses from 2.5 cm to 100 cm; six gamma-ray energies from 0.14 to 2 MeV; and 13 materials (BaF2, Bi4Ge3O12, CaF2, CsI, Cd9ZnTe10, Gd2SiO5, Ge, LaBr3, LiCaLaF6, Lu2SiO5, NaI, PbWO4, YAlO3). Two empirical formulas (cylinders and slabs) containing 20 free parameters were fit to the Monte Carlo control points. The rms deviations between the formulas and the control points was 0.026 for the cylinders and 0.027 for the slabs. The reliability of the empirical model was checked by deleting each compound in turn, fitting the model to the remaining 12, using that model to estimate the efficiencies of the deleted material, and then computing the rms deviation between that those efficiencies and the calculated efficiencies. The average of the 13 standard deviations was 0.028 for the cylinders and 0.025 for the slabs. This shows that if any compound is omitted from the fit, it could be reliably calculated from the best-fit formula that used only the other 12 compounds.
This work was supported by the U.S. Department of Homeland Security and carried out at the Lawrence Berkeley National Laboratory under U.S. Department of Energy Contract No. DE-AC02-05CH11231. N25-130:

A Thermal Neutron Detector Based on Corrugated Scintillator and Wavelength Shifting Fiber for Large Area Coverage
R. G. Cooper, R. A. Riedel, L. Crow, B. W. Hannan Neutron Scattering Division, Spallation Neutron Source, ORNL, Oak Ridge, TN, USA At the new high-powered neutron scattering facilities detector arrays can exceed 20 square meters in area. In addition spatial resolution of 5mm or less is required. To meet this need we developed a detector that uses corrugated LiF/ZnS:Ag scintillator with a wavelength shifting fiber readout. Each detector has an active area of 0.3 square meters with a spatial resolution of 50 x 5 mm. These detectors can be tiled for very large area coverage. To maximize neutron capture efficiency, position resolution, and rate capability of the detectors, several developments have been completed. These include enhancements to the scintillator, the fiber readout production techniques, the high speed single photon encoding electronics, and the pattern recognition software. The details of these enhancements will be presented on the poster.

190

N25-131:

E. B. Johnson1, C. J. Stapels1, M. McClish1, P. Dokhale1, S. Mukhopadhyay1, E. C. Chapman1, F. L. Augustine2, J. F. Christian1 1 Radiation Monitoring Devices, Inc., Watertown, MA, USA 2 Augustine Engineering, Encinitas, CA, USA Solid-state photomultipliers (SSPMs) are a compact, lightweight, potentially low-cost alternative to a photomultiplier tube for a variety of scintillation detector applications, including nuclear and medical-imaging applications. Manufacturing SSPMs with a commercial CMOS process provides the ability for rapid prototyping, and facilitates production to reduce the cost. On-chip integration of signal processing circuits is a distinct advantage of CMOS photodetectors beyond traditional phototubes. We will discuss the advances of the CMOS SSPM with integrated signal processing, which includes fabrication of large area (1 x 1 cm^2) devices and back illumination of thinned die.
N25-132:

Recent Developments for CMOS Solid-State Photomultipliers with Integrated Signal Processing

I. Sen1, D. Penumadu1, L. Miller2 1 Civil and Environmental Engg, University of Tennessee (UT), Knoxville,TN, USA 2 Nuclear Engineering, UT, Knoxville, TN, USA

Neutron Scintillator Detectors Based on Light Emitting Polymers

Title: Neutron Scintillator Detectors Based on Light Emitting Polymers Indraneel Sen, Dayakar Penumadu, Lawrence Miller A disadvantage in many existing neutron detectors is their inability to directly discriminate between neutron and gamma radiations. Such discrimination is obtained indirectly by pulse shape and pulse height analysis. Also most efficient detectors available commercially are ceramic glasses or gas filled detectors that are difficult to fabricate in large sizes. Existing detectors are structurally rigid and expensive to fabricate. What are desired therefore are devices that can be easily and economically synthesized/ fabricated in a wide range of volume, surface area and shapes. The other important criteria are efficiency of neutron detection and discrimination, and ease of implementation in public places. The uniqueness of light emitting polymers (LEP) is in the ease of forming them into flexible structures such as films, fibers, fabric etc. that emit visible light at high quantum efficiency from electroluminescence or photoluminescence. LEP can be easily blended with other polymers and additives that confer specific optical, electronic or structural properties to the solid solution. LEP can also be used in the form of a liquid scintillator dissolved in an organic or inorganic solution. Lithium6 based compounds either in molecular inclusions or in form of nanoparticles can be blended in a matrix of LEP and additives to form an effective neutron scintillation detector. Charged particles with high kinetic energy formed as Li6 nuclear capture reaction products, create secondary electrons and ions in the matrix of a detector which in turn activate the scintillator LEP. The current work presents LEP scintillator detectors that are economic to manufacture, have excellent gamma to neutron discrimination and have shown considerable light output. Synthesis of such detectors and their characterization are discussed in detail.
N25-133:

A. Osovizky1, D. Ginzburg1, M. Ghelman2, I. Cohen-Zada1, V. Pushkarsky1, E. Marcus2, A. Manor1, Y. Kadmon2, Y. Cohen2 1 Health Physics Instrumentation Department, Rotem Industries Ltd, Beer-Sheva, Israel 2 Electronics & Control Laboratories, Nuclear Research Center - Negev, Beer-Sheva, Israel The Silicon Photomultiplier is a novel and rapidly developing solid state optical sensor. In contradiction to its use for photon counting the operating conditions of the SiPM are of main concern when used as a light sensor in portable radiation detection device due to the gain dependence in temperature and operating voltage. The detection of low energy at high temperature requires an improvement of the Photon Detection Efficiency (PDE). An optimization of the PDE can be achieved by increasing the pixel size which improves the detector sensitive area fill factor. However, an increase in the pixel size reduces the dynamic range. This work describes the measurements and the results obtained for different photo-coupling configurations of a CsI(Tl) scintillation crystal with SiPM and the effect of various operating conditions on the instability of the gain. An optimization of the dominant parameters, such as noise level, resolution and dynamic range, is discussed and concluded. The dependence of these parameters in crystal dimension was examined in series of measurements using about 10 different crystal sizes and 3x3mm active area SiPM device. The energy equivalent noise level was measured over a wide temperature range and optimal operating voltage was determined. An investigation of the pixel size had been performed testing pixels from 35m to 50m to determine the most suitable to achieve the required dynamic range for energies up to 3 MeV. Reduction of sensor noise was approached with the coincidence detection method using two SiPM devices and a coincidence circuit. The improvement in noise level was tested in two configurations of the photo-sensors. The described comprehensive evaluation of the SiPM device showed the sensor performances in variety of configurations. The results emphasize the potential of the technology in radiation detection applications and the issues yet to be solved before it can present a viable alternative to the currently used technology.

Advanced Study of Novel Radiation Detector Based on Silicon Photomultiplier

191

N25-134:

J. Marchant1, B. Baumbaugh1, R. Ruchti1, B. Dolezal1, D. Howard1, J. Cleckler2, L. Hernandez2, C. Hurlbut2 1 Physics Department, University of Notre Dame, Notre Dame, IN, USA 2 Ludlum Measurements, Inc., Sweetwater, TX, USA Studies are presented of bulk PVT-based scintillators containing significant concentrations of the fluorescent dye pyrene. This material is known to form an excimer at elevated concentration, with fluorescence properties quite distinctive from that of individual pyrene molecules. The fluorescence excitation, emission, decay time, and scintillation efficiency have been studied over an extensive concentration range, from 5g/l to 225g/l. We observe significant decrease in fluorescence decay time and increase fluorescence efficiency with increasing concentration and light yield at 84% of EJ200.
N25-135:

PVT Scintillators with Very-High Fluorescent Dye Concentrations

L. A. Boatner1,2, J. S. Neal1, D. Wisniewski1,3, J. O. Ramey1, J. A. Kolopus1, B. C. Chakoumakos4, M. Wisniewska1,5, R. Custelcean6 1 Center for Radiation Detection Materials and Systems, Oak Ridge National Laboratory, Oak Ridge, TN, USA 2 Global Nuclear Security Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA 3 Institute of Physics, N. Copernicus University, Torun, Poland 4 Neutron Science Scattering Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA 5 Environmental College, Bydgoszcz, Poland 6 Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA We have recently discovered a new CeCl3(CH3OH)4 metal organic compound which represents the first example of a rare-earth metal-organic scintillator that is applicable to gamma-ray, x-ray, and fast neutron detection. This compound is only the initial member of an entire family of rare-earth-halide organic adduct compounds with potentially interesting scintillation properties for radiation detection applications in general and fast neutron detection applications in particular. Most recently, a new member of this family of metal organic scintillating materials, LaBr3(CH3OH)4:Ce, has been discovered and characterized in terms of its crystallographic structure and its scintillation properties using gamma-ray excitation. Solution-grown LaBr3(CH3OH)4 single crystals were used to determine the crystallographic structure of the base compound by means of single-crystal x-ray structural refinements, and the energy spectrum of a 4% cerium-doped lanthanum bromide-methanol adduct single crystal was obtained using 136 keV gamma-ray photons from a 1 micro Ci 57Co source. The LaBr3(CH3OH)4:Ce scintillator decay time for gammaray excitation was measured using the time-correlated single-photon-counting (Bollinger-Thomas) method. The decay time results yielded nominal values of 79 ns (45%) and 118 ns (55%) when corrections were made for the behavior at very short times. We have previously shown that large, crystals of the CeCl3(CH3OH)4 scintillator with dimensions up to 4 x 3.5 x 1.5 cm3 can be rapidly grown ( >24 hours) by slow-cooling methanol solutions that were first reacted with CeCl3. Measurements made using both Cf-252 spontaneous fission neutrons and 14.1 MeV neutrons from a D-T generator show that CeCl3(CH3OH)4 is of particular interest for combined fast-neutron and gamma-ray detection in a variety of national security and nonproliferation radiation detection activities.
N25-136:

New Scintillators for Combined Gamma-Ray/Fast Neutron Detection: Single-Crystal LaBr3(CH3OH)4:Ce and CeCl3(CH3OH)4

R. Dolenec1, H. Chagani1, S. Korpar1,2, P. Krizan1,3, A. Stanovnik1,4, R. Verheyden1 1 Dept. of Experimental Particle Physics, Jozef Stefan Institute, Ljubljana, Slovenia 2 Faculty of Chemistry and Chemical Engineering, University of Maribor, Maribor, Slovenia 3 Faculty of Mathematics and Physics, University of Ljubljana, Ljubljana, Slovenia 4 Faculty of Electrical Engineering, University of Ljubljana, Ljubljana, Slovenia

Time-of-Flight with BURLE/Photonis Multi-Channel MCP-PMT Using MCP-Out Signal

Multi-channel MCP-PMTs are considered as position sensitive detectors of single Cherenkov photons for the upgrade of particle identification systems at B-factories, where magnetic field tolerant sensors are required. They also exhibit excellent timing properties. In the case of proximity focusing RICH detector this allows to measure the time-of-flight employing the Cherenkov light produced in the sensor window. To reduce the number of electronic channels with high precision timing capabilities MCPout signal can be used to provide the timing information for the whole sensor. We have investigated the timing properties of anode and MCP-out signals on-the-bench and in a prototype used to detect cosmic rays.

192

N25-137:

Fast, Large Area CMOS Solid-State Photomultiplier for Radiation Detection

P. Dokhale, J. Christian, C. Stapels, E. Johnson, K. Shah Radiation Monitoring Devices Inc., Watertown, MA, USA Current and next generation experiments in nuclear and particle physics require sensors with fast response and high signal-tonoise ratio for detection of low intensity optical signals. Photomultiplier tubes (PMT) have been widely used for sensing light in most nuclear physics and imaging research experiments. PMTs, however, have several drawbacks that limit their use in several applications and technologies. Solid-state photomultipliers (SSPM) are compact, have high gain at low bias, fast response time and they are insensitive to magnetic fields gives a potential alternative to photomultiplier tubes for a variety of scintillation detector applications. A novel solid-state photomultiplier (SSPM) has been designed and developed by Radiation Monitoring Devices Inc., using standard CMOS technology. In this paper, we report performance of large area SSPM detector for spectroscopy and imaging applications. A detector was built by directly coupling a 6x6x5 mm3 LYSO scintillator to the large area (36 mm2) SSPM. Energy, co-incidence timing, linearity and imaging performance of the detector was evaluated. Energy resolution measured for 661.7 keV gamma rays was 9.6% (FWHM). The timing resolution measured against LYSO-PMT detector with 511 keV gamma rays (22Na) source was 700 ps. The position sensitive SSPM (PS-SSPM) was also designed, built and evaluated. A flood image was recorded with a 4x4 LYSO array (each LYSO element measuring 1.5mm x 1.5mm x 20mm) coupled to a PS-SSPM with 36 mm2 active area. All 16 LYSO elements were clearly visible and well separated from each other in the flood image. We have also studied the performance of the 36 mm2 SSPM when coupled to 5x5x3 mm3 CsI(Tl) scintillator. The energy resolution measured with CsI(Tl) scintillator for 661.7 keV gamma rays was 7.2% (FWHM).
N25-138:

W. Klamra1, T. Szczesniak2, M. Moszynski2, J. Iwanowska2, L. Swiderski2, A. Syntfeld-Kazuch2, V. Shlegel3, Y. Vasiliev3, E. Galashov3 1 Royal Institute of Technology, Stockholm, Sweden 2 Soltan Institute for Nuclear Studies, Otwock-Swierk, Poland 3 Nikolaev Institute of Inorganic Chemistry, Novosibirsk, Russia Two samples of 10x10x10 mm3 CdWO4 (CWO) and ZnWO4 (ZWO) scintillators were studied at room and liquid nitrogen temperature. At room temperature (RT) the crystals were coupled to a Photonis XP3212 photomultiplier. During experiments in liquid nitrogen (LN2) a Large Area Avalanche Photodiode (LAAPD) was used as a photodetector. Measurements of light output, non-proportionality of the light yield and intrinsic resolution as a function of gamma-ray energies were performed at both temperatures. Differences between non-proportionality curves in low energy range, below 100 keV for RT and LN2 have been observed, especially pronounced in case of CWO. Intrinsic resolution for both crystals was also changed at LN2 comparing to the data collected at RT. The intrinsic resolution was improved below 100 keV and deteriorated for higher energies. In contrast to well known BGO the tested CWO and ZWO crystals showed temperature dependence of non-proportionality and intrinsic resolution.
N25-139:

Properties of CdWO4 and ZnWO4 at Liquid Nitrogen Temperature

G. Collazuol1, A. Del Guerra2, G. Bisogni2, C. Piemonte3, S. Marcatili2, G. Llosa2 1 Scuola Normale Superiore, Pisa, Italy 2 Physis, Univerista` degli stui di Pisa, Pisa, Italy 3 IRST, Fondazione Bruno Kessler, Trento, Italy

Studies of Silicon Photo-Multipliers at Cryogenic Temperatures

The Silicon Photo-Multiplier (SiPM) is a semiconductor device consisting in a matrix of avalanche photodiodes operated in Geiger mode. High gain (~10^6) and PDE (> 80%) in detecting low photon fluxes with extreme single photon charge and time resolution (~50ps), insensitivity to magnetic fields and EM pickup, make SiPM an alternative to vacuum PMT. SiPM based detectors are becoming common in High Energy, Medical and Space physics, in particular where multi-photon detection is involved. In contrast, the use of SiPMs for few-photon applications (e.g. Cherenkov) is difficult for their high dark count rate, due to single carriers. Cooling is crucial for reducing the dark rate by several orders of magnitude. We investigated the behavior of different SiPM devices at different temperatures (T) in the range between 30K and 350K. Breakdown voltage, dark noise, after-pulsing, cross-talking, signal shape, gain and photon detection efficiency were studied as a function of T and as a function of over-voltage at fixed T. We found that the breakdown voltage is decreasing at low T in fair agreement with Baraff model. Dark count rate is also decreasing at low T. It is dominated by field-enhanced generation-recombination noise above 200K, while at lower T band to band tunnel noise is the main contribution in agreement with Hurkx model. Two exponential waveform components with different time constants can be discriminated in the signal waveform. The pixel capacitance doesn't show any dependence on T, which is the reason for the gain being constant at fixed over-voltage. Cross-talk shows no dependence on T. After-pulsing probability is increasing at low T,due to slower trapping time. The measurements in will be discussed in the talk with emphasis on the analysis, performed by comparison with the related

193

models. The quantitative agreement of data with models is promising for the possibility of tuning the processing parameters in order to reduce the dark noise in the future devices
N25-140:

E. D. Bourret-Courchesne1, G. Bizarri2, S. M. Hanrahan2, G. Gundiah2, S. E. Derenzo2 1 Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley CA, USA 2 Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley CA, USA

Growth and Scintillation Properties of Eu Activated BaXY (X,Y=Br,Cl,I)

Most Eu- activated Ba mixed halides are known photo-stimulated phosphors obtained in crystalline powder form by solid-state synthesis. Among those the Ba iodohalides BaBrI and BaClI, and the bromochloride BaBrCl crystallize in the orthorhombic structure with no weak planes. We investigated these compounds for their scintillation properties. BaBrI has the lowest band gap (estimated value of Eg = 5.3 eV). Crystals of Eu-activated BaBrxI2-x have been obtained by the vertical Bridgman technique from a stoichiometric mixture of the binary halides in sealed quartz ampoule. The melting point was determined by differential thermal analysis at 765C for x=1. The scintillation properties of Eu-activated BaBrxI2-x as a function of Eu concentration and as a function of the relative amount of halides will be presented. BaBrI has the lowest band gap (estimated value of Eg = 5.3 eV). BaBrI:Eu2+, with an estimated density of 5.0, is found to be a very bright scintillator with an estimated light output of about 80,000 photons per MeV and 98% of the luminescence decaying in less than 500 ns. The decay of the luminescence in the Ba mixed halides are found to be relatively fast compared to that of other Eu-activated scintillators such as SrI2:Eu2+. The response of BaBrI:Eu to gamma irradiation will be discussed and the results compared to other scintillator such as SrI2:Eu2+. This work was supported by the U.S. Department of Homeland Security and by U.S. Department of Energy/NNSA/NA22 carried out at the Lawrence Berkeley National Laboratory under U.S. Department of Energy Contract No. DE-AC02-05CH11231.
N25-141:

G. U. Pignatel1,2, G. Ambrosi1, P. Azzarello1, R. Battiston1, G. DiLorenzo2, M. Ionica1 1 Physics Department, National Institute of Nuclear Physics, Perugia, Italy 2 Electronic and Information Eng., University of Perugia, Perugia, Italy

Time Resolving Characterization of HPK and FBK Silicon Photomultipliers for TOF and PET Applications

In Time-of-Flight measurements, or Positron Emission Tomography experiments where two gamma rays are emitted in coincidence, the time resolution of the photon detector is of primary importance. SIPMs are very promising devices for these applications, since their intrinsic response time is very short, typically less than 1 ns. However the actual timing resolution of SIPMs is affected by the area (capacitance) of the device, by the type of electronics used to pre-amplify the signal, by the dark count rate which is detected as pure noise, and other second order effects like cross-talk and after dark pulsing. In this work we report the characteristics of different samples of HPK (Hamamatsu Photonics) and FBK (Fondazione-BrunoKessler) SIPMs, with pixel size ranging from 40 to 100 micron. In particular, we have investigated their time response when stimulated with O(100) ps pulsed laser with wavelength in the range 400 - 800 nm. SIPM performances are also compared with that of fast PIN diodes characterized with the same set-up.
N25-142:

Detection of Ionizing Radiation in Coherent Plates of Scintillating Optical Fibers

J. Marchant, M. McKenna, R. Ruchti, D. Karmgard, B. Baumbaugh, M. Vigneault Physics Department, University of Notre Dame, Notre Dame, IN, USA We have been developing intensified imaging systems using coherent fiber-optic structures of scintillation materials detection of trajectories of ionizing particles and for usage in formal and informal education settings. Functional devices have been fabricated using scintillating glass materials based on Tb2O3, and development of techniques to fabricate larger structures based on doubleclad polystyrene fiber are presented.
N25-143:

G. Gundiah1, E. D. Bourret-Courchesne2, A. Chaudhry1,3, A. Canning4, W. W. Moses1, S. E. Derenzo1 1 Life Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 3 Department of Electrical and Computer Engineering, UC Davis, Davis, CA, USA 4 Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA The past decade has witnessed enormous research efforts to develop fast scintillators based on Ce3+-doped oxides and halides. Besides Ce3+, the dipole allowed f to d transitions are also observed in Eu2+-doped compounds with luminescence decay within 1000 ns. Single crystals of Eu2+ activated SrI2 [1] and BaX2 (X=Cl, Br, I) [2] have been shown to have a high light yield, fast response and good energy resolution. 194

Scintillation Properties of Eu2+-Activated Barium Fluoroiodide

In our quest for fast and bright Eu2+-scintillators that are easy to grow and stable under ambient moisture conditions, we have investigated the mixed halide system BaFI [3]. In the 1980s, BaFI:Eu2+ was developed as a photostimulable X-ray phosphor for computed tomography [4]. We performed first-principles electronic structure calculations for Eu-doped BaFI and the relative position of the 4f and 5d levels for Eu2+ with respect to the energy gap of the host were determined. The localization of the electron in the excited state indicated this as a potential scintillator. Powder samples were synthesized by a solid state reaction. BaFI has a tetragonal, layered structure and small crystals were obtained by the vertical Bridgman technique in sealed, carboncoated quartz tubes. Under optical and X-ray excitation, the samples exhibit a narrow Eu2+ emission centered at 405 nm. The scintillation light output was determined to be ~45,000 photons/MeV at 551 keV, with 90% of the light having a decay time of less than 600 ns. An energy resolution of ~8% full width at half maximum (FWHM) has been achieved using this scintillator for 551 keV (22Na source) at room temperature. We present details of the synthesis and luminescent properties.
References: [1] N. J. Cherepy et al., Appl. Phys. Lett. 2008, 92, 083508. [2] J. Selling et al., IEEE. Trans. Nuc. Sci. 2008, 55, 1183. [3] G. Gundiah et al., manuscript under preparation. [4] M. Sonoda et al., Radiology 1983, 148, 833. Acknowledgments:This work was supported by the U.S. Department of Homeland Security and carried out at the Lawrence Berkeley National Lab under Department of Energy Contract No. AC02-05CH11231. N25-144:

R. Suarez1, J. H. Ely1, J. C. Hayes1, K. Scott2, J. I. McIntyre1, B. T. Schrom1, R. M. Williams1 1 National Security, Pacific Northwest National Laboratory, Richland, WA, USA 2 Sandia National Laboratory, Livermore, CA, USA The long-term stability of radiation detectors used in unattended fielded systems is important for system performance, accurate source calculations, and minimal technical support. We have characterized a cerium doped YAP scintillating crystal over an extended time period to test the long term stability of the scintillator, photomultiplier-tubes (PMTs), high-voltage supply, and data acquisition readout system. The metrics measured and analyzed were the efficiency, gain drift, and energy spectral features that should remain constant. An experimental campaign conducted for approximately 520 days in length demonstrated the ability of YAP, with associated electronics, to be stable and follow the expected decrease in count rate due to radioactive source decay. No evidence was found which would suggest a degradation of the YAP material. This paper analyzes and presents, in detail, the results from this experiment.
N25-145:

Long-Term Stability Analysis of Yttrium Aluminum Perovskite (YAP)

Absolute Measurements of Intrinsic Scintillation Light Yield

J. S. Salacka, M. K. Bacrania, M. P. Croce, E. A. McKigney Safeguards Science and Technology, Los Alamos National Laboratory, Los Alamos, NM, USA Substantial research effort is being directed at developing new high-performance scintillating materials for nuclear nonproliferation and homeland security applications. An essential part of this materials development research is the comprehensive characterization of the radiation response of these materials. We will present an overview of a comprehensive method that we have developed to make precise, absolute measurements of the intrinsic light yield of scintillating materials. Our method incorporates a number of measurements of the materials optical transport properties, including a direct measurement of the light collection efficiency of the scintillator (i.e. the probability of a scintillation photon escaping the material). Knowledge of the light yield of a material can to improved understanding of the underlying physics driving the scintillation process, and can also provide information for simulating and optimizing the design of scintillating detectors for a variety of applications.
N25-146:

Low Energy Measurements Using the CsI(Tl) Crystal Coupled to a Photodiodes Linear Array in Coincidence-Sum Circuitry.

C. H. de Mesquita, T. Madi Filho, M. M. Hamada Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP, Sao Paulo, SP, BRAZIL The detector of CsI(Tl) coupled with PIN photodiode constitutes one of the most promising technologies to measure gamma and X-Ray radiations. Compared with PM tubes, photodiodes offer the advantages of higher quantum efficiency, lower power consumption, compact size, improved ruggedness and they are practically insensitive to magnetic fields. However, electronic noise of the photodiode-CsI system is very influenced by the size of the photodiode surface due to its intrinsic capacitance. Thus, the system should be built with crystal and photodiode of small surface sizes, around 1x1 cm2. This kind of assembly is suitable for low energy radiation. Unfortunately, in such case, especially for low energy, like 125I (approximately 27keV) the electronic noise is responsible for the loss of the performance of the measurements. This limitation can be overcame using small photodiode and crystal, but this arrangement results low geometrical efficiency. The proposal of this work is to develop a low energy measurement system using a CsI(Tl) crystal bar coupled to a linear photodiode array, connected to charge sensitive preamplifiers and applied to a circuit of coincidence and sum of signs. This detector system is capable to reduce the electronic noise severely. In this manner it is possible to measure efficiently a low energy of 125I radiation, presenting a low signal/noise ratio. In addition, 195

the authors make considerations on the geometric aspects of the crystal size; its effect concerning the energetic resolution and the photodiode protection on radiation damage and the potential applications.
The authors express their gratitude to CNPq - Brazilian Council for Scientific and Technological Development and IAEA - International Atomic Energy Agency for the financial support. N25-147:

C. S. Williams1, R. C. Slaughter1, L. W. Burggraf1, P. E. Adamson2, M. Ross1 1 Engineering Physics, The Air Force Institute of Technology, WPAFB, OH, USA 2 Directed Energy/High Power Microwaves Division, Air Force Research Laboratory, Kirtland AFB, NM, USA

Positron Annihilation Lifetime Spectrometry (PALS) of the Positronic Dodecaborate Anion

A fast-fast PALS system incorporating two BaF2 scintillation detectors was characterized. This system was used to measure positron annihilation lifetimes and intensities for solutions of potassium dodecahydrododecaborate and carrier-free 22NaCl. Influence of electron scavenging and competition by nitrate ions was measured for various concentrations of potassium nitrate. Lifetimes and intensities were fit to a simple steady-state kinetic model.
N25-148:

8x4 SiPM Array Based Gamma Detectors for a PET/MR Fusion Imaging Application

C. Lee, H. Kim, J. Kim, B. Cha, J. Bae, J. Kim, C. Kim, G. Cho Nuclear & quantum engineering, Korea advanced institute of science and technology, Daejeon, Korea SiPM is a promising device to replace conventional PMTs which have been used in many research fields such as neutron measurements, high energy physics, low intensity light detection, PET applications. In PET applications, SiPMs can be used in high magnetic fields due to their insensitivity to magnetic fields, so simultaneous PET/MR imaging is possible with SiPMs. In this study, 8 x 4 SiPM array were designed and fabricated on 4 um thick p-type epitaxial wafers in NNFC (National Nano-Fab Center). There are 2,500 micro-cells in a 2.3 mm SiPM pixel whose operating voltage is around 32 V. A SiPM array having 2.3 mm pixel pitch was designed with poly silicon quenching resistors having high sheet resistance in order to improve its fill factor, and implantation conditions was optimized with a TCAD simulation tool to maximize avalanche triggering probability. Dark count rates (DCR), and Photon Detection Efficiency (PDE) including fill factor, triggering probability, and quantum efficiency were calculated from simulated and designed data. After fabrication and ceramic packaging, measured DCR, and PDE was compared to designed DCR and PDE. Energy resolution with Cs-137 and spatial resolution were measured with a LSO scintillation crystal and timing resolution was also measured in order to verify Time-Of-Flight (TOF) capability in PET application. Fabricated devices were also tested inside MRI facility. Obtained data were compared with measured data in off state of a MRI facility. Detailed data will be presented in the conference.
This study was supported by Radiation and Nuclear Medical Engineering Research at KAIST. N25-149:

Garnet Scintillator-Based Devices for Gamma-Ray Spectroscopy

O. B. Drury, N. J. Cherepy, T. A. Hurst, S. A. Payne Lawrence Livermore National Laboratory, Livermore, CA, USA We are developing scintillator based -ray spectrometers for safeguard applications. Improving the energy resolution of such scintillator detectors enables the development of more accurate and effective instruments. We have characterized the response of several Ce doped garnet scintillators using radioactive sources, coupling to both avalanche photodiodes and photomultiplier tubes. We have measured FWHM energy resolution below 4% at 662 keV and investigated the effects of size and choice of photodetector on the energy resolution. We present the results that will further guide designs for scintillator detectors of different sizes and applications.
This work was supported in part by the Domestic Nuclear Detection Office in the Department of Homeland Security (A. Janos) and in part by the National Nuclear Security Administration, Office of Nonproliferation Research and Development (NA-22, D. Beach) of the U.S. DOE under Contract DE-AC03-76SF00098, and performed under the auspices of the U.S. DOE by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

196

N25-150:

A. Chaudhry1,2, A. Canning3, R. Boutchko1, G. Gundiah1, E. D. Bourret-Courchesne4, S. E. Derenzo1, N. Gronbech-Jensen3,5 1 Life Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 Electrical and Computer Engineering, University of California Davis, Davis, CA, USA 3 Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 4 Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 5 Applied Science, University of California Davis, Davis, CA, USA This work presents the results of first principles electronic structure calculations for Ce3+/Li+ co-doped Ba Silicates (BaSiO3, BaSi2O5, Ba2SiO4, Ba2Si3O8, Ba5Si8O21) performed using the pseudopotential method based on the GGA+U implementation in density functional theory. Co-doping with Ce3+/Li+ allows us to dope Ce3+ onto a 2+ site such as Ba2+ thus greatly increasing the number of host materials that can be studied as candidates for bright Ce activated scintillators. The positions of the Cerium 4f and 5d states relative to the valence band maximum and conduction band minimum of the host material are determined. Qualitative predictions of the brightness of Ce3+ activated scintillation in the doped material is made based on the following criteria: (1) The size of the host material bandgap (2) The energy difference between the Valence Band Maximum (VBM) of the host material and the Ce3+ 4f level (3) The energy difference between the occupied Ce 5d excited state (Ce3+)* and the host material Conduction Band Minimum (CBM) (4) The level of localization of the excited (Ce3+)* electron state on the Ce atom. These materials have been synthesized in microcrystal form and preliminary data agrees qualitatively with our predictions.
Research supported by the U.S. Department of Homeland Security and is carried out at the Lawrence Berkeley National Laboratory under U.S. Department of Energy under Contract No. DE-AC02-05CH11231. N25-151:

First-Principles Calculations for Ce/Li Co-Doped Ba Silicates and Experimental Validation

J. Zhu1,2, Z. Zhang1, B. Zhang1,2, M. Niu1,2, T. Xu1, X. Zhang3, Q. Xie1,2 1 Department of Biomedical Engineering, Huazhong University of Science and Technology, Wuhan,Hubei, China 2 Wuhan National Laboratory for Optoelectronics, Wuhan,Hubei, China 3 Institute for Pattern Recognition & Artificial Intelligence, Huazhong University of Science and Technology, Wuhan,Hubei, China

Investigation of Timing Resolution and Energy Resolution for SiPM/PET Detectors Using the Silicon Flexible Optical Material

Silicon photomultipliers (SiPMs) attract extensive attention for detecting optical photons in high energy physics and medical imaging due to its high gain, high photon detection efficiency (PDE), low operation voltage and fast timing response. We use the silicon flexible optical material to transform the Gaussian distribution of incident light intensity into uniform in space, making the incident photons being detected by SiPM equally. In this way, we can make full use of all cells of SiPM, and more cells operating means more photons being detected for a certain pulse, which can increase the count rate of the incident photons, and improve the detection efficiency of SiPM. Furthermore, by comparing the output of SiPM in different light intensity input, we can find out the best light intensity fit for SiPM and the suitable crystal and surface treatment for positron emission tomography (PET) imaging based on SiPM. In primary experiment, we use the laser pulse as SiPM input since that its light intensity expressed as Gaussian distribution in space, and analyze the readout of SiPM using the silicon flexible optical material or not. The result is consistent with our expectation. By the use of the silicon flexible optical material, the timing resolution and energy resolution of SiPM become better than without it, when the light intensity of input is appropriated. Considering the difference between the laser output and scintillation pulse, we will apply the silicon flexible optical material to SiPM/PET detectors by coupling it to crystal directly, and evaluate its effect on the timing resolution and energy resolution in PET imaging.
N25-152:

Evaluation of the Impact of Dark Counts in Silicon Photomultiplier Multi-Readout Applications

I. F. de Castro, A. J. Soares, J. F. Veloso I3N, Physics Department, University of Aveiro, Aveiro, Portugal Noise is one of the most critical performance drawbacks of the Silicon Photomultiplier (SiPM), which has many promising features, such as high internal gain and quantum efficiency, low operational voltage and insensitivity to magnetic fields. Dark counts are highly dependent on temperature and bias voltage, increasing linearly with them. Due to the Geiger operational mode of this device, dark counts take the form of a full photo-electron avalanche. For low light level applications the dark count pulses are identical to single photon interactions and will therefore add random noise to the light sensing device. This study presents several techniques to reduce the effects of dark counts. In particular, it will focus on applications where multiple SiPM sensors are used to detect position of interaction in low light level imaging applications.
Acknowledgements: This work was supported by project GAMACAM through FEDER and ADI (Lisbon) programs.

197

N25-153:

Test in Liquid Argon of the Light Readout System for the ArDM Experiment.

V. Boccone Physik Intitut, University of Zurich & CERN, Zurich, Switzerland On behalf of the ArDM Collaboration ArDM is a new-generation WIMP detector which will measure simultaneously light and charge from scintillation and ionization of liquid argon. Our goal is to construct, characterize and operate a 1 ton liquid argon (LAr) underground detector. The project relies on the possibility to read out the VUV scintillation light and to extract the electrons produced by ionization from the liquid into the gas phase of the detector, to amplify and read out with Large Electron Multipliers detectors. The light has to be converted with wavelength shifters such as TetraPhenyl Butadiene in order to be detected by photomultipliers with bialkali photocathodes. I'll describe the light readout system R&D and report about the tests and the calibration of the prototype with liquid argon in the full size detector.
N25-154:

Highly-Scaled SRAM Immunity to MUN Based on Analysis of an Induced Parasitic Bipolar Effect

L. Liu Design, Beijing Microelectronics Technology Institute, Beijing, China Three-dimensional TCAD simulation is used to explore a new charge-collection mechanism in highly-scaled MOSEFT. The results show the charge collection with the parasitic bipolar conduction can cause an increased SEU sensitivity. Then the problem of multiple-node upset in a 0.18m 12-T SEU hardened SRAM cell is also studied .Unlike traditional multiple node charge collection in which diffusions near a single event strike collect the deposited carriers, a novel mechanism involves direct driftdiffusion collection at a NFET transistor in conjunction with parasitic bipolar conduction in nearby PFET transistor. The charge collection with the parasitic bipolar conduction compromise the SEU hardened design, thus causing upsets.
N25-155:

Contributions of Electrons and Holes to Total Collected Charge in Heavily Irradiated Si Pad and Strip/Pixel Detectors: a Comparison Simulation Study

Z. Li Instrumentation Div., Brookhaven National Lab, Upton, NY, USA A simplified approach to compare quantitatively the contributions of electrons and holes to the total collected charge in heavily irradiated Si pad detectors and strip/pixel detectors has been developed in this study. By applying a step function to approximate the weighting field and a step function to approximate the linear- and/or double junction- electric field, in the detector, one can obtain analytical solutions of total collected charge and contributions from electrons and hole for irradiated Si detectors with various electric field and weighting field profiles combinations. Although the results do not exactly replicate the situation in a real detector, they qualitatively and quantitatively explain the contributions of electrons and holes in various detectors with different segmentation- and field-profiles.
N25-156:

Space Charge Analysis in Irradiated Silicon Detectors Using Admittance Spectroscopy

C. Betancourt, G. Alers, M. Gerling, R. F. Hurley, S. Lindgren, H. F. -W.Sadrozinski, S. Sattari SCIPP, University of California Santa Cruz, Santa Cruz, Ca, USA Electrical characteristics of irradiated silicon detectors are examined using Admittance Spectroscopy. Admittance measurements taken at several frequencies, bias voltages and temperatures reveal important characteristics of deep level traps created as a result of irradiation [1]. Energy levels, majority carrier capture cross sections and lifetimes can be inferred from conductance measurements alone. Depleted bulk capacitance, and hence active thickness of the device as well as deep level trap concentrations can be determined from both capacitance and conductance measurements. Measurements were performed at various temperatures to extract activation energies. Depleted depth as a function of accelerated anneal time was calculated. When possible, the depths were compared to Charge Collection Efficiency (CCE) and good agreement is generally observed.
N25-157: The Effect of Silicon Nitride Passivation on the Electrical Properties of Neutron and Electron Radiated AlGaN/GaN HFETs

J. W. McClory, J. C. Petrosky, J. T. Moran Department of Engineering Physics, Air Force Institute of Technology, WPAFB, OH, USA Silicon nitride passivated and unpassivated AlGaN/GaN HFETs were electron and neutron irradiated and transistor and gate leakage measurements performed. Passivation reduces the overall radiation effect on gate voltage shift while increasing the effects of gate leakage. 198

N25-158: Annealing of Charge Collection Efficiency and Depletion Voltage in Proton Irradiated Silicon Detectors

R. F. Hurley1, B. Colby1, C. Betancourt1, G. Bredeson1, N. Dawson1, V. Fadeyev1, M. Gerling1, S. Lindgren1, P. Maddock1, H. F.-W. Sadrozisnki1, J. von Wilpert1, M. Hoeferkamp2, J. Metcalfe2, S. Seidel2 1 SCIPP, University of California Santa Cruz, Santa Cruz, Ca, USA 2 Department of Physics and Astronomy, University of New Mexico, Albuquerque, NM, USA

As part of the program to develop radiation-hard semiconductor detectors for the tracker regions at a future high luminosity hadron collider, we have irradiated n-type and p-type magnetic Czochralski and Float Zone silicon detectors to proton fluences up to 1.3*10^15 neq/cm^2. These fluences are predicted to be in excess for all strip layers of the LHC Upgrade detectors. The data are collected right after irradiation and after elevated temperature annealing at 60 degrees C, corresponding to several years of annealing at room temperature. As a function of bias voltage V, the following electrical parameters were measured: C-V at room temperature to extract the depletion voltage Vdep, and the charge collection efficiency in a beta source at lowered temperature to determine the efficiency voltage, i.e. the voltage at which the sensor becomes efficient at a threshold of 1 fC. The comparison of these parameters allows insight into the usefulness of planar Si sensors both as strip and pixel sensors at the LHC upgrades, and into questions of detector properties like type inversion and trapping.
N25-159:

A. K. Srivastava1, D. Eckstein1, E. Fretwurst1, R. Klanner1, V. Khomenkov1, P. Schleper1, J. Schwandt1, G. Steinbrck1, L. Feld2, W. Karpinski2, K. Klein2, J. Sammet2, M. Bergholz3, K. Borras3, G. Eckerlin3, W. Lange3, A. Mussgiller3, W. D. Boer4, P. Blm4, A. Dierlamm4, J. Erfle4, M. Frey4, F. Hartmann4, K. H. Hoffmann4, T. Mller4, H. J. Simonis4, E. C. Gil5, V. Lemaitre5, O. Militaru5, M. Fernandez6, F. J. Gonzalez6, R. Jaramillo6, A. L. Virto6, D. Moya6, T. Rodrigo6, I. Vila6, T. Bergauer7, M. Dragicevic7, M. Friedl7, S. Hnsel7, J. Hrubec7, M. Krammer7, J. Vaitkus8, E. Gaubas8, V. Kazukauskas8, S. Sakalauskas8, J. Storasta8, E. Zasinas8, J. Marczewski9, P. Grabiec9, K. Kucharski9, M. Grodner9, D. Tomaszewski9 1 Institute for Experimental Physics, University of Hamburg., Hamburg, Germany 2 Aachen University., Aachen, Germany 3 Deutsches Elektronen-Synchrotron DESY., Hamburg, Germany 4 Institut fr Experimentelle Kernphysik, Universitt Karlsruhe (TH)., Karlsruhe, Germany 5 Universit catholique de Louvain., Louvain La Neue, Belgium 6 Instituto de Fsica de Cantabria (UC-CSIC)., Santander, Spain 7 Institut fr Hochenergiephysik der sterreichischen Akademie der Wissenschaften, HEPHY., Vienna, Austria 8 Institute of Materials Science and Applied Research, Vilnius University., Vilnius, Lithuania 9 Institute of Electron TechnologyAl. Lotnikow 32/46, 02-668, Warsaw, Poland The extremely high luminosity of 1035 cm-2 s-1 of the foreseen upgrade of the LHC at CERN, the SLHC, will lead to an unprecedented high particle rate and radiation levels up to 1016 neq./cm2 in the innermost part of the tracking areas of the experiments. Therefore, tracking systems with very high granularity are mandatory. On this basis, a new sensor geometry with silicon strip sensors with short strips (strixels) is proposed for the upgrade of the CMS tracker. The R&D work described here is carried out in the framework of the Central European Consortium (CEC) for the development of silicon sensors for the upgrade of CMS tracker for the SLHC. Evaluation prototype Si sensors are currently fabricated at Hamamatsu Photonics (HPK), Japan. In order to understand the behaviour of such devices, test geometries are developed whose performance has to be optimized using 2-D and 3-D simulation of silicon sensors. We used the Synopsis TCAD device simulator to simulate the electric field and electrostatic potential distributions in the device, and current-voltage (I/V) and capacitance-voltage characteristics (C/V) of different strixel geometries. The noise performance of the various proposed design options is evaluated with particular reference to the interstrip capacitance. In this paper, we also describe a proposed new readout architecture of double metal layer Si strixel sensors. Their performance in the expected radiation environment is also simulated. This paper addresses the challenges these sensors face in an SLHC environment, based on several corresponding simulations.
Corresponding Author: Ajay Kumar Srivastava, Institute for Experimental Physics, University of Hamburg, Germany, Luruper Chaussee 149, D22761 Hamburg, DESY Bldg.,Room No. 67b/24, Tel: 040-8998-4726, Fax No. 040-8998-2170, Email: [email protected],[email protected]

Strixel Sensor Design for Large Radii of a New CMS Tracker for SLHC

199

N25-160:

A. F. Rakov1, R. H. Bartram2, A. K. Islamov1, C. L. Melcher3, U. S. Salikhbaev1 1 Dept. of Radiation Physics, Institute Nuclear Physics, Uzbekistan, Tashkent, Uzbekistan 2 Dept. of Physics, University of Connecticut, Stors, CT, USA 3 Scintillation Materials Research Center, University of Tennessee, Knoxville, TN, USA In this work we present the results of a thermoluminescence study on several oxide crystals, including Y3Al5O12 (YAG), Y3Al5O12:Nd (YAG:Nd), Lu2SiO5:Ce (LSO:Ce), Y2SiO5:Ce (YSO:Ce), Gd2SiO5:Ce (GSO:Ce), PbWO (PWO), and PbWO:La (PWO:La). A novel phenomenon involving restoration of thermoluminescence (TL) glow peaks is found to occur in some of the crystals investigated; crystals -irradiated at room temperature and subsequently stored for some time in the dark at 77K exhibit TL glow peaks in the range below room temperature. This phenomenon is caused not by a thermally or optically stimulated process, but rather as a by-product of a tunneling process. The intensity of the restored TL glow peaks measured in LSO:Ce crystals is found to be proportional both to the radiation dose and to the storage time at low temperature. A phenomenological theoretical model is proposed in which tunneling recombination occurs between deep electron and hole traps accompanied by the simultaneous ejection of an electron to the conduction band; some of these conduction electrons then repopulate shallow traps.
N25-161:

Restored Thermoluminescence in Oxide Crystals

A. F. Rakov1, R. H. Bartram2, N. I. Faleev1, A. K. Islamov1, C. L. Melcher3 1 Dept. of Radiation Physics, Institute of Nuclear Physics, Uzbekistan, Tashkent, Uzbekistan 2 Dept. of Physics, University of Connecticut, Stors, CT, USA 3 Scintillation Materials Research Center, University of Tennessee, Knoxville, TN, USA The phenomenon of the restoration of the thermoluminescence glow peaks observed in some oxide crystals (see A.Rakov et al. Restored thermoluminescence in oxide crystals. In this Book of Abstracts) is investigated in Lu2SiO5:Ce (LSO:Ce) scintillator in more details. The variation of the intensity of restored glow peaks in dependence of the storage time at nitrogen liquid temperature is studied in wide time interval. The restoration of glow peaks is found to occur at expense of deepest electron traps, which are supposed to be oxygen vacancies. The results obtained are analyzed in the frame of simple model including the interaction between two centers. This model is applied specifically to LSO:Ce , and several possible candidates are suggested for shallow traps in this material.
N25-162:

Thermoluminescent Properties of LSO:Ce Crystal

H. S. Kim1, S. H. Park1, J. H. Ha1, S. Y. Cho2, D. H. Kim2, S. H. Kim2, J. P. Park2, M. G. B. G. B. N. L. Yoo2 1 Nuclear Technology Convergence Division, Korea Atomic Energy Research Institute, Daejeon, Korea 2 Environmental Engineering, Yonsei University, Wonju, Korea Scintillation crystals convert the energy contained in x-rays and gamma rays to lights. Usually this scintillation light is collected, converted to electrons and amplified by a photomultiplier tube (PMT). The PMT functions well but has the drawbacks of being bulky and requiring a high voltage (HV) to operate it. This scintillation light can also be collected in a solid state photo-detector, such as a silicon PIN-type photodiode. The PIN photodiode is more efficient than a PMT, because it requires less bias to operate it and it is very compact. A Si PIN-type radiation detector was designed and fabricated for industrial applications, such as NDT, radiation monitoring system and medical devices in combination with a CsI(Tl) scintillator. Fabrication processes were briefly addressed. Leakage currents, photo sensitivities, and alpha responses were measured as the preliminary tests. A CsI(Tl)/PIN-type radiation detector was fabricated and tested. 7.9% FWHM for Cs-137 was achieved. The energy dependence was estimated by comparing the Co-57, Cs-137, and Co-60 energy peaks and their measured channels. The root-mean-square value for a linearity was 1. The performance characteristics of the fabricated CsI(Tl) PIN photodiode sensor were compared with a commercialized PIN diode radiation detector combined with a CsI(Tl) scintillator. And for application of a fabricated Si PIN-type radiation detector as a radon detector, alpha particles from radon and their progenies were measured in an appropriate chamber. The sensitivity was measured with a Po-218 peak and it was estimated at 0.3 cpm/pCi/L with a uranium concentrated soil source. In this study, characteristics of a fabricated Si PIN-type radiation detector for industrial applications and its experimental results were presented.
This work has been carried out under the nuclear R&D program of the Ministry of Education, Science and Technology (MEST) and under the Eco-technopia 21 Project of the Ministry of Environment (ME) of Korea. We are also supported by the BK21 program of the Korea Research Foundation (KRF).

Characteristics of a Si PIN-Type Radiation Detector for Industrial Applications

200

N25-163:

G. Ichikawa1, S. Kawasaki1, S. Komamiya1,2, Y. Kamiya2, T. Sanuki3, H. M. Shimizu4, M. Hino5, M. Kitaguchi5 1 Graduate School of Science, The University of Tokyo, Tokyo, Japan 2 ICEPP, Tokyo, Japan 3 Graduate School of Science, Tohoku University, Miyagi, Japan 4 KEK, Ibaraki, Japan 5 KURRI, Osaka, Japan Ultra-cold neutrons (UCNs) have lower kinetic energy than Fermi potentials of general materials (order of 100 neV). Therefore they can be bound by the gravitational field and the reflecting surface on the bottom. Corresponding to the quantized energies they are distributed in height with a modulation of 10 m. Recently, a series of experiments to observe the gravity induced quantum states was carried out at the Institut Laue-Langevin (ILL). The modulated position distribution was measured using a uranium-coated plastic nuclear detector (CR 39). We are planning an experiment to measure the position moduration using a CCD-based pixel detector with a magnification system for the neutron vertical distribution. CCDs have a high spatial resolution and a good background tolerance thanks to a real-time readout. These features are necessary to obtain a better resolution for the experiment. The CCD used in this experiment is a commercial back-thinned one S7170-0909 (Hamamatsu Photonics K.K.) and has 512x512 pixels with the pixel size of 24x24 m. Neutrons have to be converted to charged particles in order to be detected by the CCD sensor. To keep the high spatial resolution, a thin ( 200 nm) 10B converter is formed directly on a CCD. Because the converting cross-section is higher for slower neutrons, the detection efficiency of UCN is sufficiently high despite the thinness of the converter. For the precise measurement of positions with low-event rates, the detection efficiency and the spatial resolution of the detector are crucially important. The detection efficiencies for cold, very-cold and ultra-cold neutrons are measured to be 1.7%, 4.050.080.12% (stat.sys.) and 40.60.72.8%, respectively. The spatial resolutions were measured to be 2.90.1m for horizontal direction and 4.40.2m for vertical direction. The details of the pixel detector and its performance tests will be presented.
N25-164:

A CCD-based Pixel Detector with Micron Spatial Resolution for Ultra Cold Neutrons

A. Kok1, T.-E. Hansen1, T. Hansen1, G. U. Jensen1, N. Lietaer1, M. Mielnik1, P. Storas2 1 MinaLab, SINTEF, Oslo, Norway 2 REC Solar, Oslo, Norway

High Aspect Ratio Deep RIE for Novel 3D Radiation Sensors in High Energy Physics Applications

3D detectors with round vertical electrodes penetrating through the entire silicon substrate have drawn high interests amongst the high energy physics community in recent years, due to their unique advantages such as ultra-fast time response, edgeless capability and radiation hardness. In addition, the through-wafer electrode technology can provide the possibility to connect 3D detectors on a wafer level via 3D interconnects. These advantages, however, are solely dependent on a successful dry plasma etching process that results in vertical through holes with high aspect ratios. The through-wafer electrodes in 3D detectors are filled with highly doped polysilicon, which has a low mobility for both electrons and holes. As a result, the electrodes are rather insensitive to radiation detection, and must therefore be kept as small as possible, with a preferred diameter of less than 18 m. Meanwhile, to have a good signal-to-noise ratio, the wafer thickness must also be reasonably thick, preferably no less than 250 m. An efficient deep RIE process that results in high aspect ratio is therefore crucial in the making of 3D detectors. SINTEF MiNaLab owns a class 1000 clean room with several state-of-the-art dry etchers and has the ideal in-house facilities to develop 3D detector processing. The next 3D fabrication aims to fabricate detectors with 14 m holes through a 285 m thick substrate. Several test runs using a newly installed Alcatel AMS-200 I-Productivity etcher have been completed and preliminary results are promising with an average etch rate of 6.1m, and an aspect ratio of 20:1. In addition, high selectivity was achieved by using aluminium masks and low frequency (LF) bias. The experiments and the etching results are presented in this talk.
N25-165: Calculation of Electron Cloud Distribution in 3-D Position Sensitive CdZnTe Detectors Using MLEM Method

Y. Zhu, Z. He Dept. of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, USA Abstract A digitized pre-amp waveform readout system for 3-D position sensitive CdZnTe detectors can provide the capability to calculate the electron cloud energy deposition distribution. This paper presents the derivation of an algorithm using the MLEM method. A simulation was done to test the algorithm. The result shows that with 1keV electronic noise, the calculated energy deposition distribution of a 1243keV electron cloud is similar to the real one.

201

N25-166:

Thoeretical Energy Calibration of Multiple-Pixel Events in a Wide Band-Gap Semiconductor Detector with Pixilated Electrodes

Y. A. Boucher, Z. He Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, United States The observed energy resolution of pixilated semiconductor detectors degrades as the number of triggered pixels over which the interaction(s) occurs increases. For multiple pixel events, one of the biggest barriers to overcome during the energy reconstruction is the effects of weighting potential cross-talk. The current calibration method determines a set of correction coefficients to convert each measured signal to an estimated energy deposition as a function of lateral separation between the interaction pixels and their centroid interaction depth. Theoretically, an energy calibration procedure can be implemented that will correct for weighting potential cross-talk as a function of the individual interaction depths and lateral pixel separation, as well as variation of amplifier gain for different pixel electrodes. An analysis of this theoretical calibration has been performed to better understand characteristics of this energy reconstruction method. This paper describes the results using actual measured data on the linear relationship between multiple-pixel anode signals. During this study, a technique was discovered which can identify charge sharing and Compton scattering events underneath two neighboring pixels.
N25-167: A Vertically Integrated (3D) Rolling Shutter Mode MAPS with in-Pixel Digital Memory and Delayed Readout

Y. Degerli1, G. Bertolone2, W. Dulinski2, F. Guilloux1, F. Morel2, F. Orsini1, X. Wei2 1 IRFU/SEDI, CEA - Saclay, Gif-sur-Yvette Cedex, France 2 IPHC, IN2P3, Strasbourg, France

In this paper, a novel 3DIT rolling shutter mode monolithic binary pixel architecture, RSBPix (Rolling Shutter Binary Pixel), for charged particle tracking is presented. The process used is Chartered 130 nm CMOS process, and the 3D integration of tiers is done by Tezzaron. A 3D-2 tiers option MPW run with this process is available through FermiLab. The analog part of the pixel consists of a n-well/p-substrate diode, a first amplification stage close to the detector, a first double sampling circuitry with a series capacitor and a switch, and a discriminator. The double sampling process suppresses the offset mismatches of the amplifiers and the reference level of the detector. The discriminator is made of an offset-compensated multi-stage preamplifier followed by a latch. This approach allows to overcome process related mismatches of transistors and is compatible with lowvoltage operation. The pixel pitch is 20 m and the power dissipation of the pixel is ~120 W. The pixel is readout in 80 ns. The analog tier (Tier#1), includes an array of 40x240 pixels divided in 9 sub-arrays with different diode sizes and transistor types. For test purpose, analog signals of 8 columns - out of 40 - can be observed directly, since their outputs are available without discriminators. Analog output buffers are also included on this tier. The digital tier (Tier#2) includes a sequencer and output multiplexers for binary outputs. The digital control signals sent to pixels are synchronous with the 100 MHz nominal clock frequency and lead to a readout time of the full array in less than 20 s. At the end of the read-out chain, the digitized data are multiplexed on 8 outputs. Due to the limited design time, only a 1-bit memory was used for the pixel on this chip. Additional digital processing, like time stamping, should be compatible with such architecture. All I/O pads are on Tier#2, and the analog tier is supplied by this one. The chip size is ~1.2mm x 5.5mm.
N25-168:

F. Nachtrab1, T. Hofmann2, N. Uhlmann2 1 Cluster "Engineering of Advanced Materials", Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany 2 Development Center X-ray Technology EZRT, Fraunhofer Institute for Integrated Circuits IIS, Fuerth, Germany Only the most advanced X-ray pixel detectors offer single photon counting capabillities as well as the ability to obtain spectroscopic information about the incident X-rays (e.g. the Medipix2 or the Dectris Pilatus detectors). For high resolution imaging, the pixel sizes of such detectors are too large, but they are limited technically by integration density. We present a simple solution for a spectroscopic, single photon counting X-ray detector with a reduction of the pixel size by a factor of almost 10: the central idea of our approach is to use standard imaging sensors for industrial optical imaging as X-ray sensors. In principle each of the photodiodes can act as a direct converting X-ray sensor pixel. Image data has to be read out with high frame rate to obtain single interaction events, then all data processing usually implemented in the detector (like photon counting or spectroscopy) can be offloaded to a PC or specialized hardware. The advantages of the optical sensors are their very small pixel size (down to 2 micron side length, compared to 55 micron for the Medipix2 and 172 micron for the Dectris Pilatus), high frame rates and their relatively low price. Their main disadvantage is the low detection efficency because the fill factor may be low and because the sensor layer (the photodiode) is thin compared to classical semiconductor X-ray detectors. Results of measurements with two optical camera sensors (one CCD- and one CMOS-Chip) will be presented as well as a comparison of the two different sensors. As a summary of our results we can say that it is possible to use optical imaging sensors as photon counting X-ray sensor devices, but with a low detection efficiency. We found that image quality is considerably better using a photon counting

Simple Solutions for Spectroscopic, Photon Counting X-Ray Imaging Detectors

202

acquisition procedure compared to an integrating acquisition. The quality of the spectroscopic information depends very much on the cluster size (multiplicity) of an event but is surprisingly good.
N25-169:

Common-Grid Pixellated CdZnTe Detector System Modeling

J. C. Kim, Z. He Nuclear Engineering and Radiological Science, University of Michigan, Ann Arbor, MI, USA Common-grid pixellated CdZnTe (CZT) detectors provide performance superior to simple pixel CZT detectors. The steering grids can improve the energy resolution by steering electron clouds to the anode pixels. The signals from common-grid 3-D position-sensitive pixellated CZT detectors have been modeled to study the energy calibration and event classification. This required precise modeling of the weighting potential along the drift trajectory of electrons and electron cloud segmentation near the common grid. The detector has dimensions of 20 mm x 20 mm x 15 mm, with an 11 x 11 pixellated anode and common grid. The pixel pitch is 1.72 mm. Using 662 keV photoelectric interactions at the cathode side, the signals along a pitch line between pixels have been calculated to review the necessity of electron cloud segmentation. The CZT signals at the pixel borders will vary at the same gamma-ray interaction position according to the number of electron cloud segments. As a result, electron cloud segmentation is required to properly model gamma-ray interactions at the pixel border. For precise modeling, charge transport properties such as charge sharing, trapping, and diffusion have been considered as well as properties of VAS_UM/TAT4 ASIC such as electronic noise and shaping. The current CZT signal model is used to perform the realistic simulation in support of the Polaris project.
This work was supported by DOE NA-22 office (award number DE-FG52-06NA27499). N25-170:

Pulse Height Linearity of CdZnTe

P. B. Ugorowski, A. Kargar, D. S. McGregor Mechanical and Nuclear Engineering, Kansas State University, Manhattan, Kansas, USA CdZnTe crystals equipped with Frisch collars are studied at the Semiconductor Materials and Radiological Technologies Laboratory (S.M.A.R.T. Lab) at Kansas State University for use as room-temperature, high-resolution, semiconductor gammaray detectors. Obtaining plots of Relative Pulse Height vs. Input Photon Energy by testing with monoenergetic radioactive sources yields widely-spaced data points, but a Compton-scattering two-detector system can send photons of any desired energy into the crystal being tested, yielding a continuous curve. The linearity of the resulting curve is a limiting factor for the energy resolution of the semiconductor material.
N25-171:

F. Aezinia1, K. Iniewski2, M. J. Syrzycki1 1 School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada 2 Redlen Technologies, Inc., Sidney, BC, Canada

Modelling the Weighting Potential Cross-Talk in CZT Radiation Detectors

Decreasing the pixel size in pixellated CZT radiation detectors will increase the image resolution in nuclear medical imaging. On the other hand, using smaller pixels will lead to an increased cross-talk between channels. In this paper we propose the analytical model for the weighting potential cross-talk, which is next used to evaluate charge induction conditions in a group of neighbouring pixels in relation to detector thickness and pixel size. The modelling results will be used for more accurate estimation of the coordinates of the gamma-photon interaction.
N25-172:

K. Hayakawa1, K. Hitomi1, T. Shoji1, C. Onodera2 1 Department of Electronics and Intelligent Systems, Tohoku Institute of Technology, Senday, Japan 2 Towada Technical High school, Towada, Japan

Photoluminescence Analysis of TlBr Crystals for Radiation Detector Applications

Photoluminescence (PL) analysis is widely used for characterization of common semiconductor detector materials such as CdTe and HgI.2. Although Thallium Bromide (TlBr) emerged as a promising detector material recently, a few results of PL analysis of TlBr crystals were reported. In this study, PL analysis of TlBr crystals was performed in order to reveal correlation between the PL spectra of TlBr crystals and the detector performance. The TlBr powder with purity of 99.99% was purified by the vacuum distillation and zone melting method. After the purification, single zone pass was performed with the aim of improving the crystallinity. The TlBr crystal was cut into wafers with a diamond wire saw. Both surfaces of the wafers were polished mechanically. TlBr wafers from pure end and impure end of the TlBr ingot were analyzed with the PL method. The samples were placed in an electronic cryostat and kept at 4 K. The samples were excited by a HeCd laser (325 nm). A PL spectrum obtained from a TlBr crystal from the pure end of the ingot exhibited PL peaks at 3, 2.6 and 2.5 eV and a PL band at 1.9 eV. The PL peak 203

at 3 eV originates from direct transition emission. The PL peak at 2.6 eV is indirect transition emission relating to impurities. The origin and nature of the emission lines are still under investigation. PL spectra obtained from TlBr crystals from pure end and impure end are compared. Significant difference of the PL intensity ratio between 2.6 and 2.5 eV peaks was observed between the pure and impure crystals. The pure crystal exhibited the PL intensity ratio less than 1. A 22Na gamma-ray spectrum with a clear 511 keV full-energy peak was obtained from a TlBr detector fabricated from a TlBr crystal at pure end which exhibited the PL intensity ratio less than 1. On the other hand, poor detector performance (no full-energy peak) was obtained from TlBr detectors fabricated from TlBr crystals exhibiting the PL intensity ratio more than 1.
N25-173:

A. Schubert1,2, G. J. O'Keefe3, B. A. Sobott1,2, R. P. Rassool1,2 1 Experimental Particle Physics, The University of Melbourne, Melbourne, Australia 2 CRCBID Cooperative Research Centre for Biomedical Imaging, Bundoora, Australia 3 Centre for PET, Austin Hospital, Heidelberg, Australia Adapting a detector designed for protein crystallography to medical imaging applications requires a detailed understanding of the electrical behaviour of its sensor. At present the detector is inefficient at the higher photon energies commonly used in the medical field. Two options present themselves for increasing the detection efficiency for higher energy photons- both an increase in sensor depth, or a semiconductor of higher atomic number would deliver this outcome. Irrespective of the choice, charge collection efficiency (CCE) and pixel capacitance are of particular interest since these competing parameters play a key role in determining the sensitivity and the energy resolution of the detector. The CCE is dependent on the sharing of signal charge between neighbouring pixels, since this may lead to double counting of the hit or alternatively it may be lost to recombination processes. Sensor optimisation involves balancing the implant geometry with respect to the CCE and energy resolution. A twodimensional model of a hybrid-pixel detector sensor has been created in IseTCAD to investigate these effects. Simulations have been performed on several configurations of 300 m-1mm n-type Silicon sensors with varying p+ implants and pitches and the behaviour of the charge sharing has been quantified. The model has been validated via experimental data obtained from PILATUS II, a single photon counting hybrid-pixel detector. It is hoped that with a precise knowledge of the charge sharing distribution, it could be ultimately be exploited to deliver improved spatial resolution.
N25-174:

Theoretical Modelling of a Novel Hybrid-Pixel Detector

B. A. Sobott1, D. J. Peake1, V. Lee1, N. Kirby2, A. Schubert1, C. Broennimann3, R. P. Rassool1 1 Physics, The University of Melbourne, Victoria, Australia 2 The Australian Synchrotron, Victoria, Australia 3 Dectris Ltd, 5400, Baden, Switzerland

Evaluation of the Energy Resolution of a Prototype Sensor for the PILATUS Detector.

Position sensitive spectroscopy and protein crystallography are important applications of synchrotron radiation that frequently require the imaging of x-rays with a high degree of spatial resolution. Single photon counting detectors such as PILATUS have demonstrated an ability to meet these stringent requirements at room temperature. The challenge ahead is now to deliver improvements in the energy domain. For example, by determining the optimal sensor geometry, a substantial improvement in energy resolution is expected. To investigate this, prototype silicon sensors comprising varying p+ implant configurations have been bump-bonded to PILATUS chips and their performance investigated. These test have been undertaken on the SAXS/WAXS beamline of the Australian synchrotron. Each configuration has been characterised with monochromatic x-rays and the dependence of the energy resolution on the implant size has been determined. Results indicate that a reduction in the p+ implant can deliver a marked improvement in the overall energy resolution of the PILATUS system and therefore a reduction in the threshold for the detection of x-rays.
N25-175:

200 mm Silicon Wafer Processing for Large Area Strip Detectors

M. Christophersen, B. F. Phlips Code 7651, U.S. Naval Research Laboratory, Washington, DC, USA Large silicon detectors made of 200 mm float-zone Si wafers were developed. The single-sided silicon strip detectors have an effective active area of 156 cm^2 and 725 micrometer in thickness and were fully depleted. Basic performance was measured using a standard Am-241 source. The energy resolution obtained was 2.6 keV FWHM at 59 keV, demonstrating good charge collection.

204

N25-176:

J. Cui1, K. Andreini1, G. Nitin1, S. R. Hayashi1, W. Zhang1, M. Yamada1, H. Jiang2, G. Schweinert2, H. Chen3, G. Bindley3 1 General Electric Research, Niskayuna, NY, USA 2 General Electric Healthcare, Milwaukee, WI, USA 3 Redlen Technologies, Sidney, BC, Canada Electrical measurements of sidewall leakage and spectroscopic performance are used to quantify the degree of sub-surface damage created by cutting CZT. Subsequent step-wise polishing of sidewall surfaces tests the depth to which cutting damage extends into the crystal. CZT samples 6 to 11mm thick were cut at GE Research with outer-diameter (OD) and wire saw methods. Additional OD saw cutting was done at Disco Corporation. The cut surfaces are characterized by SEM inspection and optical profilometry. The apparent sidewall smoothness and edge chipping comparison among cutting techniques implies differences in sub-surface damage caused by the cutting tool. However, what is not directly measured by roughness data is the sub-surface damage associated by unseen fracture, induced strain and chemical contamination. For each machining technique there is a depth of sidewall surface that must be removed to create a spectroscopic-grade device. This lapping-depth effectively adds to the total kerf-loss and contributes to the cost of the CZT tile mainly through additional operator and facilities time for the lapping processes.
This work was funded by the US Dept. of Homeland Security, Domestic Nuclear Detection Office, through contract number HSHQDC-08-C00174 N25-177: Electron Mobility and Lifetime Mapping of CZT with Known Crystalline Defects by Pulsed Laser Excitation

CZT Sub-Surface Damage Assessment Using Electrical Leakage Measurements

W. Li1, J. E. Tkaczyk1, J. Cui1, K. Andreini1, Y. Z. Williams1, K. G. Harding1, H. Chen2, G. Bindley2, R. Matyi3 1 General Electric Research, Niskayuna, NY, USA 2 Redlen Technologies, Sidney, BC, Canada 3 College of Nanoscale Science and Engineering of the University at Albany, State University of New York, Albany, NY, USA

A pulsed laser method is used to map the electron mobility-lifetime product (mu-tau) in Cd1-xZnxTe (x = 0.1) pieces especially selected to include regions with crystallographic defects. Additionally, infrared imaging and X-ray diffraction are used in conjunction with the pulsed laser approach to analyze the defects. The defect regions studied include twins, misoriented secondary grains, and Te inclusions. These defect regions were observed to affect the local electron mobility-lifetime product. The test of mu-tau is an important measurement for determining the spectral quality of radiation detectors. The product of the mu-tau of the electrons with the electric field applied to the device, mu-tau x E, gives the mean drift length of the electrons. In spectral quality CZT detectors the mean drift length of the electrons usually needs to be greater than or equivalent to the thickness of the detector. The findings of the pulsed laser testing and the correlation to other analytical techniques are reported.
This work was funded by the US Dept. of Homeland Security, Domestic Nuclear Detection Office, through contract number HSHQDC-08-C00174 N25-178:

Cross-Talk Effects in CZT Pixilated Detectors

K. Iniewski, S. Awadalla, H. Chen, R. Crestani, G. Bindley Redlen Technologies, Sidney, BC, Canada Introduction: In this paper we have further developed CZT detector model presented earlier to examine pixel cross-talk effects due to surface resistive coupling. The modeling work is supported by experimental results from 5mm thick CZT detectors manufactured by Redlen. Experimental Results: Standard 5 mm thick CZT have been grown using THM method. Pixel pattern was an array of 64 pixels (2mm by 2mm) with pixel spacing of 0.46mm. In order to explore the cross-talks effects in standard material various experiments with surface passivation were conducted. These tests were used to establish correlation between inter-pixel resistivity and energy resolution (ER). Conclusions: The performed cross-talk analysis indicates that CZT pixel arrays are insensitive to pixel to pixel cross talk effects in properly fabricated samples. Only faulty surface processing might lead to observable cross-talk. However, CZT radiation detection systems can be very sensitive to digital switching signals.

205

N25-179:

M. Jeong1, M. D. Hammig1, S. Ramadoss2 1 Nuclear Engineering & Rad. Sci., University of Michigan, Ann Arbor, MI, USA 2 Galt Research LLC, Ann Arbor, MI, USA

Fabrication and Signal Acquisition of the Silicon-Based Delay-Line Radiation Detector

In order to image the radiation field, instruments measure the positions of interaction of the incident quanta, a task typically accomplished by partitioning the detector or its readout. We previously validated a structurally simple time-based measurement scheme, in which a delay-line electrode structure is used to isolate the position of charge collection. If the transmission-line electrode structure is properly designed, then the time difference between the pulses at the readout ends can then be used to measure the position at which the charge was collected. Furthermore, if the timing uncertainty is significantly less than the transit time along a single strip in that design, then the position can be isolated to one strip of the pattern or better. In this paper, we describe the fabrication techniques used to overcome the two greatest limitations of the previous detectors; namely, high leakage currents and high propagation losses. In the following, we describe, through both models and measurements, the means by which: a) one can achieve impedance matching- by conversion of the electrode into a 50 asymmetric strip line, and b) one can minimize the propagation losses- by the use of laminated and/or thicker electrodes and the use of low-loss dielectrics as a part of asymmetric strip line design. To accomplish low leakage designs, the fabrication of both delay-line PIN detectors as well as metal-semiconductor junction detectors, is described. Finally, we describe the trade-offs when employing current-sensing for semiconductor-based radiation detectors, and demonstrate the means by which the current profiles are translated into maps at which the charge is initially deposited.
We are grateful to the Domestic Nuclear Detection Office (DNDO) and the Defense Threat Reduction Agency (DTRA) for their support of our effort. We also thank for the Korea Science and Engineering Foundation Grant (No. 2005-215-D00399) for their graduate student support. N25-180:

S. Barkan1, V. D. Saveliev1, L. Feng1, M. Takahashi1, E. Damron1, C. R. Tull1, N. J. Zaluzec2 1 SIINT USA, Northridge, CA, USA 2 Electron Microscopy Center, Argonne National Lab, Argonne, IL, USA

Silicon Drift Detectors for Specialized X-Ray Applications

A 50 mm2 silicon drift detector (SDD) has been successfully applied to XRF (x-ray fluorescence), TXRF, synchrotron and microanalysis applications. Several recent detector design developments have resulted in improved performance: A. Improved Quantum Efficiency - A thicker device enables the detector to be more efficient at higher energy. The typical device thickness of 0.35 mm has an efficiency of 0.28 and 0.08 for 20 keV and 30 keV, respectively. A recent experiment with a 1 mm thick SDD showed an efficiency which matches the theoretical values of 0.6 and 0.4 for 20 keV and 30 keV, respectively. B. Improved Low Energy Performance - A significantly improved peak-to-background ratio, as well as the detection of C, B and Be has been achieved with the improved SDD with a new shallower entrance window. C. Innovative SDD for Characterization of Nano-scale Materials - The characterization of nanoscale materials at high spatial resolution has become increasingly important in state-of-the-art materials research. Instruments such as Argonne National Laboratorys (ANL) sub-angstrom electron-optical instrument and the X-ray Nanoprobe are capable of extraordinary spatial resolution to study these materials. A custom 50 mm2 SDD with a larger solid angle for use in nanotechnology characterization was designed and assembled into a scanning electron microscope (SEM) at the Electron Microscopy Center at ANL. The solid angle was improved by a factor of 25, resulting in significant improvement in the overall count rate capability.
N25-181:

Improving Pixel Detectors: Active Area Optimization and High Temperature Annealing

V. A. Fadeyev, M. Gerling, J. Wright, P. Maddock, C. Betancourt, F. Hurley, H. F. F-W. Sadrozinski SCIPP, UCSC, Santa Cruz, CA, USA The existing pixel tracking detectors of ATLAS and CMS experiments will have to be replaced after LHC luminosity upgrade (super-LHC or sLHC), due to increased radiation level and tracking performance requirements. We are studying device active area re-optimization by cutting away dead area and some of the guard rings of existing sensors. Performance of the cut sensors will be described. We also present a feasibility study to operate irradiated silicon sensors after extreme temperature excursion characteristic of a standard bump-bonding process. This is relevant for high dose radiation studies before a sLHC-quality readout ASIC is developed.

206

N25-182:

J. Hasi1, C. J. Kenney2, S. I. Parker3, C. Da Via1, S. Watts1, E. Westbrook4 1 Manchester University, Manchester, UK 2 SLAC National Accelerator Laboratory, Menlo Park, California, USA 3 University of Hawaii, Honolulu, Hawaii, USA 4 Molecular Biology Consortium, Chicago, Illinois, USA

Testing of Multi-Electrode Sensors via Active Edges and Temporary Conductors

Devices with many independent electrodes are challenging to test. This is especially true for silicon p-i-n diodes, which are organized into an array of pixels or strips. A pair of methods has been used to facilitate the testing of sensors fabricated using the planar-process for their central area and using a through-wafer process to construct doped edges. Details of the fabrication, several applications, and results will be presented.
N25-183:

D. Passeri1,2, D. Biagetti1,2, S. Meroli2, L. Servoli2 1 D.I.E.I., University of Perugia (Italy), Perugia, Italy 2 INFN Perugia (Italy), Perugia, Italy

Tilted CMOS Active Pixel Sensors for Particle Tracking Reconstruction

In this work, we propose an innovative approach to particle track reconstruction based on CMOS Active Pixel Sensor (APS) detectors. Four sensors of 256x256 pixels, featuring 10x10 m2 pixel size with fast and re-configurable read-out electronics, have been fabricated in a 0.18m CMOS bulk technology. The main idea is to use three pixel layers in the conventional way (e.g. orthogonally disposed with respect to the particle trajectory), whereas a fourth layer can be tilted with respect to the others up to 90 degrees. This would potentially allow for a better track reconstruction, thanks to the big number of reconstruction points. In order to check the suitability of the approach a dedicated tracking system has been assembled and tested at the INFN BTF, Frascati (Italy). The read-out scheme is based on external ADCs and FPGA, allowing for the parallel readout of the four detectors with a frequency up to 64MHz. Preliminary test results demonstrate the suitability of the approach, e.g. the good system sensitivity to tilted trajectories. Actually, when the "parallel" layer is precisely aligned with the beam, most of the track is included in the sensor plane, thus eventually resulting in a complete row response.
N25-184:

M. Boehnel1, P. Sievers1, J. Roth2, G. Buchholz2, O. Hupe2, U. Ankerhold2, T. Michel1, G. Anton1 1 ECAP, University of Erlangen-Nuremberg, Erlangen, Germany 2 Physikalisch Technische Bundesanstalt, Braunschweig, Germany

Time Resolved Measurement of a Pulsed X-Ray Source with the Timepix Detector

The hybrid pixelated semiconductor Timepix detector is the successor of the Medipix2 X-ray detector. It is provided with a further range of functions directly integrated in the pixel electronics which allows to perform the measurement of event times of interacting X-ray photons in the silicon sensor layer in respect to the end of each acquisition. Due to the number of pixels of the device one has 65536 independently working channels to determine the time structure of high flux X-ray sources which are operating in a pulsed mode. These X-ray sources are commonly used in medical applications and homeland security and create X-ray bursts in the range of 50 ns. We present the results of the time resolved measurement of a pulsed X-ray field and the corresponding measurement which was done in a collaboration by the Physikalisch Technische Bundesanstalt (PTB).
N25-185:

Induced Signals in X-Ray Detectors with Steering Grid Geometry

B. Kreisler, G. Anton, J. Durst, T. Michel ECAP, Unversity of Erlangen, Erlangen, Germany Oncoming pixelated X-ray detectors with semiconductor sensor layers made of CdTe or GaAs can be operated in single photon counting mode. Each charge cloud generated by an interacting photon induces a signal in the electronics which is amplified, shaped and compared to a threshold. This process needs to be very fast for high flux applications, which leads to the necessity of understanding the signal induction process. The adjoint simulation of the shape and timing of the signals can be done with the finite element software COMSOL using the convection and diffusion package. For a given sensor material, the signals depend on the geometry of the electrodes and this offers ways to optimise the size and shape of the electrodes. Even additional steering electrodes can be implemented which will be the main part of this contribution. Fully 3D simulations have been carried out to study the possible effects of steering electrodes on the signal shape. A comparison will be presented between different geometries and applied potentials.

207

N25-186:

D. Benzaid1, A. Seghour2 1 Institut des Sciences et de la Technologie, Centre Universitaire de Khemis Miliana, Ain Defla, Algeria 2 Commissariat l'Energie Atomique, Centre de Recherches Nuclaires d'Alger, Alger, Algeria The present work concern theoretical characterization of a surface barrier neutron spectrometer. The bulk of the subject is the theoretical determination of neutron detection system using one single charged particles detector. The charged particles result from a converter which allows inducing nuclear reaction with neutron. Thus, a detailed study of kinematics of nuclear reactions induced by neutrons has been conducted and a numerical application (in FORTRAN) has been developed. This study concern determination of the response function of the detector and allows us to compare detection efficiency according to the target nuclei in the converter, namely converter in boron or lithium. A study of conditions of discrimination between detected energies as a function of reaction cross sections, which we find in literatures, has been concerned
N25-187:

Theoretical Characterization of a Surface Barrier Neutron Spectrometer

A. Lymanets1, S. Chatterji2, J. M. Heuser2 1 FIAS, Frankfurt am Main, Hesse, Germany 2 Dept. CBM, GSI, Darmstadt, Hesse, Germany

The Silicon Tracker of the CBM Experiment at FAIR: Detector Developments and First in-Beam Characterizations.

The Silicon Tracking System (STS) is a key detector of the CBM experiment planned at the Facility for Antiproton and Ion Research (FAIR). It has to reconstruct the trajectories of hundreds of charged particles created in heavy-ion collisions at typical beam energies of 25 GeV/nucleon. Radiation hard sensors and fast read out electronics are needed to match the interaction rates of up to 10 MHz. A low-mass detector system is required to for a momentum measurement with about 1% resolution. First double-sided silicon microstrip detector prototypes with 50 m strip pitch developed together with CiS, Germany have been characterized in the laboratory and several in-beam tests. In a full read-out chain based on self-triggering front-end electronics they were operated at GSI in a 2.5 GeV proton beam. A demonstrator tracking station was integrated into the beam tracker of the SVD-2 experiment at IHEP, Protvino, Russia and tested in a 50 GeV proton beam. The CBM experiment will operate with a high-intensity DC beam at the FAIR synchrotron SIS-300. Fast front-end electronics and a high-throughput data acquisition system are required. The self-triggering n-XYTER read-out chip with 128 channels is being explored in the prototype system tests. We present results on the detector response and the behavior of front-end electronics in a full data acquisition chain. The high-radiation environment in the CBM experiment results in a particle fluence through the STS of up to 1015 1-MeV neq/cm2 in 6 years of operation. We present measurements of double-sided silicon microstrip detectors as well as PiN diodes irradiated up to 1014 1-MeV neq/cm2 and compare them to non-irradiated samples.
* Supported by EU-FP7 Hadron Physics2 N25-188:

A. T. Lintereur1, W. Qiu2, J. C. Nino2, J. E. Baciak1 1 Nuclear and Radiological Engineering, University of Florida, Gainesville, FL, USA 2 Material Science Engineering, Univerisity of Florida, Gainesville, FL, USA Wide band-gap semiconductors often possess the material properties necessary for high resolution room temperature gamma-ray spectroscopy. Several iodine based semiconductors, mercuric iodide and lead iodide, have already demonstrated that they have the ability to be used for radiation detection. However, there are material and growth issues associated with both lead iodide and mercuric iodide that have encouraged the exploration of alternative materials. Bismuth tri-iodide is a wide band-gap semiconductor that has similar material properties to mercuric iodide including high density, large band-gap and large atomic number. However, bismuth tri-iodide does not suffer from the material issues, such as a solid state phase transition and dissociation in air, that mercuric iodide does. Thus, bismuth tri-iodide is both easier to grow and easier to handle than mercuric iodide. Simulations have demonstrated that bismuth tri-iodide has a larger theoretical intrinsic photopeak efficiency than many of the other materials used for gamma-ray spectroscopy. A modified vertical Bridgman growth technique is being used to grow large, single bismuth tri-iodide crystals. Zone refining is being performed to purify the starting material and increase the resistivity of the crystals. Initial detectors are being fabricated using both gold and palladium electrodes and palladium wire. I-V curves have demonstrated that both gold and palladium electrodes provide ohmic contacts. Resistivities in excess of 10E9 -cm are achievable via the modified Bridgman growth technique. Cu and Fe impurities in starting bismuth tri-iodide powder are the limiting impurities in achieving increased resistance. The response of bismuth tri-iodide to radiation is being tested with both current and pulse measurements.

Development of Bismuth Tri-Iodide Radiation Detectors

208

This project is funded by DOD - Defense Threat Reduction Agency Contract HDTRA-1-07-1-0013 The first author is funded by the National Science Foundation N25-189:

J. Uher1, J. Tickner1, J. Jakubek2 1 CSIRO Minerals, Lucas Heights, NSW, Australia 2 IEAP-CTU, Prague, Czech Republic

Material Recognition in Micro Imaging Based on X-Ray Beam Hardening Correction Method

Material sensitive X-ray imaging with micron resolution is a useful tool for many applications. We describe a method for material sensitive imaging with the single-photon counting imaging detector Medipx2 that accounts for beam hardening using the signal-to-equivalent-thickness calibration method. Two images of an object are taken with two different detector energy discrimination thresholds. Both images are corrected using the signal-to-equivalent-thickness method calibrated using a single, common material. The ratio of the two resulting equivalent thicknesses is a parameter which is dependant on the material type and can be used to distinguish materials in the image. The advantage of this method is that it requires neither filtering of the Xray beam nor the setting of narrow energy response windows in the detector. Therefore, it allows for shorter exposure times compared to other methods. The method is particularly suitable for imaging samples dimensions of a few to a few hundred microns. A theoretical description of the method is given and experimental results are presented, demonstrating the ability of the technique to distinguish foils of adjacent atomic numbers over thicknesses ranging from 15 to 600 m.
N25-190:

Fabrication and Performance of a Self-Powered 4H-SiC Schottky-Type Neutron Detector.

J. H. Ha, S. M. Kang, S. H. Park, N. H. Lee, T.-Y. Song, H. S. Kim Nuclear Applied Tech. Depart., Korea Atomic Energy Research Institute, Daejon, Korea Radiation detectors based on semiconductors like Silicon Carbide (SiC), Aluminum Nitride (AlN), Boron Nitride (BN) and Gallium Arsenide (GaAs) with a large energy band gap are regarded as the most promising ionizing radiation detectors for high temperature and in harsh radiation fields. Harsh radiation environments such as a nuclear reactor core, a high energy physics experiment, or an outer space device can cause radiation damage to detectors. SiC is being studied as a useful material in harsh environments because of its radiation-resistance, high-temperature operation and high-critical breakdown voltage, and a high thermal conductivity. Present work focused on a development of a radiation resistive neutron semiconductor detector based on a wide band-gap SiC semiconductor, operating at a zero biased voltage(self-powered) by using the internal strong electric field. The self-powered neutron detector structure was a metal/n/n+ Schottky device. This device consisted of a 0.5 m inter-layer (1 x 1E18 /cm3 doped) and a 5m epi-layer (1.2 x 1E15 /cm3 doped) on a n-type substrate with a 0.08 ohm-cm resistivity and a 350 m thickness. To convert the neutron to a charged particle, a thin 6LiF convert film was evaporated onto an Au metal contact. The charge collection efficiency (CCE) at a zero biased voltage was over 80 % due to the strong internal electric field. Above 20 V of the biased voltage, the charge collection efficiency reached 100 %. Neutron detection responses were measured during a self-powered operation by using Am/Be neutron source with 2.0E7 n/s activity, and the neutron flux on the detector was 1.77E5 n/cm2/sec at 5.0MeV of an average neutron energy. In conclusion, the SiC detector with a metal/n/n+ Schottky structure device revealed a good performance for neutron detection even at a zero biased voltage operation.
This work has been carried out under the nuclear R&D program of the Ministry of Education, Science and Technology (MEST) of Korea. N25-191:

E. Guni1, J. Durst1, G. Anton1, T. Michel1, M. Fiederle2, A. Zwerger2, A. Fauler2 1 ECAP, 91058 Erlangen, Germany 2 FMF, 79104 Freiburg, Germany

Investigation of a CdTe Sensor Concerning Different Pixel- and Electrode Sizes

The quality and availability of room temperature CdTe sensor material for X-ray detection has improved highly in the last years. The FMF successfully bump-bonded an CdTe sensor (thickness:1.6mm; area:14x14mm2) with different pixel and electrode sizes to a Medipix2-MXR ASIC. With this photon counting detector we are able to investigate the influence of pixel and electrode size on the energy response spectra of this detector. The accurate knowledge of the energy response spectra is crucial when it comes to energy resolving X-ray imaging. Therefore we exposed the detector to X-rays of 241Am and 57Co. In the following analysis of the data we looked at the number of events in the photo peak (241Am), the energy resolution (241Am), the photo peak position (241Am) and the energy calibration.

209

N25-192:

E. Martin1, E. Cortina1, L. Soung Yee1, D. Flandre2, C. Renaux2 1 CP3-UCL, Louvain-La-Neuve, Belgium 2 DICE-UCL, Louvain-La-Neuve, Belgium

TRAPPISTe Pixel Sensor with 2um SOI Technology

SOI is a technology for HEP detectors, where the connection of the active circuitry with the bulk Is done by vias, allowing full charge collection with CMOS electronics on the top active layer. TRAPPISTe uses an SOI approach that will be processed at the WINFAB facility at Louvain with a 2um technology, will be available for testing before August/September 2009. The handle layer consists of a p-type substrate with low resistivity of 15-25 Ohmcm. The substrate contains a 400 nm thick buried oxide layer and a 100 nm thick silicon overlayer. An-type (4 10^17 atoms/cm3) implant with a 60x60um area is created, with a p+ guard ring of a 10um width. The circuitry is based on the 3-transistor architecture commonly used in MAPS, modified for a pipelined readout. The TRAPPISTe chip consists of an 8x8 array of pixels, with an area per pixel of 300x300um. Signal formation is computed after TCAD results in a two step process. First, a 3D interpolation is done in order to generate a regular mesh with 1um step. Second, a C++ program calculates the generated signal. An initial rough estimate of the current signal generated in the SOI bulk was done for different depletion voltages. The results obtained for the different depleted values are xV (5;10;15;25V)=49, 68, 83, 107um.The results obtained show collection times less than 9ns for a V=25V The testing will include the use of a probe station to measure the basic transfer curves of the systems. The testing set-up will be based on a PCB and a link to transmit data to a computer. A laser set-up, as well as the cyclotron facilities, will be used to characterize the sensitivity, speed and radiation hardness of the system. The grid of pixels is divided into different sections which contain different types of transistors for the same readout circuit. The transistors differ in their threshold voltage Vt and they are: standard Vt (0.46V/0.46V), high Vt (0.77V/-0.95V), low Vt (0.24V/-0.08V) and graded Vt.
N25-193:

CZT Quasi-Hemispherical Detectors with Improved Spectrometric Characteristics

V. Ivanov, L. Aleksejeva, P. Dorogov, A. Loutchanski ZRF RITEC SIA, Riga, Latvia At present time various CZT detectors of different designs and sizes are widely and successfully used for different applications due to its favorable detection properties. Among them there are hemispherical or quasi-hemispherical detectors. These detectors have rather simple design and do not require special electronics for application. Development of quasi-hemispherical detectors fabrication methods allows noticeable improves detectors spectrometric performance due to a charge collection optimization. This will allow increasing the yield of high quality detectors. For example energy resolution at 662 keV line and peak-toCompton ratio of the quasi-hemispherical detector with volume of 150 mm3 were improved from 27 keV and 2.5 to 13 keV and 4.9 correspondingly. Total absorption peaks efficiencies in all energy range remained without changes. At that, improvements were obtained without increasing of detectors high operation voltages and without application of any pulse shape correction/selection schemes.
N25-194:

T. Kishishita1, G. Sato1, H. Ikeda1, T. Kiyuna2, Y. Mito2, T. Takahashi1 1 Department of High Energy Astrophysics, Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, Sagamihara, Kanagawa, Japan 2 Acrorad Co., Ltd., Uruma, Okinawa, Japan We report on the recent development of a low-noise analog front-end ASIC for hard X-ray and gamma-ray detectors. The ASIC aims for the readout of strip detectors utilizing silicon and cadmium telluride (CdTe) as detector materials. In order to demonstrate our low-noise signal processing architectures, the 32-channel ASIC has been newly developed on the basis of the Open-IP LSI project led by JAXA. The ASIC is implemented with a TSMC 0.35-m CMOS technology with a chip size of 3 mm x 7.5 mm. Each readout channel includes a charge-sensitive amplifier, band-pass filters (fast and slow shapers) and sample-andhold circuit. It also includes a leakage current compensation circuit to accommodate to the variety of solid-state detectors. Analog outputs for 32 readout channels are held simultaneously by a properly-timed external signal and serially read out in a multiplexing scheme. In the initial evaluation tests, the equivalent noise level of a typical channel has been reached 88 e- + 8.1 e/pF (rms) for a power consumption of 3 mW per channel. We are now designing a front-end carrier board for mounting ASIC, utilizing an LTCC (Low Temperature Co-fired Ceramic) technology, to minimize the parasitic capacitance from an assembly issue. The spectral performance with combining the ASIC on the LTCC carrier board with CdTe detectors is also reported.

Performance Evaluation of Low-Noise Analog Front-End for Semiconductor Detectors

210

N25-195:

Energy Resolved X-Ray Imaging as a Tool for Characterization of Paper Coating Quality

B. Norlin, C. Fr&oumljdh, M. O'Nils, A. Fr&oumljdh, E. Fr&oumljdh, G. Thungstr&oumlm Dept. of Information Technology and Media, Mid Sweden University, Sundsvall, Sweden Energy resolved X-ray imaging can be used as a tool to analyze the variation in the chemical content of an object. In this work we have used energy resolved X-ray imaging to measure the variation in the chemical content of paper and paper coating. This is an important quality parameter for the paper industry. In order to separate the variation in coating thickness from the variation in paper thickness, energy resolution is used to separate the response of the coating from the response of the paper. The MEDIPIX2 single photon processing X-ray imaging system has been used in the measurements. The measurement results are compared to simulations with MCNP. The influence of charge sharing is discussed and the effects have been studied by comparing results from detectors with 220x220 m2 pixels and detectors with 55x55 m2 pixels. There is a trade-off between good spatial resolution obtained with detectors with small pixels and good energy resolution obtained with detectors with large pixels. The requirements on image quality, to achieve the resolution of coating distribution relevant for the application, are discussed.
N25-196:

A. Pullia1,2, D. Weisshaar3, F. Zocca2, D. Bazzacco4 1 Dept. of Physics, University of Milano, Milan, Italy 2 INFN, Milan, Italy 3 IKP, University of Koeln, Koeln, Germany 4 INFN, Padova, Italy

Cross-Talk Limits of Highly Segmented Semiconductor Detectors

The performance of highly segmented semiconductor detectors for high-resolution X-gamma spectrometry can be significantly spoiled by cross-talk effects. Such effects show up as false signals, fold-dependent spectral line shifts, pulse-shape distortion. These phenomena are particularly severe in segmented High-Purity Germanium (HPGe) gamma-ray detectors, where double-hit (fold=2) or triple-hit (fold=3) events easily occur. Two complementary electrical models are developed, which describe quantitatively the interchannel cross-talk of a monolithic detector whose cathodes (segments), and anode (core) are read out simultaneously by charge-sensitive preamplifiers. The first is here designated as Cross Capacitance (CC) model, the second as Charge Split (CS) model. The CC model builds around the capacitive parasitics Cij between the preamplifier outputs and the nighbour channel inputs. The CS model builds around the finite-value of the decoupling capacitance Cc used to read out the highvoltage detector electrode. The key parameters of the models are highlighted and ideas are shown to minimize their impact. Using a coaxial HPGe segmented detector it is found that the CC model tends to dominate over the CS model for decoupling capacitances larger than 1nF. Minimization of the parasitics Cij requires a careful design of the layout of the Printed Circuit Board (PCB) where the input transistors are mounted. With a custom design for such PCB we were able to limit Cij to about 4fF. Further cross-talk effects related to power-supply bouncing and to imperfect virtual earth at the preamplifier input will also be discussed. Eventually, an algorithm for cross-talk correction will be presented and elaborated.
N25-197: Development of Radiation Hard Silicon Sensors for the CBM Silicon Tracking System Using Simulation Approach

S. Chatterji, A. Lymanets, J. M. Heuser CBM, GSI, Darmstadt, Germany The very intense radiation environment of the planned "Compressed Baryonic Matter" (CBM) experiment at the international research centre "Facility for Antiproton and Ion Research" (FAIR) makes radiation hardness a central issue for the Silicon Tracking System (STS). STS will consist of eight stations of Double Sided Silicon Strip Detectors (DSSD) at a distance between 25 cm to 100 cm downstream of the target. It is expected that the total integrated fluence will reach 1x1015 cm-2 1 MeV neutrons equivalent which is more than expected at LHC/CERN. The major macroscopic effect of radiation damage in determining the viability of long-term operation of silicon sensors is the change in the effective charge carrier concentration (Neff), leading to type inversion. For the safe operation over full CBM life time, detectors are required to sustain very high voltage operation, well exceeding the bias voltage needed to fully deplete the heavily irradiated sensors. Thus, the main effort in the development of silicon sensors is concentrated on a design that avoids p-n junction breakdown at operational biases. Detailed calculations using the Hamburg Model have allowed the parameterization of these effects to simulate the operation scenario of the silicon detectors. Simulations are carried out to study the effect of change in Neff, as well as crucial geometrical parameters, on the breakdown performance using the PISCES code. Process simulation has also been performed using SUPREM-4 for studying the annealing behaviour of Boron implanted in silicon. The impact of crystal orientation on the total capacitance of the silicon detectors and hence on the signal/noise ratio has also been studied. We plan to do three dimensional simulation using TCAD tools from SYNOPSIS namely Sentaurus Process and Sentaurus Device packages to accurately simulate DSSD. TOPICS+KEYWORDS: Si detector, Type-inversion, Breakdown voltage, 3-D Simulation, DESIS.
Supported by EU-FP7 Hadron Physics2

211

N25-198:

L. Ratti1,2, L. Gaioni1,2, M. Manghisoni3,2, V. Re3,2, G. Traversi3,2 1 Electronics Department, University of Pavia, Pavia, Italy 2 Italian Institute for Nuclear Physics, Pavia, Italy 3 Industrial Engineering Department, University of Bergamo, Dalmine (BG), Italy

3D DNW MAPS for High Resolution, Highly Efficient, Sparse Readout CMOS Detectors

This work is concerned with the design and characterization of deep N-well (DNW) monolithic active pixel sensors (MAPS) fabricated in a vertical integration (3D) CMOS technology. These devices come as an evolution of DNW MAPS fabricated in a planar CMOS process, whose resolution, detection efficiency and charge collection efficiency have been improved by taking advantage of the increased functional density provided by 3D integration technologies. The aim of the paper is to discuss the design guidelines for the front-end electronics and the readout architecture and to present the relevant simulation results together with the first experimental outcomes from the test of 3D DNW MAPS. The work will also discuss the results from device simulations comparing the old 2D solution to the new 3D one.
N25-199:

A. Macchiolo1, L. Andricek2, M. Beimforde1, H.-G. Moser2, R. Nisius1, R. Richter2 1 Max-Planck-Institut fuer Physik, Munich, Germany 2 Max-Planck-Institut Halbleiterlabor, Munich, Germany

Development of Thin Sensors and a Novel Interconnection Technology for the Upgrade of the ATLAS Pixel System at SLHC

We present an R&D activity aimed to develop a new detector concept in the framework of the upgrade of the ATLAS pixel detector for the Super-LHC. The new devices combine thin pixels sensors with a vertical integration technology developed by the Fraunhofer IZM. The pixel sensors, with an active thickness of 75 and 150 microns, are produced by the Semiconductor Laboratory of the Max-Planck-Institut using both the n+-in-n and n-in-p technology. Thin devices combine good radiationhardness properties up to the fluences expected at Super-LHC in the inner layers of the ATLAS tracker with a modest contribution to the material budget of the detector. The results of the pre-irradiation characterization of the pixel sensors will be presented, together with the first measurements of Charge Collection Efficiency performed on micro-strip sensors produced on the same wafers and irradiated with 26 MeV protons up to fluences of 1016 cm-2 1 MeV n. eq. We envisage a demonstrator module where the thin sensors are connected to the ATLAS pixel ASIC by the novel Solid-Liquid Interdiffusion in alternative to the bump-bonding technique. In addition we plan to route the signals through Inter-Chip Vias from the ASIC wafer backside where post-processing bonding pads will be created. We will present the layout and the results obtained with a production of teststructures designed to investigate the SLID interconnection efficiency as a function of different parameters, i.e. the pixel size and pitch, and the planarity of the underlying layers.
N25-200:

Novel Fabrication Process for Edgeless Detectors on 6" SOI-Wafers

J. J. Kalliopuska, S. Eranen, T. Virolainen Micro and Nanoelectronics, VTT, Espoo, Finland The edgeless detector design offers some attractive properties for the high energy physics experiments and medical imaging studies. Construction of chips to long ladders or matrix large imaging detectors is practical with a negligible insensitive area between detectors. For charged particle detection, the design permits a large active/inactive area ratio. The active edge avoids inhomogeneous electric fields and surface leakage currents. VTT has developed a straightforward and fast process to fabricate edgeless (edge active) microstrip and pixel detectors on 6" SOI-wafers. The process excludes all slow process steps, such as polysilicon growth, planarization and additional ICP-etching. We have successfully fabricated p-on-n prototypes of edgeless detectors having deadlayer at the n-type edge of the order of few microns. The prototypes include 5x5 cm2 and 1x1 cm2 edgeless microstrip detectors with DC-, FOXFET- and PT-couplings. In addition, 1.4x1.4 cm2 Medipix2 edgeless pixel detectors were fabricated. First electrical characterization (IV and CV) gave promising results. Leakage current of the microstrips was measured to be uniform from edge-to-edge. The level of leakage current was 40-100 nA/cm2. The breakdown voltages were measured to be within 150 V to 200 V depending on the active edge distance from the closest strip. Using the novel process, we have designed n-on-n edgeless detectors having a p-type active edge. The polarity change should improve the edge response. The processing of these detectors has been started and should be finished by July. The presentation summarizes the novel fabrication process and compares it to the previous one. The electrical characterization results of both polarity type detectors are shown. The edge response of the detectors will be measured with an IR-laser. First radiation response results of the detectors with varied radiation sources are represented. 212

N25-201:

Silicon Detectors for the sLHC - Recent RD50 Results

U. Parzefall University of Freiburg, Freiburg, Germany On behalf of the RD50 Collaboration It is foreseen to significantly increase the luminosity of the Large Hadron Collider(LHC) at CERN around 2018 by upgrading the LHC towards the sLHC (Super-LHC). Due to the radiation damage to the silicon detectors used, the physics experiment will require new tracking detectors for sLHC operation. All-silicon central trackers are being studied in ATLAS, CMS and LHCb, with extremely radiation hard silicon sensors on the innermost layers. The radiation hardness of these new sensors must surpass the one of LHC detectors by roughly an order of magnitude. Within the CERN RD50 collaboration, a massive R&D programme is underway to develop silicon sensors with sufficient radiation tolerance. Among the R&D topics are gaining a deeper understanding of the microscopic defects created by radiation, testing new detector materials such as Czochralski Silicon instead of Float-Zone material, and studying planar p-type sensor that will not experience radiation-induced type-inversion. Another field of activity is the development of new sensor technologies like 3D silicon detectors especially designed for the extreme radiation levels of the innermost layers of sLHC experiments. We will report on the recent results obtained by RD50 from tests of several detector technologies and silicon materials at radiation levels corresponding to sLHC fluences. Based on these results, we will give recommendations for the silicon detectors to be used at the different radii of sLHC tracking systems.
N25-202:

Development of the Silicon Photo-Strip Sensor

D. H. Kah, J. B. Bae, H. J. Hyun, H. J. Kim, H. O. Kim, H. Park Physics, Kyungpook National University, Daegu, south KOREA We have designed and fabricated silicon strip sensors with polysilicon resistors and coupling capacitors. In this meeting, we will present for the developed the silicon photo-strip sensors which provide the position information using the capacitive readout coupling strip sensor. The radiation is converted into photo signal in a crystal scintillator and silicon photo-strip sensor read the photo signal. This developed device consists of two photo-strip sensors sandwiching opposite faces of a crystal scintillator and being oriented orthogonal to each other. This detector system is able to provide the 2 dimensional position and to measure the energy. Therefore they can be used in nuclear and medical application. The front side of the developed phto-strip sensor is a light incidence window which has very thin n+ doping layer with anti-reflective coating to prevent light losing. On the other hand, the rear side has p+ strip array for signal readout. These sensors are fabricated on a 5-in., 380m-thickness, n-type, high resistivity, oriented silicon wafer and the prototype sensors. The leakage currents of the fabricated sensors are measured as a function of the reverse bias voltages. The fabrication process of the photo-strip sensor is verified by using an alpha radioactive source and photo properties of the device are measured by a light source.
N25-203:

M. Gelin1, J. Baudot2, G. Bertolone2, A. Brogna2, G. Claus2, C. Colledani2, R. De Masi2, Y. Deerli1, A. Dorokhov2, W. Dulinski2, M. Goffe2, F. Guilloux1, A. Himmi2, C. Hu-Guo2, K. Jaaskelainen2, F. Morel2, F. Orsini1, M. Specht2, I. Valin2, G. Voutsinas2, M. Winter2 1 DSM/Irfu/SEDI, CEA Saclay, Saclay, France 2 DRS, IPHC, Strasbourg, France In the framework of the EUDET-JRA1 project (European detector R&D towards the International Linear Collider), which consists to design, realize and qualify a high resolution beam telescope based on Monolithic Active Pixel Sensors (MAPS), an intermediate chip sensor, MIMOSA22, has been delivered with success in 2008. Although this intermediate chip has fulfilled all the initial requirements of the project, however it was admitted that radiation tolerance behavior could be improved especially if the high precision telescope is used somewhere else than in the DESY testbeam infrastructure. For this purpose, a new version of a chip, MIMOSA22-bis, has been designed, using again the same process AMS 0.35 m opto as MIMOSA22. MIMOSA22-bis is based on an array of pixels covering 26.5 mm, commanded through a row-wise sequencer and surrounded by embedded electronics for slow control and data processing. The matrix is composed of 576x136 pixels divided in eighteen sub-arrays. This paper will be focused on tests in laboratory using a 55Fe source and in CERN-SPS using a 120 GeV pions beam, in order to characterize the chip with MIPs. These tests were performed before and after ionizing irradiation up to 300 kRad and behavior of the different sub arrays composed of different pixel architectures are studied.

Radiation Tolerance Study of a Digital Chip Sensor for the EUDET-JRA1 Project

213

N25-204:

A. B. Garson III1, Q. Li1, M. Beilicke1, M. Groza2, J. Martin1, K. Lee1, I. Jung3, A. Burger2, H. Krawczynksii1 1 Physics, Washington University in St. Louis, St. Louis, MO, USA 2 Physics, Fisk University, Nashville, TN, USA 3 Frederich-Alexander Universitat Erlangen-Nurnberg, Nurnburg, Germany

X-Ray and Gamma-Ray CZT Detector Development: Anode Design and Insulating Materials

Cadmium Zinc Telluride (CZT) continues to progress in quality and cost as a detector material for hard X-ray and gamma-ray photons. With our detector fabrication capabilities we have produced 2 x 2 x 1cm3 CZT detectors with alternative anode and cathode designs in an effort to maintain the excellent efficiency, energy resolution, and spatial localization acheived with pixellated CZT detectors while reducing the number of readout channels. For different contact patterns, we study the effects of the 3-D location of interaction on the detected signals using a highly collimated beam of x-rays. We also investigate the effectiveness of different insulating materials in reducing leakage currents, a major contributor to detector noise and performance.
N25-205:

Development of Active Pixel Detector with CDS in FD-SOI Technology

Y. Ikemoto Institute of Particle and Nuclear Studies, High Energy Accelerator Research Organization, Tsukuba, Japan On behalf of the SOIPIX Collaboration We have been developing a monolithic pixel detector with 0.2m Fully-depleted Silicon-on-Insulator (FD-SOI) technology. In this time, we have developed integrated active pixel detector, which is based on 3Transistors, with correlated double sampling (CDS) to enable reduction of kTC noise and correction of threshold voltage variation in source follower transistor. We report the circuit design and evaluation of its performance.
N25-206:

Charge Collection Properties of Heavily Irradiated Silicon Micro-Strip Detectors Studied by Edge Transient Current Technique (E-TCT)

G. Kramberger, V. Cindro, I. Mandi, M. Miku, M. Zavrtanik Experimental Particle Physics Department (F9), Jozef Stefan Institute, Ljubljana, Slovenia Transient Current Technique (TCT) was used to evaluate irradiated p-type silicon micro-strip detectors. A pulsed IR laser focused to a spot of to 6 m illuminated the detector edge so that the beam was parallel with the surface and perpendicular to the strips. In that way electron hole pairs were created at known depth in the detector. Scans over the entire detector thickness with 0.5 m resolution were performed. For each laser beam position the induced current shape is measured for one of the strips. The charge collection efficiency was studied as a function of laser position (depth of carrier generation), voltage, integration time and fluence up to 1016 neutrons cm-2. The shapes of the current pulses were exploited for determination of electric field profiles without relying on precise values of effective trapping times.
N25-207:

Contactless Resistance Evaluation of Pre-Processed High-Resistivity CdZnTe Wafers Using Two Layer Model

A. Cherlin, D. Braginsky, G. Litmanovich, U. El-Hanany Orbotech Medical Solutions Ltd, Rehovot, Israel We present an improved method for pre-processed CdZnTe wafers resistance evaluation. This approach is developed to allow fast and convenient way to characterize CdZnTe wafers before application of metallic contacts and probably before finishing all steps of the polishing procedure. The goal is to observe and characterize the influence and the quality of thermal annealing and polishing procedure. The original model is based on derivation of the resistance/resistivity from a characteristic relaxation time using the time dependent charge transient. The model assumes a uniform distribution of the bulk resistivity. During the measurement, the CdZnTe wafer is placed between the plain electrodes forming a structure of parallel plate capacitor. Our method is based on the model assuming 2 layers with significantly different resistances. Two relaxation times, related to two different layers are different enough to allow distinguishing between the exponentials which are functions of these relaxation times. We found experimentally that the measurements are possible only if the difference between the resistances of two layers is at least one order of magnitude. The model was validated first with RC loops of precise resistors (~100 M ~50 G) and capacitors (~0.2 pF - 1 pF), where one RC loop represents one layer of a uniform resistance. Our precision with the original one layer model was about 5%. With 2 layer model we succeeded to reach a precision of about 20%. Next, a series of measurements with CdZnTe wafers was conducted. The results demonstrate that the distinguishing between the layers is definitely possible. We present the results of the both measurements, the RC loops and CdznTe wafers, together with the analysis allowing to derive both resistances. A comparison to the I-V measurements after pixel evaporation is also given.

214

N25-208:

Comparison of Simulated and Measured Energy Response Spectra for a Medipix2 Detector Using CdTe as Sensor Material

J. R. Durst, E. Guni, T. Michel, G. Anton Novel Detectors / Medical Physics, ECAP, Erlangen Centre for Astroparticle Physics, Erlangen, Germany The new generation of pixelated photon counting X-ray detectors like the Medipix2 and the Medipix3 does not measure energy deposition directly. Instead of that the measured observable is the number of photons, which deposited an energy larger than a given threshold. The implementations in the simulation are the physics processes in the sensor layer, diffusion and repulsion of charge carriers during drift and lifetime of the charge carriers, all taking into account intrinsic doping of the sensor material. The drift of charge carriers induces mirror charges at the electrodes leading to a signal even if not all of the electrons and holes reach the electrode. This results in a partial charge collection and therefore has an impact on the energy resolution. Noise is modelled with Fano noise during energy deposition, and several noise contributions of the pixel electronics. Further, the charge summing mode of the Medipix3 is implemented. Due to the hybrid design of the Medipix detectors several combinations of an ASIC and a sensor are possible. With the implemented physics it is possible to simulate high-Z sensor materials like CdTe or GaAs in addition to silicon. This contribution compares simulated spectra from this detector class with measured spectra from a Medipix2-MXR ASIC bumpbonded to a CdTe sensor with different pixel and electrode sizes. Due to the shift of the photo peak caused by the limited lifetime of the charge carriers an improved energy calibration is needed to compare the energy response spectra.
N25-209:

Investigations of Internal Electric Field in Cadmium Zinc Telluride (CZT) Detectors

G. Yang, A. Bolotnikov, G. Camarda, Y. Cui, A. Hossain, K. H. Kim, R. James Brookhaven National Laboratory, Upton, NY, United States Cadmium zinc telluride (CZT) detectors have great potential for wide applications in medical imaging, environmental protection, national security, and treaty verification because they operate at room temperature and provide high detection-efficiency and good energy-resolution. The internal electric field distribution in CZT detectors substantially affects the transport behavior of photon-generated charges and, therefore, determines the spectral response of detectors. We employed a Pockels Effect measurement system and synchrotron X-Ray mapping technique to investigate the internal electric field distribution in different CZT detectors. The latest experimental results will be presented, and the factors affecting the internal electric field in CZT detectors will be discussed.
N25-210:

Simulation of Radiation Damage Effects on Planar Pixel Guard Ring Structure for ATLAS Inner Detector Upgrade

M. Benoit, Lounis,Dinu LAL, Orsay, France On behalf of the ATLAS planar pixel upgrade group Abstract : We use AC, DC and transient TCAD simulation to study the effects of radiation damage on active area and breakdown protection of different multi-guard ring structure planar pixel design for ATLAS inner detector upgrade. Using a model of defect introduction into Silicon band gap that agree well with simulation up to fluences of 1e15 neq/cm2 , we simulated how high radiation damage phenomena like space-charge sign inversion and double junction formation affects the efficiency of different models of multi-guard ring structure in n-type and p-type bulk. Charge collection efficiency (Transient simulation), chargesharing, pulse shape, depletion potential, potential distribution on guard ring, breakdown voltage and leakage current were extracted from simulation to compare the behaviour of the different models. We finally suggest a set of test structure inserted in a production that can be used to calibrate the defect introduction model for guard ring structures by comparing experimental data to these extracted parameters.
N25-211: Spectrometric Performances of CdTe and CdZnTe Spectrometric Semiconductor Detector Arrays at High X-Ray Flux

A. Brambilla1, C. Boudou2, P. Ouvrier Buffet1, F. Mougel1, J. Rinkel1, G. Gonon1, L. Verger1 1 LETI-MINATEC, CEA, Grenoble, France 2 XRIS, Thales Electron Devices, Moirans, France

Recently, there is a growing interest for high flux X-ray imaging detectors with energy discrimination or spectroscopy capabilities. Detectors made of compound semiconductors such as CdTe and CdZnTe have demonstrated outstanding performances for X and gamma ray spectrometry with room temperature operation . Progress in device technology allows fabricating pixellated detectors with high spatial resolution. LETI has been developping a prototype 16 channels electronic read 215

out circuit for fast X-ray spectrometry at very high count rates. The read out-circuit was coupled commercial CdTe and CdZnTe linear array detectors. The device measures the X-ray energy and builds the energy spectra on 256 bins with an electronic dead time of only 50 ns and a good energy resolution (13 % at 80 keV) for X-ray flux up to 1E7 ph.mm-2.s-1.
N25-212:

Production Chain of Isotopically Modified Ge-Diodes for the 2b0n-Search with Gerda

P. Grabmayr Kepler Centre for Astro and Particle Physics, Eberhard-Karls University Tbingen, Tbingen, Germany On behalf of the GERDA Collaboration The Gerda experiment investigates the nature of the neutrino through the search for the neutrinoless double beta decay. Due to the rareness of this decay (T1/2=1025years) large masses must investigated in radiopure, low background environment. The Gerda Phase I experiment searches with a total of 17.9 kg for the 2b0n-decay in 76Ge employing the calorimetric approach. In Gerda Phase II two new detector types will be deployed. The production chain from procurement of the isotopically modified material, reduction to metal, chemical purification, crystal pulling and diode fabrication will be tested with material depleted in 76Ge. The production steps and the specifications will be described.
This work is supported in part by the German DFG, BMBF and MPG, the Italian INFN, the Swiss SNF, and the Russian RAS. N25-213:

A. Di Vacri1, E. Bellotti2, C. Cattadori3, A. D'Andragora1, M. Laubenstein1, L. Pandola1 1 Gran Sasso National Laboratory, L'Aquila, Italy 2 Dept. of Physics, Universit di Milano Bicocca, Milano, Italy 3 INFN Milano Bicocca, Milano, Italy

Characterization of Broad Energy Germanium Detector (BEGe) as a Candidate for the GERDA Experiment

The final choice among different types of enriched Ge detectors for the second phase of GERDA experiment, devoted to the search for neutrinoless double beta decay of 76Ge presently in the commissioning phase at the Gran Sasso National Laboratory (LNGS), will be based on the relative comparison of different types of detectors. The relevant parameters are energy resolution, stability, pulse shape discrimination power and losses in obtaining diodes starting from enriched (86%) 76GeO2 powder. It must be stressed that the dimensions of the detector presently tested minimize the losses of the expensive enriched germanium in the detector construction. Broad Energy Germanium (BEGe) detectors are potential candidates since they have an electrode structure, with a small p+ contact area, that enhances the difference in the charge carrier drift times according to the interaction point offering a better rejection power of multiple-site interactions based on the pulse shape analysis. In this contribution we present the results of the measurements for the characterization of a Broad Energy Germanium (BEGe) provided by Canberra Semiconductor, Olen and presently available at the Gran Sasso National Laboratory. The pulse-shape analysis procedure, together with the results will be also illustrated.
N25-214:

Depth Reconstruction Validation in Pixelated Semiconductor Detectors

W. R. Kaye, F. Zhang, J. E. Berry, Z. He Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA The depth of interaction in pixelated CdZnTe and HgI2 detectors is determined using either the cathode-to-anode signal ratio (CAR) or the drift time information. The CAR can be used if a ray interacts only once in the detector. The drift time must be used if the ray interacts over multiple pixels. The accuracy of these reconstruction techniques is evaluated using a collimated fan beam of rays that can scan across the entire depth of the detector, from the pixelated anode to the planar cathode. First, the beam position associated with the anode and cathode of a detector is determined. To measure the anode position, the polarity of the applied bias must be flipped to ensure that electrons, rather than the holes, drift through the bulk. Once the beam is properly calibrated to determine the actual cathode and anode locations, many reconstruction properties are studied. The discrepancy between the reconstructed depth and the true collimator position is found to be small for all detectors tested, with a minor reconstruction bias for interactions near the anode side. The thickness of the inactive layer near the anode has been measured to be less than 200m for two 15mm thick CdZnTe detectors. The experiments have also been used to evaluate the time-amplitude-walk calibrations, the depth reconstruction spanning various energies, and the depth reconstruction of multiplepixel events based on the drift time.

216

N25-215:

Recent Developments and Novelties on Segmented HPGe and Si(Li) Semiconductors

B. Pirard, J. O. Beau, C. Chassaing, P. Dressler, S. Heini, M. Zuvic, M. O. Lampert CANBERRA Lingolsheim, 1 chemin de la Roseraie, 67834 Tanneries Cedex, France High-level segmentation has become a key feature in the design of high-purity germanium (HPGe) detectors, and has allowed various improvements in the fields of x-ray and gamma-ray spectroscopy, especially in terms of background rejection, interaction tracking, and imaging. We report on detection performances and prospective applications of recently developed HPGe and Si(Li) segmented detectors.
N25-216:

M. S. Jalali1, S. Mirabbasi1, K. Iniewski2 1 ECE, University of British Columbia, Vancouver, Canada 2 Redlen Technologies, Victoria, Canada

Modeling Pixel-to-Pixel Capacitance in CZT-Based Imaging Systems

Simple formulas for calculating the pixel to pixel capacitance in a CZT readout system are presented. These formulas are intended to replace the need for solving a Laplace equation to get the pixel to pixel coupling capacitance for approximate calculations. The proposed formulas while simple result in accurate estimates. The percentage contribution of the pixel capacitance to substrate and pixel capacitance to adjacent pixels to the total pixel capacitance is also provided for different pixel spacing. Such insightful results can be used for design optimization. The pixel capacitance for a standard Redlen Technologies pixel is also given.
N25-217:

S. L. Ziegler1, K. F. Ferris1, B.-J. M. Webb-Robertson1, D. M. Jones2 1 Pacific Northwest National Laboratory, Richland, WA, USA 2 Proximate Technologies, LLC, Columbus, OH, USA

Materials Availabilty Considerations in the Design of New Gamma Detectors

The design and performance of semiconducting gamma radiation detectors are based upon physical materials properties, such as band gap, density, and electron mobility; however, an important and often overlooked materials property is the global and domestic availability of the elements proposed for the new radiation detection materials. In global and domestic security scenarios, the volume of material needed for deployment of gamma detectors may be significant and availability becomes a design concern. The balance of multiple materials properties for optimized performance has been the focus of our informationbased design studies, and this practical emphasis on parameter trade-offs continues through the evaluation of availability as a design concern and investigation of correlations between availability and more traditional material properties.
N25-218:

P. Rehak1, G. Carini1, W. Chen1, G. De Geronimo1, J. Fried1, J. Keister1, Z. Li1, D. Pinelli1, D. P. Siddons1, E. Vernon1, J. Gaskin2, B. D. Ramsey2 1 Instrumentation Division, Brookhaven National Laboratory, Upton, NY, 11973, USA 2 Marshall Space Flight Center, Huntsville, AL, USA Several Arrays of Hexagonal Silicon Drift Detectors (SDDs) were designed, produced and tested. The arrays are the hart of an Xray fluores-cence spectrometer (XRS) being developed for fine elemental mapping of the lunar surface. The project is being carried out by collaboration be-tween the NASA Marshall Space Flight Center (MSFC) and the Brook-haven National Laboratory (BNL). To attain the required sensitivity for elements present at the Moon surface the total sensitive area of all SDDs must be around 500 cm2. The total power consumption of electronics and detectors must not exceed 10W which corresponds to 20 mW/cm2 or 4mW/channel having 0.2 cm2 area individual SDDs. The consumption is due to SDDs and read-out ASIC. The ASIC read - out chip dissipates 32 mW and reads 16 channels or consumes 2 mW/channel. About the same power of 2mW/cm2 is budgeted for the individual SDD. To keep the dissi-pated power on the SDD low there is no integrated field effect transistor (FET) in the SDD, rather, the readout is done by wire-bonding the anodes to the inputs of the ASIC. The ASIC provides lownoise charge amplifica-tion, high-order shaping with baseline stabilization, discrimination, and peak detection with analog memory. Our preliminary measurements show a FWHM of 153 eV at the 5.9 keV peak of a 55Fe source.

Array of Hexagonal SDDs as X-Ray Spectrometer

217

N25-219:

Performance of CdZnTe Pixellated Radiation Detectors Assembled by a New Attachment Method

P. Lu, H. Chen, S. Awadalla, G. Bindley Redlen Technologies, Sidney, BC, Canada This paper presented a new carrier board attachment method for pixellated CdZnTe (CZT) radiation detectors by using a special type of anisotropic conductive film (ACF). Due to CZT detectors high sensitivity to attachment conditions, its anode electrode attachment had always been a challenge for detector module developers. It was especially true for the detectors with short pitch or small pixel patterns. A type of ACF with very small pitch, high vertical electrical conductivity and strong mechanical strength was found to be suitable for pixellated CZT detector assembly. Attachment processes and conditions were optimized. Detector fabrication processes were improved to match the attachment requirements. Attached detector modules showed excellent spectra responses. In addition, detector modules long-term stability and reliability tests showed promising results. This ACF attachment technology had been successfully used for pixellated CZT detectors with various physical dimensions and anode pixel patterns.
N25-220:

Modified Vertical Bridgman Growth of Cd(1-x)Zn(x)Te Detector Grade Crystal in a 4" EDG

Furnace
A. Datta Center For Materials Research, Washington State University, Pullman,Washington, United States of America Achieving a high yield of CZT single crystals is one of the greatest challenges in CZT growth . Over 75 of 85-95 mm long CZT crystals were grown in a 43 zone modified low pressure Vertical Bridgman method using an Electro-Dynamic Gradient (EDG) Freeze 3" Mellen Sunfire furnace at a rate of 0.5 to 2mm/hr. The heating coils used in this furnace are made of Fe-Cr-Al (Kanthal APM). A new Electro-Dynamic Gradient (EDG) Freeze 4" Mellen Sunfire Platinum furnace has been developed to visualize the effects of increased ampoule diameter on the grain structure and spectroscopic properties of CZT. Due to low thermal conductivity of CZT, bigger diameter crystals would result in higher radial thermal gradients which affects the stress and thermal convection during the crystal formation process. Also as the power output of the 4" furnace is 25% higher , it is possible to maintain a higher and accurate axial temperature gradient which should result in better crystal structure. Consequences of these facts will be explored. The comparison in the grain structure, single crystal yield and spectroscopic properties will be discussed among crystals grown in both furnaces. The detector grade CZT crystal grown in our 3" furnace shows a bulk resistivity of 10^10 Ohm*cm, resolution of average 3% FWHM at 662 keV Cs spectra (5mm x 5mm pixelled cube,as shown below ) and a MuTau product for electrons of the order of 102 cm^2/V. The biggest single crystal detector made in our research Lab was 20 mm x 20mm x 15mm using 3" furnace. The reproducibility of the detector grade crystal using the 4" furnace will be discussed and statistics will be shown. Infrared microscopy images of Te inclusions using Olympus BX51M microscope will also be compared to see the changes in average density and percent volume of the inclusions in the samples from both furnaces.
The author thanks the US Department of Energy, NA-22, Contracts DE-FG52-06/27497/A000 and DE-FG52-08NA28769 for their financial support of this research. N25-221:

Y. Cui1, A. Hossain1, A. Bolotnikov1, G. Camarda1, G. Yang1, K. H. Kim1, R. B. James1, A. Mycielski2, D. Kochanowska2, M. Witkowska-Baran2 1 Brookhaven National Laboratory, Upton, NY, USA 2 Institute of Physics, PAS, Warsaw, Poland CdMnTe can be a good candidate for gamma-ray detection because of its wide band-gap, high resistivity, and good electrontransport properties. Our group at Brookhaven National Laboratory and Institute of Physics (Polish Academy of Sciences) has expended efforts in the growth and characterization of CdMnTe crystals for gamma-ray detection purposes. Recently, we are growing CdMnTe crystals with high purity Mn and different dopants (i.e. Indium and Vanadium). In this paper, we discuss the characterization of these newly grown crystals and show our results with gamma-ray sources.
N25-222:

Recent Results on Development of CdMnTe Gamma-Ray Detectors

Ab Initio Investigation of Layered Materials for Semiconductor Radiation Detectors

C. R. Leao, V. Lordi Lawrence Livermore National Lab, Livermore, CA, USA Semiconductors require several properties to be useful as highly sensitive radiation detectors that operate at room temperature. High density and high atomic number is required for strong absorption of radiation. A moderate band gap is needed to maximize the electronic signal created by the incident radiation, while minimizing thermal noise. High resistivity is desired to minimize noise from leakage current. Finally, long drift lengths are required to maximize carrier collection and achieve high energy resolution, implying the need for high mobilities and few carrier scattering centers. These last requirements pose the greatest 218

challenges for the fabrication of efficient radiation detectors. Experimental techniques for producing high-purity, highlycrystalline semiconductors are mature for many traditional materials, such as Si or Ge, but not for many materials of interest for room-temperature detection which have large enough atomic number and band gap. Much of the progress in producing materials to try to satisfy the above properties have been driven by trial and error, with expensive and time-consuming efforts focused on maximizing purity and crystallinity. In this work, we use predictive ab initio simulations to direct and accelerate the synthesis process, for the case of layered semiconductors GaTe and GaSe. Simulations are used to assess the limitations that hinder the application of a candidate material for use in radiation detectors, and suggest improvements. Our approach consists of identifying among the native defects and accidental impurities in the material which ones are the most detrimental. This is evaluated based on both the predicted equilibrium concentration of the respective defect as well as the strength of the carrier scattering and carrier capture induced by it. Furthermore, defect compensation schemes are suggested to mitigate the effects of the growth imperfections.
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344. N25-223:

Characteristics of Large-Volume Redlen CdZnTe Detectors

Z. He, F. Zhang, Y. F. Zhu, W. Kaye, Y. Boucher, C. Herman Nuclear Engineering and Radiological Sciences, The University of Michigan, Ann Arbor, Michigan, USA CdZnTe is currently the most promising wide band-gap semiconductor gamma-ray detector for national security and homeland security applications. In order to demonstrate that large-volume CdZnTe detectors with high spectroscopic performance can be manufactured by the THM crystal growth method and the annealing process developed by Redlen Technologies, the University of Michigan has received and tested more than thirty 2x2x1 cm^3 and six 2x2x1.5 cm^3 CdZnTe detectors manufactured by Redlen since 2008. Our test results demonstrate that energy resolution of about 1% FWHM at 662 keV can be consistently achieved on these large-volume CdZnTe detectors. The performance and characteristics of Redlen CdZnTe detectors, including energy resolutions for single- and multiple-pixel events as functions of 3-dimensional position of interactions, electron mu-tau product, electron steering and collection properties, will be reported.
This work has been jointly supported by DOD DTRA and DOE NA-22 Office N25-224:

W. Chen1, G. A. Carini1, G. De Geronimo1, J. A. Gaskin2, J. Keister1, Z. Li1, B. D. Ramsey2, P. Rehak1, D. P. Siddons1 1 Brookhaven National Lab, Upton, NY, USA 2 a. NASA/MSFC/National Space Science and Technology Center, Huntsville, AL, USA We have developed a new method to produce Thin-Window Silicon Drift Detectors (ThinWinSDD). Each individual ThinWinSDD has a thin entrance window on one side and a hexagon spiral shaped cathode around the center anode on the other side. To produce the thin entrance window of ThinWinSDD a double implantation was used, consisting of Boron ions with 10keV (1x1014/cm2) and Phosphorus ions with 50keV (4x1012/cm2), through a 500 silicon dioxide. The purpose of the second Phosphorus implantation is to compensate for the tail portion of the Boron ion implantation so that the net Boron ion distribution will result in a thinner junction in the silicon material. We will compare the test results in U3C beam line in NSLS at BNL of this newly developed method with the previous ones, where the thin junction was created by using a single implantation of Boron ions with 10keV (1x1014/cm2) through a 500 silicon dioxide.
N25-225:

Comparison of Two Different Methods to Produce Thin-Window Silicon Drift Detectors

D. H. Chivers1,2, L. Mihailescu2, S. G. Prussin1, K. Vetter1,2 1 Nuclear Engineering Department, University of California, Berkeley, Berkeley, CA, USA 2 Nuclear Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA

Space Charge Effects of Charge Clouds in Large HPGe Planar Double-Sided Strip Detectors

High-purity germanium (HPGe) double-sided strip (DSSD) detectors are used in a wide number of applications where high spatial and energy resolution of x-ray and -ray interactions. In many applications there is a relatively high probability of charge collection within the gaps between electrodes where charge loss is observed. This work investigates the basic mechanisms for the spreading of charge clouds within germanium detectors. Theoretical predictions for charge clouds produced by x- and -ray interactions result in a much greater charge spreading than expected from pure diffusion effects at drift times less than 1 microsecond. This is due to the high magnitude of space charge density resulting from the high stopping power of germanium. Gap measurements within an HPGe DSSD detector have been used to probe the charge spreading parameters. Preliminary results show that charge cloud spreading is dominated by space charge repulsion.

219

This material is based upon work supported by the U.S. Department of Homeland Security under the Grant Award Number 2008-DN-077-ARI001-02. N25-226:

D. H. Chivers1,2, L. Mihailescu2, S. G. Prussin1, K. Vetter1,2 1 Nuclear Engineering Department, University of California, Berkeley, Berkeley, CA, USA 2 Nuclear Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA

Pulse Shape Analysis Method for Estimating Charge Mobility in Large Planar HPGe Double-Sided Strip Detectors

Segmented high-purity germanium (HPGe) detectors are used in a wide number of applications where high spatial and energy resolution of x-ray and -ray interactions. In many of these applications, pulse shape analysis is performed to achieve a greater precision of position information. In order to predict correct pulse shapes within the detector, the mobility of electrons and holes need to be computed for varying magnitudes of electric field and, in some cases, varying crystallographic directions. This work is an investigation of the charge mobility within large planar HPGe detectors with segmented orthogonal strip electrodes. A number of scanning experiments have been performed yielding drift time information within the signal shapes thus enabling an estimation of charge mobility parameters. Three HPGe orthogonal strip detectors containing varying impurity concentrations and thicknesses were scanned. A number of bias voltage magnitudes were also used to ensure that a large range of electric field magnitude was represented. Preliminary results have shown consistent values for electron mobility and a lower than expected hole mobility relative to previously published values.
This material is based upon work supported by the U.S. Department of Homeland Security under the Grant Award Number 2008-DN-077-ARI001-02. N25-227:

CZT Detector for Remote Nuclear Material Measurement

S.-H. Park, Y. H. Cho, J. H. Lee, H. S. Kim, J. H. Ha KAERI, Daejeon, Korea CZT semiconductor is promising detector material for X-ray and -ray spectroscopy. The CZT detector could have high energy resolution in X-ray and -ray spectroscopy without a large cooling system. Especially, the CZT detector is very attractive choice when the detector should be small and placed closer to the radiation emitting material such as nuclear spent fuel. A -ray spectroscopy system based on CZT detector was designed and fabricated to measure the -ray in the energy range 60 ~ 700 keV. Energy spectrum of CZT detector with planar electrode show large low energy tail when high energy -ray (> 100 keV) was measured. We made three type CZT detectors (strip detector, Coplanar grid detector and Frisch collar detector) with our own electrode design to reduce the low energy tail. CZT crystals grown with High Pressure Bridgman method were used in our work. The surfaces of the crystal were polished with mechanical and chemical way, and gold electrodes were deposited. When the strip electrode detector and Coplanar grid detector were made, the electrodes were deposited with evaporation method, and when the Frsich collar detector was made, the electrodes were deposited with electroless deposition method. All detector show the reduced low energy tail in the energy spectrum. Frisch collar detector was chosen to make the -ray spectroscopy system for remote nuclear material measurement. Preamplifier designed for CZT detector was placed close to the detector, and signal line between preamplifier and amplifier was ~ 10 m. The detector system was made waterproof, and it could be used to measure the -ray from nuclear spent fuel in pool.
This work was performed under the nuclear research and development program sponsored by Ministry of Science and Technology of Korea, and supported partly by the Basic Research Supporting Program of Korea Research Foundation. N25-228:

Y. H. Cho1, S.-H. Park1, J. H. Lee1, H. S. Kim1, Y.-K. Kim2, J. H. Ha1 1 KAERI, Seoul, Korea 2 Hanyang University, Seoul, Korea CZT semiconductor is a proven material for X-ray and -ray energy spectroscopy. However, the wide application of CZT detector was mainly hindered from the uniform large crystal growing. CZT crystal was grown, impurity (indium and Cl) was doped to increase the resistance of CZT crystal, and material uniformity of each crystal was measured and compared. The crystal was grown with Low Pressure Bridgman method. The raw materials (7N Cd, Zn and Te) were vacuum sealed in a quartz amplue. In our Bridgman furnace design, the ampule was placed stationary and a heater was moved with speed of 1 mm/h during the crystal growing. The diameter of the grown crystal was 3 cm. Indium and Cl (~1018 /cm3) was doped to increase the resistance of the crystal. The ingot was cut with diamond wire saw with 1 mm thickness, and the surface was polished with mechanical and chemical way. Photoluminescence (PL) and FT-IR analysis of each CZT crystal were done to see the band structure and defect of the crystal. Gold electrode with diameter of 1 mm was deposited on both sides of the sliced crystal. I-V curve and radiation 220

Material Uniformity of Doped CZT Crystals

energy spectrum were measured with each sliced crystal. The crystal characteristics such as band structure, defects, resistance were compared along the crystal axis. The crystal uniformity was measured as the It could be seen that the crystal resistance could be increased 103 when the impurity was doped in the crystal. Our work could be helpful to obtain the uniform crystal with low pressure Bridgman method.
This work was performed under the nuclear research and development program sponsored by Ministry of Science and Technology of Korea, and supported partly by the Basic Research Supporting Program of Korea Research Foundation. N25-229:

Amorphous Semiconductor Blocking Contacts on CdTe Gamma Detectors

L. F. Voss, A. M. Conway, B. W. Sturm, R. T. Graff, R. J. Nikolic, A. J. Nelson, S. A. Payne Lawrence Livermore National Laboratory, Livermore, CA, USA Development of high resolution gamma ray detectors based on CdTe is dependent on low background noise levels. This noise is often the result of leakage current through the devices, which generally operate as resistors. Because of this, increasing the resistivity of the CdTe has been one of the primary goals of research in these materials. However, an alternate approach to achieving similar results is to increase the effective resistivity through the use of metal Schottky contacts. Researchers have also demonstrated that amorphous- crystalline semiconductor heterojunctions can improve the blocking behavior contacts for radiation detectors. In this work, we investigate the effect of amorphous Ge layers on oriented CdTe on leakage current and the gamma detection response.
N25-230:

H. Kim1, A. Churilov1, G. Ciampi1, L. Cirignano1, W. Higgins1, F. Olschner2, K. Shah1 1 Radiation Monitoring Devices Inc., Watertown, MA, USA 2 Cremat Inc., Newton, MA, USA

Thallium Bromide and Thallium Bromoiodide Gamma-Ray Spectrometer Development

Thallium bromide (TlBr) is of interest as a material for room temperature gamma ray spectroscopy due to its high density, high Z and wide bandgap. In addition, its cubic crystal structure and relatively low melting point facilitate crystal growth by melt techniques. Recent advances in material purification, crystal growth and device processing have led to mobility-lifetime products of electrons in the mid 10-3 cm2/V range enabling working detectors up to 15-mm thick to be fabricated. Although performance of TlBr devices is promising, long term detector stability at room temperature is an issue. We are investigating various compositions of the ternary compound, thallium bromoiodide (TlBrxI1-x) to vary the band gap and determine the effect of added thallium iodide (TlI) on detector performance. In this paper we report on our recent progress in TlBr gamma-ray spectrometer development as well as our initial results from TlBrxI1-x devices. Results from TlBr detectors up to 15-mm thick will be presented including depth corrected pulse height spectra . Detector stability will also be discussed. This work is being supported by the Domestic Nuclear Detection Office (DNDO).
N25-231:

Vapour Phase Growth of 100 mm Diameter Cadmium Telluride Boules on GaAs Seeds

S. A. Sakwe, J. T. Mullins, B. J. Cantwell, A. T. G. Pym, A. Basu Kromek, Sedgefield, U.K. Hetero-epitaxially seeded growth of bulk crystals of cadmium telluride and cadmium zinc telluride using a multi-tube physical vapour transport (MTPVT) growth system [1] has recently been demonstrated [2]. This technology offers a route to the production of large (several cm3), high quality, bulk single crystal boules of these materials with a high yield. Such crystals are ideally suited for the fabrication of high energy X-ray and -ray detectors. In this paper, we report the scaling of this process, for cadmium telluride growth, from 50 mm to 100 mm in boule diameter. It is possible to produce boules of usable material greater than 5 mm in thickness and so this process offers a route to the production of large area devices with high stopping power for high energy radiation. The grown crystals have been characterized by X-ray diffraction, infra-red absorption and contactless resistivity mapping. References 1. J.T. Mullins, J. Carles, N.M. Aitken, A.W. Brinkman, J. Crystal Growth 208 (2000) 211. 2. J.T. Mullins, B.J. Cantwell, A. Basu, Q. Jiang, A. Choubey, A.W. Brinkman, B.K. Tanner. J. Electron. Mater. 37 2008 1460.

221

N25-232:

Study and Characterisation of Guarded Small Pixel Contact Geometries on CdTe

M. Ayoub, P. D. Scott, F. Dierre, I. Radley, A. Basu Kromek, Sedgefield, U.K. Results from CdTe detectors using a guarded small pixel geometry have shown excellent spectral resolution which is comparable to that seen from Schottky diode devices. Unlike planar Schottky detectors, the guarded small pixel geometry operates well at lower electrical fields and does not require the use of a peltier cooler. A high spectral resolution is achieved by applying a guard voltage which both reduces leakage current and tunes the electric field inside the device. Detectors are characterised using a 57Co source and show a 122keV peak to background ratio of greater than 20 with an energy resolution, which is currently limited by electronic noise from the measurement system. We show the results from devices with different pixel to guard spacing and discuss the variation in energy resolution, count rate and leakage current as a function of guard ring bias and the relation to the electric field profile inside the device.
N25-233:

Growth of Thick CdTe:Cl Layers of Large Diameter on ZnTe Buffered Ge Seeds for Room Temperature Radiation Detection

F. Dierre, A. T. G. Pym, M. Ayoub, M. J. Winter, S. A. Sakwe, I. Radley, A. Basu Kromek, Sedgefield, U.K. We report the development of a variation of our multi-tube physical vapour transport (MTPVT) technique for the growth of CdTe:Cl layers on a Ge seed with a ZnTe buffer layer. With this development, we are able to produce reproducible 51 and 100 mm diameter bulk single crystal CdTe:Cl oriented (100), with a thickness of few millimetres. This method has the potential for larger diameter growth and is only limited by the actual diameter of the Ge seed. In this paper, we will present and discuss the basis of our technique, outlining important technological steps required for the different materials, and the characterisation carried out. The material has been characterised using X-ray diffraction, infra-red mapping, resistivity mapping by time-dependent charge measurement (TDCM). The X-ray diffraction rocking curve maps show a high level of homogeneity with widths of tens of arcseconds. A high resistivity (of magnitude 109 .cm) has been measured across the thick layers. Moreover, a low density of inclusions has been observed. The electrical performance will be discussed including -ray detection performance and mobility-lifetime product measurements.
N25-234:

F. Dierre1, M. Ayoub1, R. L. Thompson2, A. T. G. Pym1, I. Radley1, A. Basu1 1 Kromek, Sedgefield, U.K. 2 Dept. of Chemistry, University of Durham, Durham, U.K.

New Approach to Deposition of W and Mo on CdTe via the Electroless Method

Contacts deposited by the electroless method are frequently used for II-VI semiconductor materials and particularly on CdTe/CdZnTe. This chemical method of deposition creates a stronger chemical bond between the contact and the semiconductor when compared to physical deposition methods such as sputtering and evaporation. This method also forms a quasi-ohmic contact which eliminates the problem of spectral degradation due to the polarisation effect. A limited number of metallic contacts has been deposited by this method, namely Au, Pt and Pd. In addition to the intrinsic material properties, the end device performance depends strongly on the contact type, surface treatment and deposition method. The aim of this work was to study two different contacts: Mo and W deposited onto Cl doped CdTe by the electroless method. These elements are of particular interest as they both have a higher work function than CdTe and a large number of valence electrons. A two step chemical process has been used to deposit these contacts and will be outlined along with the mechanisms of deposition. The chemical composition and thickness of the interfacial layer between the contacts and semiconductor have been determined by Rutherford Back Scattering (RBS). In addition to this, some initial results of electrical performance studies will be discussed including IV measurements and -ray detection performance. These will be compared to device performance results from samples with standard electroless Pt contacts.
N25-235:

Bridging the Price / Performance Gap Between Silicon Drift and Silicon PIN Diode Detectors

D. Hullinger, K. Decker, J. Smith, C. Carter Moxtek, Inc., Orem, UT, USA Use of silicon drift detectors continues to increase as new XRF applications drive performance improvements beyond existing silicon PIN detector capability for new XRF instrumentation. Recent developments in Si-PIN detectors result in significantly increased performance. We will describe the improvements in spectral resolution, count rate capability, peak to background, and 222

detection limit improvements in silicon PIN diode detectors. These improvements result in detectors that, in some applications, offer an attractive alternative to silicon drift detectors at a reduced cost. This improvement bridges the price/performance gap between existing silicon PIN detector capability and the SDD.
N25-236:

Positron Analysis Based on High Energy X-Ray Source

Y. Yang, T. Li, Y. Zhang, P. Yang Deparment of Engineering Physics, Tsinghua University, Beijing, P.R.China Positron analysis can reveal the information of atomic level defect of materials like metals or polymers through the study of Doppler broaden 511keV spectra of positron with electrons that have different momentum distributions relating with the inner defect of materials. The usually used positrons come from isotope or accelerator positron source that irradiates the materials with positrons from outside. Because of the short range of positron in matter, only the surface of millimeter depth inside metals or polymers could be analyzed. To fulfill the demand of positron analysis inside the materials, high energy X-ray is used to produce positron. Two kinds of reaction is feasible, (1) is positron decay after the (, n) reaction of high energy photon with low photoneutron threshold nuclide, like 14N or 54Fe; (2) is the pair production. We use the first reaction in this paper. The algorithm to analyze the broaden 511keV spectra is S factor or FWHM method. Maximum likelihood expectation method is also researched to deconvolute the spectra to remove the detector response for more precise electron momentum information extraction. We setup the facility with 15MeV LINCA and HPGe detector, and then test several specimens including low carbon steel, cast iron, aluminium and copper. The experimental results of five low carbon steel specimens show appreciable correspondence between positron analysis result and tensile elongation. This indicates that the atomic level defect of low carbon steel is within the range of positron analysis. On the other hand, cast iron, aluminium and copper do not show such relation. This is perhaps due to the void in these materials is so much that positron analysis method is saturated. Keywords: positron analysis, atomic level defect, photoneutron reaction, deconvolution
N25-237:

Kodak CCD-Based Detector for Small Angle X-Ray Scattering

H.-L. Lee, T. Madden, P. Fernandaz, B. Lee, S. Seifert, J. Weizeorick, M. Molitsky BTS, XSD, Argonne National Laboratory, IL, USA The Beamline Technical Support Group at the Advanced Photon Source (APS) at Argonne National Laboratory has developed two CCD detectors for small angle x-ray scattering (SAXS): a single-CCD prototype system and a two-by-two Mosaic CCD system. Both of these systems, optimized for sensitivity, utilize the Kodak KAF-4320E CCD coupled to fiber-optic tapers, custom mechanical hardware, electronics, and software developed at Argonne. The two-by-two CCD system is now in active service in sector 12 of the APS for small-angle scattering experiment (SAXS) with good results. The single-CCD system is in the detector pool and is meant to be shared equipment for the general synchrotron user public of the APS. The signal conditioning electronics and mechanical design of the detectors will be described. SAXS data from the two-by-two CCD system will be shown and discussed.
N25-238: The Electronics in the Detector Head of the AGIPD-Detector -- a 1M Pixel, 5 MHz Camera for the European XFEL

P. Goettlicher Deutsches Elektronen-Synchrotron, 22607 Hamburg, Germany On behalf of the AGIPD-consortia The European free electron laser (XFEL) will provide trains of 12keV X-ray photons. With a repetition frequency of 10Hz 0.6ms long trains with 3000 bunches will be generated. The bunch to bunch distance is 200ns. For the experiments at the XFEL the AGIPD consortia develops a 2-dimensional 1Mega pixel detector to record individual pictures for the 200ns separated bunches. As a baseline design up to 400 pictures per train will be stored as analogue values in capacitors inside ASIC's. To keep the leakage of the charge low a fast digitizing system will be developed as board level electronic system housed in the detector head. 1024 ADC's each with 50MS/s and 14bit will be mounted in 16 modules. They generate a data rate of 16*51Gbit/s into 16 FPGA's, one per module. After first data corrections and pre-sorting the data of each module will be sent on a standard IT-link to a backend system. As links 10G Ethernet is investigated. VHDL-code for a UDP protocol was developed and implemented into the logic slices of a VIRTEX-5. Performance tests show, that 99% payload can be achieved, far above the required 55%. For the control of the detector head and its synchronization to the accelerator a data link and fast clocks are specified. A few fast bunch synchronized signals will be transmitted via dedicated differential signal lines and handled in the detector head by FPGA;s. For the slow control and the train synchronic information exchange (100ms) microcontrollers will be implemented. The information exchange will use Ethernet with the reliable protocol TCP/IP.

223

The basic building blocks have been investigated on evaluation boards and own developments. First results will be presented. After getting more experience with the performance of the different parts of the AGIPD-project custom PCB's will be developed.
The work is supported by the XFEL-GmbH. N25-239:

U. Trunk1, J. Becker2, P. Goettlicher1, H. Graafsma1, B. Henrich3, H. Hirsemann1, S. Jack1, M. Karagounis4, R. Klanner2, H. Krueger4, A. Mozzanica3, H. Perrey2, J. Pintilie2, G. Potdevin1, B. Schmitt3, J. Schwandt2, A. K. Srivastava2, C. Youngman1, X. Shi3, I. Sheviakov1, M. Zimmer1 1 DESY Deutsches Elektronen-Synchrotron, D-22607 Hamburg, Germany 2 Universitaet Hamburg, D-22761 Hamburg, Germany 3 PSI Paul Scherrer Institut, CH-5232 Villigen, Switzerland 4 Universitaet Bonn, D-53115 Bonn, Germany The European XFEL under construction in Hamburg will provide fully coherent, 100 fs long X-ray pulses, with 1012 photons at 12keV. The high intensity per pulse will allow recording diffraction patterns of single particles or small crystals in a single shot. As a consequence the 2D detectors have to cope with a large dynamic range in the images (one to 104 photons/pixel). An additional challenge is the European XFEL machine: an Electron bunch train with 10 Hz repetition rate, consisting of up to 3,000 bunches with a 200 ns spacing. This means that recorded images have to be stored inside the pixel during the bunch trains and read out between bunch trains. In order to meet these requirements, the European XFEL has launched 3 detector development projects. The AGIPD project is a collaboration between DESY, PSI and the Universities of Bonn and Hamburg. The goal is a 1000x1000 pixel detector, with 200 m pixel size and a central hole for the primary beam. The ASIC operates in charge integration mode: the output of each pixel preamplifier is proportional to the charge from the sensor generated by the X-rays. The input stage of the pixel cells will have dynamically adjustable gains. The output signal is stored in an analogue pipeline, which has to be a compromise between noise performance and the number of images. 200 to 400 images have to be readout and digitized in the 99.4 ms long bunch gap. The detector will be built of 2x8 fully depleted monolithic silicon sensors with a 2x8 array of CMOS readout chips bump-bonded to these. The interface electronics is designed to process 400 images in 99 msec, without compromising single photon sensitivity or the full dynamic range. Since the ASIC is the linchpin of the project, several MPW runs to test the different aspects in terms of radiation hardness, noise and adaptive switching are submitted. We will give a general overview and report on the current status.
N25-240:

The Adaptive Gain Integrating Pixel Detector (AGIPD): a Detector for the European XFEL. Development and Status.

Front-End Electronic System of PMT Readout for Daya Bay Reactor Neutrino Experiment

Q. Li Experimental Physics Center, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing, China On behalf of the Dayabay Collaboration As the main component of photomultiplier tube (PMT) readout electronic system for Daya Bay Reactor Neutrino Experiment, a key project proposed by Chinese physicists, a feasible electrical charge and time readout circuit scheme is studied in this thesis. It is for the readout of signals coming from both the Central Detector and the Veto Water Cherenkov Detector. Not only charge and time measurement are demanded simultaneously, but strict specifications, such as low noise disturbance and large dynamic range (16 bits), are also required at possible low cost. The readout system has features of a shorter dead time, high parallel processing capability and more flexibility, etc. By applying high density powerful FPGA technique, all data processing for both charge measurement and time measurement of the whole 16 channels and VME interfacing are completed on a single FPGA device. Introducing the front-end data processing has lowered the amount of data transfer of the whole system and will save more CPU time originally consumed from this stage. Comparing with other similar DAQ systems designed in Europe and USA, our system has a larger dynamic range and a better resolution for small signals.
N25-241:

FPGA Remote Configuration Through VME

H. T. Dong, C. Cuevas, E. Jastrzembski, J. Wilson Physics Division, Jefferson Lab, Newport News, VA, USA The custom front-end data acquisition and trigger electronics system designed for Jefferson Labs 12 Gev upgrade experiments incorporates thousands of Field Programmable Gate Array (FPGA) manufactured either by Xilinx or Altera. The functions 224

executed by the FPGA are determined by bit patterns generated from Very High Speed Integrated Circuit Hardware Description Language (VHDL) programs by compilers such as Altera Quartus or Xilinx ISE running on a personal computer (PC). The standard method to transfer these bit patterns from the PC to the FPGA or its associated Read Only Memory (ROM) is through a vendor supported Joint Test Action Group (JTAG) cable connected directly to the target circuit board. Because the JTAG commands and data are serialized, configuration of a single FPGA may take minutes. We anticipate that the functions performed by the FPGA will be modified after the complete system has been installed, so the standard method of configuration is not acceptable. A remote method of FPGA configuration that eliminates the need to connect JTAG cables to each circuit board is required. Since most custom boards in our system communicate through VME bus protocols, it is natural to use this shared data path to transport configuration data to the FPGA. The appropriate FPGA configuration bits are written as 32-bit words to the target circuit board where they are temporarily stored in random access memory (RAM). When all configuration words have been written, a local process (instantiated in an FPGA) is started that executes the proper JTAG commands to configure the FPGA over a private JTAG link on the circuit board. Unlike the JTAG transfers, the transfer of configuration words across the VME bus is very fast (20 MByte/s). Because of the parallel nature of our scheme, we can reconfigure the entire system in a time only slightly longer that of a single target board.
N25-242:

X. Cheng1, X. Tian2, R. Fan1,2, M. Zeng1 1 The department of Engineering Physics, Tsinghua Univ., Beijing, China 2 Northwest Institute of Nuclear Technology, Xi'an, China

An Ultra-High-Speed Front-End Intelligent Digital Transmission System for High-Intensity Pulsed Radiation Field Diagnosis

A Real-time Sampling System for Transient Pulsed Signals (RSTPS system) has been designed and implemented, which has been introduced in the 16th IEEE NPSS Real Time Conference, and this system is recently upgraded. Now the bandwidth is up to 448MHz with a dynamic range over 40dB. Some new functions are developed in order to make sure that the transient pulsed signal can be discriminated from the interference signals and recorded reliably in the high-intensity pulsed radiation field, which make the RSTPS system as an ultra-high-speed front-end intelligent transmission system. The most significant modification is that the sampling process of this system will be automatically pre-triggered only by the signals with designated characteristic. All the new functions are implemented in a FPGA and tested to be functional. Finally, the application of transmitting transient pulsed signals produced by the X-ray detecting in the Z-pinch experimentation is completed, to verify the proper operation of this system when being used in the high-intensity pulsed radiation field.
Thanks to the sustentation of the"Ultra High-speed Full-optical A/D conversion in Optical Communication" program, whose ID is 2007AA01Z275 in the "Chinese National Programs for High Technology Research and Development "(863 Programs). N25-243:

G. Bianchi1,2, R. Castaldi1, L. Fanucci2, G. Magazzu'1, S. Saponara2, C. Tongiani1,2, P. G. Verdini1 1 INFN - Sezione di Pisa, Pisa, Italy 2 University of Pisa - Diparimento di Ingegneria dell'Informazione (DII-EIT), Pisa, Italy

FF-LYNX: integrated Control, Trigger and Readout in Future High Energy Physics Experiments

In future High Energy Physics (HEP) experiments the systems for the distribution of the Timing, Trigger and Control (TTC) signals and for the Data Acquisition (DAQ) will share most of the requirements concerning data rates, trigger latency, robustness against transmission errors and component failures, radiation hardness and power dissipation in hardware components. The FFLYNX project aims at the definition of a general purpose protocol for the DAQ and the distribution of the TTC signals and at its implementation in radiation tolerant and low power interfaces designed and developed in standard CMOS technologies (180nm, 130nm, 90nm) and available as Intellectual Property (IP) cores to the designer of the integrated circuits for the future experiments. The FF-LYNX protocol will be described: integrated (i.e.: handled by the same protocol and the same hardware components) TTC and DAQ functions, robustness of critical signals (clock, trigger) against transmission errors, general structure of data frames, transparency with respect to different data types, easy coupling of the interfaces with the core of the host circuits, flexibility with respect to the architecture of the control and readout systems, compatibility with different link technologies and speeds. The architecture of the interfaces will be also presented and the solutions adopted for the transmitter/receiver synchronization, the arbitration between high priority and fixed latency packets (i.e.: triggers and trigger data frames) and low priority packets (e.g.: commands and raw data frames) and the reduction of the radiation effects will be described in detail. A test chip with embedded prototypes of the transmitter and receiver interfaces is under development in a CMOS commercial 0.18um technology and will be submitted to the foundry in December 2009. The architecture of the test chip and of the test bench that will be developed for its test and characterization will be also sketchily described.
The FF-LYNX project is funded by the INFN 5th Commission.

225

N25-244:

Track Reconstruction in the ATLAS High Level Trigger Using Cosmic Ray Muons

S. Ask, J. L. Lane, J. Masik School of Physics and Astronomy, The University of Manchester, Manchester, United Kingdom The large collision rate together with the complicated event signatures at the LHC implies critical requirements on the trigger system. One very important ingredient of the event selection is the track reconstruction capability. We present the track reconstruction performed in the ATLAS trigger system based on data from the inner detector and the results obtained from cosmic ray muon tracks used for the trigger decision. The results show a good overall performance, consistent with expectations.
N25-245:

X. Wang1,2, H. Gong1, Y. C. Lin1, G. H. Gong1, S. M. Chen1, B. B. Shao1 1 Department of Engineering Physics, Tsinghua University, Beijing, China 2 College of Electronic Science and Engineering, National University of Defense Technology, Changsha, China

The Control and Monitor Interface of the Trigger Board for the Daya Bay Neutrino Experiment

We have designed a local trigger board for the Daya Bay Neutrino Experiment, which is aimed to measure the neutrino mixing angle sin2213 with a precision down to 1% level at 90% C.L.. A Virtex-4 FPGA is employed to implement the trigger logic. To control the trigger logic and monitor the status of the local trigger board, an interface based on VME bus is implemented on the local trigger board. Inside FPGA, the VME bus is converted into a local bus, which can access lots of registers and RAM blocks. On the VME master board, we designed a C code program, which can control and monitor the local trigger board by accessing the registers and RAM blocks inside the FPGA of the local trigger board. This control and monitor interface is very important to guarantee the proper running of the local trigger board. This talk will present the design of the control and monitor interface.
N25-246:

V. Izzo1, A. Aloisio1,2, L. Capasso1,2, F. Cevenini1,2, M. Della Pietra3, R. Giordano1,2 1 Istituto Nazionale di Fisica Nucleare, Napoli, Italy 2 Universita' di Napoli FEDERICO II, Napoli, Italy 3 Universita' di Napoli PARTHENOPE, Napoli, Italy The number of interactions per bunch crossing for the upgrade of the Large Hadron Collider is expected to be ten times greater than the actual one. As a consequence, the ATLAS detector for SLHC foresees the use of a larger number of readout channels and also a new trigger level is under development. In order to face with such issue, we developed a new architecture for the Read Out Driver (ROD) for the ATLAS RPC Muon Spectrometer in the barrel region. Presently, each ROD board receives ATLAS RPC Muon readout data and arranges all the data fragments of a sector of the spectrometer in a unique event, sending it to the next acquisition systems. Our new design is based on the new generation Xilinx Virtex5 FPGA and it works with a clock frequency six times greater than the actual bunch crossing rate of the LHC. We also implemented the output channel of the ROD, presently based on S-Link protocol, by using the GTP transceivers inside the FPGA. We are considering to port the design from VME 64x toward the VXS standard, which offers high speed serial links.. We present an overview of our design, focusing on the newly added hardware features.
N25-247:

The ATLAS RPC ROD for Super LHC

S. Riboldi1, A. Pullia1, F. Camera1, F. Zocca2, C. Ur3, R. Isocrate3, D. Bazzacco3 1 Physics, Universita' degli Studi di Milano, Milano, Italy 2 INFN - Milano, Milano, Italy 3 INFN - Padova, Padova, Italy

A Multi-Channel Digital Acquisition System for Nuclear Spectroscopy Experiments

A multi-channels digital acquisition system for nuclear spectroscopy applications with HPGe detectors has been designed, manufactured and tested. It is based on a front-end electronics for analog signal conditioning, followed by a free running analog to digital converter. The digital output data-stream is collected and transmitted on copper wires by means of programmable logic devices and dedicated serializers. In order to satisfy the needs of actual multi-channels experiments, we implemented additional functionalities into the digital acquisition system, e.g. trigger, baseline estimator, trigger counter, self-alignment of ADC readout clock and data, etc. Provided an initial suitable electronics hardware design and by taking advantage of the potentialities of programmable logic, even though not state-of-the-art devices, it is possible to substantially upgrade, even within years of distance, the functionality of spectroscopy modules at a very small fraction of the initial investment.

226

N25-248:

A Receiver System for the TileCal Muon Signals

T. Ciodaro Xavier COPPE/UFRJ, Rio de Janeiro - RJ, Brazil On behalf of the ATLAS Tile calorimeter system The TileCal is the hadronic calorimeter of ATLAS, one of the main LHC (Large Hadron Collider) detectors, at CERN. TileCal is divided into three layers of cells, each with two readouts and different sizes and depths, labeled as A, BC and D cells. In addition, TileCal provides two signals for the ATLAS trigger system: the trigger tower and the muon signal. The trigger tower signal is made by summing up the cell readouts within a trigger region, at the TileCal adder cards. Those adder cards also provide the muon signal, which is furnished amplifying one of the two signals from each D cell readouts and is made available as a line driver for the ATLAS level-one muon trigger. In order to increase the signal to noise ratio (SNR) and, consequently, the trigger efficiency, the muon signals from the two readouts of the same D cell may be added, at the level-one receiver. Recently, during the commissioning phase of TileCal, a standalone trigger system was implemented to use the muon signal to trigger cosmic ray. A specific board was designed to add the two muon signals seeing the same D cell. Calibration tests and cosmic runs were performed to evaluate the discrimination gain of this approach. It was shown that the SNR was increased. As a consequence, the ratio between the number of cosmic events detected by the standard TileCals cosmic detection algorithm and the number of triggers given by the system was increased in ~13 %. Following this idea, a muon signal receiver was designed to provide a trigger information for the ATLAS level-one muon trigger system. The system adds the two muon signals from the same D cell and applies an energy threshold to trigger. In addition, the system is able to compensate for any signal loss in the transmission from TileCal, to mask noisy channels and to work synchronously with the experiment. Moreover, the trigger information provided is matched to the one needed by the level-one muon trigger system, to avoid any inconsistency.
N25-249:

SODA: Time Ditrsibution System for the PANDA Experiment

I. V. Konorov, H. Angerer, A. Mann, S. Paul Physic Department, Technical University of Munich, Munich, Germany The PANDA data acquisition system will operate in continuous trigger less mode, collecting about 100 GBytes of data per second from more than a thousand front-end modules. The information from all detectors is combined into data blocks, every block corresponds to one burst or to two microseconds of beam. The SODA (Synchronization Of Data Acquisition) project aims to develop a versatile network system which is, first of all, able to provide a common reference time with a precision better than 20 ps R.M.S. In addition the system synchronizes data taking with the burst structure and performs monitoring of front-end electronics and data acquisition components. Furthermore, it takes responsibility for data flow control and data throttling. The core of the system is a point-to-multipoint bidirectional optical link which is able to broadcast information from a master module to few hundred destinations and to acquire information from the destination modules via a passive optical fiber network. The first prototype system have been built and tested. The architecture, hardware implementation and performance parameters of the SODA system prototype are discussed.
This work is supported by the BMBF, the Maier-Leibnitz-Labor Garching, the Cluster of Excellence Exc153 and the EU Integrated Infrastructure Initiative Hadronphysics Project I3HP (RII3-CT-2004-506078) N25-250:

Associative Memory Design for Fast Tracker at LHC

L. Sartori INFN, Pisa, Italy We propose a new generation of VLSI processor for pattern recognition based on Associative Memory architecture, optimized for on-line track finding in high-energy physics experiments. We describe the architecture, the technology studies and the prototype design of a new Associative Memory: it maximizes the pattern density on ASICs and improves the functionality for the Fast Tracker (FTK) proposed to upgrade the ATLAS trigger at LHC.
N25-251:

A Compact PET Detector Readout Using Charge-to-Time Conversion

M. Streun, H. Larue, C. Parl, K. Ziemons Forschungszentrum Julich, Julich, Germany Position sensitive gamma detectors as used for high resolution PET systems generally contain a very large number of individual photodetector channels. In order to limit space, costs and power consumption the complexity of the readout electronics per channel should be kept low. We introduce a frontend solution based on Pulse Width Modulation that is fast and shows very good linearity. The charge from the detector is converted into a digital pulse length. Instead of analog signals, serial lines or parallel busses only a single digital line leaves each frontend channel. Due to a minimum of components the circuit can be designed very 227

compact and thus is well suited for systems with a large number of individual detectors. A prototype circuit is presently used for the readout of a 16 channel SiPM array.
N25-252: An Advanced FPGA Based Phase-Lock-Loop System as an Alternative Solution for the XFEL Timing System

A. Hidvegi1, P. Gessler2, K. Rehlich2, C. Bohm1 1 Physics Dept. / System and Instrumentation Physics, Stockholm University, Stockholm, Sweden 2 Deutsches Elektronen-Synchrotron (DESY), Hamburg, Germany

The European XFEL project requires a High-speed and very precise clock and timing distribution over large distances. A prototype system which fulfills current requirements that uses high-end components, has just been completed and is being manufactured. [1] However, the system is quite complicated and the boards are very complex, with their small micro-TCA formfactor. Therefore a new idea that could simplify the system, while possibly making it cheaper too, has been worked out. The idea is to implement an advanced PLL into programmable logic in an FPGA, which would control an external VCO. By doing so several major issues could be resolved at the same time, while making more use of modern advanced FPGAs. This work is about such a system, which could be an alternative solution to the current Timing and Triggering System for XFEL. The board is on manufacturing and measurements of clock stability, performance and comparison to the current design will be presented at the conference.
N25-253:

A. Hidvegi1, P. Gessler2, K. Rehlich2, C. Bohm1 1 Physics Dept. / System and Instrumentation Physics, Stockholm University, Stockholm, Sweden 2 Deutsches Elektronen-Synchrotron, Hamburg, Germany

Performance of the Next Timing and Triggering System Prototype for the XFEL Project

The European XFEL project requires an advanced timing and trigger distribution system. The clock must be very clean and stable, so jitter must be cleaned and drifts from all the components must be compensated for. During the project we are developing several prototypes, where we are evaluating different technologies and enhancing the performance of course. This work will present the next prototype system, that is currently being manufactured, and the latest quality measurements will be presented at the conference.
N25-254:

A Fast Hardware Tracker for the ATLAS Trigger System

M. Neubauer University of Illinois at Urbana-Champaign, Urbana, IL, USA On behalf of the ATLAS Collaboration In hadron collider experiments, triggering the detector to store interesting events for offline analysis is a challenge due to the high rates, multiplicities, and energies of particles produced. Maintaining high trigger efficiency for the physics we are most interested in while at the same time suppressing high rate physics from inclusive QCD processes is a difficult but important problem for the experiments. This is especially true for the LHC as the luminosity ramps up from its initial luminosity to its design luminosity of 10^-34 cm^-2 s^-1 and beyond. Hadron collider experiments like ATLAS search for extremely rare processes hidden in much larger background levels. Only a tiny fraction of the produced collisions can be stored on tape and an enormous real-time data reduction is needed. This requires massive computing power to minimize the online execution time of complex algorithms. A multi-level trigger is an effective solution for an otherwise impossible problem. The Fast Tracker (FTK) [1], [2] has been proposed for high quality track finding at very high rates (Level-1 output rates) as an upgrade to the current ATLAS trigger system. The FTK system will use FPGA and ASIC devices in order to complement CPUs. The FTK system addresses the combinatorial challenge with special associative memories, where parallelism is exploited to the maximum level. It simultaneously compares the tracking detector hits to pre-calculated track patterns. The system design is defined and proposed for high-luminosity studies including low-Pt B physics and high-Pt signatures for Level-2 selections: b-jets, tau-jets, and isolated high-pT leptons. We test FTK algorithms using ATLAS full simulation with WH and Hqq events at the LHC design luminosity. The reconstruction quality is evaluated comparing FTK results with the tracking capability of an offline tracking algorithm. We compare different architectures to optimize the latency and hardware system size.
[1] The fast tracker processor for hadronic collider triggers, Annovi, A et al.; ; Nuclear Science, IEEE Transactions on Volume 48, Issue 3, Part 1, June 2001 Page(s):575 - 580 [2] Hadron collider triggers with high-quality tracking at very high event rates, Annovi, A. et al.; Nuclear Science, IEEE Transactions on Volume 51, Issue 3, Part 1, June 2004 Page(s):391 - 400

228

N26: Gamma-Ray Imaging II Wednesday, Oct. 28 N26-1: 13:30-15:30 Grand Ballroom 2

Factors Influencing Time Resolution of Scintillators and Ways to Improve Them

P. R. Lecoq PH, CERN, Geneva, Switzerland The renewal of interest for Time of Flight Positron Emission Tomography (TOF PET), as well as the necessity to precisely tag events in High Energy Physics (HEP) experiments at future colliders, where high luminosity is achieved through high density trains of bunches are pushing for an optimization of all factors affecting the time resolution of the whole acquisition chain: crystal, photodetector, electronics. The time resolution of a scintillator-based detection system is determined by the rate of photoelectrons at the detection threshold, which depends on the time distribution of photons being converted in the photodetector. The possibility to achieve time resolution of about 100ps requires an optimization of the light production in the scintillator, the light transport and its transfer from the scintillator to the photodetector. In order to maximize the light yield, and in particular the density of photons in the first nanosecond, while minimizing the rise time and decay time a particular attention must be given to the energy transfer mechanisms to the activator as well as to the energy transition type at the activator ion. A particular emphasis will be put on the light transport within the crystal and the transfer to the photo-detector. Light being produced isotropically in the scintillator the detector geometry must be optimized to decrease the optical path-length to the photodetector. Moreover light bouncing within the scintillator must be reduced as much as possible. It concerns typically about 70% of the photons generated in currently used scintillators. It will be shown how photonics crystals specifically designed to couple light propagation modes inside and outside the crystal at the limit of the total reflection angle can significantly improve this situation and impact on the time resolution. Examples of production and deposition of photonics crystals on LYSO and LuYAP crystals will be shown as well a first results on light extraction improvement.
N26-2:

V. Orsolini Cencelli1, F. de Notaristefani1, A. Fabbri1, F. Petulla'1, E. D'Abramo1, R. Pani2, M. N. Cinti2, P. Bennati2, P. Boccaccio3, G. Moschini3, N. Lanconelli4, F. Navarria4 1 Sezione di Roma Tre, INFN, Roma, Italy 2 Sezione Roma I, INFN, Roma, Italy 3 Laboratori Nazionali di Legnaro, INFN, Padova, Italy 4 Sezione di Bologna, INFN, Bologna, Italy In recent years a great deal of attention was given to the Hamamatsu Position Sensitive PMTs, due to their suitability for building compact and high resolution gamma cameras. In this framework a great number of solution was proposed for the readout, that vary from the analog readout made with a resistive chain, in some cases including a hardware threshold, to fully digital readouts that read each of the PSPMT anodes individually. In order to obtain the best performace from the PSPMT, a readout that reads each channel individually must be chosen, but a great care must be also taken in designing and tuning the calculation of the point in which the gamma photon has actually interacted with the crystal. This paper describes a 5 cm by 5 cm, gamma camera, based on the Hamamtsu H8500 PSPMT and a continuous LaBr_3:Ce crystal, that makes the readout of each channel individually, up to a sampling frequency of 250K Sample/sec, and uses a new algorithm, based on mean square estimation, to reconstruct the parameters of the light distribution produced by the interaction with the gamma photon. The system has many advantages: it does not suffer from reduction of the useful field of view of the detector, that is observed in the center of gravity based approach due to the truncation of the light distribution, determined by the limited dimension of the scintillating crystal. This characteristic allows the full exploiting of the detector surface (at the price of a slight lowering of the resolution a the very borders of the crystal); moreover it is easy to compensate for the gain variations in the PSPMT anodes and in the analog frontends, avoiding spatial distortion of the gamma image; finally, for very high rate sources, the algorithm is inherently able to discriminate for double hits.

A Gamma Camera with the Useful Field of View Coincident with the Crystal Area

229

N26-3:

C. Fiorini1,2, P. Busca1, A. Gola1,2, R. Peloso1,2, A. Longoni1,2, P. Lechner3, B. Hutton4, P. Van Mullekom5, A. Pedretti6, G. Poli7, G. Lucignani8 1 DEI, Politecnico di Milano, Milano, Italy 2 Sezione di Milano, INFN, Milano, Italy 3 HLL, MPI, Milano, Italy 4 Institute of Nuclear Medicine, UCL, London, UK 5 Nuclear Fields Holland, Vortum Mullem, Netherlands 6 L'ACN, Cerro Maggiore, Italy 7 Ospedali Riuniti di Bergamo, Bergamo, Italy 8 Universita' degli Studi di Milano, Milano, Italy We are developing a new Anger camera for nuclear imaging in the framework of the HICAM (HIgh resolution CAMera) project supported by European Community. The camera is foreseen to be employed in applications where high-position resolution (~ 2.5 mm at an imaging distance of 5 cm) and camera compactness are of primary concern. The camera is based on the use of monolithic arrays of Silicon Drift Detectors (SDDs), each one with a JFET integrated in the detector chip. In this work we present the results of the characterization of a first prototype of the camera (1/4 of the final expected active area), composed by 25 SDDs of 1cm2 each in a 5x5cm2 format. The module is read-out by a single 25-channels ASIC. The prototype of the camera is presented in this work together with the results of the gamma-ray imaging characterization.
N26-4:

First Results of the HICAM Anger Camera

Three Dimensional Imaging of Hidden Objects Using Positron Emission Backscatter

D. Lee, L. C. Stonehill, M. S. Wallace, Q. Looker, M. Cowee, M. Galassi, E. Fenimore, W. V. McNeil Los Alamos National Laboratory, Los Alamos, NM, USA In this paper we present 3-D results from our work on an imaging system designed for interrogation of hidden objects. In situations where potential threat objects are located in places that are impossible to image with traditional radiography, an imaging system that requires access to only one side of the object would be beneficial. The technique uses positron annihilation, which generates back-to-back 511 keV photons. One photon is detected in a position sensitive detector while the other backscatters off the object and returns to another position sensitive detector. Using the positions for each detected photon and the time of flight, the scattering location can be assigned in 3-D. In this paper we will present results on 3-D measurements and image reconstruction from our detector system at Los Alamos National Laboratory.
N26-5:

Interpretation of Pixel-by-Pixel Spectra for Gamma-Ray Imaging Systems

M. A. Blackston, D. E. Hornback, J. A. Mullens, K.-P. Ziock Global Nuclear Security Technologies Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA We are developing analysis techniques to exploit the wealth of information obtained by gamma-ray imaging systems that measure a gamma-ray energy spectrum for each pixel in an image. The combination of spatial (image) and spectral data provides new opportunities for analysis of the structure and composition of unknown radiation emitting objects. While the joint information can be used to locate, identify and determine distributions of radioactive isotopes in a container based on line emissions, there is also information available from the spatial variations in differential absorption of emission lines by material overlying the radioisotopes that are present. In addition, the continuum radiation carries information through single Compton scatter interactions of the line emissions. By combining these types of information we are developing techniques to reconstruct the three-dimensional geometry of both the source and the surrounding materials. Much of the present effort is focused on reconstructing the Compton scattered radiation using the well-defined Compton energy-angle relationship. This is being explored through simulations of idealized imagers and a model of an extant Ge-based coded-aperture imager. The former are used to understand the base problem while the latter provide the complications presented by real-world application and can be compared to data obtained with the physical instrument.
N26-6:

B. R. Kowash1, D. K. Wehe2 1 Dept. of Engineering Physics, Air Force Institute of Technology, Wright Patterson AFB, OH, USA 2 Dept. of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA Rotating modulation collimators (RMC) have traditionally been most sensitive for locating point sources, but have also been used for imaging time dependent extended sources such as solar flares. Current methods for RMC imaging of extended sources have relied upon multiple RMC systems each with their own detectors, electronics, and mask sets. The difference between systems is that each mask pair has a different set of slit spacings so that different spatial frequency components of the source field can be 230

Extended Source Imaging Using a Single Rotating Modulation Collimator

recorded. The data from the ensemble of RMCs is then combined to produce the final extended source image. This research considers a different approach for imaging extended sources with an RMC. This method relies on a single RMC system, and takes advantage of the fact that for terrestrial imaging the separation between masks is an easily changable and free parameter. Changing the separation between the masks effectively represents changing the spatial frequency over which the RMC is most senstive. By measuring multiple data sets with different values of the mask separation parameter, a more complete map of the frequency domain can be recorded. This data is then used with a maximum likelihood expectation maximization algorithm to generate the final image. This method is validated by recording images of two standard extended sources of Co-57. Images of these sources are used to assess how this technique affects the overall resolution of the RMC image. An application is also demonstrated where an extended hot nuisance source is identified using mask separation variation method. The nuisance source is then treated as a background component and removed from the image, leaving behind a much weaker point source that may represent a target of interest. Other images of a rigid ruler source are used to assess the resolution achievable with the RMC operating in this mode.
N26-7:

Experimental Demonstration of 4 Coded Aperture Imaging with 3-D Position-Sensitive CdZnTe Detectors

S. Joshi Kaye, W. R. Kaye, Z. He Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, Michigan, USA 3-D position-sensitive CdZnTe semiconductor detectors have demonstrated 4 Compton imaging capability and excellent energy resolution at room-temperature operation. However, Compton gamma-ray imaging is not feasible at low energies due to the small Compton-scatter cross-section. This work extends the current imaging capabilities to lower energies by utilizing coded aperture masks. Five coded aperture masks are applied a single 20mm x 20mm x 15mm CdZnTe detector, one on each side except for the anode. Near-4 coded aperture imaging has been demonstrated through Monte Carlo simulation. The correct source direction is consistently identified using measured data with a single mask above the cathode side of the detector. Challenges related to depth reconstruction and electric field distortion due to space charge in the detector are discussed. The focus of this research is to image near-4 field of view using coded apertures, ultimately, combining both Compton and coded aperture imaging techniques to expand the range of gamma-ray imaging.
N27: Analog and Digital Circuits III Wednesday, Oct. 28 N27-1: 13:30-15:30 Grand Ballroom 3

Vertically Integrated Circuits at Fermilab

G. W. Deptuch, J. Hoff, A. Shenai, M. Trimpl, R. Yarema, T. Zimmerman PPD/EED, Fermi National Accelerator Laboratory, Batavia, IL, USA Early work by Fermilab with the MIT Lincoln Laboratory 3D process has led to an international consortium for the development of vertically integrated circuits. The consortium has submitted the first multi project wafer run to Tezzaron with over 25 different designs. This paper will examine the opportunities that vertical integration offers by looking at various 3D designs that have been completed by Fermilab for the MPW run.
N27-2:

K. Shimazoe1, H. Takahashi1, T. Fujiwara2, T. Furumiya3, J. Ohi3, Y. Kumazawa3 1 Bioengineering, The University of Tokyo, Tokyo,Bunkyo-ku, Japan 2 Nuclear Engineering and Management, The University of Tokyo, Tokyo,Bunkyo-ku, Japan 3 Shimadzu Corporation, Kyoto, Japan Time Over Threshold (TOT) system has advantage over pulse height measurements on its high integrity and low power dissipation because of its binary readout and circuit simplicity. However the relation between TOT and input charge is strongly nonlinear and dynamic range is limited. We propose a new dynamic TOT system which converts the pulse height to pulse width with a dynamically changing threshold. This kind of TOT system can enable wider dynamic range and improves linearity since the threshold follows the input signal and even shorten the width of TOT pulse. We show the concept of dynamic TOT system and results with discrete circuits. It can improve the dynamic range and theoretically it is possible to desired relation between TOT and input charge by using dedicated threshold function. We also designed and fabricated 48channel dynamic TOT ASIC with 0.25um TSMC CMOS technology.

A New Dynamic Time over Threshold Method

231

N27-3:

A. Pullia1,2, F. Zocca2 1 Dept. of Physics, University of Milano, Milano, Italy 2 INFN-Milano, Milano, Italy

Extending the Dynamic Range of a Charge-Preamplifier Far Beyond Its Saturation Limit: a 0.35m CMOS Preamplifier for Germanium Detectors

The dynamic range of integrated charge-sensitive preamplifiers for germanium detectors is typically limited to 5-10 MeV, owing to the intrinsically low available voltage swing of scaled CMOS technologies and to the very high sensitivity needed to achieve energy resolutions of 0.1-0.2% @ 1-2MeV. We propose a circuit technology which allows for high-resolution energy measurements, compatible with gamma-spectroscopy standards, also in the case of large input signals yielding a deep saturation of the charge preamplifier. The preamplifier has been designed and simulated in 5V 0.35m CMOS technology. The idea relies on the fact that the physical information, i.e. the charge released by the germanium crystal, is not destroyed by the saturation of the charge sensing stage. The exceeding charge, that cannot be stored on the feedback capacitance, is temporarily stored on the other capacitances present at the input node, namely the detector capacitance and the input JFET capacitance. A swift removal of this charge yields a negligible loss of charge through the resistive paths to ground due to the p-n junctions of the detector and of the JFET, and allows for a precise measure of the input signal energy. The CMOS circuit comprises a fast-reset device connected at the input node of the circuit and performing the fast de-saturation of the charge sensing stage. A first-order linear relation exists between the input charge and the reset time. By estimating the input charge through the direct measure of the reset time, a substantial increase of the energy measurement range is achieved.
N27-4:

AIDA : a 16-Channel Amplifier ASIC to Read Out the Advanced Implantation Detector Array for Experiments in Nuclear Decay Spectroscopy

D. Braga, S. Thomas Technology Department, STFC Rutherford Appleton Laboratory, Didcot, UK We have designed a read-out ASIC for nuclear decay spectroscopy as part of the AIDA project - the Advanced Implantation Detector Array. AIDA will be installed in experiments at the Facility for Antiproton and Ion Research in GSI, Darmstadt. The AIDA ASIC will measure the signals when unstable nuclei are implanted into the detector, followed by the much smaller signals when the nuclei subsequently decay. Implant energies can be as high as 20GeV; decay products need to be measured down to 25keV within just a few microseconds of the initial implants. The ASIC uses two amplifiers per detector channel, one covering the 20GeV dynamic range, the other selectable over a 20MeV or 1GeV range. The amplifiers are linked together by diodes and bypass transistors which are normally switched off. The arrival of a large signal causes saturation of the low-energy amplifier and a fluctuation of the input voltage, which forward biases the link diode to the high-energy amplifier. The bypass transistors switch on and the remainder of the charge is integrated by the high-energy amplifier. The signal is shaped and stored by a peak-hold, then read out on a multiplexed output. Control logic resets the amplifiers and bypass circuit, allowing the low-energy amplifier to measure the subsequent decay signal. We present simulations and test results, demonstrating the AIDA ASIC operation over a wide range of input signals.
N27-5:

J.-F. C. Genat1, M. Bogdan1, H. J. Frisch1, H. Grabas2, M. K. Heintz1, S. Meehan1, E. Oberla1, L. L. Ruckman3, F. Tang1, G. S. Varner3 1 Enrico Fermi Institute, University of Chicago, Chicago IL, USA 2 Ecole Superieure d'Electricite, 91192 Gif-sur-Yvette, France 3 University of Hawaii, Honolulu HI, USA Having in view Time of Flight measurements at the pico-second scale, a CMOS fast sampler chip has been developed in 130nm CMOS technology. It includes a 10-20GS/s timing generator lockable on a 40 MHz clock and four channels of 250 sampling cells able to record up to of 25 ns of analog information at 10GS/s. Each sampling cell integrates a comparator allowing a 10-bit analog to digital conversion. Preliminary results are presented in terms of sampling rate, analog bandwidth, dynamic range, linearity, analog to digital conversion and power performance.
N27-6:

A 20 GS/s Sampler Chip in 130nm CMOS Technology

Effect of Wire Resistance on Readout Noise for Large Liquid Argon Time Projection Chambers

S. Rescia, V. Radeka Brookhaven National Laboratory, Upton, NY, USA Large Liquid Argon (LAr) Time Projection Chambers (TPCs) are becoming very attractive for long baseline neutrino and proton decay experiments because of their imaging capabilities and their excellent energy resolution. In the design of large LTPCs, with wire lengths of several meters (~5m, and up to 10-20m for proposed multi-kiloton devices), the noise generated by the resistance 232

of the wire needs to be taken into account. The wire cannot be modeled simply as a capacitance connected to the preamplifier: its resistance, coupled capacitively to adjacent wires constitutes a diffusive transmission line, which injects a noise current into the preamplifier. It will be shown that for most practical cases the noise source can be modeled as an equivalent noise generator corresponding to a resistance of Rwire/3, in series to an impedance equal to the wire capacitance. This simple models allows to use the familiar formulas for the noise analysis of capacitive sources. The wire resistance noise for a stainless steel wire increases the total noise by about 20% for 5m long, 3mm spaced wires at 1 s shaping time, and becomes dominant for wire lengths >10m. It can be reduced by plating the stainless steel wire with high electrical conductivity metals.
N27-7:

FREDA: a Programmable Mixed Signal ASIC for Gas Micro-Strip Detectors Having a Wide Range of Input Capacitance

F. F. Khalid, L. L. Jones, Q. R. Morrissey, M. L. Prydderch, J. Lipp, R. Stephenson Technology, Rutherford Appleton Laboratory, STFC, Didcot, Oxon, United Kingdom The FREDA ASIC is a new mixed signal ASIC for the HOT series gas micro strip detectors. These detectors have 1024 channels each having different capacitance ranging from 2pF 80pF. This new ASIC is an upgrade to the previous programmable version of FREDA with new signal processing capabilities. It is a 16 channel ASIC each of which consists of a Peak Detect and Derandomiser, 8bit counter, 12 bit hybrid SAR ADC with multiplexed LVDS outputs in addition to a preamplifier, a programmable CR-RC shaper and a programmable gain amplifier. The peak detect derandomiser is a 3 bit analogue peak hold pipeline to store randomly arriving pulses at an average data rate of 1MHz. It also allows baseline measurements of the shaper output during low periods to activity to enable correlated double sampling. A 100MHz, 8 bit counter is used to generate a timestamp which is controlled by the peak detect signal. The 12 bit hybrid SAR ADC generates its 4MSBs from a resistor string and the rest 8 bits from a capacitor bank, it uses a 20MHz clock and takes 800ns for a single data conversion. This ASIC has been manufactured using a 0.35m CMOS process. The ASIC architecture is described including simulated and measured results.
N28: Photodetectors and Scintillation Detectors II Wednesday, Oct. 28 N28-1: 13:30-15:30 Grand Ballroom 7

D. Henseler1, R. Grazioso2, N. Zhang2, M. Schmand2 1 Healthcare, Siemens AG, Forchheim, Germany 2 Healthcare, Siemens Medical Solutions, Rockford, TN, USA

SiPM Performance in PET Applications: an Experimental and Theoretical Analysis

Silicon photomultipliers are increasingly being studied for their use in clinical and pre-clinical PET applications, both by industry and academia. Many groups have evaluated the performance of Multi-Pixel Photon Counters (MPPCs) from Hamamatsu Photonics. When coupled to typical PET scintillator crystals, these devices have shown promising results in terms of energy and timing resolution. The purpose of this paper is to analyze the main factors that determine the spectroscopic performance of SiPM based PET detectors and to provide guidelines for further optimization towards the performance levels of state-of-the-art PMT detectors. We present experimental results for the energy and timing resolution for different microcell types of Hamamatsu MPPCs coupled to single LSO crystals. For the 50 m microcell type, we show results for LSO arrays coupled to MPPC arrays with different coupling geometries. To explore the potential and the limitations of SiPM based detectors, we present a statistical signal analysis that links the detector performance to fundamental device characteristics, such as photon detection efficiency, cell density, crosstalk and afterpulsing probability and dark rate. The relative influence of each device parameter on the overall spectroscopic performance is analyzed and discussed. This theoretical analysis is carried out for several optical coupling configurations. The light distribution is modeled with the raytracing program ZEMAX, before applying the statistical model to the remaining signal chain (see Figs 1 and 2). Theoretical estimates will be given for both energy and timing resolution Our analysis concludes with a discussion of the impact of each fundamental device parameter on the spectroscopic and spatial resolution of a simple PET block detector. This way the model helps to predict the benefits of future device optimization efforts and to assign priorities to competing optimization targets.

233

N28-2:

K. Kamada1, T. Yanagida2, J. Kataoka3, A. Yoshikawa2, H. Takahashi4, K. Tsutsumi1, T. Endo1, Y. Usuki1 1 Materials Research Laboratory, Furukawa Co., Ltd., Tsukuba, Japan 2 Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Japan 3 School of Advanced Science and Engineering, Waseda University, Tokyo, Japan 4 School of Engineering, Tokyo University, Tokyo, Japan

Pr:Lu3Al5O12 (LuAG) Scintillator Read-Out Using UV-Enhanced Avalanche Photodiode.

Scintillators coupled with avalanche photodiodes (APD) have been widely used in many applications, such as high energy physics, astrophysics, and positron emission tomography (PET) imaging. Especially Si-APD have advantages such as high quantum efficiency (more than 60% in the wavelength range between 400nm and 800nm), small size, fast time response and stability in magnetic field conditions. Several groups have been developing gamma camera systems using Si-APD arrays to apply to magnetic resonance imaging (MRI)-PET scanner. In these applications, scintillators require such properties as high stopping power, high light yield, fast decay and good energy resolution. Thus, matching between the emission wavelength of scintillators and sensitive wavelengths of APD is also very important. Recently Pr:Lu3Al5O12(Pr:LuAG) attracts attention because of its interesting properties, such as Pr3+ 5d-4f emission at 310nm wavelength, high density (6.7g/cm3), high light output (around 20000 photon/MeV), very fast decay time (20ns) and good energy resolution (4.8%@662keV using PMT). On the other hand, conventional Si-APD show lower sensitivity around 310nm wavelength than that of the visible region. By improving APD sensitivity around 310nm wavelength, the coupling of Pr:LuAG with APD can be more effective for above applications. In the present work, we evaluate gamma-ray responses of Pr:LuAG with PMT, conventional Si-APD and UV-enhanced Si-APD. Furthermore we developed 64 ch. and 144 ch. UV-enhanced position-sensitive Si-APD-arrays.We will report about read-out test using Pr:LuAG scintillator arrays with these APD-arrays in my presentation.
N28-3:

Efficiency of Solid State Photomultipliers in Photon Number Resolution

S. Vinogradov, T. Vinogradova, V. Shubin, D. Shushakov, K. Sitarsky Amplification Technologies, New York, NY, USA Solid State Photomultipliers (SSPM) are widely recognized as new generation of photodetectors competitive with APD and PMT in various applications. SSPM advantages are high gain and ultra-low excess noise factor of internal amplification resulting in ability to detect single photons, high photon detection efficiency, fast response and good time resolution. SSPM drawbacks are high dark count rate, high probability of cross-talk and afterpulsing, and low dynamic range. For example, cross-talk and afterpulsing with total probability of 20% may result in elimination of SSPM advantage in excess noise factor in comparison with PMT. Many applications in Nuclear Science and Medical Imaging based on photodetector with scintillator crystal are very sensitive to its energy resolution, which is represented with respect to photodetector itself by pulse height resolution or so-called Photon Number Resolution (PNR). We will present characterization of SSPM PNR in light pulse detection taking into account nonlinearity of photoresponse due to limited number of pixels and finite reset time as well as excess noise factor of cross-talk and afterpulsing. On our opinion the dependence of PNR on number of incident photons in pulse and pulse duration time is very powerful tool for evaluation of SSPM applicability and competitiveness. Normalization of PNR relatively to ideal detector represents characterization of the detector performance in terms of total excess noise factor of photodetection or, inversely, in terms of total efficiency of photodetection. Examples of such characterization will be presented and discussed.
N28-4:

C. Piemonte1, M. Melchiorri1, A. Piazza1, A. Tarolli1, N. Zorzi1, V. Schulz2, T. Solf2, P. Fischer3 1 FBK, Trento, Italy 2 Philips Research, Aachen, Germany 3 University of Heidelberg, Heidelberg, Germany We report on the production experience and the characteristics of silicon photomultipliers (SiPMs) fabricated at FBK to be used to fully equip a preclinical positron emission tomography (PET) system. More than 700 fully working, 2x2 monolithic arrays of 4x4mm2 SiPMs have been produced. A test procedure, based on forward and reverse IV measurements, has been implemented to extract the basic properties and, finally, to select the devices at the wafer level. Methodology, results from the on-wafer tests and functional performance are shown. Besides this production, test SiPMs featuring microcells with different designs have been fabricated to find the configuration which optimizes the timing and energy resolution performance of the sensor coupled with the scintillator. Tests on these structures are ongoing and the results will be shown at the conference.

Production of Large Area Silicon Photomultipliers for a PET/MR Scanner

234

N28-5:

The Digital Silicon Photomultiplier - Principle of Operation and Intrinsic Detector Performance

T. Frach, G. Prescher, C. Degenhardt, R. de Gruyter, A. Schmitz, R. Ballizany Philips Corporate Technologies, Aachen, Germany Recently, the Silicon Photomultiplier (SiPM) gained interest as a potential candidate to replace Photomultiplier Tubes for reasons of ruggedness, compactness or insensitivity to magnetic fields. Other advantages of solid state detectors are their low operating voltage, low power consumption and large scale fabrication possibilities. Today, those SiPMs operate in an analog manner, connecting the individual Geiger-mode cells of the SiPM in parallel resulting in an analog output signal. This limits the performance of the analog SiPM due to parasitic capacitances and inductances, the influence of electronic noise and sensitivity to temperature drifts. We developed a digital SiPM (dSiPM) of 3.8mm x 3.3mm in size with readout electronics integrated next to each of the 8188 Geiger-mode cells to allow for an early digitization of the state of each cell. In addition, a low skew trigger network, the trigger logic and a time-to-digital converter are integrated on the sensor chip. In this talk, we describe the sensors principle of operation and show results on important sensor characteristics like dark count behavior, photo detection efficiency and timing resolution. The integrated electronics allow to switch off faulty cells which generate large numbers of dark counts. The dark count rate of the sensor can be significantly reduced by switching off less than 10% of all cells. The overall photo detection efficiency of the sensor, including the fill-factor, amounts to a maximum of 30% at a wavelength of 420nm and 3.3V excess voltage. The intrinsic timing resolution of the complete sensor, as determined with a picosecond laser reference, is 22ps FWHM. The results show that the digital SiPM presented here overcomes major drawbacks of analog SiPMs like high dark count rates and low yields due to faulty cells. Its integrated digital electronics enables the detection of single optical photons with very high timing accuracy.
N28-6:

S. Russo1, G. Barbarino1, R. de Asmundis2, G. De Rosa2 1 Universita' Federico II di Napoli, Naples, Italy 2 INFN sez. di Napoli, Naples, Italy

The Vacuum Silicon Photomultiplier Tube (VSiPMT): a New Concept of Photon Detector. First Feasibility Results.

The future astroparticle experiments will study both energetic phenomena and extremely rare events from astrophysical sources. Since most of these families of experiments are carried out by using scintillation phenomena, Cherenkov or fluorescence radiation, the development of photosensitive detectors seems to be the right way to increase the experimental sensitivity. We therefore propose an innovative design for a modern, high gain, silicon-based Vacuum Silicon Photomultiplier Tube (VSiPMT), which combines three fully established and well-understood technologies: the manufacture of hemispherical vacuum tubes with the possibility of very large active areas, the photocathode glass deposition and the recent Geiger-mode avalanche silicon photodiode (G-APD) for which a mass production is today available. This new design, based on G-APD as the electron multiplier, allows overcoming the limits of the classical PMT dynode chain. In this work VSiPMT first feasibility results will be presented.
N28-7:

M. Binda1, C. Fiorini1,2, D. Natali1, R. Peloso1,2, M. Sampietro1, L. Beverina3, G. Pagani3 1 DEI, Politecnico di Milano, Milano, Italy 2 Sezione di Milano, INFN, Milano, Italy 3 Dipartimento di Scienza dei Materiali, Universita' degli Studi Milano Bicocca, Milano, Italy

Squaraine-Based Organic Photodetector Coupled to a CsI(Tl) Scintillator for Gamma-Ray Detection

Low-cost technologies for the production of large area sensing devices could offer advantages in applications like medical imaging and homeland security. Organic materials might be good candidates in this direction. Aim of this work is the development of a gamma ray detector based on an organic photodetector coupled to a CsI scintillator. The active material of the developed photodetector is a blend of squaraine dyes and Phenyl-C61-Butyric-Acid-Methyl-Ester (PCBM). The architecture of this first prototype of photodetector is vertical, with a resulting detecting area of about 1.5mm x 1.5mm, and the device is characterized by an intrinsic capacitance of about 30pF and with a peak of the quantum efficiency of about 15% for wavelengths around 570nm. The current signal from the organic photodetector is fed to a low-noise transimpedance amplifier followed by a shaping filter and the Equivalent Noise Charge has been evaluated in several conditions as a function of the shaping time. The results of first tests using the photodetector-CSI(Tl) combination are here presented.

235

N29: New Detector Concepts and Instrumentation III Wednesday, Oct. 28 N29-1: 16:00-18:00 Grand Ballroom 1

Active Coded Aperture Neutron Imaging

P. A. Marleau, J. Brennan, E. Brubaker, J. Steele Radiation and Nuclear Detection Systems, Sandia National Laboratories, Livermore, CA, USA Because of their penetrating power, energetic neutrons and gamma rays offer the possibility of detecting shielded or distant SNM. Of these, fast neutrons offer the biggest advantage due to their very low occurrence in natural background. We are investigating a wholly new approach to fast-neutron imagingan active coded-aperture system that uses a coding mask made of neutron detectors. The only previously demonstrated method for long-range neutron imaging is double-scatter imaging. Active coded-aperture neutron imaging should be more efficient for improved detection speed, range, and sensitivity. We will describe our detector including design considerations and present initial results from a lab prototype.
N29-2:

N. Sullivan1, D. Beaulieu1, D. Gorelikov1, H. Klotzsch1, P. de Rouffignac1, K. Saadatmand1, K. Stenton1, A. Tremsin1, J. Legere2, J. Ryan2 1 arradiance inc, sudbury, ma, usa 2 Space Science Center, University of New Hampshire, Durham, NH, USA Prototype results, from a novel fast neutron detector technology for direct, low noise, high efficiency counting at sub-10ns timing resolution are presented. Neutron detection is accomplished through the conversion of incoming fast neutrons into a pulse of electrons, using the neutron-proton (n-p) recoil reaction within hydrogen-rich microchannel structures. The proton excites secondary electrons, which are amplified by a factor of 1000000. High gain allows event counting at very low input fluxes with very low background noise due to the absence of heavy metals and radioactive traces (e.g. Rb and K) within the plastic substrates. The prototype devices will be configured as microchannel plates (MCP) whose output is readily registered by a large variety of available technologies. The core of this technology is the thin film materials and low temperature processes used to functionalize hydrogen-rich plastic microchannel structures with conductive and high secondary electron (SE) emissive films enabling high gain detectors. The timing resolution of this fast neutron detection technology, <10ns, is limited only by the depth of neutron absorption within the MCP, better than 2ns. 1% detection efficiency for fast neutrons can be achieved since the probability of neutron recoil in the plastic selected for study exceeds this value for useable thicknesses. Successful passive SNM detection requires high efficiency in the presence of large background gamma radiation. The intrinsic sensitivity of the Arradiance plastic MCPs to gamma photons, expected to be below the 10-4 level. More complex configurations with multiple detector orientations can be used to improve gamma rejection and specify neutron source location. Demonstration of the expected high level of performance of the proposed detector may allow future development of a large area, advanced detector configurations having the ability to determine neutron energy, directionality, and SNM source location.
N29-3:

Novel Fast Neutron Counting Technology for Efficient Detection of Special Nuclear Materials

Nano-Fission Detector Program at LANL

E. Esch, S. Stange, F. Tovesson, R. Muenchausen, R. Del Sesto, F. Taw LANL, Los Alamos, NM, USA Experiments in accelerator physics today have ever-growing demands on radiation detector hardware. With larger applications and less budget for experiments, it is important for radiation detectors to be ideally suited for the task and to be cost-efficient. This paper outlines a detector development project currently carried out at Los Alamos National Laboratory. The project involves the development of a neutron fission detector to monitor neutron flux at beam lines or reactors. Neutron flux monitors are commonly used in a variety of nuclear physics applications. In nuclear reaction measurements, for example, the neutron flux is needed in order to determine the reaction cross section. Recent research on nanocomposite based -scintillator development, pioneered at Los Alamos, indicates that this approach can be extended to load fissionable nanomaterial into a matrix of scintillator material, with up to three orders of magnitude higher fissionable material loading than typical fission chambers. This will result in a rugged, cost-efficient detector with high efficiency, short signal rise time, and the ability to be used in low neutron-flux environments.

236

N29-4:

J. W. McClory1, B. D. Blasy1, D. Schultz1, J. C. Petrosky1, J. Tang2, Z. Wang2, J. I. Brand3, P. A. Dowben4 1 Engineering Physics, Air Force Institute of Technology, WPAFB, OH, USA 2 Physics and Astronomy, University of Wyoming-Laramie, Laramie, WY, USA 3 Center for Materials and Nanoscience, University of Nebraska-Lincoln, Lincoln, NE, USA 4 Physics and Astronomy, University of Nebraska-Lincoln, Lincoln, NE, USA A new solid-state material and structure for neutron detection is demonstrated. The electrical measurements show pronounced rectification properties for heavily doped Gd:HfO2 films on n-type Si(100), consistent with p-type behavior. With neutron irradiation, there is an increase in the reverse bias current and a pulse height spectra that scales with the neutron flux.
N29-5:

Neutron Detection Using Gd-Doped HfO2/Silicon Heterojunctions

K. D. Ianakiev1, J. R. Lashley1, M. R. Swinhoe1, R. M. Flemming2, A. Armstrong2, P. B. Littlewood3, L. L. Smith1, C. P. Opeil4 1 Niclear Nonprolifeartion, Los Alamos National Laboratory, Los Alamos , NM, USA 2 Semiconductor Materials & Device Sciences Department, Sandia National Laboratory, Albuquerque, NM, USA 3 Department of Physics, Cambridge Univercity, Cambridge, UK 4 Department of Physics, Boston College, Boston, USA Thermal neutron detectors based on semiconductor bulk material in which the neutron capturing material is a constituent of the crystal lattice, have unique radiation detection properties. In this case the bulk solid serves as both the neutron capturing material and provides charge transport, similar to the mechanism in 3He detectors. We are studying the feasibility of preparing bulk, solidstate thermal neutron detectors utilizing Li-containing single crystals such as LiTaO3. These crystals are commercially available in sizes up to 100 mm diameter and 1 mm thick wafers that are widely used in nonlinear optics applications. Because of the monocrystal size and nearly 20% percent lithium content, this material has the potential to produce single detectors with efficiency comparable to 3He detectors. Recent progress in photoconductivity measurements through the bulk material, deep level trap assessment and pathways for improvement of charge transport properties will be discussed.
N29-6:

Lithium Tantalate Crystals for a Solid State Thermal Neutron Detectors

V. D. Ryzhikov1, G. M. Onishchenko1, L. A. Piven1, B. V. Grinyov1, S. M. Galkin1, E. F. Voronkin1, O. K. Lysetska1, L. L. Nagornay1, T. Pochet2, C. F. Smith3 1 Institute of Scintillation Materials of STC Institute for Single Crystals, NAS of Ukraine, Kharkov, Ukraine 2 2International Atomic Energy Agency, Vienna, Austria 3 3Lawrence Livermore National Laboratory, Livermore, USA Needed now are new neutron detectors which can keep up with the crystal progress and which are small-dimensioned. Especially promising is creation of a small-sized neutron detector for portable radioactive materials detection systems using the scintillator avalanche photodiode (Sc-APD) and scintillator- mikropixel avalanche photodiodes (Sc-MAPD) . An APD combines the advantages of a PIN photodiode (small, insensitivity to magnetic fields) and those of a photomultiplier (gain and speed). The gain of APDs can be 1000 and more. APDs have moderate internal gain but still need low noise amplifiers. We have developed the low noise preamplifier. We have studied scintillator crystals CWO, GSO, ZnSe(Te,O) and CsJ(Tl), used in systems for detection of radioactive fissionable substances (RFS). Our studies have shown, that detectors based on Sc-APD (APDs were obtained from Hamamatsu) work in spectrometric mode and offer high detection efficiency throughout the visible (blue-red) part of the spectrum. These sensors have the best spectral matching with scintillator crystals: CWO, GSO, CsJ (Tl) and bias voltage of approximately 400 V. On the contrary, the detector based on the Sc-MAPD (MAPDs were obtained from Institute of the Nuclear Studies, Moscow) is a kind of photon counting devise made up of multiple APD pixels operating in counting mode. They are matching with scintillator crystal ZnSe(Te) in the red part of the spectrum and have bias voltage of approximately 50 V. Unusually high detection efficiency of fast neutrons by inorganic scintillators can be explained by the detection mechanism based on the inelastic scattering reaction (n, n) on nuclei of the scintillator [1-3]. The most evident field for this method is inspection systems for prevention of illegal transportation of RFS .
Acknowledgments. The research described in this publication was supported in part by STCU project #4115 and NATO Science for Peace project SfP-982823. 1. B.V.Grynyov, V.R.Lyubinskiy, V.G.Senchishin, V.N.Lebedev, V.V.Nekrasov, L.A.Piven, Yu.A.Borodenko. Radiometric system for detection of gamma- and neutron radiation. Patent of Ukraine 80455, G01T1/00, 3/00. Publ. 25.09.07, Biul. No.15. 2. V.D.Ryzhikov,G.M.Onhishehko,L.A.Piven,B.V.Grinyov,C.F.Smit,T.Pochet, L.L.Nagorna. Detectors for high efficiency detection of fast neutrons using inelastic scattering. Book of Abstracts (SORMA West 2008), The International Symposium on Radiation Measurements and Applications (Berkeley, California, USA, June 2-5, 2008). 3. Measurement and simulation of the neutron response and detection efficiency of a Pb-scintillating fiber calorimeter. M. Anelli, G. Battistoni, S. Bertolucci at al. NIM, A580 (2007) 368-372.

New Neutron Detectors Based on Inorganic Scintillators Using Inelastic Scattering

237

N29-7:

Gd-Bearing Composite Scintillators as the New Thermal Neutron Detectors

N. Z. Galunov, B. V. Grinyov, N. L. Karavaeva, Y. V. Gerasymov, O. T. Sidletskiy, O. A. Tarasenko Institute for Scintillation Materials, National Ac.Science of Ukraine, Kharkov, Ukraine The technology that was designed for organic composite scintillators as detectors of fast neutrons and alpha particles is applied to develop new composite inorganic detector of thermal neutrons. We use cerium-doped gadolinium silicate Ce:Gd2SiO5 and gadolinium pyrosilicate Ce:Gd2Si2O7 scintillation single crystals. To obtain composite scintillator we mechanically ground a single crystal. Then, we sieved the grains through calibrated sieves and chose the grains with necessary average size. These grains were introduced in non-luminescent silicone material Silgard-527. The composition was placed onto an optically transparent non-luminescent organic glass plate. Ce:GSO and Ce:GPS crystals were grown by Czochralski method. A 239Pu-Be source of fast neutrons with neutron flux 110^5 neutrons per second that was inside the calibrated paraffin sphere, which gives 9% thermal neutrons in 4-geometry. We have studied the samples of Ce:GSO and Ce:GPS single crystals prepared in a form of square plates 0.5 mm thick. The samples have 100 mm^2 of area of input window. The single-layer and multi-layer composite scintillators on the base of Ce:Gd2SiO5 and Ce:Gd2Si2O7 grains were studied as well. The size of grains was varied from 0.5 mm down to 0.04 mm. The results of the measurements obtained for single crystals and composite scintillators are compared and discussed. The composite Gd-bearing detectors are effective detectors of thermal neutrons and have no technological limitation on the area of the input window.
N30: Accelerators and Beam Line Instrumentation Wednesday, Oct. 28 N30-1: 16:00-18:00 Grand Ballroom 2

D-D Nuclear Fusion Using Different Sized Pyroelectric Crystals

A. M. Kovanen, D. J. Gillich, Y. Danon Mechanical, Aerospace, and Nuclear Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA Theory predicts electrons play a detrimental role in creating the conditions for D-D fusion in pyroelectric crystal accelerators. Three different pyroelectric crystal sizes were paired in six different configurations to determine if experimental results agree with theory. An experimental configuration using a 30 mm diameter tip crystal opposite a 20 mm diameter target crystal yielded better neutron results than a 20 mm diameter tip crystal versus a 30 mm diameter target crystal as predicted by theory. Two 20 mm diameter by 20 mm thick lithium tantalate crystals yielded ~1 x 10^4 neutrons per thermal cycle.
N30-2:

F. Wulf1, M. Koerfer2, H.-J. Grabosch3, W. Goettmann1 1 Div. of Experimentsysteme, Helmholtz-Zentrum Berlin fuer Materialien und Energie, Berlin, Germany 2 Div. of MEA, DESY, Hamburg, Germany 3 Div. of ZEU-LTG, DESY, Zeuthen, Germany Beam losses and beam profiles at particle accelerators can be determined by measuring the ionizing radiation outside the vacuum chamber. The next generations of free electron lasers require novel solutions for beam loss monitor and radiation detection systems. A new concept at the Free-Electron-Laser Hamburg (FLASH) and Photo Injector Test Facility Zeuthen (PITZ) was developed and put into operation. The concept based on optical fibers used as radiation detection sensor. Slow beam loss monitors determine the total ionization dose at selected positions or along the beam line and undulator sections in a response time of a few milliseconds by measuring the radiation-induced attenuation of the fiber. Fast beam loss monitors detect the Cerenkov light generated by relativistic electrons penetrating radiation hard fiber. The new beam loss position and beam profile monitor systems utilize this effect. The time response is below milliseconds with a time resolution of some nanoseconds. These monitors provide a technique to improve the beam performance and trace the stability of the over all setting parameters and the accelerator parameter drift. This paper presents the new diagnostic methods and the operation experience of the systems with a focus on the fast beam loss monitors.
N30-3:

Beam Loss Monitors for FEL Using Optical Fiber

S. D. Clarke1, S. A. Pozzi1, N. Cunningham2, S. Banerjee2, D. Umstadter2 1 Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA 2 Physics and Astronomy, University of Nebraska, Lincoln, NE, USA

Monte Carlo Characterization of a Pulsed Laser-Wakefield Driven Monochromatic X-Ray Source

The Diocles laser facility at The University of Nebraska-Lincoln (UNL) is a 100-TW, 30-fs pulsed Ti:sapphire laser system. Diocles is routinely used to accelerate electron beams to high energy (up to 800 MeV), high beam charge (~0.5 nC) and narrow energy width (~10%). These electrons may be used to scatter laser light producing monochromatic X-rays with energy and 238

spectral width that are directly related to those of the electron beam. The pulsed and directional nature of this source makes it impossible to employ traditional spectroscopy techniques for characterizing the X-ray beam. The large number of X-rays arriving at the detector in a short period of time would result in a high degree of pulse pileup making the individual X-ray energies indistinguishable. In the technique presented in this paper, a series of attenuator materials will be placed in the path of the X-rays and the integral detector response will be measured. A series of detector response measurements with different attenuators will allow determination of the incident X-ray energy by using unfolding techniques.
N30-4:

S. Friedrich1, O. B. Drury1, M. A. McKernan1, C. S. Gardner1, E. Ables1, K. W. Fong1, G. J. Yong2, R. M. Kolagani2, R. M. Bionta1 1 Applied Physics Division, Lawrence Livermore National Laboratory, Livermore, CA, USA 2 Department of Physics, Geosciences and Astronomy, Towson University, Towson, MD, USA We have built a novel cryogenic bolometer to measure the total energy of the Linac Coherent Light Source (LCLS) free electron X-ray laser that is currently being built at the Stanford Linear Accelerator Center. The bolometer consists of a 375 m thick Si absorber and a Nd0.67Sr0.33MnO3 sensor operated in its metal-insulator transition at ~180 K. As a thermal sensor, its signal is calibratable with optical lasers and is expected to contain fewer unknowns than scintillator or gas detectors in response to the ultra-short (~200 fs) ultra-bright (10^12 to 10^13 photons) X-ray pulses (0.8 to 8 keV) at LCLS. The bolometer is designed to meet the conflicting requirements of radiation hardness, sensitivity, linearity, and readout speed compatible with the LCLS pulse rate. Calibration runs with an optical calibration laser demonstrate that it meets the requirements of a signal-to-noise ratio >100 and of linearity over a dynamic range of three orders of magnitude. With LCLS just having achieved first light, we expect to present the first data on the bolometer response to the LCLS pulsed X-ray laser.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344. N30-5:

First Total Energy Measurements at the LCLS Free Electron X-Ray Laser with a Cryogenic Manganite Bolometer

Baseline Design of the Hall D Polarized Photon Beam

J. Stewart Physics, BNL, Upton, NY, USA On behalf of the GlueX Collaboration The baseline design for the JLAB Hall-D polarized photon beam will be presented. As part of the Jefferson Lab 12 GeV upgrade a new photon beam facility based on coherent bremsstrahlung will be built. The 12 GeV electron beam will pass through a 20 micron thick single crystal diamond generating photons via both incoherent and coherent bremsstrahlung. A 1.5 T 6.3 m long normal conducting dipole magnet will momentum analyze the electrons which emitted the photons. Two detector packages consisting of a scintillator PMT hodoscope and a scintillating fiber array with SiPM readout are used to detect the electrons. The photon beam generated in the crystal is a combination of coherent and incoherent bremsstrahlung, however the coherent bremsstrahlung photons have a different angular momentum relation than the incoherent bremsstrahlung photons. By collimating the photon beam after a 76 m drift distance the incoherent component can be suppressed relative to the coherent resulting in a high quality beam. The projected beam characteristics will be presented.
N30-6:

W. J. McNeil1, S. L. Bellinger1, B. J. Blalock2, C. L. Britton2, W. L. Dunn1, C. M. Henderson1, T. J. Sobering3, D. S. McGregor1 1 Dept. of Mechanical and Nuclear Engineering, Kansas State University, S.M.A.R.T. Lab, Manhattan, Kansas, USA 2 Dept. of Electrical and Computer Engineering, University of Tennessee, Knoxville, USA 3 Kansas State University, Electronic Design Laboratory, Manhattan, Kansas, USA A 1024-channel pixel array has been constructed utilizing the perforated diode neutron detector design currently produced at Kansas State University. A single pixel diode was laid out around a single trench 4 cm long and 30 microns wide and 100 microns deep. The trench is filled with 6LiF powder to provide conversion of neutrons to energetic charged particles. A pitch of 100 microns between pixels has been achieved and less than 120 micron spatial resolution has been demonstrated experimentally with a smaller, 32-channel prototype in previous work. The 1024-channel array was produced by tiling 16 chips side-by-side, each containing 64 pixels. Signal processing is handled by 16 ASICs of 64 channels each, designed at University of Tennessee. Board assembly and digital communications to PC were handled by the KSU Electronics Design Laboratory utilizing a PCI card developed at ORNL. Initial tests of this system at the KSU TRIGA reactor will be reported as well as functional test results from the SNS at ORNL.

1024-Channel Solid State 1-D Pixel Array for Small Angle Neutron Scattering

239

N30-7:

The ATLAS Beam Diagnostic Systems

M. Mikuz Univ. Ljubljana / Jozef Stefan Institute, Ljubljana, Slovenia On behalf of the ATLAS Beam Conditions Monitor The beam diagnostic system of the ATLAS detector comprises two diamond sensor based devices. The innovative Beam Conditions Monitor (BCM) is aimed at resolving background from collision particles by sub-ns time-of-flight measurement. The Beam Loss Monitor (BLM) is a clone of the LHC machine BLM system, replacing ionization chambers with diamond sensors. BCM uses 16 1x1 cm2 0.5 mm thick polycrystalline chemical vapor deposition (pCVD) diamond sensors arranged in 8 positions at a radius r 55 mm, ~1.9 m up- and down-stream the interaction point. Time difference measurements with 2.56 GHz sampling are performed to distinguish between collision and shower particles from beam incidents. A FPGA-based readout system performs real-time data analysis and interfaces the results to ATLAS and the LHC beam permit system. The diamond sensors, the detector modules and their readout system are described. Results of commissioning and timing measurements with cosmic muons in preparation for first LHC collisions are summarized. BLM utilizes 12 pCVD diamond sensors close to the beam pipe at z +/- 3.5 m. The radiation induced currents in the sensors are read by the LHC machine developed readout, averaging the current over various time constants from 40 s to 84 s. Exceeding a preset threshold for any of the readings drops the beam permit and aborts the LHC beam. Both systems provide post-mortem data dump of approximately 1000 LHC turns prior to the anomalous condition, allowing to diagnose its development and refine the BCM algorithms and BLM thresholds. The systems will be ready watching for the first LHC collisions in October and their performance with LHC beams will be reported at the conference.
N31: Semiconductor Detectors III: CZT Detectors Wednesday, Oct. 28 N31-1: 16:00-18:00 Grand Ballroom 3

Performance of Cadmium Zinc Telluride Pixel Detectors Developed for the NuSTAR Mission

F. A. Harrison, W. R. Cook, H. Miyasaka, R. McLean, V. Rana, V. Bhalereo Space Radiation Laboratory, Caltech, Pasadena, CA, USA We have developed a Cadmium Zinc Telluride Pixel detector optimized for use as a focal plane detector for astrophysical hard Xray imaging telescopes. The hybrid detector incorporates a low-noise, low-power custom ASIC developed at Caltech for the HEFT balloon program, and recently improved for homeland security and the NuSTAR space mission. Coupled to a CdZnTe sensor through a low-capacitance interconnect, we achieve spectral resolution of 1~keV FWHM at 60~keV with near-unity quantum efficiency in a device with 0.6mm pixels. In addition, the readout enables depth of interaction determination. In this presentation we discuss the detector architecture, spectral, depth determination and imaging performance.
N31-2:

Q. Li1, A. I. Garson1, M. Beilicke1, K. Lee1, P. Dowkontt1, J. Martin1, I. Jung2, M. Groza3, A. Burger3, G. D. Geronimo4, H. Krawczynski1 1 Department of Physics ,, Washington University in St. Louis, Saint Louis, MO, usa 2 Physik. Inst.,, Universitt Erlangen-Nrnberg, Erlangen, Germany 3 Department of Physics, Fisk University, Nashville, TN, usa 4 Instrumentation Division, Brookhaven National Laboratory, Upton, NY, usa

Systematic Study of the Energy Resolution and Detection Efficiency of Thick CZT Detectors as a Function of Substrate Thickness and Pixel Pitch

The techniques of crystal growth, the design of the detectors, and the electronics used for reading out the room temperature semiconductor CdZnTe (CZT) detectors have been improved considerably over the last few years. In 2008, we achieved for the first time <1% energy resolutions with detectors made from modified horizontal Bridgeman CZT grown by the company Orbotech Medical Solutions Ltd. In this contribution, we report on the systematic test of Orbotech CZT detectors with different thicknesses (0.5 cm, 0.75 cm, 1 cm) contacted with pixels at different pixel pitches (0.35mm, 0.6 mm, 1.3 mm, 1.6mm, 2.5 mm). The detectors were fabricated in the Washington University class-100 clean room using photolithography. The readout system developed at Washington University is based on the NCI-ASIC developed by G. De Geronimo at Brookhaven National Laboratory. In the contribution we discuss the dependence of the energy resolution and detection efficiency of the detectors as a function of (i) the detector thickness, and (ii) the pixel-pitch-to-detector-thickness aspect ratio.

240

N31-3:

Performance of 3-D Position Sensitive CdZnTe Detectors at Gammay-Ray Energies Higher than 1.0

MeV
F. Zhang, Z. He, W. R. Kaye The University of Michigan, Ann Arbor, MI, USA A 3-D position-sensitive CdZnTe gamma-ray detector employs a pixellated anode (11x11 pixels) and a planar cathode on a 15x15x10 mm3 or 20x20x15 mm3 CdZnTe single crystal. The signals from the 121 pixels and the cathode read out by Application Specific Integrated Circuits (ASIC) are used to reconstruct the energy depositions and 3-D coordinates of multiple gamma-ray interactions inside the detector. The best energy resolution of a given detector can be achieved by using the 3-D position sensitivity to correct the signal variations due to charge trapping and material non-uniformity. We have demonstrated excellent energy resolution of better than 1% FWHM at 662 keV in a dozen CdZnTe detectors ranging from 2.25 cm3 to 6 cm3 in size. In this study, the energy resolution and detection efficiency of several 3-D position sensitive CdZnTe detectors at gammaray energies higher than 1 MeV (up to 2.6 MeV of Th-228) will be analyzed by comparing the experimental results with GEANT4 simulation results. Possible reasons of energy resolution degradation and efficiency loss at high gamma-ray energies are discussed.
N31-4:

G. S. Camarda1, A. E. Bolotnikov1, Y. Cui1, A. Hossain1, K. Kim1, R. Gul1, G. Yang1, S. Awadalla2, J. McKenzie2, H. Chen2, R. B. James1 1 Brookhaven National Lab, Upton, NY, USA 2 Redlen Technologies, Sidney, BC, Canada Large-volume CdZnTe (CZT) detectors are readily available for radiation detection applications. In this study we measured their performance and characterized the detector material with different methods. We employed synchrotron x-ray radiation to understand the origin of the extended defects present in a number of detectors. Results on the limiting factors of the extended defects on detector performance are presented, with particular emphasis devoted to large-volume pixellated CZT detectors designed with a charge-loss correction capability.

CdZnTe Material and Detectors Characterization Using Synchrotron Radiation

Investigation of Polarisation Phenomena in CdZnTe Materials to Be Used as X-Ray Radiation Detectors by Direct Implementation of the Electric Field Distribution Profiles at Low Temperature and under in Situ X-Ray Irradiation.
N31-5:

G. Prekas, P. Veeramani, A. Lohstroh, P. J. Sellin Physics, CNRP, UNIVERSITY OF SURREY, Guildford, UK Many applications such as homeland security, medical imaging and newly developed monitoring techniques in research and industry require the operation of the CdZnTe solid state detectors under high fluxes of X-rays, and at low temperatures [1,3]. Under these conditions the spectroscopic response of the CdZnTe devices often shows severe degradation termed as polarisation that affects their stability [4,5]. By real time and direct observation of the electric field distribution, using the Pockels electrooptic effect [6,8], we have found that the polarisation is mainly attributed to the local or extended degradation of the electric field, which is mostly apparent at temperatures below 250K, as well as at high levels of X-ray irradiation at room temperature. In this paper we extend our previous Pockels studies by calculating estimates of space charge and potential distribution profiles from our measured electric field profiles as a function of bias, temperature and X-ray irradiation. In our first set of experimental results we have studied CZT materials grown by Yinnel which showed average charge transport properties, for a typical crystal we have measured ee=(4.4+/-0.1)10-3cm2V-1 and hh=(2.2+/-0.2)10-4cm2V-1. In order to investigate the polarisation phenomena further, we have extended this study to high quality Redlen CZT material. In this way we will confirm whether the polarisation phenomena solely appear in Yinnel materials, related to the material quality or, if it is a general rule for all CdZnTe crystals under low temperature and high X-ray flux conditions. We will present a comparison of measured electric field profiles and polarisation phenomena in CZT crystals supplied by different vendors, and identify possible links between the observed space charge accumulation and the known charge transport and deep trap concentrations in the materials.
References: [1] R. James et al., Semiconductors for Room Temperature Nuclear Detection Applications, Academic Press, New York, 1995, pp. 384. [2] T.E Schlesinger Et al., 2001 Mater. Sci. Eng. 32 103. [3] Cs. Szeles et al., IEEE Trans. Nucl. Sci., vol. 54, no. 4, pp. 13501358, August 2007. [4] B.W.Sturm et al., IEEE Transactions on Nuclear Science, Vol. 52, No. 5 (October 2005). [5] W. Murray, K. Krueger, M. W. Rawool-Sullivan, L. Ussery, C. Whitley, Temperature effects on CdZnTe detector performance,1998 IEEE NSS conference paper record. [6] P. De Antonis et al., IEEE Transactions on Nuclear Science, Vol. 43, No. 3, June 1996. [7] H.W.Yao et al., Mater. Res. Soc. Symp. Proc. 487 (1998), pp51.

241

[8] A. Cola at al., Nuclear Instruments and Methods in Physics Research A 568 (2006), pp 406-411. This Work was Funded by the RPSRC research grand 'HEXITEC'. N31-6:

Effect of Crystal Length on Frisch Collar Device Performance

A. Kargar Mechanical and Nuclear Engineering Department, Kansas State University, Manhattan, KS, USA On behalf of the S.M.A.R.T. Laboratory, Kansas State University In this study, the impact of crystal length on CdZnTe Frisch collar detectors spectroscopic performance are investigated. Four different Frisch collar devices were designed to have identical aspect ratios of 1.8 for simulation purposes, while three devices were actually fabricated and tested. Pulse height spectra were acquired from all Frisch collar devices with a standard calibration gamma-ray source of 137Cs and the results are presented. As known, the Frisch collar alters the weighting potential within the planar device and enhances the charge collection efficiency (CCE) distributions. Thus, the parameters affecting these distributions have great impact on the pulse hight spectrum. Device length has a dual and complex impact on CCE. Primarily, crystal (device) length L directly affects device aspect ratio (AR). Consequently, the change in AR significantly alters the weighting potential distributions and affects the CCE. Alternatively, more charge carriers are trapped in longer devices with longer traveling distances, since the mobility-life time product of the charge carriers are limited. It is shown that as a result of similarity in shape for all four devices, the weighting potential distributions resemble each other. However, due to trapping effect, the CCE profiles are not the same, and shorter devices show more uniform response to gamma rays. The CCE simulations confirms that the Frisch collar device with longer drift length, has the lowest performance.
Keywords: Frisch collar device, CdZnTe detector, CCE N31-7:

V. V. Gostilo1, A. Kozorezov2, V. Ivanov3, A. Bulycheva1, A. Owens4 1 Bruker Baltic, Riga, Latvia 2 Department of Physics, University of Lancaster, Lancaster, UK 3 RITEC, Riga, Latvia 4 Scientific Projects Department, ESA/ESTEC, Noordwijk, Netherlands We present the results of an experimental study of large (10.0x10.0x5.0 mm3) hard X- and -ray CdZnTe ring detectors. The device has a single ring electrode structure surrounding a central point anode and a guard ring on the anode plane. The currentvoltage (I-V) characteristics of the ring detectors were determined and show typical leakage currents at room temperature in the range (1-5) nA. The resistance values between the point anode and guard ring were higher than 103 GOhm for voltages (10-60) V. The resistivity was measured to be =1010 cm and the product for electrons, =3*10-3 cm2/V. The Spectroscopic performance of two ring detectors was evaluated with the standard spectrometric electronics. Both detectors exhibited good spectral resolution - the FWHM energy resolutions at 662 keV were (10-15) keV. Using a previously developed model of a ring detector1 we simulated the charge collection efficiency and line shapes for a hemispherical mode of detector operation. The distribution of the electrostatic drift field in a ring detector is non-uniform. Thus, for a given material quality the -ray detector resolution can be optimized by altering the internal field to provide as flat a charge collection efficiency from different depths as possible. This optimization procedure takes into account the exact variations of an anode weighting potential as a function of distance to absorption points, and balances differences in the induced charge through changing trapping efficiency by tuning the non-uniform drift field pattern. The optimization is achieved in practice by adjusting both the electrode pattern on the anode plane and the electrode voltage settings. We discuss and compare experimental results with model predictions and confirm the main trends for optimization. Comparison with a quasi hemispherical detectors CZT/500S with dimensions of 10x10x 5 mm3 is also presented.
1

Optimization of Large CZT Ring Detectors

A. Owens, et al, Journal of Applied Physics 102, 054505, 2007

242

N32: Radiation Damage Effects II: Scintillators Wednesday, Oct. 28 N32-1: 16:00-18:00 Grand Ballroom 7

The Radiation Hardness Test on CsI(Tl)

T. Hu Experimental Physics Center, Institute of High Energy Physics, Beijing, China On behalf of the BESIII Collaboration The BESIII electromagnetic calorimeter is composed of CsI(Tl) crystals. The radiation hardness of the CsI(Tl) crystals made by 3 manufactories has been tested. The strategy of the quality control of crystals on radiation hardness will be reported. The radiation damage effect on the absorption coefficient, the transmittance and the uniformity of light output are studied and the feasibility of monitoring the radiation damage effect by LED will be reported, too.
N32-2:

R. W. Novotny1, W. M. Doering1, D. Valery1, A. Hofstaetter2, M. Korjik3, T. Kuske1, S. Lugert4, O. Missevitch3 1 2nd Physics Institute, University Giessen, Giessen, Germany 2 1st Physics Institute, University Giessen, Giessen, Germany 3 RINP, Belarus State University, Minsk, Belarus 4 Radiation Center, University Giessen, Giessen, Germany The electromagnetic calorimeter (EMC) of the PANDA detector is based on PbWO4 crystals (length: 22Xo) of quality standard PWO-II as scintillator material to detect photons over a wide energy range with high resolution. The operation at a temperature of -25oC provides further increase of the light yield. However, the processes of radiation damage and recovery are strongly depending on the temperature and are in that case by far more sensitive to the crystal quality as known from experiences at room temperature in case of CMS/ECAL. The mass production of the final crystals has been started at the manufacturer BTCP (Russia) and the presently more than 6000 delivered crystals provide a significant base for systematic studies of performance parameters and in particular radiation hardness of PWO-II. The induced absorption over the relevant wavelength region is determined at an integral dose of 30Gy (60Co). The systematic measurements at room temperature focus on the spectral dependence, fast and slow recovery components and the correlation with the luminescence parameters. Complementary studies are performed with crystals cooled down to T=-25oC. Various experimental setups including fully assembled detector elements study the reduction of light yield and transparency as well as the recovery kinetics of several absorption regions. The results provide criteria for crystal selection for the different regions of radiation dose in the PANDA experiment. Even at low temperatures the PWO-II quality guarantees excellent energy resolution in spite of the negligible recovery processes. Concepts for monitoring and possible recovery concepts within the PANDA set-up will be discussed.
The work has been supported by BMBF and GSI. N32-3:

Limitations of the Performance of PWO-II Crystals Due to Radiation Damage

Studies of Cerium Fluoride, LYSO and Lead Tungstate Crystals Exposed to High Hadron Fluences

F. Nessi-Tedaldi, G. Dissertori, P. Lecomte, D. Luckey, F. Pauss ETH Zurich, Zurich, Switzerland High-precision calorimetry is facing a new challenge with the planning of a superLHC upgrade to collider and experiments. In particular, electromagnetic end cap calorimeters will be exposed to a high radiation environment and unprecedentedly large particle fluxes. The evaluation of different scintillating crystals for coping with that environment is urgent for upgrade planning and possibly for further R&D that might be needed on photodetectors and general calorimeter design. We will present performance results for Lead Tungstate, Cerium Fluoride and LYSO crystals exposed to various integrated fluences of 24 GeV/c protons in terms of light transmission, light output, damage recovery and activation cool-off. Qualitative studies to understand the nature of hadron-induced changes will also be presented.
N32-4:

Neutron Induced Radiation Damage Effect in Various Crystal Scintillators of Large Size

L. Zhang, R. Mao, R.-Y. Zhu California Institute of Technology, Pasadena, CA 91125, USA In this paper, we report an investigation on the radiation damage effects induced by MeV neutrons in large size crystal scintillators: PWO, LYSO:Ce, BGO and CeF3. The irradiations were carried out by using 4 to 8 MeV neutrons from a 241Am-Be source. The optical and scintillation properties of these samples, including UV excitation and emission spectra, longitudinal transmission, light output, decay kinetics and light response uniformity, were measured before and after the neutron irradiation. 243

The neutron induced photocurrent was also measured, and was used to estimate the readout noise under the neutron flux expected by an electromagnetic calorimeter at a severe radiation environment. Because of its high light output and excellent radiation resistance LYSO:Ce crystals are found to have the smallest neutron induced readout noise as compared to other large size crystals, indicating it is a good candidate material for a future crystal calorimeter in a severe radiation environment.
N32-5:

Gamma Ray Induced Radiation Damages in LSO/LYSO and PWO

R. Mao, L. Zhang, R.-Y. Zhu HEP, California Institute of Technology, Pasadena, USA This paper compares gamma ray induced radiation damage effects in two heavy crystal scintillators of large size: LSO/LYSO and PWO. Scintillation emission, optical transmission, light output, decay kinetics and light response uniformity were measured for LSO/LYSO and PWO samples of large size before and after gamma ray irradiations.
N32-6:

F. Fiedler1, H. Braess2, W. Enghardt1,3 1 Institute of Radiation Physics, Forschungszentrum Dresden-Rossendorf, Dresden, Germany 2 Siemens Medical Solutions, Erlangen, Germany 3 Oncoray, TU Dresden, Dresden, Germany In-beam PET has given valuable feedback on treatment quality over the 11 years of operation time between 1997 and 2008 of the heavy ion treatment facility at the Gesellschaft fuer Schwerionenforschung (GSI) Darmstadt. Based on this technical expertise a next generation of in-beam PET scanners will be developed. An experiment addressing the question whether the detectors and electronic components used in state-of-the-art PET-systems can be also utilized to configure a future in-beam PET scanner was performed at the medical beam line of GSI. The equipment for a new in-beam PET scanner should be tested whether it will resist the fluence of secondaries arising from the patient treatment of about 5 years. A 12C beam with an energy of E = 430.10 AMeV was stopped in a PMMA phantom. The primary particle fluence of this irradiation was equivalent to about 5300 patient fractions (3 GyE per fraction). The detectors were placed at several angles to simulate a higher secondary particle fluence, i.e. a longer time of usage. They have been exposed to a secondary particle fluence equivalent up to 13 years of usage in an in-beam PET scanner. The equipment was provided by Siemens Healthcare and tested before and after the experiment. No damage was found applying the standard test procedure.
N33: Computing and Software for Experiments IV: Software for Experimental Applications Thursday, Oct. 29 N33-1: 08:00-10:00 Grand Ballroom 1

Measurement of Radiation Hardness of PET Components

LUXSim: a Component-Centric Approach to Event Generation and Recording for Low-Background Simulations

K. Kazkaz Physics, LLNL, Livermore, CA, USA On behalf of the LUX Collaboration GEANT4 has been used throughout the physics community to simulate energy depositions in various detectors and materials. These simulations have mostly been run with a source beam outside the detector. In the case of low-background physics, however, a primary concern is the effects on the detector from radioactivity inherent in the parts themselves. From this standpoint, there is no single source or beam, but rather a collection of sources with potentially complicated spatial extent. LUXSim is a simulation framework used by the LUX collaboration that takes a component-centric approach to event generation and recording. A new set of classes allows for multiple radioactive sources to be set within any number of components at run time, with the entire collection of sources handled within a single simulation run. Various levels of information can also be recorded from the individual components, regardless of whether they were set up as sensitive detectors or not. This flexibility in both source generation and information recording is possible without a recompile, reducing the complexity of code management and proliferation of versions. Within the code itself, casting geometry variables within this new set of classes rather than as the default GEANT4 classes automatically extends this flexibility to every individual component, without any additional work required on the part of the developer. In this presentation we briefly describe the simulation philosophy behind LUXSim, detail some of its unique classes and methods, and give examples of usage.
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344.

244

N33-2:

A. Zoglauer1, G. Weidenspointner2,3, C. B. Wunderer1, S. E. Boggs1, M. G. Pia4 1 Space Sciences Laboratory, University of California at Berkeley, Berkeley, CA, USA 2 Max-Planck-Institut fuer extraterrestrische Physik, Garching, Germany 3 MPI Halbleiterlabor, Muenchen, Germany 4 INFN Genova, Genova, Italy

COSIMA - a Simulation Tool for Hard X-Ray and Gamma-Ray Telescopes Based on Geant4

Simulations are an essential tool for evaluating the performance of modern hard X-ray and gamma-ray telescopes. An especially important role play simulations of all background components encountered in the space-radiation environment originating from Earth's Albedo (photons, neutrons, etc.), from the radiation belts (protons, electrons) as well as of cosmic origin (photons, protons, alpha particles, electrons, positrons, etc.). In addition, instrument and spacecraft activation due to hadron interactions results in a delayed, internal background component, which in some cases can be dominating. We have developed a simulation tool, Cosima, capable of handling all those background components, especially the activation components, based on Geant4 and MEGAlib. It utilizes a flexible geometry and detector description, and can thus be easily applied to a variety of hard X-ray and low-to-medium-energy gamma-ray telescopes. We present the design and features of Cosima, its application possibilities, and validations based on real astrophysical measurements from e.g. TGRS.
N33-3:

S. Hauf1, M. Kuster1, M. G. Pia2,3, L. Strueder4,5, G. Weidenspointner4, A. Zoglauer6, E. Kendziorra7, C. Tenzer7, R. Chipaux8, L. Philippe8, V. Fioretti6, U. Briel5 1 TU Darmstadt, Darmstadt, Germany 2 European Organization for Nuclear Research (CERN), Geneva, Switzerland 3 INFN, Sezione di Genova, Genova, Italy 4 Halbleiterlabor, MPI, Munich, Germany 5 Max-Planck-Institut fr extraterrestrische Physik, Garching, Germany 6 Instituto di Astrofisica Spaziale e Fisica Cosmica -- IASF, INAF, Bologna, Italy 7 Institut f\, Tuebingen, Germany 8 CEA/DSM/IRFU, Centre de Saclay, Gif-sur-Yvette, France The anticipated high sensitivity of the next generation X-ray space missions, like the International X-ray Observatory -- IXO, relies on a low instrumental background, which requires optimized shielding concepts and active background reduction techniques. To optimize the tradeoff between cost, weight, and performance of the detectors and shielding components, extensive and reliable Monte-Carlo simulations are necessary. Most of the state-of-the-art approaches to estimate the prompt cosmic rays, solar proton and the cosmic x-ray induced background in space rely on simulations with the Geant4 Monte Carlo tool-kit. We will present the actual status and potential of GEANT4 to simulate the delayed background due to istotope production in the detector and shielding components. As an example we will focus on the study of two actually planned space missions: Simbol-X and the International X-ray Observatory. This work is closely related to the work of the NANO5 collaboration.
N33-4:

Progress with Geant4-based Radioactive Decay Simulation Using the Example of Simbol-X and the IXO WFI and HTRS

M. G. Pia1, G. Weidenspointner2, M. Augelli3, L. Quintieri4, P. Saracco1, M. Sudhakar1, A. Zoglauer5 1 INFN Genova, Italy, Genova, Italy 2 MPI-MPE, Munich, Germany 3 CNES, Toulouse, France 4 INFN LNF, Frascati, Italy 5 University of California at Berkeley, Berkeley, CA, USA The production of particle induced X-ray emission (PIXE) resulting from the de-excitation of an ionized atom is an important physical effect that is not yet accurately modelled in Geant4, nor in other general-purpose Monte Carlo systems. Its simulation concerns use cases in various physics domains from precision evaluation of spatial energy deposit patterns to material analysis, low background particle physics experiments and astronomy instrumentation in space science. The correct simulation of PIXE is a challenge for general-purpose Monte Carlo codes: in fact, it is intrinsically a discrete process, while all major Monte Carlo systems rely on condensed transport schemes to handle the infrared divergence of ionization cross sections. Recent developments for improving the Geant4 simulation of PIXE are presented. They include a new design of the software model, the creation of an extended and improved database of ionization cross sections for public distribution, investigations into improved particle transport schemes, and techniques to deal with infrared divergence in the context of ionization and atomic relaxation.

Recent Developments on PIXE Simulation with Geant4

245

Application results to a X-ray telescope are described; the effects of intrinsic inconsistencies introduced by the association of a discrete process to a continuous-discrete one are discussed.
The authors thank Sandro Zucchiatti for valuable discussions. N33-5:

G. Weidenspointner1, N. Kimmel1, R. Andritschke1, N. Meidinger1, L. Strueder1, R. Hartmann2, R. Richter3, A. Zoglauer4 1 MPI Halbleiterlabor, 81739 Muenchen, Germany 2 PNSensor GmBH, 80803 Muenchen, Germany 3 MPI for Physics, 80805 Muenchen, Germany 4 UC Berkeley, Berkeley, CA 94720, USA We present the status of our efforts to model the response to X rays of fully depleted, pixelized Si detectors for applications at free-electron laser sources such as FLASH or XFEL and in X-ray astronomy space missions such as eROSITA and IXO. We are pursuing a two-pronged approach to this complex problem. Experimental data on the lateral extent of the signal electron cloud upon arrival at the collecting potential minima have been obtained using pin-hole masks at various X-ray photon energies and fluxes. These experimental results were then interpreted using a suite of software tools that employ Monte Carlo simulation as well as analytical and numerical methods. Our goal is to characterize the physics of the signal electron dynamics. First results indicate that the observed lateral electron cloud size can be reproduced with an initial accuracy of better than 10% for low photon fluxes. To further improve the physical model, two aspects are of particular relevance and require further studies: the mobility of electrons at low temperatures along and perpendicular to the electric drift field, and the initial separation of electrons and holes at high charge densities (often referred to as plasma effect).
N33-6:

Modelling the Generation and Dynamics of Signal Electrons in Pixellized Si X-Ray Detectors

L. Quintieri1, R. Bedogni1, B. Buonomo1, A. Esposito1, G. Mazzitelli1, P. Valente2 1 Istituto Nazionale di Fisica Nucleare- LNF, Frascati, Italy 2 INFN Roma Universit la Sapienza, Roma, Italy

Feasibility Study of a Neutron Source at the Dafne Beam Test Facility Using Monte Carlo Codes

We present the study that we have done to estimate the feasibility of a photo-neutron source at the Da$\phi$ne Beam Test Facility, using Monte Carlo codes. In particular we describe the Montecarlo simulation results and the comparison of these ones with some important semi-empirical correlations. A neutron source at the Da$\phi$ne Beam test Facility could be useful for developping detector diagnostics for low energy and low intensity neutron flux but essentially it could be the starting point to acquire the necessary know-how for constructing high-intensity neutron sources for FEL facilities. We discuss the estimated values of the expected neutron fluxes and spectra, when the maximum beam power from the Da$\phi$ne Linac (electrons of 510 MeV ) is sent onto an optimized Tungsten target . The status of the overall design is described.
N33-7:

Design and Implementation of a Radiation Portal Monitor Multi-Lane Simulator

A. D. McKinnon, R. B. Bass, M. S. Elder, M. L. Johnson Pacific Northwest National Laboratory, Richland, WA, USA Deploying radiation portal monitors (RPMs) at U.S. ports of entry requires an understanding of a RPM systems performance at sites with a large number of RPMs. This paper describes a RPM Multi-Lane Simulator that has been designed and implemented to simulate vehicle traffic at these sites. The Simulators flexible architecture simulates vehicle traffic with its associated radiation profiles and emulates each RPMs radiation sensor panels. The RPM vendors embedded control computer firmware and supervisory software are left unchanged, thereby enabling hardware-in-the-loop testing of RPM system performance in configurations that exceed what is experienced in the field. The Simulator has proven to be a valuable and cost effective performance testing tool used by both Pacific Northwest National Laboratory and U.S. Customs and Border Protection systems integration and testing staff.
The Pacific Northwest National Laboratory is operated for the U.S. Department of Energy by Battelle Memorial Institute under Contract DEAC06-76RL01830. This work is funded by the U. S. Department of Homeland Security, U.S. Customs and Border Protection and Domestic Nuclear Detection Office under U.S. Department of Energy Contract DE-AC05-76RL01830.

246

N34: High Energy Physics Instrumentation II Thursday, Oct. 29 N34-1: 08:00-10:00 Grand Ballroom 2

The TOTEM Detector at LHC

E. Radermacher CERN, Geneva, Switzerland On behalf of the TOTEM Collaboration The TOTEM experiment is dedicated to the measurement of the total proton-proton cross-sections with a luminosity-independent method and to study elastic and diffractive scattering at the LHC. To achieve optimum forward coverage for charged particles emitted by the pp collisions in the IP5 interaction point, two tracking telescopes, T1 and T2, will be installed on each side in the pseudo-rapidity region between 3.1 and 6.5, and Roman Pot stations will be placed at distances of 147m and 220m from IP5. The telescope closest to the interaction point (T1, centered at z = 9 m) consists of Cathode Strip Chambers (CSC), while the second one (T2, centered at 13.5 m) makes use of Gas Electron Multipliers (GEM). The proton detectors in the Roman Pots are silicon devices designed by TOTEM with the specific objective of reducing the insensitive area at the edge down to a few tens of microns, in order to maximize the acceptance for protons scattered at polar angles down to a few micro-radians at IP5. The read-out of all TOTEM subsystems is based on the custom-developed digital VFAT chip with trigger capability.
N34-2:

Beam Tests of DEPFET Active Pixel Detector Prototypes

L. Andricek Max-Planck-Institut Halbleiterlabor, Munich, Germany On behalf of the DEPFET collaboration Beam Tests of DEPFET Active Pixel Detector Prototypes M.Vos, for the DEPFET collaboration The DEPFET collaboration pursues the development of ultra-thin, highly granular and precise vertex detectors for applications in future collider experiments like Belle-II or the ILC. In the depleted field effect transistor (DEPFET) concept, the signal from fully depleted silicon substrate is amplified by a Field Effect Transistor incorporated in each pixel. The latest sensor production addresses two key challenges; granularity (pixel size) and gain of the in-pixel amplification stage. In parallel, a complete module concept based on a large and thin DEPFET sensor has been developed. In July and August 2008 seven DEPFET prototype modules were tested in proton and pion beams at CERN's Proton Synchrotron (PS) and Super Proton Synchrotron (SPS), collecting over twenty million events. In this contribution the devices under test, the experimental setup and the analysis software will be described in some detail. Final results of the analysis will be presented in this contribution. The device performance was found to be excellent, with a very high signal-to-noise ratio (>135) and excellent spatial resolution (1.5 micron). The implications of these results for applications like Belle-II and the ILC will be discussed.
N34-3:

Serial Powering for Silicon Tracking at the Super-LHC

M. M. Weber PPD, Rutherford Appleton Laboratory, Didcot, UK The conventional independent or parallel powering of silicon trackers is not practical at the Super-LHC, due to the five- to tenfold increased number of electronic channels compared with the LHC trackers. Serial powering is an elegant solution for powering the SLHC inner trackers with a minimum volume of cables. The technique was first proposed for the ATLAS pixel detector, and has since been investigated extensively for powering of silicon strip detectors at the Super-LHC. Recent highlights of this R&D include: the system design of ATLAS strip supermodules, the delivery of the ATLAS ABCN-25 readout chip, and the fabrication of the general purpose SPi serial powering chip. The powering blocks of the ABCN-25 and SPi chips can be used to implement three distinct serial powering architectures, with different design philosophies. The performance of the three architectures obtained with 20 chip and 10 chip ABCN-25 hybrids will be presented. The system design of a serially powered supermodule has progressed greatly in the last two years. It will be discussed in detail with focus on grounding and shielding, protection circuitry, AC-coupling of signals, and power slow-control.
N34-4:

Development of Serial Powering for the Upgrade of the ATLAS Pixel Detector

F. Huegging, D. Arutinov, M. Barbero, A. Eyring, L. Gonella, M. Karagounis, H. Krueger, N. Wermes Physikalisches Institut, University of Bonn, Bonn, Germany Silicon tracking detectors of future particle physics experiments will have high granularity and thus extreme high power densities at low supply voltages. If this power is provided in a parallel scheme this leads to a lot of material for cables in the active area 247

yielding in a decreased measurement performance for penetrating particles. Therefore an alternative powering scheme, the so called serial powering, is proposed for the upgrade of the ATLAS pixel detector. In this concept a series of detector modules are connected in a chain at higher input voltage reducing the number of needed cables as well as the power dissipation inside the cables. Main components of this development like new on-chip voltage regulators and protection schemes are presented as well as further system aspects like DCS and HV integration and AC coupling of data transmission are discussed.
N34-5:

S. K. Dhawan1, O. K. Baker1, H. Chen2, R. Khanna3, J. Kierstead2, F. Lanni2, D. Lynn2, A. Mincer4, C. Musso4, S. Rescia2, H. Smith1, P. L. Tipton1 1 Physics, Yale University, New Haven, CT, United States 2 Physics, Brookhaven National Laboratory, Upton, NY, United States 3 Power Application Engineering, National Semiconductor Corp, Richardson, TX, United States 4 Physics, New York University, New York, NY, United States A new generation of Commercial-Off-the-Shelf (COTs) buck-type converters built using advanced short channel "high voltage" CMOS processes have the potential to operate near the interaction region of the proposed Super Large Hadron Collider (sLHC) upgraded accelerator. The benefit would be a simpler DC power distribution system and an increase of the overall power efficiency by allowing higher voltage power delivery to the front-end electronics, thus limiting ohmic losses. The devices must operate in a high magnetic field and be able to withstand both high doses of ionizing radiation and large neutron fluence. These converters are to be mounted on the same readout boards as the sensitive front-end electronics or in close proximity without introducing any additional noise due to the high switching frequencies. Radiation hardness and noise test results will be reported for application to tracking detectors and calorimeters.
N34-6:

Commercial-off-the-Shelf DC-DC Converters for High Energy Physics Detectors for the sLHC Upgrade

Diamond Detectors for Radiation and Luminosity Measurements in CMS

M. S. Hollingsworth University of Tennessee - Knoxville, Knoxville, TN, USA On behalf of the CMS BRM collaboration The Beam Conditions Monitor (BCM), a set of diamond-based detectors, provides fast, relative measurements of particle fluxes for use in the safety systems of CMS. Sudden, order of magnitude changes in the BCM readout issue non-maskable LHC beam aborts. Dangerous irradiation trends on longer timescales translate into automatic detector interlocks and injection inhibit. Operators in the LHC beam and CMS detector control room obtain real time (1Hz) readout of the BCM acquisitions. The beam radiation monitoring system also provides an independent measurement of the beam luminosity. The next generation luminosity detector, called the Pixel Luminosity Telescope, is based on pixelated mono-crystalline diamond detectors. They provide a fast occupancy information and allow particle tracking near the interaction point to distinguish trajectories originating from the proton-proton collision point and those parallel to the beam pipe. In this talk we present the use case of diamond detectors for beam radiation monitoring in CMS and first measurements with particle beams.
N34-7:

The ATLAS Muon Spectrometer

J. Dubbert Max-Planck-Institut fuer Physik, Munich, Germany On behalf of the ATLAS Muon Collaboration The ATLAS experiment at the Large Hadron Collider (LHC) at CERN is expected to record its first data from proton-proton collisions in autumn 2009. Its muon spectrometer is designed to achieve a momentum resolution of better than 10% at 1 TeV. The spectrometer consists of a system of three superconducting air-core toroid magnets and is instrumented with three layers of Monitored Drift Tube chambers (Cathode Strip Chambers in the extreme forward region) as precision detectors. Resistive Plate Chambers in the barrel and Thin Gap Chambers in the endcap regions provide a dedicated fast trigger system. In summer 2008, during the start-up phase of the LHC, the first events from single beams were recorded with the fully functional ATLAS detector. Previous and subsequent data taking with cosmic rays yielded a data set of several 100 Million events with and without magnetic field. We will report on the current status and our experience with the muon precision tracking and trigger chambers, the level-1 trigger, and the spectrometer alignment. The global performance of the muon system will be discussed and first results from combined studies with other sub-detectors will be presented. The readiness of the ATLAS muon spectrometer for first collisions will be demonstrated.

248

N35: Astrophysics and Space Instrumentation I: Component development Thursday, Oct. 29 N35-1: 08:00-10:00 Grand Ballroom 3

Performance of the Hybrid Photon Detector Cluster for MAGIC-II

R. Orito Max Planck Institute, Munich, Germany On behalf of the MAGIC collaboration MAGIC is an imaging atmospheric Cherenkov telescope (IACT) located in LaPalma, Canary islands. The first telescope, MAGIC-I, has been operated since 2004. The stereoscopic observation with the second telescope, MAGIC-II, is planned to start in 2009. Using the two telescopes with mirrors of 17m diameter, the sensitivity of MAGIC is greatly improved. To achieve higher gamma-ray sensitivity with MAGIC-II, we are planning to use a new camera consisting of photon detectors with higher performance. One of the promising candidates of the new photon detector is a hybrid photon detector with high quantum efficiency (QE). For MAGIC-II, we have developed a new HPD Hamamatsu R9792U-40 with a GasAsP photocathode in collaboration with Hamamatsu Photonics K.K.. The Hamamatsu R9792U-40 HPD has high QE(over 50% at 500nm), low after pulse(300 times lower than PMT), fast pulse shape(~2.1ns FWHM). Especially the high QE and low after pulse are important advantages to be used on IACT. By using the R9792U-40 HPD, we can obtain twice as much light output from atmospheric Cherenkov light photons, compared with the MAGIC-I camera. We have also measured a life time of the GaAsP photocathode, implemented circuits for protection from the strong light and temperature compensation. All basic performance parameters of HPD satisfy requirements to be used on IACT. We have developed the HPD cluster for the upgrade of MAGIC-II. The geometry and interface of the HPD cluster are compatible to the current PMT cluster for future possible replacements. After the performance evaluation on MAGIC-II, the central region(427 pixels) of the MAGIC-II camera will be replaced to the HPD cluster for higher gamma-ray sensitivity with lower energy threshold. In this presentation, we report on the performance and status of the HPD cluster for MAGIC-II.
N35-2:

A Novel G-APD Based Camera for Imaging Air Cherenkov Telescopes: Concept, Realization and First

Tests
M. Rissi ETH Institute for Particle Physics, Zurich, Switzerland On behalf of the FACT collaboration We present a novel camera design for the measurements of Very High Energy (VHE) Gamma Rays with Cherenkov Telescopes, based on Geiger-mode Avalanche Photodiodes (G-APD). Their promising features make them interesting candidates for the observations of the short and faint Cherenkov light flashes. In our prototype setup, the data is recorded using the Domino Ring Sampler 2 (DRS2) chip. To control gain variations, a high voltage feedback system was implemented. We will present the design, the setup and first tests of such a camera.
N35-3:

J. S. Hiraga1, S. Nakamura2, K. Miyamoto2, H. Katakura2, Y. Katada2, H. Kawano2, Y. Torigoe2, K. Fujii2, H. Tawara3 1 Cosmic Radiation Laboratory, RIKEN, Wako, Saitama, Japan 2 Yokohama National University, Yokohama,Kanagawa, Japan 3 High Energy Accelerator Research Organizatio, Tsukuba,Ibaraki, Japan

The Direct Measurement of the Signal Charge Behavior beyond 10keV in CCDs with Subpixel Resolution

Charge-coupled devices(CCDs) are widely used in X-ray astronomy as a focal plane detector for X-rays up to 10 keV. For future X-ray space missions, thick CCDs are begin developed to improve the detection efficiency of high energy X-rays beyond 10 keV. We have developed a novel multi-collimator using a barium phosphate glass, BP-1, which is valid for higher X-ray energy. Our BP-1 collimator has a large number of tapered pinholes which are randomly distributed and with ~5um in diameter. We have performed BP-1 collimator experiment using characteristic X-rays of Mo-K~(17.5,keV) as well as Cu-K(8.0,keV). We obtained the restored pixel images within a pixel for various X-ray event patterns. We measured the charge cloud shape generated by an X-ray photon beyond 10keV, for the first time. The Obtained charge cloud size for 17.5keV X-ray are 3.65+/-0.65um and 3.74+/0.66um for horizontal direction and vertical direction, respectively. This is the first verification of BP-1 collimator experiment on to CCDs.

249

N35-4:

J. Treis1, P. Lechner2, L. Andricek3, F. Aschauer4, K. Heinzinger2, S. Herrmann4, T. Lauf4, P. Majewski2, G. Lutz2, R. H. Richter3, M. Porro4, G. Schaller4, M. Schnecke3, F. Schopper4, H. Soltau2, L. Strder4, G. deVita4 1 Max-Planck-Institute for Solar System Research, Katlenburg-Lindau, Germany 2 pnSensor GmbH, Munich, Germany 3 Max-Planck-Institute for Physics, Munich, Germany 4 Max-Planck-Institute for Extraterrestrial Physics, Munich, Germany

Performance of the DEPFET Macropixel Based Focal Plane Detectors for the MIXS Instrument on BepiColombo

Radiation detectors using the combined Detector-Amplifier structure DEPFET (Depleted P-channel FET) as sensitive element are used in a variety of scientific applications, e.g. in X-ray astronomy and high energy physics. They combine excellent energy resolution, high speed readout and low power consumption with the features of random accessibility of pixels, readout-ondemand, a geometrical fill factor of 100% and excellent quantum efficiency even for low energy X-rays. DEPFET based focal plane arrays are also used for the planetary XRF instrument MIXS on the European Mercury exploration mission BepiColombo. MIXS uses a lightweight X-ray mirror system to focus fluorescent radiation from the Mercury surface on the FPA detector to measure the element abundance and distribution in Mercurys crust. The FPA needs to have an energy resolution better than 200 eV FWHM @ 1 keV and an energy range from 0.5 keV to 10 keV, for a pixel size of 300 x 300 m2. Main challenges for the instrument are the high increase in leakage current due to non-ionizing energy loss from protons from the solar wind, and the limited cooling resources due to the difficult thermal environment in the mercury orbit. To provide for the large pixel size, a new type of DEPFET based detector is used, which combines a DEPFET readout amplifier cell with a silicon drift diode (SDD) like drift ring structure. This so-called macropixel detector concept allows for pixel sizes of up to 1 x 1 mm2 and larger. In addition, it combines the charge storage capability to a silicon drift detector array, a prerequisite for building large integrating focal plane detectors. The sensors for the MIXS spectrometer were produced at the MPI semiconductor laboratory in Munich. Test devices and flight detectors have been operated, and the screening of the flight devices is in progress. The devices show excellent homogeneity and near Fano-limited energy resolution in an energy range between 0.5 and 10 keV.
N35-5:

F. F. Khalid1, M. L. Prydderch1, Q. R. Morrissey1, P. Seller1, E. Valtonen2,3, J. Peltonen2, M. Syrjsuo4, R. Vainio5, J. Huovelin6 1 Technology, Rutherford Appleton Laboratory, STFC, Didcot, Oxon, United Kingdom 2 Aboa Space Research Oy, Turku, Finland 3 Department of Physics, University of Turku, Turku, Finland 4 Space research, Finnish Meteorological Institute, Helsinki, Finland 5 Department of physical sciences, University of Helsinki, Helsinki, Finland 6 Observatory, University of Helsinki, Helsinki, Finland

Solar Intensity X-Ray Spectrometer (SIXS) ASIC Onboard the ESA BepiColombo Mission to Mercury

The new SIXS prototype ASIC has been designed for detecting and measuring solar X-rays in a Mercury orbit onboard BepiColombo. The SIXS detector consists of a CsI(Tl) core detector and 5 Silicon surface detectors. This new ASIC has undergone major modifications and has a new differential charge preamplifier without any pole zero cancellation and has a large dynamic range of 2,000e- to 1,500,000e-. The ASIC has 8 channels, each of which consists of a preamplifier, a CR-RC shaper, a peak detect hold, a comparator, a buffer, and a new 8 bit resistor string DAC which sets the programmable threshold for the comparator There are 6 channels for Si detectors with a shaping time of 1s and 2 channels for CsI detector with a shaping time of 3s, one of which has twice the gain to accommodate for low output efficiency of the detector. The channels are multiplexed to a single 10/11 bit SAR ADC which runs at a 10MHz clock and additional on chip programmable phase shifted clocks and programmable Upper and Lower threshold voltages. This ASIC has been manufactured in a 0.35m CMOS process using enclosed geometry transistors. All digital logic in the ASIC including control and ADC logic use voting logic flip-flops for SEU prevention. The ASIC architecture and design flow are described including measured results.
N35-6:

K. Lacombe1, N. Remoue1, C. Amoros1, D. Barret1, O. Gevin2, O. Godet1, J. Lande1, E. Lecomte1, O. Limousin2, F. Lugiez2, P. Mandrou1, J. Narbonne1, R. Pons1, D. Rambaud1, P. Ramon1, G. Rouaix1 1 CNRS-CESR, Toulouse, FRANCE 2 CEA-IRFU, Saclay, FRANCE ECLAIRs, a 2D coded-mask imaging telescope on the Sino-French SVOM mission, will detect and locate gamma-ray bursts, between 4 and 250 keV. The ECLAIRs detector array is an assembly of 6400 4x4 mm2, 1 mm thick Schottky CdTe semiconductor detectors, biased at -600 V and operated at -20C. The unprecedented low-energy threshold is achieved through a careful selection of the detector, a very low noise 32 channels ASIC readout, and an innovative hybridization of two ceramics, a thick film one holding the 32 CdTe detectors with the high voltage grid and an HTCC ceramic housing the ASIC. In this paper,

Development of the ECLAIRs Camera, a 6400 CdTe Array for X/Gamma detection

250

we will outline the detector selection process, describe the hybrids, and present the results from an on-going campaign to evaluate their spectroscopic performance.
Keywords: SVOM, ECLAIRs, Gamma-Ray bursts, CdTe, Hybridization N35-7:

H. Takahashi1, M. Matsuoka1, T. Mizuno1, Y. Fukazawa1, T. Yanagida2, Y. Yokota2, A. Yoshikawa2, N. Kawaguchi3, S. Ishizu3, K. Fukuda3 1 Hiroshima University, Higashi-hiroshima, Japan 2 IMRAM, Tohoku University, Sendai, Japan 3 Tokuyama Corporation, Sendai, Japan For high-sensitivity neutron measurement, we developed a phoswich detector consisting of LiYF$_{4}$ (Ce 1\%) and BGO crystals. Light outputs of LiYF$_{4}$ excited from $\alpha$-ray (simulating neutron excitation) have the faster decay-time constant $\tau \sim$ 50 ns, compared with that from the shielding BGO crystal ($\tau \sim$ 300 ns), and these waveforms output from a preamplifier are clearly distinguished each other. When the waveforms with the faster rise-time were selected, the spectrum was obtained with only the peak of 5.49 MeV $\alpha$-ray from $^{241}$Am.
N36: Nuclear Physics Instrumentation II Thursday, Oct. 29 N36-1: 08:00-10:00 Grand Ballroom 7

Study on a Phoswich Detector Consisting of Li-Composed Crystal Scintillator and BGO for Neutron Measurement

E. B. Johnson1, R. Miskimen2, C. J. Stapels1, S. Mukhopadhyay1, J. F. Christian1 1 Radiation Monitoring Devices, Inc., Watertown, MA, USA 2 University of Massachusetts, Amherst, MA, USA

Characteristics of CMOS Avalanche Photodiodes at Cryogenic Temperatures

Nuclear and high-energy physics experiments that are conducted in harsh environments, such as the milliKelvin temperatures of a helium dilution refrigerator, the high magnetic fields of several Tesla in a superconducting magnet, or the small physical dimensions of accelerator-beam target chambers, require improved photodetectors that operate in these conditions. Improvements in detector technology used in extreme environments can improve the data quality and allow new designs for experiments that operate under these conditions. Photomultiplier tubes (PMTs), the standard technology, provide excellent performance characteristics, however, in cryogenic environments, their bulk increases the heat load burden, unless light pipes are used to isolate the PMT from the cryogenic region. The critical drawback to light guides is that they loose light, particularly in systems requiring a segmented light guide for thermal isolation. In addition, a PMTs susceptibility to large magnetic fields limits their use in certain experiments. A reasonable approach to improve measurements at these extreme conditions is to employ a novel SolidState Photo-Detector concept. Using structures from RMDs CMOS SSPMs (solid-state photomultiplier) designed for room temperature operation, these photodiodes show avalanche multiplication at temperatures down to 5 K and operated in a proportional mode (below breakdown). Our results indicate that operation in Geiger mode (above breakdown) at 5 K is not feasible, as excess noise terms begin to dominate the diode response. The proportional modeoperation shows a response with reasonable gain and noise terms.
N36-2:

S. Lakshmi1, P. Chowdhury1, C. J. Lister2, S. K. Tandel1, S. Gros2, R. Farell3, M. McClish3, K. S. Shah3 1 Department of Physics and Applied Physics, University of Massachusetts Lowell, Lowell, MA, USA 2 Department of Physics, Argonne National Laboratory, Argonne, IL, USA 3 Material Science Group, Radiation Monitoring Devices, Inc., Watertown, MA, USA

Angular Correlation Measurements with Novel, Position-Sensitive Alpha and Gamma Detectors

Alpha-gamma angular correlations in the decay of 224Ra have been measured using a position-sensitive avalanche photo-diode (PSAPD) for detecting alpha particles and a planar germanium double-sided strip detector (GeDSSD) for detecting gamma rays. The excellent position resolutions of both detectors allow close positioning of the radiation source to the detectors, covering a large angular range in a single configuration, thus enhancing efficiency in a compact geometry. The PSAPD works on the principle of charge fractionation. It was operated at a bias voltage of 1400V at the onset of avalanche process to improve signal to noise. The energy and position resolution for 6 Mev alpha particles was measured to be 140keV and 400m, respectively. The GeDSSD has 14 horizontal and 14 vertical contacts of 5mm width on opposite faces. Results of improved distortion correction algorithms for the PSAPD and angular correlation results will be presented.

251

M. Kuster1,2, R. Hartmann3, D. H. H. Hoffmann1, P. Lang1, H. Braeuninger2, H. Fischer4, I. Irastorza5, K. Koenigsmann4, R. Kotthaus6, G. Lutz3, J. Morales5, A. Nordt7, J.-F. Poncet7, B. Riffaud7, J. Ruz5, L. Strueder8,2, J. Vogel4, H. Gorke9 1 Institut fuer Kernphysik, TU Darmstadt, Darmstadt, Germany 2 Max-Planck-Institut fuer extraterrestrische Physik, Garching, Germany 3 PNSensor GmbH, Muenchen, Germany 4 Institut fuer Kernphysik, Universitaet Freiburg, Freiburg, Germany 5 Laboratorio de Fisica Nuclear y Altas Energias, Universidad de Zaragoza, Zaragoza, Spain 6 Max-Planck-Institut fuer Physik, Muenchen, Germany 7 European Organization for Nuclear Research (CERN), Geneva, Switzerland 8 MPI Halbleiterlabor, Muenchen, Germany 9 ZEL, Forschungszentrum Juelich, Juelich, Germany The CAST (CERN Axion Solar Telescope) experiment at CERN searches for solar axions employing a helioscope which consists of a 9 Tesla superconducting LHC prototype dipole magnet equipped with 3 independent detection systems for X-rays in the energy range of 1 to 10 keV from axion-photon conversions inside the 10 m long magnetic field. One of 3 independent detector systems of CAST is an X-ray telescope employing a Wolter I type grazing incidence X-ray mirror system and a fully depleted pn-CCD as focal plane detector. The X-ray telescope is assembled of components available from Xray satellite missions (ABRIXAS and XMM-Newton). The use of an X-ray mirror optics greatly reduces the background by a factor of approx. 155 compared to non-focusing instruments. This vast step forward comes about by the combined use of instrumentation developed for high energy physics and for astrophysics. At present we develop a new focal plane detector for the X-ray telescope, with an improved response for photon energies E < 1 keV and a significantly lower background level compared to the existing CCD detector system. The heart of the new detector is a backside illuminated frame-store pn-CCD with a 3.26 x 1.92 cm 2 sensitive area. The application in CAST will be the first time, that such a device is operated in an environment which requires a low background. We will present first results of performance and background measurements in the laboratory and the CAST environment.
N36-4:

N36-3:

A Low Background Frame-Store PN-CCD Detector for CAST

Efficiency Calibration of a 4 Sum-Spin Spectrometer Using GEANT4

A. K. Gourishetty, I. Mazumdar, D. A. Gothe Dept. of Nuclear and Atomic Physics, Tata Institute of Fundamental Research, Mumbai, Maharashtra, India Efficiency calibration and coincidence summing correction have been performed for a 4 sum-spin spectrometer developed for measuring the multiplicity and sum-energy of the discrete low energy gamma rays emitted in heavy-ion induced fusionevaporation reactions. The spectrometer is an array of 32 conical NaI(Tl) detectors of pentagonal and hexagonal cross sections possessing both extremely high efficiency and a severe coincidence summing effect. In order to understand the effect of coincidence summing due to gamma rays emitting from radio-nuclides with a two-step cascade decay, we have carried out extensive simulations using radioactive decay module available in GEANT4 simulation package considering the array in a realistic geometry. The absolute total detection efficiency, absolute photo peak efficiency and coincidence summing correction factors have been simulated for a wide range of gamma energy from 700 keV to 2800 keV using point sources of 60Co, 46Sc, 94Nb and 24Na each of which emits two gamma rays in coincidence with almost 100% probability. The results are found to be in excellent agreement with the absolute efficiencies, both total and photo peak, simulated for mono energetic gamma rays of similar energies. In order to verify the validity of these results, we have also made measurements of efficiencies and coincidence summing correction factors using a calibrated point source of 60Co placed at center of the array. For both peak and total efficiencies, simulation and experiment typically agreed within 5%. The discrepancy between simulation and experiment for the coincidence summing correction factors was less than 1%. The approximate methods used in extracting the efficiencies and coincidence summing correction factors were found to be reliable for calibrating high-efficiency spectrometers. The detailed results of simulations and measurements will be presented and discussed.
N36-5:

F. C. L. Crespi1, V. Vandone1, S. Brambilla2, F. Camera1, B. Million2, S. Riboldi1, M. Sassi2, O. Wieland2 1 Universityof Milan / INFN Milan, Milan, Italy 2 INFN sez. of Milan, Milan, Italy In gamma spectroscopy experiments obtaining precise timing information is fundamental, for instance in heavy-ion fusionevaporation experiments to discriminate the unwanted contribution of neutrons by time of flight measurements or in experiments with radioactive ion beams to clean the gamma spectra from background radiation not coming from the target position. Next generation gamma ray tracking arrays like AGATA will cover the full 4p solid angle with large volume segmented HPGe 252

Improvement of HPGe Detectors Timing Using Pulse Shape Analysis

detectors. In these detectors a 5mm position sensitivity is achieved trough the segmentation of the outer electrode and the analysis of the current pulse shape (Pulse Shape Analysis, PSA). Large volume HPGe detectors have a time resolution limited to about 710 ns. This is due mainly to the presence of electric noise and to the fact that the rise front of the detector signal changes shape depending from the interaction position. The timing algorithm used in most of the in beam gamma spectroscopy experiments is the Constant Fraction Discriminator (CFD) which starts from the assumption of having as input a signals with a perfectly linear rise front and identical shape. The aim of this work is to investigate the possibility to eliminate the uncertainty due to the signal shape variation by using PSA techniques, thus improving the HPGe detector timing resolution. At this purpose a measurement using a coaxial HPGe detector, a LaBr3 scintillation detector and 60Co gamma source was realized. The signal shape of the HPGe is acquired when the two detectors fire in coincidence, using a 12 bit CAEN V1729 ADC VME board with a sampling frequency of 2 GHz . The RS Pulse Shape Analysis algorithm has been then applied to the experimental signals for extracting timing information and the results obtained were compared with those of a standard CFD. The results show a good improvement in timing resolution obtained applying RS algorithm (4 ns FWHM ) in comparison to standard CFD (8 ns FWHM).
N36-6:

Electronics for the MINERvA Experiment

P. Rubinov Fermi National Accelerator Laboratory, Batavia, IL, USA On behalf of the Minerva Collaboration MINERvA is new experiment designed to make precision measurements of neutrino nuclear cross-sections across a wide range of energies in support of current and future long-baseline neutrino oscillation experiments. The detector is being built using proven technologies and consists of a finely segmented, fully active scintillator tracker surrounded by ECAL and HCAL, with readout via approximately five hundred M64 multi-anode PMTs. We report on the design and operation of the electronics and data acquisition system for this detector. The electronics consists of highly integrated Front End Board mounted on each tube and connected via custom serial daisy chains to very dense VME based readout boards. This approach emphasized low cost and existing electronics and allows us to read in excess of forty thousand channels from a single VME crate, however in the experiment, we use two VME crates for ease of use.
N36-7:

Features and Final Test Results of the EXPADES Apparatus

M. Romoli Sezione di Napoli, Istituto Nazionale Fisica Nucleare, Napoli, Italy On behalf of the EXOTIC Collaboration We are performing the final tests of a new detection apparatus, named EXPADES (EXotic PArticle DEtection System), expressly developed to be used in nuclear physics experiments involving low-intensity radioactive beams. It consists of double-side segmented silicon detectors and a highly miniaturized read-out system based on a new production commercial ASIC chipset. It will be used to study nuclear reactions (scattering, direct reactions, breakup) at energies around the Coulomb barrier. The main features of the apparatus are a high granularity (832 elements, 11264 pixels), a very large solid angle covered (about 90% of 4 sr) and a wide angular range subtended (785 and 95173). A more detailed description of the apparatus and of the test results will be given in this contribution.
N37: Astrophysics and Space Instrumentation II: Instruments Thursday, Oct. 29 N37-1: 10:30-12:00 Grand Ballroom 1

The Dark Energy Camera - a New Instrument for the Dark Energy Survey

K. Honscheid Physics, Ohio State University, Columbus, Ohio, USA On behalf of the DES Collaboration The discovery that the universe is accelerating, not slowing down from the mass it contains, is the surprise that sets the initial research program of 21st Century cosmology. The Dark Energy Survey (DES) is a next generation sky survey aimed directly at understanding this mystery. DES is designed to measure the dark energy equation of state parameter with four complementary techniques: galaxy cluster counts, weak lensing, angular power spectrum and type Ia supernovae. We present an overview of the DES instrument (DECam) which will be mounted at the prime focus of the Blanco 4m telescope at CTIO. DECam includes a 3 square degree focal plane covered by 62 2kx4k CCDs, a five element optical corrector, up to eight filters, a modern readout and control system, and the associated infrastructure for operation in a new prime focus cage. We will use the 250 micron thick fullydepleted CCDs developed at Lawrence Berkeley National Laboratory (LBNL). DECam also includes design features to enhance the image quality and the efficiency of operations. DECam will be devoted to the DES for 30% of the time over five years and 253

will otherwise be available to the community as an NOAO facility instrument. We will review the status of the construction of the instrument highlighting the results of this summer's full scale integration tests.
N37-2:

Precision Studies of Cosmic Rays with the PAMELA Satellite Experiment.

E. Mocchiutti Italian National Institute for Nuclear Physics (INFN), Trieste, Italy On behalf of the PAMELA Collaboration The PAMELA satellite experiment was launched into low earth orbit on June 15th 2006. The combination of a permanent magnet silicon strip spectrometer, and a silicon-tungsten imaging calorimeter allows precision studies of the charged cosmic radiation to be conducted over a wide energy range (~100 MeV - ~200 GeV). A primary scientific goal is to search for dark matter particle annihilations by measuring the energy spectra of cosmic ray antiparticles. Latest results from the PAMELA experiment will be reviewed with a particular focus on cosmic ray antiprotons and positrons. The status of PAMELA measurements for other cosmic ray species will also be reviewed.
N37-3:

M. S. Bandstra1, E. C. Bellm1, S. E. Boggs1, D. Perez-Becker1, C. B. Wunderer1, A. Zoglauer1, M. Amman2, P. N. Luke2, H.K. Chang3, J.-L. Chiu3, J.-S. Liang3, Y.-H. Chang4, W.-C. Hung4, Z.-K. Liu4, C.-H. Lin5, M. A. Huang6, P. Jean7 1 UC Berkeley Space Sciences Laboratory, Berkeley, CA, USA 2 Lawrence Berkeley National Laboratory, Berkeley, CA, USA 3 National Tsing Hua University, Hsinchu, Taiwan 4 National Central University, Jhongli City, Taiwan 5 National Space Organization (NSPO), Hsinchu, Taiwan 6 National United University, Miaoli, Taiwan 7 Centre d'Etude Spatiale des Rayonnements, Toulouse, France The Nuclear Compton Telescope (NCT) is a balloon-borne soft gamma-ray (0.2-10 MeV) telescope designed to study astrophysical sources of nuclear line emission and polarization. NCT consists of twelve high-purity germanium cross-strip detectors (GeDs) that measure both the position and energy of gamma-ray interactions. A 10-GeD version was flown from May 17-18 2009 from the Columbia Scientific Balloon Facility in Fort Sumner, NM, with a total flight duration of 38.5 hours. The various calibrations that were performed will be summarized here but presented in other talks and posters. Here we summarize the instrument, the flight, and our preliminary science results.
N37-4:

The Spring 2009 Balloon Flight of the Nuclear Compton Telescope

T. Tanaka1, Y. Fukazawa2, J. Kataoka3, M. Kokubun4, G. Madejski1, K. Makishima5, T. Mizuno2, K. Nakazawa5, R. Sato4, H. Tajima1, T. Takahashi4, M. Tashiro6, Y. Terada6, S. Watanabe4, K. Yamaoka7 1 KIPAC, Stanford University, Menlo Park, CA, USA 2 Hiroshima University, Higashi-hiroshima, Japan 3 Waseda University, Tokyo, Japan 4 ISAS/JAXA, Sagamihara, Japan 5 University of Tokyo, Tokyo, Japan 6 Saitama University, Saitama, Japan 7 Aoyama Gakuin University, Sagamihara, Japan The Soft Gamma-ray Detector (SGD) on board ASTRO-H (Japanese future high energy astrophysics mission) is a Compton telescope with narrow field of view (FOV), which utilizes Compton kinematics to enhance its background rejection capabilities. It is realized as a hybrid semiconductor detector system which consists of silicon and CdTe (cadmium telluride) detectors. It can detect photons in a wide energy band (50-600 keV) at a background level 10 times better than that of the Suzaku HXD, which is complimentary to the Hard X-ray Imager (HXI) on board ASTRO-H with an energy coverage of 5-80 keV. Excellent energy resolution is the key feature of the SGD, allowing it to achieve good background rejection capability taking advantage of good angular resolution. An additional capability of the SGD, its ability to measure gamma-ray polarization, opens up a new window to study properties of gamma-ray emission processes. We will present the development of key technologies to realize the SGD: high quality semiconductor detectors (Si and CdTe), low noise front-end ASIC and compact packaging technology. Energy resolutions of ~1.3 keV (FWHM) for Si and CdTe pixel detectors have been measured. We also present the validation of Monte Carlo simulation used to evaluate the performance of the SGD.

The Soft Gamma-Ray Detector for the ASTRO-H Mission

254

N37-5:

The Advanced Gamma-Ray Imaging System (AGIS)

A. N. Otte SCIPP, University of California in Santa Cruz, Santa Cruz, CA, USA On behalf of the AGIS collaboration The Advanced Gamma-ray Imaging System (AGIS) is a concept for the next generation of imaging atmospheric Cherenkov telescope arrays. It has the goal of providing an order of magnitude increase in sensitivity for Very High Energy Gamma-ray (~100 GeV to ~100 TeV) astronomy compared to currently operating arrays such as CANGAROO, HESS, MAGIC, and VERITAS. After an overview of the science such an array would enable, we discuss the development of the components of the telescope system that are required to achieve the sensitivity goal. AGIS stresses improvements in several areas of IACT technology including component reliability as well as exploring cost reduction possibilities in order to achieve its goal. We discuss alternatives for the telescopes and positioners: a novel Schwarzschild-Couder telescope offering a wide field of view with a relatively smaller plate scale, and possibilities for rapid slewing in order to address the search for and/or study of Gamma-ray Bursts in the VHE gamma-ray regime. We also discuss options for a high pixel count camera system providing the necessary finer solid angle per pixel and possibilities for a fast topological trigger that would offer improved realtime background rejection and lower energy thresholds.
N38: Synchrotron Radiation Instrumentation Thursday, Oct. 29 N38-1: 10:30-12:00 Grand Ballroom 2

C. J. Kenney1, A. C. Thompson2, J. Hasi3, D. Gnani4, S. I. Parker5, E. Mandelli4, G. Meddeler4, E. Westbrook2 1 SLAC National Accelerator Laboratory, Menlo Park, CA, USA 2 Molecular Biology Consortium, Chicago, IL, USA 3 Manchester University, Manchester, UK 4 Lawrence Berkely National Laboratory, Berkeley, CA, USA 5 University of Hawaii, Honolulu, HI, USA A prototype of a direct-conversion x-ray detector for protein crystallography, SAXS, and other applications has been developed. This system would use active-edge, silicon diode sensors in a shingled array. Use of active edges is critical to avoid missing diffraction spots, which might otherwise coincide with the dead regions between sensors. We have developed a corresponding pixel readout circuit chip, which is bump bonded to the prototype sensors. The detector system would enable crystallography data to be taken with higher dynamic range and shorter frame times than is currently done with CCD detectors. A time resolution of around 16 microseconds will permit other applications such as high-speed, time-resolved small angle x-ray scattering. Measurements have been made using a synchrotron x-ray beam at the Advanced Light Source of the dynamic range, energy resolution and time resolution of the prototype system.
We'd like to thank the Advanced Light Source, Stanford Nanofabrication Facility, National Institutes of Health, and the National Science Foundation N38-2:

A High Frame Rate X-Ray Camera for Synchrotron Applications

R. L. Bates1, D. Pennicard1, J. Marchal2, C. Fleta3, G. Pellegrini3, M. Lozano3, C. Parkes1, N. Tartoni2, D. Barnett2, I. Dolbny2, K. Sawhney2, V. O'Shea1, A. MacRaighne1, D. Maneuski1, V. Wright4 1 Physics and Astronomy, The University of Glasgow, Glasgow, UK 2 Diamond Light Source Ltd,, Oxfordshire, UK 3 Instituto de Microelectronica de Barcelona, Barcelona, Spain 4 Science and Technology Facilities Council, Swindon, UK Three-dimensional (3D) photodiode detectors offer advantages over standard planar detectors in a range of applications, including X-ray detection for synchrotrons and medical imaging. The principal advantage of these sensors for X-ray imaging is their low charge sharing between adjacent pixels; which improve spatial and spectral resolution. Double-sided 3D detectors have been flip-chip bonded to Medipix2 and TimePix single-photon counting readout chips, and tested in a monochromatic X-ray beam at the Diamond synchrotron. Tests of the 3D detector's response spectrum and its Line Spread Function have shown that it has substantially lower charge sharing than standard planar Medipix2 and TimePix sensors. A 3um FWHM beam spot was scanned across the surface of a pixel to produce a charge sharing map in the planar and 3D pixels as a function of incident photon position. The 3D detector demonstrated superior charge sharing properties. Additionally, the 3D detector was used to image diffraction rings from powdered silicon, demonstrating the detector's use in a synchrotron experiment.

Synchrotron Tests of 3D Medipix2 and TimePix X-Ray Detectors

255

N38-3:

R. P. Rassool1, B. Sobott1, D. Peake1, V. Lee1, A. Schubert1, M. J. Boland2, N. Kirby2, A. Hawley2 1 School of Physics, University of Melbourne, Melbourne, Australia 2 Accelerator Physics, Australian Synchrotron, Melbourne, Australia Direct photon-counting hybrid-pixel detectors feature several advantages when compared with the current state-of-the-art CCD and imaging plate detectors. One of the key advantages of the Pilatus II detector is the ability to electronically gate the acquisition system. We have investigated the potential of sub-100ns single-frame acquisition to obtain diffraction data, synchronised with an external source. The time structure of synchrotron radiation from bunches of high-energy electrons circulating in a storage ring can be exploited for studies of fast kinetic processes involving short-lived excited species in bulk matter. Excellent agreement with the known fill-pattern in the ring has confirmed the accuracy of the obtained images and the reliability of the electronic synchronisation. A detailed simulation of the charge propagation in the silicon sensor has been performed using DESSIS and confirms the viability of reliable charge collection within the specified digital gating times.
N38-4:

The Dynamic Capabilities of the Pilatus II Hybrid Pixel Detector

G. A. Carini1, A. Dragone2, W. Chen1, J. Fried1, A. Kuczewski1, Z. Li1, J. Mead1, P. O'Connor1, P. Rehak1, K. Wolniewicz1, D. P. Siddons1, J.-F. Pratte3 1 Brookhaven National Laboratory, Upton, NY 11973, USA 2 SLAC National Accelerator Center, Menlo Park, CA 94025, USA 3 Universite' de Shrerbrooke, Sherbrooke, Quebec, Canada The detector for the X-ray Pump Probe (XPP) instrument at the Linac Coherent Light Source (LCLS) is being developed at Brookhaven National Laboratory. Ultrashort (~100fs at 120Hz) pulses from the LCLS will be used to probe the transient state of matter excited by using a fast optical laser. Depending on the experiment, X-ray absorption and emission or X-ray scattering will be recorded, requiring a 2D integrating detector. The 1024 x 1024 pixels detector is highly efficient at the energy of interest (8keV). It has ~100% fill factor, low noise, millisecond readout, single photon sensitivity and a dynamic range of more than 104 photons per pixel per frame, that is all the characteristics to fulfill the challenging requirements of the XPP instrument. The sensing element is formed by four square matrices with 512 rows and 512 columns. Each pixel of the matrix is 90 x 90 m2 and contains a JFET switch to control the charge readout. The final design of the detector and preliminary test results of the complete system will be presented.
N38-5:

The XAMPS Detector for the XPP Instrument at LCLS

J. W. Keister1, J. Smedley2, T. Rao2, E. Muller2, E. Wang3, X. Chang3, Q. Wu3, I. Ben-Zvi3, J. Distel4, D. Dimitrov5, R. Busby5 1 NSLS-II, Brookhaven National Laboratory, Upton, NY, USA 2 Instrumentation, Brookhaven National Laboratory, Upton, NY, USA 3 Collider-Accelerator, Brookhaven National Laboratory, Upton, NY, USA 4 Los Alamos National Laboratory, Los Alamos, NM, USA 5 Tech-X, Boulder, CO, USA Diamond is a unique material for x-ray energy conversion. Its high thermal conductivity and low coefficient of thermal expansion make it ideal for high heat load environments. However, its large bandgap, while offering insensitivity to visible light, makes charge trapping more likely (thermal detrapping less likely) than in silicon; energy conversion efficiency is also 3-4 times less even under the best conditions. High material strength and x-ray transmission also are potentially useful features for certain applications in x-ray science. Limitations to charge collection efficiency such as recombination and charge trapping have been investigated quantitatively using quasi-continuous tunable synchrotron radiation under flexible biasing schemes as well using detailed Monte Carlo Simulations. In the case of charge collection efficiency, the magnitude of the applied field, initial particle energy, and probe depth are adjusted. The diffusion and drift of photo-generated charge clouds are explicitly considered for the specific energy loss behavior of diamond. While recombination loss at the entrance window of diamond diodes is qualitatively similar to a treatment for an additional "dead" carbon window layer, the observed field and photon energy dependence implies that the more sophisticated model is more correct quantitatively. In addition, charge propagation in diamond is unique in that photoconductive gain is possible. Effectively, charge trapping of one carrier leads to screening of the applied field. In order to avoid photoconductive gain, either blocking contacts or explicit detrapping is required. Quantitative analysis of photoconductive gain as a function of applied field, x-ray power, waveform and photon energy offers insight into the fundamental limitations of state of the art single crystal diamond. Simple models are proposed to assist in extrapolating the observed behavior towards useful detector devices.

Charge Collection and Propagation in Diamond X-Ray Detectors

256

N39: Computing and Software for Experiments V: Bio-medical Software Thursday, Oct. 29 N39-1: 10:30-12:30 Grand Ballroom 3

P. Cerello1, F. Bagagli2,3, R. Bellotti4,5, L. Bolanos1,6, P. Bosco1,7, N. Camarlinghi2,3, C. Carpentieri2,3, R. Cataldo8,9, E. Catanzariti10,11, S. C. Cheran1, A. Chincarini12, F. De Carlo4,5, I. De Mitri13,9, G. De Nunzio8,9, M. E. Fantacci2,3, E. Fiorina7, G. Gargano4,5, G. Gemme12, I. Gori3,14, E. Lopez Torres6, A. Massafra13,9, R. Megna4,5, M. Quarta15,9, A. Retico2, S. Tangaro5 1 Sezione di Torino, INFN, Torino, Italy 2 Sezione di Pisa, INFN, Pisa, Italy 3 Dipartimento di Fisica, University of Pisa, Pisa, Italy 4 Dipartimento di Fisica, University of Bari, Bari, Italy 5 Sezione di Bari, INFN, Bari, Italy 6 CEADEN, Havana, Cuba 7 Dipartimento di Fisica Sperimentale, Univeristy of Torino, Torino, Italy 8 Dipartimento di Scienza dei Materiali, University of Salento, Lecce, Italy 9 Sezione di Lecce, INFN, Lecce, Italy 10 Dipartimento di Scienze Fisiche, University of Napoli, Napoli, Italy 11 Sezione di Napoli, INFN, Napoli, Italy 12 Sezione di Genova, INFN, Genova, Italy 13 Dipartimento di Fisica, University of Salento, Lecce, Italy 14 BRACCO Imaging, Milano, Italy 15 Dipartimento di Matematica, University of Salento, Lecce, Italy Lung cancer is the leading cause of cancer-related mortality in developed countries. Only 1015% of all men and women diagnosed with lung cancer live five years after diagnosis. However, the 5 years survival rate for patients diagnosed in the early asymptomatic stage of the disease can reach 70%. Early-stage lung cancers can be diagnosed by detecting non-calcified small pulmonary nodules with Computed Tomography (CT). Computer-Aided Detection (CAD) could support radiologists in the analysis of the large amount of noisy images generated in screening programs, where low-dose and thin-slice settings are used. The MAGIC-5 project, funded by the Istituto Nazionale di Fisica Nucleare (INFN, Italy) and Ministero dellUniversit e della Ricerca (MUR, Italy) developed a multi-method approach based on three CAD algorithms to be used in parallel with a merging of their results: the Channeler Ant Model (CAM), the Dot-Enhancement/Pleura Surface Normals/VBNA (DE-PSN-VBNA), the Region Growing (RG). The project is now entering the optimisation stage: the complete CAD system will make use of a selection and logical combination of the three modules. All the CAD algorithms were developed by means of CT images database made available by the Pisa centre of the Randomized Controlled Trial ITALUNG-CT and tested both on an independent dataset of 20 CT extracted from this database and on the 5 annotated CTs made available in the framework of the ANODE09 competition [1].
N39-2:

The MAGIC-5 CAD for Nodule Detection in Low Dose and Thin Slice Lung CT

A. Kimura1, S. Tanaka2, K. Hasegawa3, T. Sasaki4 1 Department of Electrical and Electronic Engineering, Ashikaga Institute of Technology, Ashikaga, Tochigi, Japan 2 Department of Media Technology, Ritsumeikan University, Kusatsu, Shiga, Japan 3 Ritsumeikan Global Innovation Research Organization, Ritsumeikan University, Kusatsu, Shiga, Japan 4 Computing Research Center, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan Monte Carlo simulation has been widely used for radiological simulation. In radiotherapy simulation, it applies to calculation of the dose distribution, evaluation of the beam delivery system, and so on. Geant4 toolkit is utilized to build a simulator for that purpose. Visualization tools for Geant4 are developed in order to display the complex patient image and the dose distribution calculated in such simulator. The visu-alization tools are consists of a volume visualizer and a volume visualization driver of Geant4. The volume visualization driver extracts a patient data, calculated dose distributions, particle trajectories and geometrical information of beam delivery modules in the simulation, and then saves their information as a data file. The primitive scorer of Geant4 is supported to obtain calculated physical quantities that are energy deposit, number of tracks, and so on. With the data file, the external volume visualizer is able easily to display the simulation results.

Visualization for Volume Data Scored by Geant4 Simulation

257

N39-3:

V. F. Cassola1, G. Hoff2 1 Nuclear Energy Department, Federal University of Pernambuco, Recife, PE, Brazil 2 Physics College, Pontifical Catholic University in Rio Grande do Sul, Porto Alegre, RS, Brazil

Monte Carlo Study of Glandular Dose Distribution on Radiological Procedures for Woman Who has Breast Silicone Implant

The silicone gel implant for breast have been, more and more, a common procedure among women of different ages around the world. That way it is important to know how it can affect the dose on sensitive tissues closer to the silicon implante. On this work we are studying the influence on absorbed dose distribution map into the patient, specially on tissues close to the silicone implant for diagnostic radiological procedures comparing a map of dose distribution for voxel model simulating patients with and without breast silicone implant. So three models of woman trunk considering were built considering the breasts with and without silicon implant. The three developed models were introduced into Geant4 code (version 9.2.p01) and were simulated three different parallel beam sources: frontal, lateral and rotational. The photons energy were sorted considering the distribution that represents the x-radiation spectra emitted by a tungsten target and considers additional filtration of 3mmAl at difference peak tensions. For all the studied cases the simulations considering the anterior-posterior exposures positioning of silicone implant produce an significant increase on glandular dose. For peak tension of 80kVp was calculated having a mean increase on glandular dose of 12% for the both breast simulated with silicone implante over and under the pectoral muscle. This difference decrease when the kVp increase been registered a mean value of 4,5% for the peak tension of 120kVp. There was no significant difference for the absorbed glandular dose calculated for the lateral positioning exposure, been presented differences always under 3,5% for 80kVp spectra. For higher kVp the differences presented were inferior than the statistical fluctuation due the simulation process. As conclusion the simulations realized in this work shown that the presence of silicone implante can, considering some cases, increase significantly the glandular dose due the backscattering.
N39-4:

T. Aso1, T. Yamashita2, T. Akagi2, T. Sasaki3 1 Computer Engineering, Toyama National College of Maritime Technology, Toyama, Japan 2 Accelerator Management, Hyogo Ion Beam Medical Center, Hyogo, Japan 3 Computing Research Center, High Energy Accelerator Research Organization, Ibaraki, Japan

Pencil Beam Algorithm on the Geant4 Based Simulation Framework and Verication of the Dose Distributions for Proton Therapy

Geant4 Based Monte Carlo simulation has been widely accepted for dose calculation in medical physics. In our previous work, we had developed a Geant4 based simulation framework for particle therapy system for simulating various irradiation systems. The framework realizes comparing simulated results and sharing the best physics models between different facilities. The performances had been evaluated in terms of not only dose distributions but also computational efficiency. In order to achieve statistically reliable dose calculations in reasonable time, the calculation time is still critical for the use of daily clinical routine. On contrast to MC, a PBA completes dose calculation within five minutes or less. This is suitable to get a quick overview of the dose distributions. The main uncertainty comes from the influence of Multiple Coulomb Scattering (MCS) and energy struggling on the trajectories of protons traversing complex mediums with significant density gradients, where a MC is the best solution for realistic estimation. In this work, a pencil beam algorithm has been introduced in the framework, in order to compare dose distributions with Monte Carlo simulation. The framework realizes direct comparisons of dose distributions by the PBA and the MC.In addition, the framework is able to perform as a hybrid dose calculation by connecting both schemes. The dose distributions by Geant4 and the pencil beam algorithm were compared in homogeneous water phantom and CT images of a sample patient, respectively, using an irradiation parameters at Hyogo Ion Beam Medical Center (HIBMC). The dose distributions were in good agreement but small disagreements were seen.
N39-5:

M. Huang1, G. E. Sjoden1, A. K. Al-Basheer2, M. Ghita3, W. Bolch1 1 Nuclear & Radiological Engineering, University of Florida, Gainesville, Florida, U.S.A. 2 Georgia Radiation Therapy Center, Medical College of Georgia, Augusta, Georgia, U.S.A. 3 Radiology, University of Florida, Gainesville, Florida, U.S.A.

Optimization of Electron Dose Kernels in Heterogeneous Voxelized Phantoms

In this paper, we present work on the optimization of the EDK-SN, or Electron Dose Kernel Discrete Ordinates method to compute doses and account for material heterogeneities using high energy external photon beam irradiations in voxelized human phantoms. This new EDK-SN methodology was developed by our research team, and employs CT-scanned voxelized patient phantoms, wherein doses are computed in a two-step process by projecting precomputed doses from Monte Carlo pencil beamlets projected and scaled using rapid, high fidelity discrete ordinates photon transport computations. This technique was successfully demonstrated as an accurate alternative to traditional full-physics Monte Carlo photon-electron transport calculations, but EDKSN offers a unique advantage of rapid, accurate whole body computation. The final paper will investigate the efficacy of different

258

mesh sizes and accuracy goals for the EDK-SN procedure, along with new algorithms to more accurately account for density heterogeneities in voxelized patient phantoms.
N39-6:

J. Spiga1,2,3, Y. Prezado1, E. Bruer-Krisch1, V. Fanti2, P. Randaccio2, A. Bravin1 1 European Synchrotron Radiation Facility (ESRF), Grenoble, France 2 Department of Physics, University and INFN of Cagliari, Monserrato, Italy 3 School of Environmental Sciences, University of East Anglia, Norwich, United Kingdom

The Effect of Beam Polarization in Microbeam Radiation Therapy (MRT): Monte Carlo Simulations Using Geant4

Microbeam Radiation Therapy (MRT) is an innovative experimental technique potentially able to overcome the limitations of conventional radiotherapy for infantile brain tumors. Its effectiveness seems to be related to the ability of normal tissues to tolerate a very high radiation dose in small volumes, resulting in the preservation of the tissues architecture. The effectiveness of MRT is well represented by peak-to-valley dose ratios (PVDRs), which are one of the crucial parameters associated with the outcome of the treatment. We present Geant4 Monte Carlo calculations of the dose distribution deposited by planar polarized microbeams at micrometric resolution. The simulation of the beam polarization, made possible by different libraries included in Geant4, is a crucial step in enhancing the comparability of experimental data and simulation results.
N39-7:

M. G. Pia1, E. Gargioni2, B. Grosswendt3, R. Schulte4, A. Wroe4, M. Begalli5, P. Queiroz6, D. de Souza Santos6 1 INFN Genova, Italy, Genova, Italy 2 University Medical Center Hamburg-Eppendorf, Hamburg, Germany 3 PTB, Braunschweig, Germany 4 Loma Linda University, Loma Linda, USA 5 State University, Rio de Janeiro, Brazil 6 IRD, Rio de Janeiro, Brazil

Re-Engineering a Nanodosimetry Monte Carlo Code into Geant4: Software Design and First Results

There is increasing evidence that the pattern of radiation interaction on the nanometer level is critical for the biological effects of ionizing radiation; in addition, radiation effects at the nano-scale are important for the protection of electronic devices operating in various radiation environments. In realistic use cases such small-scale systems are often embedded in larger scale ones: for instance, a component may operate within a HEP experiment or on a satellite in space, cellular and sub-cellular aggregates in real biological systems exist in complex body structures etc. The separation of the two simulation domains is due to the conceptual and technical difficulty of handling the two schemes in the same simulation environment. Achieving a conceptual approach and an architectural design where the two schemes can co-work would represent a fundamental progress in Monte Carlo simulation. Recently, a set of specialized processes for track structure simulation in liquid water has been designed and implemented in Geant4; like their equivalents in dedicated Monte Carlo codes, they operate in the rgime of discrete interactions only. As a further step towards endowing Geant4 of capabilities for simulation at the nano-scale, the so-called PTB Monte Carlo for nanodosimetry simulation is being re-engineered into Geant4. This extension of Geant4 capabilities is part of a larger scale R&D project for multi-scale simulation capable of adaptable, co-working condensed and discrete transport schemes. The project in progress reengineers physics modeling capabilities of the track-structure code originally developed at PTB to into a design suitable to collaborate with an object oriented simulation kernel; the availability of related experimental measurements from the collaborating groups allows the validation of the prototype code in a realistic experimental environment. The first results of the ongoing developments and application are presented.
N40: Photodetectors and Scintillation Detectors III Thursday, Oct. 29 N40-1: 13:30-15:30 Grand Ballroom 1

Tha NA62 RICH Detector

M. Lenti INFN - Sezione di Firenze, Firenze, Italy On behalf of the NA62 Collaboration The CERN NA62 experiment aims to measure the ultra-rare charged kaon decay into one charged pion and two neutrinos with a 10% accuracy. For the reduction of the background from the decay of the kaon in muon-neutrino, a neon-filled, RICH detector is being built, to have a good pi-mu separation between 15 and 35 GeV/c. The project will be described, together with the results from a test of a prototype equipped with 440 photomultipliers.

259

N40-2:

Search for Scintillation in Doped Lead Fluoride Crystal

R. Mao, L. Zhang, R.-Y. Zhu HEP, California Institute of Technology, Pasadena, USA An effort has been made to introduce scintillation light in lead fluoride crystals by selective doping. It was found that some rare earth ions introduced in the crystal lattice may serve as luminescence centers and make the crystal scintillating. The photo- and X- luminescence spectra, the decay time constants and the light output of these doped samples are presented. The final product of this development program may be used to build a homogeneous hadronic calorimeter with dual readout for future high energy physics experiments.
N40-3:

T. Abe1, H. Aihara1, M. Tanaka2, Y. Kawai3, H. Kyushima3, M. Suyama3 1 Physics, University of Tokyo, Bunkyo-ku, Tokyo, Japan 2 Institute for Particle and Nuclear Studies, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan 3 Electron Tube Division, Hamamatsu Photonics K.K., Iwata City, Shizuoka, Japan We report R&D status of large aperture Hybrid Avalanche Photo-Detector (HAPD). We have developed 13-inch aperture HAPD. HAPD is expected as a photo detector to replace photo-electron multiplier tube (PMT) in the next generation water Cherenkov detectors such as Hyper Kamiokande. We present the excellent performances of HAPD (~200ps time resolution and clear photoelectron separation up to 6) and the readout system dedicated for HAPD to draw out basic characteristics of HAPD. New high voltage system and 8-inch aperture HAPD are also discussed.
N40-4:

Development of Large Aperture Hybrid Avalanche Photo-Detector

An 18 Element Strip of 1-mm2 G-APDs for CMS HCAL Upgrade in a CERN Test Beam

A. H. Heering Department of Physics, Boston University, MA, US On behalf of the CMS HCAL collaboration The Compact Muon Solenoid hadronic barrel (CMS HB) calorimeter consists of 17 layers deep of scintillator in a brass absorber. Each layer of scintillator consists of 16 tiles representing one layer of the 16 towers in eta in the CMS coordinate system. Each tile has one 0.96 mm wavelength shifting fiber. In the current design we physically add 17 fibers from each layer or one full tower into one pixel of a multi-channel hybrid photo-diode (HPD). Because the HPD has only 19 pixels and reads out 16 towers, most of the depth or longitudinal shower information is lost. In the last year we have developed 18 element MAPD arrays (Micro-pixel Avalanche Photo Diodes) on 8 inch wafers with Zecotek in Singapore. A special prototype 20 pin ceramic package was made by Kyocera in Germany. By using 18 element G-APD strips (Geiger mode Avalanche photo diodes) we could readout each layer separately, so that 17 of these strips replace one HPD. Because of the large linear range requirement in the hadronic calorimeter and radiation hardness needed, we used the new generation high cell density Zecotek MAPDs with 14,000 cells per strip element [3]. Due to the higher PDE (photon detection efficiency) versus the HPD (30% vs. 12%) we got 5-7 p.e. per layer for a MIP. Because of the limited amount of electronic channels a special study was done analogically adding single elements of different strips. Non uniformity between strips was calibrated using muons and single photo-electron calibration. A third very powerful calibration tool in the HCAL is a 2 mCi Cs137 wire source that can run by each separate calorimeter tile. By looking at the DC current, due to the presence of the radioactive source moving by each tile, a total calibration was obtained for every element of each MAPD array. After calibration of the now 4 longitudinal depth calorimeter, the linearity and detector resolution was studied using pions with energy of 2 to 300 GeV interacting in multiple towers.
N40-5:

Advanced PMTs for the CTA Project

R. Mirzoyan Max-Planck-Institute for Physics, Munich, Germany On behalf of the Focal Plane Instrumentation WP of the CTA Consortium Currently the best classical PMTs with semitransparent bialkali photocathode show peak quantum efficiency (QE) of ~ (3235) %. Although the above-mentioned peak QE was demonstrated on single samples already ~ 40 years ago, one could not report any significant improvement until recently. A few years ago we have started a development program with the main PMT manufacturers Photonis, Electron Tubes and Hamamatsu, aiming to boost-up the peak QE of the (1 - 2)" size bialkali PMTs. Today we know that these efforts were successful: all of the three above-mentioned companies succeeded to boost the peak QE of bialkali PMTs to the level of (3036) %, that subsequently got the name super-bialkali [1,2]. Moreover, the Hamamtsu and Photonis succeeded to improve the QE even higher up and PMT samples with QE in excess of 40 % (these became known as ultra-bialkali type) became available. The Cherenkov Telescope Array (CTA) is planned as a next generation ground-based large instrument (after CANGAROO, VERITAS, H.E.S.S. and MAGIC) for very high energy gamma ray astrophysics. The core of the 260

CTA collaboration consists of the MAGIC and the H.E.S.S. collaborations that will merge. Also, a large number astrophysicists, mostly from European institutions join the newly setup CTA. The aim is to build an array of about 100 imaging telescopes of two or three sizes (small size: ~ 6-8m, middle size: ~ 10-12m class, large size: ~ 17-23m class) that shall provide ~10 times higher sensitivity than the currently existing telescopes. Each telescope may apply an imaging camera consisting of 1000-2000 PMTs. Currently we are starting a PMT development program with Hamamatsu (Japan) and Electron Tube Enterprises (England) aiming to develop PMTs of advanced characteristics for the CTA project (unfortunately Photonis stopped manufacturing PMTs).
N40-6:

MPPC Response Simulation and High Speed Readout Optimization

F. Retiere TRIUMF, Vancouver, BC, Canada Pixilated Geiger Mode avalanche photo-diodes and especially Hamamatsu Multi-Pixel Photon Counters (MPPC) are replacing photomultiplier tubes (PMTs) in a variety of applications. However, to become a competitive alternative to PMTs, a number of drawbacks must be overcome or accommodated: small size, lower gain (roughly a factor of 10 smaller) , large dark noise (0.5-1 MHz) and significant after-pulsing (5-20%) . While the small size has to be accommodated, the other nuisances can be dealt with when designing the readout electronics. To do so, we rely on a Monte Carlo simulation of the MPPC response developed for the T2K experiment. We focus on two applications that require excellent timing resolution: plastic scintillator readout for muon spin rotation experiments and LSO readout for Positron Emission Tomography. For the latter application, accurate simulation of the MPPC recovery is critical to understand the LSO-MPPC response because the LSO time constant is long (40 ns) compare to the MPPC intrinsic recovery (13 ns for T2K MPPC) time constant. We will show that by using the simulation information to design high speed readout electronics, we achieved excellent energy resolution (limited by LSO for PET) and timing resolution (<500 ps for PET).
N40-7:

Evaluation of Silicon Photomultiplier Arrays for the GlueX Barrel Calorimeter

C. Zorn Radiation Detection and Medical Imaging Group, Jefferson Laboratory, Newport News, VA, USA On behalf of the GlueX Collaboration The first prototype silicon photomultipliers suitable for use as the photodetector for the GlueX barrel calorimeter have been delivered and are being evaluated. These detectors are in the form of a 4x4 array of closely-packed 3x3 mm^2 detector elements. The next generation (to be delivered in the near future) will be encapsulated within a ceramic base that will allow for temperature control via an onboard peltier cooler. This will allow one to cool the sensor directly both in order to reduce the dark noise and to maintain a constant temperature so as to keep the gain stable. PDE and dark rate measurements of the two currently considered vendors indicate cooling will be necessary with one and temperature maintenance with either. A possible third vendor may also be evaluated if suitable array-style photodetectors can be delivered.
N41: Semiconductor Detectors IV: CdTe and other Wide Band Gap Materials Thursday, Oct. 29 N41-1: 13:30-15:30 Grand Ballroom 2

F. H. Ruddy1, J. G. Seidel2, P. J. Sellin3 1 Ruddy Consulting, Mt. Pleasant, SC, USA 2 Science & Technology, Westinghouse Electric Company, Pittsburgh, PA, USA 3 Department of Physics, University of Surrey, Guildford, GU2 7XH, UK

High-Resolution Alpha Spectrometry with a Thin-Window Silicon Carbide Semiconductor Detector

The potential for high-resolution alpha-particle energy spectrometry in high-temperature, high-radiation environments using thinwindow SiC radiation detectors has been demonstrated. 238Pu alpha-particle peaks separated by only 42.7 keV have been completely resolved using a SiC Schottky detector with a 400 titanium Schottky contact (entrance window). The observed FWHM for the 238Pu 5499.2-keV alpha particle is 20.9 keV or 0.37%. Factors affecting the observed alpha-particle energy resolution in SiC detectors will be discussed.

261

N41-2:

M. Fiederle1, A. Fauler1, A. Zwerger1, D. Greiffenberg1, P. Siffert2 1 FMF, Freiburger Materialforschungszentrum, Freiburg, Germany 2 Eurorad SA, Strasbourg, France

CdTe Semiconductors: Development and Evaluation of Pixel Detectors

CdTe based semiconductors are good materials for the development of high efficient X- and Gamma-ray detectors. The recent improvements of CdTe crystals and the availability of photon counting readout electronics open the possibility of processing CdTe pixel detectors with small pixels down to 55 m and high efficiency for X-ray energies above 20 keV. These small pixels make a strong demand for the material quality of the semiconductor crystals concerning homogeneity and performance. Different types of pixel detectors have been produced using our own technology with small pixels down to 55 m and high connection density up to 65,000 pixels. The detector performance of the assemblies was tested with different radiation sources and x-ray tubes. Detectors with a size of 14x14 mm and 28x28 mm were developed using the Medipix2 electronics. The pixel detectors showed a good efficiency and homogeneity of the signal response. The numbers of working pixels was above 98 percent. These small pixels are a powerful tool for the investigation of material properties on a very small scale of a few micrometers. The CdTe wafers have been extensively characterized before the processing. The Contactless Resistivity Mappings CoReMa yield a high resistivity above 5 GOhm-cm. The identification of second phase defects like Te inclusions have been done using Infra-Red Microscopy. Te inclusions with a size smaller than 15 m have been found. We will discuss the good performance of the pixel detectors regarding the concentration of defects and inhomogeneities of the used crystals.
N41-3:

S. Watanabe1,2, S.-N. Ishikawa1,2, H. Odaka1,2, S. Saito1,2, T. Fukuyama1,2, S. Sugimoto1,2, G. Sato1, M. Kokubun1, T. Takahashi1,2 1 Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency (JAXA), Sagamihara, Kanagawa, Japan 2 Department of Physics, University of Tokyo, Hongo, Tokyo, Japan We have developed CdTe diode double-sided strip detectors (DSDs) and studied their spectral and imaging performance for Xrays and gamma-rays. The double-sided strip configuration provides a fine position resolution and a large detection area with a relatively small number of the readout channels. Therefore, CdTe DSDs are very attractive for gamma-ray imaging and spectroscopy. We have developed CdTe diode double-sided strip devices having Pt cathode strips and Al anode strips, and assembled prototype CdTe DSDs. These prototypes have a strip pitch of 400 m. Signals from the strips are processed with analog ASICs (application specific integrated circuits). We have successfully performed gamma-ray imaging spectroscopy with a position resolution of 400 m. Energy resolution of 1.8 keV (FWHM: full width at half maximum) was obtained at 59.54 keV for 0.5 mm thick CdTe DSDs. The possibility of improved spectral performance by utilizing the energy information of both side strips was also demonstrated. Recently, 2 mm thick CdTe DSDs for higher energy gamma-rays and CdTe DSDs with a fine strip pitch of 60 m have become available. In this paper, we present such recent results of our CdTe diode DSDs. Moreover, we also discuss the gamma-ray response characteristics of CdTe DSDs.
N41-4:

Recent Results of CdTe Diode Double-Sided Strip Detectors

A. G. Kozorezov1, A. Owens2, F. Quarati2, A. Webb3 1 Physics, Lancaster University, Lancaster, UK 2 Advanced Studies and Technology Preparation Division Scientific Projects Department, European Space Agency, Noordwijk, The Netherlands 3 HasyLab, DESY, Hamburg, Germany We present the results of a detailed experimental study of polarisation effects in TlBr planar X-ray detectors. The experiments were carried out at the HasyLab synchrotron radiation facility by scanning a focused X-ray beam across the detector. Measurements were carried out in the range 10-100keV. Above 50 keV detector response shows a time-dependent degradation of its spectroscopic qualities: peak channel position , energy resolution and count rate. We have examined spectroscopic degradation in more detail by acquiring consecutive spectra at a fixed photon beam position within a 1 hour time domain at ~1 kcts/s and at 60 keV. Using the pump-and-probe technique we studied the spatial extent of polarisation caused by the energy deposition at a fixed spot. Fixing the beam at a chosen spot until a strong deterioration of response occurs and subsequently spatially scanning across the affected area at a much lower count rate we observe the size of the affected area. The results are : 1) the affected area is much larger than the beam size, and 2) recovers very slowly. We have developed theoretical model of non-stationary polarisation phenomena. We describe polarisation effects as a result of the charging the traps by both electrons and holes. We calculate the charge collection efficiency and the exact line shapes as functions of exposure time, count rates and photon energy. We study the onset of polarisation effects and its evolution with time until it results in stationary profiles. Comparing model simulations with experimental results we obtain important parameters characterising material quality.

Polarisation Effects in Thallium Bromide X-Ray Detectors

262

N41-5:

L.-J. Meng1, J. W. Tan1, K. Spartiotis2, T. Schulman2 1 Nuclear, Plasma and Radiological Engineerng, University of Illinois at Urbana-Champaign, Urbana, IL, USA 2 Tietotie 3, FIN-02150, Oy AJAT Ltd, ESPOO, FINLAND

Development of a Novel Energy-Resolved Photon-Counting Detector for Gamma Ray Imaging Applications

In this paper, we present the design and preliminary performance evaluation of a novel energy-resolved photon-counting (ERPC) detector for gamma ray imaging applications. The prototype ERPC detector has an active area of 4.4 cm x 4.4 cm, which is pixelated into 128 x 128 square pixels with a pitch size of 350 um x 350 um. The current detector consists of multiple detector hybrids, each with a CdTe crystal of 1.1 cm x 2.2 cm x 1 mm, bump-bonded onto a custom-designed application-specific integrated circuit (ASIC). The ERPC ASIC has 2048 readout channels arranged in a 32x64 array. Each channel is equipped with pre- and shaping-amplifiers, a discriminator, peak/hold circuitry, and an analog-to-digital converter (ADC) for digitizing the signal amplitude. In order to compensate for the pixel-to-pixel variation, two 8-bit DACs are implemented into each channel for tuning the gain and offset. The ERPC detector is designed to offer a high spatial resolution, a wide dynamic range of 12-200 keV and a good energy resolution of 3-4 keV. The hybrid detector configuration provides a flexible detection area that can be easily tailored for different imaging applications. The intrinsic performance of a prototype ERPC detector was evaluated with various gamma ray sources, and the results are presented.
N41-6:

G. Sato1, T. Kishishita1,2, H. Ikeda1, T. Sakumura3, T. Takahashi1,2 1 ISAS / JAXA, Sagamihara, Kanagawa, JAPAN 2 Department of Physics, University of Tokyo, Bunkyo, Tokyo, JAPAN 3 RIGAKU Co., Ltd., Akishima, Tokyo, JAPAN

Development of Hybrid CdTe Pixel Detectors with Low-Noise Front-End ASICs

We report on the recent fabrication of CdTe pixel detectors with our low-noise front-end ASICs. The ASICs are designed based on the Open-IP LSI project led by JAXA, and implemented by TSMC 0.35 m CMOS technology. The readout chips consist of matrices of identical 270 m x 270 m pixel cells each containing a charge-sensitive amplifier, pole-zero cancellation circuit, low-pass filter, comparator, and peak/bottom-hold circuit. Each readout channel is connected to its corresponding detector pixel using bump-bonding technique. The first version of the hybrids contains 4 x 4 pixels. The whole CdTe sensor area is 1.3 x 1.3 mm2 with the thickness of 0.65 mm. We present the basic spectral performance including pixel-to-pixel crosstalk, digital-toanalog interference, as well as charge sharing effect.
N41-7:

Preliminary Results on the Growth of Mercuric Iodide Epitaxial Crystals for Use as Radiation Detectors

M. R. Saleno, L. van den Berg, R. D. Vigil, J. L. Baker Constellation Technology Corp, Largo, FL, USA In its application as a radiation detector, mercuric iodide shares a common challenge amongst a number of competing semiconductor materials - it has been too expensive to grow and fabricate devices to be viable in the larger commercial market. Constellation Technology Corporation has developed a process for the growth of mercuric iodide homoepitaxy. Epitaxial layers with dimensions of 20mm x 20mm and a thickness of up to 9mm have been grown. These layers have been fabricated into both x-ray and gamma ray planar detectors and results have been obtained on 121 pixel (11x11) pixellated detectors. A typical ampoule-grown crystal takes up to 3 months to grow, whereas detector-size layers have been grown with the epitaxial process in just a few days. Additionally, the bulk of the epitaxial growth is in the form-factor of the final, fabricated detector. Therefore, production time and material waste (hence cost) are substantially reduced. Improvements in the mercuric iodide growth process employing epitaxial methods will be described and the results discussed in this paper.
N42: Data Acquisition and Analysis Systems II Thursday, Oct. 29 N42-1: 13:30-15:30 Grand Ballroom 3

Neutron Scattering Data Acquisition and Control Upgrades at the Lujan Center

P. S. Lewis, D. S. Barr, G. M. Cooper, K. L. Knickerbocker, A. H. Shapiro, F. R. Trouw LANSCE-Lujan Center, Los Alamos National Laboratory, Los Alamos, NM, USA The Lujan Center at the Los Alamos National Laboratorys Neutron Science Center (LANSCE) is a neutron source where research in materials, physics, and biological sciences is conducted on time-of-flight neutron scattering spectrometers. Execution of an experiment on a Lujan Center neutron spectrometer involves measurement of the neutrons scattered from the sample and 263

the control of the associated sample environment equipment. The present data acquisition and control architecture was developed in the late 1990s and put into production in 2001. We have recently reevaluated our present systems and architecture and embarked on upgrades in several key areas. These upgrades include moving from a Windows/VxWorks data acquisition environment to a uniform Linux-based environment for both host and distributed processors. We abandoned our previous higherlevel xUML-based software development methodology and moved to a more conventional approach based on open source tools. Finally, we have begun to migrate the control software for our sample environment equipment from Labview to EPICS to take advantage of the stability and networking integration it provides. Results to date have been very positive. The initial release of our new LDAC data acquisition software was used successfully on three instruments during our 2008 run cycle. We will start up in 2009 with it running on five instruments and then plan a phased roll out to the remaining instruments over the next year. Our efforts on migration of sample environment controls to EPICS are in a more preliminary stage, focused on the more common and widely used pieces of equipment such as furnaces and displexes. This migration has been aided by the recent advent of improved low end tools and platforms (e.g. softIOC) for EPICS.
Los Alamos National Laboratory is operated by the Los Alamos National Security, LLC, for the National Nuclear Security Administration of the U.S. Department of Energy under contract DE-AC52-06NA25396. N42-2:

GANDALF - Design of a High Resolution Transient Recorder for High Energy Physics

F. Herrmann University of Freiburg, Freiburg, Germany On behalf of the COMPASS Collaboration (CERN) The GANDALF 6U VME/VXS system has been designed to perform multiple electronic readout tasks in high energy physics experiments such as analog or time to digital conversions, coincidence matrix formation and fast trigger generation. As a transient recorder the GANDALF module can digitize eight analog channels with 1Gsps and has a maximum dynamic range of 14bit. Also 16 channels per module are possible at half sampling rate. The extraction of timing information is performed for both unipolar and bipolar signal shapes in a large dynamic range from a few mV up to 4V. A timing resolution of better than 50ps is achieved even for fast pulses with risetimes down to 3ns. The baselines of the 500MHz bandwidth analog input circuits are adjustable by 16bit DACs independently. Enormous numerical capabilities are provided by the implementation of a Virtex5-SXT FPGA to perform algorithms for dead-time free timing measurements and pulse shape integration. Pile-up pulses with a minimal time distance of twice the risetime can be disentangled by numerical algorithms. A fast and deep memory extension provided by QDRII+ (144Mbit) and DDR2 (4Gbit) devices buffer data for numerical processes and output. The modularity of the board design allows the implementation of a 128 channel TDC module. Here the time digitization process is done inside the V5 FPGA and reaches timing resolution of up to 125ps. The same hardware setup can be used to implement a fast 128 channel trigger matrix for complex trigger decisions. The VXS backplane bus combines the information of up to 18 GANDALF modules. Every module has 16 high speed lines for the data transfer to a central VXS switch module, which can be used to form fast global trigger decisions based on pulse shapes, time or logic patterns of the individual channels from all GANDALF modules in a single crate. Each GANDALF board has a variety of interfaces for physics data transfer, configuration and monitoring like USB 2.0, VME64x, S-Link and Ethernet.
N42-3:

The Current Performance and Upgrade Plans for the PHENIX Data Acquisition System

M. L. Purschke Physics Department, Brookhaven National Lab, Upton, NY, USA On behalf of the PHENIX Collaboration The PHENIX experiment at RHIC currently consists of 4 large spectrometer arms, two central arms and two forward Muon arms. There is a total number of about 500,000 readout channels, giving a typical event size of about 220KB/event for the largest collision systems. Over the recent years, we could significantly increase the data rate in the PHENIX data acquisition system, which is capable of rates of up to 700MB/s. The capabilities of the PHENIX data acquisition system and the triggers have continuously been improved and upgraded to keep pace with increased RHIC luminosities. Improvements have come from using event buffering in the Front End Modules (FEM's), code optimization and improved zero suppression in the Data Collection Modules (DCMs), and hardware and software improvements to the Event Builder (EvB). The data logging rates were reduced to the above levels by adding a comprehensive data compression scheme. The PHENIX collaboration is constructing 4 new detector systems in its upgrade program, in addition to 2 new trigger systems. The upgrades add a Hadron Blind Detector (HBD), a Forward Calorimeter (FoCal), and two silicon-based Vertex Detectors to the setup. The last new upgrade detector, the Forward Vertex detector, will be commissioned in 2011. This will more than double the channel count in the experiment, and at least double the per-event data size. A corresponding upgrade of the DAQ system is under way. The upgrade trigger systems consist of the instrumentation of the Muon Tracker with trigger capabilities to trigger on high264

momentum muon tracks. A resistive plate chamber (RPC) detector will add timing information, which will allow to eliminate beam-induced background from muons traversing the PHENIX acceptance, as well as background from cosmics. We will give an overview about the current detector and DAQ system, briefly describe the new detectors, and outline the road and new components for the upgrade of the DAQ.
N42-4:

A. B. Mann1, H. Angerer1, I. Konorov1, M. Kraemer1, S. Huber1, B. Grube2, J. Friedrich1, S. Paul1 1 Physics Department E18, TU Muenchen, Garching, Germany 2 Excellence Cluster Universe, TU Muenchen, Garching, Germany

The Universal Sampling ADC Readout System of the COMPASS Experiment

COMPASS is a high rate fixed target experiment located at the CERN SPS with physics programs using either a 160 GeV/c muon beam or a 190 GeV/c hadron beam. A large number of detector channels (>100000) are acquired using sampling ADC based electronics. A family of universal ADC modules was therefore developed to interface either to the APV25 analog frontend multiplexer ASIC or to sample the analog detector signals directly. The APV25 based readout interfaces to the silicon microstrip and GEM small area trackers and to a part of the RICH detector, based on MWPCs with CsI photocathodes. The direct sampling approach is implemented for the electromagnetic and hadronic calorimeters with currently 5728 channels which are continuously sampled at 80 MHz. The FPGA based data processing acquires therefore full signal waveforms for offline analysis upon a central trigger signal and implements also a first zero suppression. For the data taking period in 2009, the data processing of 1024 channels includes also the online evaluation of signal amplitudes and pulse time extraction to provide additional calorimeter energy information to the trigger system. We give an overview of the existing systems and ongoing development also for migration toward the ATCA crate standard. This work is supported by the BMBF, the Maier-Leibnitz-Labor Garching, the Cluster of Excellence Exc153 and the EU Integrated Infrastructure Initiative Hadronphysics Project I3HP (RII3-CT-2004-506078)
N42-5:

A Simple Method to Measure and Improve Linearity of Flash ADCs Used in Integrated VME ADC Modules

K. Furutaka, A. Kimura, M. Koizumi, Y. Toh, T. Kin, S. Nakamura, M. Oshima Innovative Nuclear Science Research Group, Japan Atomic Energy Agency, Tokai-mura, Ibaraki 319-1195, Japan To measure and improve differential linearity of flash ADCs used in integrated VME ADC modules, exponentially falling pulsetails were digitized and counts of the resultant digital codes were recorded. The obtained histograms were fit to a function to deduce smoothly varying curves which should correspond to that obtained in the case of ideal linearity, and were compared to the fit value to calculate code width for each code. The obtained results show zigzag shaped deviation of about 10 % in magnitude from unity in the whole range and some large spikes which are as large as 100 %. A simple method is proposed to correct the nonlinearity using the ratios in post-processors following the ADCs. In the conference, the results of the correction will be presented.
This study was the result of "Study on nuclear data by using a high intensity pulsed neutron source for advanced nuclear system" entrusted to National University Corporation Hokkaido University by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT). N42-6:

Ethernet Data Acquisition System for the Time Projection Chamber

V. J. Riot, M. D. Heffner Lawrence Livermore National Laboratory, Livermore, CA, USA Time Projection Chambers (TPC) are used in the field of particle physics to track particles from collisions produced in accelerators . The TPC is also a versatile detector that has been applied and considered for tasks such as discovering dark matter, looking for neutrino-less double beta decay, or the detection of special nuclear material. The downside to current TPC technology is the cost and size of the readout electronics. We describe here a low cost, high density, high speed data acquisition system suited for the TPC instrument developed in part by the Lawrence Livermore National Laboratory. A dual 3000 pad MicroMegas detector combined with a 6000 channel front-end analog electronics system feeding into a 62.5MHz digitizer and data processing system able to handle count rates well above 100KHz/channel and reporting data through gigabit Ethernet has been designed and prototyped. Simulation and preliminary testing is showing promises that this data acquisition approach is a good solution for the TPC instrument. We believe that this approach will allow the TPC technology to be more affordable and as such make it attractive for use in a broader range of applications.
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344

265

N42-7:

The New CALICE Data Acquisition System

M. Wing Department of Physics and Astronomy, University College London, London, UK On behalf of the CALICE-UK DAQ groups Within the CALICE collaboration, which is designing a calorimeter for the International Linear collider (ILC), a collection of UK groups (CALICE-UK) are part of the effort to prototype a highly-granular calorimeter composed of silicon and tungsten. The CALICE-UK groups have built a new data acquisition (DAQ) system for the next generation of technical prototypes to be built in the EUDET project. The system is of a generic, modular design using commercial components and so should be applicable to many detector sub-systems. Therefore, the DAQ system developed here may have consequences or applications high energy physics or any science requiring the readout of high data rates in general. In previous editions of the conference progress reports were given on the R&D towards such a system and now at this conference the final working system will be described.
N43: High Energy Physics Instrumentation III Thursday, Oct. 29 N43-1: 13:30-15:30 Grand Ballroom 7

Crystal Calorimeters in the Next Decade

R.-Y. Zhu Physics, California Institute of Technology, Pasadena, CA 91125, USA Crystal calorimeter has traditionally played an important role in precision measurement of electrons and photons in high energy physics experiments. Recent interest in calorimeter technology extends its application to measurement of hadrons and jets with dual readout. Potential application of new generation scintillating crystals of high density and high light yield, such as LYSO, in high energy physics experiments is described. Candidate crystals for the homogeneous hadronic calorimeter concept are also discussed.
This work is partially supported by DOE ADR DE-FG02-06ER41427. N43-2:

LuAG material for Dual readout calorimetry in future high energy detectors

E. Auffray, D. Abler, P. Lecoq, G. Mavromanolakis PH/CMX, CERN, Geneva, Switzerland One of the main challenges for detectors at future high energy collider experiments is high precision measurement of hadrons and jet energy and momentum. One possibility to enable such measurement is the particle flow approach that requires a complex highly segmented calorimeter system to identify and to track all particles in a jet. An alternative so-called dual readout approach consists of simultaneously recording, in an active medium, scintillation light that is proportional to total energy deposit and Cerenkov light that is proportional to the electromagnetic part only, thus allowing extracting the electromagnetic fraction of the total shower energy on an event by event basis. A promising candidate for dual readout calorimeter is Lutetium aluminium garnet Lu3Al5O12 (LuAG). Its high density, short radiation length and interaction length provide sufficient stopping power for hadronic calorimetry at high energy colliders. The fundamental absorption edge at 250nm combined with a very large optical transmittance window and a high refraction index (n=1.84) make LuAG an excellent Cerenkov radiator. Moreover cerium doped LuAG has good scintillating properties in terms of light yield and decay time. LuAG can be produced either in bulk crystal or as fiber crystal allowing to compare its performances in two different approaches of dual readout calorimetry. As bulk crystals Cerium doped LuAG can be used in the method based on the separation of scintillation and Cerenkov signal produced in homogeneous detector blocks. In fiber shape, it can be used in the method using so-called meta crystals consisting of both Ce doped and undoped LuAG crystal fibers. The undoped fibers behave as Cerenkov radiator, while Ce-doped fibers behave as scintillators. This talk reviews the properties of LuAG material in view of its use as scintillator both for bulk and fiber scintillators, and compares the two different approaches for dual readout calorimetry and presents first results obtained.
N43-3:

A Forward LYSO Crystal Calorimeter for the SuperB Project.

C. Cecchi, S. Germani University of Perugia and INFN Perugia, Perugia, Italy The SuperB project is an asymmetric e+e- accelerator of 10^36 cm^-2 s^-1luminosity capable of collecting a data sample of 5075 ab^-1 in five years of running. Development work on crystals suitable for the forward electromagnetic calorimeter, cerium doped lutetium orthosilicate (LSO) and lutetium yttrium orthosilicate (LYSO), with a a Molire radius of 2.1 cm is underway. Studies of the geometry have been made during the last year to disentangle different possible configuration and to derive the best 266

solution in terms of a budget material of the support structure of the calorimeter. Detailed simulations have been performed, allowing a preliminary design of the detector. A test beam is foreseen for the end of 2009 and study and simulations are underway on this topic. Results will be shown on geometry, dead material and mechanical structure of the forward LYSO crystal calorimeter for the SuperB project.
N43-4:

Commissioning and Operation of the CMS Electromagnetic Calorimeter

D. A. Petyt University of Minnesota, Minneapolis, MN, USA On behalf of the CMS Collaboration The CMS electromagnetic calorimeter (ECAL) is a fine-grained hermetic calorimeter consisting of 75848 lead tungstate crystals which provides fast response, radiation resistance and excellent energy resolution. It will play a crucial role in physics searches that will be conducted at the LHC. The design, construction and performance of ECAL will be reviewed, and the various stages of the assembly and commissioning of the crystal barrel and endcap detectors will be presented. Specific emphasis will be given to the overall performance of the detector following the installation of the endcap calorimeter in August 2008 and operation of the full crystal ECAL in the 3.8 Tesla magnetic field of CMS. In early 2009, the construction of ECAL was completed by the installation of the silicon pre-shower detectors. The general readiness of the detector for first beam collisions will be reviewed, together with any results relating to the overall performance of ECAL from first LHC beam data taken in 2009.
N43-5:

Commissioning of the ATLAS Liquid Argon Calorimeter

H. Ma Physics Department, Brookhaven National Laboratory, Upton, NY, USA On behalf of the ATLAS Liquid Argon Calorimeter Group The Liquid Argon calorimeter (LAr) is a key detector component in the ATLAS experiment at the LHC. It provides precision measurements of electrons, photons, jets and missing transverse energy produced in the LHC pp collisions. The LAr calorimeter has been installed in the ATLAS cavern and filled with liquid argon since 2006. The electronic calibration of the readout system, a critical system for precision measurements, has been continuously exercised in the commissioning phase, resulting in a fully commissioned calorimeter with its readout and a small number of problematic channels. A total of only 0.02% of the read out channels are dead beyond repair and 0.4% need special treatment for calibration. Throughout the last two years a large amount of calibration data has been collected. Cosmic muon data, first triggered via specially developed trigger boards on the LVL1 output of the Tile calorimeter and later with the standard ATLAS LVL1 calorimeter trigger, have been recorded at various stages of commissioning. In Sept 2008, with the first single beams circulating in the LHC ring and a near full readout of the calorimeter, events resulting from beam-gas interactions and beam-collimator splashes were recorded. We present here the LAr electronic calibration scheme, the measured stability of the pedestal, the pulse shape and the gain, and the expected calibration procedure for LHC running. We also present the calorimeter performance study based on the cosmic muon and LHC beam events. With the reconstructed muon minimum ionizing signal in the calorimeter, the uniformity of the barrel electromagnetic calorimeter can be checked. The timing alignment as measured from the data can be compared to the expectation. The high energy depositions in the calorimeter readout cells, either from catastrophic energy loss of the cosmic muons or from beam interactions, are used to validate the signal shape of calorimeter response derived from the calibration pulse.
N43-6:

Commissioning of the CMS ECAL Calibration with Muons from Cosmic Rays and Beam Dumps

T. Tabarelli De Fatis Dipartimento di Fisica, Universita' di Milano Bicocca and INFN Milano, Milano, Italy On behalf of the CMS Collaboration The electromagnetic calorimeter of the CMS experiment is designed to reach excellent energy resolution, essential for the discovery of narrow electromagnetic resonances (e.g. the SM model decay H->gamma gamma). The non uniformity of the individual channels directly contributes to the energy resolution. While the final performance will be achieved with in situ calibration exploiting physics events, a good performance at startup is already offered by the result of pre-calibration procedures involving test beam data, exposure to cosmic rays and laboratory measurements of the crystal light yield and photodetector gains. The intercalibration precision is better than 1.5% in the ECAL barrel and about 9% in the ECAL endcaps, on the average. During the commissioning phase in CMS, muons from cosmic rays and from beam dumps were used to verify whether the working conditions used during the calibration were reproduced in CMS. In particular, the measurement of the muon momentum by the tracking system allow to check the validity of the energy scale defined on a 120 GeV electron beam down to the energy released by a mip. Muons from beam dump data are used to improve the intercalibration of the ECAL endcaps. Further improvement are

267

expected with 2009 beam dump data. Test of the ECAL calibration and response stability during the commissioning phase are summarized and the projection of the performance at start-up are given.
N44: Gaseous Detectors III: GEM Applications in Particle Physics Thursday, Oct. 29 N44-1: 16:00-18:00 Grand Ballroom 3

G. Bencivenni1, D. Domenici1, M. Pistilli1, S. Cerioni1, E. De Lucia1, G. Felici1, M. Gatta1, M. Jacewicz1, F. Loddo2, G. De Robertis2, A. Ranieri2, G. Morello3, M. Schioppa3 1 laboratori nazionali di frascati - INFN - Italy, Frascati, Italy 2 INFN Bari, Bari, Italy 3 INFN Cosenza, gruppo collegato dei LNF, Cosenza, Italy We present the status of the R&D on the Cylindrical-GEM (CGEM) detector as Inner Tracker for the upgrade of the KLOE experiment at the DAFNE -Factory. The R&D activity spreads on different lines: construction and complete characterization of the full size CGEM prototype, equipped with 650m pitch 1-D longitudinal strips; extensive characterization of the single-mask GEM technology for the realization of large area GEM foils; detailed study of 2-D, XV patterned, readout strips, performed with small planar prototypes in a dedicated test at the H4-SPS beam facility, with and without magnetic field (up to 1T intensity) .
N44-2:

Status of the Cylindrical-GEM Project for the KLOE-2 Inner Tracker

The STAR Forward GEM Tracker

B. Surrow LNS, MIT, Cambridge, MA, USA On behalf of the STAR Collaboration The STAR collaboration is preparing a forward tracking detector upgrade, the Forward GEM Tracker (FGT), to provide fundamental studies of the proton spin structure and dynamics in high-energy polarized proton-proton collisions at RHIC. The FGT will focus on novel spin physics measurements in high-energy polarized proton-proton collisions at a center-of-mass energy of 500GeV, determining the flavor dependence of u and d antiquarks of the polarized sea. STAR plans to probe these polarized distribution functions using parity violating W production in the electron/positron decay mode. This upgrade will consist of six triple-GEM detectors with two dimensional readout arranged in disks along the beam axis. A two-dimensional readout board has been developed together with Compunetics Inc. based on laser-etching technology. The FGT project has completed an extensive R&D program of industrially produced GEM foils at Tech-Etch Inc. in comparison to GEM foils produced at CERN based on optical measurements, testbeam and 55Fe source measurements of a triple-GEM prototype detector using 10 X 10cm2 GEM foils. The FGT project requires large GEM foils which are currently being tested. The FGT design, the status of large GEM foil tests, the performance of triple-GEM prototype detectors based on industrially produced GEM foils along with the status of the FGT construction and the installation schedule will be presented.
N44-3:

Simulations of Space-Charge Effects and Event Deconvolution for the PANDA TPC

F. V. Boehmer, S. Neubert, S. Dorheim, C. Hoeppner, B. Ketzer, I. Konorov, S. Paul, M. Vandenbroucke, X. Zhang Physics Dept., Technichal University of Munich, Munich, Bavaria, Germany The PANDA experiment at the new FAIR facility in Darmstadt, Germany, will be investigating unresolved questions of nonperturbative QCD. The experiment is designed to make use of internal targets in the pbar storage ring HESR (design luminosity: 2 x 10 32 cm2 s1 ), requiring continuous operation of all detectors. One option for its central tracking unit is a Time Projection Chamber (TPC) with GEM readout. One of the main challenges due to the continuous operation at high rates is the buildup of ion space-charge in the chamber leading to drift distortions of the electrons on their way to the readout, compromising the track reconstruction capabilities of the detector. To be able to correct for this effect, space-charge buildup has been simulated with Monte Carlo methods based on realistic models. Spacecharge densities of up to 60 fC cm 3 in certain regions will be reached, leading to absolute drift distortions of up to 10 mm on the readout plane projection. The feasibility of a laser calibration system has been investigated on simulation level. Based on full simulations of the detector physics and response (including electronics) we were able to correct for drift distortions with an accuracy better than 200m. The effect of the drift distortions and their correction on momentum reconstruction and resolution will be presented. Another important challenge at the high rates (pbar-p annihilation rate: 2x107 s1 ) envisaged for PANDA is track deconvolution. Tracks from up to O(1000) events may be present in the TPC volume at the same time, requiring an efficient and fast method of track separation on the pre-fit level. Methods have been developed to achieve this task delivering promising results, which will be presented and discussed.

268

N44-4:

First Test Results for the TOTEM T2 Telescope

S. Lami INFN Pisa, Pisa, Italy On behalf of the TOTEM T2 Group The TOTEM T2 telescope will measure inelastically produced charged particles in the forward region of the LHC Interaction Point 5. Each arm consists in a set of 20 triple-GEM (Gas Electron Multiplier) detectors with tracking and trigger capabilities. The GEM technology has already been successfully adopted in other HEP experiments and has been considered for the design of TOTEM very forward T2 telescopes thanks to its characteristics: large active areas, good position and timing resolution, excellent rate capability and radiation hardness. Each of the four T2 half arms has been fully assembled and equipped with electronics at CERN to be systematically tested in the SPS beam line H8 in 2008/09, where also cabling and shielding were optimized with respect to noise reduction. These extensive tests were a common effort of the Pisa/Siena, Helsinki and CERN groups. After some optimization, the operation of the GEM chambers was fully satisfactory and according to expectations. Due to the lack of beam, most of the tests at the TOTEMINO facility on the H8 line have been performed with cosmic rays. These studies have been capitalized to optimize the detector noise reduction, timing properties as well as to test the whole readout and data acquisition chain. Furthermore, the tracking efficiency, the hit multiplicity and the pad and strip cluster size by crossing minimum ionizing particles have been measured and compared to a detector simulation. These results will be here presented. At the time of writing, the T2 telescope GEM detectors have been fully installed in their final position in IP5.
N44-5:

A Large TPC Prototype for an ILC Detector

G. W. P. De Lentdecker Universite Libre de Bruxelles, Brussels, Belgium On behalf of the LCTPC Collaboration A Time Projection Chamber (TPC) is a candidate for the central tracker of the future International Linear Collider (ILC) detectors. TPCs have already demonstrated very good performance in past collider experiments. However the tracking system of the ILC should have a very good track momentum resolution ((1/pt) ~ 5 10-5 /GeV/c), which is an order of magnitude more precise than in previous experiments. To achieve this resolution, the Linear Collider TPC (LCTPC) groups are pursuing R&D activities to determine the best state-of-the-art technology for the TPC using Micro Pattern Gas Detectors (MPGD) readout instead of the Multiwire Proportional Chamber (MWPC) readout. The MPGDs under investigation are the Gas Electron Multiplier (GEM) and the Micromesh Gaseous (MICROMEGAS) detectors as well as a new concept called TimePix combining a gas amplification on top of a CMOS pixel readout chip. To study those technologies, a Large Prototype TPC (LPTPC) has been built, which has a diameter of about 750 mm and a length of about 600 mm, which allows to measure tracks with up to 125 space points with pad readout. Since end of 2008, the LPTPC has been inserted into a 1.25 Tesla superconducting magnet, installed in a DESY test beam area. The LPTPC, alternatively equipped with the GEM or the MICROMEGAS readout, is exposed to an electron beam of up to 6 GeV. With both technologies the preliminary results look very promising. The LPTPC is not only a testing bed for several readout techniques based on MPGDs it is also an opportunity to understand the issues which arise when constructing such a large TPC. In this presentation, we will report on the setup, the production and the commissioning of the LPTPC as well as the first results of the test beams with the different readout technologies.
N44-6:

H. Ohshita1, S. Uno1, T. Otomo1, T. Koike2, T. Murakami1, M. Sekimoto1, T. Uchida1 1 High Energy Accelerator Research Organization (KEK), Tsukuba, Ibaraki 305-0801, Japan 2 Tokyo University of Science, Noda, Chiba 278-8510, Japan

Performance of a Neutron Beam Monitor with a GEM for the High-Intensity Total Diffractometer at J-PARC

We are developing a neutron beam monitor with a gas electron multiplier (GEM) for the high-intensity total diffractometer at JPARC. In order to analyze the basic characteristics of the GEM-based detector, a neutron irradiation test was carried out at JPARC. The wavelength-spectrum distribution obtained from the test is consistent with the calculations, and the beam profiles agree with the simple Monte Carlo (MC) simulation. Therefore we found that as a neutron beam monitor, the GEM-based detector has good two-dimensional imaging ability.

269

N45: High Energy Physics Instrumentation IV Thursday, Oct. 29 N45-1: 16:00-18:00 Grand Ballroom 7

The Status of BESIII Experiment

T. Hu Experimental Physics Center, Institute of High Energy Physics, Beijing, China On behalf of the BESIII Collaboration BESIII (BEijing Spectrometer III) has finished the installation of all the sub-detector components and starts the data taking. BESIII experiment including its design, construction and commissioning will be described. BESIII performances so far reached will also be presented.
N45-2:

The EUDET Infrastructures for Detector R&D

I. M. Gregor DESY, Hamburg, Germany On behalf of the EUDET Consortium EUDET is an initiative supported by the European Union to improve infrastructures for detector R&D, in particular for the International Linear Collider (ILC). The programme is focused on providing support for larger scale prototype experiments as well as on facilitating collaborative efforts. The project encompasses developments for vertex detectors, gaseous and silicon tracking, and highly granular electromagnetic and hadron calorimeters. Now, in 2009, the project is entering its last phase. Most infrastructures are already in place or will become available during 2009. This enables a number of interesting studies to be presented at the conference in fall 2009. This work is supported by the Commission of the European Communities under the 6th Framework Programme Structuring the European Research Area, contract number RII3-026126.
N45-3:

F. Simon1,2 1 Max-Planck-Institut fuer Physik, Munich, Germany 2 Excellence Cluster Universe, TU Munich, Garching, Germany On behalf of the CALICE collaboration

Energy Reconstruction of Hadron Showers in the CALICE Calorimeters

The CALICE collaboration has constructed highly granular electromagnetic and hadronic calorimeter prototypes to evaluate technologies for the use in detector systems at the future International Linear Collider. These calorimeters have been tested extensively in particle beams at CERN and at Fermilab.We present analysis results for hadronic events recorded at CERN with a SiW ECAL, a scintillator tile HCAL and a scintillator strip tail catcher, the latter two with SiPM readout, focusing both on the HCAL alone and on the complete calorimeter setup. Particular emphasis is placed on the study of the linearity of the detector response and on the single particle energy resolution. The high granularity of the detectors was used to perform first studies of software compensation based on the local shower energy density, yielding significant improvements in the energy resolution. The required calibration precision to achieve this resolution, and the effect of calibration uncertainties, for the CALICE HCAL as well as for a complete hadron calorimeter at ILC, has been studied in detail. The prospects of using minimumionizing track segments within hadronic showers for calibration are also discussed.
N45-4:

Test Beam Performance of the CALICE SiW Electromagnetic Calorimeter Physics Prototype

C. Calice IN2P3, LAL Orsay, Paris, France A prototype of a highly granular SiW electromagnetic calorimeter composed of approximately 10000 cells contained in a volume of 18x18x30cm^3 for an ILC detector has been examined in test beam campaigns conducted by the CALICE collaboration. Within the test beam environment a signal over noise ratio has been found to be 7.5 compared with the goal of 10 as envisaged for an ILC detector. The energy resolution of approximately 17%/sqrt{E[GeV]} is well within specifications. With the extracted linearity of approximately 1% the proof-of-principle is given that these high granular calorimeters can be operated successfully under beam conditions. The data will be further analysed in terms of exploiting the unprecedented high granularity which allows for instance the tracking of individual particles within hadronic cascades. The latter constitutes an important input to the tuning of existing hadronic shower models as available within the simulation toolkit GEANT4.

270

N45-5:

Jet Energy Corrections at CMS

A. Santocchia INFN and Perugia University, Perugia, Italy On behalf of the CMS Collaboration Many physics measurements in CMS will rely on the precise reconstruction of Jets. Correction of the raw jet energy measured by the CMS detector will be a fundamental step for most of the analysis where hadron activity is investigated. Jet correction plans in CMS have been widely studied for different condition: at stat-up simulation tuned on test-beam data will be used. Then datadriven methods will be available and finally, simulation tuned on collision data will give us the ultimate procedure for calculating jet correction. To allow for a better control of the different instrumental and theoretical effects, which enter in the jet correction procedure, a factorization in different sub-sequent steps has been implemented. Jet transverse energy is therefore corrected first for pile-up and noise offset; correction for the response of the calorimeter as a function of jet pseudorapidity relative to the barrel comes afterwards and correction for the absolute response as a function of transverse momentum in the barrel is the final standard sub-correction applied. Other effects like flavour and parton correction will be optionally applied on the Jet ET depending on the measurement requests. The energy offset, relative response and absolute response will be determined from data using minimum bias, dijet and photon+jet events, respectively. With higher integrated luminosity, the absolute correction is supplemented with Z+jet events. In this presentation we discuss the actual status of the complete CMS correction procedure with a particular focus on the data-driven methods: the main backgrounds and potential biases are discussed together with initial estimates of the systematic uncertainties. It is found that flavor and parton corrections, in particular the differences between light quark and gluon initiated jets, are important. Finally, the potential for a data-driven JEC in 1 pb-1, 10 pb-1 and 100 pb-1 scenarios will be outlined.
N45-6:

J. Yu1, A. P. White1, S. Park1, C. Hahn2, W. Kim2, K. Hong3, S. Kim4, J. R. Smith1 1 Physics, Univrsiy of Texas at Arlington, Arlington, TX, USA 2 Physics, Changwon National University, Changwon, Gyeongnam, South Korea 3 Physics, Korean Atomic Energy Research Institute, Daejeon, ChoongNam, South Korea 4 Physics, Korean National University of Education, CheongJoo, ChoongPook, South Korea

Development of the GEM Digital Hadron Calorimeter for the International Linear Collider

The physics program at the future International Linear electron-positron Collider (ILC) demands unprecedented precision in hadronic jet energy and position measurements. The Particle Flow Algorithm (PFA) that utilizes the precision charged track momentum measurements in measuring jet energies is a solution to meet this challenge. It is critical for PFA to be successful to minimize confusion terms in energy measurement by providing precise energy replacements with the measured momenta. It has been shown that 1cmx1cm level of granularity is necessary for a successful PFA. A one bit digital readout for energy measurement in hadronic calorimeter is a way to control the electronics cost at a manageable level. The High Energy Physics (HEP) group of the University of Texas has been developing a Digital Hadron Calorimeter (DHCAL) for the ILC, using Gas Electron Multiplier (GEM) as an attractive sensitive gap technology. Details on various size prototype detector development activities, including the details in mechanics, associated multi-channel readout electronics and the results on the bench using radioactive sources and in the particle beam, will be presented. Progress on developing large scale GEM prototype detector will also be presented along with the plans for testing GEM solutions in the 1mx1mx1m CALICE calorimeter stack.
N45-7:

J. Va'vra1, D. Aston1, D. Leith1, B. Ratcliff1, L. Ruckman2, G. Varner2, K. Nishimura2 1 PPA, SLAC, Menlo Park, USA 2 Department of Physics & Astronomy, University of Hawaii, Honolulu, USA

Progress in Development of the Focusing DIRC with a New ASIC-Based Waveform Digitizing Electronics

Abstract - We are developing a novel Cherenkov imaging detector, Focusing DIRC (FDIRC), for the Super B factory. The novel feature of this device is a new waveform digitizing electronics allowing a measuremnt of time and amplitude on each pixel with a sampling every 400 ps. The prototype's concept is based on the BaBar DIRC with several important improvements: (a) much faster pixilated photon detectors, (b) mirror allowing to make the photon detector smaller and less sensitive to background in future applications, (c) electronics allowing to measure the single photon resolution about 10 times better than Babar DIRC, to allow a chromatic error correction in future. In this test, the prototype was instrumented with seven Hamamatsu H-8500 MaPMTs. Waveforms from ~450 pixels were digitized with BLAB2 ASIC electronics with a sampling speed of ~2.5 GSa/s. This version of the FDIRC prototype is being tested in a large cosmic ray telescope providing muon tracks with ~1mrad angular resolution and energy cutoff of ~1.5 GeV. The prototype is used to evaluate the electronics performance to allow subsequent improvements in the ASIC design. The paper will present description of detector, electronics and the first results.

271

Medical Imaging Conference

J01: Instrumentation for Medical and Biological Research I: Radionuclide Imaging Tuesday, Oct. 27 See listing in the NSS section. J02: Instrumentation for Medical and Biological Research II: X-ray Imaging and Radiotherapy Applications Tuesday, Oct. 27 See listing in the NSS section. J03: Joint NSS/MIC 3 Tuesday, Oct. 27 See listing in the NSS section. J04: Joint NSS/MIC 4 Tuesday, Oct. 27 See listing in the NSS section. M01: Plenary 1 Wednesday, Oct. 28 M01-1: 08:00-10:00 International Ballroom Center 16:00-18:00 International Ballroom North 13:30-15:30 International Ballroom North 10:30-12:30 International Ballroom North 08:00-10:00 International Ballroom North

The Long Life of the Anger Scintillation Camera and Attempts to Replace It

G. Muehllehner The basic components of the Anger camera parallel hole collimator, NaI(Tl) crystal and photomultipliers have survived essentially unchanged for about 50 years. During this period is has been perfected and cost-reduced to the point where it becomes difficult for another technology to replace it. This presentation will describe some improvements that have been made over the years and the shortcomings that remain. It will also explore the various attempts to replace it and will describe industry attempts which have never been published. Various techniques will be highlighted that in the presenters opinion have significant merit and deserve further consideration.
M01-2:

Observations on 20 Years of LSO:Ce Scintillator Development

C. L. Melcher University of Tennessee, Knoxville, TN, USA Although powder synthesis and structural characterization of the rare earth oxyorthosilicates was achieved as early as the 1960's, the significance of Ce-doping for scintillation applications was not recognized until the 1980's. The evolution of Lu2SiO5:Ce (LSO:Ce) from an experimental powder phosphor synthesized in the laboratory to the large scale commercial production of single crystals implemented in Positron Emission Tomography spanned more than a decade and required solutions to numerous technical problems and commercial issues. These included the investigation of the scintillation mechanism, the purification and cost of raw materials, the development of a practical crystal growth process, the handling of intellectual property issues via patents and trade secrets, the development of detector processing techniques, the optimization of detector design, and the integration of detectors into imaging systems. Numerous international research collaborations contributed to the ultimate success of LSO:Ce. Research and development of LSO:Ce continues even today with recent advances based on co-doping to compensate charge traps and enhance energy transfer, thus improving light output and decay time.

272

J05: NSS/MIC Joint Posters Wednesday, Oct. 28 See listing in the NSS section. M02: Image Processing and Evaluation Wednesday, Oct. 28 M02-1: 14:00-15:30 International Ballroom Center 10:30-12:00 Palm 3, 4 & 5

Effect of Reconstruction Kernel Width on Optimal Regularization for Focal Lesion Detection in PET

D. J. Kadrmas Department of Radiology, University of Utah, Salt Lake City, UT, USA Iterative statistical algorithms offer desirable properties; however, the maximum likelihood solution provides sub-maximal performance for detecting focal lesions unless noise effects are regularized. Typical approaches include stopping the algorithm prior to convergence, penalizing objective functions, or post-reconstruction filtering. The width of the reconstruction kernel-i.e. the geometric component of the system matrix describing the projection operation and associated detector effects-dramatically affects noise correlations. This work explores the relationship between reconstruction kernel width and optimal regularization parameters that maximize focal lesion detection in PET. Experiments from the Utah PET Lesion Detection Database were used to model fully-3D oncologic whole-body FDG imaging, including multiple scans with- and without numerous lesions distributed throughout the phantom. Three reconstruction kernels were studied: a ray-driven projector with delta-function kernel; an areadriven projector with kernel width matched to the coincidence line spacing; and a volumetric projector with approximate point spread function model. Images were reconstructed to 250 iterations MLEM, storing intermediate iterations, and localization receiver operating characteristics analysis was applied using the channelized non-prewhitened observer. Maximal performance was obtained at 40, 70, and 100 iterations for the ray-driven, area-driven, and volumetric projectors, respectively. Conversely, optimal filter strength varied inversely with reconstruction kernel width. Overall, detection improved with broader kernels that more accurately modeled the tomographs actual response. These results clarify the complex relationship between reconstruction kernel width and the regularization needed to maximize focal lesion detection-important considerations when selecting regularization parameters and interpreting published results from different scanners or reconstruction methods.
R01CA107353 M02-2:

A. Borsdorf1, S. Kappler2, R. Raupach2, F. Noo3, J. Hornegger1 1 Chair of Pattern Recognition, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen, Germany 2 CT, Siemens Healthcare, Forchheim, Germany 3 UCAIR, University of Utah, Salt Lake City, UT, USA

Local Orientation-Dependent Noise Propagation for Anisotropic Denoising of CT-Images

In X-ray Computed Tomography (CT) the measured projections and consequently the reconstructed CT images are subject to quantum and electronics noise. While noise in the projections can be well described and estimated with a corresponding physics model, the distribution of noise in the reconstructed CT images is not directly evident. Due to attenuation variations along different directions, the nature of noise in CT images is non-stationary and directed. This complicates the direct application of standard post-processing methods like bilateral filtering. In this article we describe a possibility to compute precise orientation dependent noise estimates for every pixel position. This is done by analytic propagation of projection noise estimates through indirect fan-beam filtered backprojection reconstruction. The resulting orientation dependent image noise estimates are subsequently used in adaptive bilateral filters. Taking into account the non-stationary and non-isotropic nature of noise in CT images, a reduction in image noise of about 55% compared to 39% of the standard approach is achieved with much less variability over different image regions.
M02-3:

C. Wuerslin1,2, B. Yang1, F. Schick2, J. Machann2 1 Chair of System Theory and Signal Processing, University of Stuttgart, 70550 Stuttgart, Germany 2 Section on Experimental Radiology, University of Tuebingen, 72076 Tuebingen, Germany

Automatic Segmentation of Adipose Tissue in Magnetic Resonance Images of the Whole Body

The amount and the bodily distribution of different adipose tissue (AT) compartments are important indicators for the risk of obesity-related diseases and play an important role in the investigation of their pathogenesis. Magnetic resonance imaging (MRI) can be used to acquire images of the whole body, showing these compartments and their distribution. In this article, an automated segmentation algorithm is presented, being able to automatically create tissue volume profiles of the whole body for the tissue 273

classes subcutaneous AT (SCAT), visceral AT (VAT) and total tissue. The images are segmented using a fuzzy c-means algorithm, considering partial volume effects. A separation of the body into anatomic regions along the body axis is performed to define regions with VAT present. In abdominal image slices, the AT compartments are divided into subcutaneous and visceral compartments using an active contour algorithm. The slice-wise areas of different tissues are plotted against the slice position to obtain their topography. The automatically obtained tissue profiles were compared to profiles created manually by an epert. These profiles show correlation coefficients, indicating similar topography. Absolute error values were calculated for evaluation of the algorithm's absolute accuracy. These show low mean values for the classes of total tissue (4.5 %) and visceral adipose tissue (3.7 %). The deviation of total adipose tissue was higher though (9.6 %). In conclusion, the proposed algorithm enables the reliable and completely atomated creation of adipose tissue distribution profiles of the whole body from MR slices, reducing whole examination and analysis time to less than half an hour.
M02-4:

Implementing Visual Search in Human-Model Observers for Emission Tomography

H. C. Gifford, M. A. King Department of Radiology, University of Massachusetts Medical School, Worcester, MA, USA Human observers have been described as ``ideal observers with internal noise,'' but their internal noise mechanisms are not wellunderstood. As a result, studies using model observers such as the channelized Hotelling observer frequently require some fitting to human-observer data to obtain accurate predictions of human performance. This work presents initial tests of a human-model framework for emission tomography that attempts to minimize this problem with internal noise by relying less on statistical prior information about the test images. This framework is based on models of visual search for radiology, in which readings occur as a sequence involving fast, global scanning to identify candidate abnormalities followed by a lengthier focused analysis of these candidates. The initial identification of suspicious candidate sites by the model observer is directed by physical characteristics of the patient case at hand. Blob intensity and size were among the characteristics tested. Two statistical observers, a nonprewhitening (NPW) matched filter and a channelized NPW observer, were tested in the subsequent analysis of the suspicious locations. This latter stage represents an adaptive multiple-alternative forced-choice detection-localization task in which the number of alternatives can vary with the test image. Model observers based on this visual-search framework were tested against humans in an LROC study featuring hybrid SPECT lung images with Tc-99m NeoTect.
M02-5: Neurological Image Classification for the Alzheimer's Disease Diagnosis Using Kernel PCA and Support Vector Machines

M. Lopez, J. Ramirez, J. M. Gorriz, D. Salas-Gonzalez, I. Alvarez, F. Segovia, R. Chaves Dept. of Signal Theory, Networking and Communications, University of Granada, Granada, Spain An accurate and early diagnosis of the Alzheimer's Disease (AD) is of fundamental importance for the patients medical treatment. Single Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) images are commonly used by physicians to assist the diagnosis and rated by visual evaluations. In this work we present a computer assisted diagnosis tool that might aid the clinicians in this dementia diagnosis. The main problem to be faced up is the so-called small size sample problem, that is, the number of features used in the classification task is much larger than the number of available samples. We reduce the feature space dimension by means of Kernel Principal Component Analysis (PCA), which drastically reduces the feature space dimension. Images are projected onto the first m kernel principal components. Subsequently, Linear Discriminant Analysis (LDA) is used over the kernel PCA coefficients as a new linear projection that reduces to an l-dimension space taking into account the data labels, and maximizes the ratio of the between-class scatter and the within-class scatter in order to make it more reliable for classification. The resulting kernel PCA + LDA image coefficients are used to train a Support Vector Machine (SVM) classifier which is able to classify new subjects in an unsupervised manner. Several kernels were evaluated for both SPECT and PET images, using the Leave-One-Out cross-validation strategy. The proposed classification method outperforms the latest developed multivariate approaches, reaching up to 95.6% and 100% accuracy values for SPECT and PET images respectively.
This work was partly supported by the MICINN under the PETRI DENCLASES (PET2006-0253), TEC2008-02113, NAPOLEON (TEC200768030-C02-01) and HD2008-0029 projects and the Consejeria de Innovacion, Ciencia y Empresa (Junta de Andalucia, Spain) under the Excellence Project (TIC-02566). M02-6:

K. Thielemans1, E. Asma2, R. M. Manjeshwar2 1 R&D, Hammersmith Imanet Ltd, London, UK 2 Functional Imaging Lab, GE Research, Niskayuna, NY, USA

Mass-Preserving Image Registration Using Free-Form Deformation Fields

Patient movement deteriorates image quality in medical imaging. For many applications, such as respiratory movement, image registration is being used to try to combine several (noisy) images into one high quality image. Non-rigid image registration is in 274

many cases however an ill-posed problem. Even though images can be registered, the underlying deformation fields can be nonrealistic and/or non-unique. Fidelity to the meaning of the image can help to reduce such problems. In PET and SPECT, voxel values represent concentration of the radiotracer. Therefore, if an organ such as a lung gets compressed, the image values increase. Similarly, density increases, and therefore CT Hounsfield units will change. In most approaches for image registration, these changes in image values are ignored. Here we use a (penalised) least-squares objective function, modified to take the change in image values due to local stretching/compression into account by multiplying the transformed image with the Jacobian-determinant of the transformation, as for instance proposed by Ue et al (TMI 2006) for SPECT. We minimize this objective function using a free-form deformation field (i.e. parametrised by a deformation vector for every voxel). To do this, we derive formulas for the derivative of the objective function w.r.t. the deformation vectors, which allows us to use a simple gradient descent algorithm to find the penalised leastsquares solution. Our results indicate that the resulting objective function leads to unique solutions, even in uniform regions. However, regularisation is essential to prevent noise being registered as well.
M03: Image Reconstruction 1 Wednesday, Oct. 28 M03-1: 16:00-18:00 International Ballroom Center

Cone-Beam Reconstruction from a Variable-Radius, Planar Source Trajectory

F. Dennerlein, H. Kunze, J. Boese AX, Siemens AG, Healthcare Sector, Forchheim, Germany We present a practical Feldkamp-like algorithm for flat-panel cone-beam reconstruction from a planar source trajectory with varying radius. Along with the reconstruction, our algorithm also recovers the field-of-view, which can take complex shapes in the considered geometries. Numerical results of the FORBILD head phantom are presented, for a short-scan rectangular acquisition geometry.
M03-2:

S. Sra1, D. Kim2, I. Dhillon2, B. Schoelkopf1 1 Max Planck Institute for Biological Cybernetics, Tuebingen, Germany 2 University of Texas at Austin, Austin, Texas

A New Non-Monotonic Algorithm for PET Image Reconstruction

In this paper we present a new algorithm called Non-Monotonic Maximum Likelihood (NMML) for computing both penalized and non-penalized ML solutions for PET image reconstruction. We note that NMML can be easily extended to other inverse problems, e.g., transmission tomography, image restoration, etc. The distinguishing feature of NMML is its non-monotonic optimization of the likelihood, which allows it to take more aggressive steps, and therefore, to achieve rapid convergence. More importantly, we prove convergence of NMML, despite its non-monotonicity, which differentiates NMML from most other heuristic non-monotonic approaches. NMML is competitive to OSEM for reconstruction problems with low-noise. For sinograms with high-noise, NMML exhbits performance superior to OSEM, especially because it permits incorporation of regularization functions naturally and easily. Our initial results are quite promising---NMML significantly outperforms OSEM on wellconditioned data; for a more realistic setting with noisy sinograms and an ill-conditioned system matrix, it works even when OSEM fails, yielding results comparable to or better than an ordered subsets EM algorithm that can handle regularization.
M03-3:

Initial Evaluation of Direct 4D Parametric Reconstruction with Human PET Data

J. Yan, B. Planeta-Wilson, J.-D. Gallezot, R. E. Carson PET center, Yale University, New Haven, CT, USA Previously, we present a direct EM method for producing kinetic parameter images from list mode PET data, where the timeactivity curve for each voxel is described by a one tissue compartment model (1T). The initial evaluation was performed with simulation, without random and scatter effects included. By extension of our previous frame-based randoms and scatter correction methods, a practical direct 4D parametric reconstruction algorithm is now proposed and implemented for real data. Initial evaluation was performed using [11C]AFM human data. Comparisons with the 2-step approach (frame-based reconstruction followed by voxel-by-voxel parameter estimation) provide encouraging initial results. Regional analysis showed good agreement for K1 and VT values and the statistical noise for 4D was improved. These initial results suggest that direct 4D parametric reconstruction can be performed with real data, and offers the potential for improved accuracy and precision over the 2-step frame method.
We acknowledge the support of R01NS058360 (NINDS) and Siemens Medical Systems, Zhongdong Sun for programming and the staffs of Yale PET Center for the studies which formed the basis of this work

275

M03-4: Exact Iterative Image Reconstruction with a Simplified System Matrix: Potentials for GPUAccelerated Tomographic Imaging

L. Fu, J. Qi University of California, Davis, USA The quality of emission tomographic images is directly affected by the system model being used in image reconstruction. While an accurate system matrix can be computed by Monte Carlo simulation or directly measured from point source experiments, practical constraints on computation cost often force adoption of various approximations to obtain a computationally efficient system matrix. These approximations generally degrade image quality. In this work, we show that by properly constructing a surrogate imaging model, an algorithm can take advantage of the computation benefit of an approximate model, while retains the accuracy of the final solution. To validate the proposed method, we reconstruct microPET images using an accurate system matrix obtained from point source measurements, and compare the result with that obtained by the new method using a line integral approximation of the system matrix. The proposed method achieves similar objective function values, but is more computationally efficient. In particular, we show that when implemented on a graphic processing unit (GPU), a forward projection using the line integral model can be much faster than calculating an equivalent matrix vector product on the same GPU. Therefore, the proposed method allows further speed up by exploiting features of special computing hardware.
M03-5:

PET Image Reconstruction with Incomplete Anatomical Edge Information Using Level Set Method

J. Cheng-Liao, J. Qi Biomedical engineering, University of California, Davis, Davis,CA, USA Acquiring both anatomical and functional images during one scan, PET/CT systems improve the ability to distinguish abnormal uptake and to localize abnormal loci. Researchers have investigated utilizing anatomical boundary information in CT to regularize PET images. Here we propose a new approach to maximum /a posteriori/ (MAP) reconstruction of PET images with incomplete edges from CT image. An image prior is used to model both the smoothness of a PET image and the similarity between functional boundaries in the PET image and anatomical boundaries in the CT image. No smoothing is applied across functional boundaries to preserve sharp edges. Level set functions are used to track the functional boundaries. The proposed method does no assume an exact match between PET and CT boundaries, but maximizes the similarity between the two boundaries, while allowing different region definition in the PET image to accommodate possible signal and position mismatch between the two images. An advantage of the new method is that it does not require closed anatomical boundaries and can utilize incomplete edges obtained from a co-registered CT image. We conducted computer simulations to evaluate the performance of the proposed method. A digital phantom was constructed based on the digimouse data. Anatomical edges were derived from the gradient of the CT image smoothed by a Gaussian kernel. Tumors were added to the PET phantom with four different anatomical boundaries: matched, enlarged, reduced, and shifted. Compared with existing methods, the new method produced smooth images with sharp boundaries and achieved better biasvariance tradeoff. Acknowledgment: This work is supported by NIH under grant no. R01EB000194.
M03-6:

F. Dennerlein1, F. Noo2 1 Siemens AG, Healthcare Sector, Forchheim, Germany 2 Department of Radiology, University of Utah, Salt Lake City, Utah, USA

Avoiding the Backprojection Weight in Short-Scan CT Reconstruction

We present a technique to modify the power of the backprojection weight in analytic CT reconstruction formulas. The technique is applicable whenever the processing step that precedes the backprojection can be formulated as a data differentiation. To illustrate its merit, we apply our technique to the classical, ramp-filter based, short-scan fan-beam reconstruction formula to obtain a new short-scan filtered-backprojection formula with no backprojection weight. This new formula is evaluated using the Shepp-Logan phantom and shows improvement in uniformity of noise and resolution.
M03-7:

W. Huh1, J. A. Fessler1, A. M. Alessio2, P. E. Kinahan2 1 University of Michigan, Ann Arbor, MI, USA 2 University of Washington, Seattle, WA, USA

Fast kVp-Switching Dual Energy CT for PET Attenuation Correction

X-ray CT images are used routinely for attenuation correction in PET/CT systems. However, conventional CT-based attenuation correction (CTAC) can be inaccurate in regions containing iodine contrast agent. Dual-energy (DE) CT has the potential to improve the accuracy of attenuation correction in PET, but conventional DECT can suffer from motion artifacts. Recent X-ray CT systems can collect DE sinograms by rapidly switching the X-ray tube voltage between two levels for alternate projection 276

views, reducing motion artifacts. The goal of this work is to study statistical methods for image reconstruction from both fast kVpswitching DE scans and from conventional dual-rotate DE scans in the context of CTAC for PET.
This work was supported in part by NIH/NCI grant 1R01CA115870 M03-8: Direct 4D Parametric Image Reconstruction with Plasma Input and Reference Tissue Models in Reversible Binding Imaging

A. Rahmim, Y. Zhou, J. Tang Department of Radiology, Johns Hopkins University, Baltimore, MD, USA The most active area in brain PET ligand development and imaging continues to involve receptor/transporter studies involving reversible binding. The focus of this work has been to develop direct 4D parametric image reconstruction techniques for reversible binding imaging, having the advantages of moving beyond conventional independent frame image reconstruction and the ability to accurately model Poisson noise distributions in PET data (in contrast to the task of estimating complex noise correlations in reconstructed images). Based on a recent graphical analysis formulation, we developed a closed-form 4D EM algorithm to directly reconstruct distribution volume (DV) parametric images using a plasma input model. Furthermore, while previous works in the area of 4D imaging have been primarily limited to plasma input models, we sought to also develop reference tissue model schemes whereby distribution volume ratio (DVR) parametric images were reconstructed by the reference tissue model within the 4D image reconstruction task (using the cerebellum as the reference tissue in this work). The means of parameters estimated from 55 human [11C]Raclopride dynamic PET were used for simulation (22 realizations) using a mathematical brain phantom. Images were reconstructed using standard FBP or EM methods followed by modeling, as well as the proposed direct methods. Noise vs. bias quantitative measurements were performed in eight regions of the brain (grey, caudate, putamen, thalamus, corpus callosum, white, nucleus accumbens, cerebellum). Direct 4D EM reconstruction resulted in substantial visual and quantitative accuracy improvements (over 100% noise reduction, with matched bias, in both 4D schemes). Substantial improvements were also observed in the coefficient of variation (COV) of the estimated binding potential (BP) values, including even for the relatively low BP regions of grey and thalamus, suggesting the ability for robust parameter estimation even in such regions.
M04: Quantitative Imaging Techniques Thursday, Oct. 29 M04-1: 08:00-10:00 International Ballroom Center

C. Catana1, A. J. van der Kouwe1, T. Benner1, M. Hamm2, C. Michel3, B. Fischl1, B. R. Rosen1, A. G. Sorensen1 1 AA Martinos Center for Biomedical Imaging, Radiology Dept., Massachusetts General Hospital, Charlestown, MA, USA 2 Siemens Medical Solutions, Charlestown, MA, USA 3 Siemens Medical Solutions, Knoxville, TN, USA There has been recently great interest in combining PET and MRI, and a number of integrated scanners capable of simultaneous acquisition have been developed and successfully tested for small animal and human imaging. Because these scanners are not equipped with a transmission source, an MR-based attenuation correction method has to be developed. Although MRI provides excellent soft tissue contrast, tissues with very different linear attenuation coefficients (i.e., air and bone) cannot be easily distinguished using conventional MR sequences. The effect this has on PET data quantification in combined human MR-PET scanner was studied in simulations starting from segmented MR and CT datasets. Our results suggest that in order to achieve the necessary quantitation for advanced neurological studies, three main components must be identified: water-based structures, bone tissue and air-filled cavities. Furthermore, the accuracy of an MR-based AC method could, in principle, be similar to that of the segmented CT method, which is frequently used in clinical studies. Using ultra-short echo time MR sequences we demonstrated that bone/air segmentation can potentially be performed. Once the data are segmented, known linear attenuation coefficients can be assigned to each tissue class. Another factor that could affect the PET quantification in a combined MR-PET scanner is the RF coil which is located between the patient and the PET detectors. Ignoring the coil attenuation can lead to degradation of PET image quality or introduction of artifacts. An MR-based AC method implemented considering all these aspects could in principle provide accurate estimation of the activity concentration within 5%. If implemented, such a method would also have the advantage of reducing the subjects radiation exposure and the total examination time and of eliminating the co-registration errors between the emission data and attenuation map.

Towards Implementing an MR-Based PET Attenuation Correction for Neurological Studies

277

M04-2:

A. J. Chaudhari1,2, A. A. Joshi3, A. W. Toga3, R. M. Leahy4, S. R. Cherry2, R. D. Badawi1 1 Radiology, University of California, Davis, Sacramento, CA, USA 2 Biomedical Engineering, University of California, Davis, Davis, CA, USA 3 Laboratory of Neuro Imaging, University of California, Los Angeles, Los Angeles, CA, USA 4 Electrical Engineering, University of Southern California, Los Angeles, CA, USA

Atlas-Based Attenuation Correction for Small Animal PET/MRI Scanners

Small animal PET/MRI scanners producing anatomically co-registered simultaneously-acquired images of morphology, function and metabolic activity have become available and are expected to have a major impact on the pre-clinical imaging field. Attenuation correction (AC) necessary for accurate quantification of PET signals is challenging in these scanners because the measured anatomical map from MRI is based on proton densities and not on electron densities. We propose a deformable mouse atlas-based registration method for AC in small animal PET/MRI. In this method, we first match the posture of the atlas to the posture of the mouse being imaged using landmark and elasticity constraints. The asymmetric L2 pseudo-distance between the atlas surface and the extracted mouse surface is then minimized in order to register the animal surfaces. A Sobolov prior is used to ensure smoothness of the warping field. This warping field is then extended to the entire mouse volume. The computed warping transform is applied to the CT of the mouse atlas. Mutual information-based rigid registration is carried out to further align the warped atlas CT to the mouse MRI. We evaluated the proposed method for co-registered MRI and CT data acquired for a normal nu/nu mouse. Four organs were manually segmented from the animal MRI. Dice coefficients of organ overlap were measured for these organs between the manually assigned labels and those estimated by the proposed method. Dice coefficients were also measured between the skull and skeleton extracted directly from the animal CT and those estimated using the proposed method. Organ overlap of > 90 % in the head and torso region of the animal and > 60 % in the abdominal region are observed. For the skull and skeleton, the measured overlap was > 90 %. Based on published values, we derive a detailed attenuation map at 511 keV for the mouse using the warping field estimated by the proposed method.
M04-3:

Y. K. Dewaraja1, K. F. Koral1, J. A. Fessler2 1 Radiology, University of Michigan, Ann Arbor, MI, USA 2 Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, MI, USA

Quantitative I-131 SPECT Reconstruction Using CT Side Information From Hybrid Imaging

Our goal is accurate I-131 SPECT quantification for estimation of mean target absorbed dose and 3D dose distributions in internal emitter therapy. Currently, unregularized OSEM is used clinically as the iterative reconstruction algorithm for SPECT, often followed by post-filtering to reduce image noise. Indiscriminate smoothing including across region boundaries degrades quantification of activity. Using 3D OSEM with no post-filtering, we have observed good ROI quantification accuracy when using a large number of iterations, but at the expense of significant edge overshoots and considerable noise. Therefore, to achieve good estimates of both the total target activity and the activity distribution we investigated a penalized-likelihood approach to SPECT reconstruction using a modified regularizer that accounts for anatomical boundary side information obtained from CT. For the penalty function we used a spatially variant weighted Gibbs function. For both simulated and measured I-131 phantom data we compared reconstructions from 1) OSEM; 2) penalized likelihood employing CT-side information based regularization (PL-CT); 3) penalized likelihood with edge preserving regularization (no CT); 4) penalized likelihood with quadratic regularization (no CT). For both simulated and measured data the images and profiles corresponding to PL-CT were closest to the truth and did not have the edge offshoots evident with OSEM nor the blurring across boundaries evident for the methods where CT information was not used. Count recovery with PL-CT was slightly lower than OSEM for smaller tumors, but PL-CT was superior in terms of RMSE. The PL-CT reconstruction was applied to SPECT/CT data from a patient using the CT defined tumor outlines to determine the weights for the Gibbs function. These results demonstrate the potential of penalized-likelihood SPECT reconstruction accounting for CT information to provide accurate target activity and activity distribution for 3D dosimetry.
Research Support: NIH 2R01 EB001994 M04-4:

E. Rota Kops1, G. Wagenknecht2, J. J. Scheins1, L. Tellmann1, H. Herzog1 1 Forschungszentrum Juelich, Institute of Neuroscience and Medicine, Juelich, Germany 2 Forschungszentrum Juelich, Central Institute for Electronics, Juelich, Germany

Attenuation Correction in MR-PET Scanners with Segmented T1-Weighted MR Images

Attenuation correction of PET data acquired in new hybrid MR-PET scanners which do not offer the possibility of a measured attenuation correction can be done in different ways. A previous report of our group described a method which used attenuation templates. The present study utilizes a new knowledge-based segmentation approach applied on segmented T1 weighted MR images. It examines the position and the tissue membership of each voxel and segments the head volume into attenuationdiffering regions: brain tissue, extracerebral soft tissue, skull, air-filled nasal and paranasal cavities as well as the mastoid process. To examine this new approach four subjects having MR and PET (one having CT, too) were chosen. The segmented T1 278

weighted MR images were converted into attenuation maps for 511 KeV photons with coefficients of 0.098 1/cm for brain tissue, 0.146 1/cm for skull, 0.096 1/cm for soft tissue, 0.054 1/cm for the mastoid process, and 0.0 1/cm for nasal and paranasal cavities. In case of the CT volume the Hounsfield units were also converted into attenuation coefficients valid for 511 keV photons. The 4 segmented-based attenuation maps as well as the converted CT were first filtered by a 3D Gaussian kernel of 13 mm filter width and then used to reconstruct PET emission images. These were compared to PET images attenuation corrected using the conventional PET based transmission data. Relative differences (RD) were calculated from ROI values. For the subject with CT-based attenuation correction the RD exhibit a mean of 0.10%+/-1.50%, while the results from the segmented-based attenuation correction are slightly higher (1.85%+/-2.54%). Comparing the results obtained with the segmented-based attenuation correction only, the RD range from -2.23%+/-0.89% to 2.43%+/-2.07% with a maximum of 6.10% and a minimum of -3.40%. In conclusion, this new segmentation-based attenuation correction yields promising first results which are to be improved by further developments.
M04-5:

Quantification Accuracy & Partial Volume Effect of the Siemens Inveon Small Animal PET Scanner

J. G. Mannheim, M. S. Judenhofer, V. Sossi, B. J. Pichler Laboratory for Preclinical Imaging and Imaging Technology of the Werner Siemens-Foundation, University of Tuebingen, Tuebingen, Germany This study evaluates the quantification accuracy and the partial volume effect (PVE) of the Inveon small animal PET scanner. The reported numbers are especially of importance when small lesions in mouse or rodent brain studies should be quantified. To determine the influence of the emission activity contaminating the transmission data, attenuation measurements were acquired before, during and after the emission scan (at 37, 15.8 & 0 MBq). For an attenuation correction acquired at 37 MBq emission activity, the scanner shows small deviations from the true activity. However the deviation increases when including a transmission scan with a lower emission activity for attenuation correction. To determine the influence of the object size on the quantification accuracy three different sized phantoms mimicking the imaging conditions of mice and rats were measured. FBP with a segmented attenuation correction was used as reference and deviations were calculated. The deviation between the three applied reconstruction algorithms goes up to 84% for a large scanned object. The PVE was investigated in dependence of the object size and location within the FOV. The NEMA micro hollow sphere phantom including 4 different sphere sizes was measured with a sphere to background ratio of 8:1. The recovery of the biggest sphere (7.86 mm) is 60%. The smallest sphere (3.95 mm) has a recovery of only 30%. To determine the variation in partial volume effect in dependence to the FOV position, a single sphere (3.2 mm) located in a cylindrical phantom was used and moved through the FOV. The PVE changes along the transaxial FOV from 35% to 32%. The Inveon PET scanner is suitable for small animal imaging although the measurements clearly showed that for accurate quantification the protocols have to be selected and adapted to the imaging situation. The relatively large PVE causes a quantification error in small lesion mouse studies and in specific in rodent brain studies.
M04-6:

K. Wells1, B. Goswami1, E. Lewis1, A. Ab-Rahni1, J. Jones1, M. Alnowami1, M. Guy2 1 Centre for Vision, Speech & Signal Processing, University of Surrey, Guildford, Surrey, UK 2 Medical Physics, Royal Surrey County Hospital, Guildford, Surrey, UK Motion correction of the abdominal-thoracic region is one of the main research challenges in tomographic nuclear medicine imaging. We address this issue with a flexible data-driven method of motion correction. This uses marker-less stereo tracking of the anterior abdominal-chest surface and a 'virtual dissection'-based registration approach, combined within a novel particle filtering (PF) framework. The key advantage to this data driven approach is that we do not make gross prior assumptions on the configuration of the hidden state of the system, i.e. the configuration of the internal organs during the emission acquisition process. Instead, at some given time instance, we infer the hidden or unobserved internal organ configuration by using Monte Carlo sampling (and then filtering) of various propositions, or 'particles'. Such estimates are calculated using the previous state (of the internal organs) plus some noise or perturbation of the expected transition to the current state or configuration. We then compare estimated representations of the abdominal-chest anterior surface, derived from the particles or propositions, with an observation of the actual surface, derived from a marker-less stereo imaging system. By examining the differences between the estimated particle or proposition surfaces and actual observed surface data, we can infer the current configuration of the internal organs. After an update step, the process is then repeated for subsequent time points in the emission data. This allows the system to flexibly adopt previously unknown configurations of the internal organs, and thus allow for different modes of breathing (e.g. abdominal vs chest-based motion) to be represented. Preliminary results will be presented based on using the XCAT phantom to demonstrate the PF approach and the 'virtualdissection' registration process, alongside results of a parameterized anterior surface model derived from human volunteer data. 279

A Flexible Approach to Motion Correction in Nuclear Medicine

M04-7: Optical flow based estimation of multiple respiratory phases from a single CT for attenuation correction of PET Data: A phantom study

M. Dawood1,2, M. Fieseler1, F. Buther2, M. Schafers1,2, K. P. Schafers1,2 1 European Institute of Molecular Imaging, University of Muenster, Muenster, Germany 2 Department of Nuclear Medicine, University Hospital Muenster, Muenster, Germany

Respiratory motion is a major concern in positron emission tomography (PET) as it induces motion artifacts to the images. Methods for correcting respiratory motion on the non-attenuation corrected PET images have been presented by the authors before. For quantitative PET images, however, attenuation correction of the PET data with corresponding CT respiratory phases is required before motion correction can be performed in PET/CT studies. Respiratory gated CT acquisition leads to increased radiation exposure to the patients and is thus not feasible for many patients. In this study a method of transforming a single 3D CT scan to multiple respiratory phases is presented. The respiratory motion is first estimated with optical flow methods on respiratory gated PET data. The estimated motion vectors are then applied to the single CT volume to derive multiple respiratory phases. The methods are tested and validated on NCAT software based phantom data. Patient studies are already planned and the results will be added by the conference time. Using a single CT volume, eight respiratory phases were generated. The results showed that the ground truth was achieved accurately to a large extent. The difference in the position of the diaphragm between the estimated and the ground truth data was less than the resolution of the images (3.375 mm). Small errors are still seen in the areas, where no motion information is present on the PET data such as inside the lungs or the narrow edges at the end of the lower lung lobes as these are not visible on PET images. Future work should include more precise reconstruction of PET/CT data using a combined motion and attenuation correction during the reconstruction process.
M04-8: Simultaneous Measurement of Cardiac Perfusion and Permeability in Vivo with Synchrotron Radiaion Imaging

A. H. Walenta1, B. Michael2, B. Alberto3, E. Francois4, E. Raimund5, F. Stefan6, K. Oliver7, M. Joerg8, M. Stefan5, S. Brunao2, S. Hans-Werner1, W. Katrin2 1 Physics, University of Siegen, Siegen, Germany 2 Universittsklinikum, Klinik fr innere Medizin, Homburg, Germany 3 ID 17, ESRF, Grenoble, France 4 ISERM, ESRF, Grenoble, France 5 Klinik fr Kardiologie, University Essen Duisburg, Essen, Germany 6 EMBL, DESY, Hamburg, Germany 7 Medizin- Informatik, University of Applied Science, Heilbronn, Germany 8 Mielebacher Informatiksysteme, Bremen, Germany

The high temporal, spatial and contrast resolution of synchrotron radiation imaging allows the measurement of the residual density of an injected contrast medium with high precision and delivers absolute values. This was confirmed with measurements on pigs where it could be shown that the identification and characterization of the relevant compartments and their structures allows to record the time dependent contrast medium density in arteries, the myocardium and the veins. With an adapted flow model the absolute local determination of the perfusion and for extracellular contrast media (as Gadovist) the permeability through the capillary membrane could be obtained.
M05: MIC Posters 1 Thursday, Oct. 29 M05-1: 10:30-12:30 Grand Ballroom 4&5; Palm 3,4&5

Basic Evaluation of a C-Shaped Breast PET Scanner

M. Furuta, K. Kitamura, J. Ohi, H. Tonami, Y. Yamada, T. Furumiya, M. Satoh, T. Tsuda, M. Nakazawa, N. Hashizume, Y. Yamakawa, A. Kawashima, Y. Kumazawa Technology Research Laboratory, Shimadzu Corp., Kyoto, Japan For diagnosis of very small lesions of breast cancer on very early stage, a dedicated breast positron emission tomography (PET) scanner consisting of four-layer depth of interaction (DOI) detectors is now under development. We are aiming for the spatial resolution of less than 1 mm in this scanner and acquisition time is less than 5 minutes by one breast and 10 minutes in total. The C shape of this scanner allows it to be positioned closely around the breast, effectively increasing both resolution and sensitivity. The C-shaped design of the scanner accommodates a variety of patient physiques, ensuring inclusion of the entire breast into the effective field of view (FOV). The detector consists of four-layer of a 32 32 Lu1.8Gd0.2SiO5 (LGSO) crystal array, a light guide and a 64ch flat panel type 280

photomultiplier tube (H8500). The size of crystal element is 1.44 mm 1.44 mm 4.5 mm. Front-end electronics will be implemented on an Application Specific Integrated Circuits (ASIC) chip to reduce the external diameter of the detector ring. We have developed a prototype equipment of C-shaped PET scanner. Positioning for imaging can be kept without a pain for the subject. We report basic evaluation of the prototype, such as , sensitivity, count rate, resolution, and image quality of phantoms.
This study was conducted as a part of the project R&D of Molecular Imaging Equipment for Malignant Tumor Therapy Support, supported by NEDO (New Energy and Industrial Technology Development Organization). M05-4: Measurement of Lung Ventilation Using [13N]N2 and the High Resolution Research Tomograph: a Pilot Study in a Pig Model

M. C. Huisman1, J. J. Spijkstra2, L. F. van Rooij1, F. H. P. van Velden1, R. Boellaard1, A. B. J. Groeneveld2, A. A. Lammertsma1 1 Department of Nuclear Medicine & PET Research, VU University Medical Center, Amsterdam, The Netherlands 2 Medical Intensive Care Unit of the Department of Internal Medicine, VU University Medical Center, Amsterdam, The Netherlands Mechanical ventilation is used for patients with acute respiratory distress. Relative benefits of different ventilation strategies, however, are still under debate. The purpose of the present study was to develop a method for assessing effects of mechanical ventilation on pig lungs in-vivo using [13N]N2 and positron emission tomography (PET). For this study several new components were developed: (i) a mobile [13N]N2 gas delivery system, providing a constant flow of 250 MBq/min for a duration of ~15 minutes; (ii) a mechanical ventilator that produces transistor-transistor logic based gating signals for every change from inspiration to expiration (and vice versa). Together with a high resolution research tomograph (HRRT) PET scanner these components allow for the acquisition of end-inspiration and end-expiration lung ventilation images with a spatial resolution of ~3 mm. In a first experiment on a 15 kg healthy pig this setup was validated. The animal was ventilated (tidal volume 120 ml, frequency 20 strokes min-1) and a respiratory gated ventilation study was performed. Based on the gating signals indicating the change from inspiration to expiration the last 33% of the inspiration time interval were used to reconstruct an end-inspiration image (and vice versa). Volumes and total activities were ~660 ml and 42 MBq for the end-expiration phase and ~800 ml and 55 MBq for the end-inspiration phase, respectively. Each stroke 13 MBq was thus added to the lungs (12.5 MBq was expected from the ventilator settings). Subsequent macroscopic washout of the tracer occurred with a half-life of 15 s. In conclusion, in a pig model, lung ventilation can be studied using [13N]N2 and the HRRT PET scanner. This methodology is now available for clinically relevant pig models.
M05-7:

H. Kim1, C.-M. Kao1, C.-T. Chen1, H. Frisch2, J.-F. Genat2, F. Tang2, E. Oberla2, W. W. Moses3, W.-S. Choong3 1 Dept. of Radiology, University of Chicago, Chicago, IL, US 2 Enrico Fermi Institue, University of Chicago, Chicago, IL, US 3 Lawrence Berkeley National Laboratory, Berkeley, CA, US A PET detector design using continuous scintillator slab was investigated. The detector unit consists of two layers of LSO scintillator slab, each slab size of 102x102x5$mm^3$, and two large area microchannel plate(MCP) PMTs, 102x102$mm^2$, coupled to scintillator slabs. The optical photon inside scintillator was simulated using Geant4 package and the electrical signal of MCP was formed using the measured characteristics of MCP and Geant4 output. The signals from MCP were readout using the transmission line(TL) scheme with 4.25mm pitch. The multi-layers of structure enable us to extract the depth of interaction in addition to position, energy and timing in an event. The detector response was measured by impinging the pair of 511keV gamma upon the detector. As preliminary results, we obtained ~13%(FWHM) of energy resolution at 511keV and ~460ps(FWHM) coincidence timing resolution while keeping 14% detection efficiency at 511keV. The position resolution was measured ~2.3mm(RMS) at the center. Due to its simplified structure, the detector can be easily extendable to several layers of slab to increase the sensitivity. The fast timing characteristics of MCP combined with the high sensitivity of LSO makes this design reliable for TOF PET application.
M05-10:

Continuous Scintillator Slab with Microchannel Plate PMT for PET

C. W. Lerche1, N. Ferrando1, J. M. Monzo1, R. Gadea1, R. J. Colom1, F. J. Toledo1, J. M. Benlloch2, A. Sebastia1 1 Digital System Design Group, Universidad Politecnica de Valencia, Valencia, Spain 2 Nuclear Medical Physics Group, Instituto de Fisica Corpuscular, Valencia, Spain We present measurements of the timing properties of a gamma-ray imaging detector based on a single, large-sized monolithic LSO block and a position sensitive photomultiplier. The black painted surfaces and the large geometry minimize reflections and therefore the spread of the scintillation light propagation time. A timing resolution of 0.77 +/- 0.02 ns FWHM was measured. Since the analog front-end electronics also allows the determination of the gamma-ray's depth of interaction, this result can be

Timing Properties of a Gamma-Ray Imaging Detectors with Monolithic Scintillators

281

further improved. We discuss possibilities of correcting the gamma-ray arrival times for the gamma-ray's depth of interaction and compare the results to the resolution obtained with a custom made acquisition system and different time extraction algorithms.
M05-13:

J. S. Goddard1, J. S. Baba1, S. J. Lee1, A. G. Weisenberger2, A. Stolin2, J. McKisson2, M. F. Smith3 1 Measurement Science and Systems Engineering Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA 2 Thomas Jefferson National Accelerator Facility, Newport News, VA, USA 3 University of Maryland, Baltimore, MD, USA New developments have been made in optical motion tracking for awake animal imaging that measures 3D position and orientation (pose) for a single photon emission computed tomography (SPECT) imaging system. Ongoing SPECT imaging research has been directed towards head motion measurement for brain studies in awake unrestrained mice. In contrast to previous results using external markers, this work extracts and tracks intrinsic features from multiple camera images and computes pose from the tracked features over time. Motion tracking thus far has been limited to measuring extrinsic features such as retro-reflective markers applied to the mouses head. While this approach has been proven to be accurate, the additional animal handling required to attach the markers is undesirable. A significant improvement in the procedure is achieved by measuring the pose of the head without extrinsic markers using only the external surface appearance. This approach is currently being developed with initial results presented here. The intrinsic features measurement extracts discrete, sparse natural features from 2D images such as eyes, nose, mouth and other visible structures. Stereo correspondence between features for a camera pair is determined for calculation of 3D positions. These features are also tracked over time to provide continuity for surface model fitting. Experimental results from live images will be presented.
M05-16:

Intrinsic Feature Pose Measurement for Awake Animal SPECT Imaging

A. Z. Kyme1,2, V. W. Zhou1,2, S. R. Meikle2,3, K. Popovic2, M. Akhtar2,3, R. R. Fulton1,2,3,4 1 School of Physics, University of Sydney, Sydney, NSW 2006, Australia 2 Brain & Mind Research Institute, University of Sydney, Sydney, NSW 2006, Australia 3 Discipline of Medical Radiation Sciences, University of Sydney, Sydney, NSW 2006, Australia 4 Department of Medical Physics, Westmead Hospital, Sydney, NSW 2145, Australia Pre-clinical positron emission tomography (PET) is becoming increasingly important in understanding brain physiology using animal models. One of the major challenges at present is being able to perform brain PET studies without the use of anesthesia. In most cases where the animal is minimally restrained this will require some form of motion tracking to provide the necessary temporal pose information for compensation before or during image reconstruction. In previous work we have successfully demonstrated tracking of continuous movement in phantom studies and an anesthetized rat study in which the animal was moved manually. Here we report on our first successful experiments tracking the head of a fully conscious rat undergoing continuous head movement during emission and transmission PET acquisitions performed on a microPET Focus 220 scanner. The motion tracking system is based on a commercial stereo-optical motion tracking device called the MicronTracker Sx60. We have previously reported in detail on the device and its suitability for small-scale motion tracking. Here we describe the set-up and marker considerations used for motion tracking of conscious animals and report the motion data obtained. An example of the correction that can be obtained using these data for motion compensation is also presented. The tracker has been adapted to the microPET to maximize visibility of a small head marker moving through large angles. Using a marker of 18 mm x 24 mm we were able to successfully track animal head motion at least 95% of the time. Angular ranges exceeding 90 deg, and angular and linear velocities > 300 deg/s and 150 mm/s were tracked robustly at 30 Hz. Missed samples were replaced by substituting the last collected pose. These motion data enabled effective compensation of the animal movement. In future work we aim to continue optimizing motion tracking of awake animals, particularly in terms of increased accuracy within a smaller field of measurement.
M05-19:

Motion Tracking of Fully Conscious Small Animals in PET

A Monte Carlo Approach to Handle Data Scaling in Nuclear Medicine Imaging

C. Bai, R. L. Conwell Digirad Corporation, Poway, CA, USA In nuclear medicine imaging, data scaling is sometimes desired to handle the system complexity, such as uniformity calibration, head balance calibration for multiple-head systems, automatic decay compensation, etc. Since nuclear medicine data are usually saved in short integer, conventional data scaling will first scale the data in float and then truncate or round the floating data to short integer data. For example, when using truncation, scaling of 9 by 1.1 results in 9 and scaling of 10 by 1.1 results in 11. When the count level is low, such scaling may change the local data distribution and affect the application of the data. In this work, we use a low count gated cardiac SPECT study to illustrate the effect of conventional scaling by factors of 1.1 and 1.2. We then scale the data using a Monte Carlo approach using the same factors. In this approach, 9 scaled by 1.1 results in 9.9 after the multiplication. A random number evenly distributed between 0 and 1 is generated, if the random number is between 0 and 0.9, then final data will be set to 10, otherwise to 9. Finally we reconstruct and compare the images from the original data, the 282

conventionally scaled data, and the Monte Carlo scaled data. Results show that scaling of the data by 1.1 and 1.2 using the conventional approach leads to significantly different perfusion image. The image reconstructed from the original data showed a perfusion defect at the apex. The defect size was noticeably increased by scaling with 1.1 and significantly decreased by scaling with 1.2. Using the Monte Carlo approach, on contrast, the images from the scaled data appeared identical to the original image. In conclusion, conventional approaches for data scaling can lead to unexpected local count variation and significantly affect the clinical usage of the data. The proposed Monte Carlo approach minimizes the local count variation thus is better for data scaling in low count nuclear medicine imaging.
M05-22:

S. Cho1,2, D. Xia2, C. A. Pelizzari1, X. Pan1,2 1 Radiation and cellular oncology, The University of Chicago, Chicago, IL, USA 2 Radiology, The University of Chicago, Chicago, IL, USA

Utilization of Two Analytic Algorithms for Image Reconstruction in Cone-Beam CT

Cone-beam computed tomography (CBCT) is widely used in modern clinical applications: not only in diagnostic imaging tasks but also in interventional image-guidance procedures. Analytic image reconstruction algorithms have been developed in conjunction with new scanning trajectories in CBCT. The chord-based BPF algorithm basically decomposes the image into a number of chords, where a chord is defined by a line segment that connects any two points on the source trajectory, and reconstructs the image on the chords that intersect the imaged object and that collectively cover the object. It can be therefore easily applied to a general source trajectory, and is relatively robust against data truncation. Pack and Noo proposed an image reconstruction algorithm that is independent of the chords, thus enabling image reconstruction off the chords for some situations, based on a key observation that the filtered backprojection on the point can be related to the inverse Radon transform of the image at the point. Yang et al introduced a polygon plane concept to facilitate the utilization of the Pack-Noo algorithm, and we have extended Yang's method to reconstruct the image on a polygon plane with less cone-beam data thus enabling reduced CBCT scan. For some scanning configurations, use of these two algorithms in a way that the strength of each can benefit the overall image reconstruction is desirable. In this work, we present a solution to the long object problem in reverse helical CBCT using the two algorithms consecutively. The chord-based BPF algorithm reconstructs part of the volumetric image of an object from the original cone-beam data first, and the middle gap region is reconstructed by the Pack-Noo formula-based FBP algorithm from the subtracted cone-beam data by use of reprojection of the first reconstructed volumetric image.
M05-25:

The Tumor Resection Camera (TReCam), a multi-pixel gamma imaging probe for radio-guided surgery

E. Netter, L. Pinot, L. Menard, M.-A. Duval, B. Janvier, F. Lefebvre, R. Siebert, Y. Charon UMR 8165, IN2P3, CNRS, Paris 7 Diderot, Paris 11 Sud, Imagerie en Modelisation, Neurobiologie et Cancerologie - UMR 8165 CNRS, Orsay, France Using the POCI camera, we recently demonstrated the clinical impact of per-operative imaging techniques thanks to a successful clinical trial. This extensive trial was based on the sentinel lymph node (SLN) protocol for breast cancer staging. The trial included 160 patients and showed that a mini gamma camera could be routinely used to perform the preoperative lymphoscintigraphy and was of great help during surgery. The main results of this trial, which ended in 2009, will be presented first. Taking advantage of both the POCI experience and the availability of new pixelated detectors, we are developing a new hand held gamma camera TReCam (Tumor Resection Camera). The first prototype offers a 49 x 49 mm2 field of view. It combines a 15 mm thick parallel hole collimator (with an efficiency of 1.2 10-4 and a spatial resolution of 6.08 mm at 50 mm), a 3 mm thick LaBr3:Ce crystal optically coupled to a Hamamatsu H9500 flat panel multianode photomultiplier tube (MAPMT). The read-out of the MAPMT is ensured thanks to a new specific integrated circuit called HARDROC2 (HAdronic Rpc Detector ReadOut Chip). This chip was initially designed for the Digital HAdronic CALorimeter (DHCAL) of the linear collider. Here, we report the first characterization of the TReCam imaging performances (both the spatial and energetic responses). These were measured using a data acquisition set-up, that included OPERA ROC chips and the LaBr3:Ce crystal plate (other scintillators were also studied). The TReCam presents an intrinsic spatial resolution of 1.5 mm (FWHM) and an energetic resolution of 17.4% at 122 keV. The new read-out by the HARDROC2 chip is currently being implemented. The ultimate performances of the camera prototype will be presented. Results from parallel studies (correction of the position linearity, influence of the depth of interaction) using both simulated (GATE v4.0.0) and experimental data will also be discussed.

283

V. Schulz1, T. Solf1, B. Weissler1, P. Gebhardt1, P. Fischer2, M. Ritzert2, V. Mlotok2, C. Piemonte3, N. Zorzi3, M. Melchiorri3, S. Vandenberghe4, V. Keereman4, D. Wirtz5, S. Renisch5, T. Schaeffter6, P. Marsden6 1 Molecular Imaging Systems, Philips Research Europe, Aachen, Germany 2 Circuit Design, University of Heidelberg, Mannheim, Germany 3 Foundation Bruno Kessler, Trento, Italy 4 MEDISIP-ELIS-IBBT, University of Ghent, Ghent, Belgium 5 Digital/Tomographic Imaging, Philips Research Europe, Hamburg, Germany 6 PET Imaging Centre, King's College London, London, UK Simultaneous PET/MR as one of the promising hybrid imaging modalities will substantially enrich the field of molecular imaging. Compared to PET/CT, the main advantages are the outstanding variable soft-tissue contrast and the capability to acquire functional and metabolic information in a simultaneous manner at the radiation dose of PET. Nearly perfect 4D registration will further enable multi-modality motion correction techniques which results in improved quantification of the PET and fewer artifacts which increases the overall accuracy of the diagnoses. The major challenge in combining these two modalities for clinical whole body applications is the invention of an MR-hard, MR-silent, and scalable PET-detector allowing compact integration into the clinical MR system. To investigate the reliability of this new technology, we decided to approach this complex field by designing preclinical scanners as feasibility prototypes. Until now both modalities were successfully integrated for preclinical and human head application by other groups. Negligible interferences between PET and MRI have been reported. However, current technologies do not offer the measurement of the time-of-flight (ToF) information in the sub-ns range which is the key driver for highly sensitive PET and state-of-the-art in clinical PET/CT. In this presentation, we describe the design of a preclinical PET/MR prototype scanner for rabbits. The scanner uses highly sensitive and fast Silicon-Photomultiplier arrays as detector elements, direct digitization with a dedicated ASIC and integrated digital singles processing. Special measures are taken to improve the MR-compatibility of this design. Measurements of a prototype detector were performed inside a 1.5T and 3T MRI-system. Even under high gradient-slew-rates and RF-pulses disturbance energy- and coincidence timing-resolution of 18% and 530ps (FWHM) have been measured, respectively.
Acknowledgement: The presented work is part of the EU FP7 project HYPERImage, Grant Agreement N 201651. M05-31:

M05-28:

A Preclinical PET/MR Insert for a Human 3T MR Scanner

Feasibility Study of Using Solid State Photomultiplier Array with Resistor Network Readout for SPECT Detector Development

X. Sun, Y. Shao, C. J. Bircher, K. A. Lan Imaging Physics, University of Texas MD Anderson Cancer Center, Houston, TX, United States It is well known that the solid state photomultiplier (SSPM) has many advantages including high gain and robust signal output, but suffers noise dominated by high dark counts, which can be a serious issue to single gamma photon detection. There is also a technical challenge of using large size SSPM and limited readout channels to decode array of small size crystals for achieving high spatial resolution. We report the progress of using the latest commercially available an array of SSPM for developing a high intrinsic spatial resolution detector for gamma camera and SPECT applications. The initial detector consists of one 8x8 array of 1x1x3 mm CsI(Tl) crystals and was optically coupled to a 4x4 matrix of SSPM. Each SSPM (pixel) has 3x3 mm photon sensitive area, 3640 micro cells, ~1 million gain and 4-8 MHz dark counts with different biases, and ~20% PDE. The inter-pixel gap and insensitive edge around the matrix are 0.2 mm, making it suitable for detecting closely packaged crystals and tiled for large area gamma-imaging detectors. A simple resistor network based signal multiplexing board was developed to investigate the method to reduce the number of readout channels while still provide suitable imaging performance. SNR as a function of number of pixels at different signal shaping times were measured. Different light sharing among different pixels were studied. Initial crystal maps measured with a Co-57 source and 3x3 pixels have shown that all crystals can be clearly identified except those at the edge, while 4x4 pixels gave worse results due to increased noises. Our initial study have shown that 1 mm intrinsic spatial resolution can be achieved with 3x3 mm size SSPM pixels through light sharing, and certain level signal multiplexing is applicable for reducing the electronic readout channels. It is expected that this will lead to exciting development of high resolution SPECT detectors to be reported in the conference.
M05-34:

G. Liu1,2, J. Liao1, Y. Wang2, J. Qi1 1 Department of Biomedical Engineering, University of California, Davis, USA 2 College of Electrical and Information Engineering, Hunan University, Changsha, China

Hierarchical Mumford-Shah Model for Dynamic microPET Image Segmentation

We proposed an organ-by-organ segmentation method for whole-body rodent dynamic PET images based on our hierarchical Mumford-Shah model for vector-valued image segmentation (HMSMv). First, we extract the shape parameters of Time-activity curves (TAC) of VOI in a dynamic PET image via a noise-normalized principal component analysis (PCA). Then we segment 284

one organ at a time in two steps. At the first hierarchy, we segment a rough volumes of interest (VOI) blurred by partial volume, motion, and spillover effects from the background via level set method. Then, we select a seed voxel that is farthest from the VOI boundary and is the least affected by spillover and motion blur. Finally, we perform the second level set based segmentation inside the initial segmented VOI to refine the segmentation result by evolving a sphere centered at the seed voxel. Once an organ is segmented, it will be removed from the PET image by assigning zero value to the voxels inside the VOI and the segmentation algorithm will move on to another organ. The segmentation algorithm has only one regularization parameter, which is easy to choose. Segmentation results of computer simulated data of the MOBY mouse phantom and real mouse dynamic PET images demonstrated that the proposed method is efficient and robust.
US NIH grant no. R01 EB000194, National Natural Science Foundation of China grant no.60872130 , 60835004, and Chinese postdoctoral fellowship. M05-37:

T. Mizuta1, K. Kitamura2, A. Ishikawa1, A. Ohtani1, K. Tanaka1 1 Research and Development Department, Medical Systems Division, Shimadzu Corporation, Kyoto, Japan 2 Technology Research Laboratory, Shimadzu Corporation, Kyoto, Japan

Scatter-Compensated Geometrical Components for Detector Normalization in Whole-Body PET

On a PET scanner consisting of block detectors, coincidence responses to scattered radiation are different from those to trues depending on the crystal pair position within a coincidence block pair. Furthermore, these differences vary with the radial position of the coincidence block pair. In component-based normalization, a scatter-compensated crystal interference factor is thus required in addition to scatter-compensated block-profile and intrinsic crystal efficiencies In this study, we propose a scattercompensated geometrical component in normalization by using an annulus phantom which provides true and scattered radiations over a large transaxial field-of-view, and evaluated the correction accuracy using three different-sized phantoms with a wholebody PET. Results showed that the proposed normalization method significantly reduce the ring artifacts in reconstructed images with different scattered/true fractions. The proposed algorithm worked well in different scattered/true radiation conditions, and was considered to be a practical normalization method in whole-body PET.
M05-40:

G. S. K. Fung1, W. P. Segars2, T.-S. Lee1, A. I. Veress3, G. T. Gullberg4, B. M. W. Tsui1 1 Department of Radiology, Johns Hopkins University, Baltimore, MD, US 2 Department of Radiology, Duke University, Durham, NC, US 3 Department of Mechanical Engineering, University of Washington, Seattle, WA, US 4 Life Science Division, E.O. Lawrence Berkeley National Laboratory, Berkeley, CA, US

Computer Generated Coronary Arterial Tree for Modeling Regional Myocardial Perfusion Defects for the Heart Model of the XCAT Phantom

The objective of this work is to develop a realistic model of myocardial perfusion defects (MPDs) based on a recently developed computer generated coronary arterial tree (CAT) that is superimposed on the heart model of the 3D XCAT phantom. The detailed and realistic 3D model of the CAT is generated by an iterative rule-based generation method that systematically utilized anatomic, morphometric, and physiologic knowledge of the CAT. The generation methodology allows different levels of detail in the generated vasculature, from left main and right coronary artery to arterioles. As the MPD generally has direct association with coronary stenosis, an effective and logical approach to define an MPD is based on the extent of the vasculature down stream from the stenosis. Due to the gradual change of the degree of ischemia and the irregular extent of the MPD, a complicated or voxelbased description scheme is needed to define the defect region. A perfusion model is needed to define the blood-tissue exchange kinetics, from intravascular advection to extravascular diffusion, between the CAT and the surrounding myocardium. The spatial distribution of the blood from the CAT to the myocardium is realistically modeled based on the perfusion zone constraint. To simulate a realistic MPD of a pathological heart, a certain degree of stenosis at a vessel segment was defined and the subsequent reduction of the blood flow of the segment and its daughter branches were deduced based on the connection data of the generated CAT. Then, using the pre-defined perfusion model, a complete description of the affected area was automatically generated based on the degree of blood flow reduction of the affected segments. In conclusion, the 3D XCAT phantom with realistic regional MPDs in the heart model provides a useful simulation tool in the study of myocardial perfusion using different medical imaging modalities.

285

M05-43:

M. Balda1, D. Niederloehner2, B. Kreisler3, J. Durst3, B. J. Heismann2 1 Informatik, Friedrich-Alexander-University, Erlangen, Germany 2 Siemens Healthcare, Forchheim, Germany 3 Physikalisches Institut 4, Friedrich-Alexander-University, Erlangen, Germany

Lookup Table-Based Simulation of Direct-Converting Counting X-Ray Detectors for CT

Medical Computed Tomography (CT) can benefit from direct-converting counting detectors. As yet there is little expertise with this type of detectors in commercially available clinical CT systems, a precise detector model is required for developing such a system. We introduce a lookup table-based simulation of counting detectors on X-ray photon level that allows studying the influence of detector parameters and efficiently evaluating proposed designs. It uses energy-resolved sinograms of incoming Xray photons as input data and generates photon counts for each channel and reading. The effects of Poisson noise, photon interactions, pulse generation, read-out electronics and electrode signal processing are covered. The photon interaction data as well as signal characteristics are provided in the form of detector-specific lookup tables. This approach offers the precision of Monte-Carlo simulations and the efficiency of model-based descriptions. Unlike standard Monte-Carlo simulations, it is capable of simulating whole CT-scans in a reasonable amount of time on a standard workstation. Due to this efficiency, the influence of detector parameters on image quality in the reconstructed image domain can be evaluated. The simulation is verified against measured data.
M05-46:

H. Fayad1, T. Pan2, C. Roux1,3, C. Cheze Le Rest1, O. Pradier1, J. F. Clement1, D. Visvikis1 1 INSERM, LaTIM U650, Brest, France 2 Department of Imaging Physics, M.D. Anderson Cancer Center, Houston, Texas, USA 3 Institut Telecom - Telecom Bretagne, Brest, France

A Patient Specific Respiratory Model Based on 4D CT Data and a Time of Flight Camera (TOF)

Respiratory motion is an important factor leading to errors and uncertainties in radiation therapy (RT). Solutions presented to date include modeling tumor and surrounding tissues motion. Having such a model is a key point to deliver, under breathing induced motion, less dose to the normal healthy tissues and higher dose to the tumor. Many continuous motion models have been developed based on 4D CT data. All these models are reconstructed using, for instance, an external respiratory signal (RPM or respiratory belt) or the diaphragm position. Possible limitations of these models are cases where the correlation between the respiratory motion and the corresponding surrogate is less reliable. In this paper, we describe an approach based on the creation of a continuous patient specific model that takes into account respiratory signal irregularities and uses, as surrogate, a surface map acquired using a time of flight camera. This model has been validated on three patients. Our results show that using the time of flight camera surface maps for the model reconstruction and application leads to higher accuracy compared to the use of a 1D respiratory signal.
M05-49:

W. Luo1, M. Dahlbom2, C. G. Matthews1 1 Naviscan Inc., San Diego, CA, USA 2 Molecular and Medical Pharmacology, UCLA, Los Angeles, CA, USA

A Phantom Comparison Study for Measuring PEM Scanner Sensitivity

We compare four different methods to measure the sensitivity of a PEM system. The first method is the NEMA NU 2 - 2007 standards, which uses a 70 cm long line source with aluminum sleeves. The total system sensitivity at the center of the field of view is 1006 cps/MBq. The second method is the NEMA NU 4 2008 standards that utilizes a 22Na point source and steps it along the system axis. The sensitivity is 163500 cps/MBq. The third method is to scan a 500 mL saline bag placed within the field of view containing 18F activity. The system sensitivity measured is 4652 cps/MBq. The last method employs a flood field phantom covering the field of view and filled with 18F solution. The sensitivity is 2800 cps/MBq when only the activity within the field of view is included in calculation. In summary, the flood field phantom is most practical to implement and best reflects the overall sensitivity, and at the same time shows the two dimensional sensitivity profile.
M05-52:

C. G. Song1, J. H. Seo2, J. H. Han3, J. U. Kang3 1 Division of Electronics Engineering, Chonbuk National University, Jeonju, Korea 2 School of Rehabilitation Medicine, Chonbuk National University, Jeonju, KOREA 3 Dept. of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, USA

High-resolution subsurface cross-sectional imaging of an articular cartilage using a Fourier domain common path optical coherence tomography

Introduction: In this work, we explored the efficacy of the Fourier domain common path OCT (FD-CPOCT) in a shorter near infrared range for 286

higher depth resolution with a 125m bare optic fiber probe to obtain characteristic layers and detailed lesion features in the knee cartilage that can be used in various kinds of early and accurate diagnosis and degeneration detections for arthritis. Method: The schematic view of experimental set-up for the FD-CPOCT is presented in Fig 1. Its spectra data are analysed or Fourier transformed to extract the depth (distance) information from the modulated optical spectra. We used an 835nm centre wavelength with a 37nm width SLD which has a corresponding coherence length or the depth resolution of 8m. To mimic the in vivo aqueous condition in the articular cartilage joint which is originally filled with synovial fluid, we submerged the piece of ex vivo chicken knee cartilage in the 0.9% NaCl (saline) solution. Results: In Fig. 2a, we can differentiate the characteristic layers of the articular cartilage which contains superficial zone (or tangential layer), middle zone (or transitional layer), and deep zone (or radial layer) in axial direction. In Fig. 2b, we can clearly observe some of featured structures in the cartilage such as defects or cysts which are marked with arrows and fissures in the tissue in the vertical direction marked with dotted rectangles. Conclusion: We have successfully evaluated the performance of the 0.8m Fourier domain common path optical coherence tomography that can be used for diagnosing the diseases related to the articular cartilage. We were able to observe very clearly both the pathological structure of the chicken cartilage and the internal and external micro defects such as fissures and fibrillations with finer resolutions than other comparable traditional image modalities.
This work was supported by the second stage of the Brain Korea 21 Project in 2009 and a Korean Science and Engineering Foundation (KOSEF) grant funded by the Korea government (MEST) (NO. R01-2008-000-20089-0) M05-55:

High Performance SPECT Camera Design

J. Dey Div Nuc Med, Dept Radiology, University of Massachusetts Medical School, Worcester, MA, USA SPECT is primarily used for cardiac applications like myocardial perfusion imaging. However, for SPECT, sensitivity is impaired due to the need for collimation. New designs have emerged with improved sensitivity over the standard gamma cameras currently used in the clinic. Among others, the UFC from GE or the multi-pinhole design by Funk et al use multiple pinholes. The advantages of pin-hole designs are that there are no moving parts reducing manufacturing and servicing costs. We propose a novel idea to improve resolution and/or sensitivity over the flat-detector multi-pinhole designs by using curved detectors fitted to pin-holes. This will obtain improved resolution over that of a flat-detector attached to the same pin-hole. We can then trade-off sensitivity vs resolution to get increased sensitivity for similar resolution as the flat-detector (by increasing hole diameter). We derived expressions for average resolution for cone-detectors and spherical detectors. We simulated the resolution and sensitivity improvement for examples case of curved detectors for an orbit designed for cardiac application. For a conical-design, for cardiac SPECT application, the sensitivity improvement over a multi-pin-hole system by Funk et al, was 40-75%. The sensitivity improvement of the curved detector system over the clinical systems currently used in the clinic for cardiac imaging using LEHR parallel collimation was a factor of 7-9, with system resolution close to that of LEHR. Further, there are tunable surfaceparameters if application calls for greater sensitivity and/or resolution, without increasing the compactness of the design. Increasing the distance from detector to pinhole for a flat detector pinhole system and trading the resolution improvement with sensitivity, for the same field of view and orbital constraints, significantly less number of pin-holes can be accommodated, resulting in overall loss of sensitivity over a system using curved detectors.
M05-58:

A Design of Pixellated Detector Module for C-SPECT

W. Chang, H. Liang, Y. Li, J. Oldendick, C. Ordonez Radiology, Rush Univ Med Ctr, Chicago, IL, USA We have designed and tested an improved pixellated detector module for our proposed C-SPECT system, which is designed for high performance cardiac imaging. The detector system has 15 detector modules assembled in a transverse array to form an openarc. It aims to achieve: large detection solid-angle for cardiac imaging, high stopping power at 140 keV, high packing fraction, 2D position-sensitivity for the whole detector area, high intrinsic spatial resolution, high count-rate performance, and costeffectiveness. Each module is a pixellated NaI(Tl) block housed in a thin Al housing backed by a glass exit-window. The detector block, 8 (w) x 16 (l) x 1 (t) cm, is packed from a 2D array of 2-3 mm square or rectangular NaI(Tl) elements with embedded reflectors in between. In the back of each module is a light-guide that is coupled to three columns of 13 39mm PMTs in a hexagonal pattern. The two outside columns of the PMTs are straddled between adjacent modules. Digital centroid and maximum-likelihood position estimation are used and have been evaluated in simulations and experiments. Fabrication of the blocks from NaI(Tl) slabs is conducted in a dry box in our lab with a variety of reflector materials. Compared with BGO/LSO blocks, the modules required hermetic housing and glass window pose different challenges for position sensing, especially at the edges of the block. Initial measurements on a 6x6 array of 3.5x3.5x10 mm pixels yield an average energy resolution of 12 % at 140 keV. Each of the pixels is adequately resolved in the central region of the PMT array. Investigation and refinements with 287

different sized pixels and larger arrays are on-going. Due to the relative low costs of NaI(Tl) and medium-size single-anode PMTs, this modular design has the potential to be the building block of the detector system of C-SPECT.
M05-61:

Depth of Interaction Encoding Detector with Phosphor-Coated Crystals and Silicon Photomultipliers

E. Roncali, H. Du, S. Saint James, Y. Yang, Y. Wu, S. R. Cherry Dept. of Biomedical Engineering, University of California-Davis, Davis, CA, USA Introduction: in recent years, including depth of interaction information in positron emission tomography detectors has been a very active area of development. Several depth-encoding detector designs have been presented to provide discrete or continuous depth of interaction information. Among them, a solution based on the use of a phosphor-coated scintillator crystal was proposed by Du et al. The phosphor modifies the decay times of detected pulses. Measuring the variation in decay times enables to get information regarding the depth of interaction. A resolution of 8 mm has been achieved using photomultiplier tube readout of the phosphor-coated scintillator array. Here we present the results from similar experiments conducted using Multi-Pixel Photon Counter (MPPC) for the readout of phosphor-coated crystals instead of PMTs. A critical parameter in that study is the sensitivity of the photodetector to the phosphor emitted light, which peak is at 550 nm. Considering the excellent performance of silicon photomultipliers, we expected an improvement in the depth of interaction resolution. Methods: a phosphor-coated LSO crystal was coupled to a MPPC and incorporated into a depth of interaction resolution measurement setup. This crystal was irradiated side-on at five different positions and 500 pulses were recorded at each depth. Results: a variation of 14.3 ns resulted from decay times measurements with a phosphor-coated crystal whereas a difference of 3.8 ns was obtained with uncoated crystal. Conclusions: significant variations in decay times were measured with phosphor-coated crystals coupled to MPPCs. This may be improved by using larger silicon photomultipliers and optimization of the experimental setup. Pulse shape discrimination methods will be implemented to evaluate the depth of interaction. Methods to add information such as rise time variation or spectral information in pulse shape discrimination will be investigated as well
M05-64:

Development of a High-Sensitivity Radiation Detector for Chromatography

J. S. Huber, S. M. Hanrahan, W. W. Moses, S. E. Derenzo, J. P. O'Neal, G. T. Gullberg Lawrence Berkeley National Lab, Berkeley, CA, USA Radionuclide techniques will be increasingly important for the development of new biofuels, since they can both uniquely characterize metabolic pathways and image large model systems. We are investigating the application of nuclear medical imaging tools and techniques to biofuel development, using high-sensitivity chromatographic radiation detectors and compounds radiolabeled with short-lived, cyclotron-produced, positron-emitting isotopes (e.g., 11C). The application of high-sensitivity radiation detectors to standard chromatographic techniques should allow for the measurement of numerous pathway constituents whose levels normally fall below detection limits of conventional instrumentation. In this paper, we describe a proposed parallelplane PET camera designed for HPLC (high performance liquid chromatography) that should have over 100 times greater sensitivity than a conventional HPLC radiation detector; this PET camera could be used to image radioactivity in the HPLC exit tube or column. We also present results from a preliminary study, using a high-sensitivity HPLC radiation detector comprised of 8 HR+ PET detector modules arranged into a parallel plane and read out with HRRT electronics. This high-sensitivity radiation detector was placed in line after a conventional radiation detector for HPLC (a small CsI:Tl scintillator crystal coupled to a PIN photodiode and read out in current mode). If we inject 381 Ci of [11C]carfentanil into the HPLC system, we see nearly identically shaped peaks from both radiation detectors. If we inject only 2.5 Ci of [11C]carfentanil, we measure a signal to noise ratio of 5 with the HR+ radiation detector but we are unable to detect the corresponding signal with the conventional radiation detector. We have therefore demonstrated that a high-sensitivity radiation detector, using commercial PET modules, could become an important tool for fundamental biofuel research.
This work supported by the Director, Office of Science, Office of Biological and Environmental Research, Medical Science Division of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. M05-67:

M. Wedrowski1, P. Bruyndonckx1, S. Tavernier1, C. Lematre1, Z. Li1, P. Rato Mendez2, J. Manuel Perez2, K. Ziemons3 1 Vrije Universiteit Brussel, Brussels, Belgium 2 El Centro de Investigaciones Energticas, Medioambientalesy Tecnolgicas, CIEMAT, Madrid, Spain 3 Forschungszentrum Juelich, Juelich, Germany

Use of Machine Learning Algorithms for Gamma Detection in PET

Gamma detection in nearly all commercial PET scanners is based on the use of a block detectors with a large number of small scintillator pixels, each pixel typically measuring 4x4x30 mm3. One of the important factors limiting the spatial resolution in PET scanners using this approach is the uncertainty about the exact depth of interaction of the gamma ray in the crystal. Using one large and undivided piece of scintillator, and deriving both the depth of interaction and the position of the gamma interaction from the light spread in the block, allows to overcome this limitation. At the same time this allows to increase the sensitivity of 288

the scanner. This approach tends to produce poor results if simple minded methods, such as centre of gravity, are used to determine to position of the gamma interaction in the crystal block. However, if machine learning algorithms such as neural networks are used, excellent performance is obtained. With the dramatic increase of computing power in recent years, this is now a realistic approach to gamma detection in PET scanners. We will present the result of experimental work on validating this approach with small prototype PET detector modules equipped with Hamatsu S8550 APD arrays as photodetectors. Results of studies on individual detector blocks, and with two detector blocks simulating a complete scanner, will be given. Some first results obtained with silicon PMTs will also be presented.
M05-70:

Time Resolution for Scattered and Unscattered Coincidences in a TOF PET Scanner

M. Conti, H. Rothfuss Molecular Imaging, Siemens Healthcare, Knoxville, USA The system time resolution of a PET scanner is a critical parameter, particularly since scintillating material science and electronics advances have made time-of-flight PET a viable technology, developing faster and faster detector systems. Correctly measuring the system time resolution of a PET scanner and understanding the components is now more relevant for TOF PET. Unscattered and scattered events have a different behavior, in terms of time distribution and time resolution. In particular, scatter coincidences have typically longer path length and lower energy, and tend to produce larger time resolution if included in the measurement. In this work, using both experimental measurements and Monte Carlo simulation, we investigate and discuss: the contributions of scattered and unscattered events to time resolution; the choice of a method to measure time resolution and its effect on the outcome; the time distribution of scattered and unscattered events, via analysis of experimental and simulated sinograms in the time dimension.
M05-73:

Time-of-Flight PET Detector Based on Multi-Pixel Photon Counter

C. L. Kim Imaging Technologies, GE Global Research, Niskayuna, NY, USA Since Geiger-mode multi-pixel APD can have better photon detection efficiency than PMT, we proposed that it could be a suitable photo-sensor for next-generation time-of-flight PET detectors. Last year, we have presented the coincidence timing resolution of 240ps result using single channel detectors based on 3 x 3 mm2 Multi-Pixel Photon Counter (MPPC) and 3 x 3 x 10mm3 LYSO crystals. MPPC is a Geiger-mode multi-pixel APD developed by Hamamatsu Corp. In this work, we will present the coincidence timing resolution of a time-of-flight PET detector based on a 4 x 4 array of 3 x 3 mm2 50um MPPCs. Since timing resolution depends on crystal size and shape, longer LYSO crystals with 3 x 3 x 25 mm3 dimension were used to be realistic. A preliminary result showed MPPC-LYSO detector could achieve less than 350ps coincidence timing resolution in FWHM between two of these block detectors. We will present the timing result using 4 x 4 array of 3.5 x 3.5 x 25mm3 LYSO crystals too in order to understand the aspect ratio effect between LYSO and MPPC sizes. We will also discuss about its readout challenges due to increased number of channels in a SSPM block detector compared to conventional PET detector based on PMTs.
M05-76:

C. Robert1, V. Rebuffel1, G. Montemont1, L. Verger1, I. Buvat2 1 CEA-LETI MINATEC Grenoble, Grenoble, France 2 IMNC-UMR 8165 CNRS, Orsay, France

Simulation-Based Optimization of a Parallel Collimator for Scintimammography Using a New CdZnTe Gamma-Camera Architecture (HiSens)

Because of the trade-off between spatial resolution and sensitivity resulting from the collimation, conventional Anger NaI-based gamma-cameras have reached their limits. To overcome these limitations, the HiSens architecture was proposed. Based on the use of a pixelated CdZnTe (CZT) detector, the architecture is characterized by an accurate 3D localization of the interactions inside the detector using a fine sampling of the CZT detector and the Depth Of Interaction (DOI) information. In order to reduce acquisition time or to improve image quality for a given acquisition time, a large aperture collimator is associated with the HiSens architecture. The aim of the present work is to determine the characteristics of a parallel hole collimator well tuned for scintimammography applications when using the HiSens architecture. To reach this goal, a Detective Quantum Efficiency (DQE) study was first carried out. DQE curve of the conventional LEHR collimator used in scintimammography was plotted for a realistic and maximum 5 cm source-to-collimator distance. Next, the collimator septal height and the 3D detector sampling (1 to 64 pixels per collimator hole/1 or 3 virtual layers) were varied. DQE curves were plotted for each configuration and compared to the LEHR reference to determine the best collimation (i.e. the one which allows to recover or to improve the spatial resolution of the conventional LEHR collimator with the maximal sensitivity gain). Finally, results were validated using simulations. DQE computations and numerical simulations both show that well-tuned collimation combined with the accurate 3D localization of interactions in the CZT detector accomplished by the HiSens architecture allows to significantly increase the detection efficiency

289

(by a factor of 3) while maintaining or even improving the spatial resolution in geometric conditions corresponding to scintimammography.
M05-79:

Design Study of a Lower Cost Ultrahigh Resolution High-Sensitivity PET for Neuroimaging

H. Li, W.-H. Wong, H. Baghaei, Y. Zhang, R. Ramirez, C. Wang, S. Liu, S. An University of Texas, M.D. Anderson Cancer Center, Houston, Texas, USA Current clinical PET with 4-6 mm intrinsic resolution (6-9 mm practical) limits many important brain studies. The objective of this study is to use our existing technology for realizing an ultrahigh resolution high-sensitivity PET with a lower cost for neuroimaging. This neuro-PET has a 54-cm detector ring diameter, a large 21-cm axial field of view for capturing the whole brain and carotid arteries for acquiring arterial input function for quantitating imaging. The system has 131,856 LYSO small detectors (1.4 x 1.4 mm2) coupled to 924 photomultiplier tubes (19-mm) with PMT-quadrant-sharing design. We propose to use 11-mm shallow detectors to reduce the deep-of-interaction image blurring and to reduce the costly LYSO material by half, and use the effective sensitivity gained by the time-of-flight (TOF) to compensate for the sensitivity loss by the shallow detectors. Despite the tiny detector cross-section restricting light output, our preliminary study shows that a 550-ps (FWHM) TOF time resolution was achieved (enabled by the excellent timing characteristic of our PQS detectors). Hence, there would be a gain of 2x effective sensitivity for the 18-cm brain. Monte Carlo simulations show transaxial image resolutions of 1.67, 2.37 mm at 1, and 9 cm respectively. This slow resolution degrading approves reduction of image blurring by shallow detectors. The large AFOV and small detector ring diameter give this system a 2x higher sensitivity than the typical clinical PET. This camera will use our existing detector technology, transformable gantry, and production-engineering tooling developed in the last few years. The production cost of this ultrahigh resolution neuro-PET would be less than $500K. This ultrahigh-resolution PET with resolution approaching MRI and CT, also allows more meaningful image correlation, and provides ultrahigh resolution functional imaging for small brain nuclei studies, which would open new doors for functional neuroimaging and neuroscience.
M05-82:

T. Yamaya1, E. Yoshida1, F. Nishikido1, N. Inadama1, K. Shibuya2, H. Murayama1 1 Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan 2 Graduate School of Arts and Sciences, University of Tokyo, Tokyo, Japan

Influence of TOF Information in OpenPET Image Reconstruction

We have proposed an "OpenPET" geometry, which consists of two axially separated detector rings. The central point of the field of view, where the highest sensitivity is obtained, is opened. The OpenPET mainly has three applications, namely, simultaneous PET/CT, extension of the axial FOV, and in-beam PET. In our previous report, we showed that axial spatial resolution, which is degraded with the extended gap due to the parallax error, can be recovered by use of depth-of-interaction (DOI) detectors. On the other hand, image reconstruction of the OpenPET is an incomplete problem because projection data do not satisfy Orlov's condition. Low-frequency components are missing in oblique LORs, i.e., LORs with large ring differences. The gap would suffer from loss of low-frequency components because the gap, where direct LORs (i.e., LORs in the same ring) do not exist, is imaged only from the oblique LORs. In this paper, we focused on time-of-flight (TOF) information, which is expected to compensate for the loss of low-frequency components in the gap. We investigated influence of TOF information in the OpenPET image reconstruction through numerical simulations. Simulated OpenPET scanner had dual detector rings (827.0 mm in diameter and axial length of 153.6 mm each) separated by a gap of 153.6 mm. We supposed that the detectors had a DOI capability of 8 layers and had a TOF capability of 400 ps FWHM resolution. For the non-TOF case, hot-spot objects, which are commonly seen in cancer diagnosis with use of FDG, were imaged without any artifacts, but objects containing more low-frequency components suffered from strong distortion. However, these artifacts were effectively reduced by using TOF information. These results showed that TOF information can compensate for low-frequency components missing in the gap.
M05-85:

Z. Li1, G. Vandersteen1, P. Bruyndonckx2, M. Wedrowski2, S. Tavernier2, J. Schoukens1 1 ELEC, vrije universiteit Brussel, Brussel, Belgium 2 DNTK, vrije universiteit Brussel, Brussel, Belgium

Extraction of Photon Interaction Position in a Monolithic Scintillator Block by Modeling the Scintillation Light Distribution

Numerous methods have been presented to obtain the spatial information for a monolithic scintillator block. Most of them require the acquisition of reference data sets and/or parameter calibration. The proposed method in this paper is to predict the 3D 511keV gamma ray interaction position by modeling the relation between the scintillation light source and the measured photo detector pixel signals. Reflections of the scintillation light on the side surfaces are modeled by including virtual light sources. Three different photo detector types/geometries were studied using Monte Carlo simulations: two Hamamatsu S8550 APDs coupled to an 20*20*10 mm crystal, four SensL SiPMT arrays attached to an 26.1*26.2*10 mm crystal and nine Hamamatsu S10931-050P SiPMTs reading out an 22.4*24*10 mm crystal. Taking into account all degrading effects due to noise and nonlinear behavior, we obtain an average spatial resolution less than or about 1 mm in the X-Y plane and a Depth of Interaction(DOI) resolution less 290

than 2 mm in all geometries. The geometry with the four combined SensL SiPMT arrays shows the best spatial resolution. (0.8 mm FWHM and 2.9 mm FWTM in X-Y plane, 1.3mm FWHM in DOI). The method has also been evaluated using experimental data collected from an 20*10*10 mm crystal coupled to a single Hamamatsu S8550 APD. An average FWHM resolution of 1.8 mm, including a photon beam size of approximately 1 mm, was achieved without any inter-pixel gain calibration which should bring the result closer to the simulated results. Experimental results of the X-Y spatial resolution and DOI resolution, obtained with a 20*20*10mm crystal attached to two Hamamatsu S8550 APDs will also be presented.
M05-88:

A new module-level parameter Interaction Ratio (IR) to evaluate the performance of detectors for

X. Kang1,2, Y. Liu1, Z. Gu1, Y. Xia1, S. Wang1, Z. Wu1, T. Ma1, Y. Jin1 1 Dept. of Engineering Physics, Tsinghua University, Beijing, China 2 Key Laboratory of Particle & Radiation Imaging (Tsinghua University), Ministry of Education, Beijing, China A high resolution detector made from 2D modular arrays of scintillation crystals is being developed for our small animal position emission tomography (MicroPET). Their characteristic flood field response (or flood histogram) has been segmented to Crystal Look-up table (CLT), which defines the matching relation between signal position of a detected event to a corresponding detector pixel location. Whether each annihilation event could match a well-separated pixel position in CLT, has play a direct role in imaging quality and brings significant influence to the gantry overall performance. Existing estimation parameters, such as photopeak positions, energy resolution, uniformity of counts cannot be readily adapted for quantitatively evaluate such modulelevel performance. In this paper, we present an estimation parameter Interaction Ratio (IR) to evaluate of the level of miscoding. With the help of IR, we quantitatively compare the performance of different type of scintillars, different glued reflector types of crystals, different LUT of detectors. The running performance evaluation of electrical system after 72 hours without baseline correction is then being quantitatively evaluated using Interaction Ratio, showing that not only it could be quantitatively evaluate the LUT segmentation performance, but also assess the performance of electrical system. Further, We establish the relationship with module-level parameter IR and the system-level parameter intrinsic spatial resolution FWHM. As proved as experiment, IR is a simple powerful tool to evaluate the how precise of the segmentation, to quantitatively assess the performance of the crystal identification, and hence, for detector characterization and routine quality control. We have being implemented IR tool to select crystal schemes, to inspect variety of electrical system, and to evaluate the performance of detectors for our MicroPET system, which bring out the effectiveness of IR as a power estimator.
M05-91:

PET

G. Llosa1, N. Belcari1,2, M. G. Bisogni1,2, S. Marcatili1,2, G. Collazuol1,3, M. Melchiorri4, C. Piemonte4, P. Barrillon5, S. BondilBlin5, N. Dinu5, C. de La Taille5, A. Del Guerra1,2 1 Department of Physics, University of Pisa, Pisa, Italy 2 INFN Pisa, Pisa, Italy 3 Scuola Normale Superiore, Pisa, Italy 4 Department of microelectronics, FBK-irst, Trento, Italy 5 Linear Accelerator Laboratory, Orsay, France The University of Pisa and INFN Pisa are developing a small animal PET scanner with Silicon Photomultiplier (SiPM) matrices as photodetectors. The proposed PET scanner will consist of four detector heads, composed of three detection layers. Each layer will be made of a continuous LYSO crystal 4 cm x 4 cm x 5 mm and a SiPM matrix structure as photodetector. The matrices are produced at the Center for Scientific and Technological Research (FBK-irst) in Trento, Italy. The successful results obtained with the first 16-pixel matrices have lead to the fabrication of matrices with 64 (8x8) SiPM elements in a common substrate, with different layouts. The matrices tested are 12 mm x 12 mm size, with readout on two opposite sides. The SiPM elements are 1.5 mm x 1.4 mm size and they have 840 microcells of 50m x 50m size. The matrices show an excellent uniformity in the breakdown points of all pixel elements. The first tests have been performed with continuous LYSO crystals, employing the ASIC MAROC2 and test board developed at the Linear Accelerator Laboratory (LAL) in Orsay (France) as data acquisition system. The energy resolution at 511 keV is 17% without correcting for the gain variations among the different pixels. The first position determination tests have been performed. A spatial resolution about 1 mm FWHM has been obtained with center-of-gravity algorithms. Reconstruction tests with maximum likelihood algorithms are being carried out. The results will be presented.

Monolithic 64-Channel Silicon Photomultiplier Matrices for Small Animal PET

291

M05-94:

W. C. J. Hunter1, R. S. Miyaoka1, L. R. MacDonald1, T. K. Lewellen1,2 1 Radiology, University of Washington, Seattle, WA, USA 2 Electrical Engineering, University of Washington, Seattle, WA, USA

Measured Temperature Dependence of Scintillation Camera Signals Read Out by GeigerMller Mode Avalanche Photodiodes

Signals of a Geiger-Mller mode avalanche photodiode (GM-APD) are strongly dependent on junction temperature. Consequently, we are developing a temperature-controlled GM-APD-based detector for positron emission tomography (PET) whose monitored temperature can be used to dynamically account for the temperature dependence of the output signals. Presently, we aim to characterize the output-signal dependence on temperature and bias for a GM-APD-based scintillation camera. Preliminary data for a Zecotek MAPD-3N GM-APD suggest a linear dependence of breakdown voltage on temperature (slope 0.071 V/C), corresponding to a rapid variation in gain with temperature (>10%/C). Furthermore, using two MAPD-3N to read out a pair of 3.5-by-3.5-by-20 mm3 Zecotek LFS-3 scintillator, we observe a moderate decrease (~200 psec) in coincidencetime resolution of a Ge-68 point source as the temperature was lowered from 23 C to 10 C. We also investigate changes in energy resolution with temperature.
M05-97: Optimal PET Acquisition Setting of I-124 with Siemens Inveon PET: Comparative Simulation Study with F-18 and microPET R4

A. R. Yu1,2,3, J. S. Kim1, K. Kim1, Y. S. Lee1,2,3, J. G. Kim1, S. K. Woo1, J. A. Park1, W. H. Lee1, H. J. Kim2,3, G. J. Cheon1 1 Molecular imaging Research center, Korea Institute of Radiological & Medical Sciences, Seoul, Rep. of Korea 2 Department of Radiological Science, College of Health Science, Yonsei University, Wonju, Rep. of Korea 3 Institute of Health Science, Yonsei University, Wonju, Rep. of Korea

Purpose: I-124 has a long half life of 4.2 days that is suitable for imaging over several days. However I-124 has a low positron branching ratio (23%). High-energy - photons (602 keV to 1,326 keV) are emitted in cascade with the positrons. These cascade - photons degrade the image quality. In this study, noise equivalent count rate (NECR) was measured with the various energy window settings to find the optimal energy window setting in I-124 PET on newly released Siemens Inveon PET. Sensitivity and scatter fraction (SF) on Inveon PET were also assessed with the same energy windows. In addition, NECRs of I-124 on microPET R4 (R4) scanner were also assessed for the comparison. Methods: Monte Carlo simulation studies were performed to find the optimal energy window setting. System performance such as the sensitivity and scatter fraction was measured in both I124 and F-18. NEMA NU-4 rat phantom and mouse phantom were simulated. Source activity was 1 MBq to 150 MBq. NECR in I-124 PET was calculated as following equation: NECR = T2/ (T + S + 2fR + fD) Where, T, S, R, D, and f is dirty coincidence corrected true, scatter, random, dirty coincidence count rate, and average fraction of the projection taken up by the object. Simulation for NECR were repeated under four different conditions (energy window: 250~550, 250~650, 350~550 and 350~650 keV). Results and Discussion: Within an energy window of 350~550 keV, dirty coincidence fraction was dramatically reduced by 50% on Inveon. Comparing two microPET systems, Inveon has 10 times higher NECR than those of R4. This would be due to improved electronics which enable to minimize the system dead time. Considering NECR of I-124, optimal energy window was 250~650 and 350~550 keV for mouse and rat, respectively. The proposed optimal energy window setting would boost the image quality of I-124 PET.
This work was supported by grants from Nuclear R&D Program (20090078289 & M20702010002-08N0201-00200) through the Korea Science and Engineering Foundation(KOSEF) grant funded by the Korean government(MEST). M05-100: Numerical and Measurement Evaluations of a High-Sensitivity Small-Animal PET Scanner in Lesion-Detection Tasks

Y. Dong1, C.-M. Kao2, C.-T. Chen2 1 Department of Biomedical Engineering, Illinois Institute of Technology, Chicago, IL, USA 2 Department of Radiology, The University of Chicago, Chicago, IL, USA

We have shown that it's in a fundamental significance to have high sensitivity in PET imaging in terms of lesion detection task. We developed such a high-sensitivity small-animal PET prototype system, which central sensitivity is up to 30%. The compact geometry of our system yields the pronounced depth-of-interaction (DOI) blurring and degrades image resolution. By accurately modeling the DOI blurring in the system response matrix, the image reconstruction can correct the DOI blurs. However, this spatial resolution recovery introduces the amplified noise into the resulting images. The high-sensitivity performance characteristic of our prototype system is compromised by the amplified noise. Thus, how can we objectively evaluate the performance characteristics of an imaging device, such as the sensitivity, in terms of the quality of the resulting images? It is necessary to objectively determine the system performance by using clinically task-based assessment of imaging quality. We have proposed a theoretical framework by performing lesion-detection task with computer observers to determine the "equivalent sensitivity" of the prototype system with respect to a reference small-animal PET system in identical detection tasks. This 292

theoretical framework is promising to produce a significance, that is, it allows one to compare the same performance characteristic of different systems in terms of an identical lesion-detection task, and yields a solid criterion for determining the performance parameters during the system development. Therefore, in this report, we will perform numerical and measurement evaluations on the high-sensitivity performance of DHAPET in lesion detection tasks to verify the developed concept-"equivalent sensitivity", and demonstrate the benefit of high-sensitivity performance in lesion detection tasks.
M05-103:

Coincidence Imaging with Monolithic Detector Blocks for a Human Brain PET Scanner

I. Sarasola Martin, P. Rato Mendes, M. Canadas Castro, P. Garcia de Acilu, J. M. Perez CIEMAT, Madrid, Spain We are developing a novel positron emission tomography (PET) scanner dedicated to the study of the human brain, to be used in a clinical environment. In this work we have evaluated the coincidence imaging performance of the monolithic detector block design that will be implemented in our prototype scanner. Results show that the detector blocks perform according to the expected specificatins and that they are suitable for PET application. Images of point sources obtained using a simple demonstrator composed of two detector blocks and associated electronics readout will also be presented.
M05-106:

A Hoffman Brain Phantom Lesion Study with the Transformable HOTPET Camera

H. Baghaei, H. Li, Y. Zhang, R. A. Ramirez, S. Liu, C. Wang, S. An, W.-H. Wong Experimental Diagnostic Imaging, University of Texas M. D. Anderson Cancer Center, Houston, TX, USA PET is a powerful imaging device for cancer diagnosis, neurodegenerative diseases, studying normal brain functions and emotional states. However, PET research of human brain demands scanners with high image resolution. Prerequisites for actually achieving high image resolution are high sensitivity over a large range of activities and high intrinsic spatial resolution. HOTPET camera is a transformable PET scanner that can operate as a whole-body PET or as a dedicated brain PET. In the brain mode, it has spatial resolution of 2.8-mm and a large 21-cm axial field-of-view for capturing the whole brain and carotid arteries input function for quantitating neural-activities. In the brain (whole-body) mode HOTPET has transverse FOV of 39 (60) cm. We performed a phantom study of brain lesion detection capability of HOTPET in both modes using the Hoffman brain phantom. We scanned 3 lesions with inner diameter of 3.95, 4.95 and 7.86 mm. The ratio of the activity concentration in lesions to the activity in background (gray matter) ranged from 1 to 20. Visual inspection of the images showed that 7.86-mm lesion was observable for activity ratio of 2 (lesion SUV of 8 and gray matter SUV of 4) or higher. The 4.95-mm lesion was observable with activity ratio of 4 in brain mode and 5.4 in body mode. The 3.95-mm lesion was observable for activity ratio of 6 for brain mode and 10 for body mode. Rebinning HOTPET body mode data in both axial and transverse directions to generate data for a whole-body camera with crystal size of 4.6 mm (twice the size of the HOTPETs crystals) showed that the observation of a 3.95 (4.95) mm lesion would require an activity ratio of 16 (8). This study showed that using a HOTPET similar dedicated brain camera compared to the clinical whole-body PET cameras would result in reduction of the detectible tumor volume by a factor of 4 for activity ratio of 6 or for a small lesion (<5 mm) would result in reduction of required SUV for detection by a factor of 2-3.
M05-109:

M. Safavi-Naeini1, D. R. Franklin2, M. Petasecca1, M. L. F. Lerch1, G. Moorhead3, P. Dunn3, R. Kirkham3, G. De Geronimo4, A. B. Rosenfeld1 1 Centre for Medical Radiation Physics, University of Wollongong, Wollgongong, NSW, Australia 2 School of electrical, computer and telecommunications engineering, University of Wollongong, Wollgongong, NSW, Australia 3 Manufacturing and Materials Technology, CSIRO, Melbourne, VIC, Australia 4 Instrumentation Division, Brookhaven National Laboratory, Upton, NY, usa Silicon photomultipliers (SiPMs) are solid state photodetectors whose excellent gain and timing characteristics are advantageous for high resolution PET imaging. This paper presents the first coincidence measurements performed with a new combination of LYSO scintillator, SiPM detector and digital data acquisition system being developed at the University of Wollongong's Centre for Medical Radiation Physics (CMRP) in collaboration with CSIRO and Brookhaven National Laboratory. The solid-state detector is a SENSL SPMMicro 3035 SiPM device (3 mm x 3 mm) coupled to a matched LYSO scintillator in conjunction with a conventional photomultiplier tube coupled to a second LYSO scintillator. The preamplification stage for both channels is based on a fast wideband Darlington amplifier (GALI-5+) with a fixed gain of 17.5 dB. This is coupled to two inputs of a 32 channel ASIC (the SCEPTER ASIC developed at Brookhaven National Laboratory) for timing and pulse height analysis. Amplitude and timing information for each pulse is then logged using the HYMOD data acquisition board developed by CSIRO, and is analysed off-line for coincidence detection. Results are presented showing that the timing resolution achieved was approximately 12.8 ns FWHM with 24% FWHM energy resolution at 511 keV.
The authors would like to thank Mr Murray Jensen and Mr Peter Ihnat for their assistance.

Timing in Silicon Photomultiplier Detectors for Small Animal PET Imaging

293

M05-112:

X. Deng1, T. Ma2, R. Yao1 1 Department of Nuclear Medicine, The State Univ. of New York at Buffalo, Buffalo, New York, USA 2 Department of Engineering Physics, Tsinghua University, Beijing, China

Determination of the Septa-to-Crystal Position in Axial Direction for a Slit-Slat Collimator SPECT

For a SPECT scanner with a slit-slat collimator, the accurate knowledge of the septa-to-crystal position is important to obtaining desired axial resolution in the reconstructed image. To determine this parameter, we derived the geometrical model of the axial point spread function (PSF) which is a function of the septa-to-crystal position. By measuring a series of point source projections and minimizing the difference between the measured PSFs and geometrically modeled ones, the optimal estimation of septa-tocrystal position can be obtained. We confirmed the uniqueness of the estimation by analyzing multiple numerical simulations using a singular value decomposition (SVD) based approach and showed that the estimation error was less than 0.04mm. This accuracy is adequate for a SPECT system with ~ 1mm axial resolution. While we showed that the septa-to-crystal location introduced a variation of axial PSF FWHM by nearly 1mm, the impact on the axial resolution of a reconstructed phantom image was not visually significant and needs to be quantitatively assessed.
M05-115:

Multimodality Imaging Method for Identifying Metastatic Tumors in Mouse Liver Using Gated Small Animal PET/CT

S. K. Woo, K. M. Kim, T. S. Lee, J. H. Jung, Y. J. Lee, W. H. Lee, J. A. Park, J. S. Kim, J. G. Kim, G. J. Cheon Molecular Imaging Research Center, Korea Institute of Radiological and Medical Science, Seoul, South Korea Liver is the most common and critical site of distant metastasis of colorectal carcinoma. The combination of gated 18F-FDG small-animal PET and gated contrast CT image provides quantitative in vivo. The aim of this study was to obtain multimodality images with PET/CT image by using motion correction method. In the liver metastasis model, LS174T cells were injected into the spleen of nude mice. Respiratory gating was realized with the help of an external trigger device (BioVET) synchronized with the list mode acquisition. PET imaging studies was performed with a dedicated small animal R4 PET scanner. PET imaging was started 60 min after the administration of 7.4 MBq/0.1 mL of 18F-FDG via tail vein injection. Mice were placed on an acryl plate for fusion between R4 and micro CT images. Gated contrast enhanced CT images were obtained at three hours after injection of contrast agent. The CT images obtained with a micro Inveon CT scanner can be used to improve the anatomical localization of uptake in the small animal PET images. Post acquisition, the list-mode data are re-binned into respiratory gated sinograms for reconstruction. Fused image in liver tumor model showed a good correlation of images from both PET and CT. The gated PET/CT images show the better image quality, compared to that with no gating, in terms of contrast and size of interested region, particularly with high uptake of FDG and contrast. In the gated CT image, the SNR of 0 bin, 2 bin, 4 bin and 8 bin number on the CT image were 5.4, 5.5, 4.9 and 5.3, respectively. The contrast values of each bin number the CT image were 2.77, 3.93, 4.99 and 3.8, respectively. The small size of liver metastasis tumor (1 mm) was clearly visualized by gated PET/CT image. Multimodality imaging with small animal PET and CT image could be useful for the detection of metastatic tumor. Imaging of mouse liver model with 18F-FDG small animal PET and contrast CT should be considered motion correction method.
M05-118:

PEDRO: a Hybrid Small Animal Imaging System

M. R. Dimmock, J. E. Gillam, T. E. Beveridge, J. M. C. Brown, R. A. Lewis, C. J. Hall Monash Centre for Synchrotron Science, Melbourne, Australia The Pixelated Emission Detector for RadiOisotopes (PEDRO) is a hybrid emission imaging system designed for the measurement of single photon emission from small animals. The PEDRO is based on the combination of mechanical and electronic collimation (through the Compton effect). It utilizes a Compton-camera stack geometry situated behind a multiple pinhole collimator which will ultimately provide a coded aperture mask to modulate the incident radiation. The geometry is intended to maximize detection e

294

ciency over a wide range of photon energies whilst maintaining the enhanced resolution of pinhole imaging. In this way it is possible to decouple the relationship, expected in pinhole imaging, between resolution and sensitivity. This paper presents GEANT4 simulations of the geometry and preliminary results that have led to the design of an experimental system intended for both verifcation and optimization. The expected experimental geometry and hardware for the PEDRO are described, and the estimated resolution and timing characteristics of the completed system are presented.
M05-121:

A Hybrid Rotation-Translation Scan Scheme for In Vivo Animal SPECT Imaging on a Stationary

Y. Xia1, T. Ma1, X. Deng2, S. Murali2, R. Yao2 1 Department of Engineering Physics, Tsinghua University, Beijing, P.R.China 2 Department of Nuclear Medicine, University at Buffalo, State University of New York, New York, USA We investigated a novel hybrid rotation-translation (HRT) scan scheme for in vivo animal imaging on a stationary slit-slat animal SPECT system. With this hybrid scan scheme, the object is first translationally moved within the SPECT field of view (FOV) in a step-by-step manner. At each step, the object is rotated 2 to 5 times over a small angular range (20 deg). The projection sampling completeness and uniformity of the proposed HRT scan scheme were studied. By combining the translational and rotational movements, similar sampling completeness and uniformity can be achieved comparing to the conventional rotation scan scheme. Monte Carlo simulation and phantom experiments demonstrate that the reconstructed image resolution with a chosen HRT orbit is comparable to that with the conventional rotation scheme. An in vivo animal SPECT study was successfully carried out on a tumored mouse. We conclude that in vivo imaging on our stationary SPECT system is feasible with a HRT scan scheme.
M05-124:

Scanner

V. Lee1,2, G. J. O'Keefe3, A. Schubert1,2, B. A. Sobott1,2, D. Pook4, A. Scott3, R. P. Rassool1 1 School of Physics, University of Melbourne, Parkville, VIC, Australia 2 CRCBID Cooperative Research Centre for Biomedical Imaging, Bundoora, VIC, Australia 3 Centre for PET, Austin Hospital, Heidelberg, VIC, Australia 4 Ludwig Institute for Cancer Research, Austin Health, Heidelberg, VIC, Australia

Imaging Iodine125 Labelled Nanoparticle Distribution in Mice with PILATUS II.

Iodine-125 (125I) is commonly used as radioactive label in antibody studies in small animal imaging and has a dominant x-ray emission in the energy range 27 - 35 keV. In animal models, 125I based imaging lends itself readily to translation into human imaging and/or treatment using other isotopes such 131I and 124I. The PILATUS detector is a single photon counting hybridpixel detector with pixels of size 172x172um and a silicon substrate thickness of 300 um. A typical detector module is 8x5 cm and contains 100,000 pixels and arrays of up to 6x106 pixels have been deployed for use on synchrotron beamlines. The excellent spatial resolution of this detector, together with its compact design make it a suitable choice for a 125I small animal imaging system in comparison with existing clinical imaging systems. Preliminary measures are presented in which a PILATUS 100K module was used to image mice injected with I-125 labelled nanoparticles with a radioactivity ranging from 12.5 uCi to 500 uCi to determine the live biodistribution of nanoparticles in the mouse body. Incident photon direction was defined by a parallel hole collimator array. Results revealed the trapping of the nanoparticles in the liver, not reaching their tumour target destination; this was subsequently confirmed with 18F labelled nanoparticles using a PET-CT system. Tomographic datasets were also obtained; their reconstruction yielded recognisable organ shapes but resolution is limited by the collimator array. Imaging with PILATUS provided a live, non-invasive way to image for nanoparticle distribution.
M05-127:

Y.-C. Tai1, L.-J. Meng2, H. Krawczynski3, Y. Yin1, S. Komarov1, H. Wu1, J. W. Tan2, Q. Li3, A. Garson III3, J. Martin3 1 Radiology, Washington University in St. Louis, St. Louis, MO, USA 2 Nuclear, Plasma and Radiological Engineering, University of Illinois at Urbana-Champaign, Urbana-Champaign, USA 3 Physics, Washington University in St. Louis, St. Louis, MO, USA

Initial Study of a Sub-500-Micron Resolution PET Insert Device Based on Finely-Pixilated CZT Detector

In this contribution we report on the initial development of a finely pixilated CZT insert device with 350 m anode pitch that may improve the image resolution of an animal PET scanner to <500 m. In previous work, we had used a scintillator-based insert to demonstrate an image resolution of ~1 mm. We propose to develop pixilated CZT detectors with 350 m anode pitches and a readout ASIC with 2048 channels of charge-sensitive pre-amplifiers and digital readout as the core detector technology for a sub500-m resolution PET insert device. In this initial study, we first contacted a CZT substrate of 19.4 x 19.5 x 4.95 mm3 with 55 x 55 anode pixels of 250x250 m2 each with 350 m pitches. This detector has been bonded to an existing ASIC originally developed by the UIUC group in collaboration with Ajat in Finland for SPECT applications. The ability to register the timing and position of events is currently being evaluated. In order to characterize the basic performance (energy, timing, intrinsic spatial resolution and DOI resolution) of CZT detectors with the proposed 350 m anode pitches in-house, we also contacted a special 295

anode pattern with 9 anode pixels of 350 m pitches and traces connecting anodes to readout pads to readout the anode signals with our existing data acquisition systems. A Monte Carlo simulation package has been developed to model the -ray interaction with CZT and signals induced at anodes and cathode as a function of time. This will allow us to study the spatial resolution and charge sharing of pixellated CZT detectors of different thickness and pitches. Results from (1) CZT with 350 m pitched anodes bonded to the 2048-Ch ASIC; (2) CZT with a special anode pattern to mimic 9 anode pixels of 350 m; and (3) Monte Carlo simulation study of CZT with 350 m anodes will be compared to characterize the performance of such detector and its applicability to the proposed sub-500-m resolution PET imaging applications.
This research was supported by the Office of Science (BER), U.S. Department of Energy, Grant No. DE-FG02-08ER64681. M05-130:

Development of a High Resolution Image Guided Microirradiator (microIGRT)

E. W. Izaguirre, B. L. Kassebaum, J. Birch, I.-T. Su, D. Low Radiation Oncology, Washington University in Saint Louis, Saint Louis, MO, USA Small animal in vivo structural (microCT and microMRI) and functional (microSPECT and microPET) imaging has fostered the development of radiopharmaceuticals and fundamental cancer research. Similar advances in the fields of radiobiology, radiotherapy planning, radiosensitizers, and radiation therapy are anticipated if dedicated small animal micro-conformal irradiation instrumentation is developed. We will present the design and development of a small animal image guided microirradiator (microIGRT) that has been constructed using a high resolution microCT for anatomical imaging and treatment planning with an orthovoltage source for high resolution conformal microirradiation. Each of the subsystems, the microCT and the microirradiator, have separate gantries which are placed in a coaxial instrument architecture and are sequentially operated. The on board microCT subsystem is a cone beam imaging tomographic scanner constructed using a micro focus and a flat panel amorphous silicon detector. The micro focus x-ray source is an 80kVp tungsten anode and is operated at 0.1mA. The source focal spot is 75 x 75m2. The flat panel amorphous silicon detector has 1024 x 1024 pixels and the detector array has a total active surface area of 13x13mm2. The orthovoltage irradiator subsystem was designed and constructed using a 320kVp x-ray source with dual focus spots of 0.4 x 0.4mm2 and 1 x 1mm2. The smaller focal spot can be operated up to a maximum power of 800W and the large focal spot has a maximum power of 1800W. The orthovoltage beam is collimated using orthogonal jaws and exchangeable apertures. The system has been designed to deliver a maximum dose rate of 40Gy/min. An axial motorized animal bed transfers the animal from the microCT imaging subsystem to the microirradiation subsystem. Treatment planning is performed during the transfer of the small animal from the microCT imaging subsystem to the orthovoltage microirradiator.
M05-133:

M. Sibomana1, S. H. Keller1, S. Holm1, P. M. Bloomfield2, S. Blinder3, S. B. Hansen4, C. J. Michel5 1 PET Center KF 3982, Copenhagen Univ. Hosp., Rigshopitalet, Copenhagen, Denmark 2 Centre for Addiction & Mental Health, Toronto, Canada 3 Pacific Parkinson's Research Centre, Vancouver, Canada 4 PET Center, Aarhus University Hospital, Aarhus, Denmark 5 Siemens Healthcare Molecular Imaging, Knoxville, USA

Component-Based Normalization for the HRRT for Sinogram-Mode Reconstruction

The Siemens HRRT scanner is a complex PET scanner, using 8 LSO/LYSO panel detectors and Depth Of Interaction(DOI) with a total of 119908 detector crystals. The Direct Normalization (DN) is the standard method used for sinogram-mode reconstruction. We have implemented a Component-Based Normalization (CBN) method developed for listmode reconstruction. Both methods use a rotating 68Ge rod source and depending on the source activity, an acquisition of 2-5 days is required for DN while only a 1 hour acquisition is required for CBN. Even with the reduction in acquisition time for the CBN, our main requirement was that the new method should improve, or produce similar image quality as the standard method. The other advantage of CBN over DN is a method to correct for crystal efficiencies changes with count-rate due to pulse pile-up in the back layer. A first implementation of CBN was distributed as part of the HRRT Users software and tested at several sites. Two artifacts were reported using this implementation for a uniformity phantom acquisition; firstly, an octogonal artifact was visible on the sum image, and secondly, a bias was reported along radial profiles. These artifacts were not visible on the image reconstructed with the DN. We analyzed the differences between the DN and CBN normalizations and found two effects that produced the artifacts; firstly DN corrected for the offset between the rod source rotation center and the scanner FOV center whereas CBN assumed a perfectly centered rotation, and secondly the ratio between DN and CBN showed an additional solid angle effect of about 10% for LORs with small incidence angle. A second CBN implementation including the corrections for these two effects resulted in a more uniform image compared to the image reconstructed with DN.

296

M05-136:

Design and Feasibility Studies of a High-Resolution and Low-Cost Small Animal SPECT System

T. Dai, S. Wang, Y. Liu, Z. Wu, Y. Jin, T. Ma Department of Engineering Physics, Tsinghua University, Beijing, P.R.China In this study we investigate the feasibility of performing small animal SPECT imaging on a clinical SPECT scanner with a dedicate pinhole collimator. Analytical formulae are used to calculate spatial resolution and sensitivity given a pinhole collimator design scheme. Two optimal pinhole designs are proposed in terms of achieving the best trade-off between spatial resolution and sensitivity. One consists of one pinhole aperture and the other one has 5 pinholes. In the 5-pinhole design scheme, the pinholes are arranged so that the peripheral pinholes are tilted towards the center of field of view (FOV). Each pinhole only covers a portion of FOV to maximize the sensitivity in the central region of FOV and to make use of the entire detection area without much overlapping. Monte Carlo simulation studies confirm the predicted spatial resolution values from analytical formulae. Image reconstruction results of a simulated ultra-micro hot-rod phantom demonstrate that 0.75 mm hot rods can be clearly identified with the 5-pinhole collimator. Despite that the intrinsic resolution of the SPECT detector is only 3.55 mm, its large detection area allows large magnification and good spatial resolution. We conclude that high-resolution animal SPECT imaging can be performed through the proposed approach with relatively small cost for clinical SPECT users comparing to purchasing a dedicated small animal SPECT.
M05-139:

Phantom Construction Using a Conventional High Resolution Photo Printer

M. Peterson, A. Orbom, S.-E. Strand Dept. of Medical Radiation Physics, Lund University, Lund, Sweden Introduction: We are developing a method to construct phantoms for parametric evaluation of imaging systems using a conventional high resolution ink-jet photo-printer. This method has been proven to provide a flexible and easy way to produce geometries for determining efficiency, spatial resolution, linearity and homogeneity of imaging systems. Our aim is to attempt to adapt a high resolution printer and produce point- and linesources below 50 m. Methods: The printer is a Canon PIXMA iP4500 with a 9600 x 2400 dpi resolution with separate CMYK cartridges. These where emptied and refilled with radioactive ink and phantoms where printed on high resolution paper. We investigated the activity/area profile during continuous printing and the relationship between color fill and activity content. Regular printouts were also examined in a microscope to estimate the printing resolution. Results: Line and point sources can be made as small as 50 m and 30 m respectively. Radioactive ink injected into the ink reservoir gave a slow rise in activity concentration and reached a plateau. The activity concentration as a function of fill percentage was shown to have an exponential relationship. Radioactive ink injected directly into the ejection nozzle enabled phantoms with higher ink activity concentrations. Two phantoms were successfully printed and imaged by a digital autoradiography system and a -SPECT system respectively. Conclusions: We have demonstrated a working method for phantom construction using a conventional photo printer and have shown the capability to print line sources below 50 m in width and 30 m point sources. Stable values for ink activity per area makes the phantoms suitable for determining system linearity and homogeneity. Future work includes trying to increase the printed activity per area by nozzle injections and to use printouts for parametric testing of imaging systems.
M05-142: Acquisition-Duration Dependence of Reproducibility of Myocardial Perfusion in Rats with microSPECT

R. G. Wells1, M. Kordos1, J. Lockwood1, J. Strydhorst1, P. Fernando2, C. Bensimon2, T. D. Ruddy1 1 Cardiac Imaging, University of Ottawa Heart Institute, Ottawa, ON, Canada 2 MDS Nordion, Ottawa,ON, Canada

MicroSPECT/CT imaging is a powerful tool for studying cardiac disease and evaluating new therapies. A benefit of small-animal imaging is the ability to do serial imaging but multiple imaging sessions puts greater demand on camera resources and increases the need to minimize imaging time. While the increased sensitivity of modern multi-pinhole/multi-detector cameras have reduced both imaging times and required injected doses, accurate estimates of the precision of data measurements are still required to properly design experiments. In this study, we examine the impact of acquisition time on the reproducibility of ungated LV volume and myocardial perfusion in rats. Methods: We performed cardiac imaging on rats using 99mTc-tetrofosmin with a 4head 9-pinhole/head (aperture=1.5mm) small-animal NanoSPECT camera (Bioscan). 4 rats (303+/-54g) were injected with 2.0+/0.3 mCi of activity and imaged for 60min in 10min intervals. Imaging was repeated 1/week for 5 weeks. Projection data were summed to create acquisitions of 10min, 20min, 30min and 60min duration. Images were reconstructed using an iterative algorithm and a fixed filter. All images were imported into our clinical environment and analysed using 4DMSPECT (Ann Arbor, MI). Myocardial volume was adjusted for weight and myocardial perfusion was assessed using a normalized 17-segment model. Results: The standard deviation (SD) in the mean myocardial volume of the 4 rats over 5 weeks ranged from 23l at 10min to 27l at 60min (mean volume = 334l). The mean SD of the volume between rats was 43-47l. The mean perfusion in the 17 segments ranged from 61-93% and the mean within segment variability between scans decreased from a SD=4.7% at 10min to 297

3.5% at 60min duration. The mean segment SD between rats went from 5.2% at 10min to 4.4% at 60min. Conclusions: This study provides data on the variability in the reproducibility of measurement of rat myocardial perfusion as a function of scan duration.
M05-145:

Y. Gu1,2, C. S. Levin2,1 1 Electrical Engineering, Stanford University, Stanford, CA, USA 2 Radiology, Stanford University, Stanford, CA, USA

Effects of Multiple Interaction Photon Events on Measuring Position and Arrival Time in a CZTbased High-resolution Small Animal PET System

We are developing a 1 mm resolution small animal PET system using cadmium zinc telluride (CZT) photon detectors. The detectors are 40x40x5 mm^3 monolithic CZT crystals patterned with a cross-strip electrode design creating 1x5.4x1mm^3 voxels, and are capable of positioning the 3-D coordinates of individual photon interactions. Detectors are arranged to create an 8x8x8 cm^3 box-shaped field of view. Challenges exist in achieving accurate event line of response (LOR) positioning, arrival time, and coincident photon pairing for high event rates due to CZTs relatively low effective Z (high probability of Compton scattering), its finite charge drift velocity, and effects of the electrode electric field configuration on charge collection. This study attempts to study the effect of multiple-interaction photon events on the system's LOR positioning and photon sensitivity, in order to guide the design of algorithms for accurate estimation of position and arrival time for these types of events. Monte Carlo simulation was performed using a realistic mouse phantom (300 uCi uniformly distributed) centered on the system axis for a simulated acquisition time of 1 sec, during which 1013104 (100%) photons interacted at least once with the detector volume (71.6% had summed energy within 511keV+/-3%). Results yielded a mode inter-scatter distance of 1.8 mm, causing 17.4% of photons to have indistinguishable multiple interactions within the same 1x5.4x5 mm^3 detector voxel. Approximately 37.3% of photons interacted more than once in at least one of the detectors they traversed, giving rise to challenges in positioning with cross-strip readout. However, 29.7% out of the 37.3% involve just 2 interactions; we are developing an algorithm to accurately position photons and identify coincidence pairs to achieve 93.4% photon retention rate by resolving positioning ambiguity associated with these two-interaction photons.
M05-148:

S. G. Staelens1, T. Wyckhuys2, S. Deleye1, H. Hallez1, S. Vandenberghe1, B. Van Nieuwenhuyse2, K. Vonck2 1 Medical Signal and Image Processing group, Ghent University - IBBT, Ghent, Belgium 2 Laboratory for Clinical and Experimental Neurophysiology, Ghent University Hospital, Ghent, Belgium A stimulus on/off imaging study is often used to evaluate the brain's response to a presented visual, electrical or chemical trigger. Clinical software and human templates already exist but given the recent advent of ultrahigh resolution SPECT and PET, a larger need rises for a post processing platform to perform these subtraction molecular imaging studies also in small animals. We have designed such a multimodal framework to perform SPECT activation studies in rats thereby making use of CT and MRI for anatomical land marking. Our software solution is a combination of Amide, MRIcroN and a custom made Matlab implementation. We have studied the performance of different deep brain stimulations for which 6 rats were implanted with a multi-polar stimulation electrode in the right hippocampus. Each animal underwent a 99mTc-HMPAO SPECT with the Milabs U-SPECT-II and a CT scan with the GMI X-O CT before and after stimulation. Two line markers in oblique positions filled with low activity of 125I are used to register the SPECT and CT images. Afterwards the animals were sacrificed, their electrode was removed and a MRI scan was performed using the wrist coil of a Siemens Trio 3T. A semi-automated five step procedure delivers the activation map: (i) first the stimulus-on CT and the MRI are registered to the stimulus-off CT followed by (ii) the fusion of the off/on SPECT scans with their off/on CT counterparts. From the MRI, (iii) the rat brain is extracted, which is used as a mask for the calculation. Afterwards, (iv) both off/on SPECT scans are normalized and subtracted within this MRI brain mask. Finally, (v) the Z-score, representing the activation map, is achieved by dividing the result with the standard deviation of the masked stimulus-off SPECT. This semi-automated approach allows the experimental neuroscientist to draw conclusions on the location, spatial extent and intensity of the small animal brains response to the stimulus.
This work was supported by the Fund for Scientific Research Flanders (FWO, Belgium) and by Ghent University.

High Resolution SPECT for Brain Activation Analysis in Small Animals

298

M05-151:

F. D. Popota1,2, P. Aguiar3, Y. Fernandez4, C. Lois3, D. Pareto1,5, D. Ros2,5,6, J. Pavia5,6,7, J. D. Gispert1,5 1 Institut dAlta Tecnologia (IAT)-PRBB, Barcelona, Spain 2 Unitat de Biofsica i Bioenginyeria, Universitat de Barcelona, Barcelona, Spain 3 Grupo de Investigacin en Radiofisica, Universidad de Santiago de Compostela, Snatiago de Compostela, Spain 4 Centre d'Imatge Molecular Experimental CIME-CETIR, Barcelona, Spain 5 CIBER de Bioenginyeria, Biomaterials y Nanomedicina (CIBER-BBN), Zaragoza, Spain 6 Institut d'Investigacions Biomdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain 7 Servei de Medicina Nuclear, Hospital Clinic, Barcelona, Spain

Comparison of NEMA NU 4-2008 Vs NEMA NU 2-2001 for the Performance Evaluation of the microPET R4 System

In the emerging field of molecular imaging, small animal PET systems continue to play an important role in biomedical research and pharmaceutical development. In order to compare different systems, the performance evaluation has to be done by specifically standards. The purpose of this work is to compare the performance evaluation of the microPET R4 scanner according to the new NEMA NU 4-2008 standards for small animal PET systems against its previous evaluation according the adapted clinical NEMA NU 2-2001. All measurements were made for an energy window of 350-650 keV. Spatial resolution of 2.12 mm FWHM in the radial direction, 2.66 mm FWHM in the tangential direction and 2.23 mm FWHM in the axial direction was measured in the centre of the axial FOV with a 1-mm diameter Na-22 point source. Scatter fraction for mouse and rat phantom were found to be 9.5% and 22% respectively. The maximum noise-equivalent count rate (NECR) was 110 kcps at 121.3 MBq for the rat phantom 287 kcp at 129.8 MBq for the mouse phantom. The absolute sensitivity of the scanner was found to be 2% at the centre of the field of view. In some cases, the two protocols differ in the proposed methodology for the evaluation of the performance. Also, the fact that wider energy windows were applied and bigger phantoms were used, justify more the discrepancies between the results presented here and previously published ones.
M05-154:

A Multimodal Approach to Image-Derived Input Functions for Brain PET

E. K. Fung, B. Planeta-Wilson, T. Mulnix, R. E. Carson PET Center, Yale University, New Haven, CT, USA Many methods have been proposed for generating an image-derived input function (IDIF) exclusively from PET images. The purpose of this study was to assess the viability of a multimodality approach utilizing registered MR images. 3T-MR and HRRTPET data were acquired from human subjects. Segmentation of the carotid arteries was performed in MR images using a 3D level sets method. Vessel centerlines were extracted by parameterization of the segmented voxel coordinates with a polynomial curve fitted to the data. These centerlines were subsequently re-registered to static PET data to maximize the accurate classification of PET voxels in the ROI. The accuracy of this approach was assessed by comparison of the area under the curve (AUC) of the IDIF to that measured from conventional automated arterial blood sampling. Our method produces curves similar in shape to that of blood sampling. The mean AUC ratio of the centerline region was 0.37 before re-registration and 0.50 after registration. Thus, partial volume correction is still necessary. This study suggests that the combination of PET information with MR segmented regions will demonstrate an improvement over regions based solely on MR or PET alone.
M05-157:

SPECT Image Classification Based on NMSE Feature Correlation Weighting and SVM

R. Chaves, J. Ramirez, J. M. Gorriz, M. Lopez, I. Alvarez, D. Salas-Gonzalez, F. Segovia Signal Theory, Networking and Communications, University of Granada, Granada, Spain This paper shows a new computer aided diagnosis (CAD) technique for the early Alzheimer's disease (AD). The system consists of voxel-based Normalized Mean Square Error (NMSE) feature extraction, a t-test with feature correlation weighting for feature selection and support vector machine (SVM) image classification. The proposed method yields an up to 98% classification accuracy, thus outperforming recent developed methods for early AD.
M05-160:

Metal Artefact Reduction in Computed Tomography Using Nonequispaced Fourier Transform

B. Kratz, T. M. Buzug University of Luebeck, Institute of Medical Engineering, Luebeck, Germany Metal objects in the field of view of a computed tomography scanner cause several artefacts inside the image. By reducing the influence of these objects the image quality can be enhanced. We propose a metal artefact reduction (MAR) method based on nonequispaced fast Fourier transforms. By damping the transformation additional a priori knowledge may be included. The results are compared to two common polynomial interpolation methods.

299

M05-163:

F. P. Figueiras1, X. J. Gonzalez1, D. Pareto1,2, J. D. Gispert1,2 1 Institut d'Alta Tecnologia (IAT-PRBB), Barcelona, Spain 2 CIBER en Bioingeniera, Biomateriales y Nanomedicina (CIBER-BBN), Zaragoza, Spain

Partial Volume Correction Using an Energy Multiresolution Analysis

Positron Emission Tomography (PET) allows the in-vivo monitoring of functional processes in the body. However its poor spatial resolution induces Partial Volume Effect (PVE), which leads to a loss of signal in tissues of size similar to the Point Spread Function (PSF) of the imaging device and induces activity spillover between adjacent structures with different amounts of activity. PVE can be corrected using either region of interest or voxel based approaches. The aim of Multiresolution Approach (MMA) is to introduce the high frequency information of an anatomical image into the functional image in order to achieve a higher resolution PET and to decrease PVE. In this work a Partial Volume Correction (PVC) algorithm was implemented based on the wavelet transform of two co-registered images and on both images global energy analysis. Results from this work are encouraging. The quantitative results of the phantom experiments show that the combination of MMA and the energy image analysis leads to a significant recovery of lost intensity induced by PVE. The corrected human pathological image showed enhanced resolution, including all the FDG uptake information and presenting higher spatial resolution. The proposed PVC method involving this combination of analysis shows that with one single parameter, the global energy factor, it is possible to achieve a common frequency level between MRI and PET images and therefore be enable to use and replace the high frequency information in the PET image. The proposed PVC method is applicable without making any prior assumptions on the content of the functional images, and therefore it can be applied to normal and pathological datasets.
M05-166:

C. Chan1,2, S. Meikle3, R. Fulton3,4, G. Tian2, W. Cai1,2, D. D. Feng1,2,5 1 School of IT, The University of Sydney, Sydney, Australia 2 Dept. of Electronic and Information Engineering, The Hong Kong Polytechnic University, Hong Kong, China 3 Disp. of Medical Science and Brain and Mind Research Institute, The University of Sydney, Sydney, Australia 4 School of Physics, The University of Sydney, Sydney, Australia 5 Med-X Research Institute, Shanghai Jiao Tong University, Shanghai, China

A Non-Local Post-Filtering Algorithm for PET Incorporating Anatomical Knowledge

The maximum likelihood expectation maximization (MLEM) reconstruction method is known to yield noisy images at high iteration numbers because the emission tomography reconstruction is an ill-posed problem. The noise can be suppressed by postfiltering the ML estimate or imposing a priori knowledge as a constraint within a Bayesian reconstruction framework. Most of these filters and priors are based on weighting the intensity differences between neighbouring pixels within a small local neighbourhood. Therefore, they have limited information to distinguish edges from noise. We investigated the use of a non-local means (NLM) filter for post-filtering MLEM reconstructed positron emission tomography (PET) images. We further investigated the effect of incorporating anatomical side information obtained from co-registered computed tomography (CT) images into the NLM, resulting in an adaptive non-local means (A-NLM) filter which takes into account the variance within each anatomical region on the PET image. In simulated and physical phantom experiments, the A-NLM filter demonstrated superior performance tradeoff between lesion contrast and noise compared with conventional Gaussian post-filtering and NLM without anatomical prior. We conclude that the A-NLM filter has potential for improved lesion detection over Gaussian post-filtered MLEM images.
M05-169:

S. Southekal1,2, D. Schulz2, D. Schlyer2, S. Junnarkar3, S. H. Maramraju1,2, B. Ravindranath1,2, S. Stoll4, M. Purschke4, J.F. Pratte5, C. Woody2, P. Vaska2 1 Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA 2 Medical Department, Brookhaven National Laboratory, Upton, NY, USA 3 Instrumentation Division, Brookhaven National Laboratory, Upton, NY, usa 4 Physics Department, Brookhaven National Laboratory, Upton, NY, USA 5 Electrical Engineering, University of Sherbrooke, Sherbrooke, QC, Canada The RatCAP is a breakthrough innovation in preclinical neuroimaging. The complete 3D PET tomograph is the only one of its kind to facilitate the measurement of neurotransmitter function in conscious, freely moving animals. Recent studies have been carried out using a combined bolus plus constant infusion paradigm for radiotracer administration, which results in a flat binding potential (BP) vs time curve at equilibrium. This method would potentially allow the detection of changes in the specific binding of radiotracer in response to pharmacological or behavioral stimuli during a single study.We have successfully detected a 60% change in 11C-Raclopride binding due to competition with unlabelled raclopride, and effort is underway to assess the RatCAPs sensitivity to changes of a smaller magnitude. In this work, we present our attempt to assess image noise variance using multiple statistically equivalent bootstrap samples of 11C-Raclopride datasets. The uncertainty in the binding potential estimate due to factors such as injected dose and noise propagation from the reconstruction and correction methods is evaluated. A

Quantitative Dynamic Neuroimaging with the RatCAP

300

comprehensive analysis of the data acquisition and processing chain will allow us to improve our detection capability and reach a consensus on an optimized approach for quantitative neuroreceptor imaging using the RatCAP.
This work was supported by the U.S. Department of Energy (OBER) under Prime Contract No.DE-AC02-98CH10886. M05-172:

Q. Bao1, F. R. Rannou2, R. Taschereau1, D. B. Stout1, A. F. Chatziioannou1 1 Crump Institute for Molecular Imaging, University of California, Los Angeles, Los Angeles, CA, USA 2 Department de Ingenieria Informatica, Universidad de Santiago de Chile, Santiago, Chile PETbox, an integrated low-cost benchtop preclinical PET scanner dedicated to high throughput quantitative mouse studies is currently under development at the Crump Institute for Molecular Imaging, UCLA. The system employs two opposing flat panel detectors, and results a limited angle tomographic system. Iterative reconstruction based on Expectation Maximization and maximum likelihood (EM-ML) algorithm was developed with incorporation of a probability matrix, which takes into account the detection probability and crystal scattering. Point sources at different positions were simulated and showed a peak sensitivity of 4.6% at 150-650 keV. With applying a sensitivity correction map, the image values of the reconstructed point sources are within a 5.3% standard deviation, indicating good reconstruction and normalization correction. The spatial resolution of point sources is largely uniform along the coronal direction across the field of view (FOV) with worse resolution along the y direction, which is mainly due to the reduced spatial sampling in that direction. A voxelized digital mouse (Moby phantom) was simulated with the GATE software package and then iteratively reconstructed. Attenuation and non-uniform sensitivity corrections were applied. Quantification accuracy was evaluated with the reconstructed Moby data and other two analytically low pass filtered data sets. To a first approximation, reasonably good quantification accuracy was obtained for most important organs with the PETbox reconstructed image. This evaluation indicates that the integrated detector system design, reconstruction algorithm, attenuation correction and normalization correction work well, which is very important for quantitative pharmacokinetic and pharmacodynamic studies.
M05-175:

Image Reconstruction for PETbox, a Benchtop Preclinical PET Tomograph

Computer Aided Diagnosis of the Alzheimer's Disease Combining SPECT-Based Feature Selection and Random Forest Classifiers

J. Ramirez, R. Chaves, J. M. Grriz, M. Lpez, I. lvarez, D. Salas-Gonzalez, F. Segovia Signal Theory, Networking and Communications, University of Granada, Granada, Spain This paper shows a novel computer aided diagnosis (CAD) system for the early Alzheimer's disease (AD). The system consists of voxel-based Normalized Mean Square Error (NMSE) feature extraction, a \emph{t}-test with feature correlation weighting for feature selection and Random Forests image classification. The proposed method yields an up to 96\% classification accuracy, thus outperforming recent developed methods for early AD diagnosis.
M05-178:

Generating Anthropomorphic Numerical Phantoms Semi-Automatically from Magnetic Resonance

A. Sklyar1, S. Gu1, M. Gennert1, M. King2 1 Worcester Polytechnic Institute, Worcester, MA, USA 2 University of Massachusetts Medical School, Worcester, MA, USA We describe a semi-automatic approach for segmenting MRI volumes to generate anthropomorphic phantoms. We use thresholding and active contours to extract the lungs, heart, ribs, spine and body. The choice of parameter values for each type of tissue is discussed in detail. Application of the resulting phantoms for motion simulation during emission imaging and ways to add more automation to the segmentation procedure are discussed.
This work was supported in part by the National Institute of Biomedical Imaging and Bioengineering Grant No. R01 EB001457. The contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health. M05-181:

Images

A. Signoroni1, S. Masneri1, A. Riccardi1, I. Castiglioni2 1 University of Brescia, Brescia, Italy 2 H S Raffaele Institute, IBFM-CNR, Milan, Italy

Enabling Solutions for an Efficient Compression of PET-CT Datasets

In this work suitable approaches and solutions for compression of Multi-Dimensional (MD) and Multi-Modal (MM) digital data and their application to PET-CT image datasets have been investigated for telemedicine purposes. Taking advantages from cross301

modal information of MD and MM data as PET and CT offer in integrated PET-CT datasets, we have designed and developed a compression system, based on state-of-the-art wavelet coding technologies, providing efficient and reliable compression of 3D PET-CT volumes. The method is based on a PET-segmentation masking process guiding a wavelet-based selective coding for the compression of CT image voxels within PET-CT datasets and on some effective solutions for the reduction of computational and memory load. We tested the proposed algorithms on representative PET-CT clinical datasets, which were compressed and transmitted across two Nuclear Medicine Departments (San Raffaele Hospitals in Brasil and in Italy) for telemedicine purposes. For all evaluated datasets the mask algorithm handled relevant uptake values, without giving false negatives and preserving signal. Results on computational performance of the algorithm show full compatibility and feasibility with telediagnosis requirements. Preliminary results on the accuracy of diagnostic quality of CT images suggest that our system offers the possibility to reduce resource occupation and coding time and, at the same time, the possibility to improve both coding performance and reliability, suggesting a suitability for PET-CT clinical application.
M05-184: Reduction of Dental Filling Metallic Artifacts in CT-Based Attenuation Correction of PET Data Using Weighted Virtual Sinograms

M. Abdoli1,2, M. R. Ay1,2,3, A. Ahmadian1,2, H. Zaidi4 1 Department of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Tehran, Iran 2 Research Center for Science and Technology in Medicine, Tehran University of Medical Sciences, Tehran, Iran 3 Research Institute for Nuclear Medicine, Tehran University of Medical Sciences, Tehran, Iran 4 Division of Nuclear Medicine, Geneva University Hospital, Geneva, Switzerland

The streak artifacts caused by dental fillings are known to generate artifacts in attenuation maps thus leading to overestimation/underestimation of tracer uptake in resulting PET images. Correction of these artifacts is therefore mandatory to achieve accurate attenuation maps to be used by the CT-based attenuation correction procedure. Our group recently proposed a metal artifact reduction method based on the virtual sinogram concept. In spite of the satisfactory results obtained on phantom studies, the algorithm was suffering from slight imperfections such as discontinuity of corrected sinogram bins along the second dimension of the sinogram matrix (i.e. the one sampling different angular positions) and an obvious difference between corrected and non-affected sinogram bins, especially when the number of affected bins in one column of the sinogram matrix is noticeable. The purpose of this study is to present an improved method allowing to overcome the above mentioned limitations. The proposed method enhances the corrected sinogram using weighted values of the influenced bins in the corrected and non-corrected sinograms, as well as weighted values of the sinogram bins in the neighboring column of the sinogram matrix. The optimum weighting factors associated with the above mentioned three data sets (, and ) were determined by assessing different combinations of these factors having values falling in the range [0.1-0.9] and statistical analysis of the obtained images of 10 patients with dental fillings. The optimum combination achieved is =0.3, =0.6 and =0.1. The results reveal that the resulting CT images and maps preserve more detail especially in regions adjacent to metallic objects whereas the original method suffered from eliminating noticeable details. It can be concluded that the proposed MAR method using weighted virtual sinograms improves the quality of clinical CT studies thus allowing to achieve more accurate attenuation correction of PET data.
This work was supported by the Tehran University of Medical Sciences under grant No. 132/494 and the Swiss National Science Foundation under grant SNSF 3100A0-116547 M05-187:

X. Wang1, C. Li1, M. Fulham2, S. Eberl2, D. Feng1 1 School of Information Technologies, University of Sydney, Sydney, Australia 2 PET and Nuclear Medicine, Royal Prince Alfred Hospital, Sydney, Australia

PET-Enhanced Liver Segmentation for CT Images from Combined PET-CT Scanners

The use of functional (PET) information from PET-CT scanners to assist liver segmentation in CT data has yet to be addressed. In this work we implement PET data enhanced liver segmentation with CT. We utilise the difference in FDG uptake between the liver and adjacent organs to separate the liver from these structures, which have similar intensities in low-contrast CT. The relatively high normal FDG uptake, and hence high SUV of liver metabolism, allows an accurate estimation for liver segmentation in CT images. By deformable registration, the PET ROIs are mapped onto the CT images for the initial liver segmentation in CT. To overcome the different intensity values of CT images from different patients or over multiple temporal imaging sessions, the initial liver region in CT images is used to establish the accurate threshold criteria for CT liver segmentation. To prevent the deformable model from leaking into the adjacent tissues and structures, the feature images are computed to exclude and disconnect neighboring organs and tissues from liver. Our experimental results in 12 clinical PET-CT studies suggest that our algorithm is robust when dealing with livers of different shapes and sizes and from a range of different patients.
This work was supported by ARC and PolyU grants.

302

M05-190: Atanassov Fuzzy Model for Analyzing Selective Contrast Enhancement of Medical Images Using Multiresolution

M. Bhattacharya1, A. Das2 1 Information Communication Technology, Indian Institute of Information Technology & Management, Gwalior,Madhya Pradesh, India 2 Radio Physics & Electronics, University of Calcutta, Kolkata, West Bengal, India Breast cancer continues to be a significant public health problem. Screen/Film mammography is widely recognized as being the effective imaging modality for the early detection of breast cancer in women. Breast radiographs are generally examined for the presence of malignant masses and indirect signs of malignancy, such as the presence of microcalcifications. To improve the visibility of mammographic lesions image enhancement methods for digitized mammograms have been attempted by several researchers. Image enhancement using smoothing with fuzzy sets is developed for improving contrast of the pixels. Present paper intends to provide a generalized notion of fuzzy sets to achieve more flexible and intelligent contrast enhancement method of mammograms using Atanassovs intuitionistic fuzzy model. This model for medical diagnosis combines the negative and positive evidences present in many real world problems. Intuitionistic Fuzzy Set IFS, has the virtue of complementing conventional fuzzy sets, and is able to model vagueness, uncertainties as well. In the proposed algorithm, IFS approach is implemented to achieve a robust, flexible and intelligent contrast enhancement method of mammograms. Moreover, the uncertainties between the breast mass and its surroundings are well captured by IFS. Multiresolution technique has been incorporated to highlight more detail characteristics of breasts. Experimental results successfully enhance the selected mass regions of the mammograms. The overall selective contrast enhancement scheme of the proposed algorithm involves the decomposition of input images by wavelet transform, fuzzification by standard FCM with justified number of clusters, implementation of IFS approach for capturing the uncertainty between true breast mass region and its surrounding normal tissues, defuzzification and finally reconstruction of enhanced image by inverse wavelet transform
Authors kindly acknowledge Dr.S. K. Sharma of Eko X-Ray & Imaging Institute, Kolkata M05-193:

S. Somayajula1, B. Bai2, Q. Li1, R. M. Leahy1 1 Signal and Image Processing Inst, University of Southern California, Los Angeles, CA, USA 2 Dept. of Radiology, Columbia University, New York, NY, USA

Positron Range Correction Using Information Theoretic Anatomical Priors

Positron range causes a blurring in the reconstructed PET image thus limiting its resolution. In previous literature, a truncated homogeneous model was described for range correction wherein the system model incorporated spatially invariant range blurring with truncation at the object boundary. This approach did not correct the blurring in internal boundaries. We describe a method for positron range correction that uses anatomical priors based on joint entropy (JE) in conjunction with the truncated homogeneous model. Since PET tracer uptake is typically correlated with the underlying morphology, using information from a co-registered anatomical image can potentially correct the blurring caused by positron range. We use JE between feature vectors extracted from the anatomical and PET images as a prior on the PET image. These feature vectors are chosen based on scalespace theory to reflect the assumption that the boundaries in the anatomical and PET images are correlated. The goal is to correct for positron range such that the reconstructed PET image not only fits the data, but also is similar to the anatomical image in the information theoretic sense. We present simulation results for PET and anatomical images generated from a Hoffman brain phantom such that the images are identical in structure. To simulate positron range effect, we used the range kernel for Cu-60 isotope that has a range of 3.09 mm. The results for range correction using JE prior show improved quantitation and better boundary behavior compared to quadratic prior, which does not use any anatomical information.
M05-196:

S. Shokouhi1, D. W. Wildon2, S. D. Metzler3, B. S. McDonald1, T. E. Peterson1 1 Vanderbilt University Institute of Imaging Science, Nashville,TN, USA 2 Center for Gamma-Ray Imaging, Department of Radiology, University of Arizona, Tucson, AZ, USA 3 Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA

System Modeling and Image Reconstruction for SiliSPECT Using Synthetic Collimation with a Large Number of Focusing Apertures

The goal of our study is to develop dedicated system modeling and image reconstruction techniques for SiliSPECT, a smallanimal SPECT that incorporates two stationary camera heads and stacked detector acquisition. The image reconstruction was performed by combining synthetic collimation, in which multi-pinhole projection data is acquired at two magnifications simultaneously, and the ordered subset algorithm. This combination not only significantly accelerates the reconstruction process, but also enables image reconstruction with a limited number of camera views and in the presence of multiplexing (20-80 % overlap of the projection area). In simulation studies, we were able to demonstrate better reconstruction results by incorporating both magnifications in comparison to either alone. We also observe a better contrast and more accurate quantitation by increasing 303

the number of pinholes despite the multiplexing effect. Our main objective is to optimize the magnification ratios between the two detectors in such a way that we can get the most benefit from our ultra-high resolution detectors. The high detector resolution also requires a physically accurate system model that is compatible with such resolution capabilities. Based on our previous studies, we implemented an accurate geometric system model, instead of using a generic pinhole model. We will explore the feasibility of using the new system model with experimental data.
This work was supported by NIH/NIBIB R33 EB000776, NIH Grant P41 EB002035, NIH/NIBIB R01-EB-6558. a Career Award at the Scientific Interface (TEP) from the Burrough Wellcome Fund and the 2008 postdoctoral molecular imaging scholar program award from the Society of Nuclear Medicine. M05-199:

M. Blume1,2, A. Keil1, A. Martinez-Moeller3, N. Navab2, M. Rafecas1 1 Instituto de Fisica Corpuscular, Valencia, Valencia, Spain 2 Computer Aided Medical Procedures (CAMP), Technische Universitaet Muenchen, Munich, Bavaria, Germany 3 Nuklearmedizinische Klinik im Klinikum Rechts der Isar, Munich, Bavaria, Germany

Simultaneous Reconstruction of Image and Motion in Gated Positron-Emission-Tomography

We present a novel method for joint reconstruction of both image and motion in positron-emission-tomography (PET). Most of nowadays motion compensation methods consist of two steps: (i) motion estimation and (ii) image estimation. They use deformable image registration/optical flow techniques in order to estimate the motion from individually reconstructed gates. Then, the image is estimated based on this motion information. A very common image estimation approach is to deform the reconstructed gates to the reference gate and then combine them in image space. More sophisticated methods perform a completely new reconstruction. A major drawback of these methods lies in the motion estimation step, since it is completely based on the usually noisy individually reconstructed gates. As we show in a simulation study, our joint reconstruction approach alleviates this drawback and results in both visually and quantitatively better image quality. We attribute these results to the fact that for motion estimation always the currently best available image estimate is used and vice versa. Additionally, results for real dual respiratory and cardiac gated patient data are presented.
M05-202:

Modified Subset Scheme for Attenuation Weighed Reconstruction

H. E. Rothfuss, V. Y. Panin Molecular Imaging, Siemens Healthcare, Knoxville, TN, USA A new method of selecting subsets for attenuation weighed algorithms was developed to reduce observed artifacts in non symmetrical objects. Artifacts are observed primarily in the trans axial view of the images using the traditional subset selection scheme. These artifacts were determined to originate from the attenuation weighting of the data across the angular range of the data. By localizing the subset projections to sectors, the data is all similarly weighted. This greatly improves the image quality in non symmetric objects with high activity regions that are located toward the edges of the object. The results of the new subset scheme are less distorted images with fewer artifacts.
M05-205:

Comparison of Quantitative and Task-Based Optimizations of a 4D MAP-RBI-EM Image Reconstruction Method for Gated Myocardial Perfusion SPECT

T.-S. Lee, B. M. W. Tsui Radiology, Johns Hopkins University, Baltimore, MD, USA We performed quantitative and task-based optimizations of the 4D space-time Gibbs priors with a generalized potential function (GPF) for a 4D maximum a posteriori rescaled-block iterative (MAP-RBI) EM image reconstruction algorithm and compared the optimization results in terms of the maximum detectability of regional wall motion defect (RWMD) for gated myocardial perfusion (GMP) SPECT. In 4D MAP-RBI-EM, the 4D space-time Gibbs priors enforce selective smoothing in space and time domains by the GPF with its three parameters, i.e., , , and , whose values allow the GPF to degenerate into various potential functions. Realistic noise-free and noisy Tc-99m Sestamibi GMP SPECT projection data that included the effects of attenuation (A), collimator-detector response (D), and scatter (S) were simulated using the 4D Extended Cardiac-Torso (XCAT) phantom consisting of hypokinetic RWMD. The projection data sets were reconstructed using various combinations of GPF parameters in the 4D MAP-RBI-EM with ADS corrections. The trade-offs between normalized mean squared error (NMSE) in noise-free reconstructed images and normalized standard deviation (NSD) in noisy reconstructed images were determined, respectively. For those trade-offs, we performed a task-based human observer study with receiver operating characteristic (ROC) analysis. The same quantitative and task-based evaluation studies were also performed for both 3D FBP and 3D OS-EM with ADS correction followed by 4D filter over the various trade-offs. The result showed that optimized 4D MAP-RBI-EM with ADS correction showed the highest detectability and the optimum determined by the human observer study for each method lies within the range of optimal NMSE-NSD trade-offs of the corresponding method. We concluded that NMSE-NSD trade-off is a useful guide for 304

the optimization of the RWMD detection task, and the optimized 4D MAP-RBI-EM with ADS corrections provides significant improvement in detecting RWMD for GMP SPECT.
M05-208:

D. Thanasas1, E. Georgiou2, N. Giokaris1, A. Karabarbounis1, D. Maintas3, M. Mikeli1, C. Papanicolas1, L. Ragkousis1, E. Stiliaris1 1 Dept. of Physics, National & Kapodistrian University of Athens, Athens, Greece 2 Medical School, National & Kapodistrian University of Athens, Athens, Greece 3 Institute of Isotopic Studies, Athens, Greece

An Analytical Position Correction Algorithm for -Camera Planar Images from Resistive Chain Readouts

The charge limitation for peripheral FOV events detected by the most commonly used Position Sensitive Photomultiplier Tubes (PSPMTs) results to spatial distortions and non-uniformities of the obtained planar images. These effects can be corrected with newly developed sophisticated techniques operating on the charge signals from the individual wires of the multi-anode systems. However, a similar algorithmic approach for the simple case, where the resistive chain readout technique is used and, consequently, the original charge distribution information is lost, is not applicable. In this work the development of a new method to eliminate these distortion effects in the planar images for Gamma-Camera systems based on resistive chain techniques is presented. The proposed model, which incorporates an a priori knowledge of three parameters related to light diffusion inside the scintillation crystal in use, provides an accurate, analytically calculated estimate of the spatial correction as a function of the primary reconstructed planar position from the resistive chain signals. This algorithm can be used online on an event-by-event basis and can be applied to both, homogeneous and pixelated crystals. Experimental results for simple phantoms and various CsI(Tl) crystals with different thickness are presented.
This work is partially supported by the 03E287 research project, implemented within the framework of the "Reinforcement Program of Human Research Manpower" (PENED) and co-financed by National and Community Funds (25% from the Greek Ministry of Development-General Secretariat of Research and Technology and 75% from E.U.-European Social Fund). M05-211:

Simulation-Based Evaluation of OSEM Reconstruction Bias on Low Activity PET Data for the HRRT Scanner

E. Dusch, C. Comtat, R. Trebossen SHFJ/CEA, Orsay, France The ECAT HRRT is a dedicated human brain PET scanner with a spatial resolution of less than 3 mm. Because of gaps between its detector heads, the standard reconstruction method is the Ordinary Poisson (OP) OSEM3D algorithm. For very low count statistics, it has been shown by several HRRT users groups that this algorithm converges to a biased solution, in particular for low activity regions. Unbiased quantification in low activity regions such as pons or cerebellum is essential to measure accurately receptor binding in brain receptor studies with reference tissue models. It is therefore important to evaluate potential bias introduced by the reconstruction algorithm for HRRT data. We have developed a simulation tool to study the bias in the estimation of kinetics parameters. Providing a pharmacokinetic compartment model with its associated rate constants for several regions of interest of a digital brain phantom, this tool generates HRRT data with realistic noise properties, accountings for random and scattered coincidences, detector efficiencies and detector resolution. We simulated a [11C]-PE2I study with a framing ranging from 1 minute at the beginning to 5 minutes at the end of the acquisition. These data were reconstructed with the OP-OSEM3D algorithm (16 subsets, 25 iterations, with PSF modeling) and the binding potential (BPND) of PE2I was calculated using the Simplified Tissue Reference Model (STRM), with the cerebellum as region of reference and the striatum as specific region. To distinguish between bias due to the count statistics and to the finite resolution of the scanner, noise-free simulated sinograms were also reconstructed. For our framing, representative of our clinical practice, the bias in the reconstructed TACs was primarily due to resolution effects (underestimation in the striatum varying between -3 % and -10 %, moderate bias in the cerebellum). This translated into an underestimation of BPND in the striatum of the order of -7 % to -15%.
M05-214:

A. Atre1, K. Vunckx1, A. Reilhac2, J. Nuyts1 1 Nuclear Medicine, K.U.Leuven, Leuven, Belgium 2 Biospective Inc, Montreal, Canada

Evaluation of Different MRI-Based Anatomical Priors for PET Brain Imaging

Image reconstruction in emission tomography may benefit from the use of anatomical side information obtained with other imaging modalities in the same patient. If the reconstruction also uses a model for the finite system resolution, the anatomical prior may strongly improve the resulting partial volume correction. One way to handle the anatomical image, is to incorporate it as a-priori information in a maximum a posteriori reconstruction algorithm. 305

In this contribution, three different anatomical priors for maximum-a-posteriori reconstruction are studied for PET 18-F-FDG brain imaging, using a registered T1-weighted MRI image as the anatomical image. The A-MAP prior requires a (possibly fuzzy) labeling of the MRI brain voxels as gray matter, white matter and cerebrospinal fluid. The joint entropy prior does not require processing the MRI image, and uses the joint histogram to quantify the similarity between the PET and the MRI image. The prior proposed by Bowsher et al can also be applied with the unprocessed MRI image. It uses a local neighboring selection strategy to avoid smoothing over tissue boundaries. In this study, the PET SORTEO Monte Carlo simulator is used to evaluate these priors and to compare them to MLEM with resolution recovery for a realistic brain image. Comparison of simulated and measured point sources confirmed that PETSORTEO yields realistic resolution effects. The performance of the priors is illustrated on anecdotal simulations. Repeated Monte Carlo simulations are being performed to enable a detailed biase / noise analysis.
M05-217:

Fast Ordered Subset Convex Iterative Panel CT Reconstruction

Z. Burbar, I. Hong, C. Michel Siemens Medical Solutions Molecular Imaging, Knoxville, TN, USA Iterative reconstruction has several advantages over analytical cone-beam reconstruction methods for three-dimensional x-ray computed tomography. Iterative methods tend to incorporate both system and statistical models which yield a better signal to noise ratio. Iterative reconstruction also produces adequate image quality at low count, enabling low-dose X-ray CT studies, at the expense of computational overhead. This work presents an efficient implementation of an Ordered Subset Convex (OSC) algorithm. OSC algorithm requires one forward projection, exponential of the forward projection of the current image, and two back projections per iteration which makes it very computational intensive. Optimizing the forward projector starts by solving a 3D line equation with rotation by using the system geometry (symmetry). A voxel weighting method was also applied during forward projection to achieve better accuracy. Data was aligned in the Z,X,Y directions for cache optimization and sixteen voxels were forward projected at the same time using SIMD registers: four in both z and x-y directions. The back projection optimization was achieved by performing both back projections at the same time. Data were first aligned and both back projections were performed in parallel on four projection bins using SIMD. The algorithm was implemented on a small animal flat panel rotating CT from the Inveon System (Siemens Preclinical Solutions, Knoxville, TN.). Phantom data was acquired in rebin by four mode (i.e. 512 x 512 pixels per projection) with 360 projections and reconstructed with both OSC and the standard Feldkamp method. Both reconstructions were performed on a dual X5450 quad core CPUs with 32 GB RAM and 64 bit operating system. Using this fast OSC implementation, a reconstruction time of 210 s was achieved using 18 subsets and 4 iterations (i.e. 52.5 s per iteration).
M05-220:

S. Liu1, T. H. Farncombe2 1 Electrical & Computer Engineering, McMaster University, Hamilton, ON, Canada 2 Radiology, McMaster University, Hamilton, ON, Canada

Accelerated Monte Carlo Based Dual-Isotope SPECT Image Reconstruction

Simultaneous multiple-isotope imaging with photons of similar energies results in significant crosstalk of both scattered and unscattered photons from each isotope overlapping in the photopeak energy windows. In this work, a convolution-based forced detection Monte Carlo crosstalk compensation method for dual-isotope reconstruction has been developed in order to compensate for the effect of cross-talk photons. An optimization of photopeak window parameters (upper and lower limits) was performed for simultaneous I-123 and Tc-99m imaging. It is found that the cross-talk photons in one region only affect the estimation of this region itself, but do not appear to significantly contaminate the estimation of other regions. The error of overestimated activity of a certain region depends on the fraction of cross-talk photons in this region, which can be further used to determine the optimal upper and lower limits of the photopeak window.
This work has been supported by the Ontario Institute of Cancer Research under the 1 mm Challenge. M05-223:

G. L. Zeng1, F. Jing1, Q. Huang2, J. You3, G. T. Gullberg2 1 Utah Center for Advanced Imaging Research, University of Utah, Salt Lake City, UT, USA 2 Department of Radiotracer Development and Imaging Technology, Lawrence Berkeley Laboratory, Berkeley, CA, USA 3 Cubic Imaging, Billerica, MA, USA Recently derivative-backporjection image reconstruction algorithms have received significant attention in X-ray CT and in nuclear medicine imaging because of the ability to reconstruct truncated projection data. Specifically for uniformly attenuated SPECT data, algorithms using the derivative-backprojection method have been developed. This paper extends the development to that of a derivative-backprojection algorithm for non-uniformly attenuated SPECT data. It turns out that the image reconstruction 306

A Derivative-Backprojection Algorithm for Non-Uniformly Attenuated SPECT Data

algorithm can be reduced to that similar to the situation of a uniform attenuator. This has potential application for special situations in cardiac and brain SPECT. One advantage of this newly developed algorithm is its potential application in local tomography with truncated SPECT data.
M05-226:

J. Cal-Gonzalez1, J. L. Herraiz1, S. Espana1, J. J. Vaquero2, M. Desco2, J. M. Udias1 1 Fisica Atomica, Molecular y Nuclear, Universidad Complutense Madrid, Madrid, Spain 2 Unidad de Medicina y Cirugia Experimental, Hospital G.U. Gregorio Maranon, Madrid, Spain Abstract Positron range limits the spatial resolution of PET images. It has a different effect for different isotopes and propagation materials, therefore it is important to consider it during the image reconstruction, in order to obtain the best image quality. Positron range distribution was computed using Monte Carlo simulation with PeneloPET code. This simulation models positron trajectories and computes the radial distribution of the annihilation coordinates for the most common isotopes used in PET: 18F, 11C, 13N, 15O, 68Ga and 82Rb. Range profiles are computed for different positron propagation materials, obtaining one kernel profile for each isotope-material combination. These range kernels were introduced in FIRST, a 3D-OSEM image reconstruction, and employed to blur the object during forward projection. The blurring introduced can be modified in function of the material density in which the positron is annihilated, obtained for instance from a CT image. In this way, different positron range corrections for each material in the phantom are considered. The resolution and noise properties of the images reconstructed with and without positron range modelling are compared in this work. For this purpose, acquisitions of an Image Quality phantom filled with different isotopes have been simulated for the ARGUS small animal PET scanner.
M05-229:

Positron Range Effects in High Resolution 3D PET Imaging

Improved MAGIC Gel for Higher Sensitivity and Elemental Tissue Equivalent 3-D Dosimetry

X. Zhu, T. G. Reese, E. M. Crowley, G. El Fakhri Massachusetts General Hospital, Boston, MA, USA We are investigating the use of NeuroPET, the first mobile PET scanner in the world, for the in-room PET verification of proton therapy. Polymer based gel dosimeter allows truly 3D mapping of dose distribution in a phantom, therefore has a tremendous potential role in the verification of radiation fields in radiation therapy, and is a preferred phantom material. However, improvement in elemental composition (especially the oxygen to carbon, or O/C, ratio) is desired in our application. We have modified the elemental compositions of the normoxic MAGIC gel using glucose and urea and report improved elemental tissue equivalency as well as sensitivity of the MAGIC gel. The O/C ratios of our new gels agree with that of soft tissue within 1%. We tested the dose responses of our new gel formulations with MRI transverse relaxation rate (R2) measurements. Our results show that the slopes of dose response curves were 1.6 to 2.7 times larger. Further addition of urea resulted in a similar slope but with an increased interception. In conclusion, our improved MAGIC gel formulations have higher sensitivity and better elemental tissue equivalency for 3-D dosimetry applications involving nuclear reactions.
This work was supported by NIH R21-CA134812 and T32-EB002102. M05-232:

M. K. Cho1, S. Y. Park2, H. K. Kim1 1 School of Mechanical Engineering, Pusan National Univ., Busan, Korea 2 Proton Therapy Center, National Cancer Center, Goyang, Korea

On-Board Cone-Beam Digital Tomosynthesis for Image-Guided Proton Therapy

Proton therapy is greatly paid attention because the proton beam can deliver high dose to a tumor region keeping the dose to surrounding normal tissues minimally. Therefore high accurate patient positioning is very important during treatment period. For image-guided proton therapy, we investigated the feasibility of CBDT (cone-beam digital tomosynthesis) technologies in the gantry treatment room. A fully equipped x-ray projection system, which was originally operated for patient alignment, in parallel to proton-beam direction was utilized for acquiring CBDT. The CBDT system employs large-area amorphous silicon flat-panel detectors, pulse type x-ray sources and a couch having 6 DOF (degrees of freedom) in movements. Tomographic imaging performances of the system, such as signal uniformity, voxel noise, SNR (signal-to-noise ratio), linearity of CT numbers, and CNR (contrast-to-noise ratio), are evaluated by using quantitative phantoms. The overall performance of the system is demonstrated by comparing with the CT images of radiotherapy planning system. From the reconstructed phantom images, the CBCT system in the gantry treatment room will be very useful as a primary patient alignment system for image-guided proton therapy. The CBDT may provide fast patient positioning with less motion artifact and patient doses.
This work was supported by the National Cancer Center Grand (NCC-0810210).

307

M05-235:

L. Grezes-Besset1,2, J. Vandemeulebroucke1,2, P. Clarysse1, D. Sarrut1,2 1 CREATIS-LRMN, Villeurbanne, France 2 Centre de lutte contre le cancer Leon Berard, Lyon, France

Respiratory Motion Tracking from Different Viewing Angles During Lung Cancer Treatment in Radiotherapy

Respiratory motion causes errors when planning and delivering radiotherapy treatment to lung cancer patients. A precise localization of the tumor can help reduce errors and uncertainties induced by respiration. It is in this context that we propose to extract respiratory motion during treatment delivery with 3 different methods. Two methods use internal information and are based on processing a sequence of 2D cone-beam projections acquired during treatment, from different viewing angles. The first method (called Int_IB) is based on intensity variation in the lung due to air flow from respiratory, whereas the second one (called Int_BM) uses an adapted block matching algorithm. The third method (called Ext_BM) is based on abdominal height measurements. We have evaluated these methods with real data against expert-based ground-truth. As a result, we managed to extract respiratory motion regardless the method used. The signals extracted from the fluoroscopic sequence gave similar results for all angles. All the extracted signals were coherent with the expert. When comparing the 2 methods based on internal information, slightly better results were obtained with Int_BM. Results obtained with the external signal were promising. The external signal carries the advantage that it is non-invasive but it is potentially less accurate in terms of tumor localization than the internal signal. Combining both kinds of signals could provide a good compromise.
M05-238:

G. Shakirin1, H. Braess2, F. Fiedler1, D. Kunath3, K. Laube1, K. Parodi4,5, M. Priegnitz1, W. Enghardt1,3 1 Forschungszentrum Dresden-Rossendorf, Dresden, Germany 2 Siemens Medical Solutions, Erlangen, Germany 3 Oncoray, Technische Universiaet Dresden, Dresden, Germany 4 Heidelberg Ion Therapy Center, Heidelberg, Germany 5 Department of Radiation Oncology, University Clinic, Heidelberg, Germany At present, positron emission tomography (PET) is the only available technique for an in-vivo, non-invasive monitoring of the dose delivery precision in highly conformal ion beam therapy. The successful exploitation of in-beam PET at the Gesellschaft fuer Schwerionenforschung (GSI), Darmstadt, Germany during the last decade and a rising number of built or planned proton and heavier ion therapy facilities worldwide makes the development of a particle therapy PET (PT-PET) system of the next generation reasonable. The in-beam PET installation at GSI is a double-head positron scanner with a very limited solid angle which results in severe artifacts in the reconstructed images and in a low counting statistics. Thus, it is highly desirable to have larger solid angle coverage for PT-PET scanners of the next generation. However, increasing the effective area of a scanner might be limited by several requirements for the equipment of a radiotherapy treatment unit. Possible solutions for a prospective PT-PET system as well as a methodology for the evaluation of concurrent designs of the scanner taking into account the requirements of a therapy facility are discussed in this paper.
M05-241:

System Solution for Particle Therapy PET

Micro-Probe for Medical Radiation Imaging Based on Medipix2 Detector

Z. Vykydal, J. Jakubek Institute of Experimental and Applied Physics, Czech Technical University in Prague, Horska 3a/22, 12800 Praha 2, Czech Republic For medical imaging applications the emphasis is generally put on obtaining high resolution and low dose images of samples or patients. All this features can be covered by using the single photon countig devices like Medipix and advanced techniques of image reconstruction. In case that intra-operative approach is advantageous, like direct observation of dental implants, also the size of the imaging device itself plays critical role. Therefore we manufactured the miniaturized version of currently used USB interface accommodating also the Medipix2 chip itself on the same printed circuit board. The dimensions of the whole imaging device are reduced to 15 mm x 60 mm (which means that 25 % out of the total device size is actually sensitive area). Detector of this size can easily fit directly into the patients' mount. This so-called USB Lite device includes all circuits necessary for detector operation including high voltage source for detector biasing and it is fully powered from the USB bus itself. It is using USB 1.0 protocol, so the readout speed for single Medipix detector is limited to about 4 fps. As the high speed USB 2.0 version of the protocol becomes available the logical step is to upgrade existing device and therefore increase readout speed significantly. Also progress in the packaging technologies of integrated circuits allows to house rather powerful FPGAs with all the necessary IOs' in the BGA package as small as 8 x 8 mm. Use of the FPGA allows us to support perspectively all different chips from the Medipix/Timepix family with minor hardware modifications. A bread-board prototype of the FPGA-based USB 2.0 upgrade has been tested successfully.

308

M05-244:

T. Solf1, V. Schulz1, A. Thon1, P. Fischer2, M. Ritzert2, V. Mlotok2, C. Piemonte3, N. Zorzi3 1 Molecular Imaging Systems, Philips Research Laboratories, Aachen, Germany 2 Chair of Cricuit Design, University of Heidelberg, Heidelberg, Germany 3 Micro-Electro-Mechanical-Systems and Radiation Detectors, Foundation Bruno Kessler, Trento, Italy Simultaneous PET and MR imaging requires a novel type of highly integrated PET detectors. Due to geometric constraints and MR compliance a very compact detector stack was built within the HYPERImage consortium. This allows a four sides buttable detector module design with a low dead space in between. The scintillation light coming from a LYSO array is converted in a SiPM sensor tile with a high packing fraction and a high photo detection efficiency to provide sub-ns time-of-flight timing resolution. The analog signals coming from SiPM detectors are digitized close to the sensor to minimize potential crosstalk. A custom mixed-signal ASIC was integrated on a 64 channel sensor stack which is powered and controlled by an FPGA interface board. The complete sensor stack is assembled and characterized to extract the PET relevant parameters, in particular energy, timing and spatial resolution.
The presented work is part of the EU FP7 project HYPERImage, Grant Agreement N 201651. M05-247:

Solid-State Detector Stack for ToF-PET/MR

S. H. Maramraju1, S. D. Smith2, S. Rescia2, S. S. Junnarkar2, B. Ravindranath1, S. Stoll2, M. Purschke2, D. Schulz2, S. S. Southekal1, W. Lenz2, J. F. Pratte3, P. Vaska2, C. L. Woody2, D. J. Schlyer2 1 Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA 2 Brookhaven National Laboratory, Upton, NY, USA 3 Universite de Sherbrooke, Sherbrooke, Canada We have developed an MR compatible PET scanner based on Rat Conscious Animal PET (RatCAP) for simultaneous acquisition of PET/MRI images of the rat brain in a 9.4 T microMRI scanner. The PET tomograph is housed in a non-magnetic case made of delrin plastic. MRI RF coils were designed to fit inside the PET tomograph operating in full quadrature transceiver mode. A custom designed positioning device and animal bed was constructed to allow for accurate placement of the rat brain within the RF coil and RatCAP while allowing simultaneous monitoring of the animals vital signs. The simultaneous PET/MRI phantom and rat brain images were acquired in a Bruker 9.4 T (20 cm bore diameter) MRI system. We observed minimal interference between PET electronics and MRI. These preliminary results demonstrate the feasibility of performing simultaneous PET, MRI and spectroscopy studies. Additional studies, now underway, will explore the possibility of acquiring simultaneous PET/MRI whole-body images of mice.
This research was carried out at Brookhaven National Laboratory under contract DE-AC02-98CH10886 with the U.S. Department of Energy as collaboration between BNL and the State University of New York at Stony Brook. M05-250:

Simultaneous acquisition of PET/MR images of small animals at 9.4 T

A Prototype of the MRI-Compatible Ultra-High Resolution SPECT for in Vivo Mice Brain Imaging

J.-W. Tan, L.-J. Meng Department of Nuclear, Plasma and Radiological Engineering, The University of Illinois at Urbana-Champaign, Urbana,IL, USA In this study, we will present a prototype of the MRI-compatible, ultra-high resolution SPECT system for use in a pre-exisitng Siemens 3 T MR scanner. The system is based on the use of an ultra-high resolution energy-resolved photon-counting (ERPC) CdTe detector that we have recently developed. This detector consists of eight detector hybrids, each having a pixilated CdTe detector of 1mm thickness, bump-bonded to a custom-designed CMOS readout ASIC. Each detector hybrid offers an imaging area of 1.1 cm x 2.2 cm that is divided into an array of 32 x 64 pixels with a pitch size of 350 m x 350 m. A prototype SPECT system has been designed and tailored for use inside the 3.0 T Siemens scanner for tomographic imaging. It consists of a single ERPC detector head, coupled to a collimation aperture made of multi-layer pure tungsten sheets. The system is supported by a non-magnetic gantry that allows the system to be moved in and out the Siemens scanner and to be rotated around the object to obtain a sufficient angular sampling. Shielding, mechanical support and cooling are proved MRI-compatible. The imaging performance of the prototype system will be evaluated using various resolution phantoms and the results will be presented.

309

M05-253:

Respiratory Motion Correction of PET Using Motion Parameters from MR

N. Dikaios, T. D. Fryer Wolfson Brain Imaging Centre, Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK Respiratory motion during PET acquisition from the chest/abdomen leads to significant image degradation. Combined PET/MR scanners open up the opportunity to correct motion using MR data acquired simultaneously with PET. As simultaneous human chest/abdomen PET/MR images are currently unobtainable, in this preliminary study we determine motion parameters from respiratory-gated MR and then use these motion parameters in post-reconstruction registration to correct pseudo-PET images generated from the MR. The gated MR images are segmented to typical organ FDG SUV values, smoothed to mimic PET resolution, forward projected into the GE Advance geometry and reconstructed separately using OSEM implemented in STIR. The MR images are registered using a combined affine and non-rigid B-splines algorithm, with mutual information used as the cost function in a multi-resolution approach. This ITK based registration algorithm is shown to work well, both visually and quantitatively in terms of image metrics.
M05-256:

P. J. Schleyer1, C. Tsoumpas1, T. Schaeffter1, V. Schulz2, P. K. Marsden1 1 Guy's, King's and St Thomas' School of Medicine, King's College London, London, United Kingdom 2 Philips Research Laboratories, Aachen, Germany

The Effect of Bone -value Assignment and Segmentation on PET Attenuation Correction Accuracy

The ongoing development of combined PET/MR scanners has raised the need to utilise MR images for the attenuation correction of PET. MR images do not represent the attenuation of the imaged tissue, and bone is not easily distinguishable from air. The location and attenuation coefficient of bone must therefore be determined by secondary means, leading to potential errors in attenuation correction. We have investigated the nature and extent of these errors using PET/CT patient data over a range of patient weight. Bones were segmented from CT images, and patient specific and generic constant values were used to replace the bone. The effect of varying the accuracy of bone segmentation was also investigated. PET images were attenuation corrected using both modified and unmodified CTs. The effect of the CT modification was determined in the bone, soft tissue, and lungs of the PET images. All replacement values produced errors in all tissue types. Larger patient specific and generic values produced a bigger effect in the lungs and soft tissue, and were more sensitive to variation in bone segmentation accuracy. Smaller replacement values were found to produce favourable results.
M05-259:

A. Kolb1, M. Hofmann1,2,3, V. Sossi1,4, H. F. Wehrl1, A. Sauter1,5, A. Schmid1, H. P. Schlemmer5, C. D. Claussen5, B. J. Pichler1 1 University of Tuebingen, Laboratory for Preclinical Imaging and Imaging Technology of the Werner Siemens Foundation, Tuebingen, Germany 2 Max Planck Institute for Biological Cybernetics, Tuebingen, Germany 3 University of Oxford, Department of Engineering Science, Oxford, United Kingdom 4 University of British Columbia, Vancouver, Canada 5 University of Tuebingen, Department of Radiology, Tuebingen, Germany Clinical PET/MRI is an emerging new hybrid imaging modality. In addition to provide an unique possibility for multifunctional imaging with temporally and spatially matched data, it also provides anatomical information that can also be used for attenuation correction with no radiation exposure to the subjects. A plus of combined compared to sequential PET and MR imaging is the reduction of total scan time. Here we present our initial experience with a hybrid brain PET/MRI system. Due to the ethical approval patient scans could only be performed after a diagnostic PET/CT. We estimate that in approximately 50% of the cases PET/MRI was of superior diagnostic value compared to PET/CT and was able to provide additional information, such as DTI, spectroscopy and Time Of Flight (TOF) angiography. Here we present 3 patient cases in oncology, a retropharyngeal carcinoma in neurooncology, a relapsing meningioma and in neurology a pharyngeal carcinoma in addition to an infraction of the right hemisphere. For quantitative PET imaging attenuation correction is obligatory. In current PET/MRI setup we used our MRI based atlas method for calculating the mu-map for attenuation correction. MR-based attenuation correction accuracy was quantitatively compared to CT-based PET attenuation correction. Extensive studies to assess potential mutual interferences between PET and MR imaging modalities as well as NEMA measurements have been performed. The first patient studies as well as the phantom tests clearly demonstrated the overall good imaging performance of this first human PET/MRI system. Ongoing work concentrates on advanced normalization and reconstruction methods incorporating count-rate based algorithms.

Clinical PET/MRI-System and Its Applications with MRI Based Attenuation Correction

310

M05-262:

S. Yamamoto1, J. Hatazawa2 1 Kobe City College of Technology, Kobe, Japan 2 Osaka University Graduated School of Medicine, Osaka, Japan

A Phantom for Image Registration in Simultaneous Measurements of PET/MRI System

In the combined PET/MRI system, image registration of these two modalities is important because images are often displayed in fused form. We developed a new phantom for the PET/MRI system. The phantom has cylindrical shape having 60mm inner diameter, 50mm height cylindrical shape in which two types of holes and a groove are made. One type is 5-mm diameter holes which is parallel to the height direction of the phantom with which the registration of PET and MRI images in transaxial position, image sizes as well as image rotation angle are checked. The other is the angled 5mm-diameter holes with which the transaxial position and axial position are checked. The 10-mm separated grooves outside of the phantom are for mainly checking the axial position of the PET and MRI images. The phantom was tested using the integrated PET/MRI system. An F-18 solution was mixed with a black oolong tea (contrast material for MRI) and was contained with the phantom. Images are taken using the integrated PET/MRI. The transaxial images of PET and MRI showed usefulness of the straight holes in checking the image position between these two modalities. The angled holes were useful for checking the axial as well as the transaxial position. The grooves were useful for checking the axial position. With these results, it was confirmed that the developed PET/MRI phantom is useful for image registration and checking of PET/MRI systems.
M05-265:

S. Krishnamoorthy1, S. Stoll2, M. Purschke2, C. L. Woody2, P. O'Connor3, D. Schlyer4, P. Vaska4 1 Biomedical Engineering, Stony Brook University, Stony Brook, New York, USA 2 Physics, Brookhaven National Laboratory, Upton, New York, USA 3 Instrumentation, Brookhaven National Laboratory, Upton, New York, USA 4 Medical, Brookhaven National Laboratory, Upton, New York, USA

Investigation of 3D Event Positioning in a PET Detector with Continuous Scintillator and APDs

We have studied in detail the resolution properties of a prototype light-sharing detector using a single continuous 10 mm thick LYSO slab coupled with large-area APDs from Hamamatsu onto both sides. First experimental characterization of the in-plane resolution of the detector using simple Anger-logic yielded an average spatial resolution of about 3 mm FWHM across the entire face of the detector. An inherent advantage to placing APDs on both sides of the detector is the ability to measure depth-ofinteraction. The depth-of-interaction in addition to reducing parallax errors could also help improve the in-plane resolution by decreasing the light-spread blur. To further predict detector performance, a detailed, fully Monte Carlo model using GATE was also developed. Spatial resolution studies with our validated Monte Carlo model show that the spatial resolution scales with scintillator thickness, and better than 1 mm resolution is achievable with a 5 mm thick scintillator. We also implement and evaluate a maximum likelihood approach that uses the depth-of-interaction measuring capability of the detector to help locate the three-dimensional interaction point of a gamma-ray in the scintillator. Simulation studies suggest ~30% improvement in spatial resolution.
This work is supported by the U.S. Department of Energy under Prime Contract No. DE-AC02-98CH10886. M05-268:

P. Dokhale1, C. Staples1, J. Christian1, S. Cherry2, W. Moses3, K. Shah1 1 Radiation Monitoring Devices Inc., Watertown, MA, USA 2 Department of Biomedical Engineering, UC-Davis, Davis, CA, USA 3 Lawrence Berkeley Lab, Berkeley, CA, USA

Performance Measurements of a LYSO-SSPM Detector Module for Small Animal Positron Emission Tomography.

We present the performance of a compact PET detector module with a depth-of-interaction (DOI) capability based on a LYSO scintillator array coupled at both ends to CMOS solid state photomultipliers (SSPM). In this paper we present energy, coincidence timing resolution and flood imaging results for a prototype PET detector module consisting of a 4 x 4 block of LYSO scintillators, each crystal measuring 1.43 x 1.43 x 20 mm3 coupled to 4 x 4 array of SSPM (PS-SSPM) each SSPM measuring 1.5x1.5 mm2. All 16 elements in the flood image recorded with 22Na gamma-ray source were clearly visible and well separated from each other. The measured FWHM energy resolution with 511 keV gamma rays for all crystals in the array ranged between 14% 15%. The timing resolution measured for the complete detector module in coincidence with LYSO-PMT detector was 2.2 ns. The DOI resolution was measured for all crystals in the array by dual ended readout method. The average DOI resolution measured for corner crystal was ~ 2.3 mm and for middle crystal was 2.5 mm.

311

M05-271:

Imaging Reagents Study for Nuclear Medicine Using an Electron-Tracking Compton Gamma-Ray

S. Kabuki1, H. Kimura2, H. Amano2, Y. Nakamoto3, H. Kubo1, K. Miuchi1, S. Kurosawa1, M. Takahashi1, H. Kawashima3, M. Ueda4, T. Okada3, K. Ogawa5, K. Togashi3, H. Saji2, T. Tanimori1 1 Department of Physics, Gradulate School of Science, kyoto university, Kyoto, Japan 2 Department of Patho-functional Bioanalysis, Graduate School of Pharmaceutical Sciences, kyoto university, Kyoto, Japan 3 Radioisotopes Research Labaoratory, Kyoto University Hospital,, kyoto university, Kyoto, Japan 4 Radioisotopes Research Labaoratory, Kyoto University Hospital, kyoto university, Kyoto, Japan 5 Department of Electronic Informatics Faculty of Engineering, Hosei university, Tokyo, Japan We have developed an electron-tracking Compton camera (ETCC) for new imaging reagents study. Energy limitation of gamma camera is major problem for this study. However, our ETCC has a wide energy dynamic range (200-1300 keV). In this presentation, we will show the results of imaging regent study as follows: (1), Au-198 nanoparticle (412 keV) imaging for Drug Delivery System, (2), F-18-FDG (511 keV) and I-131-MIBG (364 keV) simultaneous imaging for double clinical tracer imaging, (3), Zn-65-porphyrin (1116 keV) imaging for high energy gamma-ray imaging.
M05-274:

Camera

E. Lorincz1, G. Erdei1, I. Peczeli1, C. Steinbach1, F. Ujhelyi1, T. Bukki2 1 Atomic Physics, Budapest University of Technology and Economics, Budapest, Hungary 2 MEDISO Ltd, Budapest, Hungary In Positron Emission Tomogaphs (PETs) high spatial resolution is usually achieved with block detectors using small size crystals in scintillator array and light sharing readout. The latter reduces the number of necessary photosensors and consequently the production costs. Different solutions are known for the determination of excited crystal pin in the scintillation array. We refer here to two reviews with many references [1], [2]. Our goal is to design an optical hardware module that provides solution for light sharing between electronically linked photodetectors and consequently allows precise determination of the excited pin position in a large block. The task is to rearrange uniformly the center of gravity of light distributions from each individual pin within a narrower area assigned by the centers of affected PMT's. Therefore it is most desirable to narrow somewhat the angular diameter of the light cone from each scintillator pin and to find a method for its deflection. Recently we have succeded to develop a new optical Monte Carlo simulation tool based on a parametric Zemax model [3] that is capable to predict reliably the light output of scintillator crystal pins of different configurations, i.e. geometry, surface finish and reflector material [4]. In this paper we use an extension of the model for the simulation of pins in a 21 x 21 crystal array assembled with a special light deflection module in a block detector with four 1.5 PMTs. The light sharing module consists of special double prism units on each pin as light deflectors optically coupled to a continuous light guide [5]. We present the concept of the optimized light guide, the outcome of simulations and also the validation results of first laboratory experiments with a small deflector model.
This work was supported by the National Office for Research and Technology (NKTH) grant NKFP-A1-2006-0017. [1] J.S. Karp, Against, European Journal of Nuclear Medicine, vol. 29, no. 11, pp. 1525-1528, November 2002. [2] B.J. Pichler, H.F. Wehrl, and M.S. Judenhofer, Latest Advances in Molecular Imaging Instrumentation, J. Nucl. Med., vol. 49, no. 6 (Suppl), pp. 5S-23S, June 2008. [3] ZEMAX: Software For Optical System Design, http://www.zemax.com/ [4] E. Lrincz, G. Erdei, I. Pczeli, C. Steinbach, F. Ujhelyi and T. Bkki, Light Output Analyzes of Scintillator Crystal Pins and Array for PET Detector Modules, 2008 IEEE Medical Imaging Conference (MIC) Conference Record, M10-176, pp. 4868-4871, 2008. [5] Patent pending, no. HU 09 00287, filed May 7, 2009. M05-277:

Optimized Light Sharing Module for PET Block Detectors

Study of a Positron Emission Tomograph Prototype Based on Multigap Resistive Plate Counters

P. Baesso, G. Belli, G. Musitelli, R. Nard, D. Pagano, S. Ratti, A. Vicini, P. Vitulo, C. Viviani INFN Pavia, University of Pavia, Pavia, Italia We are developing a prototype of a Positron Emission Tomograph (PET) scanner based on Multigap Resistive Plate Counters (MRPC). Although MRPC are usually intended as charged particles detectors, preliminary studies on detection efficiency to 511 keV gamma rays showed that they could be successfully used for human PET aplications, providing an economic alternative to crystals. Such studies include GEANT4 simulations and measurements using a 22Na source to test the detectors. Montecarlo studies on backscattering and peak efficiency have also been performed. Studies on suitable materials have been carried on as well. The prototype currently uder study is composed by an octagonal structure to emulate a PET ring and the signals from the MRPC are acquired by dedicated front-end boards prior to being sent to the data acquisition system. Data analysis is performed with standard imaging reconstruction algorithms. A 22Na source is used to produce the annihilation photons and test the prototype. First results on data acquisition and reconstruction are presented.

312

M05-280:

F. Taghibakhsh1,2, S. Cuddy1, T. Rvachov3, A. Reznik2,4, J. A. Rowlands1,2 1 Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada 2 Thunder Bay Regional Health Sciences Centre, Thunder Bay, ON, Canada 3 Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada 4 Department of Physics, Lakehead University, Thunder Bay, ON, Canada

Detectors with Dual-Ended Readout by Silicon Photomultipliers for High Resolution Positron Emission Mammography Applications

Reducing parallax error and increasing overall efficiency are two key factors in development of high resolution positron emission mammography (PEM) systems to detect early stages of cancerous activities in breasts. We propose silicon photomultipliers (SiPM) coupled to LYSO crystals in dual readout configuration for high timing and depth of interaction (DOI) resolution as PEM detectors. We examine the effect of surface finishing of the pixilated crystals on various detector performances such as DOI, timing and energy resolution to arrive at a proper detector design. Our experimental setup consists of single 2x2x20 mm3, and multiple 1x1x20 mm3 crystals coupled to SiPM arrays, custom designed dual readout electronics, 22Na and FDG positron sources. We used the scaled difference between the two SiPM signals for extraction of DOI information, while the sum of the signals provided energy and timing information (Fig. 1). We are investigating the effect of crystal side walls finishing on resolution and linearity of DOI measurement, as well as energy discrimination (Fig. 2 and 3). So far, our measurements indicate almost linear DOI resolution of ~0.8 mm for saw cut crystal side walls, and ~1.5 mm for polished crystals. Energy discrimination improves as DOI approaches either ends of the crystal, and resolution of better than 19% and 15% were measured for rough cut and polished crystals respectively using 22Na source; experiments with an FDG source (no lead collimator) resulted in energy resolution of better than 13%. We will present details of our experiments and results for DOI, timing and energy resolution for single and multiple detectors. The high gain and fast response of SiPM, comparable to those of photomultiplier tubes, improve timing resolution for better true coincidence detection, and the dual readout configuration provides high resolution depth of interaction information to reduce parallax error for essential for efficient and high resolution PEM.
M05-283:

J. Chamorro1, J. Aguirre1, J. Ripoll2, J. J. Vaquero1, M. Desco1 1 Unidad de Medicina y Cirugia Experimental, Hospital General Universitario Gregorio Maran, Madrid,, Madrid, Spain 2 FORTH, Institute of Electronic Structure Laser, Heraklion, Crete, Grece Fluorescence Enhanced Diffuse Optical Tomography (FDOT) retrieves 3D distributions of fluorophore concentration in small animals, non-invasively and in vivo. The FDOT problem can be formulated as a system of equations, d=Wf, where W is a weight matrix that couples the measurements (d) to the unknown spatial distribution (f) of the fluorophore concentration (forward problem). The Singular Value Decomposition (SVD) of W has been previously employed to solve the inverse problem (image reconstruction) and to study the imaging performance of FDOT. To achieve good image quality it is necessary to determine the number of useful singular values to retain. We present an automatic method that analytically calculates a threshold to select the significant singular values. We assess the validity of the threshold by analyzing the noise level of the reconstructed images. Afterwards, this work appraises the effect of different settings of the acquisition parameters (distribution of mesh points, density of sources and detectors) of a parallel-plate non-contact FDOT, in order to achieve the best possible imaging performance, i.e., minimum number of singular values of W, maximum information content in acquired measurements and minimum computational cost. We show that a U-curve based algorithm provides a suitable selection of the regularization parameter for SVD reconstruction of FDOT experiments, since this parameter falls into an acceptable range producing reconstructed images with a reasonable amount of noise and it fulfills the Picards condition thus assuring a satisfactory regularized solution. Furthermore, we conclude that the use of a mesh with lower density in the direction perpendicular to the plates achieves better performance than the usual isotropic mesh points distribution. Any increase in the number of mesh points, sources and detectors at distances shorter than the photon mean free path leads to slight improvements in image quality.
Acknowledgement This work is supported in part by Fundacin Caja Navarra (#12180), Ministerio de Ciencia e Innovacin (TEC2008-06715 and TEC2007-64731) and EU-FP7 project FMTXCT-201792. M05-286:

FDOT Setting Optimization and Reconstruction Using Singular Value Analysis with Automatic Thresholding

J. G. Brankov1, F. M. Parages1, M. N. Wernick1, Y. Yang1, T. S. Denney2 1 ECE, Illinois Institute of Technology, Chicago, Illinois 2 ECE, Auburn University, Auburn, Alabama

Estimation of Myocardium Deformation by Simultaneous Use of Tagged and Untagged Gated Cardiac MRI

In this work we present a method to estimate myocardium deformation that simultaneously uses both tagged and untagged cardiac-gated MRI image sequences. We show that this approach is more accurate and robust than traditional methods that 313

usually use only tagged cardiac-gated MRI data. Estimation of the deformation field is based on a deformable mesh model (DMM) that incorporates a myocardium surface model. The DMM is well suited for motion tracking and analysis of organs undergoing non-rigid deformation. Previously, we successfully utilized a DMM for heart motion tracking in cardiac SPECT and tagged MRI. Promising preliminary results are presented.
This work was supported by NIH grants P50-HL077100, HL065425 and HL091017. M05-289:

Regularization Methods in Iterative Algorithms for Variance Reduction on Compressed Sinogram Random Coincidences

V. Y. Panin Molecular Imaging, Siemens Healthcare, Knoxville, TN, USA The Poisson model based image reconstruction algorithms in PET require the estimation of the mean value of random coincidences. The random data are often acquired with a delayed coincidence technique, and expected randoms are estimated through variance reduction (VR) of measured delayed coincidences. In the past we have developed iterative VR algorithms that estimate singles rates from random compressed sinograms. Iterative methods have an advantage of being easily adaptated to any acquisition geometry and data compression method. Nevertheless the final estimation of single rates can be noisy in case of short scans. In the present work, we apply commonly used and specific for given task regularization methods. A monotonic sequential coordinate descent algorithm is used. Measured data from a Siemens TruePoint clinical scanner are used to validate the algorithm performance.
M05-292:

J. P. Jones1, X.-B. Pan2, J. Declerck2, M. E. Casey1 1 Siemens Molecular Imaging, Knoxville, TN, USA 2 Siemens Molecular Imaging, Oxford, UK

Synthetic 82Rb PET Images for Testing an Absolute Myocardial Perfusion Quantification Application

Two questions that arise in the development of software for quantitative myocardial perfusion (QMP) are (1) how to validate that the application determines the correct QMP, given that any patient QMP is by definition unknown, and (2) how to determine if the application is robust in the presence of realistic amounts of noise. We approach these issues by synthesizing anatomically plausible dynamic image series with known QMP and realistic noise content, thus providing a known ground truth for testing. We use the numerical NCAT phantom to create images with specified activities, then forward project using known Point Spread Function (PSF), then add an attenuation distortion and synthetic scatter. Poisson noise content was modeled from a patient study and parameterized by frame start and frame duration. To achieve realistic geometric effects, we used a norm taken from a camera with crystal efficiencies set to one. Synthetic prompt/random sinograms were reconstructed with clinically common methods. Time activity curves following a reference framing protocol were generated for 8 complete dynamic image series with different known QMPs using an ideal blood input function for the left and right ventricular chambers and a one-compartment model to predict activity in the left ventricular wall. A comparison between the expected QMP values and observed QMP values for the 8 test cases shows a strong (R2 = 0.9857) linear relationship exists with a slope of 0.89, even in the presence of realistic amounts of noise. This approach, using synthetic images, combined with other methods of testing, increases confidence in the fidelity of the results.
M05-295:

T. Hughes1, S. Shcherbinin2, A. Celler2 1 Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada 2 Radiology, University of British Columbia, Vancouver, BC, Canada

Improvement of Myocardial Perfusion Defect Severity Quantitation in Cardiac SPECT: a Simulation Study

Objectives: We aim at improving the quantitative assessment of the severity of myocardial perfusion defects in cardiac SPECT imaging. The idea of a numerical heart template is utilized, which enables a patient specific measurement of defect severity as opposed to the more traditional population-based approaches. Methods: Using NCAT we developed three male thorax phantoms with different orientations and sizes of the left ventricle. Each heart contained a small (5%) inferior wall defect with a severity of 20-80%. The SimSET code was used to perform 21 simulations modeling cardiac SPECT acquisitions with a Tc-99m radiotracer, LEHR collimator, 64x64 matrix, and 60 camera stops. A conventional method (CM) of defect severity assessment included the MLEM reconstruction with 40 iterations and a calculation of the ratio, RCM, of the average activity concentrations in the defect and the normal heart. Our template method (TM) calculates a new ratio, RTM, which is a correction to RCM by rescaling it between two reference levels corresponding to a completely non-perfused defect and a healthy myocardium. These levels are calculated by projecting and reconstructing two numerical heart templates (may be based on CT in clinical studies) with activity 314

ratios in defect to normal heart set to zero and unity, respectively. Results: The proposed TM method was more accurate and more sensitive to small changes in defect severity than CM. While CM showed no defect (RCM was equal to 0.98-1.02) in the case with 20% severity (true ratio is 0.8), TM led to RTM values of 0.84-0.92. On average, our TM technique exhibited a 17% improvement in defect to normal ratios relative the CM method. Conclusions: Our proposed method offers a patient-specific assessment of perfusion defect severity in SPECT without the limitations intrinsic to traditional methodologies (e.g. extreme heart geometries).
M05-298:

A Maximum-Likelihood (ML) Method for Combined Organ Activity Estimation and Volume-ofInterest (VOI) Registration from Planar Projections

N. Song, B. He, E. C. Frey Department of Radiology, Johns Hopkins Medical Institution, Baltimore, MD, USA Accurate organ activity estimates are essential for targeted radionuclide therapy (TRT) treatment planning. We have previously developed the Quantitative Planar (QPlanar) processing method. This method uses 3D organ volumes of interest (VOIs), an accurate model of the image-formation process, and maximum likelihood (ML) techniques to estimate organ activities from planar projections. However, this method requires the alignment of the 3D organ VOIs and the 2D planar projections. In this work, we have extended the QPlanar method by incorporating a maximum-likelihood based registration method to perform both organ activity estimation and volume-of-interest registration. We studied the accuracy of organ activity estimates using this proposed ML-based method and compared with the results from mutual information (MI) based registration followed by QPlanar activity estimation. To model the motion of organs more realistically than possible with a whole body rigid transformation, we used a motion model in which each organ moves rigidly but independently from the other organs. In this method, the 3D organ VOIs were initially registered with the planar projections using a 2D-to-3D mutual information (MI) based rigid registration method. This was followed by alternating linear ML estimation of organ activities and non-linear ML estimation of the rigid registration parameters for each organ. Image degrading factors including attenuation, scatter, and the full collimator detector response were modeled during both the registration and activity estimation steps. It was found that the new ML-based method provided improved accuracy of activity estimates for the kidneys and spleen while providing similar accuracy for the heart and liver.
M05-301:

M. E. Lyra1, K. Skouroliakou2, E. Vasoura3, A. Antoniou3 1 A' Radiology, Radiation Physics Unit, University of Athens, Athens, Greece 2 Physics, Technological & Educational Institute of Athens, Athens, Greece 3 A' Radiology, Ultrasound Unit, University of Athens, Athens, Greece

Texture Characterization in Ultrasonograms of theThyroid Gland

Object: This study tries a quantitative characterization of the thyroid tissue as far as homogeneity and echogenicity are concerned, to quantitatively evaluate thyroid texture. Digital image processing techniques offer the opportunity for texture description. Although there is no formal definition of texture, this particular method of description can quantify properties such as smoothness, coarseness and regularity. First-order and co-occurrence features are quantified and multi factor analysis is used to evaluate the optimal subset of parameters for thyroid texture. Introduction: Attempts for a quantitative evaluation of the echogenicity of the thyroid included the extraction of parameters based on gray level histograms such as mean gray-scale level (15, 64), standard deviation and highest frequency gray-level. Patients and Methods: Thyroid ultrasonograms of 40 patients were acquired for further image processing. The first order statistics computed are the mean and standard deviation of the gray levels of the image. The co-occurrence parameters were computed from the corresponding gray-level co-occurrence matrices (GLCM). Principal component analysis (PCA) is applied to reduce the number of features. In order to test the extracted components for their ability to characterize texture, a classification model is constructed using binary logistic regression. A random 70% of the ultrasonograms are used for model construction and the rest 30% for model testing. Results: The GLCMs were calculated for 14 distances horizontally, vertically and diagonally. The texture parameters correlated to the extracted components are homogeneity for small pixel separations, correlation for large separations, contrast for large separations and energy for medium separations. The ten components resulting from the PCA were used in order to construct a classification model by applying binary logistic regression.
M05-304:

J.-C. (. Cheng1, N. Agbeko2, J. O'Sullivan2, R. Laforest1 1 Dept. Radiology, Washington University, St. Louis, MO, USA 2 Dept. Electrical Engineering, Washington University, St. Louis, MO, USA

Evaluation of an Iterative Cascade Gamma Ray Correction Algorithm for Non-Standard PET Nuclides at Various Counting Statistics in High Resolution Small Animal PET Imaging

We describe an evaluation of an iterative cascade gamma ray correction algorithm which was developed to improve the quantitative accuracy of small animal PET imaging with non-standard PET nuclides. The cascade correction algorithm uses the 315

emission image and the attenuation map of the object to compute the shape or the spatial distribution of the coincidences caused by the cascade gamma rays, and the distribution is then scaled to the measured coincidences similarly to the typical scatter estimation methods. The sinogram profiles of the cascade estimate were compared between different iterations, and the cascade fraction was evaluated at various counting statistics for a given random fraction and count rate. A quick convergence for the shape of the cascade estimate was observed and found to be not sensitive to the number of counts within the frame, and a higher variation of the cascade fraction was observed for frames with lower number of counts. In addition, a higher variation of the cascade fraction is expected for frames with a higher random fraction and/or with a lower number of counts. As a result, a calibration for the scaling of the cascade estimate may need to be applied for the high random fraction and low count situations.
M05-307:

Component-Based Normalization Correction for the Clear-PEM Positron Emission Mammography

System

C. Guerreiro1,2, N. C. Ferreira2,3, R. Bugalho4, A. Trindade4, N. Matela5, M. V. Martins5, L. Mendes2,3, C. S. Ferreira4, P. Almeida5, J. Varela4 1 ESTeSC, Instituto Politecnico, Coimbra, Portugal 2 Instituto de Biofisica/Biomatematica, IBILI - Faculdade de Medicina da Universidade de Coimbra, Coimbra, Portugal 3 Instituto de Ciencias Nucleares Aplicadas a Saude, ICNAS, Coimbra, Portugal 4 Laboratorio de Instrumentacao e Fisica Experimental de Particulas, LIP, Lisboa, Portugal 5 Instituto de Biofisica e Eng. Biomedica, IBEB, Lisboa, Portugal We present a component-based normalization correction for the Clear-PEM positron emission mammography system. This system consists of two opposing parallel planar detectors that rotate around the breast. The distance between detector plates can vary to adapt to the patient and scintillation light is read at two ends of the crystals for Depth of Interaction (DOI) information. The normalization model currently accounts for intrinsic and geometric efficiencies using new methods specially developed for this purpose. Both efficiencies are calculated from data obtained with a planar source that is parallel to the two detector plates. Support for other components (deadtime and DOI) is also being developed. The whole normalization scheme is being assessed with real data using planar and cylindrical sources and the results from its validation will be presented at the conference.
M05-310:

Exact Reconstruction from Uniformly Attenuated Truncated SPECT Projection Data with Tiny a Proiri Knowledge

Q. Huang, G. T. Gullberg Department of Radiotracer Development & Imaging Technology, Lawrence Berkeley National Laboratory, Berkeley, CA, USA Kudo et al. presented a unique and stable solution to the interior problem in computed tomography (CT) given tiny a proiri knowledge on the intensity of the object distribution. In this work we advance their result to the interior problem in single photon emission computed tomography (SPECT) where a uniform attenuation is assumed. In the reconstruction, differentiated backprojection (DBP) was first performed on truncated SPECT data to obtain the truncated weighted Hilbert transform. Then with the a proiri information of a small portion of the region-of-interest (ROI), the remainder of the ROI was shown to be available using the method of the projection onto convex sets (PCOS). The algorithm is presented for a general geometry, and thus is suitable for the reconstruction of data acquired for orbits of a helical trajectory with a cone-beam or a pinhole collimator.
M05-313:

T. Humphries1, A. Saad2, A. Celler3, G. Hamarneh2, T. Moeller2, M. Trummer1 1 Department of Mathematics, Simon Fraser University, Burnaby, BC, Canada 2 School of Computing Science, Simon Fraser University, Burnaby, BC, Canada 3 Department of Radiology, University of British Columbia, Vancouver, BC, Canada

Segmentation-Based Regularization of Dynamic SPECT Reconstruction

Dynamic SPECT reconstruction using a single slow camera rotation is a highly underdetermined problem, which requires the use of regularization techniques to obtain useful results. The dSPECT algorithm (Farncombe et al. 1999) provides temporal but not spatial regularization, resulting in poor image contrast and low activity levels in organs of interest. In this paper we incorporate a semi-automatic segmentation algorithm (Saad et al. 2008) into the reconstruction process to improve the results. Following an initial reconstruction using the existing dSPECT technique, a user places seeds in the image to indicate regions of interest (ROIs). A random-walk based automatic segmentation algorithm then assigns every voxel in the image to one of the ROIs, based on its proximity to the seeds as well as the similarity between time activity curves (TACs). The user is then able to visualize the segmentation and improve it if necessary. Average TACs are extracted from each ROI and assigned to every voxel in the ROI, giving an image with a spatially uniform TAC in each ROI. This image is then used as initial input to a second run of dSPECT, in order to adjust it to better fit the projection data. We test this approach with a digital phantom simulating the kinetics of Tc99m-DTPA in the renal system, including healthy and unhealthy behaviour. Total TACs for each kidney and the bladder were calculated for the spatially regularized and nonregularized reconstructions, and compared to the true values. The TACs for the two kidneys were noticeably improved in every 316

case, while TACs for the smaller bladder region were unchanged. Furthermore, in two cases where the segmentation was intentionally done incorrectly, the spatially regularized reconstructions were still as good as the non-regularized ones. In general, the segmentation-based regularization improves TAC quality within ROIs, as well as image contrast.
M05-316:

Evaluation of Resolution Properties of Median Priors in List-Mode Reconstruction Methods for PET

M. Ahmad, A. Todd-Pokropek Medical Physics and Bioengineering, University College London, London, UK In list-mode acquisition, data is collected as a list of events with different attributes recorded for each event. List-mode reconstruction methods are used in this case. In this paper, we have contributed towards an assessment of resolution properties of list-mode image reconstruction methods, through an investigation of impulse responses, obtained using Local Impulse Response (LIR) function in Penalized-Likelihood (PL-EM) reconstruction methods with our proposed median root based (MPs) priors and have compared our results with standard quadratic priors (QPs). We have shown that better resolution can be recovered with these priors, whereas standard priors still produce a kind of over-smoothing. We have also presented results for Combine Priors (CPs), a hybrid of standard Quadratic Priors (QPs) and Median Based Priors (MPs), which can be used for better user controlled resolution characteristics, and noise resolution trade-off. Results show that our proposed priors are less sensitive to the parameter value which is a great advantage in PL-EM methods.
M05-319:

M. D. Walker1, P. J. Julyan1,2, P. S. Talbot1, T. Jones3, J. C. Matthews1 1 Wolfson Molecular Imaging Centre, The University of Manchester, Manchester, UK 2 North Western Medical Physics, The Christie NHS Foundation Trust, Manchester, UK 3 Academic Department of Radiation Oncology, The University of Manchester, Manchester, UK

Bias in Iterative Reconstruction of Low-Statistics PET Data: Benefits of a Resolution Model

Ordered-subset expectation maximization (OSEM) is an attractive and widely used method of reconstructing PET data. Several authors have reported bias when reconstructing frames containing few counts via OSEM, although the levels of bias reported varies substantially. Such bias may lead to errors in biological parameters as derived from dynamic PET data. We examine lowstatistics bias in OSEM image reconstruction of patient data and estimate the subsequent errors in biological parameter estimates. Patient listmode data were acquired during a [11C]DASB scan using a high resolution brain PET scanner (HRRT). These data were subsampled to create many independent, low-count replicates. Each replicate was reconstructed with and without the use of an image based resolution model (PSF). Time-activity curves (TACs) were then generated by Monte Carlo simulation and used to study the propagation of bias from the images into the physiological parameters of interest. Significant bias was observed when reconstructing data of low statistical quality. This bias was substantially reduced by including a PSF model (e.g. caudate head, 1.7M NEC, -6 % bias with PSF, -13 % bias without PSF). For the observed levels of bias, Monte Carlo simulations gave biases in the binding potential (BPND) of +4 and +10 % (with/without PSF). We conclude that OSEM reconstruction of dynamic PET data can yield parameter estimates of acceptable accuracy (for DASB), despite producing biased images at low statistics. The use of a resolution model is shown to reduce bias and is thus recommended. The magnitude of the bias for other tracers and methods of data analysis is yet to be evaluated.
M05-322:

Exact Step-and-Shoot Circular Cone Beam Reconstruction with a Common Line Scan

A. A. Zamyatin Toshiba Medical Research Institute USA, Vernon Hills, IL, USA Step-and-shoot circular scanning allows scanning a wide region of patient anatomy or a long object. It is also known to produce less dose than helical scanning in some cases. However, step-and-shoot circular scanning is based on approximate algorithms. When the cone angle is large, circular reconstructions suffer from cone beam artifacts, which results in jump of HU values at the boundary between the step-and shoot volumes. To solve the problem of cone beam artifacts, exact circle-and-line reconstruction method is available. However, exact circle-and-line reconstruction method in its current form cannot be applied to multiple stepand-shoot circular scans. The proposed method has two steps: 1. Collect step-and-shoot circular data 2. Collect the a single common line scan corresponding to the whole Z-extend of the scanned patient anatomy or scanned object. Note that various line view ranges are used for exact circle-and-line reconstruction corresponding to the appropriate circular scan position. Each line range is extracted from the common line scan. Note that many line views are re-used in various line ranges, which means good dose utilization.

317

M05-325:

Experimental Evaluation of Non-Uniform Object-Space Rebinning for Single Photon Emission Microscopy Imaging

N. Li, L. J. Meng Department of Nuclear, Plasma, and Radiological Engineering, The University of Illinois, Urbana,IL, USA In this paper authors present a non-uniform object space pixelation (NUOP) approach developed for a single photon emission microscope (SPEM) system, which is capable of offering an ultra-high spatial resolution for a target-region inside mouse brain. In addition, Modified Uniform Cramer-Rao bound (MUCRB) is implemented to assess the NUOP approach with different pixelation strategies. As demonstrated by Monte Carol simulation, the NUOP approach could improve the computation speed by 1-2 orders of magnitude without sacrificing resolution performance in the target-region. The further experimental estimation will be performed. NUOP will become a practical imaging tool for mouse brain study in SPEM system.
M05-328:

Y. Wu1, T. Ng2, Y. Yang1, K. Shah3, R. Farrell3, S. R. Cherry1 1 Department of Biomedical Engineering, University of California, Davis, Davis, CA, USA 2 Department of Biology, California Institute of Technology, Pasadena, CA, USA 3 Radiation Monitoring Devices Inc., Watertown, MA, USA In this work we study position-dependent timing shifts and timing resolution in position sensitive avalanche photodiodes (PSAPDs) and their effect on the coincidence window used in positron emission tomography (PET) systems using these devices. There is a delay in PSAPD signals that increases as the excitation position moves from the corner to the center of the device and the timing resolution concurrently worsens. The difference in timing between the center and corner can be up to 30.7 ns for a 14x14 mm^2 area PSAPD. This means that a PSAPD-based PET system could require a very wide coincidence timing window (>60 ns) if this effect is not corrected, although the individual crystal pairs still have full-width half-maximum (FWHM) timing resolutions better than 7.4 ns. In addition to characterizing the timing properties of PSAPDs, a correction method was developed and applied to data from a pair of PSAPD detectors. The timing offset correction reduced the timing shift of a crystal pair from 52.4 ns to 1.3 ns, improved FWHM timing resolution of the detector pair from 24.6 ns to 6.0 ns and reduced the timing window (sufficient to cover at least twice the FWHM for all crystal pairs) from 65.1 ns to 15.2 ns. The effect of PSAPD size on the timing performance was also evaluated.
M05-331: A powerful readout system for high resolution and high efficiency molecular imaging studies of cardiovascular diseases in mice

Timing Shift Study of Position-Sensitive Avalanche Photodiodes

P. Musico1, E. Cisbani2, F. Cusanno2, F. Garibaldi2, M. L. Magliozzi2, S. Torrioli2, G. Marano3, M. Musumeci3, M. Baiocchi4, L. Vitelli4, G. De Vincentis5, S. Majewski6, B. Tsui7 1 I.N.F.N. Genova, Genova, Italy 2 INFN, I.S.S Roma, Roma, Italy 3 Dipartimento del Farmaco, I.S.S Roma, Roma, Italy 4 Dipartimento id Ematologia, Oncologia e Medicina Molecolare, I.S.S Roma, Roma, Italy 5 Universita' La Sapienza, Roma, Italy 6 West Virginia University, Morgantown, WV, USA 7 Johns Hopkins Univeristy, Baltimore, ML, USA

A portable, compact, independent channel readout system for an array of Flat Panel photoMultiplier Tubes (FPMT) has been developed and tested. Using Hamamatsu H-9500 FPMT the system can handle an area of 20 x 20 cm2, with pixel size of 3 x 3 mm2, for a grand total of 4096 independent channels. The electronic system uses a custom VLSI chip as front-end and several Field Programmable Gate Arrays (FPGA): this gives high compactness, very good signal integrity (and S/N ratio), good power consumption and high flexibility. The connection with the acquisition PC is done using an high speed USB 2.0 link. A small animal SPECT detector has been equiped with this electronics. Studies of cardiovascular diseases by molecular imaging with radionculides are on going. Preliminary results on perfusion measurements on mice are available. Details on the electronic readout system will be given, together with the laboratory test results.

318

M05-334:

W. Hu1, Y. Choi1, J. Jung1, K. Hong1, J. Kang1, B. Min1, Y. Huh1, S. Shin1, H. Lim1, Y. Chung2 1 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 2 Department of Radiological Science, Yonsei University, Wonju, Korea

A FPGA-Based High Speed Multi-Channel Simultaneous Signal Acquisition Method for Positron Emission Tomography

A PET data acquisition (DAQ) system based on multiple high-speed, 8-channel DAQ cards is being developed. Each of these DAQ cards has 8-channel 14-bit 100 MHz simultaneous ADCs, a six million-gate FPGA with 104 MHz on-board clock and a 16bit 128 MB SDRAM. The SDRAM has a data transfer bandwidth of 450 MB/S, thus for raw data acquisition, only two channel ADCs can be simultaneously used at sampling rate of 100 MHz (for 8 channel simultaneous acquisition, the maximum sampling rate is about 28 MHz). In this study, data packages containing pulse arrival time, baseline, energy and position were saved instead of raw data of whole gamma signal pulse to reduce the amount of output data. Parallel signal processing containing arrival time detection, baseline calculation, pulse energy calculation and channel ID generation were implemented by model-based design (MBD) method to achieve high performance PET signal processing. Finally, data packaging and down sampling approaches were employed to achieve 8-channel simultaneous signal acquisition at 100 MHz sampling rate without exceeding the SDRAM data transfer bandwidth. To examine the functionality of the signal acquisition method, PET images were obtained using a pair of detectors. The PET detector is composed of a silicon photomultiplier (SiPM) array (3 mm x 3 mm, 4 x 4) coupled with LYSO array crystal (3 mm x 3 mm x 20 mm, 4 x 4). By rotating a pair of PET detectors for 180 degrees (6 degrees x 30 times) and simultaneously acquiring 8 channel (4 channels from each SiPM array) signals using one acquisition card, PET images of two F18 line sources were successfully acquired. The energy resolution for 8-channel gamma signals was 19% and the timing resolution for the coincidence channels was 1.5 ns. The spatial resolution for the acquired PET image was 3.1 mm. Experimental results show that, the signal acquisition method developed in this study will be useful for the development of high performance investigational PET systems.
M05-337:

F. Afzal1, T. H. Farncombe2 1 Medical Physics, McMaster University, Hamilton, ON, Canada 2 Radiology, McMaster University, Hamilton, ON, Canada

An Inexpensive Multi-Channel Digital Acquisition System for Nuclear Medicine Applications

In the past several years, there have been a multitude of new radiation detector components for use in PET and SPECT (solidstate detectors, scintillators and new photodetectors). With the introduction of these, and other, new components, the need to test them for their feasibility in imaging systems often becomes an obstacle to the further development of new systems. In an effort to ease the development process, we have designed and built an inexpensive 16 channel digital acquisition system for use in PET and SPECT applications. The system uses off-the-shelf commodity components in order to reduce costs, yet is fully featured and capable of reasonably high performance. The system consists of a commodity digital signal processor with 16 analog to digital converters, coupled to a host PC via high speed USB. The high performance DSP can be programmed for a number of signal processing functions easily. The system has been interfaced to a variety of photodetector systems and can easily be optimized for use with alternative radiation detectors. The system is scalable to multiple systems with each system operating independently with the host.
M05-340:

Point Cloud Photon Transport (PCPT) for Medical Imaging Applications

A. Sitek Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, USA We present a method for calculation of the photon propagation performed analytically in the object domain. In addition to the amount of scattered radiation, the method estimates the energy spectra of the scattered photons separately for each order of the elastic and inelastic scatter. These energy spectra can be used in the standard scatter correction in SPECT and PET, but also in novel applications that require distinguishing between first and second and higher orders of scatter. The PCPT is also applicable to X-ray Computed Tomography and can be used for model based correction for scatter and correction for beam hardening effects. The method is based on a multi-sampling point cloud image representation in which an image is defined as a set of points in 3D placed in unrestricted locations. Each point is assigned a volume, electron density, and 3D distribution (2D angular and 1D energy) of emission density for each order of scatter . This constitutes a 7D space in which calculations are performed. The algorithm is initiated by assigning 0-order of emission distribution (emission tomography) or first order of emission distribution due to scatter of primary beam (transmission tomography). By applying the order upgrade transform (OUT) the 6D distribution of n-order scattered photons is transformed to the 6D distribution of n+1-ordered scattered photons. The OUT is exactly the same for all orders of scatter. We performed feasibility and initial evaluation studies of the proposed method. Computing time for 16384 point cloud image for performing OUT was in the order of 1 minute. Compared with Monte Carlo simulation of Compton scatter (elastic scattering was not evaluated) the results agreed within a 5% for first order and 10% for higher orders margins.

319

M05-343:

K. K. Mitev1, A. S. Kirov2, Y. Y. Madzhunkov1, G. V. Gerganov1, I. Kawrakow3 1 Department of Atomic Physics, Sofia University "St. Kliment Ohridski", Sofia, Bulgaria 2 Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, USA 3 Ionizing Radiation Standards, National Research Council of Canada, Ottawa, Ontario, Canada

Study of the Influence of Photon Energy Cuts on the PET Simulation Results

The objective of this work is to study the influence of photon energy cuts on the results and speed of positron emission tomography (PET) simulations. The investigation is performed with two different MC codes: GATE and egs_pet. These codes were used to simulate the response of GE Discovery LS PET scanner to a point, 511 keV back-to-back, photon source placed at the center of a cubic 30x30x30 cm3 water phantom. The effect of the photon cuts on the simulation results is studied by making simulations with 32 energy cut values in the interval 0.3-350 keV. The dependence of the number of singles, primary and scattered coincidences on the applied energy cut is evaluated. The effect of the cuts on the simulation speed is also estimated. The results from the two codes are found to agree well within the statistical uncertainties for all the cuts applied. It is found that the simulation of atomic relaxation processes and the propagation of characteristic X rays play important role if an accurate modeling of a scanner system is to be achieved. For this scanner which is made of BGO crystals the usage of cuts below the K X-rays of Bi does not lead to statistically significant differences in the number of recorded events. Photon cuts in the interval 80-100 keV lead to less than 1.5% change in the number of singles, and less than 3% difference in the number of true and scattered coincidences. The differences are mainly due to the simulation/no simulation of Bi XK -rays in the BGO. Energy cuts higher than 170 keV result in a strong increase in the number of detected events due to the increased absorption of Compton scattered photons and are not appropriate for PET simulations. The simulation time does not change dramatically for photon cuts between 0.3 and 100 keV. In the current studies an acceleration of the order of 35% is achieved by changing the photon cut from 1 to 100 keV.
M05-346:

A Monte Carlo Study on the Spatial Resolution with Silicon-Based Detectors

J. Cabello, K. Wells Centre for Vision, Speech & Signal Processing, University of Surrey, Guildford, Surrey, United Kingdom Autoradiography (AR) is a method used to map the distribution of radiolabelled biomolecules deposited in a thin ex-vivo tissue specimen. Conventional film emulsion has been traditionally used as the imaging medium for AR but this technique presents a number of problems including low sensitivity and poor linearity among the most important. However, Silicon based imaging technologies have demonstrated some of the best overall performance in AR imaging. Work has already been published using CCD and CMOS detectors demonstrating higher sensitivity compared to film emulsion. The main problem that remains unresolved is the outstanding spatial resolution exhibited by conventional film, as yet unmatched by any other digital system. In this work we study the effect of various design parameters of CCD and CMOS detectors by means of Monte Carlo simulations using the Geant4 environment, including charge diffusion, detector noise and subsequent image correction, to address the design trade-offs between sensitivity, spatial resolution and Contrast to Noise Ratio.
M05-349:

A. S. Ayan1, R. Accorsi1,2, J. S. Karp1, S. D. Metzler1 1 Dept. of Radiology, University of Pennsylvania, Philadelphia, PA, USA 2 Dept. of Radiology, Children's Hospital of Philadelphia, Philadelphia, PA, USA

Geant4 Evaluation of the Impact of the Spatial Resolution Improvement on Contrast Recovery in a Small Animal PET System with Collimation

Two limitations on the resolution of a reconstructed PET image are sampling and detector pixel size. Using collimation (COL) that partially blocks each crystal reduces the effective crystal size. Using different COL positions increases sampling. In this study we determine the Contrast Recovery Coefficient (CRC) for a small animal PET scanner with and without COL in the transverse direction. We performed simulations of a single slice small animal PET system (205 mm diameter and 2x2x10 mm^3 LYSO crystals). The septa forming the COL were 1x2x10 mm^3 tungsten pieces covering half of each crystal transaxially. Phantoms (25 mm diameter) with one cold and 3 hot lesions with diameter D (D=0.5, 1, 2, 3 mm) were simulated with two S:B ratios (4:1; 6:1). CRC=(S/B-1)/(T-1) where S and B are mean lesion (hot or cold) and background count densities, and T is true uptake ratio. CRC was measured from reconstructions to quantify the impact of the resolution improvement. Results show COL improves CRC compared to uncollimated PET. For 1 mm hot lesions (4:1), scanned for the same duration, the collimated mean hot lesion CRC values (STD) were 0.44 (0.03) (center) and 0.24 (0.02) (off-center). Uncollimated results were 0.31 (0.01) and 0.19 (0.01), respectively, where STD is the standard deviation of measured CRCs of an ensemble of images. Although the total coincidences for the same scan time is less by about a factor of 4 in the collimated system the measured mean CRC is higher. The sensitivity loss in collimated PET manifests itself as worse STD in measured CRC and noisier images. When the collimated PET scan time is increased, the STD of measured CRCs improves and reaches that of uncollimated PET. In certain imaging scenarios, it may be possible to scan longer with a collimated PET system to make up for the sensitivity loss.

320

In conclusion, our study shows the use of COL has the potential to improve the quantification and detection capabilities of a small animal PET system.
M05-352:

Coincidences Originating from a Single Photon: An Unrecognized and Potentially Significant Source of Scatter in Small Animal PET?

A. L. Goertzen Department of Radiology, University of Manitoba, Winnipeg, Manitoba, Canada Reported scatter fractions measured for small animal PET systems are commonly much higher than one would expect based on the size of the object being imaged. The large scatter fractions are generally attributed to 'gantry scatter' and are made worse by the common use of wide energy windows to increase sensitivity. In this paper we propose the hypothesis that a significant number of the detected scattered events in animal PET systems operating with wide energy windows actually originate from the scattering of a photon in a detector and the subsequent detection of this scattered photon in another detector. These coincidences originating from a single photon form a scatter profile that is distinct from the normal low frequency profile associated with scatter in PET due to the attenuation of the scattered photons in the object being imaged. This leads to the sinogram profile of these single photon coincidences resembling a transmission sinogram. Line source emission data, for a line source in air and in a 25 mm diameter mouse sized phantom, is simulated using GATE for the geometry of a Siemens Inveon PET system. The magnitude of single photon scatter coincidences is estimated by simulating a single photon emission source in place of the positron emission source. The results suggest that for an energy window of 150-850 keV 42% of detected scattered events for the line source in air originate from single photon scatter events. For the line source in the mouse phantom, the results suggest that 19% of the detected scattered events originate from single photon scatter events. Our preliminary data suggest that coincidences from a single photon should be considered as a significant source of scatter in small animal PET systems that either operate with a wide energy window or have poor energy resolution.
M05-355:

A Simulation Study of the Charge Collection in Highly Pixilated CdZnTe Detector for PET Imaging

S. A. Komarov, Y. Yin, H. Wu, Y.-C. Tai Radiology, Washington University in St.Louis, St. Lous, USA The intrinsic spatial resolution of detectors is the primary limiting factor of the image resolution of small-bore animal PET scanners using short range positron source such as F-18. We have previously demonstrated the use of a high resolution insert device in conventional PET scanners to improve their image resolution via the virtual pinhole PET geometry. Highly pixilated room temperature semiconductor such as CdZnTe may be a good candidate for the insert detectors as the anode size can be made very small to achieve high resolution. In this study, we consider a planar CZT detector with pixilated anode as an insert detector. The anode size is mostly limited by the charge sharing between neighboring pixels. This charge sharing is affected by the following: diffusion of charge particles, original charge particles distribution after the interaction, multiple interactions of gamma rays inside the detector, electronic crosstalk, etc. To evaluate charge sharing in highly pixilated CdZnTe detectors, we developed the Monte Carlo simulation tool to guide the experiment design and validate its results. The MC tool includes a gamma tracking module and a signal processing module. The first module gives the set of interactions and energy depositions in CdZnTe detectors. The second module generates the signals from cathode and anodes pixels for every interaction. The signal is induced by the movement of holes and electrons. The movement of the particles in simulations is a drift in cathode-anode direction and thermal diffusion in the lateral direction. The results will be used to estimate the smallest anode size that can be used without significant charge sharing, guide the correction for the depth of interaction and help to model the charge sharing effect due to the diffusion and Compton scattering in the proposed system.
M05-358:

A Methodology for Optimizing the Acquisition Time of a Clinical PET Scan Using GATE

N. A. Karakatsanis, K. S. Nikita School of Electrical and Computer Engineering, National Technical University of Athens, Athens, Greece The acquisition time of a PET scan is a critical parameter when designing imaging protocols for clinical nuclear medicine studies. The statistical quality of the projection data increases when longer acquisition times are selected. However, very large scanning periods can limit the number of PET studies performed and, moreover, increase the probability of motion artifacts. The competing objectives of good statistical quality and short acquisition time are both depending on the counting rate performance of the system. The noise equivalent count rate (NECR), which measures the rate in which statistically important coincidence events are counted by a PET system, is employed in this study to quantify the counting-rate performance. Thus, higher NECR values allow for acquisition of relatively larger number of true coincidence counts at the same scanning time. NECR is directly depending, for a particular patient-scanner system, on the amount of radioactive dose injected into the patient and acquires a peak value for a certain range of dose values. Thus, a minimal acquisition time can be achieved by estimating this optimal dose range prior to a scan. In this simulation study we propose an alternative optimization methodology. Initially, a regular low dose is selected and used as a constant. Then the NECR response is modeled, using Geant4 Application for Tomography Emission 321

(GATE) simulation package, as a function of the parameters of the patients body size, the coincidence time window, the deadtime response and the energy window. Subsequently, the optimal scanning time is estimated, based on the simulated NECR, as the minimal scanning time necessary to acquire 20 million noise equivalent counts (NEC) per bed position. For this purpose, we employed a validated Biograph PET/CT scanner model, where six hypothetical dead-time responses were simulated as well as three coincidence time windows. Finally, we used three NCAT phantoms of different size, and three energy windows.
The authors would like to thank the OpenGATE collaboration for allowing us to use the GATE simulation software for the purposes of this study M05-361:

A. Mehranian1,2, M. R. Ay1,2,3, N. Riahi Alam2, H. Zaidi4 1 Dept. of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Tehran, Iran 2 Research Center for Science and Technology in Medicine, Tehran University of Medical Sciences, Tehran University of Medical Sciences, Tehran, Iran 3 Research Institute for Nuclear Medicine, Tehran University of Medical Sciences, Tehran, Iran 4 Div. of Nuclear Medicine, Geneva University Hospital, Geneva, Switzerland In quest of deeper insight into the characteristics of diagnostic x-ray spectra and better understanding of factors influencing them, many attempts have been carried out for measuring the spectra by considering as many contributing factors as possible. Nevertheless, based on what turns out from literature, quantifying the effect of anode roughness on x-ray spectra and elucidating its potential importance in total inherent filtration still remains an active research area. The necessity of further investigation motivated us to conduct a Monte Carlo-based study using the MCNP4C code to precisely characterize x-ray spectra in a surfaceroughened anode. Our modeling method is based on the concept that a rough surface can be defined into MCNPs phase space by deploying many height-differed cubes. In order to implement this concept, a computer code was written using MATLAB that defines the geometry of such a rough surface in Cartesian coordinate system using data derived from images of focal track. The images are processed such that the depth and spatial pattern of cracks can be incorporated into gray level values of image pixels. Initially four models were defined by adjusting cracks depth to 5, 10, 20 and 30 m and ignoring the roughness of intracrack regions. The spectra were simulated for pure tungsten anode at tube voltages of 50, 70, 100 and 120 kVp with total inherent filtration of 2.5 mm Al. The filtering effect of anode roughness was assessed by comparing the simulated spectra with the spectra of a perfectly plain surface anode. For tube voltages of 50, 70, 100 and 120 kVp, intensity loss of ~ 1 to 13 % was obtained for these anode models. When quantified by root mean square deviation, the variations in spectral shape were found to be more pronounced for deeper cracks and lower voltages. Further assessment is being carried out by calculating the spectra using more realistic anode models and by evaluating more spectral parameters versus anode surface roughness.
M05-364:

Quantitative Assessment of the Effect of Anode Surface Roughness on the Diagnostic X-Ray Spectra Using Monte Carlo Simulation

C.-H. Baek1,2, S. J. An1, S.-J. Lee1,2, J. Y. Hwang1,2, J. S. Kim3, Y. H. Chung1,2 1 Department of Radiological Science, College of Health Science,Yonsei university, Wonju, 220-710, South Korea 2 Institute of Health Science, Yonsei University, Wonju, 220-710, South Korea 3 Department of Radiation Oncology, Samsung Medical Center, Seoul, 135-710, South Korea

Modeling of the Photon Beam from 6 MV Varian 2100C LINAC Using GATE Simulation

The GATE(Geant4 Application for Tomographic Emission) code was employed to calculate the dosimetric performance of the photon beams from the 6 MV Varian 2100C linear accelerator. We modeled the treatment head includes the major geometric structure within the beam path such as the target, primary collimator, flattening filter, ion chamber, jaws and water phantom. The model was used to estimate the energy spectra of the photon beam, percentage depth dose and the dose profiles in the water phantom. The accuracy of the calculated results was examined by comparing them with the measured data. The simulated results agreed very well with the measured data. It has been found that the GATE code is an effective tool for simulating the clinical photon beam.
M05-367: A Modified Statistical System Model and an Accelerated Monte Carlo Simulation for a Silicon Detector Based Compton Medical Imaging System

L. Han, N. Clinthorne Medical School, The University of Michigan, Ann Arbor, MI, USA The objective of this research is to demonstrate a modified statistical system model and an approach of accelerated Monte Carlo simulation for studying the capability of a dual-planar Compton medical imaging system constructed with silicon and NaI detectors to image 364.4keV energy photons emitted from I131. The Compton imaging system is a potential medical imaging instrument for effectively observing the process of radionuclide cancer treatments. Since decoupling the tradeoff between spatial resolution and detection efficiency inherent to the absorbing collimation of conventional Anger Camera, the Compton imaging system provides improved imaging performance in both detection efficiency and spatial resolution especially for higher energy 322

photons. In the study, a modified statistical system model of the Compton system was developed considering all factors in the Compton process including Doppler broadening, energy resolution and spatial resolution of the scattering and absorbed detectors in addition to the Compton image formation process. To conquer the limitation of low simulation speed as employing the existent all-purpose Monte Carlo simulation platform due to low Compton interaction cross-section inside of the silicon detector, the developed Monte Carlo simulation involves the techniques of force detection and variance reduction to speed up the simulation. It is proved that two promotion approaches are of benefit to performance analysis and image reconstruction for the Compton imaging system.
M05-370:

R. V. Donepudi1, S. Medasani2, Z. Zhong3, R. Cesareo4, A. Brunetti4, T. Yuasa5, T. Akatsuka5, T. Takeda6, G. E. Gigante7 1 Physics, Sir.C.R.R. (A) College, Eluru-7, India 2 Physics, SDMMC, Vijayawada, India 3 NSLS, BNL, NewYork, USA 4 Istituto di Matematica e Fisica, Universita di Sassari, Sassari, Italy 5 Bio-System Engineering, Yamagata University, Yonezawa, Japan 6 Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Japan 7 Dipartimento di Fisica, Universita di Roma, Roma, Italy

Synchrotron-Based DEI-CT Images of Zebrafish Eye with and Without Contrast Agents at 20, 30 and 40 keV

Synchrotron based DEI-CT, images of small Zebrafish was acquired using 20 keV, 30 KeV and 40 keV synchrotron X-rays with contrast agents such as water, physiological saline and iodine. Recently, refraction properties of X-rays turn to be more attractive advantages for imaging over the absorption properties. Refraction is orders of magnitude more sensitive, particularly for biological or low (Z) materials. Diffraction-enhanced imaging (DEI) technique exploits these refraction properties. Diffractionenhanced images of small Zebrafish exhibited significant extinction contrast and were readily visualized by DEI in computed tomography (DEI-CT) mode, on both sides of the analyzer curve and also at the peak of the analyzer curve. Extinction properties of Zebrafish and Zebrafish eye structural features were characterized by DEI-CT. In the presence of hard tissues, which absorb a lot of X-rays, the flux emitted by the X-ray tube was considerably less for visualization of internal complex features and generates images with poor contrast. The flux emitted by the synchrotron source was much higher, and generates images with improved contrast for better visualization of the animal internal organs. Inview of this, Synchrotron based DEI-CT may be a better choice for a detailed anatomy and visualization with contrast agents We performed the experiments at the peak position of the sample, low angle side and high angle side of the rocking curve. Based on the quality of the image, we analyzed rocking curve s at multiple energies. As such there no experimental research using synchrotron X-rays in particular with DEI and the associated contrast mechanisms for this sample. Recent results provided potential source of information. This research motivated to work further in related fields.
M05-373:

A Data Adaptive Approach to the Robust Fitting of PET Data: Application to Group and Test-Retest

F. Zanderigo1, R. T. Ogden1,2, C. Chang3, S. Choy1, A. Wong1, R. V. Parsey1,2 1 Department of Molecular Imaging and Neuropathology, New York State Psychiatric Institute, New York, NY, USA 2 Department of Psychiatry, College of Physicians and Surgeons, Columbia University, New York, NY, USA 3 Department of Mathematical Sciences, New Jersey Institute of Technology, University Heights, Newark, NJ, USA Model fitting of PET time activity curves (TACs) is typically accomplished according to the least squares (LS) criterion, which is optimal for data with Gaussian distributed errors, but not robust in the presence of outliers (e.g., head motion). In contrast, quantile regression (QR) is an alternative for robust fitting that provides estimates not heavily influenced by outliers only losing a little efficiency, relative to LS, when no outliers are present. A data adaptive (DA) approach, relying on the rationale that QR should be used when outliers are present and LS otherwise, chooses LS or QR based only on the observed TAC by measuring the influentiality of each TAC point and using QR when the maximum influentiality among the TAC points is higher than a given threshold. We compare the results of applying DA to those using LS on a group of 24 and a test-retest data set of 11 controls studied with [11C]-DASB to investigate whether DA improves the parameter estimates accuracy and the test-retest experiments reproducibility. In group analysis, DA decreases the standard deviation (SD) of the radioligand distribution volume (VT) estimates (relative improvement range: 0.12% - 2.59%), while keeping the within-group average VT values almost unchanged. This ~2.5% decrease in the estimated outcome SD would result in a ~5% decrease in the number of subjects needed to maintain the same statistical power. Applied to test-retest, DA reduces the VT estimates percent difference (PD) by a quantity in the range 0.01% - 4.39% in the 63% of the regions for the majority of test-retest pairs. DA selected QR over LS in 61% and 66% of the cases in group and test-retest analysis, respectively. The practical import of this reduction in the SD and PD of estimated outcomes, achievable by simply implementing a different fitting approach, is smaller sample size to detect differences between groups and enhanced test-retest experiments reproducibility (i.e. improved occupancy studies sensitivity).

Analysis

323

B. Ravindranath1, S. S. Junnarkar2, M. L. Purschke2, S. H. Maramraju1, S. S. Southekal1, S. P. Stoll2, J. F. Pratte3, P. Vaska2, C. L. Woody2, D. J. Schlyer2 1 Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA 2 Brookhaven National Laboratory, Upton, NY, USA 3 Universite de Sherbrooke, Sherbrooke, Quebec, Canada PET is a powerful imaging technique which can be used to non-invasively quantify physiological processes in vivo. By combining PET images of tissue over time with an accurate measure of the amount of radioactivity in the blood, unique kinetic parameters can be obtained which, when used in combination with appropriate mathematical models, can be used to characterize physiological process for the diagnosis and characterization of diseases, and patient response to a given course of pharmacologic, radiation or behavioral therapy. However, the gold standard for determining the input function is invasive in nature and involves arterial blood sampling. In order to determine the input function non-invasively, a wrist scanner capable of measuring arterial blood radioactivity was developed. The feasibility of such an approach was proven using a prototype planar imaging device. A significant upgrade in the wrist scanner has been made in a new detector ring that is capable of fully 3D tomographic imaging. Input functions derived from this system and its comparison with arterial blood samples will be presented.
M05-379:

M05-376:

3D Tomographic Wrist Scanner for Non-Invasive Determination of Input Function

K. Berradja1, M. Bentourkia2 1 Imaging, University of Mostaganem, Mostaganem, Algeria 2 Dept of Nuclear Medicine and Radiobiology, Universite de Sherbrooke, Sherbrooke, Quebec, Canada

Kinetic Modeling of Brain FDG Data with Input Function Derived from Images by Independent Component Analysis

The input function (IC) is mandatory in quantitative positron emission tomography (PET) imaging. The measurement of IC can be processed by manual or automatic blood sampling. In both cases the samples need to be analyzed for radiotracer concentration in plasma and cross calibrated with the PET scanner. These procedures are cumbersome and introduce uncertainties. In repetitive small animal imaging, this procedure becomes almost impossible. Several approaches have been reported to non-invasively extract IC from the images among them factor analysis and independent component analysis (ICA). In the present work we extract IC from the internal artery in human brain images by means of ICA and we compare the regional metabolic rates of glucose (rCMRG) calculated with IC derived from ICA and by manual blood sampling. By decomposing the images into tissue and blood images, ICA allows to correct for spillover from tissue to blood. Meanwhile, the blood images of the brain arteries are subject to partial volume effect (PVE) and the concentration of the radiotracer in the plasma is needed instead of blood radioactivity. We correct for these two effects by normalizing the blood curve extracted from ICA blood images with three plasma samples. The results show a good decomposition of the slices of brain images by means of ICA in tissue and blood images. The normalized ICA-IC appeared to have comparable shape and intensity as the plasma sampled IC. rCMRG values as calculated with the sampled plasma IC and ICA-IC presented a relative difference of 3%. A part of this difference might be attributed to the differences in blood and image sampling times where the integration of the two ICs affect the value of the perfusion during the kinetic modeling. In conclusion, ICA is a strong approach in image decomposition to extract blood curves, while a normalization is mandatory to correct for PVE and blood to plasma ratio.
M05-382:

Noise Reduction and Contrast Enhancement for Small-Dose X-Ray Images in Wavelet Domain

M. Sakata, K. Ogawa Hosei University, Faculty of Science and Engineering, Tokyo, Japan An important issue in diagnostic imaging is to achieve high quality images with the smallest possible radiation dose. However, xray images acquired with a small radiation dose are affected by the quantum noise. In this paper, we propose a new method to obtain high quality images acquired with a small dose by enhancing the contrast of degraded images adequately. The proposed method consisted of the following five steps: (1) pre-processing of an original image with a gamma correction, (2) wavelet transform of the image, (3) noise reduction by a modified BayesShrink method in the wavelet domain, (4) contrast enhancement of denoised wavelet coefficients, and (5) reconstruction of a contrast enhanced image by an inverse wavelet transform. The validity of the proposed method was confirmed by comparing images acquired with a standard dose and those acquired with a small dose (1/10 of the standard dose) and processed with the proposed method.

324

M05-385:

G. Mettivier1, P. Russo1, N. Lanconelli2, S. Lo Meo2 1 Department of Physics, INFN and University of Napoli, Napoli, Italy 2 Department of Physics, Alma Mater Studiorum - University of Bologna and INFN, Bologna, Italy

Evaluation and Correction of Scattering in Cone-Beam Breast Computed Tomography: a Monte Carlo and Experimental Phantom Study

Several research groups are working on pendant-geometry Cone-Beam Breast Computed Tomography (CBBCT) as a new methodology in breast tumor diagnosis. The technique is made possible thanks to improvement in flat panel technology and cone-beam reconstruction algorithms. One of the principal problem in CBBCT scanning, and in general in Cone-Beam Computed Tomography (CBCT) scanning, is the presence of a great scatter fraction in the projection images. This fraction is usually much larger than the one occurring in X-ray mammography with a compressed breast and a low-energy X-ray beam and significant also with respect to Fan-Beam Computed Tomography (FBCT), where the cone angle is about ten times lower. The high Scatterto-Primary Ratio (SPR) reduces the contrast and increases image artifacts and CT number inaccuracy. Therefore, the study and the reduction of this component is an important and actual challenge. Recently, several studies using both a Monte Carlo (MC) simulation method or empirical techniques have been reported on the evaluation of the SPR as a parameter to quantify scatter in CBBCT. In this study, MC simulations of PMMA cylinders of different diameters and a PMMA breast phantom are used to evaluate the 2D scatter distribution in projection images in CBBCT, with irradiation geometry and technique factors that reproduce the experimental setup used for validation measurements. The simulations were performed by varying X-ray beam energy and filtration but with the same exposure level of 1 mGy air kerma at the isocenter. This study will allow us to optimize the choice of the beam spectrum and to develop a scatter correction algorithm.
M05-388:

Directionally Selective Regularization for Sliding Preserving Medical Image Registration

D. Ruan Radiation Oncology, Stanford University, Stanford, CA, U.S.A Medical image registration aims to recover physical deformation between source and target observations. It is an ill-posed optimization problem, and regularization energies are often used to encourage desirable characteristics in the solutions. Motion induced deformations are generally smooth, with possible sliding along certain organ/structural boundaries. Conventional regularizations tend to over-smooth motion boundaries and fail to address the sliding issue. Some previous studies on slidingpreserving regularization have demonstrated the potential to achieve more physical results with regularization designs, but have revealed numerical stability issues for 3D registration: 1D sliding singularities may occur as an artefact by relaxing the curl of deformation. Observing that sliding often occur along organ boundaries, which can be inferred quite reliably from CT images, the current study takes advantage of such prior information to obtain a superset of the sliding interface locations. The normal directions of the potential local sliding interfaces are obtained along the intensity gradients. Depending on whether knowledge about local sliding direction is available, a local coordinate system is constructed and the local deformation is presented with respect to such orthonormal basis. Certain entries in the local deformation Jacobain are relaxed from the smoothness regularization to allow for sliding across the motion interface. Connection is drawn between entries in the curl of the deformation field and the relaxation terms. The assumption on the set relationship between intensity discontinuity and motion interface is valid for clinically relevant sliding-inclusive registration problems, such as head and neck or thoracic cases. A preliminary test was conducted to register two thoracic CT volumes at end inhale and exhale stages. Even in the most pessimistic inference scenario, the proposed method produced promising results.
The author thanks Drs. Marc Kessler, Jeff Fessler and Selim Esedoglu for motivating this work. M05-391:

P. L. Rossi1, S. Masetti1, A. Turco1, D. Falchieri1,2, N. Lanconelli1,2, D. Bianchini1, L. Roma3, A. Margotti2, M. Zuffa2, G. Baldazzi1,2 1 Dep.t Physics, University of Bologna, Bologna, ITALY 2 INFN, Bologna, ITALY 3 Dep.t Radiology, S.Orsola-Malpighi University Hospital, Bologna, ITALY Quasi-monochromatic Multi-Energy CT imaging can produce many upgrades to polychromatic ones, in terms of background subtraction as for tissues recognition, that it seems to be the fundamental task to investigate the pathology growing like angiogenesis. To analyze these capabilities, a CT prototype with quasi-monochromatic X-ray source was developed at the University of Bologna-Physics Department. The small bandwidth energy peaks are produced by Bragg diffraction on HOPG crystal and a triple energy reconstructing algorithm permits to erase the projection error, i.e. noise due to background lack of homogeneity. After the acquisition of three quasi-monochromatic images of patient at proper energies, we can emphasise the signal of the investigated tissue erasing projection errors and the operator can select to: 1) exalt images of contrast mediums injected in vessels, for example iodine or 2) separate contribution of different tissues, reconstructing point-by-point their density, 325

Small Animal Imaging with a Quasi-Monochromatic Multi-Energy CT

electron density or effective Z map. Tests performed on dedicated phantoms show that triple energy technique drastically reduces background signals, making possible the visualization of very low concentration of iodine contrast medium (also 10 time smaller than traditional ones) and the tissues separation. Beneath the possibility to separate different signals, our M.E.C.T. becomes a very interesting X-ray apparatus to extract information on physic-chemical structure of tissues. For this reason, images of small animal (in particular, fishes and mice) were acquired, in order to analyze sensitivity and resolution, also testing the applicability in tumour growth study and angiogenesis development on mouse. Aim of this work is to discuss imaging proprieties of quasimonochromatic CT, showing both the images of dedicated phantoms and the in-vivo images of small animals.
M05-394:

An Iterative Reconstruction Method for Multiple Moving Objects

X. Jin, L. Li, Z. Chen, Y. Xing Engineering Physics, Tsinghua University, Beijing, China The reduction of motion artifacts is one of the crucial problems in CT reconstruction, especially in medical diagnosis and therapy where the motion of organs or the moving of injured patients while scanning can not be neglected. It is already known that when there is only one object in the projection area and its motion can be described with single affine transformation function, the accurate reconstruction can be achieved through motion compensation algorithms. However when there are multiple objects obeying different affine transformations, the algorithm will suffer from artifacts. Since multiple objects model is much a better approximation to the actual physical situation than single object model, it is more significant to practical applications. To solve this problem, we proposed a new iterative method. Delivered from accurate FBP motion compensation method for single moving object, the iterative one for multiple objects reconstructs every object separately using motion compensation method in each iteration, and uses forward projection to amend sinograms for next iteration. The method shows a fast speed of convergence and suppresses the artifacts significantly.
This work was supported in part by the NSFC grant 10575059 & 60772051. M05-397:

Automated Dual-Exposure Technique to Extend the Dynamic Range of Flat-Panel Detectors Used in Small-Animal Cone-Beam Micro-CT

A. Sisniega, J. J. Vaquero, M. Abella, I. Vidal-Migallon, E. Lage, M. Desco Unidad de Medicina Y Cirugia Experimental. Hospital General Gregorio Maranon, Madrid, Spain This work presents an approach to extend the dynamic range of X-ray flat-panel detectors for cone-beam micro-CT by using two different acquisitions of the same sample, taken at two different X-ray photon fluxes with the same X-ray beam peak energy and filtration. Photon flux for the first scan is chosen as the maximum possible not saturating the detector in the low attenuation areas. The second exposure level is calculated from the data acquired in the first exposure, assuming that the detector response to the incoming radiation is linear. To combine both datasets, we model the response of a detector pixel, as well as the overall gain of the detector. Each pixel of each dataset is weighted according to its value. Pixels whose value lies within the high gain region of the detector, receive the higher weights. The weighted Joint Probability Density Function (JPDF) is obtained by assuming each dataset follows an independent Gaussian distribution. The final pixel value is calculated by means of a maximum-likelihood estimation performed on the JPDF. The algorithm has been tested imaging two different phantoms on a small-animal cone-beam CT. We have performed comparative experiments using the proposed dual-exposure technique and a regular single-exposure acquisition, following a fully automated procedure. In both cases the dose delivered to the subject was the same. Image quality was assessed from the image noise level and the presence of artifacts on the reconstructed slices, at different dose levels. The dual-exposure data show higher signal-to-noise ratio. Also, image artifacts are noticeably reduced in the combined dataset. Since the proposed technique enhance the quality of the data extending the dynamic range of the detector without increasing the delivered dose, it is particularly suitable to image samples which contain both low and high attenuating regions. Furthermore, the extension of the technique to use more than two exposure levels is straightforward.
This work is partially funded by the european project FP7-201792 and with grants from the Ministerio de Ciencia e Innovacin, project TEC200806715-C02-01. M05-400:

K. Oh1, M. Kim2, M. Yun1, Y. Kim2, S. Nam1 1 Biomedical Engineering, INJE University, Gimhae, South Korea 2 Medical Image and Science, INJE University, Gimhae, South Korea

Mercury(II) Iodide Flat Panel Radiation Detectors for Simultaneous Acquisition of Static and Moving Image

Mercuric iodide deposited on flat panel thin film transistor (TFT) array is one of the best alternate photoconductive materials for direct digital X-ray detectors for both static and moving image application in medical imaging. The mercuric iodide is coated 326

onto the array by a Particle-In-Binder (PIB) process and scaled up to the 7inch 8.5inch size required in common medical imaging application. A TFT array with a pixel pitch of 139microns was used for detector. Mercuric iodide coating thickness around 200 microns was tested with beam energy between 40kVp and 100kVp utilizing exposure ranges typical for both static and dynamic imaging. Detector performances were evaluated by obtained images. Mercuric iodide deposited on flat panel thin film transistor (TFT) array is shown to exhibit high sensitivity to X-rays, excellent spatial resolution and high Detective Quantum Efficiency (DQE). Especially it is quite suitable for moving image because of low image lag. Resolution tests on resolution target phantoms showed that resolution is limited to the Nyquist frequency for the 139 microns (resolution ~3.6lp/mm) pixel detectors. The ability to operate at low voltages (~100V) gives adequate dark currents for most application and allows low voltage electronics designs. Also the detector can use exceptionally low dose-rate X-ray illumination because of the very high X-ray sensitivity, which exceeds any other known X-ray detector material. The fabricated detector represents the most advanced photoconductor material available today for flat panel, high resolution, x-ray, medical detector, which alternates conventional a-Se technology.
This research was financially supported by the Ministry of Knowledge Economy (MKE) and Korea Industrial Tech-nology Foundation (KOTEF) through the Human Resource Training Project for Strategic Technology M05-403:

R. G. Figueroa1, E. G. Pena1, W. H. Gerber2, J. F. C. A. Veloso3 1 Physisc Deparment, Universidad de La Frontera, Temuco, Chile 2 Physic Institute, Universidad Austral, Valdivia, Chile 3 I3N, Physic Deparment, University of Aveiro, Aveiro, Portugal

Optimization of Planar Arrangement Micro Collimators (PAMC) for Images with X-Rays.

We study the characteristics of a Planar Array Micro Collimator (PAMC) for optimization of an X-ray image of the chemical elements. The image is generated by the X-ray fluorescence induced in a sample by a beam spread from an X-ray tube. Detection of the energy (E) and position, (x, y) of the fluorescent radiation generated in the plane of the sample with a MicroPatterned Gaseous Detector (MPGD) using a Micro Hole & Strip Plate (MHSP) + PAMC and a adequate electronic readout, allows obtaining a two-dimensional map image of the chemical elements associated with the sample under study. The image quality is strongly influenced by the characteristics of collimation system used, it is therefore necessary to know the distribution of radiation within the collimated MPGD and determine their effect on the intensity and spatial resolution of the image (x, y) and energy (E) as well as its ability to obtain an image of a map of chemical elements in real time. Model was established theoretical basis and simulates the operation and java and GEANT4, in which different parameters are considered for PAMC such as: materials, distribution of the holes, diameter (20 to 500 m) and lengths of the micro holes (100 to 1000m), including distance, location and separation between the detector and sample and location of the PAMC. While this study focuses on the case identified above, using a detector PAMC + MHSP a microstructure can be applied to other MPGD or other position sensitive detector used in Nuclear Medicine, or other medical or industrial applications where improved spatial resolution is required.
This work was supported by the National Fund for Scientific and Technological Development of Chile FONDECYT M06: PET/SPECT instrumentation 1 Thursday, Oct. 29 M06-1: 14:00-15:30 International Ballroom Center

M. D. Haselman1, S. Hauck1, T. K. Lewellen2, R. S. Miyaoka2 1 Electrical Engineering, University of Washington, Seattle, WA, USA 2 Radiology, University of Washington, Seattle, WA, USA

FPGA-Based Pulse Parameter Discovery for Positron Emission Tomography

Modern Field Programmable Gate Arrays (FPGAs) are capable of performing complex digital signal processing algorithms with clock rates above 100MHz. This, combined with FPGAs low expense and ease of use make them an ideal technology for a data acquisition system for a positron emission tomography (PET) scanner. The University of Washington is producing a series of high-resolution, small-animal PET scanners that utilize FPGAs as the core of the front-end electronics. For these next generation scanners, functions that are typically performed in dedicated circuits, or offline, are being migrated to the FPGA. This will not only simplify the electronics, but the features of modern FPGAs can be utilizes to add significant signal processing power to produce higher resolution images. In this paper we report how we utilize the reconfigurable property of an FPGA to self-calibrate itself to determine pulse parameters necessary for some of the pulse processing steps. Specifically, we show how the FPGA can generate a reference pulse based on actual pulse data instead of a model. We also report how other properties of the photodetector pulse (baseline, and event triggers) can be determined automatically by the FPGA.

327

M06-2:

R. Vinke1, S. Seifert2, D. R. Schaart2, H. T. van Dam2, F. J. Beekman2,3, H. Loehner1, P. Dendooven1 1 KVI - University of Groningen, Groningen, The Netherlands 2 Delft University of Technology, Delft, The Netherlands 3 University Medical Centre Utrecht, Utrecht, The Netherlands

Optimization of Digital Time Pickoff Methods for LaBr3-SiPM TOF PET Detectors

Scintillation detectors based on LaBr3:Ce crystals and silicon photomultipliers (SiPMs) are promising for time-of-flight (TOF) positron emission tomography (PET). LaBr3:Ce is a fast and bright scintillator, while SiPMs have low transit time jitter and high gain. We focus here on the optimization of digital signal processing (DSP) time pickoff methods for such detectors. Several methods are compared, including conventional leading edge (LE) estimators, pulse fitting routines and statistical least square estimators. The analysis is performed on digitized waveforms originating from two bare 3 x 3 x 5 mm3 LaBr3: 5% Ce crystals, coupled directly to 3 x 3 mm2 SiPMs. A high bandwidth preamplifier provided an energy and a timing signal. Timing signals were digitized at 8 GS/s and 10 bit resolution. The noise on the digitized timing signal is mainly introduced by the digitizer itself. High gain amplification of the timing signal minimizes the noise-to-slope ratio of the pulse rising edge on which the time pickoff is performed. As a result, the noise contribution to the timing resolution is negligible and all time pickoff methods show similar coincidence timing resolutions of 100-105ps FWHM. It is shown that linear fitting and extrapolation to the baseline of the pulse rising edge gives rise to incorrect time pickoff, resulting in artificially good timing resolutions below 100 ps. In a second series of measurements, timing analysis is performed on large monolithic 18 x 16 x 20mm3 LaBr3:Ce (5%Ce) crystals, coupled to 4 x 4 arrays of SiPMs. In addition to the time pickoff analysis, a correction is made for the depth-of-interaction related time walk. This is achieved by determining the 3D photo-conversion location inside the crystal using Maximum Likelihood Estimation (MLE). A simplified simulation shows that this time walk correction can improve the coincidence timing resolution by a factor of 1.5.
M06-3:

M. A. N. Korevaar1,2, M. C. Goorden1,2, J. W. T. Heemskerk1,2, D. R. Schaart2, F. J. Beekman1,2,3 1 Department of Nuclear Medicine, University Medical Center, Image sciences institute, Utrecht, Netherlands 2 Department of R3, Applied Sciences, Radiation Detection and Medical Imaging, Delft, Netherlands 3 Molecular Imaging Labs (MILABS), Utrecht, Netherlands

Cramer Rao Lower Bound Optimization of an EM-CCD Based Scintillation Gamma Camera

Gamma cameras based on Charge Coupled Devices (CCDs) coupled to scintillators can be used for high resolution imaging of gamma photons. The performance of the gamma camera will be influenced by many parameters, including temperature, gain, pixel size, Quantum Efficiency QE) and noise. Since the choice of parameters may have a non-trivial influence on spatial and energy resolution and experimentally testing of all possible settings is often impossible, a reliable method that predicts gamma camera performance is desirable. In this paper we apply Cramer Rao (CR) lower bound calculations to predict the spatial and energy resolution of a CCD-based gamma camera. Here we investigate the influence of temperature, gain, pixel size, QE and noise but this approach can also be applied to other parameters. Our set-up consists of an E2V EM-CCD optically coupled to a CsI(Tl) scintillator. We assume an isotropic light spread distribution and an EM-CCD noise model that takes into account EMgain, readout noise, multiplication noise, clock induced charge (cic) and dark current noise. We have experimentally verified the resulting output probability distribution which is used as the input for the CR lower bound calculations. The CR lower bound for spatial and energy resolution are calculated as a function of gain and temperature. The CR results show that good performance is obtained in a temperature range up to -20 degrees Celsius and we find an optimum value for the EM-gain. We conclude that CR lower bound calculation is a useful tool to optimize gamma camera performance for many parameters, including temperature, gain, pixel size, QE and noise.
M06-4:

Preclinical and Clinical PET Detector Design Considerations Using Silicon

H. Peng, P. Olcott, C. Levin School of Medicine, Stanford University, Palo Alto,CA, USA We are developing a new high-resolution PET block detector using the silicon photomultiplier (SPM) for both clinical (~3 mm resolution) and preclinical (~1 mm resolution) PET/MRI applications. As a new type of photon detection device, each SPM comprises thousands of microscopic avalanche photodiodes operated in Geiger mode; This detector exhibits the advantages of compact size, high gain and the ability to operate in a strong magnetic field. In this work, we investigated coupling arrays of 3 mm and 1 mm scintillation crystals to a 4x4 array of 3x3 mm SPM pixels. For the 3.2 mm crystal pitch array, 2x2 crystals were coupled one-to-one to a 2x2 portion of the 4x4 SPM array charge without any multiplexing). The 511 keV photopeak energy resolution (the average for four SPM pixels) is 15.3+/-0.2% FWHM. The individual crystals in the array can be clearly resolved with average peak-to-valley ratio of 23.1+/-0.8. We studied acrylic plastic light diffusers of different thickness for sharing light from 1 mm pitch crystals to the 3 mm pixels of the SPM array. For the case of a 6x6 array of 1 mm crystals directly coupled to 2x2 pixels of the SPM array with only optical multiplexing, individual 1 mm crystals were not well resolved as the size of the SPM (3 mm) is too coarse to resolve the crystals of finer size (1 mm). There was also no distinguishable 511 keV photopeak. When we coupled an 8x8 array of 1 mm crystals 328

through a 1.5 mm thick light diffuser to a 3x3 portion of the 4x4 SPM array, an electrical multiplexing was implemented along with the optical multiplexing to further reduce 9 readout channels down to 4. All 64 crystals are resolved with an average peakto-valley ratio of 4.48+/-2.06. The 511 keV photopeak energy resolution of the global energy spectrum (after the normalization per 1 mm crystal) was 21.2+/-0.4% FWHM.
M06-5:

T. Kormoll1, D. Kunath1,2, W. Enghardt1,2 1 OncoRay - Radiation Research in Oncology, Technische Universitt Dresden, Dresden, Germany 2 Institute of Radiation Physics, Forschungszentrum Dresden-Rossendorf, Dresden, Germany Experience during percutaneous ion beam therapy at Gesellschaft fr Schwerionenforschung (GSI), Darmstadt, Germany showed that in-vivo dose monitoring can contribute to the treatment's success. It has been studied whether a similar system, independent of treatment device and based upon the techniques of positron emission tomography (PET) can be implemented for high energy photon therapy. Simulation shows a strong correlation between the distribution of the annihilation points of positrons emerging from pair production and the dose. Design parameters of a detector system have been assessed which is capable of registering annihilation radiation during treatment. A test device system has been assembled which can be operated with various scintillator crystals and avalanche photodiodes (APD). A custom digital signal processing system has been implemented. Phantom experiments at an Oncor Impression (Siemens, USA) accelerator showed a severe background that does not stem from scattered primary photons but from leakage radiation from the treatment head. It was not possible to shield the detectors sufficiently and thus there was no success in identifying annihilation radiation.
M06-6:

3D In-vivo Dosimetry for Photon Radiotherapy Based on Pair Production

Determining Block Detector Positions for PET Scanners

L. A. Pierce, R. Miyaoka, T. Lewellen, A. Alessio, P. Kinahan Radiology, University of Washington, WA, USA We present an algorithm for accurate localization of block detectors in a positron emission tomography (PET) scanner. Accurate reconstruction of PET images requires precise knowledge of the physical position and orientation of the detectors. However, in some systems, e.g. small animal scanners, block detector positioning and orientation can have relatively large tolerances, leading to implicit errors in the coincidence line-of-response (LOR) positioning. To compensate we utilize a rotating point source phantom where the rotational angle of the phantom is used to precisely determine the location of each scintillator crystal within a detector block. The aggregate block positions are then applied to the system model to determine the true location of each LOR. Images reconstructed with the more accurate LOR positioning demonstrate improved image fidelity.
M07: Image Reconstruction 2 Thursday, Oct. 29 16:00-18:00 International Ballroom Center

M07-1: MAP Reconstruction of Fourier-Rebinned PET Data Using Estimated Sinogram Blurring Function from Point Source Scans

M. S. Tohme, J. Qi Biomedical Engineering, University of California, Davis, Davis, CA, USA The accuracy of the system model that governs the transformation from the image space to the projection space in positron emission tomography (PET) greatly affects the quality of reconstructed images using iterative methods. For efficient computation in reconstruction, the system model in PET can be factored into a product of geometric projection and sinogram blurring function, where the former is often computed based on analytical calculation, and the latter is estimated using Monte Carlo simulations. To further speed up reconstruction, fully 3D PET data can be rebinned into a stack of 2D sinograms and then be reconstructed using 2D iterative algorithms. In previous work, we proposed an approach to estimating the sinogram blurring function from experimental scans of point sources. In this study, we apply this sinogram blurring estimation to Fourier rebinned (FORE) PET data and incorporate the estimated sinogram blurring function in the subsequent 2D maximum a posteriori (MAP) reconstruction algorithm. Physical phantom experiments were performed to compare the proposed method to existing approaches, including a 2D MAP without sinogram blurring model and a 2D MAP with a Monte Carlo based blurring model. The results show that the proposed method produces images with improved contrast and spatial resolution.
This work is supported by NIH grant R01 EB005322

329

M07-2: High Resolution PET Image Reconstruction for the Siemens MR/PET-Hybrid BrainPET Scanner in LOR Space

J. J. Scheins1, L. Tellmann1, C. Weirich1, E. Rota Kops1, C. Michel2, L. G. Byars2, M. Schmand2, H. Herzog1 1 Forschungszentrum Jueluch GmbH, Institute of Neuroscience and Medicine, Juelich, Germany 2 Molecular Imaging Division, Siemens Medical Solutions USA, Rockford, TN, USA

The Siemens MR/PET-hybrid BrainPET scanner is based on Avalanche Photo Diode (APD) technology to allow the operation of the BrainPET scanner within the field of a Magnetic Resonance (MR) gantry. In this way, PET listmode data and MR sequences can be simultaneously acquired. The BrainPET has 32 detector heads containing 6 detector blocks in the axial direction. Each block consists of a matrix of 12x12 crystals with dimensions of 2.5x2.5x20 mm^3. In total, the system comprises 27648 individual crystals resulting in approx. 260 millions Lines-of-Response (LOR). The specific system design, including the APDbased electronics as well as the geometry in terms of gap structure and comparatively small system diameter cause some particularities in the measured LOR data which have to be accurately characterized. For the normalisation the LOR detection efficiencies are decomposed into involved crystal efficiencies and a LOR-dependent component describing geometrical variations. Variance-reduced corrections are fitted by using Ge68 plane source measurements. In addition, a count rate dependent pile-up correction becomes necessary. For the attenuation correction a template-based method using MR images is used. Besides the accuracy of the data corrections also the specific reconstruction algorithm significantly influences the image quality. Here, the library PRESTO is utilized for the fully-3D iterative reconstruction. PRESTO provides very fast projectors based on the volumeof-intersection weighting using a memory-resident, pre-calculated system matrix, thus allowing the accurate consideration of all detected LORs without angular compression or rebinning. The high computational effort results in artefact-free images of high quality in terms of resolution and minimal noise.
M07-3:

A. Andreyev1, A. Sitek2, A. Celler1 1 MIRG, University of British Columbia, Vancouver, BC, Canada 2 Dept. of Radiology, Harvard Medical School and Brigham and Women's Hospital, Boston, MA, USA Compton camera (CC) has been proposed as an imaging tool in astronomy, industry, homeland security and medical imaging where source location can be identified based on detection of Compton scattered and photoabsorbed photons. However, due to the geometrical complexity of the problem, image reconstruction from CC data is difficult when using techniques such as filtered backprojection (FBP) or maximum likelihood expectation maximization (MLEM). In this paper, we propose a novel stochastic origin ensembles (SOE) approach based on Markov chains (previously implemented and tested for PET) to the reconstruction of CC data. The advantages of this method are that it is universal (i.e. works with any camera geometry), parallelizable, and does not require forward- and back-projection operations nor voxelization of the image space. During image reconstruction the origins of every measured events are randomly assigned locations on the conical surface (which is the CC analog of lines-of-response in PET). Therefore, the image is defined as an event ensemble holding the coordinates of all possible event origins. During the course of the reconstruction origins of the events are moved and the acceptance of the new event origin is determined by the acceptance probability which is proportional to the change in the event density. For example, if the event density at the new location increases, it is always accepted. As a result, the reconstructed image converges to the quasi-stationary state which can be voxelized and displayed. Implementation of corrections for detector energy resolution can be done at almost no additional computational cost, which is a major advantage of SOE over other reconstruction methods. Our preliminary results indicate that for distributed sources the NMSE errors, measuring the difference between the true and the reconstructed image, are typically below 5%.
M07-4: Optimized Weighting for Fourier Rebinning of Three-Dimensional Time-of-Flight PET Data to NonTime-of-Flight

Stochastic Image Reconstruction Method for Compton Camera

S. Ahn1, S. Cho2, Q. Li1, R. M. Leahy1 1 University of Southern California, Los Angeles, CA, USA 2 Massachusetts General Hospital, Boston, MA, USA

Fully three-dimensional (3D) time-of-flight (TOF) PET scanners provide the potential for significant signal-to-noise ratio (SNR) improvement in clinical PET. Therefore, TOF PET technology is attracting increasing interest and commercial TOF systems are in production or development. However, fully 3D TOF PET image reconstruction is challenging due to the huge data size. One solution to this problem is to rebin TOF data into a lower dimensional format. We have recently developed Fourier rebinning methods for mapping TOF data into non-TOF formats and achieved substantial SNR advantages over sinograms acquired without TOF information. However, such mappings for rebinning into non-TOF formats are not unique and optimization of rebinning methods has never been investigated. In this paper we address the question of optimal rebinning in order to make full use of TOF information and consequently to maximize the image quality. We present a unified framework based on an approximate generalized projection slice theorem to derive approximate mappings between any data spaces and then, within the framework, 330

we derive an optimal rebinning method based on best linear unbiased estimators. We demonstrate by Monte Carlo simulations that the optimal rebinning method achieves significant variance reduction compared to other rebinning methods and the data acquisition without TOF information in both rebinned sinograms and reconstructed images.
M07-5:

Attenuation Artifacts and Time-of-Flight PET

T. G. Turkington, J. M. Wilson Duke University Medical Center, Durham, NC, USA Attenuation artifacts in PET are seen when attenuation correction (AC) is not performed and when AC is performed but is based on an incorrect attenuation map. PET attenuation artifacts are generally more profound than the straightforward differences in photon attenuation through surrounding tissue from one source location to another. Effects such as the apparent radioactivity in gas pockets in the body or concave contours of the body surface, distortions, and a pronounced body contour are all in addition to the expected nonuniformities due to depth in the body. We have investigated the effects of time-of-flight (TOF) reconstruction on PET artifacts. Uniformity without AC (NAC) was investigated in a large tapering F-18-fillled phantom that increases from a small end to the size of a large patient. Throughout the length of the phantom, the NAC images with TOF were more uniform that the non-TOF images. A whole-body phantom with oval cross section was imaged with uniform F-18 background and a 5.5 cm air-filled sphere at one section, and a 5.5 cm hot (8x background) sphere at another section. In NAC images, the air-filled lesion was artificially hotter than background with non-TOF, but comparable to background with TOF. When this section was corrected with a uniform AC (as if the gas pocket had moved before transmission scan), the non-TOF image showed the sphere to be artificially hot (hotter than background) whereas the TOF images reduced the artifact. In the section with the hot sphere, the NAC distortions typical near the bladder (higher counts anterior and posterior to bladder; depleted areas lateral to bladder) were greatly diminished with TOF images. While AC is necessary for quantitation in PET, the use of TOF and NAC may be useful in providing more interpretable images in some situations were an attenuation map is impossible to obtain, or introduces errors of its own. TOF may also lessen some artifacts in corrected images.
M07-6:

Motion-Compensated Fully 4D PET Reconstruction Using PET Data Supersets

J. Verhaeghe, P. Gravel, R. Mio, R. Fukasawa, P. Rosa-Neto, J.-P. Soucy, C. J. Thompson, A. J. Reader Montreal Neurological Institute, McGill University, Montreal,QC, Canada Patient head movement not only inhibits the full potential of high spatial-resolution neurological PET imaging, but it also significantly degrades fully 4D reconstruction when using temporally-extensive basis functions. In this case a single movement in just one of the time frames propagates to impact the reconstruction of all other time frames. Here we propose a novel motioncompensation strategy through the use of PET data supersets and demonstrate its application to fully 4D reconstruction. The richly-sampled superset is populated by considering head motion as equivalent to shifts in scanner position. This requires the positioning of the list-mode events into the superset and the creation of time-dependent normalization factors for the superset. A great advantage of this approach is that the attenuation factors for the superset need only be computed once for the reference position. This approach adapts readily for use with existing fully 4D reconstruction methods with the only modification being the introduction of time-dependent normalisation factors. Using simulated as well as real high-resolution PET data from the High Resolution Research Tomograph (HRRT), we demonstrate both the detrimental impact of head motion in fully 4D PET reconstruction as well as the efficacy of our proposed motion-compensation method. This is an important step to realising the potential of fully 4D reconstruction methods for patient studies.
M07-7:

PET Projection Data Supersets for Reconstruction with Acquisition Motion

J. Verhaeghe, A. J. Reader Montreal Neurological Institute, McGill University, Montreal,QC, Canada Motion during PET data acquisition is either introduced by design (e.g. couch wobble, detector rotation, bed translations) or through motion of the subject (e.g. rigid body motion of the head, respiratory motion). At present such effects are generally considered very differently in PET imaging, with intentional motion regarded as beneficial for enriching data sampling whilst subject motion is regarded as a detrimental effect to be eliminated. In this paper we propose the concept of PET projection data supersets leading to a generalization of PET image reconstruction in the presence of any type of known acquisition motion. This could open the way to optimal and unified exploitation of all sources of motion in image reconstruction. Our work includes rigid body and elastic object motion. As such, conventional motion-compensation techniques are special cases of our proposed method. We propose different approaches for the practical implementation of the PET data superset and reveal both the connections and differences between them. Using simulated data, we demonstrate the PET data superset concept and its potential to exploit object motion to the benefit of reconstruction using simulated data.

331

M07-8:

J. Yankam Njiwa1, C. Baltes1, D. Ratering1, M. Rudin1,2 1 Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland 2 Institute of Pharmacology & Toxicology, University of Zurich, Zurich, Switzerland For the sake of short acquisition time and increased spatial resolution of reconstructed images, several methods have already been proposed for Magnetic Resonance Imaging (MRI) that acquire partial or undersampled k-space data and use this known information about the imaged object to deal the reconstruction problem. This paper presents the development of a new methodology for reconstructing parallel Magnetic Resonance Images from reduced undersampled scans in k-space. It exploits the combined use of the analytic image reconstruction and SENSE processing to further accelerate the acquisition time. Assessments were made by experiments on mouse and rat brain images and compared with conventional Homodyne SENSE reconstruction. While being simple and not requiring any phase correction of the reconstructed image, satisfactory results are obtained with an increase of acceleration factor.
M08: X-ray imaging 1 Friday, Oct. 30 M08-1: 08:00-10:00 International Ballroom Center

Analytic Image Concept Combined to SENSE Reconstruction

Improved Contrast and Spatial Resolution with Single Photon Counting (SPC) for an Area X-Ray Imager, the Newly Developed High-Resolution Micro-Angiographic Fluoroscopic (MAF) Detector

A. Jain, A. Kuhls-Gilcrist, D. R. Bednarek, S. Rudin Toshiba Stroke Research Center, SUNY at Buffalo, Buffalo, NY, USA Although in radiological imaging, the prevailing mode of acquisition is to integrate the energy deposited by all x-rays absorbed by the imaging detector, much improvement in image spatial and contrast resolution could be achieved if each individual x-ray photon were detected and counted separately. In this work we compare the conventional energy integration (EI) mode with the new single photon counting (SPC) mode for a recently developed high-resolution Micro-Angiographic Fluoroscopic (MAF) detector, which is uniquely capable of both modes of operation. The MAF has 1024x1024 pixels of 35 microns effective size and is capable of real-time imaging at 30 fps. The large variable gain of its light image intensifier (LII) provides quantum limited operation with essentially no additive instrumentation noise and enables the MAF to operate in both EI and the very sensitive low-exposure SPC modes. We used high LII gain with very low exposure (<1 x-ray photon/pixel) per frame for SPC mode and higher exposure with lower gain for EI mode. Multiple signal-thresholded frames were summed in SPC mode to provide an integrated frame with the same total exposure as EI mode. A heavily K-edge filtered x-ray beam (avgerage energy of 31 keV) was used to provide a nearly monochromatic spectrum. The MTF measured using a standard slit method showed a dramatic improvement for the SPC mode over the EI mode at all frequencies. Images of a line pair phantom also showed improved spatial resolution with 11 lp/mm visible in SPC mode compared to only 8 lp/mm in EI mode. In SPC mode, images of human distal and middle phalanges showed the trabecular structures of the bone with far better contrast and detail. This improvement with the SPC mode should be advantageous for clinical applications where high resolution is essential such as in mammography and extremity imaging as well as for dual modality applications, which combine nuclear medicine and x-ray imaging using a single detector.
NIBIB RO1 EB002873 NINDS RO1 NS43924 Toshiba Equipment Grant M08-2:

M. Stampanoni1,2, S. A. McDonald1,3, F. Marone1, C. David1 1 Swiss Light Source, Paul Scherrer Institut, Villigen, Switzerland 2 Institute for Biomedical Engineering, ETH Zuerich, ETH Zuerich, Switzerland 3 Dept. of Radiology, University of Lausanne Medical School, Lausanne, Switzerland

Advanced X-Ray Phase Contrast Tomographic Imaging Using a Grating Interferometer.

The use of conventional absorption based X-ray microtomography can become limited for samples showing weak attenuation contrast. However, a wide range of samples studied in biology and materials science can induce significant phase shifts of the Xray beam. Hard X-rays grating interferometry allows to overcome some of the problems that have impaired the wider use of phase contrast in X-ray radiography and tomography. A grating interferometer has limited requirements on monochromaticity and on spatial coherence, which get though more stringent as high sensitivity and therefore larger fractional Talbot distances are required. In addition the instrument is scalable up to large fields of view. These characteristics, combined with the fact that such instrument is particularly sensitive to shallow phase gradients, makes it an ideal technique for soft tissue studies in the fields of medical and biological imaging. We recently implemented grating-based Differential Phase Contrast (DPC) imaging into TOMCAT, a beamline dedicated to tomographic microscopy and coherent radiology experiments at the Swiss Light Source. Developments have been made focusing on faster and more efficient data acquisition as well as a pipelined data post-processing to enable a high-throughput of samples. Optimization of the data acquisition procedure enables a full phase volume (1024 x 1024 332

pixels x 1000 projections x 9 phase steps 9000 projections in total) to be acquired in 20 minutes (with a pixel size of 7.4 m). The sensitivity for such an experiment is in the order of 1 mg/cm3. Several new features have been implemented like local-DPC, widefield-DPC, fast-DPC and dark-field imaging. We will show the advantages of these new features by a case study which focus on the visualization of soft tissue structures in a rat brain. Currently, on-going developments aiming at the transfer of this technology (including the new features) onto conventional, X-ray tube-based, systems will also be presented.
M08-3:

M. Giordano1, C. Neukirchen1, M. Bertram1, W. Mali2, M. A. Viergever3, E.-J. Vonken2 1 X-ray Imaging Systems, Philips Research Laboratories, Aachen, Germany 2 Department of Radiology, University Medical Center Utrecht, Utrecht, The Netherlands 3 Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands

Perfusion Estimation in the Peripheral Vasculature Using C-Arm X-Ray Systems

Peripheral arterial occlusive disease is caused by occlusions or blockage in the peripheral vasculature and can lead to hypoperfusion of more distal parts (e.g. the lower legs). When this occurs, the tissues are not sufficiently fed by blood and can become ischaemic. Treatment options include minimally invasive techniques such as percutaneous transluminal angioplasty (PTA). The outcome control of these procedures today is limited to visual inspection of blood flow through the recanalized vessels. The quantity of blood that effectively supplies the tissues, the perfusion, can only be measured using diagnostic modalities (CT or MRI). In this paper we introduce a framework for the quantitative assessment of perfusion in peripheral body parts on interventional Carm systems. We propose to estimate the average density of contrast along the direction of the acquired x-rays, assuming homogeneity along the rays. This quantity allows the computation of TACs in the tissues, that may be of clinical interest, if measured before and after the intervention, enabling direct and quantitative outcome control of complex procedures. Our method is based on a two-scan acquisition protocol. First a rotational soft tissue scan is made so as to allow extraction of the 3D anatomy of the perfused area (vascular territory and tissue) by means of segmentation. Then a dynamic X-ray perfusion sequence is acquired from a fixed angular view. For quantitative analysis, each projected ray is normalized by the effective length of contrast material along the direction of the ray. For the computation of the normalization factors, the perfused areas are segmented using a gaussian mixture model. The method was evaluated in a simulation study using a CT perfusion dataset of the lower leg of a patient affected by arterial occlusions. Results have shown that quantitative perfusion assessment and also detection of hypo-perfused areas with this method are possible.
This work received financial support of the European Community under a Marie-Curie Host Fellowship for Early Stage Researchers Training, MEST-CT-2005-020424. M08-4:

Empirical Beam Hardening Correction for CT

Y. Kyriakou, D. Prell, W. A. Kalender, M. Kachelriess University of Erlangen, Institute of Medical Physics, Erlangen, Germany Due to x-ray beam polychromaticity and scattered radiation attenuation tends to be underestimated and cupping and beam hardening artifacts become apparent in the reconstructed CT images. In order to correct for higher order effects due to a mixture of materials of different energy dependence we propose a strategy that does not require any knowledge about spectra or attenuation coefficients. The empirical beam hardening correction (EBHC) segments materials other than water using a simple thresholding of the volume. Then a forward projection is performed. The resulting sinogram entries are squared and reconstructed to yield a second volume. Both volumes are then linearly combined with a combination weight determined to maximize the flatness of the new and corrected volume. EBHC was evaluated using phantom and patient data acquired with a modern conebeam spiral CT and a C-arm CT scanner. Although no physics apart from the initial segmentation threshold enter the correction process dark streaks due to beam hardening where largely removed. EBHC appears to be a superior alternative to iterative beam hardening correction algorithms and does not tend to over- or undercorrect the data. Unlike conventional iterative methods, it further requires no assumptions on the spectra, detector response or the type of material and therefore can be applied to any CT image without a priori information.

333

M08-5: Stationary-gantry Tomosynthesis System for Online Image Guidance in Radiation Therapy Based on a 52-Source Cold Cathode X-Ray Tube

J. S. Maltz1, F. Sprenger2, J. Fuerst3, A. Paidi1, F. Fadler3, A. R. Bani-Hashemi1 1 Oncology Care Systems Group, Siemens Medical Solutions USA, Inc., Concord, CA, USA 2 XinRay Systems, LLC., Research Triangle Park, NC, USA 3 Components and Vacuum Technology, Siemens AG Healthcare, Erlangen, Germany

We present the design, simulation and first images of an imaging system that employs a compact multiple source x-ray tube to produce a tomosynthesis image from a set of projections obtained at a single tube position. The electron sources within the tube are realized using cold cathode carbon nanotube technology. The primary intended application is tomosynthesis-based 3D image guidance during external beam radiation therapy. The tube, which is attached to the gantry of a medical linear accelerator (linac) immediately below the multileaf collimator, operates within the voltage range of 80 to 160 kVp and contains a total of 52 sources that are arranged in a rectilinear array. This configuration allows for the acquisition of tomographic projections from multiple angles without any need to rotate the linac gantry. The x-ray images are captured by the same amorphous silicon flat panel detector employed for portal imaging on contemporary linacs. The field-of-view (FOV) of the system corresponds to that part of the volume that is sampled by rays from all sources. The present tube and detector configuration provides an 8 cm x 8 cm FOV in the plane of the linac isocenter when the 40.96 cm x 40.96 cm imaging detector is placed 40 cm from the isocenter. Since this tomosynthesis application utilizes the extremities of the detector to record image detail relating to structures near the isocenter, simultaneous treatment and imaging is possible for most clinical cases, where the treated target is a small region close to the linac isocenter. The tomosynthesis images are reconstructed using the simultaneous iterative reconstruction technique (SART), which is accelerated using a graphics processing unit (GPU). We present details of the system design, simulations of imaging performance, as well as reconstructed tomosynthesis images of phantoms imaged using the prototype.
M08-6:

A Novel CT Perfusion Protocol for Quantitative Contrast Material Mapping

B. J. Brendel, E. Roessl, J.-P. Schlomka, A. Thran, R. Proksa Sector Medical Imaging Systems, Philips Research Europe - Hamburg, Hamburg, Germany In conventional CT assisted perfusion imaging (perfusion CT, PCT) the quantitative assessment of diagnostically important parameters (e.g. cerebral blood flow (CBF) and cerebral blood volume (CBV) for brain perfusion) is based on the CT numbers in the acquired images. These CT numbers are distorted by a number of artifacts, especially by beam hardening. The distortion due to beam hardening can be avoided by acquiring the first images immediately after injection of the contrast agent with a dual kVscan, and decompose the projections into base component line integrals representing Compton scattering and photo effect. These base component line integrals can be used to estimate the contrast material line integrals from the projections of the following CT scans, done in a conventional manner. The contrast material line integrals can then be reconstructed conventionally to get a quantitatively correct iodine map, which can be used to calculate perfusion parameters more accurately. Based on simulated data, we show that this technique indeed has the potential to completely eliminate errors in the estimation of the contrast material concentration due to beam hardening.
M08-7: The Price of Tomography: SNR Comparisons of Acquisition Strategies for X-Ray Fluorescence Imaging

P. J. La Riviere1, L.-J. Meng2 1 Dept. of Radiology, The University of Chicago, Chicago, IL, USA 2 Dept. of Nuclear and Radiological Engineering, The University of Illinois, Urbana-Champaign, Champaign, IL, USA

We have been working to develop and implement faster synchrotron-based X-ray fluorescence computed tomography (XFCT) methods for the mapping of trace metals in biological samples. As practiced now, the XFCT image acquisition process is slow (on the order of 1 or more hours per slice), which severely limits the ability to perform 3D imaging and to image multiple samples for sake of comparison and improved experimental statistics. This is because the data needed for tomographic reconstruction are acquired line-by-line in a first-generation tomographic geometry. Recently, we have been demonstrated the feasibility of using emission tomography systems for synchrotron X-ray fluorescence computer tomography. The proposed detection system combines high-resolution semiconductor detectors with multiple-pinhole apertures. The most promising acquisition mode involves using a sheet-collimated beam to illuminate a single plane through the object, which can then be directly imaged through the pinholes without need for tomographic reconstruction. Monte Carlo and experimental studies demonstrated the potential for order-of-magnitude improvements in SNR or imaging speed using this strategy. Some have found this surprising since the use of a pinhole collimator greatly reduces the detection efficiency of the system. In the present paper, we derive analytic figures of merit based on the ideal observer signal to noise ratio for the two geometries. These FOMs provide simple analytic insight into the reason for the improvement obtained with the pinhole-based system: the low

334

detection efficiency is more than compensated for by avoiding the ill-posedness of tomographic reconstruction. The derived FOMs also provide a basis for performing system optimization.
M08-8:

W. C. Barber1, E. Nygard1,2, J. S. Iwanczyk1, T. J. Beck3, J. C. Wessel1,2, N. Malakhov1,2, G. Wawrzyniak1,2, N. E. Hartsough1, T. Gandhi1, K. Taguchi3 1 DxRay Inc., Northridge, CA, USA 2 Interon ASA, Asker, Norway 3 Johns Hopkins University, Baltimroe, MA, USA We have developed the fastest photon counting x-ray imaging arrays to date with maximum output rates above 6 million counts per second per pixel. In order to provide the arrays with adequate performance for clinical dual-energy applications, we have designed and fabricated a fast ASIC for data readout from pixellated direct-conversion CdTe sensors. The ASICs and CdTe sensors that comprise the photon counting detectors are vertically integrated so that the readout is contained within the active area of the pixels allowing for the fabrication of large area arrays. Previously we have tiled pixels optimized for CT into a full 32-row multi-slice CT system used to generate the first photon counting clinical CT patent images. The pixels optimized for CT have demonstrated a maximum output count rate of 6 million counts per second per pixel with an energy resolution of 12% at 60 keV and a noise floor of 20 keV. Recently we have developed novel CdTe pixels which significantly reduce the noise floor to 10 keV allowing for lower energy imaging applications. We report on a characterization study of a prototype large area detector used for the generation of two and three dimensional DEXA images. The prototype CdTe array has 256 pixels with about a 1 mm pitch. We have measured several important detector parameters including the maximum output count rate, energy resolution, and noise performance. Additionally, the relationship between the output and input rate has been found to fit a non-paralyzable detector model with a dead time of 160 nsec. A maximum output rate of 3.5 million counts per second per pixel has been obtained with a low noise floor of 10 keV and an energy resolution of 10% at 60 keV. All detector noise counts are less that 10 keV which is sufficiently low for clinical DEXA. In conclusion, our results demonstrate fast output count rates with good energy resolution and an extremely low noise floor for the application of photon counting arrays to clinical DEXA systems.
The authors would like to thank Peter Weilhammer, Danielle Moraes, Pierre Jarron, and Jan Kaplon of CERN for their assistance in the development of the ASIC. M09: MIC Posters 2 Friday, Oct. 30 M09-2: 10:30-12:30 Grand Ballroom 4&5; Palm 3,4&5

Large Area Photon Counting X-Ray Imaging Arrays for Clinical Dual-Energy Applications

E. Y. Sidky1, M. Levine1, X. Pan1, I. S. Reiser1, R. M. Nishikawa1, R. H. Moore2, D. B. Kopans2 1 Radiology, University of Chicago, Chicago, IL, USA 2 Massachusetts General Hospital, Boston, MA, USA

Adaptive Image-Reconstruction Algorithms Applied to Digital Breast Tomosynthesis

Recent interest in "compressive sensing" has renewed interest in image-reconstruction algorithms that employ total variation (TV) in their corresponding optimization problem. Much of the challenge of utilizing the ideas of ``compressive sensing'' has come in developing efficient algorithms for the optimization. These algorithms generally have an adaptive step-size control to balance the data fidelity with the TV-norm. One such algorithm that we have been developing uses a combination of Steepest Descent (SD) and Projection onto Convex Sets (POCS). As the algorithm involves adaptive control, it is called adaptive SDPOCS or ASD-POCS. In this work, we investigate ASD-POCS with a class of image norms, and explore their impact on microcalcification detection in digital breast tomosynthesis. Comparisons with other algorithms such as expectationmaximization (EM) and conjugate-gradient (CG) algorithms will be presented at the meeting.
M09-5: Performance of a Prototype Semiconductor Brain PET Scanner Featuring High Energy Resolution with MAP Based Reconstruction

K. Matsuzaki1, W. Takeuchi1, Y. Morimoto1, T. Ishitsu1, Y. Ueno1, K. Kobashi1, N. Kubo2, C. Katoh2, T. Shiga3, N. Tamaki3 1 Central Reseach Lab., Hitachi, Ltd., Tokyo, Japan 2 School of Medicine, Hokkaido University, Sapporo, Japan 3 Graduate School of Medicine, Hokkaido University, Sapporo, Japan A prototype semiconductor brain PET scanner has high energy resolution feature of CdTe detectors (4.1% FWHM at 511 keV). Although energy resolution is much higher than conventional PET scanners, it is not avoidable to decrease sensitivity. To confirm potential of the narrower energy window setting, physical performance of CdTe PET with 490-530 keV energy window was measured by phantom studies. And the image quality was evaluated with using MAP (maximum a posteriori) based iterative 335

reconstruction method. Physical performances were measured with NEMA NU2 1994 standards. Image quality was evaluated in hot sphere phantom, which has 8, 10, 20, 30, 40, and 50 mm diameter hot spheres and hot by cold ratio was 8:1 (38.4 kBq/ml in hot). The image was measured for 30 min. and reconstructed by FBP and MAP based reconstruction method (hereinafter, called DP). To evaluate image quality, relative recovery coefficient (RC) and coefficient of variance (CV, calculated as SD divided by average) were measured in each hot sphere. All experiments were conducted with 390-530 keV as standard energy window (SEW) and 490-530 keV as photopeak energy window (PEW) setting. In PEW, images were reconstructed without scatter correction. As results, NECR (2R) at 0.2Ci/cc are 25kcps in PEW and 45kcps in SEW. The scatter fraction are 12% in PEW and 39% in SEW. RC of 20 mm diameter are 0.87 in PEW-DP, 0.86 in SEW-DP, 0.81 in PEW-FBP, and 0.80 in SEW-FBP. CV of 20 mm diameter are 0.055 in PEW-DP, 0.053 in SEW-DP, 0.17 in PEW-FBP, and 0.13 in SEW-FBP. The performance of CdTe PET with photopeak energy window was evaluated in several phantom studies. Although NECR in PEW decrease to about half times of SEW, RC in PEW is better than one in SEW, and CV is almost same between PEW and SEW, when we use MAP based reconstruction. That means, the CdTe PET scanner with narrower energy window setting is feasible for clinical use with high recovery coefficient and it has good potential for quantitative imaging.
M09-8:

K. J. Hong1, Y. Choi1, J. H. Kang1, W. Hu1, J. H. Jung1, B. J. Min1, S. H. Shin1, Y. S. Huh1, H. K. Lim1, Y. H. Chung2, P. Hughes3, C. Jackson3 1 Department of Nuclear Medicine, Sungkyunkwan University, Samsung Medical Center, Seoul, South Korea 2 Department of Radiological Science, College of Health Science, Yonsei University, Wonju, South Korea 3 SensL, Cork, Ireland Geiger-mode avalanche photodiode (GAPD) has been demonstrated to be a high performance PET sensor because of high gain, fast response, low excess noise and magnetic field insensitivity. The purpose of this study is to develop a PET for human brain imaging using 4x4 array of large size GAPD. PET detector modules were designed and built to develop a prototype PET. The PET consisted of 8 pairs of LYSO-GAPD block detectors arraged in a partial ring, covering arc of 80, with an inner diameter of 330 mm. The LYSO arrays consisted of 4x4 array of 3x3x20 pixels, which were 1-to-1 coupled to a 4x4 array of 9 GAPD pixels (SensL, Ireland). The GAPD arrays were fabricated to maintain the the variations of 511 keV photopeak position of the 16 pixels within 20%. The signals of the each module were amplified by a 16-ch preamp with differential outputs and then sent to a position decoder circuit (PDC), which readouts digital address and analog pulse of the one interacted channel from 64 signals of 4 preamplifier boards. The PDC output signals were fed into FPGA-embedded DAQ boards. The analog signal was sampled with 100 MHz, and arrival time and energy of the digitized signal were calculated and stored. PET imaging was performed by rotating an object with a step-and-shoot acquisition with 60 projections over 180. The coincidence data were sorted and each projection was normalized and reconstructed by OSEM. The average energy and time resolution of 16 LYSO-GAPD block detectors for 511 keV was 20% and 2.2 ns, respectively. Activity distribution patterns of hot&cold-rod phantoms were well imaged without distrortions, and rods down to a diameter of 3.2 mm were resolved. Currently, a full-ring PET system consisting of 72 detector modules having 12.9 mm axial and 330 mm transaxial FOV is being developed. These results demonstrate that high performance PET could be developed using the GAPD-based PET detectors, analog and digital signal processing methods designed in this work.
This study was supported by a grant of the Mid-Term Industrial Technology Development Program, the Ministry of Knowledge Economy (10024198), by a grant of the Industrial Source Technology Development Programs, the Ministry of Knowledge Economy (10030029), and by a grant of the Radiation Technology Development Program through the Korea Science and Engineering Foundation funded by the Ministry of Education, Science and Technology (2007-00321), Republic of Korea. M09-11: An Ultrahigh Resolution SPECT System Based on a Novel Energy-Resolved Photon-Counting CdTe Detector

Development of PET Using 4x4 Array of Large Size Geiger-Mode Avalanche Photodiodes

L.-J. Meng, J. W. Tan Nuclear, Plasma and Radiological Engineering, University of Illinois at Urbana-Champaign, Urbana, USA In this paper, we present the development and the imaging performance of an ultrahigh resolution SPECT system that is based on a novel energy-resolved photon counting (ERPC) CdTe sensor. This system consists of four ERPC detectors installed on a rotating gantry. The object supported by a horizontal bed remains stationary during an imaging study, whilst the cameras are rotated around the object to obtain a sufficient angular sampling. Compared to the I-EMCCD detector used in a single photon emission microscope (SPEM) system that we previously developed for mouse brain studies, the ERPC detector has a slightly sacrificed spatial resolution (350 um versus 160 um), but offers a dramatically improved energy resolution (4keV versus 40keV). This ensures an excellent rejection of scattered gamma rays and allows simultaneous use of multiple tracers labeled with different radioisotope. The ERPC detector has a very small dead area (at least on two opposite sides), which allows close packing of multiple detectors and therefore results in an improved angular sampling. A prototype SPECT system with a single ERPC detector was previously evaluated for SPECT phantom imaging, which has demonstrated a nice imaging performance. In this effort, we will fully populate the system with four ERPC camera heads. The performance of this new SPECT system will be 336

evaluated in a side-to-side comparison to the SPEM system. This would be help to assess the relative importance of intrinsic imaging resolution versus energy resolution and therefore provide guidelines for the development of future SPECT instrumentation. The results from this study will be presented.
M09-14:

B. Singh1, B. C. Stack2, S. Thacker1, V. Gaysinskiy1, S. Cool1, G. Entine1, V. V. Nagarkar1 1 RMD, Inc., Watertown, MA, USA 2 University of Arkansas for Medical Sciences, Little Rock, AR, USA

In Vivo Imaging of Lingual Cancer in a Rabbit Model Using a Hand-Held Imaging Beta Probe

Advances in radiopharmaceuticals and clinical understanding have escalated the use of intra-operative gamma probes in surgery, providing benefits that include increased specificity in tissue obtained for biopsy, minimally accessed incisions, and the reduction of inpatient hospital utilization with improved patient recovery. At present, most probes on the market are non-imaging gamma probes that suffer from the lack of ancillary information of the surveyed area, such as the clear delineation of malignant tissue. Also, the highly penetrating gamma radiation present locally or even in remote parts of the body increases the background radiation level and limits the practical use of these probes. To address these limitations, we have developed a novel, hand-held digital imaging probe to be used in surgery or in the clinic in conjunction with beta emitting radiopharmaceuticals such as FDG (F18), I131 and P32. This imaging beta probe (IBP) allows the real-time imaging of a surveyed area with higher spatial resolution and sensitivity and greater convenience than existing instruments. The hand-held IBP consists of a thin microstructured CsI:Tl film optically coupled to a highly sensitive electronmultiplying charged coupled device (EMCCD) via a flexible fiberoptic conduit. We report the results of our experiments using this hand-held IBP for rapid and accurate identification and delineation of lesions in a rabbits tongue. VX cancer cells were injected into the tongues of 8 New Zealand white rabbits, and positron emission tomography (PET) scans were performed after 7 and 12 days post-injection to determine the uptake of radioisotope in the tongue and the lymph nodes. Beta imaging scans using the IBP were performed on the tongue and the lymph nodes in-vivo and ex-vivo after 12 days post-tumor injection. The results of these experiments and the clinical usefulness of the IBP are discussed in the paper.
This work was partially supported by NIH grant number 2 R44 CA096030-03A1. M09-17:

Evaluation of Noise Properties in PSF-Based PET Image Reconstruction

S. Tong, A. Alessio, P. Kinahan Dept. of Radiology, University of Washington, Seattle, WA, USA The addition of point spread functions (PSF) in PET system modeling produces images with unique noise properties, which have not been fully studied. This work offers a systematic evaluation of noise and signal properties in different combinations of reconstruction methods and parameters. We evaluate 2 algorithms: OSEM+LOR (OSEM with exact line of response modeled) and OSEM+LOR+PSF (OSEM+LOR with a measured PSF included), in combination with the effects of 4 post filtering parameters and 1-10 iterations. We used a modified NEMA IQ phantom, which was filled with 68Ge and consisted of 6 hot spheres of different sizes with a target/background ratio of 4:1. The phantom was scanned 50 times in 3D mode on a clinical system to provide independent noise realizations. Data were reconstructed with OSEM+LOR and OSEM+LOR+PSF using different parameters. With access to multiple realizations, 3 metrics are adopted to quantify the noise properties in reconstructed images. Ensemble noise is calculated across independent realizations, and represents a more accurate estimate of true noise in the image. Pixel to pixel variability and NEMA style noise are defined on a single realization, and these measures are more commonly used in practice for simplicity. We also evaluate signal to noise performance with accepted measures (recovery coefficient, NPW filter), and study the relations between different metrics. From the analysis, a linear correlation is observed between NEMA noise and ensemble noise for all different combinations of reconstruction methods and parameters, suggesting that NEMA style noise is a reasonable surrogate for ensemble noise when multiple realizations are not available in practice. The addition of PSF reduces pixel to pixel variability for unfiltered images by roughly 35%, while it does not reduce NEMA or ensemble noise. When noise is measured across realizations, OSEM+LOR+PSF offers slightly improved (~7%) signal to noise performance across a range of parameters.
This work is supported by a grant from GE Healthcare and NIH grant CA74135.

337

M09-20:

A. Pourmoghaddas1,2, R. G. Wells2 1 Physics Department, MSc candidate - Carleton University, Ottawa, Canada 2 Cardiac Imaging, The University of Ottawa Heart Institutue, Ottawa, Canada

Motion Compensation Requires Increased Attenuation-Correction Accuracy in Cardiac PET/CT Imaging

The use of single helical CT scans is a commonplace for attenuation correction of PET images. However, single helical CT scans only capture a short portion of the respiratory cycle while the PET image corresponds to an average over multiple respiratory cycles. In this work, we study the combined effect of respiratory gate alignment and attenuation correction. The NCAT phantom was used in conjunction with SimSET to generate a 17-frame Rb-82 respiratory-gated dynamic acquisition. Each frame consisted of 10 respiratory phases, where a 5s respiratory cycle was assumed and the translation of the heart was set to be 2 cm in the SupInf direction. Sinograms were reconstructed with attenuation correction based on 3 different transmission maps: intensitymaximum, temporal-averaged and respiratory phase-matched. Algorithms developed in-house to detect and fit the heart for kinetic analysis were modified to fit an ellipsoid to the myocardial wall activity for each phase. Image volumes of each phase were translated to align the centers of the fitted ellipsoids. Motion-blurred images were created by averaging the images for each respiratory phase without alignment. Motion-free image sets, were created by selecting a single respiratory gate for each dynamic frame. Regional mean percentage errors (rMPE) were calculated for tracer uptake and blood flow. The STD of rMPE for each polar map was used as a measure of image uniformity. Without MC, AC with the max-intensity map shows improved uniformity, and rMPE of < 21% in blood flow. AC with phasematched and average CT had similar distribution and both have rMPE < 36%. MC reduces the rMPE in all cases, but with the greatest improvement coming for phase-matched AC. With MC, the mean rMPE is 4% for phasematched, 16% for average CT and 9% for max-intensity AC. The uniformity of the images also increases with MC to: 1% with phase-matched, 4.5% with averaged, and 4.2% with max-intensity AC.
M09-23:

K. Mueller1, F. Xu2 1 Computer Science, Stony Brook University, Stony Brook, NY, United States 2 Siemens Corporate Research, Princeton, NJ, United States

Optimal Sampling Lattices for High-Fidelity CT Reconstruction

We show that the familiar Cartesian lattice, while convenient for signal processing and representation, is sub-optimal when it comes to signal fidelity. We explore various applications of optimal sampling lattices, such as the Hexagonal and Body Centered Cartesian (BCC) lattices for 3D Computed Tomographic (CT) Reconstruction, both in terms of the (2D) detector and the (3D) reconstructed object. We find that BCC lattices compare favorably with CC lattices for both CT data acquisition and reconstruction. For example, they increase the recovery and detectability of small features, such as small tumors in the brain.
Supported by NIH grant R21 EB004099-01 M09-26:

D. Sowards-Emmerd1, K. Balakrishnan2, J. Wiener1, L. Shao1, J. Ye1 1 Nuclear Medicine/Imaging Physics and System Analysis, Philips Healthcare, San Jose, CA, USA 2 Nuclear Medicine/Imaging Physics and System Analysis, Philips Healthcare, Cleveland, OH, USA

CBCT-Subsystem Performance of the Multi-Modality Brightview XCT System

As the clinical needs are better understood in the recent years, it is apparent that a cost-effective SPECT/CT is needed for most nuclear medicine clinical applications. Philips Nuclear Medicine recently developed a flat panel based SPECT/CT system called Brightview XCT [SNM 09 Sowards-Emmerd]. The BrightView-XCT system utilizes a flat-panel X-ray detector to perform conebeam computed tomography (CBCT) localization and attenuation correction to supplement SPECT imaging. A unique off-set geometry is used to maximize the CT FOV while minimizing gantry footprint. The XCT images are reconstructed with FDK through a GPU backprojection implementation to speed-up the processing time with typical time of 1 min for each XCT rotation. An advanced scatter correction and optimized Anti-Scatter Grid reduce the effects scatter inherent in CBCT. In addition, filtering and profile matching methods are used to further improve the image quality. The purpose of this paper is to investigate the initial CT performance from the system. The Catphan and Virtually Human Pelvis phantom were scanned at doses ranging from lowdose attenuation correction to high-resolution bone protocols. With the bone protocol, high-contrast resolution was measured at >15 lp/cm at 10% MTF, uniformity as better than +/- 20 HU, and the 5 mm, 1% contrast insert was discernable at 5 mm slice thickness. For a typical body localization protocol, soft tissue boundaries were clearly delineated in the pelvis phantom. Finally, clinical patient data was acquired. In summary, the images produced can be successfully used for localization and open opportunites for other applications.

338

M09-29:

T. Koike1, S. Uno2, T. Murakami2, M. Sekimoto2, T. Uchida2 1 Tokyo University of Science, noda,Chiba, Japan 2 High Energy Accelerator Research Organization(KEK), Tsukuba,Ibaraki, Japan

Development of New Type Gaseous Gamma Camera with GEM

We are developing a new gaseous detector with Gas Electron Multiplier (GEM) as a gamma camera for biomolecular analysis and medical applications. The conventional gamma cameras using NaI(Tl)/PMT scintillation detector have inherent limitations in spatial resolution, sensitivity, and detector system is very expensive. Gamma cameras using GEM have been shown to offer potential improvements in basic performance as compared with conventional gamma cameras at spatial resolution and system costs. In our camera, a solid converter (gold) is used for the detection of gamma-ray. We have constructed a small field of view prototype gamma camera with 10cm x 10cm GEM foils. Our system consists of the gamma conversion layers and the amplification layers. The GEM foils of conversion layer are coated by gold with 3um thickness on the both surfaces. A GEM technology is a well consolidated way to amplify a few electrons to obtain a detectable signal. In order to obtain higher detection efficiency, and several gold-coated GEM foils are stacked. Gold-coated GEM foils were manufactured and a simple test was carried out to confirm the principle. In order to know the performance of the prototype gamma camera, gamma-ray irradiation test was performed with radioactive source of 99m-Tc (140keV). 2D images were obtained using a pinhole collimator. More tests will be done before the conference.
M09-32:

N. Inadama1, H. Murayama1, F. Nishikido1, T. Yamaya1, E. Yoshida1, K. Shibuya2, Y. Yazaki3,1, H. Osada3,1 1 Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan 2 Graduate School of Arts and Sciences, Tokyo University, Tokyo, Japan 3 Graduate School of Sciences, Chiba University, Chiba, Japan

Proposal of a 8-Layer DOI Detector Composed of Same Scintillation Crystal Elements

Previously, we developed a 4-layer depth of interaction (DOI) detector. It is composed of four layers of a scintillation crystal array and a position sensitive photomultiplier tube (PS-PMT) and by removing some reflectors between the crystal elements in the array, we control scintillation light distribution on the PS-PMT so that the responses of all crystal elements are discriminated. We also proposed a 2-layer DOI encoding method by using the crystal elements cut as a triangular prism and proved its capability. Combining these two methods, here we propose an 8-layer DOI encoding method. The method makes it possible that the 8-layer DOI detector is consisted of same scintillation crystal elements, that is to say, any scintillation crystals. The method was demonstrated with the Lu2xGd2(1-x)SiO5 (LGSO) crystals. The obtained results indicated the possibility of the proposed method.
M09-35:

Evaluation of Accuracy and Precision of Geometric Mean Based Activity Estimation Methods

B. He, E. C. Frey Department of Radiology, Johns Hopkins Medical Institutions, Baltimore, MD, USA Accurate estimation of organ activity is essential for targeted radionuclide therapy. The traditional method uses geometric mean (GM) methodology to estimate activity from whole body scans. However, many physical effects limit its quantitative accuracy. A variety of compensations have been proposed. The goal of this work was to systematically evaluate the effects of these compensations on the accuracy and precision using a simulated phantom population. For comparison, we also applied the previously-developed quantitative planar (QPlanar) and SPECT (QSPECT) methods. A realistic population of NCAT phantoms (N=49) was used with activity distributions that modeled clinically observed In-111 ibritumomab tiuxetan distributions and anatomies of 7 patients. The projections were simulated using parameters appropriate for a GE VH/Hawkeye SPECT/CT system. GM methods with various combinations of compensations for scatter, body thickness, overlap/background activity, and volume were evaluated and compared to the QPlanar and QSPECT methods. The results showed that the GM methods provided more accurate estimates for larger organs than smaller ones. The accuracy generally improved when more compensations were applied. However, these methods were still significantly worse than the QPlanar and QSPECT methods. Due to the subjectivity involved in organ ROI definition and patient dependence of the validity of the underlying approximations in the compensation methods, the precision for the GM methods was much worse than with the QPlanar and QSPECT methods. We conclude that GM-based methods can provide relatively accurate activity estimates for large, high-uptake organs when appropriate compensations are applied. However, they were generally relatively inaccurate and imprecise for small organs. The QPlanar and QSPECT methods provided much more accurate and precise estimates for all organ sizes, but required longer computation time and definition of organ regions in 3D.

339

M09-38:

Evaluation of List Mode-Driven Respiratory and Cardiac Gating in PET

F. Buether, M. Dawood, O. Schober, K. P. Schaefers European Institute of Molecular Imaging, University of Muenster, Muenster, Germany Different intrinsic, list mode-based methods of cardiac and respiratory gating in PET that do not require additional hardware are investigated in a cardiac FDG patient study comprising 29 patients and compared to traditional gating methods based on video monitoring and ECG for respiratory and cardiac gating, respectively. One of the intrinsic methods, the sensitivity method, is based on the fact that the geometric sensitivity of a 3D PET scanner is inhomogeneous, therefore a moving positron source inside the scanner's field of view should result in varying counting rates. Another approach is the center of mass method, in which the axial center of mass and the standard deviation of the measured coincidences in a small time frame is determined. Motion should be detectable by changes in the center of mass/standard deviation. We demonstrate that these methods lead to well resolved PET images in the case of respiratory gating. List mode-based cardiac gating is shown to be successful in the case of high FDG uptake in the heart.
M09-41:

C. Tsoumpas1, C. Buerger1, A. P. King1, V. Keereman2, S. Vandenberghe2, V. Schulz3, T. Schaeffter1, P. K. Marsden1 1 Div. of Imaging Sciences, King's College London, London, UK 2 IBBT, University of Ghent, Ghent, Belgium 3 Research Laboratories, Philips, Aachen, Germany We have implemented a scheme for simulating realistic dynamic PET data from real MR acquisitions. This toolkit uses a series of MR acquisitions, image registrations and segmentations. PET images are simulated assigning typical values to the segmented images, and potentially manually inserting additional lesions. The data are simulated using analytic forward-projections (including attenuation) with STIR, providing a fast and simple simulation toolkit that can be used to efficiently study the performance of different reconstruction and correction approaches for dynamic PET data.
This project is supported by the European Union under the seventh framework program (201651). M09-44:

Simulation of Dynamic PET Data from Real MR Acquisitions

Kinetic Parameter Estimation for a Heterogeneous Tumor Model

Y. Lin, Q. Li, R. Leahy Signal and Image Processing Institute, USC, Los Angeles, CA, United States We investigate a heterogeneous kinetic model of tracer uptake in PET for the purposes of improved characterization of small or inhomogeneous tumors. Using a mixture model we allow each voxel to be a combination of two kinetic processes representing a tumor and a normal tissue time activity curve. We then use a novel nonlinear least squares procedure to estimate the kinetic parameters of these processes from a region of interest consisting of multiple mixture voxels with unknown fractions of these two processes. Through investigation of the Cramer Rao lower bound (CRLB) we determine conditions under which the two processes can be accurately identified. We also perform Monte Carlo simulations to show that the CRLB's are good predictors of the variance of the rate parameters estimated using nonlinear least squares.
M09-47:

Limits of Ultra-Low Dose CT Attenuation Correction for PET/CT

T. Xia, P. E. Kinahan Department of Radiology, University of Washington, Seattle, WA, USA We present an analysis of the effects of ultra-low dose X-ray computerized tomography (CT) based attenuation correction for positron emission tomography (PET). By ultra low dose we mean less than approximately 5 mAs or 0.5 mSv total effective whole body dose. The motivation is the increased interest in using respiratory motion information acquired during the CT scan for both phase-matched CT-based attenuation correction and for motion estimation. Since longer duration CT scans are desired, radiation dose to the patient can be a limiting factor. In this study we evaluate the impact of reducing photon flux rates in the CT data on the reconstructed PET image by using the CATSIM simulation tool for the CT component and the ASIM simulation tool for the PET component. The CT simulation includes effects of the x-ray tube spectra, beam conditioning, bowtie filter, detector noise, and bean hardening correction. The PET simulation includes the effect of attenuation and photon counting. Noise and bias in the PET image were evaluated from multiple realizations of test objects. We show that techniques can be used to reduce the mAs needed for CT based attenuation correction. The limiting factor, however, is not the noise in the CT image but rather the bias introduced by raw CT sinogram elements with no detected flux. These results constrain the methods that can be used to lower CT dose in a manner suitable for attenuation correction of PET data.
Supported by NIH Grant R01-CA115870

340

M09-50:

BetaBox: a PSAPD Based Charged Particle Camera with a USB Data Acquisition System

N. T. Vu, R. W. Silverman, M. E. Phelps, A. F. Chatziioannou Molecular and Medical Pharmacology, UCLA Crump Institute for Molecular Imaging, Los Angeles, CA, USA The BetaBox is a low cost and compact system for the detection and imaging of charged particles. It was designed for the quantification of radiolabeled probes at multiple locations in microfluidic platforms. Previous work has demonstrated the use of a position sensitive avalanche photodiode (PSAPD) camera for microfluidic based assays. The previous systems required the use of nuclear instrumentation modules for signal processing and a PCI data acquisition system which were both cumbersome and expensive. In order to reduce the cost and increase the portability of previous designs, we have developed a beta camera based on a low cost, off the shelf USB data acquisition module. The self-contained system uses compact printed circuit boards for pulse shaping and event triggering, contains all the necessary power supplies and is packaged inside a single enclosure with dimensions less than 26 x 26 x 11 cm3. The compact design allows the BetaBox to be used on a lab bench top and connected via a single USB connection to a standard PC for data acquisition and system control. Preliminary performance of the BetaBox for dynamic imaging of charged particles is reported in this work. Using a 204Tl source placed directly on top of the BetaBox the absolute sensitivity was measured to be 43%. Images and a calibrated energy spectrum were also acquired with a printed [18F]FDG grid source. With this grid, 1mm spaced points were easily discernable within the central region of the FOV. The BetaBox is also capable of acquiring dynamic images using timestamp information for each event. The relative timing between each of the events can be measured to an accuracy of 1 second.
M09-53:

A. Z. Kyme1,2, S. R. Meikle2,3, C. Baldock1, R. R. Fulton1,2,3,4 1 School of Physics, University of Sydney, Sydney, NSW 2006, Australia 2 Brain & Mind Research Institute, University of Sydney, Sydney, NSW 2006, Australia 3 Discipline of Medical Radiation Sciences, University of Sydney, Sydney, NSW 2006, Australia 4 Department of Medical Phsyics, Westmead Hosptial, Sydney, NSW 2145, Australia

Understanding and Compensating for Refraction Errors in Stereo-Optical Tracking During Small Animal PET / SPECT

Pre-clinical positron emission tomography (PET) and single-photon emission tomography (SPECT) are becoming increasingly important in understanding brain physiology using animal models. Restricting animal movement so that it remains within the scanner field-of-view typically is achieved using some form of barrier (eg. transparent tube). A potential limitation arises in such circumstances due to the interface between tracker and target. When refractive interfaces exist between the image plane and target point, the assumption of standard photogrammetric triangulation is violated. Failure to account for this is expected to lead to triangulation errors and inaccuracy in pose estimation of a tracked marker. However, to our knowledge, this issue of refraction error arising in motion tracking of small animals has not been investigated in detail in the literature in spite of such set-ups being used. We have used simulations to understand the dependencies of the triangulation error and obtain some preliminary error estimates for the magnitude and direction of errors expected in small animal scanner geometries. We have also demonstrated that even a simple interface with vector normal aligned with the optical axis of the tracker leads to pose estimates which can result in noticeable distortion when applied in motion correction. Triangulation errors derived from the three-dimensional (3D) target positions measured by the tracker followed the same trend as the simulations but were consistently greater. Based on our results, we propose extending previous 3D ray-tracing methods for refraction to include accounting for the changing pose of the interface relative to the tracked object. This would enable analytic compensation for refraction during real-time motion tracking of small animals. Having the flexibility to track moving animals through an interface offers potential advantages such as increased freedom of movement and reduced need for animal training.
M09-56:

K. Shimazoe1, K. Fujita2, H. Mori1, M. Ohno1, H. Takahashi1, T. Momose1, M. Fukuda3 1 The University of Tokyo, Tokyo, Japan 2 Japan Atomic Energy Agency, Ibaraki, Japan 3 HOYA Corporation, Tokyo, Japan Positron imaging for cancers in endoscope inspection is promising to detect overlooked small or unclear carcinomas since it enables double inspection of structure by optical image and function by radio image. There have been researches about positron probes with scintillation detector. We have investigated, designed and fabricated Si based positron detectors, which has an advantage in its high energy resolution and sensitivity. Simulation with EGS is executed to determine the optimal detector, detector thickness and available condition in human body. The optimal thickness of Si detectors is 1.0mm and the thickness of obstruction should be less than 0.6mm. We also fabricated a Si detector with diameter of 2.2mm and thickness of 1mm with a cable. Fabricated Si detectors is confirmed to successfully detect the emitted positrons. We are now in the stage to implement Si detector inside the endoscope.

Development of a Positron Probe for Multimodality Endoscopy

341

M09-59:

J. Dang1, P. Lasaygues2, S. Tavernier1, S. Mensah2 1 Vrije Universiteit Brussel, Brussels, Belgium 2 CNRS, Laboratory of Mechanics and Acoustics, Marseille, France

Development of Anthropomorphic Phantoms for Combined PET-Ultrasound Breast Imaging

As part of the development a combined PET-Ultrasound multimodal scanner for breast imaging by the Crystal Clear collaboration (The ClearPEM-Sonic project) we have developed and tested a phantom that can be used for making realistic images with both modalities. In the first step we measured the propagation velocities of the acoustic waves, the attenuation coefficient and the elasticity (Youngs modulus) for several series of different samples based on gelatine and agar mixtures. We determined which preparations reproduce the acoustical and elastic properties of different body tissues of interest in breast imaging such as fat tissue, glandular tissue, fibrous tissue and carcinomatous tissue. In the Second step we have built phantoms where we added a small amount of FDG during the preparation of the phantom such as to give the different parts in the phantom activities similar to what is usually present in the breast during PET imaging. The phantom was than imaged on a Philips Gemini TF PET/CT and on a US scanner from ATL, HDI 5000. The images were superposed using rigid transformations to produce combined PET/US images. We also evaluated the performance of our phantoms for US elastographic imaging. Details on the procedure for producing the phantoms will be given.
We thank Dr. A. Bossuyt and Dr. H. Everaert from the department of Nuclear Medicine and Dr. J. De Mey and Dr. C. Breuck from the department of Radiology of the University hospital of the Vrije Universiteit Brussel for allowing us access to their equipment. M09-62:

M. Ito1,2, J. S. Lee2,3,4,5, M.-J. Park2, K.-S. Sim1, S. J. Hong5,6 1 Department of Physics, Korea University, Seoul, Korea 2 Departments of Nuclear Medicine, Seoul National University, Seoul, Korea 3 Biomedical Sciences, Seoul National University, Seoul, Korea 4 Brain and Cognitive Sciences, Seoul National University, Seoul, Korea 5 Institute of Radiation Medicine, Seoul National University, Seoul, Korea 6 Department of Radiological Science, Eulji University, Gyeonggi-do, Korea

Effect of Crystal Size on the DOI Resolution in New DOI Encoding Method Using Light Spread Within Monolayer Pixelated Crystal Array with Single-Ended Readout

We have proposed a new depth-of-interaction (DOI) encoding method for a monolayer pixelated scintillation crystal array with a single-ended readout scheme. Simulation showed that the method provided the DOI information as a function of direction and extent of 2D light dispersion within the partially transparent crystal array, and estimated the excellent DOI resolution and crystal separation. In this work, Monte Carlo simulation was used to estimate the effect of crystal size for the DOI resolution. The results indicated that 1.5 x 1.5 x 20mm3 and 1.0 x 1.0 x 24 mm3 crystals are the optimal size to provide better DOI response with the consideration of the tradeoff between the DOI resolution and the detection efficiency. The DOI resolutions were 1.7 mm for 1.5 x 1.5 x 20 mm3 and 1.8 mm for 1.0 x 1.0 x 24 mm3. On the other hand, since the extent of light dispersion can be controlled by the index of the material between crystal and grid, the light spread method can apply to the crystal array of wide size range by the appropriate material between crystal and grid.
M09-65:

A. L. Goertzen1, D. B. Stout2, C. J. Thompson3 1 Department of Radiology, University of Manitoba, Winnipeg, Manitoba, Canada 2 Crump Institute for Molecular Imaging, University of California, Los Angeles, Los Angeles, California, USA 3 Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada

Validation of a New Method to Measure the Energy Spectrum of Coincidence Events in PET

PET system energy response is typically characterized in singles detection mode yet there are situations in which the energy spectrum of coincidence events might be different than the spectrum measured in singles mode. Examples might include imaging with isotopes that emit a prompt gamma in coincidence with positron emission, imaging with low activity in a LSO/LYSO based camera, in which the intrinsic activity is significant, and high scatter situations in which the two 511 keV photons have different scattering probabilities (i.e. off-centre line source). Being able to accurately measure the energy spectrum of coincidence events could be used for validating simulation models, optimizing energy discriminator levels, and examining scatter models and corrections. For a typical PET system operating in coincidence mode the only method available for estimating the energy spectrum is to step the lower and upper level discriminators (LLD and ULD). Simple measurement techniques such as using a narrow sliding energy window or stepping the LLD only will not yield a spectrum of coincidence events that is accurate for cases where there are different energy components contributing to the spectrum. In this work we present a measured validation of a new technique to measure the energy spectrum of events that contribute to coincidences in PET. The measurement technique uses 342

a linear combination of two sets of coincident count measurements, one made by stepping the LLD and one made by stepping the ULD. Measured data were acquired on a Siemens Inveon camera for sources of just the intrinsic 176Lu activity and for a 207 uCi 68Ge source. Monte Carlo simulation with GATE was used to generate simulated data equivalent to the measured data runs. The results show that our energy spectrum calculation method accurately recovers the beta/gamma spectrum of the 176Lu intrinsic activity present in the LSO scintillator crystals as well as accurately modeling the 68Ge coincidence energy spectrum.
M09-68:

M. A. N. Korevaar1,2, M. C. Goorden1,2, J. W. T. Heemskerk1,2, R. Kreuger2, F. J. Beekman1,2,3 1 Department of Nuclear Medicine, Image sciences institute, Utrecht, Netherlands 2 Department of R3, Radiation Detection and Medical Imaging, Delft, Netherlands 3 Molecular Imaging Labs (MILABS), Utrecht, Netherlands

Statistical versus Analytical Scintillation Detection Algorithms for CCD Based Gamma Cameras

Gamma cameras based on Charge Coupled Devices (CCDs) and micro-columnar CsI(Tl) scintillators can reach high spatial resolutions. However, the gamma interaction probability of these scintillators is low (typically <30% at 141 keV) due to the limited thickness of presently available micro-columnar scintillators. Continuous scintillators can strongly improve the interaction probability but suffer from increased light spread compared to columnar scintillators. In addition, for both types of scintillators gamma photons incident at an oblique angle reduce the spatial resolution due to the variable depth of interaction (DOI). To improve spatial resolution and spectral characteristics of these detectors, we have developed a statistical detection algorithm that uses a depth-dependent light spread model and an experimentally validated noise model. The algorithm uses a maximum likelihood (ML) estimator to determine the scintillation position and energy. Here we investigate i) the advantages of using a statistical method instead of an analytical method when both algorithms are based on the same photon transport model and ii) the influence of the choice of light spread model on the performance of the algorithm. The performance of the statistical algorithm is compared to that of a previously developed analytical algorithm, the multi-scale algorithm (Korevaar et al., 2009). The light spread models we use are a simple Gaussian distribution and an isotropic model that includes Fresnel reflections. The algorithms are compared for an Electron Multiplying CCD (EM-CCD) optically coupled to CsI(Tl) scintillators of different thicknesses. The statistical algorithm yields a 25% better energy resolution and for oblique incidence a 16% better DOI corrected spatial resolution than the best analytical algorithm. We conclude that the statistical algorithm significantly improves the performance compared to the analytical algorithm.
M09-71:

Analytic Derivation of Pinhole Collimator Sensitivity for General Source Distribution

Y. Li, J. Oldendick, C. E. Ordonez, W. Chang Department of Diagnostic Radiology, Rush University Medical Center, Chicago, IL 60612, USA Pinhole collimators are widely used for SPECT imaging of small organs and animals. There also has been renewed interest in using pinhole collimation for clinical cardiac SPECT imaging, which uses multiple pinhole arrays to achieve high sensitivity and complete data sampling. Overall sensitivity of a pinhole array is critical in determining a systems performance. Conventionally, a point source model has been used to evaluate the overall sensitivity and optimize the system design. This model is simple and ideal, but far from realistic. This work addresses the use of more realistic source models to assess the sensitivity performance of pinhole systems. The optimization of pinhole array designs using simulations that are based on ray-tracing or Monte Carlo methods is time-consuming. As an efficient alternative, we derived an analytical formula for the sensitivity of pinhole collimation with a general source distribution model using spherical harmonics. As special cases of this general model, we also provide the pinhole sensitivity formulas for line segment, disk and sphere sources. These results show that the point source model is just the zero-order approximation of the other source models. The point source overestimates or underestimates the sensitivity when the line segment or disk source lies in a plane that is parallel or perpendicular to the pinhole aperture plane, respectively. The sphere source yields the same sensitivity as a point source at the center of the sphere when attenuation is not a factor. The calculated sensitivities based on these formulas show good agreements with separate Monte Carlo simulations in simple cases. The general and special sensitivity formulas derived here are useful for design and optimization of SPECT systems that utilize pinhole collimator.
M09-74:

V. Fanti1,2,3 1 Dept. of Physics, University of Cagliari, Cagliari, Italy 2 INFN, Cagliari, Italy 3 CERN, Geneva, Switzerland On behalf of the AX-PET Collaboration

First Results from the AX-PET Demonstrator

The Axial-PET Demonstrator is based on a new concept in PET detectors, with LYSO crystals aligned along the z coordinate (patient's axis) and WLS (Wave-length shifter) strips placed orthogonal to them. This kind of structure permits to avoid parallax errors due to different depths of interaction of the photons in the crystals, to register the three coordinates of the impinging 343

photon and to reconstruct Compton events. In this way both the spatial resolution and the sensitivity can be highly improved. Moreover, as both the LYSO crystals and the strips are readout via Geiger mode APDs (Hamamatsu MPPC), the detector is insensitive to magnetic fields and is then suitable to be used in a combined PET/MRI apparatus. A complete Monte Carlo simulation of the Demonstrator has been developed based on Geant4 GATE software. Dedicated reconstruction software, needed due to the particular geometry arrangement, is under development. A pre-module with two LYSO-crystal rows (eight crystals each) and two WLS-strip rows (26 strips each) has been built and tested with a Na-22 source. The two final modules, each composed by 48 crystals (six rows) and 156 WLS strips (six rows) are being built. The results on the performances of the system obtained with a Na-22 point source (0.25 mm diameter) will be reported.
M09-77: Impact on the Spatial Resolution Performance of a Monolithic Crystal PET Detector Due to Different Sensor Parameters

X. Li1, C. Lockhart2, T. K. Lewellen2, R. S. Miyaoka2 1 Physics, University of Washington, Seattle, WA, USA 2 Radiology, University of Washington, Seattle, WA, USA

The performance characteristics of a monolithic crystal PET detector utilizing a novel sensor on the entrance surface (SES) design is reported. This design utilizes an array of micro-pixel avalanche photodiode (MAPD). MAPDs are one manufacturers version of Geiger-Muller mode APDs (GM-APD) that can provide signal gain similar to a photomultiplier tube (PMT). Since these devices are still under active development, their performance parameters are rapidly changing. Using a multi-step simulation process, we investigated how different MAPD parameters affect the performance of monolithic crystal PET detector. These parameters included gain variability between different channels; gain instability; and dark count noise. The detector simulated was a 49.6 mm by 49.6 mm by 15 mm LYSO crystal detector readout by a 16 by 16 array of 2.8 mm by 2.8 mm MAPD elements. To reduce the number of signal channels that need to be collected, the detector utilizes row-column summing. A statistics based positioning method is used for event positioning. Of the variables investigated, the dark count noise had the largest impact on the intrinsic spatial resolution. Gain differences of 5-10% between detector calibration and detector testing had a modest impact on the intrinsic spatial resolution performance and led to a slight bias in positioning. There was no measurable difference with a gain variability of up to 25% between the individual MAPD channels. Based upon these results we are planning to cool our detectors below room temperature to reduce dark count noise and to actively control the temperature of the MAPDs to reduce drifts in gain over time.
M09-80:

H. Kim1, C.-M. Kao1, C.-T. Chen1, H. Frisch2, F. Tang2, J.-F. Genat2, E. Oberla2, W. W. Moses3, W.-S. Choong3 1 Radiology, University of Chicago, Chicago, IL, US 2 Enrico Fermi Institue, University of Chicago, Chicago, IL, US 3 Lawrence Berkeley National Laboratory, Berkeley, CA, US

A Design of PET Detector Using Microchannel Plate PMT with Transmission Line Readout

A simulation study has been conducted for a PET detector design. The detector unit consists of 24x24 array of pixelated LSO, each pixel with 4x4x25$mm^3$, and two large area microchannel plate(MCP) PMT, 102x102$mm^2$, coupled to both side of scintillator. The crystal pitch was 4.25mm and reflective media was inserted between crystals. The optical photon inside scintillator was simulated using Geant4 package and the electrical signal of MCP was formed using the measured characteristics of MCP and Geant4 output. The signals from MCP were readout using the transmission line(TL) scheme. The readout scheme at both side of scintillator enable us to extract the depth of interaction in addition to energy and timing in an event. The detector response was measured by impinging the pair of 511keV gamma upon the detector. As preliminary results, we obtained ~12%(FWHM) of energy resolution at 511keV and ~375ps(FWHM) coincidence timing resolution while keeping 36% detection efficiency at 511keV. The position resolution was measured ~4.25mm, which is the pitch between LSO pixels. We found that the energy asymmetry and time difference from both side MCPs can be exploited for depth of interaction by observing strong correlation between them. The fast timing characteristics of MCP combined with the high sensitivity of LSO makes this design suitable for TOF PET application.

344

A. Studen1, D. Burdette2, E. Chesi2, V. Cindro1, N. H. Clinthorne3, E. Cochran2, B. Grosicar1, K. Honscheid2, H. Kagan2, C. Lacasta4, G. Llosa4, V. Linhar4, M. Mikuz1,5, V. Stankova4, P. Weilhammer2, D. Zontar1 1 Josef Stefan Institute, Ljubljana, Slovenia 2 Ohio State University, Columbus, OH, USA 3 University of Michigan, Ann Arbor, MI, USA 4 CSIC/IFIC, Valencia, Spain 5 University of Ljubljana, Ljubljana, Slovenia This paper reports the characterization of a detector module, the building block to be used for the MADEIRA PET probe prototype. The prototype will be used in synchronization with a conventional PET ring, amplifying the basic image with a subset of events with high spatial resolution. For image improvement, the crucial parameters are the spatial and timing resolution of the probe, while the energy resolution can be used in event classification. The prototype is made of high-resistivity silicon detectors, 1~mm thick with 1040 square pads with a size of 1~mm. The spatial resolution of the probe equals to the pad size. The pads are read out with VATAGP7, a Gamma Medica-Ideas designed application specific integrated circuit (ASIC). The ASIC provides a logic trigger signal and an analog output for every out of 128 input channels. A simplified module with 256 pads was used for initial tests. The measurement of the timing resolution was performed, using a positron source and a timing reference detector. A similar measurement was performed with a test pulse, ie. a charge injection into the ASIC, to study inherent resolution of the first-stage electronics. We calibrated the analog output of the ASIC using standard gamma sources (241-Am, 57-Co, 133-Ba), and we determined the overall energy resolution. The comparators were calibrated by a test pulse, looking at the rate of triggers versus threshold. The gain variation was compensated with internal 3-bit DACs. The spatial resolution of the probe equal to the pad size of 1~mm suffices the requirements posed by the PET application. The timing resolution is manageable and the energy resolution is excellent.
The work was (partially) carried out within the Collaborative Project "MADEIRA" (www.madeira-project.eu), cofunded by the European Commission through EURATOM Seventh Framework Programme (Grant Agreement FP7-212100). M09-86:

M09-83:

Performance of the MADEIRA PET Probe Prototype

B. J. Min1, Y. Choi1, N.-Y. Lee2, J. H. Jung1, K. J. Hong1, J. H. Kang1, W. Hu1, K. Lee3, Y. B. Ahn4, J. Joung5 1 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 2 School of Computer Aided Science, Institute of Basic Science, Inje University, Gimhae, Korea 3 Department of Radiologic Science, Korea University, Seoul, Korea 4 Department of Electronic Engineering, Konkuk University, Seoul, Korea 5 Molecular Imaging, Siemens Medical Solutions USA, Inc., Hoffman Estates, IL 60195, USA

A Miniature SPECT Using Multi-Pinhole Collimator with Vertical Septa

We had previously reported a simulation study on a novel multi-pinhole collimator (MP) which could provide improved angular sampling and enlarged imaging field of view (FOV) compared to low energy high resolution parallel-hole collimator (LEHR) using same size detector. The aim of this study was to develop a miniature SPECT to verify the proposed MP having lead vertical septa. A detector having 70 mm x 70 mm active area was consisted of a 6 mm thick NaI(Tl) crystal coupled to a 127 mm diameter position sensitive PMT. A 49-pinhole MP with 2 mm diameter pinhole and 40 mm focal length was fabricated to evaluate the performance compared to a typical LEHR. Additionally, a detector having 50 mm x 50 mm active area with 25pinhole MP and LEHR was investigated to evaluate enlarged imaging FOV. Planar spatial resolution, sensitivity, and resolution hot- and cold-rod phantom images were acquired. Images were reconstructed by the use of a dedicated MLEM algorithm with an unmatched projector/backprojector pair. Spatial resolutions and sensitivity were 7.6 mm FWHM and 0.25 cps/Ci at 100 mm distance, respectively when the 49-pinhole MP was used. The values were 6.5 mm FWHM and 0.25 cps/Ci with LEHR. Although the detector size was smaller than the phantom size, MP allowed to image entire phantom images while LEHR provided the images with truncation artifact. The reconstructed images using 60 and 30 projections with both 49-pinhole MP and LEHR showed a similar quality image to the image using 120 projections. However, the reconstructed images with 10 projections using 49-pinhole MP showed higher quality than those with LEHR. The reconstructed images using the MP provided high quality images with enlarged imaging FOV even when insufficient angular sampling data were used which would be useful to develop a stationary SPECT.

345

M09-89:

L. A. Eriksson1,2,3,4, M. Conti1, C. L. Melcher2, D. W. Townsend5, M. Eriksson3, M. E. Casey1, B. Bendriem1 1 Molecular Imaging, Siemens Healthcare, Knoxville, Tennessee, USA 2 Scintillation Materials Research Center, University of Tennessee, Knoxville, Tennessee, USA 3 Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden 4 Dept. of Physics, University of Stockholm, Stockholm, Sweden 5 Dept. of Medicine and radiology, University of Tennessee Medical Center, Knoxville, Tennessee, USA The main performance parameters in positron camera system design are sensitivity and spatial resolution. This paper concerns sensitivity, which is a function of the scintillation material, the solid angle subtended by the detector array, and the scintillator packing fraction. The solid angle can be increased by extending the axial extent of cylindrical detector systems. Commercial positron camera systems are usually based on rings of detector blocks with lutetium oxyorthosilicate, LSO:Ce, as the scintillator of choice. By adding more block detector rings, the solid angle increases while the detection efficiency remains fairly constant assuming the same crystal thickness. It has been shown that Ca co-doping of LSO:Ce reduces the scintillation decay time to ~ 30 ns with a light output over 30000 ph/MeV. This improvement may give a time-of-flight (TOF) advantage with time resolution of 500 to 600 ps. If we couple the count rate sensitivity of a large axial field-of-view (AFOV) system to the TOF sensitivity increase, we have the means to create examination times in the sub minute range with no compromise in image quality. In the present study we have compared the existing Siemens molecular CT (mCT) systems to future 6, 12 and 20 block ring systems with and without the TOF advantage. The mCT 4 block ring system has been used as a reference. The time for acceptable image quality with this system is then extrapolated to other systems based on planar sensitivity. However, the planar sensitivity is related to the solid angle, and reaches saturation for large AFOVs. This implies that there is an upper count rate sensitivity limit. A 20 block ring system may cover a 70 cm examination range at a certain planar count rate and could provide acceptable quality images in approximately 10 seconds by combining a high planar sensitivity count rate provided by the multi-ring feature, the high stopping power of LSO and the TOF gain due to the improved timing resolution.
M09-92:

Towards Sub Minute PET Examination Times

F. Garibaldi1, M. Baiocchi2, E. Cisbani1,2, F. Cusanno1, S. Colilli1,2, R. Fratoni1,2, F. Giuliani1,2, M. Gricia1,2, M. Lucentini1,2, M. L. Magliozzi1,2, G. Marano2, M. Musumeci2, F. Santavenere1,2, L. Vitelli2, G. De Vincentis3, S. Majewski4 1 INFN Roma gr. Coll. Sanita', Rome, Italy 2 Istituto Superiore di Sanita', Rome, Italy 3 University La Sapienza, Rome, Italy 4 University of West Virginia, Morgentown, USA SPECT systems using pinhole apertures permit radiolabeled molecular distributions to be imaged in vivo in small animals. Nevertheless studying cardiovascular diseases by means of small animal models is very challenging. Specifically, submillimeter spatial resolution, good energy resolution and high sensitivity are required. We designed what we consider the optimal radionuclide detector system for this task. It should allow studying both detection of unstable atherosclerotic plaques and monitoring the effect of therapy. The ideal system for these very challenging tasks should be an open and flexible one, to be integrated in multimodality with other detectors (MRI, optical). This is not possible with the system based on Anger Camera. We have studied three possible layout: 1) 8 detectors 100 x 100 mm2 with CaI(Na) 0.8 mm pitch coupled to Hamamatsu H9500 (3 x 3 mm2). 2) 8 detectors made of 100 x 100 mm2 LaBr3(Ce) continuous 6 mm thick. 3) 8 detectors 50 x 50 mm2 CsI(Na). Results on these studies as well as the ones ongoing on the use of Silicon PhotoMultipliers (SiPm) that will allow integrating the system in a MRI scanner, will be described during the Conference.
M09-95:

High Resolution and High Efficiency Open SPECT Detector for Molecular Imaging Studies of Cardiovascular Diseases on Mice

A Laser-Based Method and Apparatus for Precise Detector Block Positioning and Finding Misalignments of Detector Blocks in Positron Emission Tomographs

H. Sabet, K. Ishii, S. Matsuyama, K. Nakazawa Dept. of Quantum Science & Energy Engineering, Tohoku University, Sendai, Japan A method and apparatus for detector block positioning and finding misalignments of detector blocks in Positron Emission Tomographs (PET) is developed. In this work, the apparatus is designed for use in FinePET (small animal PET which was developed at Tohoku University, Japan). A laser sensor is used to measure the distance to the surface of detector blocks. A very fine dedicated holder for the laser sensor is designed and constructed to hold and carry the laser sensor in the gantry. Measuring the distance from center of the gantry to all detector blocks, the equation of surface of all detector blocks can be derived. Having the complete set of surface equation for all detector blocks and comparing them with the ideal position of detector blocks, an error vector for position of all detector blocks is generated. According to the error vector, either the detector blocks can be moved to the ideal position or the coordinate of detector blocks used by image reconstruction program can be modified. This approach

346

can be utilized to improve the spatial resolution of PET scanner. Furthermore, using the apparatus in PETs with transformable gantries can guarantee a perfect detector block positioning in order to prevent any undesired misalignment.
M09-98:

E. Lage1, J. L. Villena1, G. Tapias1, I. Vidal-Migallon1, M. Abella1, A. Sisniega1, N. P. Martinez2, D. Ros2, M. Desco1, J. J. Vaquero1 1 Laboratorio de Imagen Medica, Hospital G. U. Gregorio Maranon, Madrid, Spain 2 Unidad de Biofisica y Bioingenieria, Universidad de Barcelona- IDIBAPS, Barcelona, Spain This work reports on the development and assessment of the r-SPECT system, a new laboratory small-animal scanner consisting on two opposed small-field-of-view gamma cameras mounted on a rotatory gantry. Each detector is built around a position sensitive photomultiplier (Hamamatsu H8500), a 30x30 NaI(Tl) scintillator array with elements of 1.4x1.4x6 mm3 and electronics for pre-processing and matching the detector signals to an in-house developed data acquisition system. The camera components are enclosed in a lead-covered case with a receptacle that allows easy collimator exchange (parallel hole or pinhole with different tungsten apertures) to optimize the detectors configuration depending on the study requirements. System performance has been assessed following the guidelines described on the NEMA NU-1 2007 for different imaging scenarios: low energy and high resolution parallel-hole collimator (LEHR) and a 0.75 mm pinhole collimator with a 60 degree aperture angle. Intrinsic energy resolution is 10% at 140 KeV. Planar spatial resolution using the parallel collimator ranges from 1.7 mm at the detector surface to 4.2 mm at 45 mm radius of rotation (ROR) with sensitivity of 3.5 cps/Ci/detector (20% energy window). Planar spatial resolution using the pinhole collimator ranges from 1 mm at 10 mm to 2.4 mm at 45 mm. Sensitivity using this collimator ranges from 3 cps/Ci/detector at 15 mm to 0.5 cps/Ci/detector at 45 mm (20% energy window). Measured tomographic spatial resolution in typical mouse studies (pinhole collimator, at 24 mm) is < 1.3 mm (FDK reconstruction). Spatial resolution in typical rat studies (LEHR collimator, at 32 mm) is < 2.1 mm (OSEM-2D, 10 subsets, 5 iterations). This SPECT was conceived for compactness and cost-effective routine small-animal imaging, but even though this design constrain acquisitions of living mice and rats obtained with the system demonstrate its ability to provide useful high-resolution images for in vivo research.
M09-101:

R-SPECT: a Compact Gamma Camera Based SPECT System for Small-Animal Imaging

R. Ter-Antonyan1, R. J. Jaszczak1,2, J. E. Bowsher3, K. L. Greer1, S. D. Metzler4 1 Dept. of Radiology, Duke University Medical Center, Durham, NC, USA 2 Dept. of Biomedical Engineering, Duke University, Durham, NC, USA 3 Dept. of Radiation Oncology, Duke University Medical Center, Durham, NC, USA 4 Dept. of Radiology, University of Pennsylvania, Philadelphia, PA, USA

Multi-Geometric Collimation for High Sensitivity Brain SPECT

Certain neurological disorders such as Parkinsons disease are diagnosed by imaging centrally located brain structures such as basal ganglia and striatum. High sensitivity imaging of these structures is essential for the early-stage diagnosis of the disorder. We simulated a multi-geometric collimator that combines a short-focal-length astigmatic half-cone-beam collimator, designed to efficiently image the bottom half of the brain and especially the centrally located brain structures, and a spatially-variablefocusing collimator, designed to image the rest of the brain without truncation. We demonstrate that a multi-geometric collimator may offer significant sensitivity increase in imaging of the centrally located brain structures compared to conventional generalpurpose parallel-hole and fan-beam collimators. As a result, the noise is reduced in the reconstructed images and the image quality is improved. Multi-geometric collimation shows a strong potential for improved regional tomography of the brain and early-stage diagnosis of neurological disorders such as Parkinsons disease.
This work was supported by the National Institute of Neurological Disorders and Stroke of the National Instututes of Health (NIH) under Grant R01 NS054797. M09-104:

D. R. Patlolla1,2, J. E. Breeding2, W. F. Jones2, J. L. Everman2 1 Electrical and Computer Engineering, University of Tennessee, Knoxville,TN, USA 2 Siemens Healthcare Molecular Imaging, Siemens Medical Solutions USA Inc., Knoxville,TN, USA

A GPU-Based Architecture for Improved Online Rebinning Performance in Clinical 3-D PET

Online rebinning is an important and well-established technique for reducing the time required to process PET data. However, the need for efficient data processing in a clinical setting is growing rapidly and is beginning to exceed the capability of traditional online processing methods. High-count rate applications such as Rubidium 3-D PET studies can easily saturate current online rebinning technology. Real-time processing at these high count rates is essential to avoid significant data loss. In addition, the emergence of time-of-flight (TOF) scanners is producing very large data sets for processing. TOF applications require efficient online rebinning methods so as to maintain high patient throughput. Currently, several new hardware architectures like Graphics Processing Units (GPUs) are available to speedup data parallel and number crunching algorithms. In comparison to the usual 347

parallel systems, such as multiprocessor or clustered machines, GPU hardware can be much faster and above all, it is significantly cheaper. The GPUs have been primarily delivered for graphics for video games but are now being used for High Performance computing across many domains. In this study, we investigate the suitability of the GPU for PET Rebinning algorithms and also implemented them on the same.
The authors wish to thank the following: Catalina Olarte. M09-107:

V. W. Zhou1,2, W. C. Leung3, A. Kyme1,2, S. R. Meikle2, R. Fulton4 1 School of Physics, Sydney University, Sydney, Australia 2 Discipline of Medical Radiation Sciences and Brain & Mind Research Institute, Sydney University, Sydney, Australia 3 Centre for Medical Radiation Physics, University of Wollongong, Wollongong, Australia 4 Department of Medical Physics, Westmead Hospital, Sydney, Australia The LOR rebinning technique for motion correction in PET imaging has been shown to compensate effectively for rigid motion. In the LOR rebinning each individual event, represented as an LOR, is spatially transformed to compensate for motion of the object during the scan. The main limitation of this technique is that some motion-corrected events which fall out of the acceptance range of sinograms are discarded, reducing count statistics in the affected sinogram bins and potentially leading to artifacts in the reconstructed image. In this paper we describe a compensation method for lost events in LOR rebinning that uses single slice rebinning to assign events that would otherwise be lost to a sinogram. The experimental verification is presented.
M09-110:

Compensation for Lost Events in LOR Rebinning Motion Correction for PET

F. Gallivanone1, A. Stefano2, C. Canevari3, C. Messa1,3,4, M. C. Gilardi1,3,4, I. Castiglioni4 1 Universita' Milano Bicocca, Milan, Italy 2 LATO s.r.l, Cefal (PA), Italy 3 Scientific Institute H. San Raffaele, Milan, Italy 4 IBFM-CNR, Milan, Italy Partial Volume Effect (PVE), as a consequence of the PET limited spatial resolution, affects quantification of radiotracer uptake for small lesions. AIM Aim of this work was to develop a method for PVE correction of PET-CT oncological studies, based on curves of Recovery Coefficients (RC) vs measured Lesion-to-Background ratio (L/Bm) using both Operator-Dependent (OD) and Operator-Independent (OI) techniques to estimate 18F-FDG metabolic uptake within measured volumes (Vm). MATERIALS AND METHODS. Measurements were performed on the PET-CT scanner Discovery STE (GE Medical Systems). Measured radioactivity concentration (Cm) and Vm were estimated by OD (ODmax: maximum value for Cm and CT volume for Vm, ODman: average in manual drawn ROI for Cm and manual defined PET functional volume for Vm) and OI (OI60%: average in a 60% thresholded ROI for Cm and thresholded defined PET functional volume for Vm) techniques. RC curves were obtained by different measurements with the NEMA 2001 IQ phantom and correction was applied to validation phantoms miming different body districts. PVE correction was applied to 16 oncological patients (follow up studies). RESULTS The underestimation of Cm confirmed the severity of the error. RC curves were obtained for each technique, fitting RCs vs. measured lesion dimension at different L/Bm. % residual error between GS and PVE corrected Cm was calculated for validation phantom. ODmax was found accurate (>80%) for breast phantom. ODman and OI60% were found with an accuracy >50% and >76%, respectively. In patient studies PVE correction was found to modify significantly SUV quantification. In the considered patient population, our analysis showed that an accurate SUV quantification with PVE correction provides an trustworthy diagnostic confidence. CONCLUSION Results suggest that the proposed PVE correction method are suitable in a clinical setting. Further investigations will be devoted to optimize RC curves and to enlarge the clinical cases.
M09-113:

PVE Correction from PET-CT PVE Affected Images

Strategies to Reduce Artifacts and Improve Accuracy in Multiplexed Multi-Pinhole Small Animal

SPECT
F. P. DiFilippo, S. Patel Dept. of Nuclear Medicine, Cleveland Clinic, Cleveland, OH, USA Multi-pinhole collimation is an established technique for improving image quality in small animal SPECT. However multiplexed data introduces ambiguity in image reconstruction and potentially causes image artifacts. We investigate several strategies to optimize data sampling and minimize artifacts, including varying the pinhole configuration, introducing 1D and 3D bed motion, and providing a CT-based seed for iterative image reconstruction. Through computer simulations, we assess the merits of these strategies in terms of quantitative accuracy, image noise, and lesion detection. Overall image quality is best with an imbalanced and irregular multiplexed pinhole configuration, which provides better lesion detection than the non-multiplexed case and with negligible artifact. Use of the CT-based seed for reconstruction significantly improves quantitative accuracy. Axial bed motion, and to a lesser degree transaxial bed motion, helps to eliminate truncation artifact and multiplexing artifact. Further work is 348

ongoing to study additional configurations. For mouse studies, optimal system design depends on distribution, dose, and imaging task. Study-specific simulations should be run to optimize the configuration. A multi-pinhole collimator designed with an irregular configuration of pinhole inserts that may be selectively blocked with lead plugs offers performance and flexibility for various mouse imaging applications.
M09-116:

B. S. McDonald1,2, S. Shokouhi2,3, T. E. Peterson2,3 1 Department of Physics and Astronomy, Vanderbilt University, Nashville, TN, USA 2 Institute of Imaging Science, Vanderbilt University, Nashville, TN, USA 3 Department of Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, USA The goal of this project is to build a multi-pinhole SPECT system with specific capabilities and purpose: high spatial resolution and sensitivity, motionless, and small field-of-view, such as for imaging targeted regions of mice and rats with I-125-labeled radiotracers. High space-bandwidth product silicon double-sided strip detectors (DSSDs) form the basis of the system. A single camera head consists of two DSSDs and a multi-pinhole aperture with pinholes tilted toward a common focal point. Stacking detectors increases overall detection efficiency and permits a unique method of reconstruction based on the synthetic-collimator concept. Because of the high spatial resolution, which was shown to be equivalent to the 59-micron strip pitch, the front detector can be placed close to the aperture to limit multiplexing. The back detector's projections have better resolution but also have more multiplexing. The system consists of two camera heads with their aperture planes orthogonal to one another. Neither the camera heads nor the object is rotated. Combining simulated simultaneously-acquired, four-projection data into an iterative reconstruction algorithm has produced good tomographic results. We have completed assembly of the camera-head housings and optimized detector operation. We will present updates on the detector characteristics, the camera imaging performance, and calibration methods for SPECT measurements.
M09-119:

Imaging Performance Measurements of SiliSPECT

A. Orbom1, P. Golubev2, V. Avdeichikov2, K. Ljunggren1, B. Jakobsson2, S.-E. Strand1 1 Dept. of Medical Radiation Physics, Lund University, Lund, Sweden 2 Dept. of Physics, Lund University, Lund, Sweden

Evaluation of a Two-Detector E-E Technique to Improve Isotope Separation in Digital Autoradiography

Introduction: In ex vivo autoradiography, imaging several isotopes simultaneously gives the ability to use several types of targeting molecules in the same animal thereby reducing the necessary number of animals. Separation of the isotopes is possible by using both physical half-life and emission energy spectra. For energy separation there is a trade off where a thicker detector can accurately measure higher energy particles but where scattering will deteriorate spatial resolution. Using a thinner, highresolution detector, separating isotopes with emissions above 300 keV becomes difficult with limited statistics since the measured dE/dx spectra will be very similar. We propose to use a two-detector system (E-E) to measure the energy of particles that escape the first detector to improve isotope separation. The purpose of this work is to investigate the viability of this concept. Methods: For the E detector, a double sided silicon strip detector is a good choice and in our test setup we used one with a thickness of 300 m. The E detector could be a thicker silicon detector or a fast scintillator, Initially, we used a 1000 m thick single sided silicon strip detector. Using this setup we measured samples of 125I, 131I and 90Y and compared the recognition patterns in the E detector when the E detector was left out or included in the trigger. Results: The 125I signal mostly disappeared with the E detector condition, and the 131I signal was reduced compared to 90Y, a difference that can be used to separate 131I and 90 Y. Conclusions: Early results indicate that a two-detector system can improve isotope separation by absorbing more energy from high-energy betas. Data analysis and GEANT4 simulation is currently ongoing and results will be presented at the meeting. Future work will also include testing the principle using a DSSD+scintillator setup.
M09-122:

G. Montemont1, C. Comtat2, F. Mathy1, O. Monnet1, R. Trebossen2 1 CEA-LETI, Grenoble, France 2 CEA-SHFJ, Orsay, France

Experimental Evaluation of a High Resolution CdTe-Based PET System

Small animal PET imaging requires both high resolution and high sensitivity in order to detect small amounts of radiopharmaceutical with precision. Semiconductor detector such as Cadmium Telluride (CdTe) makes it possible to localize gamma ray interactions in 3D at submillimetric scale, enabling the design of PET system architectures with innovative geometries. Indeed, thank to a good knowledge of the depth of interaction (DOI) inside the detection unit, it become achievable to obtain a homogeneous spatial resolution in the whole field of view (FOV) and then to maximize the solid angle coverage, and then the line-of-response (LOR) coverage. We are developing a CdTe PET detection module, based on the stacking of individual detectors, associated with read-out ASICs. To validate experimentally this design, a small test bench featuring two detection heads was built. A dedicated electronics associating the readout ASIC and FPGA-based processing boards was developed. 349

Experiments were carried out with point sources (using 200 micrometer capillary tubes) and micro-Derenzo phantoms. First results show the potential of this technology to achieve homogeneous spatial resolution in the whole field of view.
This work has been supported by French National Research Agency (ANR) through TecSan program (project Topase-Med n ANR-06-TECS009). The authors wish to thank all people who contributed to this project and especially BiospaceLab team for their help on the mechanical part. M09-125:

ZIPI: a Single Detector Insert to Locally Improve the Sensitivity and Spatial Resolution of Small Animal PET

S. St. James, J. Zhou, J. Qi, S. R. Cherry Biomedical Engineering, University of California, Davis, Davis, Ca, U.S.A. In small animal PET, the spatial resolution depends on several factors, among which the crystal width is a dominating factor. The concept of virtual pinhole PET uses the principle that two detectors with different intrinsic spatial resolution may be used in coincidence. The intrinsic spatial resolution achieved with these two detectors will vary, with the measured spatial resolution being dominated by the detector that an object is closest to. We propose building a single ultra-high resolution depth-ofinteraction (DOI) detector and integrating this detector into the existing microPET II scanner to locally improve the spatial resolution and sensitivity in a volume of interest. This is different from previous work where many high resolution detectors were used in a ring or partial ring configuration inside of an existing scanner. In many small animal PET studies the region of interest is already known prior to imaging. Examples of this include xenograft tumor studies and heart imaging. We propose that when these studies are performed with the single detector insert, the region of interest (tumor, heart) would be placed close to the zoom-in PET insert (ZIPI), resulting in locally improved spatial resolution in the axial and tangential directions. The detector that we propose for the ZIPI is based on a LSO array made up of 56 x 56 crystals, each of dimension 0.22 mm x 0.22 mm x 20 mm. This array would be read-out at the back end using a Hamamatsu M64 PMT and at the front end by a 20 mm x 20 mm position sensitive avalanche photodiode. Provided that the individual crystals may be read out, this detector when used with the microPET II scanner will locally improve the spatial resolution of the system. Monte Carlo simulations using GATE have been performed to assess the improvements to sensitivity when incorporating the ZIPI detector into microPET II. At the center of the field of view (1 cm from the face of ZIPI) the sensitivity is improved by 22 %.
M09-128:

C. M. Lee1, S. J. Hong2, H. S. Yoon1, M. Ito3, S. I. Kwon1, S. K. Park1, K. S. Sim3, J. S. Lee1 1 Department of Nuclear Medicine, Seoul National University, Seoul, Korea 2 Department of Radiological Science, Eulji University, Gyeonggi-do, Korea 3 Department of Physics, Korea University, Seoul, Korea

Spatial and Energy Resolutions of a Hexagonal Animal PET Scanner Based on Single-Layer LGSO Crystal and Flat-Panel PMT

The aim of this study was to explore both spatial and energy resolution of the PET scanner that we have recently developed. Based on 6 detector modules with 1-layer LGSO crystals, this scanner has a hexagonal figure in order to image small animals. A 22 Na point source was employed to estimate radial and tangential spatial resolutions. To assess the energy resolution, uniform 18F cylindrical phantom was scanned. A software-based spectrum analysis method on list-mode data was used to assign local energy window centered on the photo-peak position for every single crystal. For the image reconstruction, ML-EM algorithm was used. The radial and tangential spatial resolutions were 0.99 mm and 1.13 mm, respectively. The energy resolution averaged over each PMT ranged from 13.3% to 14.3%, which gave an average value of 13.8%. These results show that this system is promising in small animal imaging with excellent spatial and energy resolutions.
M09-131:

Photon Interaction Rate Studies for a Semiconductor-Based High-Resolution Small Animal PET

Y. Gu1,2, C. S. Levin2,1 1 Electrical Engineering, Stanford University, Stanford, CA, USA 2 Radiology, Stanford University, Stanford, CA, USA We are developing a 1 mm resolution 3-D positioning small animal PET system using cadmium zinc telluride (CZT) photon detectors. The detectors are 40x40x5 mm^3 monolithic CZT crystals patterned with a cross-strip electrode design with 1 mm anode strip pitch, 5.5 mm cathode strip pitch, and are capable of positioning the 3-D coordinates of individual photon interactions. The system has an 8x8x8 cm^3 box-shaped field of view formed by 16 layers of axially stacked square-shaped detector rings. For signal readout, detectors are arranged into modules each comprising a stacked pair of CZT detectors. Module signals are read out using a commercially-available application specific integrated circuit (ASIC). This study attempts to quantify the flux and spatial distribution of photon interactions within the systems detector volume, and specifically, how they translate to data acquisition bandwidth specifications that are compatible with the readout ASIC capabilities. 350

System

Monte Carlo simulation was performed using a realistic mouse phantom (300 uCi uniformly distributed within) placed at the system center, and a simulated acquisition time of 1 sec (11181701 counts above a 15keV energy threshold). The flux of photon interactions was discretized according to anode and cathode detection volumes to enable calculation of expected hit rates on individual anode and cathode strips. Simulation results showed maximum per anode and cathode strip hit rates of 5694/s and 59190/s respectively. The hit rate was seen to reduce by up to an order of magnitude for electrode strips near the corners and away from the system center. Based on this data and the ASICs estimated maximum of 138.14k triggers/s throughput, each ASIC is expected to be capable of reading 30 anodes with an average hit rate of ~4500 independent triggers per second. Careful load balancing across the ASIC for cathode readout will allow 6 ASICs to provide adequate readout bandwidth for every detector module throughout the system.
This work was supported in part by the U.S. National Institute of Health under Grant No. R01 CA120474. M09-134:

Y. Yang1, Y. Wu1, S. St. James1, H. Du1, J. Qi1, R. Farrell2, P. A. Dokhale2, K. S. Shah2, S. R. Cherry1 1 Department of Biomedical Engineering, University of California at Davis, Davis, CA, USA 2 Radiation Monitoring Devices Inc., Watertown, MA, USA

Effects of Reflector and Crystal Surface for Depth Encoding Small Animal PET Detectors

We have been developing depth encoding PET detectors by using dual-ended readout of finely pixelated LSO arrays with PSAPDs to simultaneously obtain high spatial resolution and high sensitivity. In this work we compare performance of detectors made of different crystal surfaces and inter-crystal reflectors and try to find the best compromise between crystal identification and DOI resolution. The first comparison was performed on four 5x5 rectangular LSO arrays with crystal size of 1.5x1.5x20 mm3. The crystal surfaces are either polished or unpolished (leave as cut), and the inter-crystal reflectors are either the enhanced specular reflector (ESR) or Toray lumirror E60 film (Toray). The results show that all but the polished ESR array can be used as depth encoding PET detector and the two arrays using Toray reflector provide both better crystal identification and DOI resolution. The second comparison was perform on seven tapered LSO arrays consisting of 13x13 crystals of 0.514x0.514 mm2 at the small end, 0.514x0.914 mm2 at the large end and 20 mm long and using different crystal surface and inter-crystal reflectors. For arrays that use ESR reflector the flood histograms are better in the second cut direction than the first cut direction. For arrays use both ESR and Toray reflectors, flood histograms are improved but DOI resolution is degraded as more crystal surfaces are polished. For the array use Toray reflector and with all crystal surface polished, all crystals can be clearly resolved from the flood histogram and a DOI resolution of 2.6 mm is obtained.
M09-137:

The Design of a Hybrid Small Animal Imaging System

M. R. Dimmock, J. E. Gillam, T. E. Beveridge, J. M. C. Brown, R. A. Lewis, C. J. Hall Monash Centre for Synchrotron Science, Melbourne, Australia In order to successfully perform small animal single-photon emission imaging the detection system is required to provide both high efficiency to emitted radiation and excellent spatial resolution. The convergence of these detection characteristics, between which there is generally a direct trade-off, often leads to sacrifices in either efficiency or resolution. The Pixelated Emission Detector for RaddiOisotopes (PEDRO) is a single-photon emission imaging project for small animal volumes and will perform both mechanical and electronic collimation simultaneously. It is developing a mechanically collimated modular Compton camera configuration that will be able to be optimized over a wide range of photon energies (30keV to 511keV), and provides a number of other imaging advantages. The spectral resolution required to perform Compton collimation will also enable multi-isotope studies. It is intended that the information obtained from both mechanical and electronic collimation be combined in the final reconstructed image. Currently simulations for optimization of PEDRO are being performed, and a Phase I prototype system is under construction. In this study the design considerations for this prototype imaging system are described. The experimental system is planned to both validate simulations and perform proof-of-principle measurements, and while it is not the optimum geometry, it will provide an excellent testbed for the PEDRO concept.
M09-140:

Noise Propagation in Multipinhole SPECT Calibration

K. Vunckx, J. Nuyts Nuclear Medicine, K.U.Leuven, Leuven, Belgium For small animal imaging using a clinical gamma camera, a calibration scan is required to determine the geometrical camera parameters with sufficient accuracy such that high spatial resolution can be achieved in the reconstruction image. It was shown previously that a single pinhole SPECT system, requires a measurement of a calibration phantom with 3 point sources of which two inter-point distances are known, while multipinhole SPECT only requires 2 point sources without any knowledge about their distance with respect to each other. However, anecdotal tests with measured data indicated that if the distance between the point sources is not fixed, the calibration of the multipinhole system is very senstive to noise. In this work, we carried out simulations to assess the effect of fixing the inter-point distances on the estimated calibration parameters. The simulations revealed that fixing 351

the inter-point distances indeed strongly reduces noise propagation, which on its turn improves the reconstruction image quality. In addition, it was found that in multipinhole SPECT, similar robustness can be obtained by fixing the distance(s) between the pinhole apertures instead of those between the point sources. This has the advantage that the calibration scan can easily be done simultaneously with the animal or phantom scan by positioning (at least) two easily discernable point sources inside the field of view (FOV). Not including any distance information or reducing the number of point sources decreases both the calibration and the reconstruction accuracy.
M09-143:

J. Y. Hwang1,2, S.-J. Lee1,2, C.-H. Baek1,2, M. Ito3, S. J. Hong4, J. S. Lee5, Y. H. Chung1,2 1 Department of Radiological Science, College of Health Science, Yonsei University, Wonju, 220-710, South Korea 2 Institute of Health Science, Yonsei University, Wonju, 220-710, South Korea 3 Department of Physics, Korea University, Seoul, 136-701, South Korea 4 Department of Radiological Science, Eulji University, Seongnam, 461-713, South Korea 5 Department of Nuclear Medicine, Seoul National University College of Medicine, Seoul, 110-744, South Korea

Monte Carlo Simulation of Four-Layer DOI Detector with Relative Offset in Animal PET

We have built a four-layer detector to obtain the depth of interaction (DOI) information in which all four layers have a relative offset of a half crystal pitch with each other. The main characteristics of the detector, especially the energy and spatial resolutions, strongly depend on the crystal surface treatments. As a part of the work for the development of an animal PET, we have investigated the effect of crystal surface treatment on the detector performances by Monte Carlo simulation, in order to optimize the surface condition of crystals composing a four-layer detector. The proposed detector consists of four LYSO layers with a crystal dimension of 1.5 mm x 1.5 mm x 7.0 mm. A simulation tool, DETECT2000, was used and validated against the experimental results, flood images acquired by prototype module. Spatial resolution and sensitivity were simulated by varying the surface treatment in crystals, POLISH, PAINT, and METAL. In this study, the optimal surface condition of the four-layer crystals was derived for small animal PET with a view towards achieving high sensitivity as well as high and uniform radial resolution.
M09-146:

P. Major1, G. Hesz2, A. Szlavecz2, G. Nemeth1 1 research/development, Mediso Ltd., Budapest, Hungary 2 Dept. of Control Engineering and Information Technology, Budapest Univ. of Technology and Economics, Budapest, Hungary The PET component of NanoPET/CT is a high resolution small animal PET system, in which every detector module contains 81 x 39 LYSO crystal needles with the size of 1.12 x 1.12 x 13 mm3. Two Hamamatsu H-9500 position sensitive photomultipliers (PSPMT-s) are optically coupled to each crystal matrix using a thin BK-7 optical glass lightguide, allowing 95 mm AFOV. However, the gain of the 256 anodes of the PSPMT are not uniform, hence the local energy scale of the different anodes can differ by a factor of three to one. On the other hand, the active area of the PSPMT-s is always smaller than the real size of those, therefore the light collection effeciency for the crystal needles at the middle of the module (between the two PSPMT-s) is restricted, resulting in further non-uniformity effect. Consequently, the commonly used 350-650 keV global energy window for the data acquisition could cause the loss of 511 keV (photopeak) events. We have developed and implemented an application called LUT-QT to solve the crystal positioning and the gain uniformity correction. Considering the number of crystal elements in the detector ring (37908 pcs), all of the methods were designed to run automatically, although all of them allows the possibility of manual modification in every steps. This method can be applied for every PET detector module based on scintillation crystal matrix and multianode PSPMT-s.
The authors would like to thank Gergely Nemeth, Sandor Torok, Zoltan Nyitrai, Balazs Domonkos, Gergely Patay, and David Volgyes for their help and for useful discussions. This work has been supported by Mediso Ltd. and National Office for Research and Technology (NKTH) grant NKFP-06-A1-PETCT-06. M09-149:

Local Energy Scale Map for NanoPET/CT System (Summary)

M. Carles1, A. Ros1, C. W. Lerche2, F. Sanchez1, A. Sebastia2, J. M. Benlloch1 1 IFIC (CSIC-U.V.), Valencia, SPAIN 2 ITACA, UPV, Valencia, SPAIN

Detection Efficiency and Spatial Resolution for a Continuous Scintillation Crystal - Interface Continuous Scintillation Crystal System in Positron Emission Tomography (PET)

In this work we study the effects on the detection efficiency, spatial resolution, and depth of interaction (DOI) determination, for a continuous scintillation crystal - interface - continuous scintillation crystal (CIC) system. Due to the 511 keV of the gammarays employed in Positron Emission Tomography (PET), thick scintillators are required in order to achieve a sufficiently high detection efficiency. However, using thick scintillation crystals for detectors that do not provide information about DOI of the recorded gamma-rays results in a parallax error which leads to uncertainties in the parameterization of the line of response (LOR) 352

and an error due to image compression. These errors increase with the crystal thickness and will give rise to an inhomogeneous spatial resolution. DOI determination in a PET system allows minimizing parallax error and permits to discard one of the main disadvantages of thicker scintillation crystals.We report how the combination of the CIC system and the DOI determination method affects energy and spatial resolution.
M09-152:

S. A. Kis1, I. Lajtos1, M. Emri1, G. Opposits1, T. Bukki2, G. Hegyesi3, J. Imrek3, I. Valastyan3, J. Molnar3, D. Novak3, L. Balkay1 1 University of Debrecen, Institute of Nuclear Medicine, Debrecen, Hungary 2 MEDISO Ltd., Budapest, Hungary 3 Institute of Nuclear Research of the Hungarian Academy of Sciences, Debrecen, Hungary The examination of small animals using high resolution in-vivo imaging methods is a crucial tool in pharmaceutical research projects. To suffice this demand, a full ring small animal PET camera (MiniPET-II) has been built in our Institute as part of an R+D project. In this work we determined the performance parameters of the MiniPET-II scanner. The measurements and data evaluation for this purpose were based on the National Electrical Manufacturers Association (NEMA) NU-4 standards. The spatial resolution is varied between 1.5 to 2.1 mm at 5 and 25 mm radial distances. The system sensitivity was 3.18%. The counting rate capability, expressed in noise equivalent counting rate (NEC), was shown to peak at over 185 kcps at 63 MBq using a rat phantom. The scatter fraction at the same acquisition was 19.4%. Evaluations of image quality and quantitation accuracy were also performed using the NEMA NU-4 required image-quality phantom and animal studies.
This project was supported by the National Fund for Research and Development tender NKFP-A1-2006-0017. M09-155:

Performance Test of the MiniPET-II Small Animal Scanner According to the NEMA NU-4 Standards

Multivariate Approaches for Alzheimer's Disease Diagnosis Using Bayesian Classifiers

M. Lopez, J. Ramirez, J. M. Gorriz, D. Salas-Gonzalez, I. Alvarez, F. Segovia, R. Chaves Dept. of Signal Theory, Networking and Communications, University of Granada, Granada, Spain In the diagnosis of the Alzheimer's Disease (AD), the evaluation of cerebral functional images such as PET (Positron Emission Tomography) and SPECT (Single Photon Emission Computed Tomography) is usually done through visual ratings performed by experts. Several efforts have been recently done to develop an automatic tool that might aid the experts in this dementia diagnosis. A classification system for the diagnosis of Alzheimers disease was shown in this work. The PET and SPECT images undergo a Principal Component Anlysis (PCA) transformation which maximizes the scatter of all the projected samples and allows a strong dimension reduction by projecting data onto the principal components, solving the well-known small size sample problem. Since the learning set is labeled, it makes sense to use this information to build a more reliable method for reducing the dimensionality of the feature space. Linear Discriminant Analysis (LDA) is a class specific linear method for dimensionality reduction, and it is applied over the PCA coefficients so that dimension is still reduced and samples are maximally separated according to their class labels. The classification is done using a Bayesian classifier, which uses the a priori information to classify new samples. The proposed methodology has shown to perform better than previous multivariate approaches both for PET and SPECT images, reaching 100% and 96.7% accuracy rates respectively.
This work was partly supported by the MICINN under the PETRI DENCLASES (PET2006-0253), TEC2008-02113, NAPOLEON (TEC200768030-C02-01) and HD2008-0029 projects and the Consejera de Innovacin, Ciencia y Empresa (Junta de Andaluca, Spain) under the Excellence Project (TIC-02566). M09-158:

F. Segovia1, J. M. Gorriz1, J. Ramirez1, D. Salas-Gonzalez1, I. A. Illan1, M. Lopez1, R. Chavez1, C. G. Puntonet2 1 Teoria de la Seal, Telemtica y Comunicaciones, Universidad de Granada, Granada, Espaa 2 Arquitectura y Tecnologa de Computadores, Universidad de Granada, Granada, Espaa Dementia type Alzheimer (DTA) has become in an important disease that affects more and more people. Unfortunately Alzheimer disease has no cure and it only possible to mitigate its effects, thus, an early diagnosis is crucial. Diagnosis is traditionally performed by doctors in a visual way from tomography images of the brain which is very subjective and prone to error. SPECT (Single Photon Emission Computed Tomography) is a widely used functional technique to obtain these images that provide an activation map displaying the local intensity of the regional cerebral blood flow (rCBF). In order to improve the prediction accuracy especially in the early stage of the disease, computer aided diagnosis (CAD) tools are desirable. Usually, CAD methods for Alzheimer diagnosis are based on classical classifiers trained using all voxels of SPECT images as features. The advantage of this approach is that no prior knowledge about the disease is needed but it suffers the so-called small size sample problem, due to the fact that images represent large amounts of features and most imaging studies have relatively few subjects (generally less than 100). In this work we use a model-based clustering approach which allows us to avoid the small size 353

Automatic Selection of ROIs Using a Model-Based Clustering Approach

sample problem. For this, we generate an image with differences between normal and DTA subjects and we extract from it the most relevant information by clustering. This information is used to select a reduced set of features of each patient. To test the goodness of this method we have used some classifiers and the Leave-One-Out cross-validation technique reaching an accuracy rate up to 94.5%.
M09-161:

X. Jin1, T. Mulnix2, B. Planeta-Wilson2, R. E. Carson1 1 Dept. of Biomedical Engineering, Yale University, New Haven, CT, USA 2 Dept. of Diagnostic Radiology, Yale University, New Haven, CT, USA

Accuracy of Head Motion Compensation for the HRRT: Comparison of Methods

Motion correction in PET has become more important as system resolution has improved. The purpose of this study was to evaluate the accuracy of two motion compensation methods, event-by-event motion compensation with list-mode reconstruction (MOLAR) and frame-based post-reconstruction motion correction using image registration. Motion compensated image reconstructions were carried out with simulated HRRT data, using a range of motion information based on actual motion data. ROI analyses in high contrast regions were performed to evaluate the accuracy of both motion compensation methods. Our study showed that MOLAR can reliably correct for all reasonable head motions. Over all motions, the average ROI count was within 2.13.1% of the reference no motion value. The location of the ROI centroid was found to be within 0.80.4 mm of that of the reference image. Frame-based motion compensation was able to correct for small (<3mm translational and 5 rotational) within-frame motions, but image registration accuracy begins to deteriorate for medium (<10mm translational and 15 rotational) and large (10-50mm translational and 15-30 rotational) motions. The average centroid location of the ROI had an offset error of 1.61.0 mm for medium motions and 2.70.9 mm for large motions. The average count in the ROI dropped by 6.66.9% for medium motions, and by 34.211.5% for large motions. We conclude that event-by-event based motion correction works accurately for all reasonable motions, whereas frame-based motion correction is accurate only when the within-frame motion is small.
We thank Zhongdong Sun for programming support and the staff of the Yale PET Center for the studies which formed the basis of this work. M09-164:

T. Hughes1, S. Shcherbinin2, A. Celler2 1 Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada 2 Radiology, University of British Columbia, Vancouver, BC, Canada

A New Approach in Patient Motion Correction for Cardiac SPECT: a Simulation Study

Objective: Patient motion artifacts created in cardiac SPECT imaging can lead to misinterpretation of the images, resulting in false diagnoses. This simulation study proposes a new technique for patient motion correction (MC), where we utilize a modified template projection/reconstruction (TPR) algorithm to perform a voxel-by-voxel correction to the original image. Methods: Using NCAT, we developed two female phantoms with large breasts containing a non-beating heart (heart: background=5:1). Phantom 1 had a healthy heart, and phantom 2 had a heart with a small (10%) perfusion defect in the lateral wall (severity=50%). The SimSET code was used to perform simulations for both phantoms modeling cardiac SPECT acquisitions with Tc-99m, 128x128 matrix, and 60 camera stops (20s each). In addition to two standard (no motion) acquisitions (ST) for each phantom, seven acquisitions with different degrees of phantom motion were created by manually shifting a selected number of projections in a given direction (motion ranged from 8 to 22 mm). MC images (MCI) were created using a modified TPR, where the projected template is adjusted to match the motion detected in the experimental projections by aligning the center of mass in each projection. All reconstructions were performed using OSEM with resolution recovery and attenuation correction. Results: For all simulated movements, the MCI images exhibited improvements in both standard deviation (SD) and mean accuracy relative to the uncorrected experimental reconstructions (ER). On average, the accuracies calculated for ER, MCI and the ST reconstructions were 68%, 77%, and 76%, respectively. The average SD for ER, MCI and the ST reconstructions were 5.2, 4.0, and 4.0, respectively. Conclusions: Our proposed technique offers a voxel-by-voxel motion correction, which provided improved image accuracy and standard deviation of counts relative to the uncorrected images and, in many cases, the images created without patient motion.
M09-167:

R. E. Ansorge1, S. J. Sawiak2, G. B. Williams2 1 Cavendish Laboratory, University of Cambridge, Cambridge, UK 2 Wolfson Brain Imaging Centre, University of Cambridge, Cambrdige, UK

Exceptionally Fast Non-Linear 3D Image Registration Using GPUs

3D non-linear image registration has many important applications in medical image processing. For example following disease progression or comparing genetically modified animal models. MRI images of the brain are often used for these investigations and individual data sets contain large numbers of voxels (256x256x256 is typical). Unfortunately the high resolution deformation maps needed, require very lengthy computations. For example a cubic B-Spline based deformation map with control point 354

spacings of a few voxels in each dimension can require days to compute on a single processor. Typically PC clusters have been used to speed the process but these are expensive, require cooled server rooms and dedicated technical support. We present new results from a demonstration program, AIRWC (Accelerated Image Registration with CUDA), which uses an NVIDIA GTX285 Graphics processor card to speed computation of both 12-parameter affine and 3D cubic B-Spline based image registrations. The performance of this program is very encouraging; typically affine registrations require a few seconds and high resolution B-Spline registrations having up to 131x131x131 control points require ~6 minutes. The latter case corresponds a speedup of ~750 over our conventional code running on a single processor. Results are presented for both human and mouse brain imaging. A high resolution average brain was computed from 14 human scans by iteratively registering to an average target. In all about 112 high-resolution B-Spline registrations were performed in about 9 hours using a DELL XPS630i PC running Windows Vista. The GPU coding was carried out using NVIDIA's C++ like CUDA programming language and some details are presented. Overall we have found the code development to be rather straight forward and thus we expect that the exploitation of GPUs for image registration will grow rapidly.
We acknowledge the help of colleagues at the Wolfson Brain Imaging Centre in providing our images. M09-170:

Impact of Partial Volume Correction in Whole-Body PET Imaging : a Computer-Aided Detection

C. Lartizien1, S. Tomei1, A. Le Pogam2, S. Marache-Francisco1, D. Visvikis2 1 CREATIS - CNRS UMR 5220 - INSERM U630 - INSA Lyon - Lyon University, Villeurbanne, France 2 LATIM-INSERM U650, Brest, France Partial volume effects (PVEs) are well-known consequences of the limited spatial resolution in emission tomography which can affect the quality of PET oncology images both qualitatively and quantitatively and thus impact the diagnostic task. There exists a wide variety of PVE correction methods but they are not all well suited to oncology images. In an attempt to obviate the need for drawing or segmenting ROIs as well as to produce PVE-corrected images, Boussion et al have proposed a novel voxel-wise PVE correction using the iterative Lucy-Richardson deconvolution methodology in combination with a dedicated wavelet-based denoising algorithm (LR-W). A refinement of this method based on the use of ridgelets instead of wavelets (LR-R) has also been recently proposed by Le Pogam et al. The purpose of this study is to evaluate the impact of these PVE correction methods on clinical detection performances. This study is based on a series of realistic simulated whole-body FDG images including spherical lesions of calibrated diameters and uptakes. Detection performance is evaluated based on a computer-aided detection system that we are developing for whole-body PET/CT images. Comparison of the detection performance achieved with these PVE correction methods is performed visually based on the analysis of the detection maps derived from the CAD system and quantitatively based on a Free-Response Receiver Operating Characteristic (FROC) analysis. Visual analysis of the detection maps indicates equivalent detection sensitivity but some differences in the spatial distribution of false positive detections. The CAD system indeed generates many false detections at the organs edges with PVE corrected images. Quantitative results based on the FROC analysis indicate similar detection performance. Performances achieved with the PVE corrected images are likely to be improved when also correcting the normal (ie non pathological) images database for PVE during the CAD training phase.
M09-173:

Study

J. Arjona1, P. H. Pretorius2, H. Gifford2, J. G. Brankov1 1 ECE Dept., Illinois Institute of Technology, Chicago, IL, USA 2 Dept. of Radiology, University of Massachusetts Medical School, Worcester, MA, USA

Generalization Performance Evaluation of the Internal Noise Models for CHO

In this work we compare the generalization performance of different internal noise models for the channelized Hotelling observer (CHO) in terms of predicting human observer (HO) detection performance. Here the generalization performance is measured as CHO accuracy in predicting HO over new independent test samples. A HO study for a detection task is a gold standard but it is a costly and time consuming procedure. Therefore, the development of a numerical observer (NO) surrogate is highly desirable. The CHO, the most widely used NO, typically employs channels (to extract image features) and an internal noise model to predict HO detection performance. The properties of the CHO have been studied for many years but little attention has been given to CHO generalization performance. Specifically, in this work we evaluate CHO agreement with HO where the different internal noise models parameters are adjusted using HO data and images reconstructed with a broad set reconstruction technique, then tested on images reconstructed by an equally broad set but a different reconstruction technique, which were not available during CHO adjustment.
This work was supported by NIH grant HL065425 and HL091017.

355

M09-176:

W. Luo1, M. T. Harz2, C. G. Matthews1 1 Naviscan, Inc, San Diego, California, USA 2 Fraunhofer MEVIS, Bremen, Germany

Improving Lesion Detectability of a PEM System with Post-Reconstruction Filtering

We present a method to quantify the image quality of a positron emission mammography (PEM) imaging system through the metric of lesion detectability. For a customized image quality phantom, we assess the impact of different post-reconstruction filters on the acquired PEM image. We acquired six images of the image quality phantom on a Naviscan PEM scanner using different scan durations which gave differing amounts of background noise. The image quality phantom has dimensions 130 mm x 130 mm x 66 mm and consists of 15 hot rod inserts with diameters of approximately 10 mm, 5 mm, 4 mm, 3 mm, and 2 mm filled with activities ratios of 3.5, 6.8 and 12.7 times the background activity. One region of the phantom had no inserts so as to measure the uniformity of the background noise. Lesion detectability was determined for each background uniformity and each activity ratio by extrapolating a fit of the recovery coefficients to the point where the lesion would be lost in the noise of the background (defined as 3 times the backgrounds standard deviation). The data was reconstructed by the systems standard clinical software using an MLEM algorithm with 5 iterations. We compare the lesion detectability of an unfiltered image to the image after applying one of five common post-reconstruction filters. Two of the filters were found to improve lesion detectability: a bilateral filter (9% improvement) and a Perona-Malik filter (8% improvement). One filter was found to have negligible effect: a Gaussian filter showed a 1% decrease in lesion detectability. The other two filters tested were found to worsen lesion detectability: a median filter (8% decrease) and a stick filter (7% decrease).
M09-179:

Y. Bentoutou1, N. Taleb2, A. Bounoua2, C. Serief1 1 Centre des Techniques Spatiales, Arzew, Oran, Algeria 2 Electronics, Djilali Liabes University, Sidi bel abbes, Algeria

An Elastic Registration Technique for Reducing Patient Motion Artifacts in Digital Subtraction Angiography

This paper presents an elastic approach for the nonrigid registration of digital X-ray angiography images. In particular, the results of the proposed registration algorithm are compared to those obtained using rigid and affine registration techniques. The method is based on local similarity detection by means of template matching according to a combined invariants-based similarity measure, which is insensitive to intensity changes. The algorithm has been successfully applied to the fully automated registration of several clinical data sets. The results clearly indicate that the proposed elastic registration strongly improves the quality of subtraction images and provides more useful clinical images than rigid or affine registration.
M09-182:

F. P. Figueiras1, X. J. Gonzalez1, D. Pareto1,2, V. Gomez3,2, J. Llop3,2, J. D. Gispert1,2 1 Institut d'Alta Tecnologia (IAT-PRBB), Barcelona, Spain 2 CIBER en Bioingeniera, Biomateriales y Nanomedicina (CIBER-BBN), Zaragoza, Spain 3 CIC BiomaGUNE. Unidad de Imagen, San Sebastian, Spain

An Evaluation of Simultaneous Dual-Tracer Technique for PET Static Studies

The simultaneous dual-tracer technique (SDTT) consists in separating out each radiotracer using the difference in their radioactive decay rates. In this study, SDTT was performed at two different levels, using different inicial sources: the dynamic sinogram (SM) and directly the dynamic image (IM). The aim of this study was to assess to what extent the experimental settings could influence the SDTT results. SDTT was tested with a phantom consisting of six eppendorf tubes that were filled with different relative concentrations of 18F-FDG and 13N-NH3 radiotracers. The precision of SDTT depends on the chosen rebinning algorithm and the total time acquisition. Frame duration does not influence SDTT results. The results encourage the use of SDTT as it presents a high potential for simultaneously examining two different biological processes at the same time utilizing rodent PET scanners.
M09-185:

Automatic Segmentation of Distal Radius Using Active Shape Model in DXA

S. Cho, J. Kwon, Y. Ro Image and Video system laboratory, KAIST, Daejon, Korea Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture. Osteoporosis is a common disorder affecting a large number of adults. In the United States more than 25 million people are afflicted, particularly women. The bone mineral measurements in forearm based dual x-ray absorptiometry (DXA) have come to be regarded as an essential criterion for evaluating a patients risk of osteoporotic fracture. The main technique employed for measuring the mass of bone mineral is the automatic segmentation of bone area located in a region of interest. Active shape model (ASM) have been used in most of the 356

previous works on bone segmentation. ASM is an iterative algorithm that tries to fit the shape model to the contours of the vertebrae in the image. ASM has the advantage that they make use of prior knowledge of vertebral shape and appearance and therefore, do not need to rely completely on the information in the images. The proposed distal radius segmentation system consists of mean shape training, chamfer matching, DXA decomposition, and modified ASM process. In ASM, we need to find a set of adjustments which moves each landmark toward a better position. However, original ASM uses an adjustment along a normal to the model boundary toward the strongest image edge, with magnitude proportional to the strength of the edge. To solve this problem, we use DXA decomposition image. In the decomposition image, soft tissue component is eliminated; this offers the advantage of considering only the weight parameter at the edge between bone and background. In order to verify the effectiveness of the proposed segmentation system within distal radius DXA image, experiments were performed with DXA. Experimental results show that the proposed bone segmentation system reveals precise segmentation performance in multiple edge situation.
Acknowledgments : This paper is supported by the development of diagonosis for multi x-ray absorptiometry project (A080782) of Ministry of Health and Welfare in the Republic of Korea M09-188:

Image Magnification Using FREBAS Transform

S. Ito, Y. Harada, Y. Yamada Department of Information Systems Sciences, Utsunomiya University, Utsunomiya, Tochigi, Japan Super-resolution is a method of restoring an image by extrapolating the signal beyond the band-limited receiver system. The Gerchberg-Papoulis algorithm based on the Fourier transform has not been effective, but the Gerchberg-Papoulis method based on the convolution integral can expand the signal band easily. The FREBAS transform is a scale transformation based on the convolution integral. In this study, image magnification with super-resolution effects using FREBAS transform was investigated. FREBAS allows optical scaling of images, and, therefore, we have studied the relation of a scaling parameter D of the FREBAS transform and the improvement of resolution. The image magnifying procedure using FREBAS transform is that the original FREBAS signal which was obtained by an original image by FREBAS transform is expanded by the zero padding outside the signal. The real part of magnified image which was obtained by inverse FREBAS transform of the signal is FREBAS transformed again, and central region is replaced by the original FREBAS signal as true data. As a result, we can obtain the magnified images which were sharpened by super-resolution effects after few times of iterations. It was shown that spatial resolution in the central domain is improved when D is an even number and on the other hand the peripheral domain is improved when D is an odd number.
The present study was supported in part by a Grant-in-Aid for Scientific Research(C) (No. 19560416) by the Japan Society for the Promotion of Science, by a Grant in Eminent Research at Utsunomiya University, and by a Grant-in-Aid at Utsunomiya University Center for Optical Research and Education. M09-191:

J. M. Mukherjee1, H. C. Gifford1, M. A. King1, L. Shao2, X. Song2, J. Wang2 1 Radiology, University of Massachusetts, Worcester,MA, USA 2 Philips Medical Systems, Milpitas, CA, USA

Fast Shift-Variant Blur Compensation Within Iterative Reconstruction for Fan-Beam Collimator

In this paper we have modeled the spatially-varying blurring of a fan-beam collimator as a function of both distance and lateral position across the imaging plane within iterative reconstruction. We demonstrate that an incremental blurring technique can be used to more efficiently model such non-stationary blurring than the straightforward non-stationary convolution methodology by a factor of 1.5 in speed and has 99% less memory usage. This method can therefore be used for efficient variable-blur compensation in 3D for collimator geometries in which the detector response function varies across the imaging plane as well as with distance from the face of the collimator.
This work was supported by a grant from Philips Medical Systems. M09-194:

Rotational Convolution and SPECT Post Processing

G. L. Zeng Utah Center for Advanced Imaging Research, University of Utah, Salt Lake City, UT, USA Convolution is a common mathematical tool to perform shift-invariant filtering. However, in many imaging systems, the point spread function (PSF) is not shift invariant. Usually an iterative algorithm is used to de-blur the non-stationary PSF. This paper presents a new idea of rotational convolution to compensate for the spatially variant PSF. This method first further blurs the image with rotational convolution so that the resultant image has a stationary PSF, then deblurs the resultant image using an

357

efficient shift-invariant decovolution technique, for example, the Wiener filter. The proposed method is illustrated and implemented for SPECT (single photon emission computed tomography) applications.
M09-197:

Motion-Corrected PET Reconstruction Using an External Motion Field

G. Delso, A. Martinez-Moller, R. A. Bundschuh, S. G. Nekolla, S. I. Ziegler Nuklearmedizin, TU Muenchen, Muenchen, Germany The combination of MR and PET scanners can provide a powerful tool for clinical diagnosis and investigation. One of the most interesting features of a combined MR/PET scanner would be the possibility of acquiring simultaneously data from both modalities. This could provide valuable information on physiological motion, which could be incorporated in the PET reconstruction procedure to achieve better uptake positioning and delineation. This paper presents a prospective study on how a non-rigid motion field such as what could be obtained in a whole-body MR/PET scan can be incorporated in a maximumlikelihood PET reconstruction framework. For this purpose, a motion-corrected MLEM algorithm is presented and tested with simulated, phantom and real data sets.
M09-200:

S. Gu1, J. E. Mcnamara2, K. Johnson2, H. C. Gifford2, H. Pretorius2, A. Sklyar1, M. A. Gennert1,2, M. A. King2 1 Computer Science, Worcester Polytechnic Institute, Worcester, U.S.A 2 Radiology, University of Massachusetts Medical School, Worcester, U.S.A Non-rigid body motion is a form of patient motion which degrades SPECT imaging accuracy in addition to respiratory and rigidbody motion. We proposed a deformation correction method to correct non-rigid body motion. However, there was no clinical application of correction was performed. In this paper, we describe an experiment based on a series of two clinical SPECT acquisitions performed in the same patients, one without and the second with body motion. This allowed the use of reconstructions of the first to serve as a standard for judging the reconstructions of the second. A comparison is made between the no motion baseline reconstruction and the reconstructions of the acquisition with motion for no correction, rigid body motion correction, and non-rigid body correction techniques. The result of this experiment shows that deformation correction can improve the SPECT imaging.
This work was supported by the National Institute of Biomedical Imaging and Bioengineering by Grant R01-EB001457. M09-203:

Improved Motion Correction in SPECT Imaging Using a Non-Rigid Motion Model

A. A. Isola1,2, U. van Stevendaal1, D. Schaefer1, W. J. Niessen2,3, M. Grass1 1 Philips Research Europe - Hamburg, Hamburg, Germany 2 Biomedical Imaging Group, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands 3 Faculty of Applied Sciences, Delft University of Technology, Delft, The Netherlands We present a reconstruction method for helical computed tomography, which allows to compensate for artifacts caused by cardiac motion. The method requires a motion-vector field, which is determined from a 4D gated non-compensated region of interest reconstruction. Given the estimated motion-vector field, a cardiac-motion compensated analytical reconstruction algorithm is performed, which results in a significant improvement in image quality.
This work received financial support of the European Community under a Marie-Curie Host Fellowship for Early Stage Researchers Training, MEST-CT-2005-020424. M09-206:

Cardiac-Motion Correction for Helical CT

E. Meyer1,2, F. Bergner1, R. Raupach2, M. Kachelriess1 1 Institute of Medical Physics, University of Erlangen-Nuernberg, Erlangen, Germany 2 Siemens Healthcare, Forchheim, Germany

Normalized Metal Artifact Reduction (NMAR) in Computed Tomography

Severe artifacts degrade the image quality and the diagnostic value of CT images if metal objects are present in the field of measurement. The standard method for metal artifact reduction replaces affected projection data by interpolated data. Often, linear interpolation is used. However, simple sinogram interpolation introduces new artifacts and lacks accuracy close to metal objects. Recently, a method was presented, which uses a simple length normalization of the sinogram prior to interpolation to better preserve the contrast between air and water-equivalent objects [J. Mueller and T. M. Buzug, Proc. SPIE, Vol. 7258:1Y1 1Y8, 2009]. However, contrast between water and bone, for example, is still impaired. We introduce a generalized normalization technique, which concisely preserves details of different materials. It is based on a forward projection of a ternary image, which is obtained from a multi-threshold segmentation of the initial image. Simulations and measurements are performed to evaluate 358

our normalized metal artifact reduction method (NMAR) in comparison to standard MAR with linear interpolation and MAR based on simple length normalization. We find considerable improvements in particular for bone structures with metal implants. The improvements are quantified by comparing profiles through images and sinograms for the different methods using simulated data. NMAR clearly outperforms both other methods. We also obtain promising results by applying NMAR to clinical data. This is demonstrated with a scan of a patient with two hip implants. NMAR is computationally inexpensive. Therefore, our normalization technique can be used as a supplement for any conventional sinogram interpolation-based MAR method.
M09-209:

Simultaneous Crosstalk Compensation for an ROI Activity Estimation in Dual-Isotope SPECT

Imaging
S. Shcherbinin, A. Celler Radiology, The University of British Columbia, Vancouver BC, Canada Simultaneous model-based compensation for dual-isotope SPECT (DI-SPECT) imaging considers data acquired in both energy windows as a single set, models crosstalk effects from both isotopes and includes them simultaneously into the reconstruction. Being very appealing from the theoretical point of view, it requires substantial computer resources. However, if the investigated object can be divided into regions of interest (ROI) and only the average activity in each ROI needs to be reconstructed, the dimension of the imaging problem tremendously decreases. This approach can be applied to any clinical DI-SPECT study where the boundaries of investigated organ(s) can be determined from structural modalities. In this work, we investigate an accuracy of this strategy for DI-SPECT/CT studies. In our technique, the elements of imaging matrix modeling contribution of each photon to each energy window are calculated by uniformly filling each ROI with unity activity and projecting it. For the projection step, our Analytical Photon Distribution (APD) method incorporating effects of attenuation, scattering, and resolution loss was employed. The computed noiseless APD-based projections were re-scaled using empirical calibration factors. For validation, we performed phantom experiments on clinical SPECT/CT system (Infinia-Hawkeye, GE Healthcare), where four 30ml plastic containers (cylindrical bottles) filled with different concentrations of Tc-99m and In-111 activities modeled ROIs. From these phantom experiments, our algorithm recovered the absolute activities in each ROI with errors 1.7-9.4% (In-111 activities) and 5.9-21.4% (Tc-99m activities). The decreased number of unknowns in ROI-based DI-SPECT reconstruction allowed us to use data from only 6 projections (camera stops) without affecting the quantitative accuracy of the results. The improvement of the downscatter estimation and CT-based ROI determination are key elements of our plan to optimize the presented method.
M09-212:

Log-likelihood-based rule for image quality monitoring in the MLEM-based image reconstruction

A. Gaitanis1,2, G. Kontaxakis3,4, G. Spyrou2, G. Panayiotakis1, G. Tzanakos5 1 Medical School, Department of Medical Physics, University of Patras, Patras, Greece 2 Biomedical Research Foundation of Athens Academy, Athens, Greece 3 ETSI Telecomunicacin, Ingeniera Electrnica, Universidad Politcnica de Madrid, Madrid, Spain 4 Networking Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Madrid, Spain 5 Department of Physics, Division of Nuclear & Particle Physics, University of Athens, Athens, Greece We address here the problem of the noise deterioration of the quality of the reconstructed images when employing the maximum likelihood expectation maximization (MLEM) algorithm for iterative image reconstruction in positron emission tomography (PET). It is observed that despite the fact the cost function (log-likelihood) is monotonically increasing, the image quality deteriorates after a certain optimum point during the iterative process. The principal aim of the work is the discovery of a rule that would directly link the quality of the reconstructed images at each iteration with the cost function (log-likelihood) that is being maximized. We assume that the true image corresponds to a likelihood value in correlation with the data acquired, which, when achieved, it makes no sense to continue iterating looking for higher likelihood levels. In this work we study the hypothesis that there is a direct correlation of the log-likelihood of the true image (a quantity that is not known a priori in real PET scans) with certain properties of the pixel updating coefficients (PUC) in the MLEM algorithm. We show here that the minimum value of the PUC for the non-zero pixels in the true image can be directly correlated with the value of the log-likelihood that corresponds to the true image. We are currently exploring the dependence of the correlation parameters with the activity distribution level in the source and the scanner configuration, with the goal to formulate a robust stopping criterion for the MLEM algorithm, Preliminary results employing noise-free synthetic 2D PET sinograms have been carried out and are presented. Further work is ongoing for the validation of the proposed methodology on realistic data generated with the GATE platform.

for PET

359

M09-215:

A Fully Corrected Fan Beam OSEM Algorithm for Dirty Emitters

N. Agbeko, J. O'Sullivan, R. Laforest, J.-C. Cheng Washington University In St. Louis, Saint Louis, MO, USA Non-standard radionuclides used in PET imaging introduce additional constraints into the imaging model that need to be properly accounted for in order to ensure quantitative accuracy. One of these constraints is the complex decay scheme whereby in addition to the two 511keV gamma rays, additional photons of different energies are emitted. We developed a scatter and correction algorithm that simultaneously estimates the cascade and scatter in projection data. Cascade gamma ray coincidences are modeled in a manner similar to scatter correction, but with no angular dependence between the cascade gamma ray and the two 511keV gamma rays. A second constraint is a reduction in spatial resolution due to the long positron range from the generally high positron energy. Accurate modeling of positron range effects requires the use of density-dependent thereby spatially-variant blur to account for the inhomogeneity of the subjects. We present simultaneous cascade and scatter correction implemented in an iterative image reconstruction algorithm based on maximizing the log-likelihood function using a log-cosh potential as the regularization term. The algorithm is implemented using fan-beam projection data and operates on ordered subsets for maximum speed. Accurate modeling of the scatter, cascade and positron range effects will be demonstrated to lead to higher accuracy in the measurement of activity concentrations in small animal PET imaging.
M09-218: Interpolation for the Gap-Filling of the HRRT PET Sinograms by Using the Slices in the Direction of the Radial Samples

U. Tuna, S. Peltonen, U. Ruotsalainen Department of Signal Processing, Tampere University of Technology, Tampere, Finland The ECAT High Resolution Research Tomograph (HRRT) (CTI PET Systems, Knoxville, TN, USA) positron emission tomography (PET) scanner is one of the cornerstones in the nuclear medicine imaging field. The HRRT PET scanner is constructed by using octagonally placed eight detector blocks. Each detector block is separated by detector-free regions from each other. These gaps between the detector blocks introduce missing parts to the acquired PET data. Without the estimation of the missing parts of the sinogram data, the methods which require full sinogram dataset give undesired results. Previously, we proposed the iterative discrete cosine transform (DCT) domain gap-filling method which gave better quantitative and qualitative results than the previously published gap-filling methods. However, the gap-filling methods published so far estimate the missing parts only on the transaxial slices while they are ignoring the information from the neighboring transaxial slices. In this study, we propose a non-iterative gap-filling method for the compensation of the HRRT PET sinogram data by using the slices in the direction of radial samples. Therefore, the inter-transaxial slice information is engaged in the gap-filling process. We compared the results of this new method with the DCT domain gap-filling results. We used 3D numerical phantom sinogram at eight different Poisson noise levels for checking the quantitative results. For visual assessment, we employed a [11C]-raclopride human brain study. After the missing parts were estimated, we applied 2D filtered backprojection which obligates full sinogram dataset. The results showed that the non-iterative interpolation by using the slices in the direction of radial samples gave similar results as the DCT domain data estimation method. However, its simplicity and the improved estimation time over the DCT domain gap-filling method are the notable advantageous properties of this non-iterative gap-filling approach.
This work was supported by the Academy of Finland, (application number: 129657, Finnish Programme for Centres of Excellence in Research 2006-2011) and by the Graduate School in Electronics, Telecommunication and Automation (GETA), Finland. All authors are with the Department of Signal Processing, Tampere University of Technology, P.O. BOX 533, FIN-33101, Tampere, Finland (email: [email protected]). M09-221:

L. Li1, H. Hu2, Z. Chen1, K. Kang1, L. Zhang1 1 Department of Engineering Physics, Tsinghua University, Beijing, China 2 Tsinghua University, Beijing, China

Without a Priori Knowledge Solving the Interior Problem in CT Using Two Scans

Recently, region-of-interest (ROI) reconstruction problem has been solved exactly for imaging configurations satisfying some geometrical conditions, especially the interior problem while the ROI is totally contained inside the object. The exact reconstruction of the interior problem has been also proved to be theoretically possible under few restrictions. One of these restrictions is the requirement of a priori knowledge of the object on a limited region inside the ROI. However, this kind of a priori knowledge is not easily satisfied in practice. In the paper, we focus on solving the interior problem without a priori knowledge. A novel data sufficiency condition is presented that unique and stable interior ROI reconstruction can be obtained without a priori knowledge. It needs only the line integrals passing through this ROI and another small FOV2 located outside the object. A reconstruction algorithm is developed which can be considered an extension of the DBP-POCS (differentiated backprojection-projection onto convex sets) method, where we add the total-variation (TV) minimization method into the each iterative step to reduce the noise in the interim reconstruction images. Finally, we present experimental results using real data to 360

illustrate the new data sufficiency condition and the good stability of the algorithm. In practice, the required projection data can be collected by two scans. In each scan, the FOV respectively covers the ROI and FOV2 by adjusting the x-ray beam width with a collimator, which can effectively reduce the detector area and radiation dose.
M09-224:

M. Sibomana1, S. H. Keller1, C. Svarer2, O. V. Olesen1,3, F. Andersen1, S. Holm1, L. Hoejgaard1 1 PET Center, KF 3982, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark 2 NRU, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark 3 IMM, DTU, Copenhagen, Denmark

New Attenuation Correction for the HRRT Using Transmission Scatter Correction and Total Variation Regularization

In the standard software for the Siemens HRRT PET scanner the most commonly used segmentation in the -map reconstruction for human brain scans is MAP-TR. Problems with bias in the lower cerebellum and pons in HRRT brain images have been reported. The main source of the problem is poor bone / soft tissue segmentation in these regions and the lack of scatter correction in the -map reconstruction. In this paper we describe and validate the new TXTV segmentation method included in the HRRT_U 1.0 user software aimed at solving the bias problem. Its improvements fall in two parts: Firstly, it introduces the use of scatter correction, and secondly, it uses a simple threshold in combination with nonlinear total variation -map filtering. To validate the new transmission -map reconstruction with scatter correction and TXTV segmentation, we compared 50 RoIs average on emission images reconstructed with the new and MAP-TR -maps to images reconstructed with co-registered CTbased -map considered a gold standard on four 18F FDG scans. The comparison show a significant decrease in the bias in cerebellum and pons, and a better agreement between images reconstructed with the new method and CT-based -maps.
M09-227:

C.-C. Liu1, Y.-J. Tsai1, C.-H. Hsu2, I.-T. Hsiao1 1 Department of Medical Imaging and Radiological Sciences, Chang Gung University, Tau-yuan, Taiwan 2 Dept of Biomedical Eng. and Environmental Sci, National Tsing Hua University, Hsin-Chu, Taiwan Dual-head planar positron emission tomography (PET) without rotating the camera found its application in breast cancer detection recently due to the properties of high efficiency, high resolution and low cost. However, the image quality for planar PET along depth axis is degraded due to limited angle coverage. One way to alleviate this limitation is applying weighting factor during the reconstruction. The purpose of this work is to explore different weighting approaches in reconstructing the planar PET images. The two weighting methods considered in this work were geometric weighting and data weighting. Geometric weighting factor was computed from geometric LOR ratio within FOV, and data weighting factor was calculated from backprojection of the measured data. The dimension of planar PET system is 12.0cm8.0cm and a varying distance between the detector planes. The total LOR number is 24002400. The projection data was then reconstructed by weighted MLEM with 5, 10, 20, 50, and 100 iterations with non-weighting, geometric weighting, and data weighting, separately. The preliminary results showed that data weighted reconstruction can improve diffused distortion in the depth direction and lead to faster activity recovery in the early iteration. Future work should include various physical effects as well as Poisson noise modeling to further validate the effectiveness of the weighted reconstruction.
M09-230:

Weighted Image Reconstruction in Planar PET

N. Govindarajan1, V. Nazaryan1, C. Keppel1, P. Gueye1, A. Weisenberger2, S. Majewski2 1 Medical Physics, Hampton University, Hampton,VA, USA 2 Radiation Detector and Medical Imaging Group, Thomas Jefferson Laboratory, Newport News,VA, USA

Real Time Imaging and Dose Verification for Intracavitary Breast Brachytherapy

Accelerated Partial Breast Irradiation (APBI) using, for instance, the Hologic MammoSite applicator is becoming popular due to the good dose distribution, increased volume of treatment, improved cosmetic outcome, and relatively short five day treatment procedure. A major concern with this treatment modality is the balloon to skin spacing. If the skin to balloon distance is less than 7mm, the procedure may not be recommended due to adverse skin reactions leading to poor cosmesis. Studies show that this issue may prevent more than 4,000 patients from taking advantage of BCT every year. The skin spacing issue can be overcome by partially shielding the radiation dose to the skin by introducing a thin layer of high Z ferrous material inside the balloon catheter. A slight magnetic field will be applied externally to arrange the powders internally under the segment of surface of the balloon where the skin spacing is less than 7 mm, thereby protecting the skin from radiation damage. The uniformity and the thickness of the metal powder deposition must be measureable and reproducible. For these measurements we inject a Tc99m based imaging agent into a breast phantom, then use a high resolution miniature gamma camera to image activity from the uptake of injected Tc99m to get an outline of the void formed by the afterloader balloon and to also image the activity of the radioactive seed at the center of the balloon. The two images will be fused to get the relative positioning of the cavity and balloon. The relative position of the Ir192 seed to the tissue in the vicinity of the breast cavity will serve as base for calculations of the dose delivered to tissue by using a variation of the standard conjugate planar imaging method. These images may also be used to 361

determine the uniformity, position, thickness of metal shield. This technology will allow for reliable non-invasive dose measurement based on measured in-situ positions of the cavity balloon (with Tc99m), and 192Ir seed.
M09-233:

J. E. Kim1, M. W. Kim2, Y. S. Kim3, K. M. Oh3, M. S. Yun3, C. Kwon2, K. T. Nam4, S. H. Nam5 1 Kaya University, Radiological Science, Gimhae 621-748, South of Korea 2 Inje University, Medical image and Science department, Gimhae 621-749, South of Korea 3 Inje University, Biomedical Engineering, Gimhae 621-749, South of Korea 4 Division for Applied Robot Technology, Korea Institute of Industrial Technology, Cheonan-Si Chungnam 330-825, South of Korea 5 Inje University, Biomedical Engineering and Medical Image Research Center, Gimhae 621-749, South of Korea In order to get a higher resolution of the image for a patient, the direct conversion mode which X-ray converts to electric charge are better than the indirect conversion mode which X-ray converts to light. But, the flat panel X-ray image detectors for direct conversion mode are not yet commercially available. In this study, we fabricated the samples for X-ray image film by using polycrystalline mercury iodide and carried out a feasibility study on the radiotherapy application. A sheet of copper (1 mm thickness) was attached onto the radiotherapy X-ray image film detector by insulating tape. The characteristics of the samples such as X-ray sensitivity, reproducibility, X-ray response on dose and dose rate for 6 MV beam was investigated. Dark currents were measured as a function of applied electric fields. The attenuated fraction of samples exposed with 6 and 10 MV beams was also measured as a function its thicknesses. By data analysis, the reproducibility of electric current measurements was 1.75% in the coefficient of variation. The measured values of electric charge revealed linearity with respect to the dose and dose rate, respectively. On the other hand, as the dose increases, the X-ray sensitivity decreased due to polarization effect. And the attenuated fraction for a 6 MV beam was 0.75% at the sample with the thickness of 380m. Therefore, the fabricated samples for X-ray image film provides the feasibility of radiotherapy imaging application. Moreover, it is found that the thickness of polycrystalline mercury iodide is very important factor for radiotherapy imaging, and a proper thickness having the attenuated fraction of about 1% should be greater than about 400m.
This research was financially supported by the Ministry of Knowledge Economy(MKE) and Korea Industrial Technology Foundation (KOTEF) through the Human Resource Training Project for Strategic Technology M09-236:

Analysis of the signal for high resolution Digital X-ray Detector under Radiotherapy irradiation condition

M. Petasecca1, M. L. F. Lerch1, D. Cutajar1, D. Franklin1, J. Green1, M. Weaver1, J. Jakubek2, M. G. Carolan1,3, M. R. Conway4, S. Pospisil2, T. Kron5, M. Zaider6, A. B. Rosenfeld1 1 Centre of Medical Radiation Physics, University of Wollongong, Wollongong - NSW, Australia 2 Istitute of Experimental and Applied Physics, Czech Technical University in Prague, 12800 Prague 2, Czech Republic 3 Illawarra Cancer Care Centre, Wollongong - NSW, Australia 4 Sydney Eye Hospital Faculty of Medicine, The University of Sydney, Sydney - NSW, Australia 5 Peter MacCallum Cancer Centre, Melbourne VIC 8006, Australia 6 Memorial Sloan-Kettering Cancer Centre, NY, USA Australia is one of the countries most affected by melanoma. Melanoma and squamous cell carcinoma are the most common ocular malignancies in adults. Disease morbidity and cost to the community is significant with almost 50% of patients losing vision or the eye due to the disease and/or the treatment [1,2]. Currently, most popular eye plaques used for treating ocular cancers incorporate I-125 seeds or are uniformly coated with a radioactive material such as Ru106/Rh106. Planning for plaque radiotherapy is typically performed using manual charts or by Treatment Planning Software (TPS) (e.g. BEBIG GmbH, Berlin, Germany). Optimum treatment can only be achieved by customisation of plaques for each patient however, there are currently no procedures with the ability to precisely customize the dose distribution by taking into account factors such as tumour shape, location, and proximity to vital structures. Clinicians also require reliable instrumentation and procedures to verify that desired dose pattern has been achieved for QA, once a proposed plaque configuration has been designed. Experimental dosimetry in ocular radiotherapy, in which tumours are small and can be close to critical structures such as the optical nerve, demands dosimetry systems with high spatial resolution and detector dynamic range due to the very steep dose gradients involved. In this work, a fast and reliable three-dimensional dosimetric imaging technique is tested using an eye phantom. The silicon pixelated detector MEDIPIX2 was used to perform the experiments that consisted of the measurement of several 2D dose distributions across the eye phantom.

Three-Dimensional Dosimetry Reconstruction in Eye Brachytherapy

362

M09-239:

Experimental Validation of a Monte Carlo Model to Predict EPID Images for Online Verification in Radiotherapy

D. Lazaro-Ponthus, T. Lamotte, J.-C. Garcia-Hernandez, S. Legoupil LIST, CEA, Gif-sur-Yvette, France Electronic portal imaging devices (EPIDs) were increasingly used this past decade to provide images which can then be compared to a predicted dose image for pretreatment or in vivo dosimetric verification in external beam radiotherapy. Monte Carlo (MC) simulations were recently shown to be a powerful and reliable tool to predict accurately portal images and are more and more often integrated into treatment planning systems. The goal of the present study is twofold: validate a portal prediction model for open fields against measurements and demonstrate the reliability of the MC code PENELOPE for this purpose. Experimental portal images were acquired at 12 MV using a Saturne 43 accelerator and a fluoroscopic Lynx2D EPID (SDD = 150 cm), for different open fields (4x4, 10x10 and 15x15 cm2) and both without and with an object (30x30x30 cm3 water cubic phantom) in the beam. Firstly, in-air portal images of a 10x10 cm2 open field were simulated for four different geometries of the EPID and on-axis profiles were compared to a measured profile. This comparison validated that the most adapted EPID model was a three-layer model consisting in a copper layer, a GOS layer and a water layer whose thickness was adjusted to correctly reproduce the amount of backscattered radiation observed experimentally within the EPID structure and on the lab walls. Secondly, portal images were simulated using the previously defined model for the different open fields, without and with the water phantom in the beam. The gamma analysis was then used to compare simulated and measured images and on-axis profiles. A good agreement was found both on profiles and images, showing that the model was able to predict reliably portal images for open fields, with and without object in the beam. This study also demonstrated that the MC code PENELOPE is an appropriate tool to model portal images, which can be used to compute EPID images for dosimetric verification and to develop dose reconstruction methods.
Acknowledgments: We would like to thank people of the LNHB laboratory for their useful help to the experimental part of this work and J. Guillon (FIMEL) to provide us with the description of the Lynx2D EPID. This work was funded by MEDICEN, in the framework of the MINIARA project. M09-242:

Designing the Scintillation Module of a Pixelated Gamma Camera : Ths Spatial Spreading Behaviour of Light.

E. Netter, M.-A. Duval, B. Janvier, F. Lefebvre, L. Menard, L. Pinot, R. Siebert, Y. Charon UMR 8165, IN2P3, CNRS, Paris 7 Diderot, Paris 11 Sud, Imagerie en Modelisation, Neurobiologie et Cancerologie, Orsay, France Designing the scintillation module of a mini gamma camera requests good understanding of how the scintillation light signal spreads within the crystal and how photons are finally collected on the PMT. For a given scintillation process, the key parameters are the geometry of the crystal, its optical coating and the interface between the crystal and the PMT. In order to optimize the design of our gamma camera TReCam, we studied the shape of the spatial distribution of the scintillation light for different combinations of these key parameters. Our approach was to produce computer simulations, for which the optical simulation transport software DETECT2000 was used. Three major parameters were investigated : optical coatings for the crystal (aluminum, Teflon, corner cube retro-reflectors), the thickness of the crystal (ranging from 1 mm up to 5 mm) and the interfaces (air, optical grease). In addition to the FWHM of the light spot distribution, specific figures of merit were implemented to further characterize photon spreading. In order to evaluate the influence of these parameters on the imaging performances of the TReCam, the whole scintillation module (including the PMT) was also simulated with GATE v4.0.0 based on GEANT 4.9.1.p02. Here, we report our first results for this extensive study. Teflon was found to deposit 30% more photons compared to other coatings on a longer distance from the hottest pixel, but was much more sensitive to the thickness of the crystal (33% of scintillation photons were lost between 1mm and 5mm, versus less then 2% for others). Using optical grease, 99% of the optical photons were collected onto the PMT, however, the output signal of the light broadened by a factor of 50%, whatever optical coating was used. Further investigations, such as the study of the gamma interaction position must be done before discussing the ideal configuration. All these subjects will be presented and simulated data will be compared to experimental data.
M09-245:

Attenuation Correction of Coils and MR Bed for a Sequential PET/MR System

D. Pal, Z. Hu, N. Ojha, C. H. Tung, J. Kaste Nuclear Medicine, Philips Healthcare, Highland Heights, OH, USA PET/MR is a multimodality imaging system that utilizes MR images for attenuation correction of the PET images. The MR images are co-registered with the PET images and that may introduce several interesting clinical applications. The MR system utilizes several RF coils for MR imaging in addition to the MR bed that cannot be removed during the PET scan to avoid patient movement. These coils are not visible in the MR image and hence attenuation correction of the coils and the MR bed is a challenge. It is proposed to use a template for the rigid portions of the coils and the bed. The template can be introduced during 363

the PET image reconstruction. This work evaluates the need for the attenuation correction of the coils and the bed. The evaluation is performed using a PET/CT scanner utilizing different phantoms such as uniform cylinder and IEC phantom. The evaluation is performed using a CT attenuation map. Ignoring the coil can lead to non-uniformity in the images and global bias in the estimated counts ranging from 3 to 30 % depending on the coil type. However it is demonstrated that CT is not appropriate to generate the template for the coils and the bed due to the lack of a proper scaling from the CT energy to the 511 kev energy, especially for the dense components of the coils and the bed. It is proposed to use a long transmission scan of the coils and the bed to generate the template. The accuracy of this approach is demonstrated using an IEC phantom study with the bed. The PET image reconstructed with the CT attenuation map has artifacts that are removed by using a transmission scan generated map. This approach will be evaluated for all the coils utilized in the PET/MR system.
The authors would like to acknowledge Gary Muswick for helping to acquire the transmission scans and helpful discussions. M09-248:

A Fiducial Marker Based Technique for Alignment of Simultaneously Acquired PET and MRI

Images
J. E. Mackewn, G. Charles-Edwards, E. J. Somer, P. Halsted, M. Kelly, S. F. Keevil, T. Schaeffter, P. K. Marsden Guy's, King's and St Thomas' school of medicine and dentistry, London, UK Combined MRI and PET systems are currently being developed by a number of research groups and also commercially. Most systems have been designed to permit the simultaneous acquisition of MRI and PET data. The approach taken by many groups, is to build an MRI compatible PET insert, that works inside a standard MRI scanner. Unlike other multi-modality imaging systems such as PET/CT, the physical location of the PET scanner within the MRI scanner may vary each time the PET scanner is removed and replaced in the MRI scanner. In order to produce fully aligned PET and MRI images over the same region of the subject, the PET scanner location within the MRI FOV is required, as well as the transformation between the PET and the MRI images. We have developed such a technique for our single slice pre-clinical MR-compatible PET system. The method uses MRI visible markers attached to the PET scanner at a known position with respect to the PET FOV. In our configuration the markers must be positioned close to MRI compatible PET gamma shields which are used to improve the signal to noise ratio of the PET images and result in local distortion of the main magnetic field, although they impose little distortion over the useful FOV of the PET system. We have assessed the accuracy with which the MRI visible markers can be used to select the location of the PET imaging slice and to align the PET and MRI images in-plane, using simultaneously acquired data of various phantoms. The slice location accuracy and the in-plane registration appear to be accurate to <1mm. The method of registration will be used in our preclinical simultaneous PET/MRI studies as it will be essential for situations in which there is little similarity between the PET/MRI images.
M09-251:

S. Yamamoto1, M. Imaizumi2, Y. Kanai2, Y. Hasegawa2, I. Higuchi2, M. Takasawa2, M. Aoki3, E. Sugiyama3, E. Shimosegawa2, J. Hatazawa2 1 Kobe City College of Technology, Kobe, Japan 2 Osaka Univiersity Graduated School of Medicine, Osaka, Japan 3 Hitachi Metal, Neomax Company, Saga, Japan We developed an integrated PET/MRI (iPET/MRI) system for small animals by designing a new MR-compatible PET and a permanent magnet open MRI. In the MRI, a tunnel is opened at the yoke of the magnet. The position sensitive photo-multiplier tube (PSPMTs) of the MR-compatible PET are positioned backside of the yoke where the magnetic field is low enough. Scintillators for the PET system is positioned at the center of the MRI magnets and the scintillation photons are changed its direction by slanted light guides and are fed to the PSPMTs by 75cm long optical fiber bundles. The PET detectors employed two types of LGSO crystals (1.9mm x 2.2mm x 6mm and 7mm) with different decay times (33ns and 43ns) for depth-of-interaction (DOI) detection. Sixteen optical fiber based block detectors are arranged in 112mm diameter ring. The transaxial field-of-view (FOV) of the PET system is ~80mm and axial FOV is 21mm. The axial FOV can be enlarged by a z-motion of the PET detector ring during MRI acquisition. The tranaxial and axial resolution at the center of the PET system was 2.9mmFWHM and 2.4mm FWHM, respectively. Absolute sensitivity was 1.5% at the center of axial FOV. Phantom images showed that there was no artifact in either PET or MRI images. We could successfully obtain simultaneously measured small animal images using the iPET/MRI system. We confirmed that the iPET/MRI system has good performance and can be used for the simultaneous measurements of small animals.

Development of an Integrated PET/MRI System for Small Animals

364

M09-254:

S. Nicol1, S. Karkar1, D. Benoit1, F. Cassol-Brunner1, P. Descourt2, C. Morel1 1 CPPM, Aix-Marseille Universite, CNRS/IN2P3, Marseille, France 2 LaTIM - U650 INSERM, Brest, France

Design and Construction of the ClearPET/XPAD Small Animal PET/CT Scanner

In this study we present a small animal PET/CT scanner based on the prototype high resolution PET scanner ClearPET demonstrator developed within the Crystal Clear Collaboration and the prototype X-ray hybrid pixel detector XPAD3 developed at CPPM. Using the GATE Monte Carlo simulation platform, the complete system has been simulated. Previous measurements and the simulations have both shown the necessity of shielding the PET detectors against object scattered X-rays in order to operate both modalities simultaneously. The final design including the PET detectors appropriately shielded, the XPAD3 detector and a collimated RTW X-ray tube will be presented together with experimental measurements in order to demonstrate the feasibility of acquiring simultaneous data in presence of an X-ray beam scattering on a mouse phantom embedding a Ge-68 point source. Performance of the PET scanner are evaluated with regards to the type and thickness of the shielding material and to the X-ray spectral distribution.
Special acknowledgments to the mechanical design team and mechanical workshop at CPPM that have been seriously involved in the ClearPETXPAD realization. M09-257:

B. Ravindranath1, S. S. Junnarkar2, M. L. Purschke2, S. H. Maramraju1, X. Hong3, D. Bennett3, K. Cheng3, D. Tomasi2, S. S. Southekal1, S. P. Stoll2, J.-F. Pratte4, P. Vaska2, C. Woody2, D. J. Schlyer2 1 Biomedical Engineering, Stony Brook University, Stony Brook, NY, USA 2 Brookhaven National Laboratory, Brookhaven, NY, USA 3 Aurora Imaging Technology, Inc., North Andover, MA, USA 4 Universite de Sherbrooke, Sherbrooke, Quebec, Canada The BNL breast PET system is designed to be a high-resolution PET insert that can operate inside an MRI enabling simultaneous PET-MRI imaging of the breast. A prototype system was assembled and tested successfully for PET image acquisition capability and MRI image quality. We have now assembled prototype II that has 24 2.2 x 2.2 15 mm3 LYSO crystals coupled directly nonmagnetic APD arrays. The detector has an inner diameter of 100 mm and an axial extent of 18 mm. It has a measured average resolution ranging from 1.11-2mm from the center to the edge of its field of view. Details and imaging results of the newly assembled system using a mini deluxe resolution phantom and future plans of obtaining simultaneous PET-MRI measurements using this system are presented.
M09-260:

Results from Prototype II of the BNL Simultaneous PET-MRI Dedicated Breast Scanner

X. Zhang1, F. Chen1, Y. Li1, Q. Wei1, H. Zhang2, Y. Qi1 1 Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai, China 2 Department of Radiation Medicine, the Ohio State University, Columbus, OH, USA

Unified Geometric Calibration and Image Registration for Detached Small Animal SPECT/CT

The aim of study is to develop a unified mathematical optimization method to determine the complete geometric and registered parameters of both cone-beam CT and pinhole SPECT for misalignment calibration and image registration of these detached systems. The common projection equations are derived for cone-beam CT and pinhole SPECT imaging geometry. Complete geometric parameters are determined by point object phantom with a priori relative position information. The optimal configuration of calibration and registration phantom is designed to constrain the optimization problem to converge to the unique solution. The cost function is structured as the least-squares about residual error. The implementation of nonlinear estimation utilizes the medium-scale trust region method. The registration procedure is implemented by the nonlinear optimization similar to the geometric calibration, and five parameters are sufficient to register the detached systems for the image fusion. To validate the method presented, we evaluate the proposed optimization method for geometric calibration and registration with point object (or sources) phantom by computer simulations based on our Micro-CT and Micro-SPECT imaging systems. The preliminary results from the simulation studies show that the unified estimation for geometric calibration and registration is considerable robust. In conclusion, we have presented a unified optimization method to estimate complete parameters for geometric calibration and image registration of detached cone-beam CT and pinhole SPECT. The validity of this method is effective and robust for the imaging systems. The performance of geometric calibration and image registration from experiments studies for our Micro-CT and Micro-SPECT systems will be presented in the full paper.

365

M09-263:

Monte-Carlo Simulations of LYSO-SiPM Modules

J. Pulko, M. Hohberg, S. Fuerst, S. I. Ziegler Nuklearmedizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universitaet Muenchen, Munich, Germany In this paper, a simulated signal from a SiPM detector with 400 cells/mm2 and 1600 cells/mm2 coupled to a 1 x 1 x 20 mm3 scintillation crystal is investigated using Monte-Carlo simulations. A model is introduced in which light is generated in the scintillation crystal by gamma quanta with different energies and transmitted through the crystal, the optical adhesive, and epoxy layer onto the surface of the SiPM. From the photons that reach the active surface of the detector and trigger an avalanche, a mathematical signal, considering the optical crosstalk, afterpulses and recovery time of cells was constructed. Two different aspects of analysis of the constructed mathematical signal were compared, signal integration and peak detection. The analysis concentrated on evaluating the influence of various crystal and SiPM characteristics. Decay constants of the crystal, afterpulse probability, crosstalk probability, and recovery time of the SiPM were varied and their effect on energy resolution of the crystalSiPM combination was quantified. The model introduced in this study was able to reproduce measurement results of two SiPM configurations, thus yielding valuable information for optimizing future detector modules since it can incorporate the specific characteristics of each component.
M09-266:

S. Jeon1, K. S. Song1, D. H. Lee1, B. Kim1, Y. G. Hwang1, Y. Huh1, Y. Park2, N. Jang3, C. Seo4 1 Advanced Medical Device Research Center, Korea Electotechnology Research Institute, Ansan, Kyonggi-do, Korea 2 Research Center, Hanwha L&C, Daejeon, Korea 3 School of Mechatronics Engineering, Changwon National Univ., Changwon, Gyeongnam, Korea 4 Dept. of Media, Soongsil Univ., Seoul, Korea A photon counting based detection and x-ray system for full field digital mammography is designed and optimized in a scanned multi-slit geometry. The detection system described in this work is based on a side illuminated silicon microstrip detector with a specific readout electronics operated in single photon counting mode. The feature of the silicon detector has an n+-n-p+ structure and is equipped with guard-ring. A CMOS front-end for photon counting mode operation contains a preamplifier with automatic compensation for detector leakage current, a discriminator, and counter based in a shift register. The chip is designed and manufactured in a 0.18um CMOS technology. The x-ray system is designed to generate a quasi-monochromatic x-ray beam by using multilayer reflector. The monochromatic spectrum at 26.1keV with a full width at half maximum (FWHM) of 5.7keV is obtained using a general x-ray tube with Mo anode. The reflectivity is about 60% and dispersion angles are 0.2 degree for a length of 100mm reflector. This work concerning the design and optimization of detection system for the scanning full field digital mammography will be described and the preliminary results obtained the prototypes will be presented. Moreover, the feasibility study of quasi-monochromatic x-ray in the mammography field will be performed and investigated.
M09-269:

Design and Initial Performance of Photon Counting Based Detection System for Digital Mammography

B. W. Miller1, J. W. Moore1, L. R. Furenlid1,2, H. H. Barrett1,2, M. E. Gehm3 1 College of Optical Sciences, University of Arizona, Tucson, Arizona, USA 2 Dept. of Radiology Research, University of Arizona, Tucson, Arizona, USA 3 Dept. of Electrical and Computer Engineering, University of Arizona, Tucson, Arizona, USA

Novel Applications of Rapid Prototyping in SPECT, PET, CT, X-Ray, and Gamma-Ray Imaging

Advances in 3D rapid-prototyping printers, 3D modeling software, and casting techniques are allowing for the fabrication of cost-effective, custom components in SPECT, PET, CT, X-ray, and gamma-ray imaging systems. Applications extend to new fabrication methods for custom collimators, pinholes, calibration and resolution phantoms, mounting and shielding components, and imaging apertures. Details of the fabrication process for these components are presented, specifically the 3D printing process, cold casting with a tungsten epoxy, and platinum lost-wax casting.
The Center for Gamma-Ray Imaging is supported by NIBIB Grant P41-EB002035-5. We would like to thank Wei Ren Ng at the Laboratory for Engineering Non-Traditional Sensors (LENS), University of Arizona, for assistance with the Objet Eden350 rapid prototyping printer in aperture/pinhole development and fabrication.

366

M09-272:

Y.-S. Kim1, M.-S. Kim1, S.-H. Jung2, K.-M. Oh1, J.-W. Shin3, S.-H. Nam1,4 1 Bio-Medical Engineering, Inje University, Gimhae 621-749, Korea 2 Medical Imaging&Science, Inje University, Gimhae 621-749, Korea 3 R&D Center of LISTEM, Seoul, Korea 4 Bio-Medical Research Center, Seoul 411-709, Korea

Characteristic Study of Multi-Layer Using Hybrid Method for Digital X-Ray Detector

Recently digital X-ray detector, often called Active-Matrix Flat Panel Detector (AMFPD), have been researched by using either the direct or the indirect conversion method in order to form the digital image. In this paper, we introduced the concept of Hybrid method using phosphor layer to complement a weak point of direct conversion method having low conversion efficiency about X-ray. The Hybrid receptor scheme consists of a photoconductor layer, a phosphor layer and top/bottom electrodes to collect charges produced in the photoconductor layer. A light reflective layer was deposited under the phosphor as lower layer. Mercury Iodide(HgI2) is used as photoconductor layer to detect X-ray and light from the phosphor, and Cesium Iodide(CsI) is used as phosphor layer to convert x-ray to light. We fabricated a photoconductor (HgI2 200) and a phosphor(CsI 50) using PIB(Particle In Binder) method. As top and bottom electrodes, ITO (Indium Tin Oxide 1) layers were deposited to photoconductor layer and light reflective layer(Al 15~25um) in order to improve photon conversion efficiency. As results, higher X-ray sensitivity of Hybrid method exhibited comparing with that of direct conventional digital X-ray detector. The Hybrid method shows about 1.5 times higher sensitivity than the direct conversion method. This effect is caused by simultaneous detection of electric signal induced by the direct X-ray absorption in photoconductor layer and by the light absorption produced in phosphor layer. The Hybrid method can solve problems such as lower conversion efficiency of photoconductor, breakdown because of high voltage application in the direct conversion method as well as low fill fator of TFT, low electric image signals and low SNR in indirect conversion method. Keywords: Digital x-ray detector, photoconductor, phosphor, Hybrid method, PIB
This research was financially supported by the Ministry of Knowledge Economy(MKE) and Korea Industrial Technology Foundation (KOTEF) through the Human Resource Training Project for Strategic Technology M09-275:

Investigation of Multi-Anode Microchannel Plate PMT for Time-of Flight PET

W.-S. Choong Lawrence Berkeley National Laboratory, Berkeley, CA, U.S.A Multi-anode microchannel plate PMTs (MCP-PMTs) are evaluated for time-of-flight (TOF) PET detectors. The primary advantage of an MCP lies in its excellent timing properties (fast rise time and low transit time spread), thus making it a good candidate for TOF applications. In addition, the Planacon MCP-PMT from Photonis has good position resolution provided by its 64 anodes arranged in an 8 by 8 array. The size of each anode is 5.9 mm2 with a pitch of 6.5 mm between the anodes. The Planacon MCP-PMT also has a compact profile and a small dead area around the periphery of 4 mm wide, thus allowing detector modules to be tiled together with minimal dead area. Two different pore sizes of the MCP (10 m and 25 m) are evaluated. Single electron responses such the rise time, pulse width, and transit time spread of the Planacon MCP-PMTs are measured We couple an LSO crystal to the Planacon MCP-PMTs to measure the energy resolution and timing resolution when excited with 511 keV annihilation photons from a Ge-68 source. We compare the results with similar measurements performed with a fast timing conventional PMT (Hamamatsu R-9800).
This work was supported in part by grant No. R21EB007081 from the National Institute of Biomedical Imaging and Bioengineering, and in part by the Director, Office of Science, Office of Biological and Environmental Research, Medical Science Division of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. M09-278:

T. Y. Song1, H. Wu1, S. A. Komarov1, S. B. Siegel2, Y.-C. Tai1 1 Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO, USA 2 Siemens Preclinical Solutions, Knoxville, TN, USA

A SiPM Block Detector for a Sub-Millimeter Resolution PET Insert for the Siemens Inveon Scanner

We report in this work a prototype SiPM-based block detector based for a new PET insert device specifically designed for the Siemens Inveon PET scanner in order to provide sub-millimeter resolution PET images with high system sensitivity. 36 SiPM photosensors were arranged to a 6 by 6 array and optically coupled with a 20 by 20 LSO array through a light guide. The LSO pixel size was 0.80.83 mm3. The package of each SiPM photosensor measures 1.92.40.8 mm3 with a sensitive area of 11 mm2. The SiPM sensors were equally spaced in both directions, and the pitch of sensitive area was 3.07 mm. SiPM block detector was mounted on a readout board consisted of a resistive charge division network and preamplifiers. The 20 by 20 LSO crystals of 0.8-mm pixel size can be resolved in the flood image. Mean and standard deviation of individual crystals energy resolution were 20.4% FWHM and 3.9% respectively. Timing resolution obtained from a previous prototype block detector 367

with 3 by 3 SiPM was 380 ps FWHM. The results demonstrate that the proposed SiPM-based block detector is suitable for the proposed PET insert application and should offer sub-millimeter resolution PET image when it is integrated into a Siemens Inveon PET scanner. Since the SiPM is immune to strong magnetic field, this detector technology can potentially be used for MR compatible PET applications, as well.
M09-281:

S. Motomura1, T. Fukuchi1, Y. Kanayama1, H. Haba1, Y. Watanabe2, S. Enomoto1,3 1 Metallomics Imaging Research Unit, RIKEN Center for Molecular Imaging Science, Kobe, Japan 2 Molecular Probe Dynamics Laboratory, RIKEN Center for Molecular Imaging Science, Kobe, Japan 3 Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan

Three-Dimensional Tomographic Imaging by Semiconductor Compton Camera GREI for Multiple Molecular Simultaneous Imaging

We have investigated the possibility of three-dimensional (3D) tomographic imaging by semiconductor Compton cameras that we have been developing for multiple molecular simultaneous imaging. As a significant feature of Compton cameras, multidirectional projection of the -ray source distribution can be obtained even by a fixed-angle imaging with a single Compton camera unit. We have already implemented a 3D tomographic image reconstruction method applicable for both single- and multiple-unit Compton cameras, and succeeded in 3D tomographic imaging with a single Compton camera unit. In order to enlarge the field of view and to obtain more accurate images, we are constructing an array of multiple semiconductor Compton camera units.
M09-284:

P. J. La Riviere1, Z. Xie2, H. F. Zhang2 1 Dept. of Radiology, The University of Chicago, Chicago, IL, USA 2 Dept. of Electrical Engineering, The University of Wisconsin Milwaukee, Milwaukee, WI, USA There has recently been development of high-frequency photoacoustic microscopy (PAM) systems with demonstrated capabilities to image biological tissue at a microscopic scale. The imaging depths achievable (a few mm) are shallower than in low-resolution photoacoustic imaging, but deeper than conventional optical microscopy. PAM usually employs a focused singleelement high-frequency ultrasonic transducer and a spatially overlapped optical illumination. The integration of PAM with existing optical microscopic modalities such as confocal microscopy, two-photon microscopy, and optical coherence tomography has been hampered because existing PAMs require mechanical scanning of the ultrasonic-optical assembly, which is not compatible with other optical microscopic modalities. Recently one of us (H.Z.) has developed a laser-scanning OR-PAM (LSOR-PAM) to demonstrate the feasibility of employing optical scanning in PAM. A lateral resolution of 7.8 m and a circular field-of-view with a diameter of 6 mm were achieved in an optically clear medium. In LSOR-PAM, the ultrasonic detector is kept stationary and only the laser light is raster-scanned within the FOV during data acquisition. Further improvements in image quality and the development of image quality metrics would benefit from the development of an analytical system model for this imaging device. The present study is focused on the development of such a model and its validation.
M09-287:

System Model for Laser-Scanning Photoacoustic Microscopy

G. Wagenknecht1, E. Rota Kops2, L. Tellmann2, H. Herzog2 1 Central Institute for Electronics, Research Center Juelich, Juelich, Germany 2 Institute of Neurosciences and Medicine, Research Center Juelich, Juelich, Germany

Knowledge-Based Segmentation of Attenuation-Relevant Regions of the Head in T1-Weighted MR Images for Attenuation Correction in MR/PET Systems

Attenuation correction (ATC) is an important prerequisite for quantitative analysis of PET images. In contrast to PET-CT systems, MR images from MR/PET systems do not provide unambiguous values which can be directly transformed into PET attenuation coefficients (AC). The same relaxation- and proton density-dependent intensity values must be assigned to different AC depending on the position and tissue membership of the voxel examined. Thus, the new method aims at segmenting such attenuation-differing regions of the patients head based on its T1-weighted MR data for correct PET-based brain analysis. Attenuation-differing regions of the head are basically brain tissue, extracerebral soft tissue, bone, air-filled craniofacial nasal and paranasal cavities and the mastoid process in the temporal bone. Since air-filled and bone regions cannot be separated by their intensity values, the new knowledge-based segmentation approach utilizes the regions' anatomical position to each other and its rough shape in addition. It consists of 3 steps: 1. tissue classification; 2. separation of the brain and extracerebral region; 3. segmentation of the extracerebral region into soft tissue, the mastoid process, craniofacial cavities, and bone. It was applied to 19 MPRAGE images (9 M, 10 F) acquired at a Siemens 3T MR/PET TimTrio system. Even though the region variability was large in this study, the method yields excellent results in separating brain tissue, extracerebral soft tissue, craniofacial cavities, the mastoid process, skull, and craniofacial bones. 3D bone renderings yielded results looking like CT-based renderings. For one subject, the segmentation quality was compared to CT and PET ATC was compared to measured ATC. Rel. differences 368

calculated from ROI values yield promising first results (mean 1.85% 2.54%). Since the new method is not based on special sequences, templates or atlases to be registered it could be a good alternative to those methods for ATC in brain regions.
M09-290: Simultaneous 3D Imaging of Bone and Vessel Microstructure in a Rat Model: Measurement of Vascular-Trabecular Interdistance

M. Langer1,2, R. Prisby3,4, Z. Peter2,5, A. Guignandon3, M.-H. Lafage-Proust3, F. Peyrin1,2 1 European Synchrotron Radiation Facility, Grenoble, France 2 INSA Lyon; CREATIS-LRMN, CNRS UMR 5220; INSERM U630, Universite de Lyon, Lyon, France 3 INSERM U890, Universite de Lyon, St-Etienne, France 4 Dept. of Kinesiology, University of Texas, Arlington, TX, USA 5 GTE Department, University of Paris 10, Ville d'Avray, France

We developed a new method based on synchrotron radiation microtomography allowing the simultaneous analysis of bone and vessel microstructure. A new parameter for the quantitative characterization of the 3D relationships between the two structures is proposed. It consists in measuring the distance to bone in each point in the vascular structure by means of a 3D discrete distance transform. We define the vascular-trabecular interdistance (VTI) as the mean value of this distance. Calculation of VTI requires data which contains information on the bone and vessel microstructure simultaneously. Classically, parameters are calculated on bone and vessels separate, such as bone volume, degree of mineralization in bone, and vessel thickness. The VTI relates to both bone and vascular structure and can be interpreted as representing the spatial organization of vessels relative to bone. This method is applied to a study of the effect of treatment with parathyroid hormone (PTH) on vascularization in rat models. 3D images of rat femurs posthumously injected with contrast agent and embedded in MMA were acquired using synchrotron radiation mictrotomography. By statistical analysis it is found that treatment with PTH significantly decreases VTI.
This work was supported by French ANR project BLAN06-1_144534 M09-293:

Implementation and Evaluation of a Model-Based Downscatter Compensation Method for Quantitative I-131 SPECT

N. Song, D. Yong, B. He, E. C. Frey Department of Radiology, Johns Hopkins Medical Institution, Baltimore, MD, USA I-131 is an important radionuclide for targeted radiotherapy, and accurate estimates of organ activities are essential in the patient specific treatment planning. However, I-131 has high energy photon emissions (637 keV and 722 keV) which can interact in the patient or detector and be detected in the photopeak energy window. The resulting contribution to the projection data is often referred to as down-scatter and can reduce the quantitative accuracy of I-131 SPECT images. The goal of this study was to develop a model-based downscatter compensation method and evaluate its efficiency in terms of the accuracy and precision of organ activity estimates from quantitative I-131 SPECT. In this study, we combined our previously developed model-based downscatter estimation method with iterative reconstruction-based compensations for patient attenuation (A) and scatter (S) and the full (D) collimator-detector response to form a comprehensive compensation method. The realistic 3D NCAT phantom was used to model the anatomies and organ activity distributions based on real patient scans. For each phantom, 50 sets of noisy projections with independent and clinically realistic noise were generated using a previously validated Monte Carlo simulation code. These projection data were then reconstructed using the proposed compensation method. We computed the mean and standard deviation of the errors in the activity estimates for organs (heart, lung, liver, kidneys, spleen and bone marrow) from the reconstructed SPECT images and compared them with the results from SPECT images reconstructed with only A, S and D compensations. We found that 29% of the total counts in the I-131 photopeak energy window resulted from downscatter contamination. This downscatter resulted in overestimation of organ activity estimates. After downscatter compensation the mean error of activity estimates were less than 5% for heart, liver and kidneys with the residual errors resulting from partial volume effects.
M09-296:

Effects of Defects on Partial Volume Compensation in Emission Computed Tomography Cardiac

Imaging
Y. Du, E. C. Frey Department of Radiology, Johns Hopkins Medical Institutions, Baltimore, MD In emission computed tomography, partial volume effects (PVEs) caused by the limited spatial resolution reduce image quality and quantitative accuracy. We have developed a partial volume compensation (PVC) method for cardiac imaging using VOI maps defined from x-ray CT images. However, x-ray CT cannot identify myocardial uptake defects and thus provides a single myocardial volume-of-interest (VOI). The goal of this work was to study how accuracy of PVC is affected by the presence of uptake defects. The 4D XCAT phantom was used to generate phantoms with cardiac defect in the anterior, inferior, lateral, or 369

septal wall. Three defect to normal activity ratios, 0.11, 0.5, and 0.75, were used. The images were blurred to model SPECT and PET spatial resolution. Two PVC strategies were used. In PVC1, both defect and normal regions were included in the same VOI. In PVC2, the defect and normal regions were treated as independent VOIs. As expected, PVEs caused underestimation of the normal myocardial activities. Both PVC1 and PVC2 greatly reduced the normal region PVEs, and activity estimates were close to the truth. The errors of defect activity estimates strongly depended on the defect severities. Severe and mild-to-moderate defect produced over- and under-estimation, respectively. The defect/normal ratios were also overestimated due to PVEs with the severe defects having the largest error. PVC1 improved the estimates for mild, but not moderate and severe, defect. PVC2 greatly improved the activity estimates for all degrees of defect. Both PVC1 and PVC2 improved estimates of defect/normal ratios with the best results given by PVC2. In conclusion, the PVEs in myocardial defects were different from those in normal regions and depended on defect severity. PVC can reduce PVEs in both defect and normal regions. However, the improvement will be limited if the defect and normal regions are included in same VOI. The best PVC strategy is to treat defect region as an independent VOI.
M09-299:

R. Taschereau1, Q. Bao1, F. Rannou2, A. Chatziioannou1 1 Molecular and Medical Pharmacology, University of California Los Angeles, Los Angeles, CA, USA 2 Departamento de Ingenieria Informatica, Universidad de Santiago de Chile, Santiago, Chile

Video Image Based Attenuation Correction for PETBox, a Preclinical PET Tomograph

PETBox is a new simplified bench top PET scanner for pre-clinical imaging. Because PETBox detectors can collect very oblique emission angles, attenuation correction is necessary. We have developed a method by which two optical views obtained from video cameras located on the top and side of the PETBox are combined to create a 3-dimensional estimate of the subject that is used to produce attenuation data. The top camera is located at about 8 cm from the mouse with a varifocal (4-12mm) lens. The side camera is about 4 cm from the mouse with a 2.5 mm focal lens. We first apply barrel-distortion correction to the side image to fix deformation caused by the wide-angle lens. Both images are then segmented into mouse and background regions. To facilitate segmentation and distinction between all types of mice, a backdrop of colored translucent acrylic is placed on the side of PETBox. A 3D volume representing an approximation of the subject is created by replicating each of the two images in one direction: top-down direction for the top-view image, and left-right direction from the side-view image. A final 3D image is created with voxels belonging to the intersection of both projections. We have also studied a second variant of this method in which a third view (elliptical cross section) is used and replicated along the long-axis of the mouse to further refine the final shape. These two variants are compared to a reference attenuation-corrected file. The reference attenuation-corrected file shows corrections up to 60%, justifying the need for this type of correction. A comparison of the two variants shows differences up to 18% (first variant) and 7% (second variant) with respect to the reference file. The second variant offers a better match, with most organs within 3%. In both variants, the lungs have the largest error (30%) because the model over corrects (assumes water instead of air). This can be taken into account when reporting activity for that organ.
M09-302:

M. Ljungberg1, K. Sjostrand2, J. Richter2, L. Edenbrandt2 1 Medical Radiation Physics, Lund University, Lund, Sweden 2 Exini Diagnostics AB, Lund, Sweden

Quantification of Skeletal Involvement from Whole-Body Imaging: Validation by Monte Carlo Simulations

Quantification of metastatic bone involvement can be an important prognostic indicator of survival or useful for treatment response monitoring. An automated computer program for analysis of whole-body (WB) bone scan images, determining skeletal involvement of metastases was developed. The methods involves estimating volumetric structures (metastases and skeletal structures) from planar images, an ill-posed problem. We describe a method for validating that the automated program is sufficiently accurate based on Monte Carlo simulations. A computer XCAT phantom was used to simulate male and female patients with different degree of metastatic disease. We used the Monte-Carlo program SIMIND to simulate WB bone studies with a scintillation camera system equipped with a LEGP collimator. The simulations mimic common clinical 99m-Tc MDP procedures, both anterior and posterior images where produced. Three metastases containing 115 voxels each was placed in each of 14 unique skeletal regions (e.g. skull, pelvis, humerus). Metastases was assigned a high, a medium and a low intensity. High defined a value close to the 80th percentile of the empirical distribution of maximal metastasis intensities calculated from a training set, low was close to the 20th percentile and medium close to the median intensity. The total number of skeleton voxels in the phantom was calculated together with the size of the simulated metastases (expressed as skeletal involvement). The true skeletal involvement for all simulated metastases was 0.019%. The mean calculated involvement for all metastases classified as low, medium and high was 0.018%, 0.022%, and 0.030% respectively. 3/42 metastases were not detected due to regionally unusual intensity distributions. The results provide adequate information for determining the accuracy. Simulation of scintillation imaging using digital phantoms provide a controlled environment for validating processing methods on realistic images with known characteristics.

370

M09-305:

Fundamental Phantom Experiments and Simulations Aimed at Quantification of Myocardial Blood Flow Using 3-D PET

Y. Hirano, H. Watabe, K. Koshino, H. Iida Department of Investigative Radiology, Advanced Medical Engineering Center, National Cardiovascular Center Research Institute, Suita, Oska, Japan 2D-Dimension Positron Emitter Tomography (PET) is useful for the quantification of myocardial blood flow. On the other hand, 3-D PET is difficult to use for the purpose due to the scatters and randoms (accidental coincidence) much lather than these of 2D. Though there are some theories of scatter correction, these are not considered scatters originated from outside field of view (FOV). However, we should use 3-D PET which can reduce the inspection time or dose. The purpose of this study is to figure out how much contribution to the quantification by scatters or randoms and these generation process. First, in order to know how many scatters or randoms, phantom experiments of four cases have investigated. Cylindrical phantom with cold volume is placed in FOV (Case1). The diameter of phantom is 20 cm and the length is 20cm. Additional cylindrical phantoms are placed in outside FOV (Case2). Case3, 4 are same as Case2 but Pb shields (thickness 6mm) are put under or on the outside FOV phantoms to suppress the outside FOV activity. Next, Monte Carlo simulation are performed with almost same geometries after the simulation reproduce the experiments. From result of experiments and simulations, influence of outside FOV activity can be clarified. Finally another Monte Carlo simulation is performed with numerical human model aiming at actual inspections. This simulation is focused on cardiac PET. The heart placed in center of FOV. We assumed activity is distributed uniformly on whole body. From these experiments and simulations, contribution to the reconstructed image or quantification is small, especially in case of introduction of shielding system. This encourages us to use conventional scatter corrections. But we must design protocols for each inspection, because tracer and activity of actual inspections are different from this study. Then, result of this study or simulation techniques will help us to design protocols.
M09-308:

R. Prasad1, M. R. Ay2, O. Ratib1, H. Zaidi1 1 Radiology/Nuclear Medicine, Geneva University Hospital, Geneva, Switzerland 2 Department of Medical Physics, RCSTIM & Tehran University of Medical Sciences, Tehran, Iran

CT-Based Attenuation Correction on the FLEX Triumph Preclinical PET/CT Scanner

Positron Emission Tomography (PET) imaging has emerged as a very valuable molecular imaging modality for quantitative measurements of biochemical processes in the clinical and preclinical imaging domains. However, PET imaging suffers from various physical degrading factors including photon attenuation, which can be corrected using CT-based attenuation correction (CTAC) on combined PET/CT systems. The attenuation map is calculated by converting CT numbers derived from low-energy polyenergetic x-ray spectra to linear attenuation coefficients at 511 keV. In this study, we implemented and evaluated tube voltage (40-80 kVp) dependent attenuation correction on the FLEX Triumph preclinical PET/CT scanner. The impact of both bilinear and quadratic polynomial calibration curve methods obtained at different CT tube voltages was assessed. Correlation coefficients and best fit equations were calculated for both techniques. Phantom and rodents PET images were used to access image quality and quantitative accuracy. It was observed that the slopes of the bilinear calibration curves for CT numbers greater than 0 HU increase with increasing tube voltage. Significantly higher correlation coefficients were obtained using the quadratic polynomial calibration curve method. Acquired phantom and rodents PET data were successfully corrected for photon attenuation using the CTAC procedure thus enabling more accurate quantitative analysis of biochemical processes in vivo.
This work was supported by the Swiss National Foundation grant No 31003A-125246. M09-311:

F. P. Vidal1,2, D. Lazaro-Ponthus2, S. Legoupil2, J. Louchet1,3, E. Lutton1, J.-M. Rocchisani1,4 1 APIS, INRIA Salcay -- le-de-France, Orsay, France 2 LIST, CEA Saclay, Gif-sur-Yvette, France 3 Artenia, Chtillon, France 4 UFR SMBH, Paris XIII University & Avicenne hospital, Bobigny, France Fully 3D tomographic reconstruction in nuclear medicine requires high computing power and leads to many challenges. The trend today is to use more general methods that can integrate more realistic models (application-specific physics and data acquisition system geometry). To date, the use of such methods is still restricted due to the heavy computing power needed. Evolutionary algorithms have proven to be efficient optimisation techniques in various domains, including medicine and medical imaging. However the use of evolutionary computation in tomographic reconstruction has been largely overlooked. In previous work, we showed that an artificial coevolution strategy (also called ``Parisian evolution'') based on the ``fly algorithm'' can be used to reconstruct the 3D distribution of radioactive emitters in Single Photon Emission Computed Tomography (SPECT). In this abstract, we propose a computer-based algorithm for fully 3D reconstruction in Positron Emission Tomography (PET) based

PET Reconstruction Using a Cooperative Coevolution Strategy

371

on the same approach and evaluate its relevance. Realistic models describing the physics of PET could be integrated in the reconstruction loop while taking advantage of artificial evolution to reduce the computing time.
This work has been partially funded by Agence Nationale de la Recherche (ANR), France. M09-314:

Simple and Robust Energy-Based Scatter Correction from List-Mode PET Data

W. A. Worstell, L. V. Romanov, H. A. Kudrolli, P. Kulinich, J. D. Nevin, T. S. Toole, C. Worth PhotoDetection Systems, Inc., Boxboro, MA We have implemented an energy-based scatter correction approach for PET which operates on list-mode data including event energies, and which makes use of prior knowledge of unscattered 511keV photon event spectra from an earlier calibration. The calibration spectra are further adjusted based on singles event spectra to compensate for small energy shifts and resolutionbroadening effects (such as event pile-up) in a given dataset. The proposed method is simpler than previously presented multipleenergy-window methods in that it spectrally separates events associated with sets of lines-of-response into just two species, unscattered pair (true) events and events containing at least one scatter (scatters). The sets of lines of response for which combined spectra are collected, and for which the two species are separated, correspond to those collected by a given pair of block detectors within the scanner. This data space segmentation matches the natural granularity of the energy-domain response function for a scanner made up of block detectors. The resultant algorithm is robust against gain shifts and calibration inaccuracies which may otherwise limit the accuracy of energy-based scatter correction methods for PET. Model-based scatter corrections for PET are typically long on computationally demanding radiation transport calculations (usually with simplifying approximations with associated inaccuracies) and short on incorporating device response characteristics which distinguish real systems from idealized models. By basing PET scatter correction on observable spatially-variant energy spectra, we accurately incorporate out-of-field activity and multiple scatter effects without any fitting or scaling to match lines of response outside the object being imaged. The resultant separated spectra can then be used in an energy-dependent weighting strategy so as to maximize the signal-to-noise of the resultant image.
M09-317:

R. Clackdoyle1,2, D. Ghosh Roy1, M. Defrise3, C. Mennessier1, M. S. Ould Mohamed1 1 CNRS Laboratoire Hubert Curien, Jean Monnet University, Saint Etienne, France 2 Dept. of Radiology, University of Utah, Salt Lake City, Utah, USA 3 Div. of Nuclear Medicine, Free University of Brussels, Brussels, Belgium

Two-Dimensional Region-of-Interest Reconstruction: Analyzing the Difference Between Virtual Fanbeam and DBP-Hilbert Reconstructions

This work addresses theoretical advances classical (2D) tomographic image reconstruction. During the past several years, inversion formulas have been established that allow ROI reconstruction from incomplete (yet sufficient) data. Such reconstructions have important consequences in certain practical situations, such as truncated projections. The precise relationship between the largest ROI that can be reconstructed and the incompleteness of the sinogram is a complex question which has still not been completely answered in the 2D case. These relationships are inherent to the system and have consequences for iterative/statistical reconstruction methods, because they describe which part of the reconstructed image is determined completely by the data; the other parts of the image will have been more heavily influenced by the regularisation method or by the nature of the objective function. Our understanding of the nature of reconstruction from incomplete yet sufficient data relies mainly on formulas obtained from the virtual fanbeam (VFB) method and from the DBP-Hilbert method. The purpose of this work is to provide a structure in which to examine the inherent differences in these two approaches. Using a common reconstruction problem, we reformulate VFB and DBP-Hilbert reconstruction formulas into weight functions that are applied in the sense of an inner product to the sinogram. A common regularization is used for the Hilbert transform in both methods. Unlike the usual Fourier windows used in analytic methods, the regularization we used is applied locally to the singularity to avoid the regularization obsuring the nature of the reconstruction. The weight functions clearly show how truncated projections are being correctly handled. The dissimilarity in the weight functions of the two methods illustrates fundamental differences in managing incomplete data, and suggests that many other such methods exist.
This work was partially supported by grant number NT05-1 45428 from the Agence Nationale de la Recherche, France M09-320:

Iterative Blob-Based Super-Resolution Reconstruction with Wavelet Denoising

E. Y. T. Ho School of EEE, Nanyang Technological University, Singapore, Singapore It has been shown previously that incorporating blob-based basis functions into super-resolution reconstruction can guarantee better results and save computational time, and we are able to use a lower number of low-resolution datasets for the super372

resolution reconstruction. Although blob-based basis functions are effective on suppressing image noise during the reconstruction compare with the ordinary pixel-based reconstruction, this is only limited to image reconstruction from low-resolution datasets which are not excessively corrupted with random noise. For raw image datasets with excessive noise, we can use wavelets to perform pre-processing noise reduction before the datasets are used for the super-resolution reconstruction. Wavelet denoising can effectively separate image noise from useful image features which is sometimes necessary for pre-processing before image reconstruction.
M09-323:

S. Angeli1,2, E. Stiliaris1 1 Dept. of Physics, National & Kapodistrian University of Athens, Athens, Greece 2 Dept. of Bioengineering, Imperial College, London, UK

An Accelerated Algebraic Reconstruction Technique Based on the Newton-Raphson Scheme

The idea presented in this paper is based on the Newton-Raphson root-finding method. The proposed algorithm follows the iterative approach of the traditional Algebraic Reconstruction Technique (ART) with the introduction of a new correction methodology, similar to the Newton-Raphson scheme. The definition of the derivative in this method causes a quadratic rate of convergence speed, which results to a respectable drop of the number of iterations needed to minimize the quadratic deviation. The major issue was the definition of a cost function and its derivative, the equivalent root of which would lead to the accelerated reconstruction of the image. It contains the squared difference of the measured and the reconstructed projections in the appropriate matrix notation. Using this function in a Newton-Raphson based ART (NR-ART) the reconstruction time reduced drastically (at least one order of magnitude) for a 128x128 pixels image. Apart of the formalism, the quality of the proposed reconstruction and its convergence with respect to the traditional ART is discussed in this paper.
The financial support by the program KAPODISTRIAS (Special Account for Research Grants) of the National and Kapodistrian University of Athens is gratefully acknowledged. M09-326:

Selection of Point Source Positions in Small Animal PET for System Matrix Generation

J. Schirmer, S. Ziegler Nuclear Medicine, Klinikum rechts der Isar, Munich, Germany The system response matrix required by iterative reconstruction algorithms can be obtained by physically measuring the system response to a collimated point source or by calculations using Monte Carlo simulations or analytic models. For the measurement of the system response, it is important to determine the position and number of point source measurements required to obtain an accurate system matrix. The aim of this work is to investigate how changes in position and number affect reconstructed images. This study is performed with Monte Carlo simulations of a small animal PET tomograph in 2D. Two million annhilation events at each voxel are simulated for one quarter of the field of view (FOV), which has dimensions 70 x 70 voxels and a voxel size of 0.5mm2. System matrices are constructed by taking into account the information at every simulated voxel (4900 voxels), 2nd voxel (1225 voxels), 5th voxel (196 voxels), 7th voxel (100 voxels) and 10th voxel (49 voxels). In the sparse system matrices, the information at the missing voxels is obtained by performing a linear interpolation. The effect of losing voxels (i.e. measurements) is investigated by reconstructing the uniformity and recovery coefficient phantoms of the NEMA Standards Publication NU 42008. Both uniformity and recovery coefficients decrease as the information from voxels is lost. The percentage standard deviation of the uniform phantom changes from 2.64% (every voxel) to 3.98% (every 10th voxel) at two MLEM iterations. As well, a substantial decrease in image quality is seen when every 5th, 7th or 10th voxel is considered. A simple linear interpolation could not recover the missing information. Further techniques for interpolation are currently being investigated, such as the use of a DRF analytic model to predict the optimum position and number of measurements required for accurate system matrix generation.
M09-329: Optimum Real-Time Reconstruction of High-Rate Gamma Events for Anger Cameras, with the Use of a GPGPU

S. Pedemonte1, A. Gola1,2, A. Abba1, C. Fiorini1,2 1 Dipartimento di Elettronica e Informazione, Politecnico Di Milano, Milano, Italy 2 INFN, sez. di Milano, Milano, Italy

Aim of the HICAM project is to build a new, compact gamma-ray imager with a submillimeter spatial resolution, based on the Anger Camera principle. The system is composed of a detection module with front-end ASICs, an acquisition board and a host PC, where real-time data processing and image reconstruction is implemented. The detector is based on an array of 100 Silicon Drift Detectors (SDDs) of 1 cm2 each in a 10 x 10 cm2 format, coupled to a single scintillator crystal. The position of the interaction with the crystal and the energy of the incident radiation are obtained through the use of a MLE algorithm, that optimally exploits the information given by the detectors. Moreover it is possible to modify the algorithm in order to determine 373

the depth of interaction of the gamma photon into the crystal. The MLE algorithm, on the other hand, requires a large amount of calculations per event. In order to process the events in real-time we have implemented the MLE algorithm on a GPGPU, obtaining a processing rate of 150000 events/second, considering a FOV of 512x512x10 points (calculation of the z coordinate is performed). In the paper we discuss the derivation of the algorithm, its performance for what concerns spatial resolution and distortion and the speed of its implementation on the GPGPU.
M09-332:

N. Belcari1,2, F. Attanasi1,2, G. Franchi3, S. Moehrs1, V. Rosso1,2, A. Santos4,5, F. Spinella2, G. Sportelli4,5, A. Del Guerra1,2 1 Department of Physics, University of Pisa, Pisa, Italy 2 INFN, Pisa, Italy 3 AGE Scientific, Capezzano Pianore, Lucca, Italy 4 CIBER-BBN, Zaragoza, Spain 5 BIT, Universidad Politecnica de Madrid, Madrid, Spain We have developed a flexible and expandable acquisition system specifically designed to work with modular detectors. As a first application the system will be used for the data acquisition of a dual head PET system for positron emission mammography (PEM) where each head is subdivided into independent detector modules. A first advantage of this modular approach consists in the spreading of the coincidence events, and then the data flow, among the modules, thus reducing both the system dead time and the probability of electronic pile-up in the detector head. The PEM system we have developed consists of two planar heads (10cmx10cm each), each head made up of a matrix 2x2 independent detector modules. Each module is a square 64 anodes photomultiplier tube coupled to a matrix of 23x23 LYSO crystals (1.9mmx1.9mmx16mm pixel, 2.0 mm pitch). A multiplexed system based on resistive network with the 88 input and 4 Anger-like output signals signals is used for the readout of each PMT. The signals are then pre-amplified and sent in parallel to a FPGA-based DAQ board that is connected to a mainboard with a main FPGA. Data are then transferred to a host PC via a single USB 2.0 connection. The system is able to work with up to 9 + 9 modules. The coincidence system is based on a combinatorial network designed to detect the time coincidence and to trigger the acquisition system. In our system each module of a single head is put in time coincidence with every module of the opposing head. The network is also designed for random correction using the delayed window technique. The PEM and the new acquisition system will be described. The performance of the acquisition system, with a special attention to count rate capability, random measurement and pileup rejection will be also presented. The possible use of this acquisition system for other geometries such as ring or rotating scanners, as well as SPECT systems, will be discussed.
M09-335:

A Flexible Acquisition System for Modular Dual Head PET Systems for Positron Emission Mammography

fMRI Data Analysis Using a Novel Clustering Technique

J. M. Gorriz University of Granada, Granada, Spain On behalf of the SiPBa group We present a novel method for the analysis of fMRI data. It is based on a model-based clustering approach and the expectationmaximization (EM) algorithm for Gaussian Mixture Model (GMM) fitting. Our method includes the grey-level of each voxel in the process unlike other proposed approaches which only perform clustering on a set of activation maxima by means of the Activation Likelihood Estimation. By considering only the intensity levels inside the clusters, where each cluster is then represented by a single Gaussian with a certain center, shape and weight, the resulting feature space has a significantly reduced dimensionality with respect to former analysis approaches using the voxel intensities directly as features. With this feature extraction method one avoids the so-called small sample size problem that suffers multivariate approaches for functional imaging analysis. In addition, this type of procedure for dimensional reduction has been successfully apply to the problem of medical diagnosis. Therefore, this technique is particularly applicable for the diagnosis of neuro-degenerative diseases like for instance Alzheimer's disease.
Thanks to The Wellcome Department of Cognitive Neurology for providing the database used in this paper. This work was partly supported by the MICINN under the PETRI DENCLASES (PET2006-0253), TEC2008-02113, NAPOLEON (TEC200768030-C02-01) and HD2008-0029 projects and the Consejera de Innovaci\'{o}n, Ciencia y Empresa (Junta de Andaluca, Spain) under the Excellence Project (TIC-02566). Furthermore it was supported by a fellowship within the Postdoc-Programme of the German Academic Exchange Service (DAAD).

374

M09-338:

C. D. Winant1, C. Mari Aparici1, S. Bacharach1, B. W. Reutter2, A. Sitek3, G. T. Gullberg2 1 Department of Radiology, UCSF, San Francisco, CA, USA 2 Life Sciences Division - Center for Functional Imaging, LBNL, Berkeley, CA, USA 3 Department of Radiology, Brigham and Women's Hospital - Harvard Medical School, Boston, MA, USA

Analysis of Dynamic SPECT-CT Measurements of the Arterial Input Function in Human Subjects

Measurement of the Arterial Input Function [AIF] is essential to deriving quantitative estimates of regional myocardial blood flow using kinetic models. Accurate measurements have been possible with a wide range of radiotracers in both research and clinical PET-CT imaging. However, accurate measurements of the AIF with dynamic SPECT or SPECT-CT have posed various challenges; foremost being that imaging a rapidly evolving radiotracer distribution with a slowly-rotating single- or dual-head SPECT scanner yields temporally inconsistent projection data. Most image reconstruction algorithms operate on the assumption of temporally consistent projection data. Other non-negligible artifacts arise from the relative low count rates, increased attenuation, and greater collimation-induced blurring in SPECT images compared to PET. We present a measurement of the AIF in human subjects from dynamic SPECT-CT measurements of 99mTc-tetrofosmin concentration in the left atrium imaged with a Philips Precedence SPECT-CT scanner. The 2-minute infusion is imaged in a series of eight back-to-back 180-degree continuous-mode acquisitions (or rotations), with the dual camera heads. A set of 36 projections (128 by 128 3.19-mm2 pixels) is acquired in each rotation each over a time span of 54 seconds yielding a total acquisition time of 438 seconds. We compute the AIF with traditional image-based analysis and discuss the error induced by inconsistency and other effects with two validations. The first method derives SPECT-like dynamic (inconsistent) projection data from selected forward projections, chosen from modeled SPECT acquisitions, of existing dynamic 94Tc-MIBI PET images. The second validation method uses a database of SPECT measurements of an anthropomorphic phantom to generate both PET- and SPECT-like projections. We discuss potential improvements in measuring the AIF using projection-based spatiotemporal algorithms.
This work was supported in part by the University of California Discovery Grant Program (Contract DIG06-2110-UC) and Philips Medical Systems. M09-341:

Simulation Study on the Diverging SPECT System with a Semiconductor Detector

T. Kusayanagi, K. Ogawa Hosei University, Faculty of Science and Engineering, Tokyo, Japan Single photon emission CT (SPECT) systems with a semiconductor detector have several advantages compared with a SPECT system with a NaI(Tl) scintillation detector in terms of energy resolution and spatial resolution. On the other hand, one of the disadvantages of the detector is a small field-of-view (FOV), and thus we proposed a new SPECT system with a diverging collimator to enlarge the FOV. In this SPECT system with the diverging collimator, we should consider a truncation problem in the axial direction. And to cope with this problem, we adopted an FBP-type cone-beam reconstruction algorithm with Radon space interpolation, which was proposed by Yang et al. in 2006. This method reconstructs more accurate images than Feldkamp, Davis, Kress (FDK) method or Grangeat's method. In this study we evaluated the images reconstructed with Yang's method applied to the data acquired with a diverging collimator. The results obtained with simulations showed the feasibility of the SPECT system with a semiconductor detector and a diverging collimator.
M09-344:

T. Carlier1,2, P. Descourt3, L. Maigne4, D. Visvikis3, M. Bardies2, I. Buvat5 1 Dept. of Nuclear Medicine, University Hospital - Nantes Hotel Dieu, Nantes, France 2 U892, Inserm, Nantes, France 3 U650, INSERM, Brest, France 4 Dept. of Corpuscular Physics, CNRS, Clermont-Ferrand, France 5 IMNC UMR8165, CNRS, Orsay, France To speed up Monte Carlo simulations of Single Photon Emission Computed Tomography (SPECT) scans using iodine-131, a tabulated modeling of the detector response has been incorporated within the GATE simulation toolkit based on the use of the Angular Response Function (ARF). In this work, we validate the ARF methodology within GATE for I-131 simulations and demonstrate the practical feasibility of the simulation of I-131 SPECT patient acquisitions. Methods: The Siemens Symbia T equipped with a high energy collimator was considered. Planar acquisitions of I-131 point and plane sources in air were simulated using GATE without (sGATE) and with the ARF model (ARF-GATE). Profiles through the projections and root mean square differences (RMSD) between ARF-GATE and sGATE projections were calculated. The statistical distributions of the simulated projections were also investigated. A I-131 Lipiocis SPECT scan was also simulated and the simulated and acquired projections were compared. Results: Profiles across the ARF-GATE et sGATE projections agreed well for the point and plane sources, with RMSD of 9%, similar to those obtained between two independent sGATE projections. The ARF-GATE and sGATE projections were Poisson distributed. About 36 times less photons were needed with ARF-GATE than with sGATE to get 375

Efficient Simulations of Iodine 131 SPECT Scans Using GATE

images of equivalent statistical quality for the plane source. Overall, ARF-GATE produced images indistinguishable from the sGATE images in 140 less time. For the patient simulations, simulated projections of visually comparable quality as acquired projections were obtained in 100000 s. The expected computational time efficiency was estimated at 90 when using ARF-GATE instead of sGATE. Conclusions: GATE including the ARF model makes it possible to speed up GATE simulations by a factor ~ 100 without loss of accuracy. Simulations of patient I-131 SPECT scans become feasible in about 2 days or less using reasonable computational resources (small cluster with at least 20 CPUs).
M09-347:

S. J. Lee1, J. S. Baba1, J. S. Goddard1, A. G. Weisenberger2, S. Stolin2, J. McKisson2, M. F. Smith3 1 Oak Ridge National Laboratory, Oak Ridge, TN, US 2 Thomas Jefferson National Accelerator Facility, Newport News, VA, US 3 University of Maryland, Baltimore, MD, US

Effects of geometric and motion tracking errors on awake small animal SPECT

A series of simulation studies were performed to evaluate effects of geometric and motion tracking errors for an awake animal SPECT system. List-mode SPECT data generated with a custom Monte Carlo program and experimental mouse motion SPECT data were reconstructed by MLEM with Siddon's ray tracing. To better understand the impact of motion tracking and geometric parameter errors on system reconstructed data, up to a 1 mm offset was applied separately to each for evaluation. Without any motion tracking or geometric error, the motion compensated reconstruction successfully generated volumes without any degradation or distortion. The results show that motion tracking errors propagate through the SPECT reconstruction. However, the accuracy of the current motion tracking system (0.1mm x,y,z, 0.1 degree rotational angles) minimally and negligibly affects motion compensated reconstruction. In addition, the results indicate that geometric error both directly and indirectly impacts motion compensated reconstruction quality and that an error in pinhole transaxial position determination produces the most distortion among all cases investigated. Furthermore, geometric errors are revealed to have more impact on reconstruction quality than equivalent tracking errors.
M09-350:

L. Volokh1, Y. Grobshtein1, H. H. Barrett2, M. A. Kupinski2 1 Molecular Imaging, GE Healthcare, Haifa, Israel 2 College of Optical Sciences, The University of Arizona, Tucson, AZ

Singular Value Decomposition Analysis of a Novel Clinical Ultra-Fast Cardiac SPECT Camera

Myocardial perfusion imaging is currently the most frequently used nuclear medicine test. In this study, we analyze the new ultra-fast cardiac system Discovery NM 530c developed by GE Healthcare. The Discovery system uses an array of Cadmium Zinc Telluride (CZT) pixilated detectors and a multi-pinhole collimator, simultaneously imaging all cardiac views with no moving parts during acquisition. High system sensitivity together with high intrinsic resolution of the CZT detectors allow acquiring high quality SPECT studies within short time. In this work, we pursue analysis of the system using singular value decomposition techniques. SPECT imaging systems are often modeled by a linear operator H that maps from voxels in object space to detector pixels in measurement space. Practically relevant choices in the number of voxels used to represent discretized object space and the number of pixels used to measure image space make the H large. As a result, for conventional rotational SPECT imaging systems using parallel collimation, implicit representation of the mapping provides basis for projection and back-projection operators used in the course of tomographic reconstruction. For the Discovery NM 530c system, we calculate and store the H matrix in sparce format. The singular vectors and singular value spectra of H provide useful tools for assessing system performance. At the same time, large size and sparseness make conventional singular value decomposition (SVD) methods impractical. For SVD analysis of the Discovery system we used an iterative power methods SVD algorithm (Lanczos) designed to operate on very large sparsely stored matrices to calculate the singular vectors and singular values. We present results of the SVD of the system and arguments on usability of higher-order values. Finally we demonstrate measurement and null space calculations for several relevant objects.
M09-353:

P. Crespo1,2, J. Reis1,3, M. Couceiro1,2, N. Ferreira4,5, R. Ferreira Marques1,3, P. Fonte1,2 1 Physics Department - University of Coimbra, LIP - Laboratrio de Instrumentao e Fsica Experimental de Partculas, Coimbra, Portugal 2 Department of Physics and Mathematics, ISEC - Instituto Superior de Engenharia de Coimbra, Coimbra, Portugal 3 Physics Department - University of Coimbra, Faculty of Science and Technology, Coimbra, Portugal 4 Faculty of Medicine - University of Coimbra, Institute of Biophysics-Biomathematics, Coimbra, Portugal 5 University of Coimbra, ICNAS - Instituto de Cincias Nucleares Aplicadas Sade, Coimbra, Portugal

Whole-Body Single-Bed Time-of-Flight RPC-PET: Simulation of Axial and Planar Sensitivities with NEMA and Anthropomorfic Phantoms

A single-bed, whole-body positron emission tomograph based on resistive plate chamber detectors was proposed (RPC-PET). It has been shown that RPC-PET with an axial field-of-view (AFOV) of 2.4 m is feasible and yields an absolute sensitivity 376

enhancement of 20 (4.5) in respect to 16-cm AFOV PET systems with or without time-of-flight (TOF). These results, however, do not correlate directly with lesion detectability. It is the planar (slice) sensitivity that dictates the exposure time necessary to obtain enough statistics to detect a lesion. This planar sensitivity is currently obtained with a NEMA NU-2 2001 line-source phantom. After validating our simulations with measurements published for existing scanners (GE Advance, Siemens TruePoint), we study here by simulation the axial sensitivity profiles of state-of-the-art PET scanners, and compare the results with RPCPET. Planar sensitivity results indicate that RPC-PET is expected to outperform 16-cm AFOV scanners by a factor 5.6 for a 70cm long scan. If the axial extent is elongated to 1.5 m (approximately head to mid-legs), the sensitivity gain increases to 12.2. But PET systems with larger AFOV must provide a wider acceptance angle. This results in a larger solid angle coverage, but also in larger attenuation effects. In order to quantitate these competing effects, we have studied both point- and line-sources immersed in a water phantom. For 1.5-m-long scans, the planar sensitivity drops over one order of magnitude in all scanners. In this scenario, RPC-PET outperforms 16-cm AFOV scanners by a factor of 38 (6) with or without considering the TOF benefit. RPCPET has a sensitivity gain of 13 when compared to 16-cm AFOV scanners with 600 ps FWHM TOF resolution. We further extended our studies to an anthropomorfic phantom with a realistic biodistribution. Work is in progress and results will address the issues of trues and scatter sensitivities, as well as scatter fraction and noise equivalent sensitivity profiles.
This work is supported by Fundao para a Cincia e a Tecnologia, Lisbon, Portugal, under grants nb. SFRH/BPD/39223/2007 (P. Crespo) and SFRH/BD/42217/2007 (M. Couceiro). M09-356:

B. McIntosh1, A. L. Goertzen1,2 1 Dept. of Physics and Astronomy, University of Manitoba, Winnipeg, MB, Canada 2 Dept. of Radiology, University of Manitoba, Winnipeg, MB, Canada

Validation of GATE Simulations of the 176Lu Intrinsic Activity in LSO PET Detectors

The effects of the intrinsic activity of lutetium-based scintillators such as LSO and LYSO used in PET imaging has been well documented and is not a concern in routine scanning. However, this intrinsic activity can become problematic when using a wide energy window or in low count rate scenarios such as in cell trafficking studies in small animal imaging. To date, no systematic validation of Monte Carlo simulations of the effects of the intrinsic 176Lu activity has been performed, making it difficult to incorporate them into the design and simulation of a proposed scanner. This study seeks to validate GATE simulations of the 176 Lu intrinsic activity in LSO based detectors against data gathered from a bench-top characterization of Siemens Inveon detectors. Measurements from two opposing detector modules were acquired using NIM based electronics and a PC based data acquisition (DAQ) card. The detectors were characterized by determining the count rate due to intrinsic events in singles and coincidence mode vs. detector separation. The energy spectra for singles and coincidence measurement operation were generated for each crystal in the detector. Monte Carlo simulations were performed using GATE. The simulations reproduced the geometry of the bench-top measurements made with the two detectors, modelling the intrinsic activity of the 176Lu as an ion source located within the scintillator crystals. Initial measurements show good agreement between the simulated and measured results. Intrinsic coincidence count rates are in good agreement, within a range of 1.3 to 15 % depending on the distance of the two detectors. Simulated energy spectra also agree well with measurements. The bench-top setup results require further refinement to improve the accuracy of measurements at low energy levels which will be done before completing validation of the simulation results.
M09-359: GATE Simulations of Human and Small Animal PET for Determination of Scatter Fractions as a Function of Object Size

A. Konik1, M. Madsen2, J. Sunderland2 1 Physics, University of Iowa, Iowa City, IA, USA 2 Radiology, University of Iowa, Iowa City, IA, USA

Scatter fraction (SF), which depends predominately on object size and density, is a useful parameter in determining the level of scatter correction required for a PET study. To determine SF as a function of object size, GATE simulations of Inveon and ECAT HR + PET scanners were performed. SF values were obtained for mouse, rat and human NEMA phantoms with varying diameters from 2-5 cm, 3-9 cm, and 10-56cm, respectively. The phantom and line-source lengths were held constant, as specified in NEMA. The ratio between the radius and the line-source distance to the phantoms center was also kept constant as in NEMA: 0.8, 0.7 and 0.45 for mouse, rat and human phantoms, respectively. SF results obtained in this work (at a 350-650 keV energy window) agreed with NEMA results found in other studies: 8.6% and 17.5% for mouse and rat phantoms (Inveon) and 36.1% for human phantoms (ECAT). Depending on the diameter, SF values varied between 7-15% for mouse (2-5 cm); 12-25% for rat (3-9 cm) and 26-46% for human phantoms (10-56 cm). Additionally, various sizes of XCAT and MOBY digital phantoms were simulated in ECAT and Inveon scanners. The anatomical data were segmented into 4 different attenuation regions (i.e., air, lung, water and rib bone) whereas the emission data were linearly transformed into radioactivity distributions. The SF values for an XCAT human phantom with a fixed height of 180cm, and varying diameter of 21-45 cm ranged from 30-40%, whereas MOBY mouse/rat phantoms of lengths 3.5-11.9 cm and diameters 2.1-7.2 cm ranged from 7-20%. The modified NEMA simulations indicated that for small diameter rats and for all mouse sizes, SF was lower than reported in 2D human PET (~15%) scanning. As well, SF values for MOBY phantoms up to 5 cm in diameter and 7.7 cm in length were below 15%. These results suggest that, 377

curve-fitting based scatter corrections may be sufficient for small animal PET imaging, versus sophisticated model-based scatter corrections.

M09-362:

S. A. Komarov1, H. Wu1, D. B. Keesing2, J. A. O'Sullivan3, Y.-C. Tai1 1 Radiology, Washington University in St.Louis, St.Louis MO, USA 2 BME, Washington University in St.Louis, St.Louis MO, USA 3 ESE, Washington University in St.Louis, St.Louis MO, USA

Investigation of Compton Scatter in a Clinical PET/CT Scanner with a High Resolution Half Ring PET Insert Device

We have developed a high resolution PET insert device for a clinical PET/CT system to significantly improve its image resolution within the targeted imaging field of view (FOV). For the rest of the FOV, the insert provides limited improvement for image resolution while it causes significant attenuation and scatter of the annihilation gamma rays. In order to employ all the coincidence events registered by the scanner and the insert detectors, adequate corrections for the attenuation and scatter of the events must be developed. We have developed a set of Monte Carlo (MC) simulation tools to characterize scatter in the proposed system. The Single Scatter Simulation (SSS) model was implemented and tested for a hypothetical PET insert system, and its results compared to scatter contribution simulated by MC. Results show good agreement between the SSS and MC. The SSS model was subsequently used to derive scatter estimation for the existing prototype PET insert system and will be used in a statistical image reconstruction framework.
M09-365:

A Full-System Simulation Chain for Computed Tomography Scanners

S. Kappler, D. Niederloehner, S. Wirth, K. Stierstorfer Siemens Healthcare, Forchheim, Germany Image quality and dose reduction are fields of continuous research and development in Clinical Computed Tomography. This requires screening new approaches such as innovative detector technologies, spectral filtering, data compression or novel CT algorithms. Resulting image quality properties like sharpness, image noise, homogeneity and image artifacts need to be assessed and compared to state-of-the-art solutions. We present a chain of spectral simulation tools covering full CT-systems with X-ray source, filtering, patient or phantom interaction, energy integrating or quantum counting detectors, data processing and CT image reconstruction. Modelling of the relevant physics processes is described in detail for each of the stages involved. Finally we present CT images from simulated scans showing excellent agreement with data from a real CT scanner.
M09-368:

Mathematical Modeling of a New Detector for PET on the Base of Homogeneous Scintillator

V. Y. Pedash, A. I. Ivanov Institute for scintillation materials NAS of Ukraine, Kharkov, Ukraine For the last several years the problem of an error in LOR restoration caused by relatively thick scintillation elements was discussed intensively. Several solutions for completely pixelized scintillation crystal were proposed. As an examples detector on the base of two different scintillation materials with different decay time and detector on the base of conventional crystal with two photodiodes on both input and output facets could be mentioned. The main idea in all solutions is to get radial coordinate from the detector. In this work we moved from pixelized crystal toward homogeneous one with complicated position sensitive photo detector in order to get 3D information about each flash. Theoretically the problem of LOR position uncertainty in such system will be solved automatically. We studied detector on the base of crystals with square, rectangular and trapezoidal crosssection. As the first step, the dependencies of the signals from each PMT when flash position moves only along one coordinate were calculated. It was shown that for crystal with rectangular cross-section any changes in flash position have resulted in changes in PMT signals. That is exactly what is needed for the system for 3D restoration of the flash position. However, the changes in PMT signals when flash moves along Z coordinate are relatively small, and when flash moves along radial coordinate changes in PMT signals are rather abrupt. Detector of this type showed the best results. According to mathematical calculations it allows to reconstruct precisely angular and Z coordinate and radial coordinate less accurate.
This work was sponsored by CRDF, project UAE2-1672-KK-02

378

M09-371:

A. A. M. Meneses1,2, C. J. G. Pinheiro3, L. M. Gambardella2, R. Schirru1, R. C. Barroso3, D. Braz1, L. F. Oliveira3 1 Nuclear Engineering Program, COPPE/UFRJ, Rio de Janeiro, RJ, Brazil 2 Dalle Molle Institute for Artificial Intelligence, Lugano, TI, Switzerland 3 Physics Department, State University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil

Neural Computing for Quantitative Analysis of Human Bone Trabecular Structures in Synchrotron Radiation X-Ray microCT Images

Prevention and treatment of Osteoporosis in elderly patients is critical and important since this disease became a major public health problem. It is well known the fact that osteoporotic fractures may occur as a result of a combination of the degeneration of trabecular structures and low bone mass. Therefore, the quantitative analysis of human bone trabecular architecture might be useful for treatment and diagnosis of this disease. Synchrotron Radiation X-Ray MicroComputed Tomography (CT) enables magnified images with a high space resolution that allows detailed analysis of the trabecular structure. In the quantitative analysis of medical images of human bone, it is possible to use filters and binarization, nevertheless these techniques may cause loss of information. This paper describes the alternative application of Neural Computing (Artificial Neural Networks) to the quantitative analysis of human bone trabecular structure in Synchrotron Radiation CT images obtained at the Synchrotron Radiation for Medical Physics (SYRMEP) beamline of the ELETTRA Laboratory at Trieste, Italy. Results demonstrate that, despite the complexity of the trabecular architecture, the ANNs have considerable success in the recognition of bone pixels for the quantitative analysis and that its use is compatible to the characteristics of Synchrotron Radiation images.
M09-374:

Spectral Pharmacokinetic PET Reconstruction

A. McLennan, S. M. Brady Engineering Science, University of Oxford, Oxford, UK The estimation of physiological parameters from Dynamic PET (dPET) data usually involves first independently reconstructing a series of contiguous temporal images before the parameters are estimated. This indirect procedure leads to low SNR due to using only a subset of the temporal data when reconstructing each image. Our work concentrates on the estimation of parameters directly from the dPET temporal projections. We represent the voxel-wise Time Activity Curves (TACs) as the convolution of an Input Function with a sum of exponentials, and use spectral analysis to reformulate the Ordered Subset Maximum-Likelihood Expectation-Maximisation iterative algorithm (OSEM). The main benefit of spectral analysis applied to dynamic PET reconstruction is that no particular pharmacokinetic model needs to be specified a priori, with only the input function needing to be sampled at scan time. We test our algorithm on highly realistic SORTEO generated data.
The authors would like to thank the support of Siemens Molecular Imaging, the Engineering and Physical Sciences Research Council (EPSRC) and the Department of Trade and Industry (DTI) for funding this research project. M09-377:

Tissue Spillover Correction for Dynamic Pinhole SPECT Studies of Fatty Acid Metabolism in the Rat Heart

B. W. Reutter, R. Boutchko, R. H. Huesman, A. C. Sauve, G. T. Gullberg Radiotracer Development & Imaging Technology, Lawrence Berkeley National Laboratory, Berkeley, CA, USA The goal of this research is to quantitatively compare fatty acid metabolism in the hearts of Wistar-Kyoto (WKY) normal rats and spontaneously hypertensive rats (SHR) as a function of age, and thereby track physiological changes associated with the onset and progression of heart failure in the SHR model. The fatty acid analog, I-123-labeled BMIPP, was used in longitudinal dynamic pinhole SPECT imaging studies performed on two WKY normal rats and two SHRs every seven months for 21 months. In previous work, we addressed issues associated with reconstructing dynamic data acquired with a slowly rotating camera. In this work, we address quantitative effects of limited spatial resolution that result in underestimation of metabolic rate from compartmental models. In particular, blurring of activity between the left ventricular blood pool and surrounding myocardial tissue decreases contrast between blood input and tissue uptake time-activity curves (TACs). Standard compartmental modeling straightforwardly accounts for spillover of blood activity into tissue volumes. However, accounting for spillover of tissue activity into blood volumes is more problematic. Because of tissue spillover, there is no reconstructed voxel that contains a pure blood TAC. Thus, we developed a method to jointly estimate the pure blood input along with compartmental model parameters from Bspline TACs reconstructed directly from dynamic SPECT projection data for 11 studies. Tissue spillover correction improved the contrast between blood input and myocardial uptake curves for all studies and visually improved the fit of the compartmental model for some studies. Estimates of metabolic rate of I-123-BMIPP increased by an average of 103 +/- 55% across all studies, compared to estimates obtained without spillover correction. Thus, the tissue spillover correction method resulted in improved quantitative dynamic imaging of fatty acid metabolism in the rat heart, even with slow camera rotation.

379

This work was supported by the National Institutes of Health of the U. S. Department of Health and Human Services under grants R01-EB00121, R01-EB007219, and R01-HL71253; and by the Director, Office of Science, Office of Biological and Environmental Research of the U. S. Department of Energy under contract DE-AC02-05CH11231. M09-380:

Dual Energy CT with Mismatched Rays

C. A. Maass, R. Grimmer, M. Kachelriess Institute of Medical Physics, University Erlangen-Nuremberg, Erlangen, Germany Dual energy CT (DECT) generally allows a higher order correction of beam hardeing artifacts than it can be applied to usual single energy scans. In most cases rays that are acquired with different energies do not match geometrically. Then the typical proceeding to obtain images with dual energy specific information is to linearly combine reconstructed images of each energy. This leaves the possibility of higher order beam hardening corrections unused. We propose an iterative algorithm that allowes to improve material decomposition and beam hardening artifacts for scans that were acquired by mismatched rays. The algorithm is assessed using simulated data without noise (to enhance the visibility of beam hardening artifacts). In order to use completely mismatched rays for our evaluations 3D simulations were performed with two orthogonal circular source trajectories. Two iterations were found to be capable to reduce beam hardening artifacts remarkably.
M09-383:

X. Wang1, D. Meier2, B. M. Sundal2, P. Oya2, G. E. Maehlum2, D. J. Wagenaar2, B. M. W. Tsui1, E. Frey1, B. E. Patt2 1 Johns Hopkins University, Baltimore, MD, USA 2 Gamma Medica - Ideas, Oslo/Sherbrooke/Northridge, Norway/Canada/USA We describe a digital line-camera that can be used for computed tomography (CT) with energy resolved x-ray photon counting (XPC) capability. The camera is based on pixellated cadmium telluride (CdTe) radiation sensors and application specific integrated circuits (ASICs). Each camera pixel simultaneously counts x-rays in multiple energy bins providing the capability for color x-ray imaging and energy weighted x-ray CT. The camera can be used in pre-clinical x-ray micro-CT, where energy resolved photon counting can improve signal-to-noise and contrast in the image and reduce radiation dose in the subject. We began testing the camera using 80-kVp polychromatic x-rays illuminating a rotating phantom. We measured the intensity of xrays transmitted through the phantom and reconstructed tomographic images. This article describes the functionality and performance of the camera, and presents initial results from x-ray micro-CT with phantoms.
M09-386:

A Digital Line-Camera for Energy Resolved X-Ray Photon Counting

A Preliminary Investigation of Compressive-sensing Image Reconstruction from Flying-focal-spot CT Data

D. Xia, J. Bian, X. Han, E. Sidky, X. Pan Radiology, The University of Chicago, Chicago, IL, USA In advanced clinic CT scanners, a flying-focal-spot (FFS) technique, which allows for view-by-view deflections of the focal spot, is implemented for improving the sampling density in the rotation direction, (i.e., xFFS) the longitudinal direction (i.e., the zFFS), or the combinations of the two. Studies have been conducted for image reconstruction from the data obtained with the FFS technique by use of the one-step algorithms. Results of the studies demonstrate that the utilization of the FFS technique can lead to a reduction of aliasing artifacts in the reconstructed images. The increase of the sampling density through multiple illuminations in FFS can result in the increase of the radiation dose to the imaged subject when the imaging dose for each FFS is fixed. In this work, we develop a compressive-sensing-based algorithm that reconstructs an image through minimizing the total variation (TV) of the image, subject to the data condition, from simulated data as well as actual data acquired with a clinical CT scanner. The results of the study demonstrate that aliasing artifacts observed images reconstructed by use of the one-step algorithms can be eliminated effectively in images reconstructed with the TV-minimization algorithm even from a single-xFFS data.
M09-389:

G. S. K. Fung1, J. Xu1, K. Taguchi1, S. Kawamoto1, E. Fishman1, T. G. Flohr2, K. Stierstorfer2, B. M. W. Tsui1 1 Department of Radiology, Johns Hopkins University, Baltimore, US 2 Siemens Health Care, Forchheim, Germany

Optimization of kVp Settings and Metal Filter for Maximum Differentiation of Kidney Stones using Dual-Source CT

The objective of this study is to optimize the low-kVp and high-kVp x-ray spectra for maximum differentiation of kidney stones using dual-source CT. Three common kidney stones with different chemical compositions and densities were simulated. They were embedded in a cylindrically shaped water-filled phantom with a diameter of 30 cm. Noise-free projection data of the phantom were simulated using the DRASIM simulation package with dual-kVp x-ray sources. The low-kVp and high-kVp x-ray 380

spectra were varied by changing the kVp settings from 60 to 150 kVp and by adding different metal filters with different thicknesses. The common constraint was constant total energy fluence for the output x-ray spectra after filtration. The projection data were reconstructed using the filtered backprojection method after beam-hardening correction in the sonogram domain. The Hounsfield unit (HU) values of the simulated kidney stones were determined from the dual-kVp CT images and the HU ratio (HUR) was employed as the material index. The overall difference of HUR (DHUR) between different kidney stone types was used as the figure-of-merit for material differentiation. Among the filter materials tested, tin (Sn) and molybdenum (Mo) were the best in terms of maximum DHUR improvement of more than four times better than that from other filter materials tested. Better kidney stone separation was found with larger kVp difference. With 0.5 mm Mo filtration, the 60/150 kVp setting achieved three times higher DHUR than the 80/140kVp setting. Further studies on the effect of noise, radiation dose, and practical constraints are need. In conclusion, optimization of kVp settings and metal filtration in terms of both material and thickness, which modify the shape and the degree of overlap of the two x-ray spectra, allows maximum differentiation between different kidney stones using dual-source CT. The optimization technique can be applied to other applications with different tissue types.
M09-392:

Distribution of Absorbed Dose in Cone-Beam Breast Computed Tomography: a Phantom Study with Radiochromic Films

P. Russo, T. Coppola, G. Mettivier, M. C. Montesi, A. Lauria Department of Physics, INFN and Universit di Napoli Federico II, Napoli, Italy Cone-Beam Breast CT (CBBCT) with dedicated scanners, as investigated by several authors and, recently, by our group, after the pioneering work of Boone et al., is a promising technique for diagnosis of breast tumor at the screening and/or at the clinical level. CBBCT holds potential for a better diagnosis through a 3-D visualization of tissue lesions, so avoiding the blurring effect of normal tissue structures underlying and overlying the lesion. This situation occurs in mammography, which is a 2-D projection imaging technique. For better localization, common practice in conventional X-ray mammography is to take (at least) two views for each breast (craniocaudal and mediolateral oblique). The corresponding Mean Glandular Dose for regulatory guidelines should not exceed 2.5 mGy (in EU) per view in a standard breast compressed to 53 mm (45 mm PMMA equivalent). The main interest of CBBCT is related to the possibility of providing a 3D visualization of tissue lesions with respect to the 2D visualization of conventional mammography. In addition, a distinctive advantage of breast X-ray tomography over breast X-ray mammography is that in the former case, a more uniform distribution of the radiation dose delivered to the organ can be achieved, since the X-ray beam irradiates the organ in a full 360 deg rotation. A more homogeneous distribution of the total dose in CBBCT is a qualitative parameter of interest, still to be quantitatively assessed experimentally. We have taken on this task and have investigated the possibility of using a radiochromic film placed at midplane in a PMMA phantom to record the dose distribution with high spatial sampling and ease of operation.
M09-395:

Development of Multimirror for the Parallel X-Ray Beam of a Laboratory Based X-Ray Source

T. Nguyen, K. Kim, I. Jeon School of Mechanical Systems Engineering, Chonnam National University, Gwangju, Republic of Korea Observing clearly the internal structure of an object is one of the most important targets of using the parallel X-ray beam. A cluster of the parallel beam can be obtained from a dispersive laboratory based X-ray source after reflecting the X-rays through the multimirror containing a numbers of mirror layers piled up one another with small gap between them. In this paper, a parabolic optical multimirror using the depth-graded multilayer coating has been proposed to create a parallel Xray beam by the total external reflection. The reflective X-ray mirrors require super-smooth surfaces and highly accurate profiles. A replication method was used to fabricate the mirrors based on the master cores. The ultra-precision polishing tool was used to finish the surface of the cores. A coupled W/Al bilayer with gradually changed thickness was coated on the smooth surface of the cores in order to increase not only the incidence angle but also the intensity of the output parallel beam. The maximum length of mirrors is limited under 4 inch. The thickness 2.71 mm of the parallel beam was obtained using five mirrors located at the position 1200 mm far from the X-ray source at the wavelength of 0.124 nm.
M09-398:

L. P. Nogueira1, R. C. Barroso2, C. J. Pinheiro2, D. C. Braz1, L. F. Oliveira2, G. Tromba3, N. Sodini3 1 Nuclear Instrumentation Laboratory, Federal University of Rio de Janeiro/COPPE, Rio de Janeiro, Brazil 2 Physics Institute, State University of Rio de Janeiro, Rio de Janeiro, Brazil 3 Sincrotrone Trieste SCpA, Trieste, Italy

Mapping Lead Distribution in Bones by Dual-Energy Computed Microtomography with Synchrotron Radiation

Bone is already known to be the repository of most lead that is not excreted from the body. In adults the half life of lead in trabecular bone is several years and in cortical bone probably decades. There are good evidences that in men and women the risk for hypertension increases as bone lead concentrations increase. The purpose of this study was to demonstrate the feasibility of using dual-energy subtraction to enhance the intrinsic lead distribution in contamined bones. Dual-energy subtraction imaging is 381

an effective technique proposed, which allows removal of contrast between pairs of materials by a combination of two images acquired at different energies (low and high energy). In this way it is possible to improve the contrast of details of interest, by removing the background. All samples (n=8) were imaged using synchrotron radiation micro-CT developed on the Synchrotron Radiation for Medical Physics (SYRMEP) beamline which is one of the bending magnets of the ELETTRA Synchrotron Radiation Facility. The micro-CT system is based on 3D parallel tomographic acquisition. This set-up consists of a high precision rotary table mounted upon high-accuracy translators and cradles. This allows precise alignment ofthe rotation axis of the sample with the detector pixels. Each sample was imaged at two different energies, one just above (16.10 keV) and the other just below (15.75 keV) a specific electron binding energy of the lead. For evaluation of the potential of this procedure we studied the contrast on a set of images obtained by applying the dual-energy procedure. The subtracted images enhanced the presence of different levels of lead, which allows to map its distribution in contaminated bones. This study has shown that micro-CT using dual energy subtraction is potentially feasible and could provide a material selective image reconstruction.
This work was supported by ICTP-ELETTRA Users Programme. The authors thank the Brazilian agencies CNPq and FAPERJ for finantial support. We also thank the SYRMEP group, and particularly Franco Zanini and Diego Dreossi for help during image acquisition and reconstruction. M09-401:

Y. Kubota1, K. Aoki2, H. Nagahashi2, S.-I. Minohara3 1 Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Yokohama-shi, Kanagawa-ken, Japan 2 Imaging Science and Engineering Laboratory, Tokyo Institute of Technology, Yokohama-shi, Kanagawa-ken, Japan 3 National Institute of Radiological Sciences, Chiba-shi, Chiba-ken, Japan

Pulmonary Motion Tracking from 4D-CT Images Using a 3D-KLT Tracker

We propose a new method for lung-motion tracking and its quantification from 4-dimensional X-ray computed tomographic (4DCT) images. This method uses an enhanced 3D-KLT tracker. An advantage of our method is that it can find many feature points (regions) for tracking that are not restricted to the bifurcation points of bronchi or vessels. The feature point extraction algorithm depends only on image gradients. Moreover, our method adopts a hierarchical tracking based on pyramidal image structure. This provides robustness for large movements of the objects. Lung motion is quantified by tracking a large number of feature points in the lung. In this paper, we first evaluate the performance of our proposed method for artificial 4D-CT images and then describe quantification results of real 4D-CT images. Our experimental results clearly show that lung movement is not a simple translation but an oval pattern.
M09-404:

Reduction of Artifacts Due to Multiple Metal Objects in Computed Tomography

K. Y. Jeong, J. B. Ra KAIST, Daejeon, South Korea A CT image including metallic objects suffers from strong metal artifacts such as shades and streaks. Most metal artifact reduction algorithms based on filtered back-projection have tried to make a metal-free projection data by modifying the original data on the metal-traces in the sinogram. Although those algorithms are able to reduce metal artifacts, they tend to produce new artifacts and distort non-metallic regions. In this paper, we propose a novel algorithm for reducing metal artifacts while preserving the non-metallic region with minimum distortion. In the proposed algorithm, we first generate a metal-free image by removing metallic regions in the initially reconstructed image and interpolating the intensities of the metallic regions with those of non-metallic surrounding pixels. Based on the metal-free image, we obtain a reprojection data. We then modify the original projection data by combining the reprojection data with the projection data whose metal-traces are replaced with linearly interpolated values in the sinogram. Meanwhile, since we find that shade artifacts of large area are directly related with the underestimated intensity values in the overlapped regions of metal-traces, we modify intensity values in the overlapped regions by interpolating them along the metal-trace directions. The final reconstructed image is then obtained by using this modified projection data. The proposed algorithm is compared with the existing algorithms, and it is shown that the algorithm can provide much better performance by effectively reducing shade artifacts between metallic objects without introducing the shape distortion of non-metallic objects.

382

M10: PET/SPECT instrumentation 2 Friday, Oct. 30 14:00-15:30 International Ballroom Center

T. Zeniya1, Y. Hirano1, T. Sakimoto2, K. Ishida1, H. Watabe1, N. Teramoto1, H. Kudo3, K. Minato2, J. Hatazawa4, H. Iida1 1 Dept. of Investigative Radiology, Advanced Medical Engineering Center, National Cardiovascular Center Research Institute, Suita, Osaka, Japan 2 Graduate School of Information Science, Nara Institute of Science and Technology, Ikoma, Nara, Japan 3 Graduate School of Systems and Information Engineering, University of Tsukuba, Tsukuba, Ibaraki, Japan 4 Dept. of Nuclear Medicine and Tracer Kinetics, Osaka University Graduate School of Medicine, Suita, Osaka, Japan

M10-1: Conceptual Design of High Resolution and Quantitative SPECT system for Imaging a Selected Small ROI of Human Brain

We designed a concept of high resolution and quantitative SPECT for imaging a selected small region-of-interest (ROI) of human brain. This system is aimed at achieving high resolution less than 1 mm and will apply for imaging neurons and evaluating drug delivery system. Pinhole or cone-beam collimators are useful for high-resolution imaging of small field-of-view (FOV). However, when the object is large like humans, the projection data are truncated by radioisotope outside small FOV. Because of this truncation, on the reconstructed image, the artifact appears and the voxel value was overestimated. This hampers quantitative assessment of physiological functions. We are introducing the new truncation compensated 3D-OSEM (TC-3DOSEM) reconstruction method. The truncated data can be successfully reconstructed within FOV by fulfilling the condition that FOV contains the region with known value. In addition to small FOV detector, we are introducing the parallel-hole collimator attached middle-sized FOV detector covering the whole of brain, to acquire the untruncated data and provide the known value in small FOV, even if it has low resolution. For imaging with high resolution, we are using LaBr3(Ce) as scintilator of the detector and position-sensitive photomultiplier tube (H8500, Hamamatsu, Japan). We are also incorporating attenuation and scatter corrections, and resolution recovery into 3D-OSEM reconstruction, to obtain the image with high resolution and high quantitation. And also, for proof of our concept, we performed preliminary experiment using pinhole SPECT and brain phantom. FOV contained the region outside the phantom, that is, zero count as the known value. The truncated data were reconstructed by TC-3DOSEM. The reconstructed image without artifact and overestimation was obtained with high resolution. This preliminary experiment suggested feasibility of high resolution and quantitative SPECT for imaging a selected small ROI of human brain.
M10-2:

Design and Development of a New PET Detector with Both DoI and ToF Capabilities

V. C. Spanoudaki, C. S. Levin Radiology, Stanford University, Stanford, CA, USA We are investigating a detector design for clinical PET which will combine depth of interaction (DoI) and time of flight (ToF) features. Such a detector advance is needed in whole-body clinical PET systems that have adjustable system diameter to optimize photon sensitivity (and potentially dose) for different size patients. This design shows promise for improvement of lesion detectability through enhancement of SNR in the reconstructed image, as well as uniform spatial resolution throughout the FoV. Monte Carlo simulations show that a 3-fold or worse spatial resolution degradation is observed at the edges of the FoV of cylindrical PET systems with diameters below 40 cm. We are considering two alternative DoI/ToF detector designs with individual and dual ended crystal readout (3-20 mm long and 3 mm wide crystals). In order to more accurately process intercrystal scatter events, these detectors will also have capability to localize the 3-D coordinates of individual 511 keV photon interactions. For extraction of ToF information, Silicon Photomultipliers (SiPMs), capable of single electron resolution, are used and a first photoelectron-based time pick-off method is facilitated by ultra fast electronics. Initial results from SiPM intrinsic timing measurements yielded 480 ps resolution (limited by the pulser). Optical simulations of the two designs demonstrated an improvement in photon flight time larger than 50% for rough over polished crystal surfaces. A dependence of the mean but not RMS flight time on DoI (30% difference for 4 mm DoI difference) is evident for rough crystal surfaces and long crystals. Our first studies show that a detector design employing 3 mm long crystals (3 mm DoI resolution) with unpolished surfaces results in the best trade-off between light collection efficiency (>64%) and minimum mean photon transit time in the crystal of ~70 ps. Monte Carlo modeling of a complete clinical DoI/ToF tomograph with 3D positioning capabilities is also performed.
M10-3:

Development of a Novel High-Sensitivity Dedicated Brain PET Scanner: NeuroPET

S. Adler, P. Domigan, O. Johnson, P. Juels, H. Kudrolli, P. Kulinich, D. Lazuka, P. Monteverde, J. Nevin, L. Romanov, S. Starsja, J. Taggart, T. Toole, S. Walker, B. Worstell, C. Worth PhotoDetection Systems, Inc., Boxboro, MA, USA We have developed a compact, movable, dedicated 3D PET brain scanner (NeuroPET, an FDA Cleared medical device) targeted towards imaging of Alzheimers disease pathology as well as that of other neurological and psychiatric indications. The system can perform brain PET scans in a ten by fourteen foot space with office electrical power. We have performed standardized 383

NEMA (National Electrical Manufacturers Association), ACR, and Hoffman phantom-based testing of this new instrument, which can be used to compare its performance with existing general-purpose scanners. We have also generated human amyloid radiotracer PET images and have compared these with those of the same subjects on conventional general-purpose PET scanners. The NeuroPET system has produced comparable results with significantly lower dose to the patient due to its higher system sensitivity.
M10-4:

Characterization of the Clear-PEM Breast Imaging Scanner Performance

J. Varela LIP, Lisbon, Portugal On behalf of the ClearPEM Collaboration We present results on the characterization of the Clear-PEM breast imaging scanner. Clear-PEM is a dual-head Positron Emission Mammography scanner using APD-based detector modules that are capable of measuring depth-of-interaction (DOI) with a resolution of 2 mm in LYSO:Ce crystals. The full system comprises 192 detector modules in a total of 6144 LYSO:Ce crystals and 384 32-pixel APD arrays readout by ASICs with 192 input channels, which represents an unprecedented level of integration in APD-based PET systems. The system includes Frontend and Data Acquisition electronics and a robotic gantry for detector placement and rotation. The software implements calibration (energy, time and DOI), normalization and image reconstruction algorithms. In this work, the scanner main technical characteristics, calibration strategies and the spectrometric performance in a clinical environment are presented. Images obtained with point sources and extended uniform sources are also presented. The first commissioning results show 99.7% active channels. After calibration, the dispersion of the channels absolute gain is 15.3%, which demonstrate that despite the large number of channels the system is rather uniform. The mean energy resolution at 511 keV is 15.9% for all channels, and the mean DOI constant is 5.9%/mm, which is consistent with a 2 mm DOI resolution, or better. The coincidence time resolution at 511 keV, for a energy window between 400 and 600 keV, is 5.2 ns FWHM. The image resolution measured with point sources was found to be of the order of 1.3 mm FWHM. The DOI capability was found to have a strong impact on the image sharpness. Images of extended uniform 68Ge sources, corrected for sensitivity and for the artifacts due detector dead spaces, have good uniformity.
M10-5:

W. W. Moses1, S. Buckley2, C. Vu1, W.-S. Choong1, N. Pavlov2, Q. Peng1, C. Jackson2 1 Lawrence Berkeley National Laboratory, Berkeley, CA, USA 2 SensL, Cork, Ireland We present the design for OpenPET, an electronics readout system designed for prototype radiotracer imaging instruments. The critical elements are that it has sufficient performance, channel count, density, and power consumption to service a complete camera without sacrificing performance, and yet be simple, flexible, and customizable enough to be used with almost any detector or camera design. There are 4 types of custom circuit boards in the system. Each Detector Processing Board (DPB) processes 16 analog signals and creates digital Singles Event Words for each gamma ray interaction. Singles Event Words are passed to a Processing Support Board, which multiplexes the Singles Event Words from 8 DPBs. An optional multiplexer board can provide a further layer of multiplexing for Singles Event Words. The Coincidence Board searches through the Singles Event Words for pairs that are in time coincidence; when it does so, it passes a Coincidence Event Word to the Host PC. Alternatively, it can multiplex and pass unaltered Singles Event Words to the Host PC. A unique feature of this system is that each analog input is processed independently. Each input can be configured to accept signals of either polarity as well as either differential or ground referenced signals. It is digitized by a continuously sampled (80 MHz) ADC, which is processed by an FPGA to extract pulse height information. A leading edge discriminator creates a timing edge that is time stamped with <0.5 ns accuracy by a TDC implemented inside the FPGA. A single FPGA services all 16 analog channels, and so information from multiple channels can be processed by this FPGA to do Anger logic for crystal lookup, DOI calculation, calibration, etc. As this processing is software controlled, it can be modified / customized easily. The system will be open source, meaning that all technical data (specifications, schematics and board layout files, source code, and instructions) will be publicly available.
This work was supported in part by the Director, Office of Science, Office of Biological and Environmental Research, Medical Science Division of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231, and in part by the National Institutes of Health, National Institute of Biomedical Imaging and Bioengineering under grant No. R01-EB006085.

OpenPET: a Flexible Electronics System for Radiotracer Imaging

384

M10-6:

M. Meissani1, M. Canadas2, P. Arce3 1 IFAE, Barcelona, Spain 2 Medical Physics Applications, CIEMAT, Madrid, Spain 3 Basic Research, CIEMAT, Madrid, Spain

Performance of PET Scanner Based on Pixelated Solid-State Detector

We propose a novel PET scanner design, Pixel-PET, which uses pixelated room temperature solid-state-detector (CdTe for example) coupled to readout front end electronic, instead of using scintillating crystals coupled to photon-sensitive sensor. The advantages of Pixel-PET when compared to PET based on crystal detectors are the followings: A- The detector can be segmented into millimetre pixels size, something which is not possible with scintillating crystals B- CdTe detector can achieve a FWHM energy resolution less than 1% at 511keV photons. This allows us to eliminate most of the scattered events and thus keep the golden events C- Pixel-PET design allows an adequate depth of absorption; using CdTe detectors achieve high detection efficiency for 511keV photon, something that is not possible with PET based on crystal detectors D- When used in MRI, the field inside the CdTe is // to the field of the MRI. Therefore and the device can operate in high magnetic field E- The unit-detector (see fig.1) of Pixel-PET can be designed to have trapezoidal parallelepiped instead of rectangular parallelepiped geometry, so that it can be arranged to form a flat-panel detector, instead of cylindrical, to be used as Gamma Camera Preliminary simulation results for Pixel-PET shows that the spatial resolution is about two times smaller than the one of a stateof-the-art crystal PET, with a 2.5 bigger sensitivity for CdTe and 7.5 for TlBr, and a scatter fraction of 2-3% of true events compared to almost 95% in crystal PET. It is easy to conclude that one expects for Pixel-PET the same quality image as crystal PET but with very little radiation dose, thanks for its ability to eliminate scattered events and keep only the golden events. Pixel PET will have a great impact on brain imaging with dual imaging modality (PET+MRI) for its ability to reduce scattered events and to stand strong magnetic fields
M11: Plenary 2 / Multimodality Instrumentation and Techniques Friday, Oct. 30 M11-1: 16:00-18:00 International Ballroom Center

Challenges and Solutions for Imaging Stem Cell Transplantation

K. Wilson Stanford University, Palo Alto, CA, USA The discovery of human embryonic stem cells (hESCs) has dramatically increased the tools available to medical scientists interested in regenerative medicine. However, direct injection of hESCs, and cells differentiated from hESCs, into living organisms has thus far been hampered by significant cell death, tumor (teratoma) formation, and host immune rejection. Understanding the in vivo hESC behavior after transplantation requires novel imaging techniques to longitudinally monitor hESC localization, proliferation, and viability. Molecular imaging, and specifically bioluminescent and positron emission tomography reporter gene imaging, has given investigators a high-throughput, relatively inexpensive, and sensitive means for tracking in vivo cell proliferation over days, weeks, and even months. This advancement has significantly increased the understanding of the spatio-temporal kinetics of hESC engraftment and proliferation in living subjects. In this talk the many challenges, and solutions, for imaging stem cell transplantation, and their implications for future clinical therapies will be discussed.
M11-2:

S. Chen1, D. Meier2, J. Xu1, J. Yu1, D. Wagenaar3, B. M. W. Tsui1 1 Radiology, Johns Hopkins Medical Institutions, Baltimore,MD, United States 2 Gamma Medical-Ideas Inc., Oslo, Norway 3 Gamma Medical-Ideas Inc., Northridge, United States The purpose of the study is to investigate the feasibility of simultaneous dynamic SPECT/MR imaging of small animals. The stationary SPECT camera used here has relatively high sensitivity and is fully compatible with MRI scanners. The SPECT camera consists of three rings of 8 CZT detector modulesand fits inside MRI scanners with a minimum bore diameter of 12 cm. A sleeve collimator with 24 pinholes of 2 mm diameter is used which provides high geometrical efficiency up to ~2.5%. We measured the count-rate performance of the SPECT camera for both inside and outside the bore of a Philips 3T MRI scanner. We developed calibration methods for the SPECT camera that can be applied inside and outside of the MRI scanner. We generated Monte Carlo simulated Tc-99m Sestamibi myocardial perfusion SPECT data acquired from the digital MOBY mouse phantom and corresponding simultaneously acquired MR images. The iterative maximum a posterior (MAP) expectation-maximization (EM) image reconstruction method was applied to reconstruct the SPECT projection data from a typical dynamic study using the corresponding co-registered MR images as the prior information.Our results show the SPECT camera can achieve a count rate of ~105 cps for the activity concentration of 0.0176 mCi/ml in a mouse-size uniform cylindrical phantom with negligible count rate loss due to detector dead time. There is little difference between the count-rate performances inside and outside the 3T MRI 385

Feasibility of Small Animal Dynamic SPECT Inside MRI

scanner. The standard deviation of a region of interest in the images reconstructed by MAP-EM method is ~1/3 of that reconstructed by the conventional maximum-likelihood expectation-maximization (ML-EM) method. In conclusion, the stationary SPECT camera allows simultaneous dynamic SPECT/MR imaging of small animals. The MAP-EM reconstruction method using the registered MR images as the prior provide SPECT images with significantly reduced noise level and artifacts from the SPECT/MR data.
The work is supported by NIH SBIR grant R44 EB 0067112-02. We would like to thank Dr Michael Schaar for his help on MRI experiments. M11-3:

MR-Based Motion Correction in Simultaneous PET-MR

S. Cho, B. Guerin, T. Reese, J. Ouyang, C. Catana, G. El Fakhri Radiology/Nuclear Medicine, Mass. General Hospital and Harvard Medical School, Boston, USA Accurate motion correction is very important for quantitative PET studies. Specifically, in thoracic imaging including cardiac studies, the respiratory and cardiac non-rigid motions cause inaccurate lesion/perfusion defect detection and activity estimation. PET-MR provides a promising approach to continuously compensate for motion during PET acquisitions. The main goal of our project is to develop a quantitative PET reconstruction algorithm in a simultaneous PET-MR scanner by accurately correcting patient motion using SPAMM tagged MR. We have developed a novel PET reconstruction method where the motion information is incorporated into the system matrix to correct for motion. We derive the motion information from tagged MR using the HARmonic Phase (HARP) tracking algorithm. We extend HARP to thoracic and cardiac PET studies in order to estimate lung and cardiac motion. We also develop a method to evaluate the motion tracking algorithm with realistic MR simulations. We show the feasibility and usefulness of our motion corrected PET reconstruction method in realistic simultaneous PET/MR simulations and physical acquisitions
M11-4: Development of a Detector Module for Combined PET/CT or Combined Photon Counting/Standard CT Based on SiPM Technology

A. Persson1, A. Khaplanov1, B. Cederwall1, C. Bohm2 1 Department of Physics, Royal Institute of Technology, Stockholm, Sweden 2 Department of Physics, Stockholm University, Stockholm, Sweden

Recent developments make it possible to utilize SiPMs for quantifying high radiation fluxes in current mode as well as in pulse mode measurements for counting and characterizing individual gamma-ray or X-ray photons when coupled to scintillators. This opens new possibilities for multimodal medical imaging by enabling common radiation sensors for PET, standard CT and photon counting CT. Currently, two separate detector systems are required for combined PET/CT. The advantages of an integrated PET/CT system include increased patient throughput, higher image fusion accuracy due to perfect PET-CT sensor alignment and reduced system cost. In this work we present a novel detector design based on dual-mode readout of SiPM-based radiation sensors that enables imaging systems where conventional CT imaging is combined with PET, as well as in CT systems where both conventional integrating (high-flux) and photon counting (low-dose) operation can be selected. Such dual-mode SiPMs coupled to scintillators in medical imaging can be used to detect and characterize single primary X-ray or gamma-ray photons up to fluxes of millions of photons per mm^2 per second in pulse mode. Current mode operation allows the dynamic range to be extended to the much higher rates found in standard CT imagers and beyond. For this purpose the selection of the optimal scintillator is crucial as is the design of the dual SiPM readout electronics. A prototype detector has been developed and studied from the point of view of the energy and timing resolution required for the photon-counting application as well as the flux-tocurrent characteristics, essential for the current mode.
M11-5:

M. S. Judenhofer1, H. F. Wehrl1, F. C. Maier1, J. G. Mannheim1, D. Bukala1, G. Tabatabei1, A. Schmidt1, N. Kemmler1, A. Sauter1, C. Calaminnus1, G. Reischel2, B. J. Pichler1 1 Laboratory for Preclinical Imaging and Imaging Technology of the Werner Siemens-Foundation, University of Tuebingen, Tuebingen, Germany 2 Department of Radiopharmacy, University of Tuebingen, Tuebingen, Germany PET/MRI offers the possibility to acquire simultaneously multiple diagnostic parameters such as PET tracer uptake and apply various MR techniques. This can be used to correlate sites of PET tracer uptake with areas of MR spectroscopy or contrast enhanced MR images. Furthermore, dynamic PET data like [O-15]water perfusion can be correlated with corresponding MRASL (arterial spin labeling) measurements. An investigation of mutual interference of PET tracer on MR spectroscopy was performed in SMA560 brain tumor bearing mice. Spectra pre and post application of [C-11]Choline tracer were acquired and compared. In addition, CSI maps and contrast enhanced images were acquired. A comparison showed no influence of the PET tracer on the MR spectra of choline indicating that the sensitivity of MRI may be to low to detect the small amounts of used PET tracer. Fusing PET with CSI maps or post contrast MR images showed only partial matching of PET and MRI data revealing that 386

Simultaneous in Vivo Measurements with a Combined Animal PET/MRI

PET and MRI may not be measuring the same metabolic processes (deposited choline content vs. elevated choline turnover). In a healthy Balb/c mouse perfusion measurements have been performed using [O-15]water PET and ASL MRI. To change the perfusion rate of the animal, two simultaneously PET and MRI acquisition were performed using anesthesia with pure oxygen (O2) and carbogen (CG). To generate perfusion values from PET data a second PET measurement was performed at the heart to obtain an arterial input function. ASL-MR data showed a slight increase in perfusion from [69+-23] to [101+-27] ml/100g/min during CG breathing. The PET cortex perfusion for O2 was [94+-25]ml/100g/min and [148+-20]ml/100g/min for CG. Besides differences in the absolute perfusion values between ASL-MR and PET, we observed differences in the relative perfusion increase between O2 and C breathing (46% MR - 57% PET). These differences may be caused by diverse modeling approaches and assumptions, and are currently further investigated.
M11-6:

Z. Hu1, N. Ojha1, S. Renisch2, V. Schulz3, I. Torres4, D. Pal1, G. Muswick1, J. Penatzer5, T. Guo1, P. Boernert6, C.-H. Tung1, J. Kaste1, L. Shao1, M. Morich5, T. Havens1, P. Maniawski1, W. Schaefer7, R. W. Guenther7, G. A. Krombach7 1 BU NM, Philips Medical Systems, Cleveland, Ohio, USA 2 Digital Imaging, Philips Research, Hamburg, Germany 3 Molecular Imaging Systems, Philips Research, Aachen, Germany 4 Institute of Imaging & Computer Vision, RWTH Aachen University, Aachen, Germany 5 BU MR, Philips Medical Systems, Cleveland, Ohio, USA 6 Imaging Systems & Intervention, Philips Research, Hamburg, Germany 7 Department of Diagnostic Radiology, University of Technology, Aachen, Germany Integration of whole-body PET and MR imaging has recently gained significant interest because of several potential advantages over PET/CT. MR imaging offers outstanding soft-tissue contrast with the capacity to acquire functional data, all without radiation exposure. However, unlike transmission sources or CT, MRI does not allow direct measurement of attenuation coefficients of photons generated in PET. We describe a fully automated whole-body MR-based attenuation correction (MRAC) technique. We also address other technical challenges for effective attenuation correction (AC) in a whole-body PET/MRI, including compensation for MR-image truncation and correction for the patient table, RF coils and accessories. For MRAC, a dedicated imaging sequence was developed which is also appropriate for anatomical localization, thereby improving clinical workflow. MR images were segmented into air, lung and soft tissue. Lung identification was based on region growing technique and outer contour identification was done using a threshold technique. This implementation was chosen to assure reproducibility and stability of results in a wide patient population. MRAC accuracy was evaluated on patient datasets collected on separate PET/CT and MR systems. PET data was reconstructed using the 3-segment MRAC and compared to PET images reconstructed with CT-based attenuation correction (CTAC) using standard clinical PET/CT protocol. Visually, the images reconstructed with MRAC closely resemble CTAC images. Preliminary quantitative analysis demonstrated < 10% error in SUV measurement in ROIs around heart, liver, and bladder. It should be noted that this error did not include potential contributions from MR-image truncation and attenuation correction of table/coils/accessories.
M12: X-ray imaging 2 Saturday, Oct. 31 M12-1: 08:30-10:00 International Ballroom Center

MR-Based Attenuation Correction for a Whole-Body Sequential PET/MR System

J. Hsieh1, B. Senzig1, S. Woloschek1, S. Aluri1, T. Benson2, X. Wu2, D. Okerlund1, B. Li1 1 GE Healthcare, Waukesha, WI, USA 2 GE GRC, Niskayuna, NY, USA

CT Spectral Projection Imaging

Purpose: Dual-energy (DE) CT has received much attention. Despite its successes, radiation dose for DECT is significantly higher than that of a DE x-ray radiography (DEXR). We investigate a spectral projection imaging (SPI) approach to identify locations and characteristics of suspected pathology. Methods and Materials: In SPI, the gantry remains stationary while the table travels continuously similar to a scout. By switching the tube voltage rapidly (>2kHz) with a synchronized DAS sampling, overlapped samples are obtained for both low- and highkVp settings. These samples undergo calibration steps similar to CT to remove non-ideal acquisition conditions. The calibrated samples are processed to produce material-density projections. Because the rapid sampling between two kVps (0.5ms) is less than 0.4% of that of DEXR (>125ms), it eliminates potential mis-registration artifacts that often appears in DEXR. Experimental Results: A chest phantom was scanned with regular scout, and overlapped anatomies make the visualization of small structures difficult. SPI images were then generated for soft-tissue density and bone density. By separating the high-density structures from the soft-tissue, visibility of vascular details is significantly improved. To further test its material decomposition capability, we scanned a torso phantom with planted kidney stones. By generating Ca and water material pairs, uric acid stone can be clearly differentiated from the Cystine and Struvite based on its absence in the calcium-density image. Conclusion: We present a SPI approach that utilizes the DE acquisition to provide additional information to DECT. We show the 387

advantages of SPI over DEXR in terms of motion suppression, better material separation, and reduced beam-hardening effects. Phantom and clinical experiments show that SPI is capable of accurately classifying different materials in the body to provide excellent information at very low dose.
M12-2:

Radiation Dose Reduction with Voltage Modulation in Dual Energy CT

Y. Zou Image Quality, Toshiba Medical Research Institute USA, Inc., Vernon Hills, IL 60061, USA In this paper, we propose to apply the voltage modulation in dual energy CT. With a pre-reconstruction decomposition, the projection data in each view are decomposed into the path lengths of the basis materials that are independent of spectra. The inconsistency due to voltage variation in a single energy CT is avoided. The voltages in each view can be adapted to the subject to achieve the optimal dose efficiency. Combined with the current modulation, the patient dose can be reduced and the image quality can be preserved.
M12-3:

A new Method for Cupping and Scatter Precorrection for Flat Detector CT

R. Grimmer, C. Maass, M. Kachelriess Insitute of Medical Physics, University of Erlangen-Nuernberg, Erlangen, Bavaria, Germany Scatter and beam hardening are prominent artifacts in x-ray CT. Currently, there is no precorrection method that inherently accounts for tube voltage modulation and shaped prefiltration. We generalized a method for self-calibration based on binary tomography of homogeneous objects to use this information to preprocess scans of other, non-binary objects, e.g. to reduce artifacts in medical CT applications. Further on we extended the method not only to handle beam hardening but also scatter and to allow for detector pixel-specific precorrections. This implies that our calibration technique handles varying tube voltage and shaped prefiltration. We propose a method that models the beam hardening by using a rational function (7 parameters per projection value) while scatter is modeled using the pep-model (3 parameters per projection value). A smoothness constraint is applied to the parameter space to regularize the underdetermined system of non-linear equations. The parameters determined are then used to precorrect CT scans. Our algorithm was evaluated using simulated data of a flat panel scanner with tube voltage modulation and bow tie prefiltration and real data of a flat detector cone-beam CT scanner. The results were compared to standard reconstructions with a conventional empirical precorrection function applied to all detector channels. In simulation studies our correction model proved to be nearly perfect and the algorithm showed its abilities by correcting the beam-hardening and scatter effects. The relative error between the estimated parameters and those used for simulation was below 1.7%. Reconstructions of measured data showed significantly less artifacts than the standard approach.
M12-4: Dynamic Iterative Beam Hardening Correction (DIBHC) for an Optimized Assessment of Cardiac Perfusion in ECG-Correlated CT

P. Stenner1, B. Schmidt2, R. Raupach2, T. Allmendinger2, T. Flohr2, M. Kachelriess1 1 Institute of Medical Physics, University of Erlangen-Nuernberg, Erlangen, Bavaria, Germany 2 Siemens, Healthcare Sector, Forchheim, Bavaria, Germany

In cardiac perfusion examinations large concentrations of iodine pass through the ventricle and cause significant beam hardening artifacts. This leads to an underestimation of the time-attenuation curve (TAC) thereby yielding incorrect perfusion parameters. Beam hardening corrections are either implemented as simple precorrections which cannot account for higher order beam hardening effects, or as iterative approaches that are based on segmenting the original image into material distribution images. Conventional segmentation algorithms, however, fail to clearly distinguish between iodine and bone. We propose a new algorithm, DIBHC, that calculates the time-dependent iodine distribution by analyzing the time series of a cardiac perfusion examination and by determining the voxel changes for each time frame t (typically N=30 ECG-correlated scans distributed over a total scan time T=20 s). These voxel dynamics are due to changes in contrast agent. This prior information allows to precisely distinguish between bone and iodine, and is key to DIBHC where each iteration consists of a multi-material (soft tissue, bone, iodine) polychromatic forward projection, a rawdata comparison and a filtered backprojection. Simulations with a semianthropomorphic dynamic phantom and clinical scans using a dual source CT scanner (2x128 slices, 100 kV, 180 mAs, 0.28 s) have been carried out. The uncorrected images suffer from beam hardening artifacts that appear as dark bands connecting large concentrations of iodine in the ventricle and bony structures. The CT-values of the affected tissue are typically underestimated by up to 20 HU. One iteration of DIBHC greatly reduces these artifacts yielding CT-value deviations of only 1 HU for the simulations and 14 HU for the measurements. The accuracy for determining the baseline of the TAC has improved considerably (by 30 %).

388

M12-5: Low-dose Kilo-voltage Cone-beam CT Image Reconstruction by Constrained Total-variation Minimization: Experience with Clinical Data

X. Han1, J. Bian1, E. Pearson2, S. Cho1,2, E. Y. Sidky1, C. A. Pelizzari2, X. Pan1,2 1 Radiology, The University of Chicago, Chicago, IL, USA 2 Radiation and cellular oncology, The University of Chicago, Chicago, IL, USA

Current clinical radiation therapy makes extensive use of linear-accelerator integrated, kilo-voltage (KV) cone-beam CT (CBCT) as an image guidance tool. Compared to mega-voltage (MV) CT, the lower operating energy range offers improved soft-tissue contrast, but also causes more X-rays absorbed by the patient and increases imaging dose. An imaging technique capable of yielding clinically useful information at a minimum cost of patient dose is highly desired. We recently developed an interative reconstruction method based on constrained minimization of image total variation (TV), and showed its success in a series of numerical simulations. However, real patient data from clinical scans contain more complex anatomical structures, and are contaminated by higher inconsistency, which pose challenges to applying TV-minimization algorithm for clinical applications. Therefore in this work, we investigate and demonstrate the performance of the TV-minimization algorithm using real patient data. A clinical on-board KV CBCT system with a low-dose protocol was used to scan a patient at the head-neck region. We then extracted a small fraction of the total 641 projections and used our TV-minimization algorithm to reconstruct the image. The reconstructed image was then compared with images reconstructed by the FDK algorithm from the same under-sampled data and from the full data. The results showed that for images reconstructed from 30 - 100 projections, the TV-minimization algorithm successfully recovered most anatomical structures and the streak artifacts seen in FDK images were significantly suppressed.
M12-6:

B. De Man1, A. Caiafa1, Y. Cao1, K. Frutschy1, D. Harrison1, L. Inzinna1, R. Longtin1, V. B. Neculaes1, J. Reynolds1, J. Roy1, J. Short1, J. Uribe1, W. Waters1, X. Zhang1, Y. Zou1, R. Senzig2, N. Pelc3 1 GE Global Research Center, Niskayuna, NY, USA 2 GE Healthcare, Waukesha, WI, USA 3 Stanford University, Stanford, CA, USA Third-generation CT architectures are approaching fundamental limits. Dose-efficiency is limited by finite detector efficiency and by limited control over the X-ray flux spatial profile. Increasing the volumetric coverage comes with increased scattered radiation, conebeam artifacts, Heel effect, wasted dose and cost. Spatial resolution is limited by focal spot size and detector cell size. Temporal resolution is limited by mechanical constraints, and alternative geometries such as electron-beam CT and dualsource CT come with severe tradeoffs in terms of image quality, dose-efficiency and complexity. The concept of multi-source inverse-geometry CT (IGCT) breaks through several of the above limitations, promising a low-dose high image quality volumetric CT architecture. In this paper, we present recent progress with the design and integration efforts of the first gantrybased multi-source CT scanner.
This work was funded partially by NIH grant R01 EB006837. M13: MIC Posters 3 Saturday, Oct. 31 M13-3: 10:30-12:30 Grand Ballroom 4&5; Palm 3,4&5

Multi-Source Inverse-Geometry CT: from System Concept to Research Prototype

J. Bian1, X. Han1, K. Yang2, N. Packard2, E. Sidky1, J. Boone2, X. Pan1 1 Department of Radiology, The University of Chicago, Chicago, IL, USA 2 Department of Biomedical Engineering, University of California Davis, Sacramento, CA, USA

A Feasibility Study of Breast CT Imaging with Substantially Lowered Radiation Dose

In the last several years, prototype CT scanners for dedicated breast imaging have been built and under evaluation. Most of these breast CT systems employ circular-scanning configurations for acquiring projection data at a relatively large number of views (i.e., typically around 400 to 500). Because the total dose of a breast CT scan is generally controlled to be at the same level as that in a typical two-view mammogram. Therefore, the signal-to-noise ratio (SNR) of the data at each view in breast CT is relatively low, and it is generally considered a challenging task to reconstruct breast images from such low-SNR data collected at a number of views significantly lower than 400 to 500 views. In this work, we investigate image reconstruction from data acquired from a small number of views in breast CT. A motivation of the work is that it is always desirable to further reduce the radiation dose used in breast CT imaging provided that images of practical utility can be reconstructed from such data. This is particularly true if the breast CT is considered for the general screening purpose. Inspired by the compressive sensing observation, we have developed algorithms for image reconstruction from fan- and cone-beam data collected at highly sparse views through minimization of the total variation (TV) of the image, subject to the condition that the estimated data is consistent to the measured data. In our study, we will reconstruct images by 389

using the TV-minimization algorithm from phantom and patient data acquired with a dedicated breast CT scanner at numbers of views that are substantially lower than what are currently used in breast CT imaging. The aim of the study is to investigate and demonstrate the feasibility of reconstructing breast images from such low-number-of-view data, and thus of further reducing the imaging dose substantially in breast CT scans.
M13-6: Cross-strip capacitive multiplexing and electro-optical coupling for silicon photomultiplier arrays for PET detectors

P. D. Olcott1,2, H. Peng1, C. S. Levin1 1 Radiology, Stanford University, Stanford, CA, USA 2 Bio-engineering, Stanford University, Stanford, CA, USA

A key component for the development of simultaneous, PET/MR is a PET block detector that has a low number of readout channels, non-magnetic components, and little or no mutual influence between PET and MR systems. We have developed a differential multiplexing circuit for silicon photomultipliers (SiPM) that uses capacitors instead of resistors in a way that preserves their coincidence time performance. We demonstrated that a 4 x 4 array of 3 mm x 3 mm SiPM devices can be multiplexed into four signals with excellent spatial, energy (15.9 +/- 0.4% FWHM at 511 keV) , and timing resolution (1.4 ns FWHM) using a variety of scintillation crystal designs. Output signals from the multiplexing circuit can directly drive telecommunication-grade lasers without using active amplifiers to transmit the energy and fast timing information of the scintillation block detector out of the MR, using multi-mode optical fibers, rather than coaxial cables, using a custom designed laser alignment block. This multiplexed, laser coupled block detector has a significant reduction in the number of readout channels while having a very low electrical footprint. These two technologies will be a key enabler of SiPM technology for high resolution small animal and clinical PET/MR.
This research has been funded in part by a GE research grant, and the 2009 SNM Predoctoral Molecular Imaging Award. M13-9:

A. Kolb1, E. Lorenz2, D. Renker3, R. Grazioso4, N. Zhang4, D. Henseler4, B. J. Pichler1 1 University of Tuebingen, Laboratory for Preclinical Imaging and Imaging Technology, Tuebingen, Germany 2 Max Planck Institute for Physics, Muenchen, Germany 3 Paul Scherer Institute, Villigen, Switzerland 4 Siemens Medical Solutions, MI, Knoxville, USA The focus of this study was to build a Geigermode Avalanche Photodiode (G-APD) based PET block detector with a high multiplexing factor and to determine desirable structure specifications for future G-APD detectors. In order to determine the most appropriate structure specifications for the block detectors, G-APDs from various manufacturers (Hamamatsu, SensL, Zecotek) will be compared. Most G-APD manufacturers use a standard silicon n-on-p type structure with the exception of Hamamatsu and Zecotek which use a p-on-n structure. Typically, p-on-n devices have a higher quantum efficiency in the blue-light region (<470nm) than n-on-p devices which can significantly increase the G-APD photon detection efficiency. The signal-to-noise ratio of p-on-n devices is additionally enhanced by having a typically lower dark count rate than n-on-p devices [1]. These two effects make p-on-n type G-APDs more suitable in PET detector applications using LSO and similar scintillators emitting in blue light. The newly built block detectors using p-on-n type G-APDs will be compared to a previously built detector using n-on-p G-APDs from SensL. Moreover, block detectors will be built with a 4x4 array of G-APDs in order to readout a 15x15 LSO block of 1.5 mm x 1.5 mm x 10 mm crystals. Initially, G-APDs from two different vendors were coupled to a single LSO crystal (3 mm x 3 mm x 20 mm) and were evaluated in terms of energy and timing resolution. The G-APDs tested were the Hamamatsu S10391 and the Sensl SPMMicro3035 both having an active area of 3 mm x 3mm. A prototype detector was built using a 4x4 array of S10391 G-APDs coupled to a 12 x 12 LSO crystal block with an individual crystal size of 1.5 mm x 1.5 mm x 10 mm was used. The LSO crystal block/G-APD array combination was evaluated in terms of energy and timing resolution yielding an energy resolution range of 16-22% and a timing resolution of 1.4 ns in coincidence with a LSO/PMT detector.
[1] Renker, D. & Lorenz, E. Advances in solid state photon detectors. Journal of Instrumentation, P04004 (2009).

Development of G-APD-Based PET Block Detectors

390

M13-12:

N. Auricchio1,2, G. Di Domenico1,2, L. Milano1,2, R. Malaguti2, M. Ionica3, E. Fiandrini3,4, G. Zavattini1,2 1 Department of Physics, University of Ferrara, Ferrara, Italy 2 INFN, Sezione di Ferrara, Ferrara, Italy 3 INFN, Sezione di Perugia, Perugia, Italy 4 Department of Physics, University of Perugia, Perugia, Italy

Measurements for the SiliPET Project: a Small Animal PET Scanner Based on Stacks of Silicon Detectors

In this paper we propose a new scanner for small-animal positron emission tomography (PET) based on stacks of double sided silicon detectors. Each stack is composed of 40 planar detectors with dimension 60x60x1 mm3 and 128 orthogonal strips on both sides to read the two coordinates of interaction, the third being the detector number in the stack. Multiple interactions in a stack are discarded by an exclusive OR applied between each plane detector of a stack. In this way we achieve a precise determination of the interaction point of the two 511 keV photons. The reduced dimensions of the scanner also improve the solid angle coverage resulting in a high sensitivity. Preliminary results were obtained with MEGA prototype tracker (11 double sided Si detector layers, each with a thickness of 0.5 mm and a strip pitch of 470 m), divided into two stacks 2 cm apart made of respectively 5 and 6 prototype layers, placing a small spherical 22Na source in different positions. We report on the results, spatial resolution, imaging, spectral and timing performances obtained with double sided silicon detectors, manufactured by ITC-FBK, having an active area of 3x3 cm2 and a strip pitch of 500 m. Two different strip widths of 300 m and 200 m, and two thicknesses of 1 mm and 1.5 mm, equipped with 64 orthogonal p and n strips on opposite sides were read out with the VATAGP2.5 ASIC, a 128-channel general purpose charge sensitive amplifier. Each channel is composed of a low-noise/lowpower preamplifier, a shaper (peaking time of 3.1 s) with sample/hold, multiplexed analogue readout and a fast shaper (peaking time of 1.1 s) that gives a trigger signal.
M13-15:

J. Kataoka1, M. Koizumi2, H. Ikeda3, H. Matsuda1, T. Miura1, M. Yoshino1, S. Tanaka2, H. Ishibashi2, Y. Ishikawa4, N. Kawabata4, Y. Matsunaga4, S. Kishimoto5, H. Kubo6 1 Research Institute for Science and Engineering, Waseda University, Tokyo, Japan 2 Department of Physics, Tokyo Institute of Technology, Tokyo, Japan 3 Institute of Space and Astronautical Science, JAXA, Kanagawa, Japan 4 Solid State Division, Hamamatsu Photonics K.K, Shizuoka, Japan 5 Photon Factory, High Energy Accelerator Research Organization, Ibaragi, Japan 6 Department of Physics, Kyoto University, Kyoto, Japan We report on the development of versatile APD-based PET modules with time-of-flight capability. The module consists of a LYSO scintillator matrix optically coupled with a position-sensitive avalanche photodiode (APD) array, and front-end circuits (FEC) directly connected to the rear-end of the APD package. Each APD device has a monolithic 16x16 (or 8x8) pixel structure with an active area of 1.0 (or 4.0) mm^2 for each pixel. Time resolutions of 155 ps and 214 ps (FWHM) were obtained for 1.0 mm^2 and 4.0 mm^2 APD pixels, respectively, measured by the direct detection of 10 keV X-rays. The FEC carries two identical analog ASICs specifically designed for the APDs in TSMC 0.35 m CMOS technology. Each ASIC consists of 32-channel charge-sensitive amplifiers, band-pass filters, differentiators, pulse-height and timing discriminators, and two-channel time-toamplitude converters. The noise characteristic of the ASIC, mounted in a low temperature co-fired ceramics (LTCC) package, is 560 + 30 e/pF with an electric timing resolution of 484 ps (rms). The overall dimension of the module (including APD-array, LYSO matrix and FEC) is 30x30x80 mm^2. The variation of signal amplitude was less than 20 % among all pixels. The average energy resolutions of 11.7+-0.7 % and 13.7+-1.1 % were obtained for 662 keV gamma-rays, measured with 8x8 and 16x16 arrays, respectively. An attainable spatial resolution is < 0.8mm (FWHM) for 16x16 array in a reconstructed image. These results suggest the APD-based PET module can be a promising device for future applications, especially for high resolution MRI-PET and TOF-PET.
M13-18:

Versatile APD-based PET Modules for High Resolution, Fast Medical Imaging

X. He1, B. Gallas2, E. C. Frey1 1 Radiology, Johns Hopkins University, Baltimore, MD, USA 2 Division of Imaging and Applied Mathematics, Food and Drug Administration, Silver Spring, MD, USA Multi-class ROC analysis has remained an open problem since the introduction of binary ROC analysis in the 1950s. Previously, we have developed a paradigm for three-class ROC analysis that extends and unifies the decision theoretic, linear discriminant analysis, and the probabilistic foundations of binary ROC analysis in a three-class paradigm. One critical element in this paradigm is the equal error utility (EEU) assumption. This assumption allows us to reduce the intrinsic space of the three-class ROC analysis (5D hypersurface in 6D hyperspace) to a 2D surface in the 3D space of true positive fractions (sensitivity space). In this work, we show that a 2D ROC surface fully and uniquely determines the complete descriptor for the performance of a three391

Three-Class ROC Analysis the General Decision Theoretic Solution

class classification task : the triplet of likelihood ratio distributions, assuming such a triplet exists. To be specific, we consider two classifiers that utilize likelihood ratios, and we assume each classifier has continuous and differentiable 2-D sensitivity-space ROC surfaces. Under these conditions, we prove that the classifiers have the same triplet of likelihood ratio distributions if and only if they have the same 2-D sensitivity-space ROC surfaces. Thus the 2-D sensitivity surface is a complete descriptor of optimal three-class task performance.
M13-21:

Impact of Fully 4D Reconstruction on Kinetic Parameter Estimates

P. Gravel, J. Verhaeghe, A. J. Reader Montreal Neurological Institute, McGill University, Montreal, QC, Canada The spatiotemporal noise reduction offered by fully 4D reconstruction has potential to reduce noise in parametric images for PET when compared to sequential 3D reconstruction of dynamic scans. Since these images are composed of voxels representing estimated kinetic parameters, these estimates should be as close to the ground truth as possible to achieve accurate interpretation of the images. Conventionally, the bias and variance of these estimates is calculated based on simulated datasets, the latter defining the ground truth. One caveat of these simulations lies in the difficulty of accurate modeling of the data acquired with a real PET tomograph. This work uses a list-mode data temporal re-sampling method to assess the impact of fully 4D reconstruction on kinetic parameter estimation. In this way we effectively combine temporal simulations with real data. Using this approach, the potential of fully 4D reconstruction (with respect to the choice and number of temporal basis functions as well as the number of iterations) can be compared to conventional reconstruction, and in turn the parameterization of 4D reconstruction can be optimized for the specific kinetic parameters of interest.
M13-24:

GPU Accelerated Statistical Image Reconstruction for Compton Cameras

V.-G. Nguyen, S.-J. Lee, M. N. Lee Dept. of Electronic Engineering, Paichai University, Daejeon, Korea Although statistical methods are known to have great potential to improve quantitative accuracy in Compton camera reconstruction, the use of these methods has been hindered by the computational load due to repeated operations of conical projection and backprojection. In this paper we propose GPU (graphics processing unit) accelerated methods that can dramatically improve the performance in computation time. Since the conventional ray-based backprojection method is inefficient for GPU, we develop a fully voxel-based backprojection method using an approximated approach to calculating the intersecting area of a cone in the voxel for the system matrix. For each voxel, our method first measures the scattering angle determined by the detected position pair and the voxel location, and chooses the nearest cone which was predefined for forward projection. If the cone crosses the voxel, the intersecting area of the cone is approximately calculated by multiplying the three factors; (i) the number of the rays on the cone surface passing through the voxel, (ii) the percentage of the normal distance from the center of the voxel to the cone surface, and (iii) the maximum intersecting length of the ray in the voxel. Our experimental phantom studies with the OS-EM (ordered subsets expectation maximization) algorithm indicate that the GPU-based method is roughly 50 times faster in computation time per iteration than the CPU-based method. According to our experimental results, for acceptable images from OSEM with 64 subsets and 8 iterations, the CPU-based method takes about 2.3 hours, whereas the GPUbased method takes only 2.7 minutes. In summary the GPU-based method is inevitably more practical than the CPU-based method for Compton camera reconstruction.
M13-27:

A MR Compatible Brain PET Using Tileable GAPD Arrays

J. H. Jung, Y. Choi, K. J. Hong, J. H. Kang, W. Hu, B. J. Min, Y. S. Huh, S. H. Shin, H. K. Lim, D. S. Kim, H. B. Jin Dept. of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea The aim of this study is to develop a MR compatible PET that is insertable to MRI and allows simultaneous PET and MR imaging of human brain. The brain PET having 72 detector modules arranged in a ring of 330 mm diameter was designed. Each PET module composed of 4x4 matrix of 3x3x20 LYSO crystals coupled to a tileable 4x4 array Geiger-mode avalanche photodiode (GAPD) and designed to locate between RF and gradient coils. Signals of the each module were fed into preamplifiers using flat cable of 3 m long, and then sent to a position decoder circuit (PDC), which readouts digital address and an analog pulse of the one interacted channel from preamplifier signals. The PDC outputs were fed into FPGA-embedded DAQ boards. The analog signal was digitized, and arrival time and energy of the signal were calculated and stored. All electronics were located outside MR bore to minimize signal interference between PET and MR. Cross-compatibilities of the PET module and MR were examined. A proof-of-principle PET consisted of 8 pairs of the PET module covering arc of 80 was constructed and the PET performance were evaluated outside MR bore. The differences of the PET module performance measured inside/outside MR bore were negligibly small. The SNR of various MR sequence phantom images acquired with/without the PET module were also similar. Energy, time and spatial resolutions of the PET were 21%, 2.0 ns and 3 mm, respectively. Activity distribution patterns of hot and cold rod phantoms were well imaged without distortion, and rods down to a diameter of 3.2 mm were resolved. A full ring PET including mechanics and magnetic shielding to operate the PET inside MR bore is being designed and 392

constructed. MR compatibility of the insertable PET is under investigation. Preliminary experimental results demonstrate that MR compatible high quality PET imaging is feasible using the GAPD arrays, electronics, signal processing method and MR insertable PET design schemes developed in this study.
M13-30:

Development of PET Detectors Using Monolithic Scintillation Crystals Processed with Sub-Surface Laser Engraving Technique

T. Moriya, K. Fukumitsu, T. Sakai, S. Osuka, T. Okamoto, H. Takahashi, M. Watanabe, T. Yamashita Hamamatsu Photonics K.K., Hamamatsu, Japan New monolithic scintillation detectors for PET have been developed, where the crystals are processed using sub-surface laser engraving (SSLE) technique. When high intensity light pulses of short duration from a laser are focused into a scintillation crystal, they induce multi-photon absorption in the focal point and result in refractive index changes or micro-cracks inside the crystal. By applying the SSLE technique to a monolithic scintillation block, fine segmentation in the crystal can be formed without inter-pixel gaps. We have fabricated various graven patterns inside monolithic LYSO crystal blocks by using a YAG laser. The engraved crystals were coupled with a position-sensitive PMT and their 2D position histograms were measured for uniform gamma-ray irradiation. Preliminary results suggest that it is possible to fabricate high performance PET detectors using SSLE technique.
M13-33:

K. Shi1, S. Astner1, M. Souvatzoglou2, I. Miederer2, J. Wilkens1, F. Nuesslin1, M. Molls1, S. Ziegler2 1 Dept. Radiooncology, Technical University Munich, Munich, Bavaria, Germany 2 Dept. Nulear Medicine, Technical University Munich, Munich, Bavaria, Germany

Quantitative Assessment of Hypoxia Kinetic Models by a Cross-Study of Dynamic 18F-FAZA and 15O-H2O in Head and Neck Tumor

Kinetic modeling is one important method to assess the underlying physiology behind tracer uptake in molecular imaging. Although there are many well developed models which cover a broad range of applications, it is still challenging to quantitatively assess mathematical models with consideration of clinical applications and their biological nature. Tumor hypoxia is considered as one main resistance factor of both radio- and chemotherapy. Besides an increase of oxygen consumption, hypoxia is usually observed with a decreased supply of oxygen, which is strongly related to perfusion within a tumor. It is expected that hypoxia occurs more likely with a decreased blood supply. Under this assumption, we made an experiment to compare different hypoxia kinetic models. Dynamic PET images of hypoxia tracer 18F-FAZA and perfusion tracer 15O-H2O were acquired and Thorwarth model, reversible and irreversible two-tissue compartment model, Logan plot and Patlak plot were applied to model the process of tracer transport and accumulation under hypoxic condition. With the cross analysis between these two specific tracers, it is shown that kinetic modeling delivers significantly different information than static measurements. Different models have a large variation under the same condition and they even lead to opposite physiological meaning when interpolated. Our result shows that irreversible two compartment corresponds better to the expectation of negative correlation between hypoxia and perfusion.
M13-36:

J. Jones1, E. Lewis1, M. Guy2, K. Wells1 1 Centre for Vision Speech and Signal Processing, University of Surrey, Surrey, United Kingdom 2 Department of Physics, University of Surrey, Surrey, United Kingdom

A Virtual Dissection Based Registration to Model Patient Specific Respiratory Motion

Improvements in PET/SPECT image resolution are fuelling the effort to reduce corruption due to voluntary and involuntary patient motion. A single affine transformation covering an entire torso has proved to be incapable of accurately expressing the correction needed. Rather than resort immediately to non-rigid registration techniques the possibility of doing a piecewise rigid transformation is being investigated as part of a wider motion correction approach which will include stereo tracking of the patient during acquisition and the use of a particle filter to account for variations in respiratory motion. A virtual dissection of the upper torso is achieved by first segmenting, from low-dose CT data, the major organs. We then use the ribs, being the more easily segmented features, as a framework for dividing the residual tissue volume. This produces 37 distinct surface components which are then individually registered using the Iterated Closest Point algorithm. The registrations are then expressed as motion fields extending throughout each organ and, in the case of ribs, beyond the ribs into the surrounding tissue. A single mosaic motion field is constructed using the virtual dissection of the CT data which is then applied to the corresponding gated SPECT images. This approach has been developed and tested using the NCAT virtual phantom, and using the RMDP software to derive synthetic CT and corresponding SPECT image data. Comparison is made between this scheme and an alternative using a conventional non-rigid registration. The virtual dissection based approach shows promising results.

393

M13-39:

Adaptive SPECT Imaging with Variable System Modeling Techniques

L.-J. Meng, N. Li Nuclear, Plasma and Radiological Engineerng, University of Illinois at Urbana-Champaign, Urbana, IL, USA This paper presents an adaptive imaging procedure for use with a single photon microscopic (SPEM) system. The SPEM system consists of multiple ultrahigh resolution camera heads, each coupled to a focusing micro-pinhole aperture to provide an excellent spatial resolution for a small target region. In this work, we explored the use of adaptive system modeling (ASM) techniques to improve the imaging performance. For SPEM studies, in which one is interested primarily in a local target region, we would use smaller pixels in the target region and other regions that are known to be important to the reconstruction of the target region, and to use larger pixels for the rest of the object-space. This non-uniform pixelation process is adjusted in real time in response to user input and the imaging information being collected. The adaptive system modeling (ASM) technique leads to a significantly reduced number of unknowns in the reconstruction/inverse problem and therefore delivers a greatly reduced variance in the reconstruction of the target region. In this study, we will use both Monte Carlo simulation and experimental studies to demonstrate the benefit of using the ASM technique. The results from these studies will be presented.
M13-42:

Energy Dependent X-Ray Phase-Contrast Imaging Using Grating-Interferometry

P. Bartl, T. Michel, G. Anton ECAP, University of Erlangen-Nuremberg, Erlangen, Germany We present a complete simulation framework for X-ray phase-contrast imaging inheriting the wave-like as well as the particle behavior of photons. The developed tool includes the modeling of a partially coherent X-ray source, the propagation of the X-ray photon through samples, and a full implementation of a detector, which in particular can be chosen to be the Medipix- or Timepix-detector. Thus the simulation is capable to physically model a grating-based interferometric imaging system. Results on such a setup concerning spatial and temporal coherence will be shown. The extent of the used energy range herein was chosen to coincide with the medical imaging range. Samples consistent of elements and structures similar to biological tissue were implemented to demonstrate the applicability on medical imaging. Another key component of the setup was the implementation of a Medipix-detector aiming to describe forthcoming energy-resolved measurements. The simulation splitting the influences of each imaging component advances a profound understanding of X-ray phase-contrast imaging and additionally can be used to optimize the grating-interferometry towards a broader medical application.
M13-45:

K. Taguchi1, H. Kudo2, W. C. Barber3 1 Radiology/Med Imag Phys, Johns Hopkins University, Baltimore, MD, U.S.A. 2 Comp Sci, Sys and Info Eng, University of Tsukuba, Tsukuba, Ibaraki, Japan 3 DxRay, Inc., Northridge, CA, U.S.A. Photon counting x-ray CT (PCXD) detectors with energy discrimination capabilities offer various advantages over the current intensity integrating detectors, such as improved accuracy in material decomposition, reduced image noise, etc. The major problem of PCXDs, however, is the limited speed. The required count rate for clinical x-ray CT scanners is ~1 x 10^9 counts-persecond-per-square-millimeters (cpspmm2), while the operational count rate of the world fastest PCXD is merely 3.5 x 10^5 cpspmm2. In this study, we propose to enable PCXD-based clinical CT imaging by using the following four methods. First, we designed bowtie filters for various sizes of objects. Second, we derived analytical PCXD models of the pulse shape, pulse pileup effect, and count rate loss for both paralyzable and non-paralyzable detector types. Third, we integrated the PCXD models into a penalized maximum-likelihood sinogram restoration method. The algorithm provided multiple sinograms of basis functions through the material decomposition process. Fourth, we applied an exact interior region-of-interest reconstruction algorithm to reconstruct density distributions of the basis functions. Computer simulations provided excellent results. The newly designed bowtie filter reduced the required count rate down to 1.7 x 10^7 cpspmm2. The PCXD models showed an excellent agreement with the results of Monte Carlo simulations. The penalized maximum-likelihood method allowed us to compensate for the spectrum degradation caused by the limited count rate, and thus, improve the operational count rate to 3.5-7.0 x 10^6 cpspmm2. Thus, almost all points inside the object except for those near skins had accurate projections sufficient for exact reconstructions. With the air cavity as a priori, Kudos algorithm allowed for accurate reconstruction of density distributions of the basis functions. We concluded that the proposed method enabled PCXD-based clinical x-ray CT imaging.
This work was supported in part by NIH R44 EB008612 and the start-up fund of DMIP in JHU.

Enabling Photon Counting Clinical X-Ray CT

394

M13-48:

Y.-C. Tai1, D. B. Keesing2, H. Wu1, T. Y. Song1, S. Komarov1, J. A. O'Sullivan3 1 Radiology, Washington University in St. Louis, St. Louis, MO, USA 2 Biomedical Engineering, Washington University in St. Louis, St. Louis, MO, USA 3 Electrical and Systems Engineering, Washington University in St. Louis, St. Louis, MO, USA

Correction Techniques for a Half-Ring PET Insert System to Locally Enhance the Image Resolution of a Clinical PET/CT Scanner

We have previously developed a half-ring PET insert system that can be integrated into a clinical PET-CT scanner to improve its image resolution within the central region of the FOV. A statistical reconstruction framework was established to locally enhance the image resolution near the insert detectors while preserving the entire imaging FOV of the PET scanner. In testing this unique capability, we realized that conventional correction techniques such as normalization, attenuation and scatter corrections need to be redeveloped for this complex system. In this work, we present correction techniques developed for this system and their initial results. The attenuation correction contains two components: correction for the insert device itself is based on forward projection of the AutoCAD drawing of the insert gantry with the attenuation coefficients of the insert components experimentally measured by a transmission scan or CT. The attenuation correction for the subject is based on CT images, same as regular PET/CT scans. Scatter correction is modified from the commonly employed single-scatter-simulation technique, including models of the highly attenuating LSO detectors in the insert. Normalization is based on a maximum-likelihood estimation of detector efficiency and other geometric correction factors, which are compounded to compute normalization factor for each LOR. Default random correction based on delayed window technique was validated using experimental data. Initial results suggest that the proposed new correction techniques adequately model the targeted physical effects. When these techniques are fully integrated into the already established statistical reconstruction framework, we anticipate that the image uniformity and contrast will be improved, and so is the tumor detectability in phantom studies.
This work was funded by the NCI of NIH (CA110011) and the Susan G. Komen for the Cure (BCTR0601279). M13-51:

S. L. Bowen1, F. Godinez1, L. Fu1, J. Qi1, R. D. Badawi2 1 Department of Biomedical Engineering, UC Davis, Davis, CA, U.S.A. 2 Department of Radiology, UC Davis Medical Center, Sacramento, CA, U.S.A. Dedicated breast PET/CT is expected to have utility in the local staging, surgical planning, assessment of therapy response, and detection of residual disease for breast cancer. The reconstruction of accurate quantitative breast PET images may significantly improve patient care in such. We present a validation of data correction schemes for a custom built dedicated breast PET/CT scanner via phantom scans. A variance reduced randoms estimate was computed from delayed coincidences using the Casey and Hoffman method. Attenuation factors were calculated by using a CT based segmentation scheme. Live-time was estimated on a projection-by-projection basis with a multi-component model using singles and coincidences estimated from digital counters alone and with the PET DAQ system, respectively. Raw data was subject to a 350-650 keV energy window and reconstructed with FBP or a Maximum a posteriori (MAP) algorithm. To assess the accuracy of randoms and dead-time corrections as a function of activity a decay experiment was performed with an 18F-FDG filled cylinder and residual error analysis performed according to NEMA. A high statistics low activity scan was acquired for a three component phantom, composed of a large chamber (background) (inner diameter = 8.9 cm) and two internal cylinders as a preliminary validation of the attenuation correction method. Concentration ratios were set as follows: 1:9:0 (background : hot cylinder : cold cylinder), and contrast recovery coefficients were calculated according to NEMA. Residual error for the randoms and dead-time corrections was 11% at a phantom activity with singles flux comparable to that measured during patient imaging. With attenuation corrected MAP images hot rod contrast improved significantly compared to uncorrected images, but still suffered from contamination by scatters. In the future we plan on implementing a scatter correction scheme and rigorously validating the CT based attenuation correction.
M13-54:

Quantification with a Dedicated Breast PET/CT Scanner

A. T. Cebula1, M. T. Studenski1, A. Samarin2, D. R. Gilland1 1 Dept. of Nuclear and Radiological Engineering, University of Florida, Gainesville, FL, United States 2 Dept. of Radiology, University of Florida and North Estonia Medical Center, Talllin, Estonia

Evaluation of a Bedside SPECT System Based on Cardiac Phantom Defect Detection

This paper discusses the ability of a prototype mobile SPECT system to detect a cardiac defect. The system is designed for general purpose imaging including both low and high energy emitters in order to perform, for example, myocardial perfusion (Tc-99m-sestamibi) and viability (F-18-FDG) studies of immobile patients. The evaluation was conducted using an anthropomorphic thorax phantom for both Tc-99m and F-18 activity distributions. For both cases, high count, low noise data were acquired with 1 mCi in the myocardium and 2 mCi in the liver to simulate out-of-field activity. High count projection data were acquired using 25 projection angles over 180 degrees for both energies. Low count projection data were generated using binomial thinning to reduce the count level by a factor of four. Reconstruction of the projection data was performed using 50 395

iterations of MLEM. Line profiles through the myocardial defect region were generated and the reconstructed images were compared. The results demonstrate the capability of the system to image a defect in a thorax phantom for both Tc-99m and F-18 activity distributions.
M13-57:

A Staggered Array of Pinhole Cameras for Dedicated Breast SPECT

C. R. Tenney, H. K. Dhah, J. W. McCurley Physics, California State University, Fresno, Fresno, CA, USA A staggered detector geometry is proposed for pinhole camera SPECT imaging of breast cancer. Several such systems are simulated in two dimensions, with reconstructed images evaluated for spatial resolution, noise, and artifacts. Designs for dedicated breast SPECT scanners generally involve moving a detector around the breast. Multiple detectors have greater detection efficiency. With enough detectors, a static array can be used. Imaging detectors are expensive, but advancing technology continues to reduce costs. However, for a given object size and camera size, a limited number of cameras can be placed around the object. This in turn limits the achievable combination of sensitivity and spatial resolution. A second row of pinhole detectors may be placed so as to image through small gaps between the detectors in the first row. This could double the number of cameras, although the sensitivity would be less than doubled as the second row of cameras would be farther from the object. This arrangement would also have better tomographic ray coverage than a single row. The front row of cameras could also be moved closer to the object, for improved spatial resolution and/or sensitivity. Parts of the object's periphery might not imaged by the front row of cameras, but these would be imaged by the second row, albeit with less sensitivity and resolution. A staggered pinhole array could be built in a planar arrangement, suitable for imaging a compressed breast. Slit-slat collimators could be used in such a configuration. Other applications suggest themselves as well, e.g. small animal imaging. We evaluate several staggered arrays and compare them with corresponding single-row arrays. Initial studies are in two dimensions, representing either a central slice from a pinhole array, or a single plane in a slit-slat array (neglecting axial blurring). In simulation, test objects are imaged using different camera arrays, and the projections are used to reconstruct images for analysis.
This work is supported in part by a Faculty-Student Collaborative Research Seed Grant from CSUPERB, the California State University Program for Education and Research in Biotechnology. M13-60:

E. Roncali1, Y. Yang1, M. McClish2, P. Dokhale2, C. Stapels2, E. Johnson2, J. Christian2, K. S. Shah2, S. R. Cherry1 1 Dept. of Biomedical Engineering, University of California-Davis, Davis, CA, USA 2 Radiation Monitoring Devices Inc., Watertown, MA, USA Due to their excellent overall performance characteristics, silicon photomultipliers are promising for numerous emerging applications and technologies, and are particularly suitable for constructing compact detector modules such as for small animal PET scanners. Designing new silicon photomultipliers with enhanced features (higher photodetection efficiency and larger active area) has been a very active area of development for the last recent years. Here we present the characterization of position sensitive CMOS solid state photomultipliers (PS-SSPMs) using LSO and LYSO scintillator crystals. Spatial and timing resolutions were measured on LYSO crystals. Flood imaging of LYSO and LSO crystals has also been performed. Depth of interaction (DOI) resolution measurements have been performed using dual-ended readout of LYSO crystals by PS-SSPMs. Pulse shape analysis for DOI-encoding also was evaluated using signals recorded by a phosphorcoated LSO crystal read out by a PS-SSPM. 0.5 mm crystals could be resolved in LSO and LYSO arrays and an intrinsic spatial resolution of 70 m was measured for a highly collimated light source. 1 ns timing resolution and 15% energy resolution was measured on a 1.5 x 1.5 x 20 mm3 LYSO crystal. A depth of interaction resolution of 2.5 mm was obtained with a dual-ended readout setup. Experiments conducted with a phosphor-coated LSO crystal showed a relationship between the depth of irradiation and the pulse shape. Amplitude and decay times were found to be greater and longer when interactions occurred at the near end of the crystal. Development of larger devices is in progress and new PS-SSPM elements will be tested.
M13-63:

Position-Sensitive Solid State Photomultipliers for PET Imaging

S. Yamamoto1, M. Honda2, K. Shimizu3, M. Senda3 1 Kobe City College of Technology, Kobe, Japan 2 National Center of Neurology and Psychiatry, Tokyo, Japan 3 Institute of Biomedical Research and Innovation, Kobe, japan

Development of PET-Hat: Wearable PET System for Brain Research

We developed a low cost, small size, wearable brain PET system for brain research named PET-Hat. The PET-Hat employed double counter-balanced systems for the mechanical supports. The detector ring can freely move with the counter balanced systems. These motions enable subject to move relatively freely with the PET during acquisition. The detector consists of a GSO block, a tapered light guide, and a flat panel photomultiplier tube (FP-PMT). Two types of GSOs are used to form depth-of396

interaction (DOI) detector to minimize the ring diameter. The tapered light guide is used to increase the size of the GSO blocks and reduce the number of FP-PMTs. Sixteen detector blocks are arranged in a 280mm diameter ring. Trans-axial and axial fieldof-view (FOV) are 22cm and 5cm, respec-tively. Energy resolution of the block detectors was ~15% FWHM and timing resolution was ~4.6nsFWHM. Transaxial resolution and axial resolution at the center of the FOV were ~4.5mmFWHM and ~3.5mmFWHM, respectively. Sensitivity was ~1% at the center of the axial FOV. Hoffman brain phan-tom images are successfully obtained. We conclude that the PET-Hat is promising, low cost, small size, wearable brain PET system for brain functional studies.
M13-66:

J. S. Scheuermann1, S. Surti1, J. A. Kolthammer2, J. S. Karp1 1 Radiology, University of Pennsylvania, Philadelphia, PA, USA 2 Philips Healthcare, Cleveland, OH, USA

Evaluation of a fully 3D, big bore TOF PET scanner with reduced scatter shields

Traditionally, PET scanners have annular lead shielding at the axial ends that extends beyond the crystals to reduce the number of photons from outside of the axial field of view (FOV) hitting the detector. In recent years there has been a trend toward reducing the end shielding in order to increase the port diameter of the PET scanner. The reduction of the lead shielding could have performance effects that could affect overall image quality. The Univ. of PA has both a Gemini TF Big Bore and a standard Gemini TF, with the major difference between the two systems being the reduction of the end shielding on BigBore. As expected, we saw an increase in scatter fraction for a given energy gate in BigBore versus TF. To compensate for the increase in scatter fraction, the low energy threshold was raised; resulting in a 6% decrease in sensitivity and 10% decrease in peak NEC with a large diameter phantom. Timing resolution measurements at low count rates are better on BigBore and timing resolution increases similarly as a function of singles rate for the two systems. Singles rate increases with activity more on Big Bore, such that in the clinical activity range, timing resolutions of the systems are similar. While there are differences in the performance of BigBore as compared to TF, it is expected that the overall image quality of studies obtained on BigBore will be comparable to those obtained on TF. While we are not expecting significant effects on images quality, it is important to understand how the differences in the performance measurements and the changes made to compensate for those differences may affect image quality. The system default settings will be investigated to see if they can be further optimized to increase sensitivity and peak NEC while maintaining quantitative accuracy. Image quality phantom measurements will be performed to evaluate the quantitative effects of the increased demand on the data corrections caused by the removal of the scatter shields.
M13-69:

A Study on the Timing Performance of PMT-Quadrant-Sharing LYSO Position-sensitive Detector Blocks for Time-of-flight PET

S. An, H. Li, S. Liu, R. A. Ramirez Jaramillo, Y. Zhang, C. Wang, H. Baghaei, W.-H. Wong Experimental Diagnostic Imaging, Univ. of Texas M.D. Anderson Cancer Center, Houston, U.S.A. Time-of-flight (TOF) information can improve the imaging quality of positron emission tomography (PET) camera by calculating the original positron-electron annihilation position. In this work, we measured the preliminary TOF resolution of nonTOF-optimized LYSO arrays for TOF PET using our PMT-Quadrant-Sharing (PQS) technology which has higher light collection efficiency (no external light guide). We investigated signal-processing techniques that may achieve better time resolution. Digital oscilloscope Tektronix DPO7054 was used to capture the waveforms. The timing methods, Linear Fitting (LF) to the pulse rising time, Constant Fraction Discriminator (CFD) and Leading edge Discriminator (LD), were studied. A new Time-Energy (T-E) correction using Multivariate Least Mean Square Fit was applied to minimize time walk caused by the variation of signal shape. The time resolution as a function of the depth of interaction was also studied. The average block-to-block coincident timing resolution of 13*13 (4*4*20 mm3 detector pixel) array with large PMT R9779 (39-mm diameter) was measured to be 432 ps (FWHM). Another 13*13 crystal array with ultrahigh spatial resolution 1.4*1.4*10mm3 detector pixels coupling to small 19mmdiameter poor-timing XP1912 PMTs was also studied and achieved 551 ps. This study shows that our regular PQS detector blocks (without optimization for TOF) have good TOF time resolution suitable for TOF PET systems, in addition to higher spatial resolution and reducing PMT cost.
M13-72:

J. K. Poon1, S. St. James1, S. R. Cherry1, R. D. Badawi2 1 Biomedical Engineering, University of California, Davis, Davis, CA, USA 2 Radiology, University of California Davis Medical Center, Sacramento, CA, USA

An Experimental Evaluation of Triangular Prism Shaped LSO Crystals for Large Detector Blocks

Introduction: Conventional block detector modules in clinical positron emission tomography (PET) scanners consist of a matrix of square or rectangular prism shaped scintillator crystals read out with four photomultiplier tubes (PMT). To reduce the number of electronic channels and PMTs for a detector block without sacrificing detection efficiency and surface area coverage, we propose using triangular prism shaped crystals that form triangular shaped blocks, which can be read out with only three PMTs. The objectives of the study are to characterize and compare the performance of triangular prism crystals against square prism 397

crystals based on the following parameters: (1) light output, (2) energy resolution, and (3) the ratio of acceptable to total counts. Methods: Each set of LSO crystals consists of six optical-grade polished prisms that are 20 mm in length. The equilateral triangular and square base prisms each have a set with a base area of 9 mm2 and 64 mm2. A crystal was coupled to a PSPMT with optical grease and bench-top NIM electronics were used to perform logic operations and pulse shaping. Crystals were experimentally tested 6 times in random order and the mean values of each parameter were used to calculate the overall average parameter values for each set of crystals. Results: The mean energy resolution, mean photopeak position, and mean ratio of acceptable to total counts of the square prism were improved by factors of 4.7%, 7.3%, and 2.3% compared to the triangular prism respectively for the 9 mm2 prism base. However, the larger 64 mm2 base prisms showed minimal difference in performance when comparing the three parameters. Conclusions: The experimental data shows triangular prism crystals could potentially be used for triangular shaped block detectors, since the performance penalty in light output, energy resolution, and photofraction compared to square prism crystals is insignificant.
M13-75:

K. Ogawa1, T. Ishikawa1, K. Shuto2, H. Kobayashi2, T. Nakahara3, N. Shigematsu3 1 Hosei University, Faculty of Science and Engineering, Tokyo, Japan 2 Toshiba Medical Systems, Tochigi, Japan 3 School of Medicine Keio University, Tokyo, Japan

Development of a Prototype Semiconductor Gamma-Camera System

We made a prototype gamma camera system with CdZnTe semiconductor detectors that work at room temperature. The effective field of view of this detector is 195 156 mm2 with a pixel size of 2.46 2.46 mm2. We evaluated the performance of the detector in terms of the energy resolution, intrinsic spatial resolution, system spatial resolution and system sensitivity. The performance of the detector was evaluated with methods based on the NEMA standard (NU1). The results showed that the mean energy resolution was 5.58 % FWHM for 99mTc, the intrinsic spatial resolution 2.5 mm FWHM, the system spatial resolutions 7.1 mm FWHM (without scatterer) and 7.4 mm FWHM (with scatterer), and the system sensitivity 46.3 cps/MBq.
M13-78:

Performance Estimation of High Resolution SPECT for the Head by Monte Carlo Simulation of Scintillation Lights

Y. Hirano, T. Zeniya, H. Watabe, H. Iida Department of Investigative Radiology, Advanced Medical Engineering Center, National Cardiovascular Center Research Institute, Suita, Oska, Japan We have been developed a high resolution Single Photon Emis-sion Computed Tomography (SPECT) for the head aiming at clinical assay. This SPECT has two kinds of detectors and rotated simultaneously. One detector views whole head with parallel collimator, the other detector views local region with extremely resolution (~1mm) using pinhole or cone beam collimator. The detector consist of NaI scintillator (15cm x 20cm), 12 flat panel type multianode PMTs (H8500 manufactured by HAMAMATSU, 5cmx5cm) and collima-tors. The PMT has 64 small anodes (5.2mm x 5.2mm). High resolution has achieved by the Anger method using 768 anodes. Currently the detector with parallel collimator has installed. On the other hand, we have plan to use LaBr3(Ce) scintillator for the local region to improve resolution. Higher resolution is expected using LaBr3(Ce) because of large amount of scintillation lights which is about 1.6 times larger than that of NaI. Hence, we estimated how much resolution will be obtained using LaBr3(Ce) and the influences by the differences of optical properties between NaI and LaBr3(Ce) such as transparency and refractive index. Scintillation lights are simulated with Monte Carlo method. We made comparison of position resolution between NaI and LaBr3(Ce) with another detector system after validateing this simulation. LaBr(Ce) has better resolution than that of NaI. The position resolutions of NaI and LaBr3(Ce) are 2mm, 1.7mm respectively. In addition, we investigated what is the main factor contributing position resolution by changing optical properties or scintillator and size of materials using this simulation. In order to make high resolution system, we should estimate performance by some simulations previously and this simulation will be useful to design detector system.
M13-81:

Evaluation of a Cardiac SPECT System Using a Common Set of Solid-State Detectors for Both Emission and Transmission Scans and a Ultras-Low Dose Lead X-Ray Transmission Line Source

C. Bai, H. Babla, J. Kindem, R. L. Conwell Digirad Corporation, Poway, CA, USA We developed a new cardiac SPECT system (XACT) with transmission-based attenuation correction (AC). Three solid-state detectors are configured to form a triple-head system for emission scans and reconfigured to form a large field-of-view single head system for transmission scans. The transmission line source is the collimated fluorescence X-ray emission from a lead piece when it is fluxed by a narrow X-ray beam from an X-ray tube. When the X-ray tube operates at 160 kVp and 2.0 mA, the line source has a flux density of 80 times that of conventional isotopic sources. High quality transmission scans can hence be completed in as short as one minute with insignificant patient dose (~5 Sv). For evaluation, we first scanned an anthropomorphic phantom with a uniform cardiac insert. We then scanned the phantom with two 45-degree defects (one full and 398

one 50%) in the cardiac insert. We also scanned an ACR phantom and visually assessed the image quality. Results showed that image uniformity and defect contrast were improved by AC in the anthropomorphic phantom studies. The 17-segment scores of the images of the uniform cardiac insert were, average and standard deviation, 78 and 6 before and 88 and 3 after AC. The inferior to anterior wall ratio and the septal to lateral wall ratio were 0.99 and 1.16 before and 1.02 and 1.00 after AC. The defect contrast of the full and 50% defect was 0.514 and 0.136 before and 0.580 and 0.150 after AC. The ACR phantom images with AC showed correction of the bowling effect in the image without AC due to attenuation. The reconstructed attenuation coefficient of water at 140 keV was 0.151 +/- 0.004/cm and 0.151 +/- 0.002/cm in the liver and cardiac regions of the anthropomorphic phantom, and 0.151 +/- 0.002/cm in the uniform region of the ACR phantom. In conclusion, the XACT system generated accurate attenuation maps with one-minute transmission scans. Phantom studies showed that AC improved image quality and quantification over no AC.
M13-84:

Improved Reconstruction of Planar Scintigraphic Iimages from a Continuous Crystal with a Pad Segmented Photomultiplier Readout

A. Perrotta INFN Bologna, Bologna, Italy On behalf of the ECORAD collaboration In the field of scintillation gamma camera, continuous crystals may improve spatial resolution with respect to scintillation arrays of pixellated crystals. However, the typical algorithms used in the reconstruction of the planar image from the light collected by those continuous crystals loose linearity, and therefore resolution, close to the crystal edges. This has a negative impact if good resolution over the whole field of view is needed, in particular when one has to readout several nearby crystal tiles and the point to be imaged is located in the boundary region between them. In this work we propose a new algorithm that improves the spatial resolution and reduces linearity distortions in the whole field of view of the camera. It calculates the position from a single scintillation event by fitting the charge collected on a multi-anodes photomultiplier tube (MA-PMT) to a charge distribution function that takes into account the light reflections and the truncations at the crystal edge. The algorithm operates by using the CERN root libraries, initially developed for high energy physics applications. It has been applied on a MC generated sample where a compact gamma camera is simulated with a 48x48x4.0 mm3 LaBr3:Ce continuous scintillation crystal, coupled to a MAPMT Hamamatsu H8500. The results show a substantial improvement in spatial linearity, and therefore resolution. Moreover, by taking into account light reflection and truncation, the method can automatically recover the light losses at the crystal boundaries and provide a more equalized total energy measurement throughout the whole field of view.
M13-87:

Feasibility Study of Using Solid State Photomultiplier Array for PET Detector Development

C. J. Bircher, Y. Shao, X. Sun, K. Lan Physics, University of Texas MD Anderson Cancer Center, Houston, TX, USA Solid State photomultipliers have the potential to dramatically improve the performance of Positron Emission Tomography. Their low form factor, high intrinsic gain, fast timing, and good energy resolution will allow SSPMs to advance simultaneous PET/MRI and dual ended readout for depth of interaction measurements. In this study we investigated the imaging capability of the first tillable SSPM array with large area pixels for PET application. The array has 16 pixels in a 4x4 matrix. Each pixel has 3640 total micro-cells, 3x3 mm sensitive area, and about one million amplification gain. Both inter-pixel gap and insensitive edge are around 0.2 mm, which improves the photon detection efficiency and makes it possible to closely tile multiple arrays for designing large area detector. Its basic performance of gain, dark counts and signal linearity were measured. Energy and resolution was measured with a 2x2x10 mm LYSO crystal with a Na-22 source are ~20%. To evaluate its imaging capability, a 12x14 array of 1.4x1.4x10 mm LSYO crystals were optically coupled to the SSPM array with light sharing among different pixels. Either with a individual signal readout of each SSPM pixel or with a position sensitive readout based on a signal multiplexing circuit with a resistor network for current sharing, all crystal were clearly identified. Other detector performance such as Depth-Of-Interaction, detector with multiple SSPM arrays, and test inside MRI for PET-MR dual modality imaging will be reported as well. Our initial studies have indicated that, with appropriate detector design and electronics, substantial imaging performance and innovative imaging design approaches can be achieved for PET clinical and preclinical applications.

399

M13-90:

N. Belcari1,2, F. Attanasi1,2, V. Rosso1,2, A. Santos3,4, F. Spinella2, G. Sportelli3,4, A. Del Guerra1,2 1 Department of Physics, University of Pisa, Pisa, Italy 2 INFN, Pisa, Italy 3 BIT, Universidad Politcnica de Madrid, Madrid, Spain 4 CIBER-BBN Zaragoza, Zaragoza, Spain

A Novel Random Counts Estimation Method for PET Using a Symmetrical Delayed Window Technique and Random Single Event Acquisition

We have developed a new method for estimating the distribution of the random counts in PET based on the delayed window technique. The new method follow a different approach in using the data that can be recorded when a delayed coincidence is found. The advantage of the new method stays in an easier implementation with respect to the standard technique and in the removal of some constraints connected to the acquisition of data in the delayed side. The implementation of the new method has been specifically designed for a dual head modular Positron Emission Mammography (PEM) system we are developing but the proposed technique is also applicable in a more general case, where the random count distribution is obtained by recording a random LOR to be subtracted from the prompt signal on-line, or stored as a separate file for later processing. To test the new method we have developed an acquisition system using a dual head PET. Each head consists in a single module of the large system we are developing. The acquisition system is based on a pair of DAQ boards plugged into a mainboard. Each DAQ board is managed by a FPGA and converts the four signals from the PMTs with four 12 bit ADCs. The results of the conversion are stored in an interfacing FIFO accessible by a fast prototyping mainboard. The mainboard hosts the two DAQ boards and an Opal Kelly XEM3005 board for system management and data transfer to PC. We have used a dedicated coincidence board that can be used in both standard and new delayed window methods. Thanks to the capability of on-line reconfiguration the system allows seamless firmware interchange and then can be switched between the standard and novel delayed window methods for a quick comparison. The new technique will be presented together with a model verification simulation and experimental measurements. A detailed comparison between the standard and the new technique in terms of spatial distribution and noise properties will be reported.
M13-93:

A Comparison of the Performance of High QE Photomultiplier Tubes to Conventional Photomultiplier Tubes

H. Liang, J. Oldendick, Y. Li, C. Ordonez, W. Chang Nuclear Medicine, Rush University Medical Center, Chicago, USA Our objective is to evaluate the performance of high quantum efficiency (HQE) photomultiplier tubes (PMTs) over normal quantum efficiency (NQE) PMTs for use in gamma cameras. Energy resolution, position sensitivity and a cathode scan have been evaluated. Measurements were preformed with 4 HQE Hammatsu R6231-100 PMTs and 4 NQE Hammatsu R1306-19 PMTs, and all PMTs were 2 round tubes. Energy resolution was measured using a 2(d) X 2(l) round NaI(Tl) crystal and was measured for each of 4 NQE and 4 HQE PMTs. Position sensitivity was measured using a NaI(Tl) slab. For the position sensitivity measurement, 2 PMTs were arranged side by side along the length of the crystal slab. Identical set-ups were used for both HQE and NQE PMTs measurements. Position sensitivity was determined using the digital centroid method. For all measurements, a collimated 140 Kev photon beam from Tc99m was directed normally onto the crystal from underneath with the PMTs mounted on top of the crystal exit window. We ran a cathode scan using a collimated LED light source (=420nm) and a reference PMT on both an HQE and an NQE PMT. The light source was scanned across the entire surfaces of the PMTs in a 2D grid using 1mm steps. The results for the energy resolution measurement showed that the average over 4 HQE PMTs had an energy resolution of 8.8% while the average over 4 NQE PMTs had a 9.9% energy resolution at 140 Kev. Our results for the slab experiments show that the HQE PMTs had an average 12.6% increase in position sensitivity relative to the NQE PMTs. The cathode scan showed an average improvement of 23% in the noise. In conclusion, the HQE PMTs performed significantly better than the NQE PMTs in both energy resolution and position sensitivity. A cathode scan of an HQE and an NQE PMT demonstrated a significantly lower noise for the HQE PMTs over the NQE PMTs. Additional cathode scans and further results will be presented at the meeting.
M13-96:

A. Ohtani1, K. Tanaka1, T. Mizuta1, Y. Inoue1, K. Kitamura2, H. Tonami2, J. Ohi2 1 Medical System Division, Shimadzu Corporation, Kyoto, Japan 2 Technology Research Laboratory, Shimadzu Corporation, Kyoto, Japan

Optimization of Enhanced Energy Window on a Whole-Body DOI PET System

The use of Depth of Interaction (DOI) detectors is one effective technique for the improvement of whole-body PET imaging. Then we have developed whole-body PET system with DOI detectors since 2006, and achieved a less than 3mm (FWHM) uniform spatial resolution regardless of the radial position in the filed of view (FOV). Furthermore, it was simulated that DOI PET system can effectively acquire true events using a different energy window for each layer. In this study, we measured sensitivity and noise equivalent count rate (NECR) of various energy window widths, and confirmed improvements of these 400

performances. The DOI detector consists of dual-layer GSO crystals, and the DOI detectors are arranged in a circular detector ring, with a diameter of 664 mm. We evaluated sensitivity and NEC performance (NEMA NU2-2001 standard) of this DOI PET system. As results, it was confirmed that these performances were more improved by the enhanced energy window without the energy region of second layer from 300 keV to 400 keV.
M13-99:

H. Ngo1, K. Dinelle2, S. Blinder2, N. Vafai2, G. Topping1, V. Sossi1,2 1 Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada 2 Pacific Parkinson's Research Centre, Vancouver, BC, Canada Subject motion during brain Position Emission Tomography (PET) scans can reduce image quality, and may lead to incorrect biological outcome measures (e.g. binding potentials), especially during analysis of dynamic data sets. This is particularly relevant when imaging with high resolution scanners such as the HRRT (Siemans Medical Solutions). Frame to frame (FtoF) image realignment is simpler to implement and requires fewer computing resources than methods that corrects for motion during data reconstruction and has been shown to lead to a significant improvement in the accuracy of the dynamically-derived biological variables. An ongoing problem is a lack of objective criteria to validate the accuracy of frame-to-frame realignment. We have developed a Quality Control protocol that automatically flags inadequate realignment. The protocol is based on comparison of motion transformation matrices obtained from two independent sources: motion transformation matrices obtained with image based realignment, AIR in the current implementation, and motion transformations obtain with the Polaris, an external motion tracking device. A distance metric between motion vectors determined with the two methods was defined and a realignment was flagged as unsatisfactory when the distance was greater than a predefined threshold. Since the two methods rely on independent motion assessment tools, the chance of both algorithms giving consistently wrong estimates is very unlikely. Preliminary tests of the quality control protocol performed with phantom and human studies show that the algorithm is capable of correctly identifying residual frame-to-frame mis-alignments, thus providing an objective quality control metric. In addition, this quality control protocol can aid in the evaluation of a realignment protocol, such as identifying conditions when a realignment procedure might be superior or, for example, an optimal choice of target frames in image-based realignment.
M13-102:

Quality Control Protocol for Frame-to-Frame PET Motion Correction

Feasibility Study of Entire Whole-Body PET Scanners Based on the OpenPET Geometry

E. Yoshida, T. Yamaya, F. Nishikido, N. Inadama, H. Murayama Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan A conventional PET scanner has a 15-25 cm axial field-of-view (FOV) and images a whole body using about six bed positions. An OpenPET geometry can extend the axial FOV with a limited number of detectors. The entire whole-body PET scanner must be able to process a huge amount of data, effectively. In this work, we study feasibility of the full 3D entire whole-body PET scanner using the GATE simulation. The OpenPET has 12 block detector rings with the ring diameter of 840 mm and each block detector ring consists of 48 DOI (depth-of-interaction) detectors. Total axial length is 955 mm. Open gap setting of 318 mm provides the continuous axial FOV. The OpenPET has higher loss of single data than a conventional PET scanner at grouping circuits. NECR of the OpenPET decreases by loss of single data. But, loss of single data is improved by separating the axially arranged detector into two parts. Also, multiple coincidences are found to be important for the entire whole-body PET scanner. The entire whole-body PET scanner with the OpenPET geometry promises to provide a large axial FOV with the open space and to have sufficient performance values.
M13-105:

T. Hasegawa1, Y. Sato2, E. Yoshida3, T. Yamada4, K. Oda5, Y. Wada6, T. Yamaya3, H. Murayama3, T. Takeda1, K. Saito1 1 Allied Health Sciences, Kitasato University, Kanagawa, Japan 2 National Institute of Advanced Industrial Science and Technology, Ibaraki, Japan 3 National Institute of Radiological Sciences, Chiba, Japan 4 Japan Radioisotope Association, Tokyo, Japan 5 Tokyo Metropolitan Institute of Gerontology, Tokyo, Japan 6 RIKEN Center for Molecular Imaging Science, Hyogo, Japan

Evaluation and Calibration of PET scanners with a Specially Designed Point-like Radioactive Source

Background: Conventional techniques for evaluating and calibrating PET scanners depend on scatter and attenuation correction because they are based on phantom measurements. Although an extrapolation method with multiple absorber pipes for a line radioactive source, proposed by D. Bailey et al., is useful to remove dependence on scatter and attenuation, the long line source limits its application range. Objective: The aim of this study is to propose a point-like radioactive source with multiple absorber capsules, for which evaluation and calibration can be independent of scatter and attenuation correction. Methods: The source consists of a small spherical radioactive part and a set of successively sized aluminum absorber capsules. Considering 18F or 22 Na, the inner and outer size (diameter and length) of the innermost capsule were set 1 and 3 mm, respectively. In measurement, the source was located at several positions in the field-of-view of a research PET scanner. In reconstructed images, saturated 401

region-of-interest (ROI) values were used to evaluate performance characteristics and determine a calibration constant through extrapolation. Results: Performance was evaluated with the point-like radioactive source and a calibration constant was determined successfully. The validity of the extrapolation technique for the point-like source was supported by Monte Carlo simulation with Geant4. Conclusion: The proposed source is useful for evaluating and calibrating PET scanners. Further studies are expected to refine the design of the source and improve the practicability and traceability of the evaluation/calibration technique.
M13-108:

A Low-Cost Coincidence System with Capability of Multiples Coincidence for High Count-Rate TOF or Non-TOF PET Cameras Using Hybrid Method Combining AND-logic and Time-mark Technology

C. Wang, H. Li, H. Baghaei, Y. Zhang, R. Ramirez, S. Liu, S. An, W.-H. Wong Exper. Dignostic Imaging, University of Texas, MD Anderson Cancer Center, Houston,TX, USA This study is to evaluate a new developed coincidence system for TOF and Non-TOF PET cameras with optional coincidence of multiples. In the front-end electronics, every gamma event is synchronized by a 125MHz system clock and generating a trigger associated with a time-mark given by an 8-bit high-resolution TDC (68.3ps/bin). The clock synchronized trigger and time-mark are sent to this coincidence system for a 2 level coincidence. The first level is like a snapshot, using an AND-logic to pre-sort the coincidence pairs to reduce the data rate for the second level fine sort. These pre-sorted coincidence pairs are within 2 clock cycles, hence, an 8-bit short time-mark will be enough for the second level fine sort. The second level coincidence uses 42 sorting units in parallel to reduce the dead time and accept multiples. This FPGA-based coincidence system for modular PET camera offers reprogrammable flexibility and expandability. The coincidence time window size and time offset of each module pair can be programmed independently and multiples coincidence can be turned on or off. This coincidence system is low-cost reconfigurable for the different module based PET camera. We used GATE to generate gamma events by model the detectors and phantoms for coincidence system evaluation. The performance of count rate and NECR of this system are compared with ideal system (no dead time, time window sorting directly) with and without multiples.
M13-111:

Wavelet Optimization and Noise Reduction for LEGP Collimator in SPECT Simulated Images

F. Babapour Mofrad, A. Abbaspour Tehrani-Fard, B. Ebrahimi, D. Sardari, M. Pouladian, A. Ebrahimi Science and Research Branch, Islamic Azad Univ., Tehran, Iran The Poisson noise in SPECT sinogram is a major cause to compromise the quality of images. A new method of noise reduction is Wavelet, that it is a new timefrequency analysis method. In this paper, the SPECT simulated images de-noised by wavelet. In order to, the SIMIND program is used to simulated SPECT images using 99mTc radionuclide in the Triangular parallel holes collimator of type LEGP, then the images simulated are de-noised by 118 types of wavelet tests and 3 types of test low pass filters, after they are compared together. Finally, the best type of wavelet is compared to the best type of low pass filter, that results demonstrated a significant improvement with the wavelet, as compared to the low pass filter. So, wavelet increased the quality of the nuclear medicine images without decreasing contrast.
M13-114:

R. S. Miyaoka1, X. Li2, C. Lockhart1, T. K. Lewellen1 1 Radiology/Nuclear Medicine, University of Washington, Seattle, WA, USA 2 Physics, University of Washington, Seattle, WA, USA

New Continuous Miniature Crystal Element (cMiCE) Detector Geometries

Continuous miniature crystal element (cMiCE) detectors are a potentially lower cost alternative to high resolution discrete crystal designs. We report on the intrinsic spatial resolution performance for two cMiCE PET detector designs with depth of interaction (DOI) positioning capability. The first detector utilizes a 50 mm by 50 mm by 8 mm LYSO crystal coupled to a 64 channel, multi-anode PMT. It provides 4 layers of DOI information. The crystal has beveled edges along two of it sides to improve the detector packing fraction when placed in a ring geometry. For this truncated crystal design, the front surface of the detector is 45 mm by 50 mm and the useful imaging area of the detector is 45.6 mm by 46.6 mm. The second detector utilizes a 50 mm by 50 mm by 15 mm, rectangular geometry, LYSO crystal coupled to a 64 channel, multi-anode PMT. It provides up to 15 layers of DOI information. The expected useful imaging area of the crystal is 44.6 mm by 44.6 mm. In addition to evaluating the X, Y positioning resolution, we also did testing to evaluate the DOI positioning capabilities of the 15 mm thick detector. The average intrinsic X, Y spatial resolution for the 8 mm thick, truncated crystal detector was 1.33 +/- 0.31 mm FWHM. The average intrinsic X, Y spatial resolution for the 15 mm thick crystal detector was 1.74 +/- 0.35 mm FWHM. In addition, the average DOI spatial resolution for 56 test points spanning a 26.4 mm by 12.2 mm region of the crystal was 4.80 +/- 0.36 mm. We believe the 8 mm thick truncated crystal design is suitable for mouse imaging while the 15 mm thick crystal design is more suited for human organ specific imaging systems (e.g., breast and brain).

402

M13-117: Preliminary Experimental Results of a Quasi-Monolithic Detector with DOI Capability for a Small Animal PET

S.-J. Lee1,2, C.-H. Baek1,2, J. Y. Hwang1,2, Y. Choi3, Y. H. Chung1,2 1 Department of Radiological Science, College of Health Science, Yonsei University, Wonju, Gangwon, Korea 2 Institute of Health Science, Yonsei University, Wonju, Gangwon, Korea 3 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea

In our previous work, the new detector module with depth of interaction (DOI) based on a quasi-monolithic crystal array and maximum-likelihood position-estimation (MLPE) algorithm was designed and the methodology for 3D event positioning in the detector was established by using Monte Carlo simulation. The aim of this study was to evaluate the performance of our detector module experimentally. The detector module consisted of a 1D array of eight LYSO crystals of 2.0 (axial) 20.0 (trans-axial) 10.0 (depth) mm3 optically coupled to a Hamamatsu H8500C flat panel position sensitive photomultiplier tube (PSPMT). The PSPMT was read out using a resistive charge divider, which multiplexes 64 anodes into 8(X) + 8(Y) channels. Gaussian-based MLPE methods have been implemented using experimentally measured detector response functions (DRFs). The results demonstrated that the detector module could identify the position of the gamma ray interaction inside the crystal with a resolution of 2.0 mm in all three directions. The new DOI detector for a small animal PET was developed and verified experimentally with a view towards achieving high sensitivity as well as high and uniform radial resolution.
M13-120:

D. Burdette1, D. Albani1, E. Chesi1, N. Clinthorne2, E. Cochran1, K. Honscheid1, S. S. Huh2, H. Kagan1, M. Knopp3, C. Lacasta4, M. Mikuz5, P. Schmalbrock3, A. Studen5, P. Weilhammer1 1 Dept. of Physics, The Ohio State University, Columbus, OH, USA 2 Div. of Nuclear Medicine, University of Michigan, Ann Arbor, MI, USA 3 Dept. of Radiology, The Ohio State University, Columbus, OH, USA 4 Dept. de Fisica Experimental, Inst. de Fisica Corpuscular (IFIC), Valencia, Spain 5 Faculty of Mathematics and Physics, University of Llubljana, Llubjana, Slovenia The image resolution for small animal PET systems using high energy positron emitters, such as 68Ga and 94mTc, is dominated by the positron range. The blurring caused by the positron range can be reduced by either embedding the field of view of the PET ring within a strong magnetic field or using an iterative reconstruction algorithm that statistically corrects for the range of the positron. This study investigates improvement in PET image quality using both techniques simultaneously.
M13-123:

A Study on PET Image Quality Using Both Strong Magnetic Fields and a ML-EM Positron Range Correction Algorithm

Per Ring Dead Time Correction for Preclinical PET

M. Chen, M. E. Casey Siemens Molecular Imaging, Knoxville, TN, USA This study investigated dead time correction for each detector ring of a preclinical PET scanner. It is noticed that the peak-valley structure in the axial sinogram profile is dependent on count rate. Thus the standard normalization acquired under one count rate will not completely correct the peak-valley structure in a scan acquired under another count rate. Dead time correction parameters were derived for each individual detector ring to compensate for the count rate dependent detector sensitivity and to adjust the standard normalization by this function.
M13-126:

N. G. Johnson-Williams1, R. S. Miyaoka2, X. Li2, T. K. Lewellen2, S. A. Hauck1 1 Department of Electrical Engineering, University of Washington, Seattle, WA, USA 2 Department of Radiology, University of Washington, Seattle, WA, USA

Design of a Real Time FPGA-Based Three Dimensional Positioning Algorithm

Abstract- We report on the implementation and hardware platform of a real time Statistic-Based Processing (SBP) method with depth of interaction processing for continuous miniature crystal element (cMiCE) detectors using a sensor on the entrance design. The cMiCE detector module was modeled as being composed of a 49.6 mm by 49.6 mm by 15 mm LYSO crystal coupled to a 16 by 16 array of micro-pixel avalanche photodiode sensors. The algorithm allows an intrinsic spatial resolution of ~0.65 mm FWHM in X and Y and a resolution of ~0.95 mm FWHM in Z (i.e., the depth of the crystal) based upon DETECT2000 simulation results that include the effects of Compton scatter in the crystal. We have previously reported on a Field Programmable Gate Array (FPGA) SBP implementation that provided a two dimensional (2D) solution of the detectors intrinsic spatial resolution. This new implementation extends our work to take advantage of parallel data paths and three dimensional (3D) look up tables to provide a 3D positioning solution that improves intrinsic spatial resolution. Resolution is most improved along the edges of the crystal, an area where the 2D algorithms performance suffers. Further changes to the algorithm have improved 403

throughput and latency by more than 10%. A pipelined FPGA implementation is able to process events in excess of 350k events per second, which is greater than the maximum expected coincidence rate for an individual detector. In contrast to all detectors being processed at a centralized host, as in the current system, a separate FPGA is available at each detector, thus dividing the computational load. These methods allow SBP results to be calculated in real-time. A prototype design has been implemented and tested using a reduced word size due to memory limitations of our commercial prototyping board. A custom board is currently under development that will allow implementation of the full seven stage algorithm with full arithmetic accuracy.
M13-129:

Theoretical Design of the High-Resolution Zoom-in Detector for Targeted PET Imaging

J. Zhou, S. St. James, S. Cherry, J. Qi Biomedical Engineering, University of California, Davis, CA, USA We proposed to integrate a high-resolution detector into the existing microPET II scanner for high-resolution PET imaging of a target region. The high-resolution zoom-in detector will be able to improve the spatial resolution and lesion detectability over the original microPET II system. In this paper we evaluate the performance of the proposed system with high-resolution detectors of different crystal length and DOI resolution. The purpose is to find the optimal design of the high-resolution detector. We focus on the system performance for lesion detection based on computer observers. In general, with a fixed crystal length (or DOI resolution), the system performance improves as the DOI resolution (or crystal length) increases. However, when the DOI resolution is linearly proportional to the crystal length, our theoretical results show that there is an optimal crystal length that maximizes the lesion detectability.
M13-132:

J. Deng1, S. Yan1, H. Yu2, G. Wang2, M. Chen1 1 Molecular Imaging/ Preclinical Solutions, Siemens Medical Solutions, Knoxville, TN, USA 2 2Biomedical Imaging Division, VT-WFU School of Biomedical Engineering and Science, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA In preclinical micro-CT, a circular scanning trajectory is widely used, and the Feldkamp algorithm is usually the standard reconstruction method to recover the images. This imaging mode has the advantages of fast reconstruction speed and good intraslice resolution for images near the central plane. However, it shows cone-beam artifacts for images at relatively large cone angles. To address this problem, a spiral scanning trajectory is employed for the micro-CT. A spiral reconstruction algorithm was utilized to reconstruct the images and compared with the standard Feldkamp algorithm. The results show that the spiral approach delivers images with reduced cone beam artifacts.
M13-135:

A Study on Spiral Cone Beam Scanning Mode for Preclinical Micro-CT

A. Vandenbroucke1, J. Lee2, V. C. Spanoudaki1, F. W. Y. Lau1, P. D. Reynolds1, C. S. Levin1 1 Radiology, Stanford University, Stanford, CA, USA 2 Mechanical Engineering, Stanford University, Stanford, CA, USA

Temperature and Bias Voltage Studies of a Large Area Position Sensitive Avalanche Photodiode

We are constructing a high resolution, high sensitivity PET detector system with depth of interaction capability. The detectors are built from modules comprising LSO crystal arrays coupled to Position Sensitive Avalanche Photodiodes (PSAPDs). The entire system will have 4,608 densely packed dual LSO-PSAPD modules. The performance of the large area (1x1 cm2) PSAPDs in our system depends on bias voltage and temperature. Coincidence data was obtained by placing a Na-22 source between an LSO crystal coupled to a PMT and an LSO-PSAPD module. The bias voltage dependence shows that the energy resolution remains constant around 14.1+/-0.5% FWHM at 511 keV and deteriorates by about 5% at lower bias voltages. Crystal identification capability stays constant over the range of interest. Optimal coincidence time resolution of 2.63+/-0.02ns FWHM was observed around 1740 V. Timing resolution decreases from there by about 10 % for a 20 V change. The gain increases by a factor 2 for every 35 Volt and can be described by two exponentials. The point where those two exponentials intersect corresponds to the beginning of the avalanche breakdown. The module's temperature was varied using a thermoelectric cooler coupled to a heatsink. In general, decreasing the temperature of a PSAPD improves performance. The coincidence time resolution improved from 5.88+/-0.05 ns FWHM at 39 degrees to 1.96+/-0.03 ns at 5 C. 511 keV energy resolution improved from 14.43+/-0.01% at 39 C to 11.82+/-0.01% at 5 C. PSAPD gain increases by 20% every 5 degrees. The rate of gain increase is even larger (10% per degree) at the lowest temperatures. Most of the observed behaviour is attributed to the PSAPD, since the light output of LSO varies only slightly with increasing temperature. In summary, we present the performance variation of a large area PSAPD as a function of temperature and bias voltage. These parameters are of extreme importance in densely packed systems needed for state-of-the art PET.
This work was supported in part by National Institute of Health grants R01 CA119056 from NIBIB and R33 EB003283 from NCI, and the Belgian American Educational Foundation

404

M13-138:

M. I. Peterson1, B. W. Miller2, K. Ljunggren1, S.-E. Strand1 1 Department of Medical Radiation Physics, Lund University, Lund, Sweden 2 Center for Gamma Ray Imaging, University of Arizona, Tucson, USA

Construction of a Pre-Clinical High Resolution Tomographic Scintillation Camera System

Objectives: We are constructing a High Resolution Tomographic Scintillation Camera System for small animal imaging. Material and methods: The system consists of a CsI(Tl) scintillator coupled to a military surplus image intensifier with a CCD camera focused on the output window. The system operates in photons counting mode and the spatial resolution is improved by centroid calculations for every photon interaction. The system can be mounted on an optical rail together with either a pinhole or a parallel-hole collimator and a rotating table. Results: The system has a registered intrinsic resolution of 160 m. With a 1 mm pinhole the FWHM is 1.6 mm and 2.2 mm for planar and SPECT respectively. With a 0.380 mm pitch parallel-hole collimator the FWHM was 0.7 mm. System sensitivity for pinhole and parallel-hole is 1.2 and 0.9 cps/MBq respectively. Discussion and future work: The system is capable of acquiring images with sub-millimeter resolution with parallel-hole collimator. Since our system is mounted on an optical rail with mobile parts the resolution for pinhole can be increased by optimizing the pinhole diameter and/or the object/pinhole pinhole/detector distance for each application. Future work includes the optimization of the pinhole geometry, implementation of coded aperture collimator for detailed ex-vivo imaging of organs and tissue samples.
The authors would like to thank Professor Harrison H. Barrett and our collaborators at the Center for Gamma Ray Imaging. M13-141:

Searching for a Precise System Response Function for a Single Photon Emission Microscope System

G. Fu, L.-J. Meng NPRE, University of Illinois at urbana-champaign, Urbana,IL, USA Having a precise system response function (SRF) is one of the keys for aching an ultrahigh SPECT imaging resolution. We have recently developed a single photon emission microscope (SPEM) system for ultrahigh resolution imaging of mouse brain. Despite its excellent imaging performance for use with I-125 labeled tracers, the spatial resolution achieved for Tc-99m studies has stayed well below expectation, especially with relatively thick (2 mm) scintilaltor. This is primarily due to the severe discrepancy between the system model and the actual system response. Even with detailed modeling of factors such as depth of interaction, photon penetration through pinhole knife edge, the resultant point source projection remains different from the measured one in both shape and magnitude, which become the major limiting factor. In this study, we carried out detailed experiments to determine the underlying causes for the discrepancy. Adding to the current system modeling, we will further include factors such as (a) the potential variation in columnar CsI(Tl) crystal density along the depth direction, (b) the escape of fluorescence x-rays, (c) the variation in the distribution of photoelectrons as a function of interaction depth and (d) the small variation in the actual pinhole profile. The resultant SRF will be validated by comparing to the ones experimentally measured with CsI(Tl) crystals of 0.25 mm, 0.5 mm, 1 mm and 2 mm thickness and a single crystal YAP scintillator of 1mm thickness. This study will provide crucial insight for improving the system modeling and therefore deliver an improved spatial resolution with the SPEM system. The resultant experimental images of various phantoms will be presented.
M13-144:

J. Donnard1, N. Arlicot2, R. Berny1, H. Carduner1, S. Chalon2, A. Faivre-Chauvet3, P. Leray1, E. Morteau1, N. Servagent1, D. Thers1 1 Subatech, Nantes, France 2 U930, Inserm, Tours, France 3 U892, Inserm, Nantes, France The beta autoradiography is commonly used in preclinical research. This technique is able to locate, in 2 dimensions with a high spatial resolution, molecule labeled with beta emitters. We report on the development of an integrated and transportable device base on the PIM (Parallel Ionization Multiplier) structure able to perform the image of 10 microscope slides in the same time. Thin slices of organs can be labeled with beta emitters of low energy like 3H / 14C or Auger electrons emitters such as 125I / 99mTc. The measured spatial resolution over the whole area of 18x18 cm is 30 microns Full Width at Half Maximum in 2D with a 3H labeling. The sensitivity threshold is estimated to be close to 0.005 cps/min/mm. The use in proportional counter mode is suitable to perform on-line imaging and monitoring. New modalities such as large area imaging of entire rat slices or imaging of beta emitters of high energy like 131I or 18F are now under development. In case of labeling with beta emitters of high energy, the spatial resolution is estimated to be around 200 microns. First results obtained with these new configurations will be presented.

The PIMager: a New Tool for High Sensitive Numerical Beta Autoradiography

405

M13-147:

K. H. Tainter1, S. Lokitz1, C. Vascoe1, J. Zhang2, B. J. Coco2, K. Iwata2, D. J. Wagenaar3, J. M. Mathis4 1 The Biomedical Research Foundation of Northwest Louisiana, Shreveport, LA, USA 2 Preclinical Imaging, GE Healthcare, Waukesha, WI, USA 3 GammaMedica-Ideas, Northridge, CA, USA 4 Cell Biology, Anatomy and Gene Therapy Program, LSU Health Sciences Center, Shreveport, Shreveport, LA, USA The recent availability of pre-clinical PET/SPECT/CT tri-modality systems provides an opportunity to image animals injected with mixed modality (PET/SPECT) isotopes. However, before these procedures can be truly useful, some understanding of the effect of different modality radiotracers on image quality and the resulting noise characteristics must be investigated. Our goal in this study was to characterize effects on image resolution that might arise in one modality because of the presence of another (different modality) isotope. A phantom, consisting of a set of side-by-side micro-capillary tubes, were imaged on a trimodality (SPECT/PET/CT) pre-clinical scanner. Each tube was filled with a SPECT isotope and imaged while varying acquisition time, amount of activity, and distance between line sources. Images were evaluated to determine the visualization threshold at which objects became apparent in cross-sectional and MIP images (both that the objects were line sources and that the lines were two separate objects). These experiments were repeated in the presence of a PET isotope. After re-evaluating resolution, the signal to noise characteristics of the image as a whole were assessed and related to amounts of each isotope as well as total activity in the field of view. PET acquisition in the presence of SPECT isotopes was investigated using a contrast phantom with spheres filled with PET, SPECT, and mixed isotopes in backgrounds of mixed SPECT and PET isotopes. The noise properties of the background and spheres were measured as well as the visibility of the spheres. Finally, SPECT, PET, and CT images were acquired of a mouse after injection of both a PET and SPECT isotope. Acquisition of SPECT and PET images in the presence of different modality tracers (with careful considerations of the activity and acquisition parameters) seems feasible.
M13-150:

Mixed Isotope Effects : Image Quality in Multimodality PET/SPECT Preclinical Imaging

Y. Qi1, C. Zhao1, Q. Dai1, M. Liu1, H. Zhang2 1 Shanghai Institute of Applied Physics, Shanghai, China 2 Department of Radiation Medicine, The Ohio State University, Columbus, USA

Performance Comparison of Two Compact Charge Multiplexed Readout Schemes for Multi-Anode Position Sensitive Photomultiplier Tubes

The aim of this study was to develop a high performance compact readout electronic for a high-resolution modular gamma camera. The modular camera is based on a pixellated NaI(Tl) crystal with 1.2mm x1.2mmx5 mm pixel size and 1.4 mm pixel pitch coupled to a 5cm square Hamamatsu H8500 flat-panel position-sensitive photomultiplier tube (PSPMT) with 8x8 matrix anodes. Two compact multiplexed readout circuits have been developed and evaluated. One is discretized positioning circuit (DPC) which sums all the charge from 64 anodes into a 2D resistor-chain network and divides the charge proportional to 4 outputs. The other is symmetric charge division circuit (SCDC) which divides the charge from the 64 anodes into 8-row and 8column resistive networks. Then a subtractive resistive readout scheme with the truncated center-of-gravity (TCOG) positioning method was utilized for the SCDC readout signal processing and further reduces the readout channels from 16 to 4 outputs. The performance of the modular camera with two compact multiplexed readout schemes was evaluated by a flood image of Cs-137 source. The preliminary results show that both readouts can provide fair good spatial resolution. However, the SCDC scheme with subtractive resistive readout could significantly reduce the non-linearity and the compress effect of positioning to maximize the crystal identification. The useful field-of-view of the imaging detector was also improved by about 20% as compared to the conventional DPC readout. In conclusion, the SCDC scheme with subtractive resistive readout is a better approach for high performance compact readout electronics for the development of high-resolution modular gamma cameras.
M13-153:

Comparison Between Parallel Hole and Rotating Slat Collimation with a Contrast Phantom Using an Analytical Method

L. Zhou, K. Vunckx, J. Nuyts Dept. of Nuclear Medicine, K.U.Leuven, Leuven, Belgium In single photon emission computed tomography (SPECT) imaging, image quality is always limited by the trade-off between the spatial resolution and the system sensitivity. In contrast to the conventional parallel hole (PH) geometry, rotating slat (RS) collimation can overcome this limitation by a much higher geometric efficiency due to its capability of in-plane photon collection. However RS needs an extra reconstruction step, which makes it suffering from increased noise propagation. In our previous work, the PH and the RS collimator systems were compared by investigating the contrast-to-noise ratio (CNR) of the central point of a homogeneous phantom using two analytical methods. In this study, one of the analytical methods was extended to evaluate CNR in a region of interest (ROI) as well as the system performance in a lesion detectability task. A contrast phantom was used to compare these two collimation systems. The comparison was done by calculating the gains (RS over PH) in three figures of merit (FOMs): CNR in a pixel of interest, CNR in a region of interest, and the signal-to-noise ratio (SNR) of a channelized Hotelling observer(CHO). All FOMs were calculated at a fixed spatial resolution. 406

The comparison was mainly performed for planar imaging. The main results are: (1) By doing iterative reconstructions, it was verified that the analytical method can give accurate predictions for CHO-SNR. (2) The gains of all FOMs show the same tendencies in the system comparison. (3) PH only outperforms RS for cold spot imaging with a large phantom, while RS is better for all the other cases. (4) For system optimization, the optimal collimator apertures depend on the FOM to be optimized, however the ratio of the apertures (PH/RS) is always around square root of 2, a value that we derived analytically in our previous work.
M13-156:

Technique to Distinguish Signal from Statistical Noise in PET Imaging

J. Hamill, M. Conti Molecular Imaging, Siemens Healthcare, Knoxville, TN, USA In PET imaging, the physician interpreting the images sometimes cannot distinguish signal, i.e. image features indicating disease, from noise, i.e. image artifacts. This situation can lead to misdiagnosis. In iterative reconstructions, focal hot spots in the image could indicate small lesions but could also be due to statistical fluctuations in the PET measurement. The problem is common in PET imaging of the liver. A repeat scan is the safest way to be determine if the hot spot is real, since a lesion would be seen again in the repeat scan whereas a noise artifact would not be expected in the same location, but this option is not normally practical. Alternatively, one can transform the measurement into two or more new data sets that are statistically independent of each other, reconstruct those data sets, and compare the images. A hot spot at the same location in all images would indicate the presence of a lesion, whereas the absence of a hot spot in one of the new images would indicate a statistical noise artifact. We have studied two approaches. (1) The list mode PET measurement was statistically divided into two smaller lists of half the size. Each new list was then histogrammed into a sinogram. (2) The original sinogram was divided into two sinograms, one with even-valued angle indices and one with odd angle indices. An iterative reconstruction was performed on the original data set and on each of the new ones, allowing a three-way comparison of the original image and two independent ones, using human observers and machine vision. We tested the two methods in phantoms with spherical objects < 1 cm diameter, and in human PET/CT studies of the liver. ROC analysis was used to measure machine vision's ability to distinguish actual focal lesions from noise artifacts. In many clinically realistic cases, this statistically based approach is able to distinguish signal from noise without a repeat scan.
M13-159:

G. Wagenknecht1, A. Poll1, M. Losacker1, I. Blockx2, A. van der Linden2 1 Central Institute for Electronics, Research Center Juelich, Juelich, Germany 2 Bio Imaging Lab, University of Antwerp, Antwerp, Belgium

A New Combined Live Wire and Active Surface Approach for Volume-of-Interest Segmentation

Volume-of-interest (VOI) segmentation is an important prerequisite for quantitative analysis of structure and function in multimodal tomographic images of humans and small animals. A suitable VOI segmentation method should allow user interaction to position the VOI and be automated to support the user as much as possible. The new method combines a 2D live wire (LW) with a 3D active surface model (ASM) approach integrated into MITK to provide the GUI for user interaction and visualization. 2D LW allows the user to initialize the VOI close to the desired position by segmenting regions of interest (ROIs) in a few sample slices. ROIs are segmented by setting only a few points on its boundary contour. Contour pieces in between are determined automatically with LW using gradient-based image features. The ASM start mesh is generated from these sample ROIs by first building a binary start volume which is then transformed into a triangular mesh. The 3D ASM yields the final VOI segmentation by deforming this mesh until internal and external forces are in balance. The external generalized gradient vector flow (GGVF) field attracts the mesh to the edges and the Taubin filter approach steers the internal smoothness of the mesh, avoiding volume loss. The ASM approach was optimized to be fast in calculating the GGVF field and to get improved voxelization results. Quantitative results were obtained as misclassification rates based on spherical phantoms with a cylindrical element in the middle. ROIs were segmented in sample slices with LW or polygons to build the start mesh. User interaction could be reduced significantly with LW-based compared to polygon-based initializations while yielding similar or better final results. Optimization steps have improved the computation time and segmentation quality. A set of MRI images from different human brain and small animal studies were segmented to show the excellent performance of the method in segmenting different kinds of VOIs.
This study is part of the EU-funded EMIL program (LSHC-2004-503569). Thanks are due to Dr. K. Buehler, VRVis, Wien, Austria, for providing the MRI data set of Fig. 5.

407

M13-162:

A. Pourmoghaddas1,2, R. G. Wells2 1 Physics Department, MSc candidate - Carleton University, Ottawa, Canada 2 Cardiac Imageing, The University of Ottawa Heart Institutue, Ottawa, Canada

Improvement in Defect Size and Contrast Using Respiratory Motion Correction in Cardiac PET/CT

During long PET acquisition times, respiratory motion has been shown to degrade cardiac image quality. In this study, respiratory-gated dynamic images were corrected for motion blurring to evaluate the impact of respiratory-motion correction on abnormal studies. The NCAT phantom was used in conjunction with SimSET to generate a 17-frame Rb-82 respiratory-gated dynamic acquisition. Each frame consisted of 10 respiratory phases, where a 5s respiratory cycle was assumed and the translation of the heart was set to be 2 cm in the Sup-Inf direction. Two lesions were separately simulated wherein phantom activity values were set to zero for 50mm and 30 mm of the lateral wall. All images were reconstructed with phase-matched attenuation correction. Algorithms developed in-house to detect and fit the heart for kinetic analysis, were modified to fit an ellipsoid to the myocardial wall activity. After fitting each phase separately, image volumes of each phase were translated to align the centers of the fitted ellipsoids and thus compensate for motion. The dynamic data from a single respiratory phase was used to simulate a motion-free image set and a 50% of maximum threshold in this image was used to define the defect location for all images. Image uniformity was measured as the pixel standard deviation within the polar map representation of the normal myocardium. Average line profiles drawn through the defect areas were used to determine defect contrast. The mean percentage error was calculated for only the defect area. The corrected images show improvement in defect size and contrast. Image uniformity is improved by up to 20% in the Anterior-Lateral, Inferior-Septal and apical regions where most motion-averaging effects occur. A %6 absolute reduction in the mean percentage error is seen for the large defect, and a 3% reduction for the smaller defect, after applying motion compensation. Defect contrast shows an improvement of %6 with respect to the motion-free images.
M13-165:

The Utah PET Lesion Detection Database

D. J. Kadrmas Department of Radiology, University of Utah, Salt Lake City, UT, USA Task-based assessment of image quality is a challenging but necessary step in evaluating advancements in PET instrumentation and algorithms. We have been developing methods of evaluating observer performance for detecting and localizing focal warm lesions using experimentally-acquired whole-body phantom data designed to mimic oncologic FDG PET applications. This work describes a new resource of experimental phantom data that is being developed to facilitate lesion detection studies for the evaluation of PET reconstruction algorithms and related developments. A new large custom-designed thorax phantom has been constructed to complement our existing medium thorax phantom, providing two whole-body setups for lesion detection experiments. The new phantom is 70% larger and has a removable spine/rib-cage attenuating structure that is held in place with low water resistance open cell foam. Several series of experiments have been acquired, with more ongoing, including both 2D and fully-3D acquisitions on tomographs from multiple vendors, various phantom configurations and lesion distributions. All raw data, normalizations, and calibrations are collected and offloaded to the database, enabling subsequent retrospective offline reconstruction with research software for various applications. The offloaded data are further processed to identify the true lesion locations in preparation for use with both human observers and numerical studies using the channelized non-prewhitened observer. These data have been used to study the impact of improved statistical algorithms, point spread function modeling, and time-of-flight measurements upon focal lesion detection performance, and studies on the effects of accelerated block-iterative algorithms and advanced regularization techniques are currently ongoing. Interested researchers are encouraged to contact the authors regarding potential collaboration and application of database experiments to their projects.
R01CA107353 M13-168:

J. Cabello1, A. Metaxas2, A. Bailey2, I. Kitchen2, K. Wells1 1 Faculty of Electronics and Physical Sciences, Centre for Vision, Speech & Signal Processing, University of Surrey, Guildford, Surrey, United Kingdom 2 Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, United Kingdom Autoradiography (AR) is a method used to map the distribution of radiolabelled biomolecules deposited in a thin tissue specimen. After the sample has been exposed to autoradiographic film, subsequent analysis is performed on the digitised autoradiogram. This is currently undertaken by manually segmenting those regions of functional importance, usually showing significant high levels of radioligand density. In terms of throughput, this is considered as one of the main bottle-necks in AR that remains unsolved. A possible approach to semi-automatically warp a digitised mouse brain atlas on to the autoradiogram is presented here. The algorithm is comprised of two steps: first those regions with significant high and low levels of ligand density are automatically segmented, by using a region grower-based approach, and secondly the atlas is elastically warped on to the

Elastic Atlas Registration of Beta- Autoradiograms Using Scattered Data Interpolators

408

autoradiogram using the previously segmented regions as landmarks. Whilst principally applied here to AR data, the method is also applicable to to the registration of PET and SPECT functional image data.
M13-171:

H. Yu1, J. Hoppin2, K. Harlin2, J. McDonald3, P. Kuehl3, T. Anderson1, C. Lackas2, B. Gershman4, G. Candelaria1, J. Y. Hesterman5, J. P. Norenberg1 1 College of Pharmacy, University of New Mexico, Albuquerque, NM, USA 2 inviCRO, LLC, Boston, MA, USA 3 Lovelace Respiratory Research Institute, Albuquerque, NM, USA 4 Northeastern University, Boston, MA, USA 5 Bioscan, Inc, Washington, DC, USA

Implementation of a 3D Topographic Thinning Model for Assessing Aerosol Deposition of Radioactive Assays in Small-Animal CT/SPECT Imaging

Nuclear imaging techniques are often employed to assess the deposition of radiolabeled particles in the lung delivered via nebulizer to both humans and animal models. These imaging studies enable an assessment of the nebulizer properties as well as the lung function of the subject. This work presents a detailed description of three different topographic thinning or onion model approaches used to quantify deposition in SPECT scans of rodent lungs. The techniques developed here were created in conjunction with a large-scale (32 animal) inhalation study looking at the relationship between particle size and deposition. The reliable and accurate segmentation of lungs in 3D CT images and the precise measurement of their volume is the basis for the onion analysis of the labeled particles in SPECT of lungs of the small-animals. In addition to the novel lung modeling approaches, this work also presents the development and evaluation of a new robust and fully-automated lung segmentation method for large high-resolution 3D CT small-animal image data sets. The fully-automated segmentation method reported here integrates multi-threshold segmentation, 3D region splitting and growing, morphological operation, and knowledge-based classification. To adapt to the variability of the large data sets, robust customized features are used in classification. Results of the rodent uptake study are presented including images, deposition plots and an evaluation of automated 3D segmentation compared to manual segmentation references. The onion-model analysis for SPECT data and the new, robust 3D CT fully-automated segmentation method presented here provides a reliable tool to quantify and visualize the deposition of radiolabeled particles in three dimensions.
M13-174:

J. M. Wilson1, T. G. Turkington1,2 1 Graduate Program in Medical Physics, Duke University, Durham, NC, USA 2 Department of Radiology, Duke University, Durham, NC, USA

TOF-PET Small-Lesion Image Quality Measured over a Range of Phantom Sizes

The hypothesis of this study is that TOF image quality improvement over non-TOF PET can be parameterized using small lesions in a phantom that represents a range of body sizes. A set of 46 1.0-cm spherical inserts was divided into 6 groups and positioned in a fillable, tapered phantom that represents a range of body dimensions (an oval cross section from 38.5 x 49.5 cm to 6.8 x 17.8 cm, a length of 51.1 cm). The hot spheres were positioned in a uniform warm background (6:1 radioactivity concentration) and were distributed in 6 regions of different cross-sectional areas: 6 spheres at 306 cm2, and 8 spheres at 426, 574, 754, 964, and 1187 cm2. The PET/CT study was acquired on a Discovery 690 PET/CT (GE Healthcare), which uses LYSO crystals (4.25 x 6.3 x 25 mm3) in three rings of 6 9 blocks. Three bed positions, spanning 36 cm, were each acquired for 5 min. Images were reconstructed using OSEM with 16 subsets and 110 iterations for TOF data and with 24 subsets and 110 iterations for non-TOF data. The 50-cm field of view (FOV) was reconstructed to a 128x128 matrix with attenuation, normalization, decay, scatter, and singles-based random corrections, and z-axis smoothing along with a post-filter were applied. An automated algorithm was used to locate and place regions of interest on the hot spheres in the CT, and then apply the ROIs to the PET images to measure signal and background variability (image noise). When comparing the small lesion signal-to-noise ratios between TOF and non-TOF images, a trend emerges. This phantom, representing a continuum of body sizes, provides a way to parameterize the TOF improvement.
M13-177:

X.-B. Pan , T. H. Schindler2, O. Ratib2, S. Nekolla3, J. Declerck1 1 Siemens Molecular Imaging, Oxford, UK 2 Hpitaux Universitaires de Genve, Geneva, Switzerland 3 Klinikum rechts der Isar der Technischen Universitaet Muenchen, Munich, Germany

Effect of Reorientation on Myocardial Blood Flow Estimation from Dynamic 13NH3 PET Imaging

Objectives: Estimation of the myocardial blood flow (MBF) from dynamic PET cardiac acquisition using 13NH3 or 82Rb involves a series of processing steps. Some of them can not be completely automatic, and manual adjustment would introduce inter and intra operator variability. Reorientation is one of the main factors for introducing this variability. The aim of this study is to assess the effect of reorientation variation on the MBF estimation. 409

Methods: 58 13NH3 dynamic datasets (rest and stress) from 29 patients were reoriented with an automatic algorithm. For each dataset a series of 20 random reorientations were generated from the automatically computed result by perturbing the two left ventricle (LV) rotation angles within +/-5 degrees and center within +/-5mm translation using a uniform random distribution. MBF represented by quantitative myocardial perfusion (QMP) value were computed using syngoCirculation package, where dynamic 13NH3 studies were modeled by a two tissue compartment model. The variation of QMP with the reorientation parameters was assessed using the coefficient of variation (CV). Results: The mean, standard deviation and max of CV for the 58 datasets with reorientation and center perturbation are computed for the three AHA vessel regions: LAD (4.8%, 3.3%, 21.1%), LCX (4.5%, 3.3%, 20.3%), RCA (5.6%, 5.4%, 30.6%), all regions combined (4.5%, 3.3%, 17.1%). The large variation presented in some cases is due to the BIF being incorrect for some reorientation angles. Conclusion: The robustness of QMP estimated by the syngoCirculation application was assessed with regards to the LA reorientation parameters (angle and center). This reorientation study simulates some aspect of inter and intra observer variability. Manual adjustment of the reorientation is not expected to cause large variation on the QMP estimation on the three vessel regions or globally, if the placement of the BIF ROI is appropriate.
Acknowlegement The datasets used in this study are provided courtesy of the Hpitaux Universitaires de Genve, Switzerland. M13-180:

H. Lee1, K. Lee1, Y. Kim1, J. Joung2, K. Moon3, S.-K. Joo3, K.-M. Kim4, G.-J. Cheon4 1 Radiology science, Korea University, Seoul, Korea 2 Advanced Research, Siemens Medical System, IL, USA 3 School of Electrical Engineering, Korea University, Seoul, Korea 4 Molecular Imaging Research Center, Korea Institute of Radiological & Medical Sciences, Seoul, Korea

Image Registration for PET/CT and CT Images with Particle Swarm Optimization

Image registration is a fundamental task in image processing used to match two or more images. It gives new information to the radiologists by matching images from different modalities. The objective of this study is to develop 2D image registration algorithm for PET/CT and CT images acquired by different systems at different times. We matched two CT images first(one from standalone CT and the other from PET/CT) that contain affluent anatomical information. Then, we geometrically transformed PET image according to the results of transformation parameters calculated by the previous step. We have used Affine transform to match the target and reference images. For the similarity measure, mutual information was explored. Use of particle swarm algorithm optimized the performance by finding the best matched parameter set within a reasonable amount of time. The results show good agreements of the images between PET/CT and CT. We expect the proposed algorithm can be used not only for PET/CT and CT image registration but also for different multi-modality imaging systems such as SPECT/CT, PET/MR and so on.
M13-183:

D. Merhof1, P. Markiewicz2, J. Declerck1, G. Platsch3, J. Matthews2, K. Herholz2 1 Siemens Molecular Imaging, Oxford, UK 2 Wolfson Molecular Imagin Centre, The University of Manchester, Manchester, UK 3 Siemens Molecular Imaging EU, Erlangen, Germany

Classification Accuracy and Robustness of Multivariate Analysis Applied to ECD SPECT Data in Alzheimer's Disease Patients

With increasing life expectancy in developed countries, there is a corresponding increase in the frequency of diseases typically associated with old age, in particular dementia. In recent research, multivariate analysis has shown potential for classification between Alzheimer's disease (AD) patients and healthy controls. In this work, the feasibility of multivariate analysis using Principal Components Analysis (PCA) and Fisher's Discriminant Analysis (FDA) of Single Photon Emission Computed Tomography (SPECT) data is investigated. In order to obtain robust an reliable results, bootstrap resampling is applied and the robustness and classification accuracy of PCA/FDA are investigated. The results indicate that PCA/FDA on SPECT data enables a robust differentiation between AD patients and healthy controls based on three principal components, with a classification accuracy of 88%.
This work is funded by Siemens Molecular Imaging, Oxford, UK. The ECD SPECT data was provided by the Clinic of Nuclear Medicine, University of Erlangen-Nuremberg, Erlangen, Germany.

410

M13-186:

S. David1, M. Hatt1, N. Boussion1, P. Fernandez2, M. Allard2, O. Barrett2, D. Visvikis1 1 Laboratoire de Traitement de l'Information Medicale (LaTIM), U650 INSERM, Brest, France 2 Service de Medecine Nuclaire, Groupe Hospitalier Pellegrin, Bordeaux, France

Multi-Tracer PET Image Fusion Using Fuzzy Logic: a Feasibility Study

Imaging tumor metabolism by fluorodeoxyglucose positron emission tomography (FDG-PET) is now well established as a suitable tool for patient monitoring and early assessment of tumor response. However, radiotracers targeting other tumor characteristics like proliferation, hypoxia and apoptosis offer additional valuable information for a complete response assessment. Fusing all available measurements associated to each tracer may facilitate interpretation of such information to assess the tumor response. In this study, we tested the feasibility of a multi-tracer image fusion method based on fuzzy logic. The fusion process consists of two steps: after putting two different images in a common mathematical framework, fuzzy maps reflecting each tracer distribution are derived. The second step consists of the fusion of both fuzzy maps with a combination operator in order to underline differences and similarities of both tracer distributions. Three approaches, a voxel-by-voxel, mean and median fusions were compared. The evaluation of the fusion process is obtained by computing classification errors of the fused map. The various fusion methodologies were tested on simulated datasets of multi-tracer images of tumours. For the different tumor volumes simulated, distribution associated to each tracer were correctly identified in the fused map using either voxel-by-voxel or median fusion. Because of the additional blur introduced in the resulting fused map, the mean approach is not appropriate. By comparing the classification error of voxel and median approaches, the median fusion seems to be more robust and reproducible. Future work will investigate the extension of the fusion process to more than two tracers, and validation of the method to simulated and clinical datasets in the context of therapy assessment in oncology applications.
M13-189:

ROC Analysis of 3D X-Ray CT Performance for Lesion Detection

C. Shi, Y. Xing Department of Engineering Physics, Tsinghua University, Beijing, China Task based performance and image quality evaluation has been widely researched in recent years. Most of works focused on 2D nuclear medicine imaging. Since X-ray CT dosage has been pushed to lower level and 3D imaging has been a trend, it would be intuitive to study the task based performance for 3D X-ray CT systems. In this work, we investigate the application of ROC analysis of lesion detection from 3D X-ray CT images. Using the methodology of classical ROC analysis, we studied the difference in performance of detecting hot and cold lesion in X-ray CT. Moreover, to explore the advantage of 3D imaging, we proposed our method of combining information from orthogonal slices and quantitatively evaluated the improvement of performance we can obtain compared with using 2D images only. Our experimental results demonstrated that detection of cold lesion is much more difficult than detection of hot lesion. Also, volumetric images will improve the detection performance.
M13-192:

A Physical Phantom Evaluation of an Absolute Quantitation of Simultaneous Tc-99m/In-111 SPECT

S. Shcherbinin, A. Celler Radiology, The University of British Columbia, Vancouver BC, Canada In dual-radionuclide Tc-99m/In-111 SPECT imaging, the contamination of In-111 photons into Tc-99m energy window can destroy both the quality and accuracy of Tc-99m image. In this study, we propose a model-based reconstruction strategy for a practical and quantitatively correct Tc-99m/In-111 dual-isotope SPECT imaging. In our approach, instead of time-consuming modeling of downscatter, we re-scale the second-order scatter distribution, analytically calculated for both In-111 photopeaks. Specifically, our algorithm includes the following steps: (i) Quantitative reconstruction of In-111 image (with corrections for attenuation and resolution loss and with analytically modeled self-scatter) from both In-111 photopeak energy windows. (ii) Calculation of the spill-down component (SDC) using extrapolation of the calculated in the previous step noiseless distributions of second-order self-scatter to the Tc-99m energy window using a quick simulation of energy spectra and calibration measurements for all three energy windows. (iii) Quantitative reconstruction of Tc-99m with SDC included in the forward step of the OSEM algorithm. A series of experiments with a clinical hybrid SPECT/CT system (Infinia Hawkeye, GE Healthcare) were designed to examine how accurately our algorithm can reconstruct Tc-99m activity from the projection data containing various levels of In-111 contamination. We used four 30ml plastic containers with different ratios of Tc-99m and In-111 activities placed inside the Jaszczak phantom. The absolute activities for these containers were reconstructed with errors 0.8-7.0% (In-111) and 1.6-13.1% (Tc-99m). Especially, in one of our experiments, the developed SDC compensation technique decreased the absolute activity errors from 44.8% to 8.2% (container #1 filled with Tc-99m activity only); from 61.5% to 1.6% (container #2 with Tc99m to In-111 concentration ratio of 2:1), and from 132.8% to 13.1% (container #3 with Tc-99m to In-111 concentration ratio of 1:2).

411

M13-195:

Anatomy Assisted MAP-EM PET Image Reconstruction Incorporating Joint Entropies of Wavelet Subband Image Pairs

J. Tang, A. Rahmim Department of Radiology, Johns Hopkins University, Baltimore, MD, USA A promising approach to incorporate high resolution anatomical information (e.g. MRI, CT) into functional PET image reconstruction is to utilize anato-functional joint entropy (JE), a component of the standard mutual information (MI) measure. Nevertheless, a potentially significant factor regarding standard MI/JE as applied to images is that spatial information is not considered; i.e. a voxel is classified merely based on its intensity information, and structural spatial information from its neighborhood is neglected. In this work, we have implemented an anatomy assisted MAP-EM algorithm wherein the JE measure is supplied by spatial information generated using wavelet analysis. This approach has the benefit of utilizing some theoretical advantages of wavelets, including the ability to decompose an image of certain size into downsampled subbands, allowing concise representation of the image into individual wavelet-derived feature images, while preserving original information. The proposed MAP-EM algorithm involves calculation of derivatives of the JE measures with respect to PET image intensities, which we have shown can be computed very similar to how wavelet reconstruction is performed. Using simulations of a mathematical human brain phantom with activities generated based on a clinical FDG study, it was observed that compared to conventional EM reconstruction, the proposed MAP-EM algorithm exhibited improved quantitative performance, as assessed by bias (NMSE) vs. noise (NSD) trade-off curve comparisons in a number of regions of interest
M13-198:

I. Hong1,2, Z. Burbar2, C. Michel2, R. Leahy3 1 Dept. of Com. Eng, Korea Polytechnic University, Seoul, Korea 2 Molecular Imaging, Siemens Healthcare, Knoxville, TN, USA 3 Signal and Image Processing Institute, Univ. of Southern California, LA, CA, USA

Ultrafast Preconditioned Conjugate Gradient MAP Reconstruction for Fully 3-D microPET

Iterative 3D PET reconstruction represents a very computational challenge due to the large number of lines of response (LOR) collected for each data set. This iterative 3D reconstruction also needs a lot of iterations to achieve an acceptable PET reconstructed image. A Preconditioned Conjugate Gradient (PCG) method was previously shown to have faster convergence rate than expectation maximization (EM) type algorithms. For the microPET, imagea suffer from crystal penetration blurring due to small scanner radius. An exact 2D blur model is needed to achieve high resolution image. A Preconditioned Conjugate Gradient (PCG) method is described for reconstruction of high-resolution 3D images from the microPET Inveon small-animal scanner from Siemens. The projector pair is used as part of a factored system matrix that takes into account detector-pair response by using shift-variant sinogram blur kernels, attenuation correction, and detector efficiency corrections. The system matrix for geometric projection is based on depth dependent solid angle calculation in combination with a spatially variant detector response model. The mircoPET PCG is combined with OSEM to accelerate convergence. This reconstruction model achieves a high resolution animal image; however, it took an hour to reconstruct a frame. Therefore, we describe an ultrafast forward and back projector pair based on Symmetry and SIMD projector (SSP). The proposed method produces similar quality images when compared to those obtained with the software package from Siemens and requires order of magnitude less computation.
M13-201:

K. Zeng1, B. De Man1, J.-B. Thibault2, Z. Yu3, C. Bouman3, K. Sauer4 1 CT/X-ray Lab, GE Global Research Center, Niskayuna, NY, USA 2 Applied Science Laboratory, GE Healthcare, Waukesha, WI, USA 3 School of electrical engineering, Purdue University, West Lafayette, IN, USA 4 department of electrical engineering, University of Notre Dame, Notre Dame, IN, USA Iterative reconstruction (IR) has recently been proposed to improve multiple aspects of image quality over conventional filtered backprojection (FBP) in X-ray computed tomography (CT). FBP reconstruction and its corresponding reconstruction kernels have been optimized for decades to provide the best possible image quality. IR does not have the notion of reconstruction kernels but uses other mechanisms to change the image resolution and image noise. This paper presents one computationally efficient technique to enhance the spatial resolution of IR images reconstructed from high resolution scans, based on the introduction of an enlarged voxel footprint in the forward model, combined with a band-suppression filter designed to eliminate any undesirable over- or under-shoot artifacts that may arise from the use of the enlarged voxels. The proposed technique achieves higher spatial resolution than high resolution FBP with significantly lower noise. Results are shown on both phantom and clinical patient data.

Spatial Resolution Enhancement in CT Iterative Reconstruction

412

M13-204:

PET Image Reconstruction Using LOR-OSEM with a 3D Spatially Variant System Matrix

D. B. Wiant, J. D. Bourland Radiation Oncology, Wake Forest Univ. School of Medicine, Winston-Salem, NC The use of PET imaging in radiation therapy treatment planning and evaluation is limited by the spatial resolution of PET images. Improved resolution has been observed in images reconstructed with iterative algorithms using system matrices (SM) that include spatially variant point spread functions (PSF). In this work we model a PSF that is spatially variant in three dimensions for the GE Discovery ST PET scanner in 2D acquisition mode and use it derive a SM. The PSF was sampled over a 5 mm grid that covered the area corresponding to one block and block gap in half of the image planes (6336 points) using a 0.5 mm 68Ge point source. This information was used to determine PSF blurring kernels that are spatially variant in the radial r, depth d, and axial directions z. These kernels, with a geometric component, were used to create SMs for non-uniformly spaced lines of response that are used by LOR-OSEM reconstruction algorithms. A point phantom and a 220 mm diameter cylindrical phantom containing either hot spheres or cold spheres were scanned and reconstructed using LOR-OSEM with a 1) geometric SM, 2) geometric + PSF(r,z) SM, 3) geometric + PSF(d,z) SM, and 4) geometric + PSF(r,d,z) SM to directly compare the effect of PSF spatial dependencies on reconstructed image quality. We found that all PSF blurring models offer better resolution and bias-noise ratios than LOR-OSEM alone. Reconstructions that use PSFs modeled with depth dependence show a finer resolution of point sources than LOR-OSEM or LOR-OSEM+PSF(r,z) at radii > 150 mm. Near the center of the FOV all of the PSF methods showed similar resolution and bias-noise in reconstructions of point sources and hot and cold spheres. The use of a PSF dependent on r, d, and z gave improved resolution at radial locations > 150 mm. However, the increased ill-conditioning introduced by PSF blur and non-uniform LORs led to the appearance of Gibbs artifacts near the borders of spherical objects in the hot and cold phantoms,
This work was supported by NCI T-32 CA1133267 M13-207:

R. Grimmer1, T. Berkus2, M. Oelhafen2, P. Kunz2, M. Kachelriess1 1 Institute of Medical Physics, University of Erlangen-Nuernberg, Erlangen, Bavaria, Germany 2 Imaging Laboratory GmbH, Varian Medical Systems, Baden-Daettwil, Switzerland

Cone-Beam CT Sequence Scan Reconstruction with Improved Dose Usage and Scan Coverage

For circular cone-beam CT one scan covers often not the complete z-range of interest. If this is the case two or more circle scans are made. These sequence scans are typically reconstructed by separately reconstructing each circle scan followed by combining the resulting partial volumes. This image-based concatenation method uses only those data that are needed for each partial volume, the contribution of rays to neighboring volumes is ignored, redundancies are not used, and dose is wasted. We developed and evaluated an algorithm that uses all rays that run through a voxel by appropriately weighting the rays followed by filtered backprojection. This leads to improved dose usage and increases the overlap region of neighboring volumes, potentially leading to reduced artifacts in this region. Alternatively, our approach allows to increase the table increment in-between adjacent circle acquisitions and thereby the scan coverage, without impairing image quality or increasing dose. To evaluate our method we use the geometry of the Varian OBI flat panel detector CT scanner. Simulated and measured data are processed at varying table increment values and an evaluation of image noise, spatial resolution and artifacts has been performed. The method shows good image quality on simulated phantom data as well as on clinical patient scans. In this paper the algorithm demonstrates its ability to extend the z-range of sequence scans, to improve the image quality in the overlap region, and turns out to be usable on devices in the clinical practice.
M13-210:

F. Bergner1, T. Berkus2, M. Oelhafen2, P. Kunz2, M. Kachelriess1 1 Institute of Medical Physics, University Erlangen-Nuernberg, Erlangen, Germany 2 Varian Medical Systems, Baden-Daettwil, Switzerland

A Comparison of 4D Cone-Beam CT Algorithms for Slowly Rotating Scanners

The study compares several algorithms for the 4D reconstruction of cone-beam computed tomography (CBCT) data that were recently proposed and which can be used for slowly rotating devices. In this case the imaging units are mounted to linear particle accelerators (LINAC). The algorithms are the conventional phase-correlated reconstruction, the McKinnon/Bates-Algorithm, the prior image constrained compressed sensing (PICCS) algorithm, the total-variation minimization (TV) algorithm, and our autoadaptive phase-correlation (AAPC) algorithm. For each algorithm the same motion-affected rawdata are used and the reconstruction results compared to each other regarding their noise and artifact levels, as well as temporal resolution, and computational complexity and convergence. These criteria result in a discussion of the advantages and disadvantages of each algorithm. The temporal resolution is best in the algorithms which exclusively use data from a single motion phase only. The iterative algorithms show lower noise and artifact levels but are computationally complex and therefore have a limite usage in the clinical

413

application. Algorithms which include image enhancements beside a faster reconstruction represent a suitable trade-off for the clinical workflow.
This work was supported by a grant from Varian Medical Systems, Palo Alto, CA. The FDK reconstruction pipeline, simulation, and components were provided by RayConStruct GmbH, Nuernberg, Germany. M13-213:

Synthetic Zooming of Tomographic Images by Combination of Lattices

N. Dixit, N. V. K. Medathati, J. Sivaswamy CVIT, International Institute of Information Technology, Hyderabad,Andhra Pradesh, India We propose a method for synthetic zooming of tomographic images by applying super resolution technique on reconstructed data via a combination of rotated lattices (CRL). The proposed method consists of 2 steps: (i) Sinogram data is filtered and backprojected on to two lattices, which are rotated versions of each other and (ii) the samples from the two lattices are interpolated to find the upsampled image. Square and hexagonal lattices have been investigated for CRL. Results of subjective and objective evaluations of the proposed method on analytic phantoms are presented and compared with direct upsampling of data reconstructed on a single square lattice. The proposed method shows qualitative and quantitative improvement over direct upsampling in general with the combination of hexagonal lattices yielding better results than square lattices.
M13-216:

H. Hsieh1, K. Lin1,2, C. Hsu3, I. Hsiao1,2 1 Department of Medical Imaging and Radiological Sciences,, Chang Gung University, Taoyuan, Taiwan 2 Molecular Imaging Center, Chang Gung Memorial Hospital, Taoyuan, Taiwan 3 Dept. of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsin-Chu, Taiwan Stationary multi-pinhole SPECT system allows acquisition of multiple projections at the same time such that the high-resolution kinetic information can be obtained without rotating the camera. In this case, image reconstruction can be performed on the data collected without rotating detector heads and enable dynamic data acquisition as PET does. In stationary multi-pinhole SPECT system, angular sampling is constrained by number of pinholes. The limited angular scanning is not sufficient to provide angular sampling data around the object, and finite angular sampling and insufficient sampling coverage can result in streak artifact and/or geometric distortion. We are developing a stationary multi-pinhole SPECT system based on a clinical three-head SPECT system. A ring type collimator covered by the 3 detector heads was designed with 27 pinholes focusing on FOV center to achieve high sensitivity and high resolution. The goal of this paper is to evaluate different reconstruction approaches on image resolution from sparse or insufficient projection data in the stationary multi-pinhole SPECT system from simulated data. These methods include OSEM, OSEM with TV optimization, MAP-TV with smoothing prior, and MAP-TV with edge-preserving prior. The results showed the benefit on restoring both the effect on finite angular sampling and insufficient sampling coverage from TV optimization, but noisier on reconstruction image than the method above. In conclusion, for the stationary multi-SPECT reconstruction, OSEM without TV seems to generate compatible image to those with TV optimization. Further study on OSEMTV is needed for the TV optimization parameters as well as the weighting between TV and OSEM reconstruction. In addition, MAP reconstruction with edge-preserving and smoothing prior without TV optimization should be studied as well.
This work was supported by Grants NSC 97-2314-B-182-029-MY3 from NSC, Taiwan, and CMRPD34005 from the Research Fund of Chang Gung Memorial Hospital, Taiwan. M13-219:

Performance Evaluation on Reconstructions in a Stationary Multi-Pinhole SPECT

List-Mode Wavelet-Based Multiresolution Image Reconstruction for Compton Imaging

M. Frandes, V. Maxim, R. Prost CREATIS-LRMN, University of Lyon, Lyon, France Iterative image reconstruction of data measured by a Compton scattering camera must overcome various difficulties, e.g. a large amount of data, noise arising from both low counts recorded and imaging detector response, etc. Several approaches were proposed trying to attain an efficient Compton scattered data reconstruction. A forward step is introduction of list-mode data acquisition, which represents a mandatory way of events measurement allowing to preserve a full precision of measured events as well as to gain computation efficiency. Iterative approaches include e.g. ML based algorithms, Bayesian methods, etc. Image estimation by ML criterion brings noise amplification when an iterative algorithm is applied, so a regularization step is needed. The proposed solution is a regularization technique using wavelet-based thresholding of the OSEM correction factors, called WOSEM. Methods of wavelet thresholding are efficient for edge-preserving image de-noising. The threshold used is the universal parameter proposed by Donoho et al. Improvements include estimation of level-dependent threshold, which should admit wavelet coefficient corrections according to the expected noise level.

414

M13-222:

Z. Wang1,2, L. Zhang1,2, Z. Huang1,2, Z. Chen1,2, K. Kang1,2 1 Engineering Physics, Tsinghua Univercity, Beijing, China 2 Key Laboratory of Particle & Radiation Imaging (Tsinghua University), Ministry of Education, Beijing, China

Linear Partial Derivative Matrix for Iterative Algorithm to Reconstruct Refractive Index from Refraction Angle Data

X-ray phase contrast imaging (PCI) has great potential in the medical diagnosis, material science et al. Diffraction enhanced imaging and grating-based imagingare the two significant methods for PCI, currently. The commonness of the two methods is that the projection data is the partial derivative of the refractive index, which is also the refraction angle. Both of the two methods are named for differential phase contrast imaging. Analytical algorithms to reconstruct the refractive index have been proposed. The basic idea is that after a one-dimension Hilbert transform, the refraction angle projections can be used to reconstruct the refractive index by conventional filtered back-projection algorithm (FBP). In spite of these algorithms, there is lack of an iterative algorithm to reconstruct the refractive index directly from the refraction angle data. In this paper, an algebraic iterative algorithm is proposed to reconstruct the refractive index from the refraction angle data. Our algorithm is a complementarity to the analytical reconstruction algorithms and it inherits the benefits of iterative algorithms that can work well with in-complete projection data. The proposed algorithm is validated by numerical simulation and the results show that it can reconstruct the refractive index accurately and the iteration is convergent and steady.
M13-225:

O. V. Olesen1,2, M. Sibomana1, S. H. Keller1, F. Andersen1, J. Jensen3, S. Holm1, C. Svarer4, L. Hoejgaard1 1 PET Center, KF 3982, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark 2 IMM, DTU, Copenhagen, Denmark 3 Dept. of Electrical Engineering, DTU, Copenhagen, Denmark 4 NRU, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark

Spatial Resolution of the HRRT PET Scanner Using 3D-OSEM PSF Reconstruction

The spatial resolution of our HRRT Siemens dedicated brain PET scanner was measured using a point-source phantom with high statistics. Further, it was demonstrated how the newly developed 3D-OSEM PSF reconstruction can improve the resolution in reconstructed images with high signal-to-noise-ratios. A micro drop of 18F-FDG solution (800-300kBq) was drawn into a 20l capillary tube with an inner diameter of 0.3-0.7mm. The scans were reconstructed using either the ordinary 3D-OSEM or the improved 3D-OSEM PSF algorithms (16 subsets and 6 iterations). A zoom factor of 5 was used to decrease the size of the reconstructed voxels (0.24mm/pixel in the transaxial axes) to fit a Gaussian distribution to the intensity profiles and the Full Width Half Maximum (FWHM) was determined. The resolution of the HRRT is centrally (r < 60mm) homogenous with a FWHM of 1.4mm for 18F-FDG in air. This is where the main part of the brain is located if the patient has been positioned correctly. The 1.4mm resolution is obtained using the newly develop 3D-OSEM PSF reconstruction algorithm, which is a significant improvement over 3D-OSEM reconstruction without PSF.
M13-228:

L. Wan1,2, Y. Zhang3, X. Cao4, S. Zeng1,2, Q. Xie1,2 1 Department of Biomedical Engineering, Huazhong University of Science and Technology, Wuhan, Hubei, China 2 Wuhan National Laboratory for Optoelectronics, Wuhan, Hubei, China 3 Department of Nuclear Medicine,Union Hospital, Huazhong University of Science and Technology, Wuhan, Hubei, China 4 School of Mathematics and Statistics, Huazhong University of Science and Technology, Wuhan, Hubei, China In this work, we propose a novel positron emission tomography(PET) attenuation correction method with attenuation map library. Our concept is summarized as follows: Firstly, we construct a library containing a large number of attenuation maps for the specific object from different patients. Secondly, by characterizing the statistical information of these maps, we build the statistical template to represent these objects. The statistical template is used to implement attenuation correction for this kind of objects. In this work, we initially evaluate the performance of this method. To construct the attenuation map library, we generate 30 digital phantoms with 30 attenuation maps. The phantoms are all consisted of three parts and the linear attenuation coefficients of same part from the 30 phantoms follow Gaussian distribution. To reduce the complexity, the phantoms are with absolutely same geometrical characterization. By averaging the 30 attenuation maps, we build the statistical template of attenuation map for these phantoms. We employ the Monte-Carlo simulation to generate the PET emission data. The results are compared with the conventional PET transmission scan method and current PET/CT method. Key words: Attenuation Correction, Attenuation Map Library, Statistical Template

An Investigation of Attenuation Correction with Attenuation Map Library in PET Imaging

415

M13-231:

Determination of Absorbed Dose Measurement Between ART Phantom and CADPLAN of 10 MV X-

Ray
R. Abdullah Clinical Oncology Unit, University Malaya Medical Centre, Kuala Lumpur, Malaysia On behalf of the Mohd. Moktar Bin Mat Nudin@Mat Nor A study about inter comparison of absorbed dose measurement with CadPlan 3-Dimensional system treatment planning and thermoluminescent dosimeter (TLD-100) by using ART anthromorphic phantom with high energy X-ray was done to achieve homogeneity of dose throughout the target volume of 5%. The accuracy and precision in dose calculation by using computerized treatment planning systems for high energy photon has been determine throughout the average of dose comparison. The measured dose from TLD-100 in the phantom is compared with the calculated dose of CadPlan 3-D system treatment planning which the dose is delivered with an error of less than 5%. The result showed that percentage deviation of target volume is 1.46%. This means that the planning of the treatment dose through CadPlan 3-D system treatment planning is accurate, because of the maximum dose was given to the treated volume without increasing the risk of tissues. Wedge 300 and Multi-Leaf Collimator were used in this study with the accuracy within 5%, to avoid any complication of the radiobiological risk to the organs at risk (eyes = 1.6%). The dose of spinal cord was under-dose with 31.01%. This was because of the existent of air-gaps in between the slices of phantom, which had influenced the reading of the TLD-100. As a conclusion, CadPlan 3-D system treatment planning is suitable for the radiotherapy treatment. Key words: CadPlan 3-D System Treatment Planning, TLD-100, target volume, homogeneity, high energy photon
M13-234:

M. Bati1, J. Burger2, V. Cindro1, G. Kramberger1, I. Mandi1, M. Miku1,3, A. Studen1, M. Zavrtanik1 1 Experimental Particle Physics Department, Institute Joef Stefan, Ljubljana, Slovenia 2 Department of Radiophysics, Institute of Oncology, Ljubljana, Slovenia 3 Department of Physics, University of Ljubljana, Ljubljana, Slovenia

Localization of High Dose Rate Ir-192 Source During Brachytherapy Treatment Using Silicon Detectors

A system for in vivo tracking of 1 Ci 192Ir source during brachytherapy treatment has been built using high resistivity silicon pad detectors as image sensors and knife-edge lead pinholes as collimators. The sensors consists of 256 pads arranged in 32 by 8 grid with pad size 1.4 1.4 mm2 and 1 mm thickness. The sensors are made with two metal layers, enabling connection of readout electronics (VATAGP3_1 chips) at the edge of the detector. With source self-images obtained from a dual-pinhole system, location of the source can be reconstructed in three dimensions in real time, allowing on-line detection of deviations from planned treatment. The system was tested with 1 Ci 192Ir clinical source in air and plexi-glass phantom. The movements of the source could be tracked in a field of view of approximately 20 20 20 cm3 with absolute precision of about 10 millimetres. Positions of the source, relative to the first measured source position, could be mapped with precision of around 1.5 millimeter. In the contribution more details about sensors, readout system and reconstruction algorithm will be described. Results of both real measurements with clinical source and simulation will be presented.
M13-237:

H. M. T. Thomas1, D. Devadhas2,3, S. Purnima1, S. Balukrishna1, B. P. Ravindran1 1 Department of Radiotherapy, Christian Medical College, Vellore, Tamilnadu, India 2 Department of Nuclear Medicine, Christian Medical College, Vellore, Tamilnadu, India 3 Department of Radiology, Thomas Jefferson University, Philadelphia, Pennsylvania, USA

Evaluation of Radiotherapy Treatment Planning with Mega-Voltage Cone Beam CT

The rationale for MV CT for patient treatment planning are prospective use in treatment planning for consecutive phases in 3D conformal and intensity modulated radiotherapy, patients with metal implants, adaptive radiation therapy and single fraction palliative treatment, this modality has potential for further growth. This study is to validate the MV cone beam CT images generated by proprietary 3D reconstruction software based on FDK algorithm for megavoltage treatment planning. The projection images were acquired based on half scan cone beam technique from the EPID system of a commercial linear accelerator. It was noted that the treatment planning system did not use the rescale slope and intercept provided in the DICOM header of the CT images for HU calculations. Rather it calculated the HU from the pixel values of the images. So the pixel values in the MV CBCT images were rescaled to represent a pixel value equivalent to that of the kV CT images. Validation of CT number was done using standard HU value calculation and with a calibration phantom with known inserts. The CT numbers calculated with both methods were comparable to the CT number of kVCT. The whole brain was delineated as the target volume and forward planning with parallel oppose field with micro leaf collimator shaping and 6 MV beam was performed on both kV and MV CBCT images. 6 points were identified within the target volume on both datasets. Relative dose was calculated. Dose volume histograms and volume comparisons of the target were done. The volumes of the target were measured on kV and MV CBCT as 940.9 cc and 1020 cc respectively. The relative dose for the points in the target volume in MV CBCT was within 3% 416

difference from dose calculated with kV CT. The results of this study show that MV CBCT images could be effectively used for planning in a commercial treatment planning system, if the necessary corrections mentioned are incorporated.
M13-240:

J. E. Baciak1, Y. Feng2, Z. Li3 1 Nuclear and Radiological Engineering, University of Florida, Gainesville, FL, USA 2 DCH Cancer Center, Tuscaloosa, AL, USA 3 University of Florida Proton Therapy Institute, Jacksonville, FL, USA

Utilization of Photon Imaging for in-Vivo Dose Verification and Localization in Proton Therapy

In this study, we investigate the use of in-vivo Compton imaging to improve the accuracy of proton dose delivery and verification. Proton therapy delivers high dose to tumors while reducing dose to healthy tissue by making use of the Bragg Curve associated with charge particle energy loss. However, since the energy of proton beam is generally selected to stop the beam completely inside the body, dose verification by measurement transmitted particles is not feasible. In the process of proton delivery, there are a number of characteristic photons that are generated by the variety of interactions between protons and atoms. The current method to providing dose estimation is to utilize PET imaging after the procedure is complete. However, this technique presents challenges since the patient must be transported to the PET facility and this can take 15-20 minutes after the procedure. Thus, the quality of the PET image in providing dose information is poor due to the fast decay of the positron emitting isotopes in the patient. The design of a compact Compton imaging system that can be placed in the proton treatment room will allow for in-vivo photon measurements that can reconstruct the proton delivery and estimate dose. We simulate the proton beam to generate the output photon spectrum from the patient. Utilization of ordered subset maximum likelihood imaging reconstruction algorithms have shown that we can produce an image with 5 mm position resolution with only 20 iterations, and further improvements in precision can be achieved by optimization of the imaging system design. Simulations of cylindrical sources show that it is possible to image the emerging photons without significant loss in position resolution for the 511 keV annihilation photons created within a patient during actual treatment (potentially in real time). Other higher-energy characteristic photons can also be imaged.
M13-243:

Evaluation of positron probe-guided neurosurgical resection using gel phantoms

M.-A. Park, W. Wu, S. C. Moore, A. Golby Brigham and Women's Hospital, Boston, MA, USA Maximal surgical excision is currently the best treatment for brain tumors; however, delineation of resection margins is difficult because tumors resemble normal brain and may infiltrate. We have created phantoms to evaluate the use of a positron probe for detection of residual tumor tissue and, eventually, for training surgeons for margin resection. Phantoms were prepared using a mixture of 5% (4%) porcine gelatin and 5% (4%) agarose to resemble normal (tumor) brain tissues. To generate a boundary that could be visually discriminated, as is the case for patient surgery, we added green dye to both tissues, but with a higher concentration for the tumor. FDG was added, with a tumor-to-normal tissue uptake ratio of 2.7. In the center of the cylindrical phantom, we positioned a mold which was later filled with the tumor gel. Before adding the tumor, a CT image (CT0) of the phantom was acquired on a CT system with 0.12mm pixel size, which was used to measure the volume of the tumor. Warm tumor gel was then poured into the mold, which yielded a seamless boundary. A neurosurgeon resected the tumor, first, based solely on the appearance and texture of the tumor. A CT scan (CT1) was acquired immediately after the first resection, following which the surgeon used the positron probe to localize and excised residual tumor. The probe consisted of a 3mm diameter plastic scintillator with a long cylindrical handle containing a photomultiplier tube. A final CT image set (CT2) was obtained after the second resection. Resected tumor volumes were measured for each surgical procedure using the CT images. For two phantoms containing tumors of different size, the surgeon removed 52% (60%) of the small (large) tumor, respectively. With the aid of the probe, he was able to remove 70% (97%) of these tumors. We have shown that gel phantoms can improve surgeons' ability to delineate tumor margins, and can serve as valuable training tools for positron-probe guided surgical resection.
M13-246:

Development of a Next Generation MR Compatible PET-Detector for Small Animal Imaging

K. Lankes, M. S. Judenhofer, M. Hossain, B. J. Pichler Laboratory for Preclinical Imaging and Imaging Technology of the Werner Siemens-Foundation, University of Tuebingen, Tuebingen, Germany Multimodality imaging is an emerging field in both, biomedical research and clinical diagnosis. Instead of PET/CT what is already established, our group focuses on the next generation of imaging modalities: the combination of positron emission tomography and magnetic resonance imaging (PET/MRI). Our current development is concentrating on a new PET device which fits into a 7 Tesla animal MRI system (Bruker) and has a large field of view (FOV) in axial and transaxial direction to allow whole body mouse and multi-bed rat scans. A full-ring of the next generation PET-MRI system consists of 16 PET detector cassettes, with 3 LSO-APD detector blocks each, arranged in axial direction. A detector block uses a 15 x 15 LSO crystal array 417

coupled over an in-house developed tapered light guide to a 3 x 3 avalanche photodiode (APD) array. The APD signals are read out with a 9-channel charge sensitive preamplifier (Siemens). Three of these detector blocks are placed in axial direction on a master electronic board, forming an axial FOV of about 72 mm. To read out the detector blocks we are using the QuickSilver (QS) electronic (Siemens). Therefore a signal shaping circuit is needed to adapt the requirements of the QS. An efficient and fast responding cooling system is required to enable the readout of a 15x15 scintillation crystal block. Our decision was to develop a cooling system based on pressurized air and cyclone tubes. With that we could reduce the temperature from 32C down to 12C. With this setup we improved the energy resolution in the center of the block from 20% to 16% (FWHM). Currently we are assembling the individual detector cassettes to build up a ring of MR compatible PET detectors.
M13-249:

Imaging Tests with Silicon Photomultipliers Made from MPPC Arrays in Magnetic Fields up to 14

Tesla

S. Majewski1, J. Proffitt2, J. McKisson3, R. Raylman1, A. Stolin3, S. Velan1, A. Weisenberger3 1 Center for Advanced Imaging/Radiology, West Virginia University, Morgantown, WV, USA 2 Adaptive I/O Technologies, Inc., Blacksburg, VA, USA 3 Radiation Detector and Imaging Group/Physics, Jefferson Lab, Newport News, VA, USA

We have constructed and tested several variants of small compact imaging photomultipliers made of 4x4 to 6x12 arrays of individually read 3x3mm2 Hamamatsu MPPC devices. The 3x3mm2 units were intentionally not arranged in the tightest possible arrangement, in an effort to lower the cost and complexity of the device. The package includes low power on-board input stage amplifiers for each SiPM element and a sum amplifier placed on an additional but distant electronics board. Signal and imaging operation in the MRI fields of 3, 7 and 14 Tesla were performed. Results show that the devices can perform well in the tested fields with only small percentage decrease in the signal amplitude, and the only encountered issue was the temperature stability, which was the main contributing factor to the operational stability. Even at the highest tested field of 14 Tesla the effect of magnetic fields on signal amplitude was less than 10%. Once temperature correction factors are experimentally derived (and temperature reasonably stabilized), MPPCs, are very good candidates for MRI compatible photosensors. The important remaining issue is the effect of the MPPCs and their electronics on the MRI operation, and we are performing now these interaction studies now, with results expected by the time of the conference.
M13-252:

V. Keereman1, S. Vandenberghe1, J. De Beenhouwer1, R. Van Holen1, S. Staelens1, V. Schulz2, T. Solf2 1 MEDISIP, Ghent University-IBBT-IBiTech, Ghent, Belgium 2 Molecular Imaging Systems, Philips Research Europe, Aachen, Germany System design research for upcoming PET-MR scanners has mainly focussed on the effect of the high magnetic field on PET performance and on the influence of the PET scanner inside the MR bore on MR image quality. However, the presence of MR components close to the PET detectors could also have an influence on PET performance. We have investigated these effects in a simulation study of the preclinical PET-MR insert of the HYPERimage project. Simulations were performed with the ProcessGATE extension of the GATE simulation framework, which makes it possible to determine the fractions of total scatter caused by different components. A preclinical insert was simulated inside a clinical MR scanner. All components of the clinical system and the preclinical insert were modeled in realistic dimensions and materials. The PET detector consisted of 10 detector blocks on a 100 mm radius cylinder, each containing a 44 (tangential) by 72 (axial) array of LYSO crystals. The crystal dimensions were 1.3 x 1.3 x 10 mm. A uniform cylinder (radius 5 mm, length 100 mm) filled with 1 MBq of 18F was simulated. The energy window was set to 250 - 750 keV. The simulated time was 1s yielding one million simulated decays. Only 47% of detected singles were unscattered. The clinical system and precinical insert accounted for respectively 38% and 15% of scattered photons. On the coincidences level the influence of the clinical system was much smaller (17%), while the scatter effect of the insert increased (20%). In the clinical system the gradient coils scatter the largest fraction of photons (58%). In the insert over 65% of scatter is caused by the table and the RF screen. In conclusion, it is clear that putting MR components within or close to the FOV of a PET scanner can cause significant scatter. Setting a smaller energy window could eliminate part of the scattered photons. The scattering effect of the MR components should however also be taken into account in the design phase.
This work was supported by the EU FP7 project HYPERimage (Grant Agreement N 201651).

Scatter Effects of MR Components in PET-MR Inserts

418

M13-255:

S. Vandenberghe1, V. Keereman1, S. Staelens1, V. Schulz2, P. Marsden3 1 MEDISIP, ELIS IBBT, Ghent University, Gent, Belgium 2 Molecular Imaging Systems, Philips Research Europe, Aachen, Germany 3 PET Imaging Centre, Guy's, King's and St Thomas' School of Medicine, London, UK

Effect of Geometrical Constraints on PET Performance in Whole Body Simultaneous PET-MR

Introduction: Simultaneous PET-MR scanners are being developed for whole body imaging. An important modification will be geometry PET scanner design which is determined the space constraint imposed by the MR scanner. This limits the geometry of the PET scanner and the possibility for end shielding. The aim of this study is to determine the effect of modified geometry and reduced shielding on the PET performance with regards to scatter and random coincidences. Materials and methods: All data were simulated using the GATE Monte Carlo simulation tool. The reference system for the simulation was the Gemini TF scanner with LYSO crystals ( Philips Medical systems). This system has a diameter of 90 cm and end shields with an inner diameter of 70 cm. The energy resolution of the system was 12 % and based on this system a whole body PET scanner was designed that would fit inside a 3T MR scanner. This system was simulated without end shielding and with limited end shielding (60 cm diameter). The object used was the 70 cm long NEMA scatter phantom. Results: The reduction to a diameter of 70 cm results in an increase of the amount of trues by 28 %. The relative scatter fraction increased from 33 % to 36 % for the 70 cm diameter system without end shields. The introduction of short shields resulted in a small reduction (2-3 %) of scattered and random coincidence fraction. More detailed analysis about origin of the events showed that in the new design 85 % of scattered events originates from inside the FOV, while 90 % of the random coincidences is caused by outside FOV activity. Conclusions: Given the good energy resolution of current systems, end shields only play a limited role in the reduction of scatter. The end shields are only blocking a limited part of the scattered outside FOV activity. The combined effect quantified by the Noise equivalent counts even results in a clearly higher value than the current PET design due to large increase in true events.
The presented work is part of the EU FP7 project HYPERImage, Grant Agreement N 201651. M13-258:

Y. Huh1, Y. Choi1, W. Hu1, J. Kang1, J. Jung1, K. Hong1, G. Lim2, B. Min1, S. Shin1, H. Lim1 1 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 2 Department of Center for Molecular and Cellular Image, Samsung Biomedical Reserach Institute, Seoul, Korea A silicon photomultiplier (SiPM) array-based MRI compatible PET is being developed. The purpose of this study is to develop a PET signal acquisition method for simultaneous PET/MRI imaging. The detector for this MRI compatible PET was composed of a 3 mm x 3 mm x 20 mm LYSO scintillator coupled with a 3 mm x 3 mm silicon photomultiplier (SiPM) array. PET signal acquisition was implemented by a FPGA-based data acquisition (DAQ) system. RF noise of MRI has large effect on PET signal acquisition when PET detectors are placed inside MRI. In this study, to reduce the amount of data to be acquired by DAQ system, a FPGA-based method which uses a maximum rise window and an integration (energy) window for real time RF noise rejection has been developed. To evaluate this acquisition method, a PET detector was placed both outside and inside a 7-T small animal MRI (BioSpec 70/30 USR, Bruker Biospin). When it was placed inside MRI, two types of RF pulses (Gradient Echo Sequence and T1 Sequence) were transferred to the MRI scanner. For 5 minutes PET data acquisition, the saved data amounts for outside MRI, inside MRI with gradient echo sequence and T1 sequence were 56.6 MB, 56.6 MB and 402.5 MB. However, acquired energy resolution for outside MRI, inside MRI with gradient echo and inside MRI with T1 sequence were very similar (20%, 19.8%, 20.3%), thus this method can be used for simultaneous PET/MRI imaging. Currently, we are trying to further reduce the saved data amounts for inside MRI with T1 sequence. We believe this method will be useful for the development of MRI compatible PET systems.
M13-261:

A FPGA-Based PET Data Acquisition Method for Simultaneous PET/MRI Imaging

J. H. Kang1, Y. Choi1, K. J. Hong1, J. H. Jung1, W. Hu1, G. H. Im2, B. J. Min1, S. H. Shin1, Y. S. Huh1, H. K. Lim1 1 Dept. of Nuclear Medicine, Samsung Medical Center, SungKyunKwan University, Seoul, South Korea 2 Dept. of Center for Molecular and Cellular Image, Samsung Biomedical Research Institute, Seoul, South Korea For developing a hybrid PET/MR system, it is important to carefully shield the PET components from MR interference and to prevent the distortion of MR image by the PET conductive structures. The purpose of this study is to characterize the crosscompatibility of PET components and MR which is important for designing a hybrid PET-MR and for acquiring high quality images from both modalities. Experimental studies performed by placing the PET module inside and outside RF-coil to examine the mutual interference at different location in MRI. The PET module which consists of LYSO, GAPD and preamplifier was illuminated with Na-22(6 Ci) and the MR images of a CuSO4-filled phantom were acquired under various MR imaging sequences while the PET module was mounted at the MRI iso-center. In addition, MR image quality as a function of Cu shielding thickness and area was examined to quantify the effect of magnetic shielding material on MRI. The thickness and area 419

Characterization of Cross-Compatibility of PET Components and MRI

of copper was varied from 0 to 200 and from 30 to 120 , respectively. The PET detector counting rate was decreased up to 80% and the significant artifacts were observed in MR image when the PET module was placed inside RF-coil. In contrary, no obvious degradation of the MR image quality was observed by placing the PET module outside RF-coil. Shielding material (Cu) thickness did not affect the homogeneity but SNR of MR images changed from 150 to 110 when the copper thickness changed from 0 to 200 . The SNR and homogeneity were considerably changed from 270 to 29 and from 88 to 69 when 30 and 120 shielding area were employed. The temperature of shielding material was risen by ~ 1 when large area shielding (120 ) was used. In summary, our preliminary results demonstrate that the cross-compatibility between PET and MRI would be minimized by placing the PET module between RF and gradient coils. Thin and segmented shielding material is preferable to improve the MR compatibility.
M13-264:

J. W. T. Heemskerk1,2, M. C. Goorden2, M. A. N. Korevaar1,2, E. Van der Kolk2, R. Kreuger2, P. Dorenbos2, F. J. Beekman1,2,3 1 Image Sciences Institute, University Medical Center Utrecht., utrecht, netherlands 2 Radiation detection and medical imaging, Technical University Delft, delft, netherlands 3 Molecular Imaging Laboratories, utrecht, netherlands Charge Coupled Devices (CCDs) coupled to scintillation crystals can be used for high resolution imaging of X-rays and gammarays. In contrast to photomultiplier tubes (PMTs), which are prominent in scintillation gamma detection, CCDs are integrating devices. This enables, contrary to traditional PMT-based gamma cameras, the application of a whole new class of scintillators with longer decay times. For our recently developed gamma cameras based on Electron Multiplying CCDs (EM-CCDs) which are read out at 50 frames/s, scintillators with decay times up to several milliseconds can be used. In this paper we investigate the performance of our CCD-based gamma camera with three different scintillators; two relatively slow scintillators, Gd2O2S:Tb,Ce and CdWO4, are compared to the often used CsI:Tl. These new scintillators were selected because of their excellent light yield and/or high density. The thickness of different crystals was tuned to have equal capture efficiency (66%) at 140 keV. We found that compared to CsI:Tl, use of Gd2O2S:Tb,Ce improves the intrinsic spatial resolution by approx. 24 % (148 m compared to 194 m); use of CdWO4 improves the spatial resolution even by 36 % (124 m). However, both these crystals have an energy resolution that is worse than that of CsI:Tl. We believe that this is due to either the lower light yield (in the case of CdWO4) or the ceramic nature of the scintillator (for Gd2O2S:Tb,Ce). Based on these first results we conclude that the application of scintillators materials that were not considered previously because of their relatively long decay time offers new possibilities for improving CCD-based gamma cameras.
We would like to thank P. Rodnyi and E. Gorokhova for kindly providing the Gd2O2S:Tb,Ce sample. M13-267:

Comparison of Scintillators for an EM-CCD-Based Gamma Camera

The "X'tal Cube" PET Detector: 3D Scintillation Photon Detection by a 3D Crystal Array Using

Y. Yazaki1,2, H. Murayama2, N. Inadama2, H. Osada1,2, F. Nishikido2, K. Shibuya3, T. Yamaya2, E. Yoshida2, M. Suga4, T. Moriya5, M. Watanabe5, T. Yamashita5, H. Kawai1 1 Graduate School of Sciences, Chiba University, Chiba, Japan 2 Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan 3 Graduate School of Arts and Sciences, Tokyo University, Tokyo, Japan 4 Graduate School of Engineering, Chiba University, Chiba, Japan 5 Central Reserch Lab., Hamamatsu Photonics K.K., Shizuoka, Japan ABSTRACT- We have proposed a depth of interaction (DOI) PET detector named X'tal cube, in which a number of Multi-Pixel Photon Counters (MPPCs) are coupled on various positions of six surfaces of a segmented scintillation crystal block. There are no reflectors within the block, and the areas among MPPCs on the surface are covered with reflector. Each MPPC is thin and light solid-state photo-detector so that the crystal block can be closely placed on the PET detector ring. To study the characteristics of the Xtal cube, we constructed a crystal block consisting of six layers of a 6 x 6 segmented crystal array with Lu2xGd2(1-x)SiO5 (LGSO) crystals. Each crystal is 3.0 x 3.0 x 3.0 mm3. We examined crystal identification performance for different MPPC arrangements on the block surfaces, where we used 3-dimensional (3D) Anger-type position calculation. The preliminary experiments showed the possibility of a 3D detector having isotropic resolutions for PET.
M13-270:

MPPCs

Proposal of a Threshold Type Cherenkov PET Detector Without Pulse Height Measurements

T. Mitsuhashi, H. Kawai, H. Nakayama, Y. Saito, A. Suzuki, M. Tabata, K. Tomioka physics, graduate school of science.Chiba university, Chiba,Chiba, Japan We have developed the new production methods of silica aerogel with the refractive indices between1.15 and 1.25. Since the velocity of the photoelectric electron from 511keV photon is 0.866c and the maximum velocity of the Compton electron is 0.8c, 420

only photoelectric electrons occur the Cherenkov light inside the radiator with n~1.2. We are constructing the simple prototype PET detectors using these silica aerogels. Since there are no efficiency for the Compton events, the pulse height measurements are not need. The data acquisition system becomes very simple.
M13-273:

Possibility Analysis of Si-PM Based DOI Detector Using Pulse Shape Analysis for PET

S. Yamamoto Kobe City College of Technology, Kobe, Japan Geiger-mode avalanche photo-diode (silicon photomultiplier: Si-PM) was tested whether it can be used for depth-of-interaction (DOI) detector based on decay time differences of the scintillator using pulse shape analysis. Hamamatsu Si-PM (MPPC: S10362-33-025C was used for the experiments which has 3mm x 3mm sensitive area. The Si-PM was optically coupled to two GSO crystals with difference Ce concentration, 1.5 mol% (decay time of 35ns) and 0.5 mol% (~50ns), optically coupled in DOI direction. Also the Si-PM was tested with two LGSO with different Ce concentration, 0.025 mol% (33ns) and 0.75 mol% (43ns). Pulse shape and energy spectra for Cs-137 gamma photons (662 keV) were measured digitally using dual integration method. In both combinations of scintillators, pulse shape spectra were successfully obtained with Si-PM. Si-PM optically coupled with GSO crystals with different Ce concentration showed good separation in the pulse shape spectrum; peak-to-valley (P/V) ratio was 3.6. Si-PM with LGSO crystals showed moderate separation; P/V ratio was 1.25. With these results, it is confirmed that Si-PM can be used for DOI detectors based on the decay time difference of scintillators using pulse shape analysis.
M13-276:

H.-C. Liang1,2, M.-L. Jan1, W.-C. Lin1, J.-L. Su2, L.-T. Huang1 1 Physics Division, Institute of Nuclear Energy Research, Longtan, Taiwan ROC 2 Department of Biomedical Engineering, Chung-Yuan Christian University, Taoyuan, Taiwan ROC

Improving Precision of Position Estimation by Photopeak-Index Method for a DOI Imaging Detector

In the application of a positron breast planar imager, the detector separation distance is suggested to be as close as possible to maximize the acceptance angle and also to improve the sensitivity. However, this causes the parallax error to become more crucial and degrades the image quality. A scanner which offers the depth-of-interaction (DOI) information of the incident gamma events can abate parallax error to gain image quality improvements. In our earlier study, a design of a position-of-interaction imaging detector, which can provide 3D information of the event positions, was presented. This long stripe crystal with ends-read design can not only reduce the parallax error but also requires less photon detectors in forming a larger imaging area. However due to scintillation photon attenuation and poor energy resolution, preliminary results showed poor precision for the estimated event positions along the long axis of crystals. In this study, a photopeak-index method is introduced to improve the accuracy and precision of gamma-ray-interaction position estimations of a DOI detector with long crystal.
M13-279:

R. Pani1, P. Bennati2, R. Pellegrini1, M. Cinti1, S. Nourbakhsh3, P. Pani3, V. Orsolini Cencelli4, F. de Notaristefani4, F. Navarria5, S. Lo Meo5, A. Perrotta5, N. Lanconelli5, G. Moschini6, P. Boccaccio6, R. Scafe'7 1 Dept. of Exp. Medicine, Sapienza University of Rome and INFN, Rome, Italy 2 EDEMOM PhD shool, Roma Tre University of Rome and INFN, Rome, Italy 3 Dept. of Physics, Sapienza University of Rome and INFN, Rome, Italy 4 Dept. of Electronic Engineering, Roma Tre University of Rome and INFN, Rome, Italy 5 Dept. of Physics, University of Bologne and INFN, Bologne, Italy 6 Legnaro National Laboratory, INFN LNL, Padua, Italy 7 Casaccia, ENEA and INFN, Rome, Italy Knowledge of the Depth-Of-Interaction (DOI) in detector is crucial for small ring diameter PET scanner but it is also very important for single photon emission. In particular, for applications where very high spatial resolution is required and position distortions caused by slant collimators can strongly affect the final response of the 3D reconstructed image. Continuous scintillation crystals are, in principle, the most suitable for DOI determination based on the measurement of scintillation light width of cones corresponding to each gamma ray interaction. In this work, we propose an analysis based on a Monte Carlo GEANT4 studies and on some experimental measurements at 140keV of a LaBr3:Ce continuous crystal coupled to a Flat Panel PMT Hamamatsu H8500. The basic idea is that, the high light output of LaBr3:Ce can reduce statistical uncertainties related to the light distribution width determination. In addition, It could be crucial of a more precise measurement of scintillation light distribution by the 8x8 anode array of MAPMT. Monte Carlo modelling scintillation light transport of a LaBr3:Ce crystal allowed to verify the possibility of the depth interaction determination at 140 keV photon energy based on the scintillation light cone width discrimination. A DOI spatial resolution of about 2.5-2.8 mm resulted by MC analysis if position and Standard Deviation (SD) are calculated from the raising to the 2nd power truncated light distribution. In addition, from the selection of MC events on SD window of scintillation light widths resulted an improvement of spatial resolution of about 20%. It has been confirmed by experimental data involving 1.0 mm spatial resolution. According to Monte Carlo results we can assume that such 421

Investigation of Depth Dependent Response of Continuous LaBr3:Ce Scintillation Crystals

images are produced at deeper distance from the crystal entrance. DOI spatial resolution can be further improved minimizing optical window and by the use of Ultra High Quantum Efficiency photocathode.
M13-282:

N. D. Fitzgerald1, F. O'Sullivan1, G. Newman2 1 Statistics, University College Cork, Cork, Ireland 2 Neurology, Albert Einstein Medical Research Institute, Philadelphia, USA

Cerebral Perfusion Maps from Dynamic Contrast MRI Data Utilizing Rician Statistics.

Bolus tracking of contrast agent with MRI is a well established technique for measurement of local cerebral hemodynamic parameters flow, mean transit time (MTT) and volume. When performed on a voxel-by-voxel basis, it allows development of hemodynamic parameter maps useful for assessment of ischemic damage following stroke and tumor characterization in cancer. The analysis of the acquired dynamic data requires the use of deconvolution to reconstruct the residue function (R) of the contrast agent. Measurement of the tissue time course and the arterial input function are obtained by T2 or T2* weighted sequences. Reconstruction of R provides estimates of flow, volume and mean transit time. The raw MRI scan signal intensity is well approximated by Rician statistics. The standard approach to estimation involves logarithmic transformation and least squares deconvolution. At low signal to noise this approach is not be efficient and as an alternative this work adopts an iterative reweighted non-linear least squares (IRWNLLS) algorithm to incorporate Rician statistics, impose constraints on the residue function and optimize for tracer arrival delay. The algorithm is implemented on a voxel-by-voxel basis and cerebral maps for the hemodynamic parameters flow, MTT and volume are presented. In addition, an automatic segmentation technique which takes into account both spatial and temporal variation is presented. This segmentation technique is shape driven, choosing only voxels that correlate highly with a well-known arterial input function template.
Supported by the Irish Health Research Board. M13-285:

P. Guerra1, J. Aguirre2, J. E. Ortuno1, M. J. Ledesma3,1, J. J. Vaquero2, M. Desco2, A. Santos3,1 1 Bioengineering, Biomedical Research Center in Bioengineering, Biomaterials and Nanomedicine, Zaragoza, Spain 2 Medical Imaging Lab, Hospital General Universitaria Gregorio Maraon, Madrid, Spain 3 Biomedical Image Technologies, Universidad Politecnica de Madrid, Madrid, Spain Optical tomography relies on reasonable modeling of light propagation in scattering tissue, being the radiative transfer equation (RTE) and the diffusion approximation, being the latter the most common approach as it admits analytical solution for certain geometries. As an alternative to complex mathematical models, computer simulation of photon transport is a versatile and flexible option. This work presents a multipurpose Monte Carlo (MC) code for photon transport in voxellized biological tissues. The simulator improves existing codes with the support of multiple laser sources and detectors, photon polarization, timeresolved photon tracking as well as fluorophore absorption and reemission at longer wavelenghts. The MC code has been profiled and variance reduction techniques have been introduced in order to speed up simulations; in particular improving efficiency when fluorochromes are present in the simulation. These capabilities enable the use of the simulator to study different optical techniques such as early photon tomography (EPT) or diffuse optical tomography (DOT). The simulator has been validated with results reported by previous authors and with our own analytical results from the diffuse approximation. Further experimental validation of the simulator with optical phantoms is in progress. The ultimate goal is to evaluate the benefit in the reconstructed volumes of incorporating in the system matrix a-priori information obtained with other modalities, such as CT or MR, using the simulator to solve the forward problem, instead of the current approach where the same system matrix, derived from a uniform tissue slab, is used in every reconstruction.
This work is supported by the Spanish Ministry of Science and Innovation under grant number TEC2008-06715-C02-02. M13-288:

Multipurpose Monte Carlo Simulator for Photon Transport in Turbid Media

T. Ichihara1, R. T. George2, J. A. C. Lima2, A. C. Lardo2 1 Dept. of Radiological Technology, Fujita Health University School of Health Science, Toyoake, Aichi 470-1192, Japan 2 Div. of Cardiology, Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA

Quantitative Analysis of First-Pass Contrast-Enhanced Myocardial Perfusion Multidetector CT Using a Patlak Plot Method and Extraction Fraction Correction During Adenosine Stress

PURPOSE: The purpose of this study was to develop a quantitative method for myocardial blood flow (MBF) measurement that can be used to derive accurate myocardial perfusion measurements from dynamic multidetector computed tomography (MDCT) images by using a compartment model for calculating the first-order transfer constant (K1) with correction for the capillary transit extraction fraction (E). MATERIALS AND METHODS: Six canine models of left anterior descending (LAD) artery stenosis were prepared and 422

underwent first-pass contrast-enhanced MDCT (Aquilion 64, Toshiba Medical Systems Corporation, Otawara, Japan) perfusion imaging during adenosine infusion (0.14-0.21 mg/kg/min). K1, which is the first-order transfer constant from left ventricular (LV) blood to myocardium, was measured using the Patlak plot method applied to time-attenuation curve data of the LV blood pool and myocardium. The results were compared against microsphere MBF measurements, and the extraction fraction of contrast agent was calculated. K1 is related to the regional MBF as K1=EF, E=(1-exp(-PS/F)), where PS is the permeabilitysurface area product and F is myocardial flow (in units of ml/g/min). Based on the above relationship, a look-up table from K1 to MBF can be generated and Patlak plot-derived K1 values can be converted to calculated MBF. RESULTS: The calculated MBF and microsphere MBF showed a strong linear association. The extraction fraction for dogs as a function of flow (F) was E=(1-exp(-(0.253F+0.7871)/F)). CONCLUSION: Regional MBF can be measured accurately using the Patlak plot method based on a compartment model and look-up table with extraction fraction correction from K1 to MBF.
M13-291:

J. Zeintl1, A. H. Vija2, A. Yahil3, J. Hornegger1, T. Kuwert4 1 Chair of Pattern Recognition, University of Erlangen, Erlangen, Germany 2 Molecular Imaging, Siemens Medical Solutions USA, Inc., Hoffman Estates, IL, USA 3 Image Recon LLC, Stony Brook, NY, USA 4 Clinic of Nuclear Medicine, University of Erlangen, Erlangen, Germany

Quantitative Accuracy of Slow-Rotating Dynamic SPECT Imaging

We investigate biases of time-activity measurements relevant for quantitative dynamic SPECT/CT imaging when slow-rotating dual-headed gamma camera systems, like the Symbia series in combination with OSEM3D (Flash3D) with scatter and attenuation correction are used. The goal is to establish a baseline and show potentials and limitations of clinical dual-headed gamma camera systems for quantitative 3D imaging of dynamic processes. We use simulations of a SPECT/CT system to estimate absolute quantitation errors in time-activity measurements. We systematically assess dependencies of these errors on signal to noise ratio, sampling frequencies and activity change rates using time-activity profiles such as MAG-3 renal activity function. In addition, a physical phantom is developed to measure dynamic processes on a clinical SPECT/CT system. The phantom consists of a cylindrical chamber placed in a large cylinder phantom and connected to a smart peristaltic pump. SPECT/CT acquisitions are conducted varying the flow rates of the pump and the sampling rates of the SPECT system. Absolute activity concentrations are calculated by cross calibrating the imaging system with the well counter and using correction factors derived from simulations. Results from simulations show underestimation of activity concentrations between 20.449.88% and 20.613.29% for 7.5 and 120 seconds per time frame. Phantom experiments show estimation errors of -9.7314.40% with an SPECT imaging speed of 30 seconds per timeframe, when aforementioned calibrations are applied. Conclusion: With slowrotating dual headed SPECT/CT systems like the Symbia series accurate time-activity estimation is possible in phantoms. Absolute time-activity estimations are robust against changes in signal to noise ratio, sampling frequency, and activity change rate provided that object boundaries can be determined precisely using e.g. co-registered CT images.
M13-294: Background Correction Strategies for List-Mode Submillimetre 3D Image Reconstruction Algorithm Applied to the High Resolution quad-HIDAC PET System

L. Ortega Mynez1, T. Kosters2, H. J. Ochoa1, F. Wubbeling2, K. Sachfers2 1 Departamento de Electrica y Computacion, UACJ, Cd. Jurez, Chih., Mexico 2 Department of Mathematics and Computer Science, University of Munster, Munster, Germany

Two novel methods for the inclusion of the background estimation within the reconstruction process are presented. The additive and subtractive models for background correction are very well known procedures included within the reconstruction process. In a previous work, estimation of list-mode scatter events was introduced. One of the methods proposed includes the background estimation using an additive model, in which the estimation is performed by adding the result of the forward step over a background image previously reconstructed and the forward step over the estimated image. The background image was reconstructed using the scatter and random list mode files. The second method proposed, changes the model 1/q, typically used in event by event reconstruction, to be (1-(m/b))/q; in which the numerator is the event probability of being a true or a background event. A F18 small mouse NEMA phantom is used to assess the methods along with a clinical list mode data from a small mouse scan, both sources were acquired from the quad-HIDAC PET scanner. Comparison between both models is presented demonstrating the performance of these methods.

423

M13-297:

K. L. Perez1,2, S. J. Cutler2,3, P. Madhav2,3, M. P. Tornai1,2,3 1 Medical Physics Graduate Program, Duke University, Durham, NC, USA 2 Deparment of Radiology, Duke University, Durham, NC, USA 3 Department of Biomedical Engineering, Duke University, Durham, NC, USA

Towards Quantification of Dedicated Breast SPECT Using Non-Traditional Acquisition Trajectories

Quantification of radiotracer uptake in lesions gives valuable information to physicians in deciding patient care or determining treatment efficacy. While more routinely done in PET, quantification is still less common for routine SPECT imaging. Physical processes and, with systems capable of 3D trajectories, reconstruction artifacts can yield an incorrect absolute activity of the tracer. Differences in the obtained activity value from each trajectory are investigated to determine if the acquisition trajectory can be used for quantification. For these experiments, a fillable 600mL breast and 2.3mL lesion phantoms containing aqueous 99mTc pertechnetate were imaged with a dedicated dual-modality SPECT-CT scanner. SPECT images were collected using a compact CZT camera with various 3D acquisitions including vertical (VAOR), 45 degree tilted (TPB), and complex sinusoidal (PROJSINE) trajectories. Collimator and detection efficiencies of the SPECT camera were incorporated into the OSEM iterative reconstruction. Attenuation correction was implemented using a uniform attenuation coefficient matrix, but in the next steps will be done with scaled volumetric CT images. In addition, a Compton Window scatter correction method was applied with an empirically determined k value of 0.3 and a scatter window ranging from 113 to 133 keV, abutting and below the 8% photopeak window. A line source was used to determine a scaling factor of 0.07 to correct reconstructed data to activity. The resulting calculated lesion activity in the image was found to be within 20.5% (+/- 9.9%) of the dose calibrator measured activity value. The scaled breast background values were double the dose calibrator measured activity value indicating that further investigation of the linearity of the method is needed. As far as we are aware, this is the first time that absolute quantification has been applied to SPECT data acquired with non-traditional, non-circular trajectories.
This work is supported by W81XWH-08-1-0192, W81XWH-06-1-0765, W81XWH-06-1-0761 and NIH R01-CA096821. M13-300:

J.-S. Lee1, K.-H. Su1, W.-P. S. Tam1, R.-S. Liu2, S.-J. Wang2, J.-C. Chen1,3 1 Department of Biomedical Imaging & Radiological Sciences, National Yang-Ming University, Department of Biomedical Imaging & Radiological Sciences, National Yang-Ming University, No. 155, Taipei, Taiwan 2 National PET/Cyclotron Center, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Road, Taipei, Taiwan 3 Department of Education and Research, Taipei City Hospital, 145 Zheng Zhou Road, Taipei, Taiwan

Measurement of Input Function in Rodents Using Wavelet Packets Based Sub-Band Decomposition Independent Component Analysis.

Purpose: The purpose of this study was to acquire an input function or whole-blood time-activity-curve (wTAC) by wavelet packets based sub-band decomposition independent component analysis (WP-based SBICA). Methods: There were three types of materials in the experiments: (1) simulated dynamic rat images without motion artifact, (2) simulated dynamic rat images with motion artifact, (3) in four real Sprague-Dawley rats, arterial blood samples were obtained at 0, 8, 15, 30, 60, 120, 180, 300, 450, 600, 900, 1500, 2400s following the injection of 18F-FDG. The wTACs were determined in a well counter and they were treated as the reference values. Each wTAC was estimated by WP-based SBICA. We also used FastICA estimated wTAC to compare with the results of our method using the normalized root mean square error (NRMSE) and error of area under curve (EAUC). Results: In the simulated study without motion artifact, the NRMSEs for the two methods were 0.1169 and 0.1610 for WP-based SBICA and FastICA, respectively. The EAUCs were 0.0690 and 0.0961 for WP-based SBICA and FastICA, respectively. In the simulated study with motion artifact, the NRMSEs for the two methods were 0.3259 and 7.4742 for WP-based SBICA and FastICA, respectively. The EAUCs were 0.2948 and 6.6889 for WP-based SBICA and FastICA, respectively. In the realistic rat study, the averaged NRMSEs for the two methods were 0.3942 and 0.5325 for WPICA and FastICA, respectively. The averaged EAUCs were 0.2584 and 0.6456 for WP-based SBICA and FastICA, respectively. Conclusion: Our results show that the accuracy of estimation of the input function by using WP-based SBICA outperforms that obtained by FastICA.
M13-303:

Quantitative Analysis of the Spine Using Curve Modeling

O. Hay, I. Hershkovitz Anatomy, Tel Aviv University, Tel Aviv, Israel Spine curvature and posture are important to sustain healthy back. We propose new method for analyzing spine curvature in 3D, using CT imaging. The proposed diagnosis involves new concept of spinal curve assessment using the spinal canal centerline, novel segmentation method, and applying modeling techniques to spine curvature as means of establishing a baseline for normal and pathological configuration of the spine. The spine curve modeling is based on the spinal canal and manifest many improvements over existing methods (which are based on the vertebral bodies, usually 2D in nature, and in most cases concerns only a segment of the spine). The method presented is 3D (volumetric), covers both global and local deformations, overcome 424

small bone deformations, and requires minimal user interaction in clinical setup. spinal canal extraction is carried out using 3D deformable objects to overcome the difficulty of weak or no edges with the intervertebral disks and lumbar nerves. The spinal canal segmentation is based on mesh deformation using both edge and texture information (as external forces), and strong shape constraint (as internal forces). Employing texture edge forces based on JSEG, alongside gray-value edges, overcomes the weak edges, while using shape constrains overcomes incorrect segmentation where no edge could be found. Mesh adaptation is optimized with on-the-fly force calculation, together with adjustable mesh resolution, to improve performance and quality of segmentation. Initial clinical results on 52 individuals clearly show that the presented method can be reliably utilized for identifying and quantifying spine mal posture (e.g. scoliosis). The results also suggest that there is a correlation of posture problems and low back pain. As such, it may greatly assist orthopedic surgeons of the spine and provide new insight into nonspecific low back pain diagnosis.
The authors would like to thanks Dr. Natan Peled, Dr. Robest Shechner, and Matti Schnapp for their kind support in acquiring the data and establish validation ground truth, and Prof. Ehud Rivlin M13-306:

Evaluation of Miscellaneous ROI Placements on 18F-FDG PET/CT Quantification: 1D to 4D

J. Zhang, N. C. Hall, R. Layman, M. V. Knopp Radiology, The Ohio State University, Columbus, Ohio, USA Purpose: To evaluate the impact of miscellaneous ROI tools on quantitative assessment of Standardized Uptake Value (SUV) and Metabolic Active Tumor Volume (matV) in F18-FDG PET/CT. Methods: Miscellaneous ROI tools from 1D to 3D were enabled with and without rotation functionality using vendor workstations (Siemens Leonardo, GE AW, Philips EBW, MIMVista) and inhouse developed software. An advanced 4D isosurface fusion tool was developed for ROI effectiveness validation. 50 target lesions of various sizes, shapes, heterogeneity and anatomical region were used for evaluation and quantification as well as advantages and disadvantages of the tools compared. Results: To determine tumor SUVmax and matV, 3D VOI approaches show the advantages of high efficiency and convenience. Due to that tumor diversity and heterogeneousness, the exterior outlining based tools with regular geometry may cover unexpected regions into the target ROI. The interior growing based Region-grow approaches are helpful in dealing with lesions with irregular shapes however regions may over-spread to non-target areas. The developed ROI tools with rotation functionality revealed a significant improvement (p>0.01) for accuracy of ROI placement. 3D VOI approaches can lead to false representations of the true SUV and matV by including additional tumors/hot areas that might not be seen in current MPR view approach, while the developed 4D matV fusion tool gives a clear map of the tumor homogeneity/heterogeneity and can easily detect the true findings to further validate the effectiveness of placed VOIs. Conclusions: Tumor diversity and heterogeneousness gives ROI placement a very important role in F18-FDG-PET/CT quantitative assessment. Most current ROI tools have limitations when dealing with these complicated issues and potentially leading inaccurate quantification. We found that an advanced VOI approaches with rotation and isocontour integration markedly improves the accuracy and robustness.
This study was supported by the Wright Center of Innovation in Biomedical Molecular Imaging (ODOD AGMT TECH 03-051 ) and Imaging Response Assessment Teams (NIH/NCI P30; CA16058). M13-309:

M. Costin1,2, D. Lazaro Ponthus1, S. Legoupil1, P. Duvauchelle2, V. Kaftandjian2 1 LIST, SSTM/LID, Atomic Energy Commission, Saclay, France 2 CNDRI, INSA, Lyon, France

A Multiresolution Image Reconstruction Method in X-Ray MicroCT

We propose a method for multiresolution image reconstruction in X-ray micro computed tomography (CT). The method uses a zoom-in CT setup and it targets non-destructive testing (NDT) or medical applications, like material characterization or small animal imaging studies, respectively. Our main purpose is to recover an overall image of the sample with a coarse resolution, and a fine resolution for a region-of-interest (ROI). In order to allow general settings, the ROI can be freely chosen inside the object, but the sample must be placed in both scanning positions with the center of the ROI on the beam axis. We acquire two sets of projections, related by a zoom factor zr which is computed as the ratio between the magnification factors at the two positions. The projections are combined into a single enlarged set and decomposed line wise with a discrete wavelet transform (DWT). Two subimages are obtained by backprojection, which serve for the final image recovery. By comparing our method to the standard FBP reconstruction, we obtain very small values of the mean squared error and a similar robustness to noise, with two important advantages. First, the number of acquired samples can be reduced with a factor up to zr-1, therefore the exposure and acquisition times are reduced, and secondly the reconstruction speed is improved with an important factor. Moreover, with our method the file storage requirements can be lowered when only the subimages are saved.

425

M13-312:

Geometric calibration and image reconstruction of a stationary MR-compatible MicroSPECT

camera

J. Xu1, S. Chen1, J. Yu1, D. Meier2, D. Wagenaar2, B. E. Patt2, B. M. W. Tsui1 1 Radiology/Medical Imaging Physics, Johns Hopkins University, Baltimore, MD, USA 2 Gamma-Medica Ideas, Inc, Northridge, CA, USA

We developed a geometric calibration method for a stationary MR-compatible MicroSPECT camera for simultaneous SPECT/MR imaging of small animals. The instrumentation details are presented in [1]. The parameters to be calibrated are the center positions of 24 CZT detectors and the 24 matching pinholes, in total 6 x 24 = 144 unknowns. In its current configuration of the SPECT camera, the calibration point sources are often truncated or overlapped in the projection image that renders existing methods inapplicable. We propose a 2-step method to overcome this difficulty. First, a Poisson likelihood method is developed to estimate the centroids of the point source projections. The point spread function (PSF) of the pinholes is modeled as a 2D Gaussian distribution, where the mean is the unknown centroids and the variance the shape of the PSF. The overall distribution from multiple point sources can be described by a Gaussian mixture model (GMM). The acquired photon count in a projection bin follows a Poisson distribution with mean related to the unknown GMM parameters. The ML solution of the GMM parameters given the point source projection image can be obtained by the expectation maximization algorithm applied to grouped and truncated data. The point source centroids estimates and their variances are fed to the second step of a weighted least squares estimation of the 144 geometric parameters. Experimental data were acquired from a prototype MicroSPECT camera using 4 non-coplanar point sources for geometric calibration. The calibrated parameters reduced the projection matching error compared with the default values. We present reconstructed images of phantom experiments using the calibrated parameter values. The proposed 2-step method is effective in dealing with point source projection truncation, overlap, and low count statistics, and it can be used to provide accurate geometric parameters for artifacts-free image reconstruction of the stationary MicroSPECT camera.
[1] Dirk Meier, Douglas J. Wagenaar, Gunnar Mhlum, Bjrn Sundal, Bradley E. Patt, Mark Hamamura, Werner W. Roeck, Seung- Hoon Ha, Orhan Nalcioglu, Fellow, Si Chen, Jingyan Xu, and Benjamin M.W. Tsui, "A Stationary SPECT Camera for Simultaneous Small Animal SPECT/MRI", Proceeding of the IEEE, Joint Nucl. Sci. Symp. and Med. Imaging Conf. 2009. M13-315:

Count-Rate Dependent Component-Based 3D PET Normalization Using Singles Events

L. V. Romanov, H. A. Kudrolli, P. Kulinich, J. D. Nevin, T. S. Toole, W. A. Worstell, C. Worth PhotoDetection Systems, Inc., Boxboro, MA, US We have implemented a component-based normalization method for 3D PET scanning which makes use of a prescaled sampling of singles events. These prescaled singles events are included in the list-mode emission event datastream generated by our PET scanner, with the list containing both measured energy and crystal-of-interaction information for each single. Our normalization method is capable of correcting for run-specific (e.g. count-rate dependent) changes in detection efficiency between acquisitions or even within a single acquisition. The factorized component-based approach uses singles events to measure both the triggerlevel efficiency for gammas contributing to coincidence pairs, and the small variation in crystal-selection efficiency within block detectors at higher trigger rates due to rate-dependent effects. Ancillary measurements utilizing calibration sources which generate coincident gamma pairs are used to estimate geometric correction factors for coincidences, where these correction factors share the transaxial and axial block structure symmetries of the detector geometry. These and additional coincidence normalization factors are separately derived for transaxial and axial geometric effects, in factorized form. Live fraction effects for coincidences are modeled based on singles and coincidence rates in a manner consistent with the underlying segmented coincidence event data acquisition system, and are incorporated in the normalization process. An extension to this methodology for continuous bed motion makes use of a virtual detector concept but is otherwise common to the static case, and is implemented in our large-bore whole-body PET scanner. Finally, energy-based event weighting, which maximizes image signal-to-noise in the presence of scatter events, is supported in a consistent way within our weighted-event normalization process.
M13-318:

Variance-Reduced Randoms Correction for 3D PET Using Prescaled Singles Events

H. A. Kudrolli, P. Kulinich, J. D. Nevin, T. S. Toole, W. A. Worstell, C. Worth PhotoDetection Systems, Inc., Boxboro, MA, US We have developed and tested a variance-reduced randoms correction method which avoids the need for delayed coincidence windows, accurately estimates both the spatial and spectral distributions of randoms events with low variance, and provides redundant methods for confirmation of randoms subtraction accuracy. By estimating the energy and timing spectral properties of randoms events and subtracting these from the measured coincident event spectra, it provides quantitatively accurate inputs for event-weighting methods which maximize resultant image signal-to-noise. Our method is based on collection of prescaled singles events, singles counter rate measurements for each block within the detector, and a combination of data acquisition live fraction measurements and estimations based on measured rates and known coincidence processor live fraction characteristics. Live 426

fraction estimates both for randoms and for true+scatter events are an integral part of the randoms correction process. We have implemented this method both for static acquisitions and for the case of continuous bed motion, where the latter makes use of a virtual detector construct into which data is binned based on measured bed positions. Time-varying randoms estimation within long acquisitions of rapidly decaying radioisotopes is supported by measuring singles rates and single event distributions over multiple sub-intervals within a given acquisition, and integrating the resultant randoms estimates. Results are presented for measurements from decaying activity distributions which confirm linear dependence of the resultant sinograms on activity after live fraction correction, with stable sinogram spatial distributions across varying activities.
M13-321:

F. P. Jansen1, D. Beque2, H. Qian1, G. Bal1 1 GE Research, Niskayuna, NY, USA 2 GE Global Research, Munich, Germany

Efficient Algorithm for Modeling Keel-Edge Pinhole Response

Multi-pinhole collimators allow projection of multiple views of a subject at the same time, which potentially improves the resolution and sensitivity of a gamma camera over a limited field of view. One method to reduce the problem of overlapping projections is the use of a narrow acceptance (cone) angle for the pinhole, but this has the drawback of a very abrupt change in sensitivity profile, which can generate artifacts in the reconstruction. By contrast, a keel edge pinhole contains a cylindrical section, which gives it a smoother roll-off of sensitivity, and is easier to manufacture accurately since the diameter is defined by a conventional drill rather than by accurately counter-drilled cones. A drawback is the complex shape of the point spread function (PSF) and sensitivity profile, which needs to be determined accurately for faithful reconstruction, and which needs to take into account mechanical errors in the imaging system (calibration). We describe a very efficient method for accurately calculating the shift-variant point spread function of a gamma camera with multiple keel-edge pinholes. The resulting PSF is computed for every voxel/pinhole/gantry-angle combination and stored in a highly compact format that permits extremely fast reconstruction using conventional hardware. A test case with a 32x32x32 voxel image matrix, 17 pinholes and 60 view angles (1020 projections) with 128x92 pixels resulted in a system matrix of 1.4 GB with a reconstruction speed of 2.2 sec / iteration (EM) on a single 3 GHz CPU. We have validated the calculations with point source data obtained on a gamma camera fitted with a multi-pinhole collimator, and have confirmed both the sensitivity and PSF shape. At the conference we will compare reconstructed phantom images obtained with this method against traditional algorithms.
M13-324:

T. Koesters1, M. Fieseler1, M. Dawood1, F. Buether2, K. P. Schaefers1 1 European Institute of Molecular Imaging, University of Muenster, Muenster, Germany 2 Department of Nuclear Medicine, University of Muenster, Muenster, Germany

A Nested EM-Based Motion Detection And Correction Reconstruction Algorithm for PET Imaging

Intrinsic motion is a source of image degradation in quantitative PET leading to blurring artifacts. Due to the long acquisition time of PET the patient has to breath resulting in organ movement. In order to get a quantitative image the reconstruction algorithm has to consider this movement. Existing correction methods can be classified in image based and data based approaches depending on the space where the correction is performed. For both approaches it is necessary to reconstruct the single gates prior to the motion estimation leading to a high number of reconstructions on the one hand and motion estimation on low statistic images on the other hand. Recently, a method has been proposed to overcome these problems by combining motion estimation and motion correction in one framework. In this work we propose to combine the two established methods, EM for image reconstruction and Optical Flow for motion estimation, in a nested algorithm. Instead of reconstructing single gates and deriving the motion estimation afterwards a 4D image containing the data of all gates is used. Once a reconstruction step of the 4D image is done the motion between the different gates is estimated. The derived motion vectors are then included in the next reconstruction step. Therefore all gates are reconstructed using the full statistic and the motion vectors are updated iteratively leading to a smoother motion estimation. Finally, the resulting 4D image contains all gates reconstructed with all acquired data.
M13-327:

A Hybrid Algorithm for Randoms Variance Reduction

C. C. Watson Siemens Healthcare Molecular Imaging, Knoxville, TN, USA Random coincidences are a significant source of noise in positron emission tomography; consequently, there is considerable interest in developing algorithms that reduce the data variance due to randoms correction. Two approaches are commonly used: estimation of the randoms rates from measured singles rates in the detector pairs, or estimation from sums over multiple lines of response in the delayed coincidence sinogram. We describe a new technique that combines these two approaches. It has the advantage of using all available randoms and singles data for improved precision relative to block sum techniques, and improved accuracy relative to fan sum algorithms. Results are shown for a Siemens Biograph TruePoint PET/CT.

427

M13-330:

M. Iatrou1, R. Manjeshwar1, S. Wollenweber2, S. Ross2, C. Stearns2 1 ImT, General Electric Global Research Center, Niskayuna, NY, USA 2 MI, General Electric Healthcare Technologies, Milwaukee, WI, USA

Out-of-Field Scatter Estimation in 3D Whole Body PET

Many implementations of model based scatter correction (MBSC) are based on the single scatter simulation (SSS) formulation within the scan field-of-view (FOV). A fully 3D approach that models both the axial and trans-axial scatter components can accurately model scatter from hot regions in neighboring slices and outside the scan FOV resulting in greater quantitative accuracy. Herein we discuss how to incorporate the estimation of out-of-field scatter in fully 3D MBSC. MBSC with Out-OfField (OOF) scatter estimation requires activity and attenuation data from an axial FOV that extends beyond the scanned FOV. We show through Monte Carlo simulations the required extent of the emission and attenuation data for OOF scatter estimation, we discuss the required data collection latency in estimating OOF scatter, and we present simulation and measurement results. The results demonstrate that MBSC and MBSC with OOF scatter estimation perform equally well the further we are from a region with high activity. In contrast, for slices within the imaged FOV that are closer to the frame with high activity, MBSC with OOF scatter estimation estimates scatter more accurately.
M13-333:

A. B. Mann1, S. Paul1, A. Tapfer2, V. C. Spanoudaki3, S. I. Ziegler2 1 Physics Department E18, TU Muenchen, Garching, Germany 2 Klinikum rechts der Isar, Nuklearmedizinische Klinik und Poliklinik, TU Muenchen, Munich, Germany 3 Molecular Imaging Instrumentation Laboratory, Stanford University, Palo Alto, USA

A Computing Efficient PET Time Calibration Method Based on Pseudoinverse Matrices

The time calibration of PET detector systems is an important task to achieve optimal system performance and time resolution. As detector channel delays may vary with environmental parameters and electronic settings, fast and efficient computing methods are required to allow repeated calibration during the operation. We describe a time calibration method determining a least squares solution via pseudoinverse matrices, with on-the-fly generation of the pseudoinverse and thus avoiding calculation of the singular value decomposition. Hence, computing times are in the range of 1--8s for a system with 1000 detector channels, with only negligible memory requirements for the calculation. Additionally, by using segmented or "fan-sum" LOR geometries, the statistics in the initial LOR time difference histograms can be improved, resulting in more precise input values for the matrix equation.
M13-336:

W. Hu1, Y. Choi1, J. Jung1, K. Hong1, J. Kang1, B. Min1, Y. Huh1, S. Shin1, H. Lim1, Y. Chung2 1 Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 2 Department of Radiological Science, Yonsei University, Wonju, Korea An improved simple digital timing method has been developed for positron emission tomography (PET). The so-called initial rise interpolation method is based on an important characteristic of gamma signal: a properly preamplified and sampled gamma signal pulse can be characterized to arrive with an initial rise from baseline and then to go up with a maximum rise. Pulse arrival time is obtained by calculating the intersection of the initial rise line with the baseline for each gamma signal pulse. In this study, a FPGA-based data acquisition (DAQ) card was used for data acquisition and processing. We measured coincidence timing resolution of two types of recently developed 3 mm x 3 mm PET sensors (a fast and a slow SiPMs) using 3 different digital timing methods: initial rise interpolation, digital CFD and maximum rise interpolation. The experimental results show that, initial rise interpolation method provided the best timing resolution for both types of SiPMs: 0.7 ns FWHM for fast SiPM and 1.5 ns FWHM for slow SiPM (digital CFD: 1.5 ns FWHM and 2.2 ns FWHM; maximum rise interpolation: 1.8 ns FWHM and 2.7 ns FWHM). Based on these experimental results, we concluded that the improved simple digital timing method is reliable and useful for the development of high performance PET.
M13-339:

An Improved Simple Digital Timing Method for Positron Emission Tomography

The Study on the Accuracy of the Random Coincidence Estimation by Delayed Window

Y. Zhang, H. Li, S. Liu, S. An, C. Wang, R. Remirez, H. Beghaei, W.-H. Wong Dept. Experimental Diagnostic Imaging, the Univ. of Texas, M. D. Anderson Cancer Center, Houston, TX, USA With the advance of new technology, PET systems nowadays have much higher sensitivities than before with the adoption of faster scintillators, faster electronics, larger solid angle detectors and full 3D data collection. All of these lead to the increase of random coincidences that will affect the image quality. Accurate random correction is important for the quantitative imaging in PET applications. Delayed window method is the most popular way for random coincidences estimation. However, this method is not mathematically accurate. The difference between the delayed window randoms and prompt window randoms can be significant in certain conditions. We studied the details of these differences using Monte Carlo simulations since only the 428

simulation could separate the real random events from the true and scattered events. Three phantoms are used in the study, including the NEMA1994 20-cm long phantom, NEMA2001 70-cm long phantom and a 10-cm long, 1-cm diameter small cylinder phantom. The PET system used in this study is a 12-module LYSO camera with 20-cm axial FOV and 54-cm detector ring diameter. Two policies about how to handle the multiple events are used: one is to reject all multiples, and the other is to keep all good pairs. Random rates and NECR as the function of activities from prompt window and delayed window are obtained. The simulation results show that 1) the take-all-goods policy underestimates the randoms rate and the reject-allmultiples overestimate the randoms, which results overestimation of NECR for take-all-goods policy and underestimation of NECR for reject-all-multiples; 2) The discrepancy is more significant for smaller phantom than bigger phantom. The study shows the intrinsic discrepancy for random coincidence estimation with the delayed window method. When the phantom is relatively small compare to the FOV dimension of the PET system, the discrepancy is big enough to produce un-negligible errors.
M13-342:

W. C. J. Hunter1, R. L. Harrison1, S. B. Gillispie1, L. R. McDonald1, T. K. Lewellen1,2 1 Radiology, University of Washington, Seattle, WA, USA 2 Electrical Engineering, University of Washington, Seattle, WA, USA

Parametric Design Study of a Long Axial Field-of-View PET Scanner Using a Block-Detector Tomograph Simulation of a Cylindrical Phantom

We aim to optimize the design of a long axial field-of-view (AFOV) positron emission tomography (PET) scanner for use in whole-body PET-imaging studies. Given the large parameter space for PET scanner design, we conduct this study in two phases. Presently in our first phase, we use the noise equivalent count rate (NEC) divided by length of the AFOV as a coarse indicator of PET-scanner performance. From this first phase, we select a subset of scanner designs that we can then evaluate in terms of accuracy for a whole-body quantitation task.
M13-345:

A. B. Loehr1, J. R. Durst1, T. Michel1, G. Anton1, P. Geithner2 1 Novel Detectors / Medical Physics, ECAP, Erlangen Centre for Astroparticle Physics, Erlangen, Germany 2 CV IBD TP, Siemens Healthcare, Erlangen, Germany

Comparison of Recent Experimental Data with Monte Carlo Tools Such as RoSi, Geant4 and Penelope

The modeling of complex X-ray tube setups or photon detectors in the lower energy range of a few tens of keV requires a reliable simulation code which is also easy to use. ROSI (ROentgen SImulation, http://www.pi4.physik.uni-erlangen.de/Giersch/ROSI) was developed for that purpose and represents an object-oriented open source Monte Carlo simulation code that is based on the C++ library LSCAT-Gismo. The library uses the well established EGS4 code and its low energy extension LSCAT. We have worked thoroughly on validating the simulation results by comparing them to recently acquired experimental data and also cross checking with other established Monte Carlo tools such as Penelope and Geant4. One experimental setup was used to measure the x-ray dose emitted by a thin target. The setup was then implemented in ROSI and Geant4 and allowed a conclusion on the suitability of the implementation of the emission angle of the bremsstrahlung to agree with experimental data. The implementations in RoSi was found to be unsuited for the lower energy range and only after substituting it with the routine used in Penelope its agreement with the measurement was very good. A different experimental setup was used to measure the x-ray spectra from various thin and thick materials. The results from the simulations and the comparison with experimental data, allow a good insight in the accuracy of the implementations of the physics processes such as the generation of bremsstrahlung and the apropriate cross sections. A further experimental setup was used to measure the backscattered electrons from various materials. Again the setup was implemented in the simulation tools and the output was compared to the experimental results. A good agreement was found between simulation and experiment. Results for all three experimental setups and the related simulation results will be shown at the conference.
M13-348:

A. Ferrero1, J. K. Poon2, R. D. Badawi3 1 Dept. of Biomedical Engineering, KTH, The Royal Institute of Technology, Stockholm, Sweden 2 Dept. of Biomedical Engineering, UC Davis, Davis, CA, USA 3 Dept. of Radiology, UC Davis Medical Center, Sacramento, CA, USA

Characterization of the Scatter Fraction Arising from Different Sized Objects - a Simulation Study

The simplest approach to estimating scatter fraction for a uniform cylinder is to make one measurement using a radial displacement for the line source thought to give a value for scatter fraction that is representative of the whole phantom. In this work we use the SimSET simulation package to determine the appropriate line source radial displacement required to estimate scatter fraction for a range of phantom sizes suitable for small animal and whole-body PET scanners. We show that the displacement is close to a linear function of phantom radius and is only weakly dependent on scanner size or detector material. The optimum displacement position appears to be at approximately three-quarters of the phantom radius from the center. The positions prescribed by the current NEMA NU-4 standard for animal scanners are very close to this value but that for the NU-2 429

standard for whole-body scanners is not, resulting in an over-estimate of the scatter fraction for a uniformly filled 20 cm cylindrical phantom.
M13-351:

The Relevance of Peak-to-Valley Ratio and Signal-to-Noise Ratio for Anger-Logic-Based PET Detector Designs

H. Peng, C. Levin School of Medicine, Stanford University, Palo Alto,CA, USA A fundamental challenge for high resolution PET detector design is achieving a miniscule crystal size without greatly increasing the number of readout channels. To address this issue, commercial PET manufacturers have used several methods to design Anger logic block detectors. A widely used figure of merit for characterizing a block detector is the peak-to-valley ratio (PVR) of the crystals seen in the flood histogram. The overlap between two adjacent peaks in the flood will cause degradation of spatial resolution due to decoding uncertainties, also known as the block effect. In addition, to what extent the detectors signal-to-noise ratio (SNR) (associated with both scintillation light output and photodetectors) limits the detectors performance is of particular importance for novel block detector design based on avalanche photodiode (APD) and silicon photomultiplier (SPM) devices. We present a general model for the Anger-logic PET block detector design and studied the relevance of the PVR and SNR, in the context of the detectors modulation transfer function (MTF). For a given crystal and optical diffuser configuration (2.5x2.5x20 mm3 crystal, 9 mm thick optical diffuser), as the SNR decreases from 100 to 5, the PVR in the flood histogram decreases from 15.6 to 1.1 and the position uncertainty increases from 0.5 mm to 2.8 mm. To be able to accurately relate the position uncertainty to the physical crystal dimension, the MTF of the block detector was derived by taking the Fourier transform of the coincidence spread function (CPSF). As the PVR varies from 14 down to 2, corresponding to different optical diffuser configurations, the spatial frequency (corresponding to the half of the FWTM of the MTF) decreases by ~25% from 0.80 to 0.58 (cycles/mm) for 1 mm crystals, and decreases ~10% from 0.31 to 0.28 (cycles/mm) for 2.5 mm crystals.
M13-354:

J.-B. Michaud1, S. Rechka1, C.-A. Brunet1, R. Lecomte2, R. Fontaine1 1 Electrical and Software Engineering, University of Sherbrooke, Sherbrooke, Quebec, Canada 2 Nuclear Medicine and Radiobiology, University of Sherbrooke, Sherbrooke, Quebec, Canada

Monte Carlo Results from Neural Networks as an Alternative to Compton Photons LOR Analysis

Sensitivity in PET can be improved by lowering the energy threshold, at some expense of image quality. This paper presents Monte-Carlo results from our proposed method, where an alternative to explicit but statistically uncertain Compton-kinematics analysis uses neural networks to compute the correct Line-Of-Response (LOR) for simple triple-coincidences. Correctly analyzing those coincidences, involving 1 photoelectric 511-keV detection plus 2 511-keV-sum detection arising from Compton scatter, would account for a worthy sensitivity increase, depending on scanner geometry. A custom GATE adder was coded to allow exact accordance analysis of the Compton kinematics at the singles level, a feature not supported in the built-in adder because of centroid merge of electronic energy depositions consequent to photonic ones. This in turn allows better training possibilities for the neural network itself which, because of statistical uncertainties inherent to real scanner data, can only be trained to recognize the correct LOR from simulation data. A 55-mm diameter, 8-ring, 128-GSO-detector rings scanner was used to assess performance in worst-case conditions. The detector dimensions were 2.7 mm x 2.7 mm x 20 mm with an APD-like energy resolution of 35% FWHM. The resulting trained neural network has 6 neurons arranged in 2 layers. For various pointsource positions inside the scanner, discrimination of the correct LOR failed in approximately 22% of the time, while a cylindrical source yielded an error rate of 19%. Sensitivity increase ranged from 32 to 75%. Various other conditions were also simulated with similar results. Preliminary error dependence assessment shows contribution from, in decreasing order of importance, spatial localization within the field-of-view, detector form factor, undistinguishable theoretical cases, and energy resolution.
M13-357:

LuCaS2: Efficient Monte Carlo Simulations of Serial PET Scans for Assessing Detection and Quantification Methods Used in Patient Monitoring

S. Stute, H. Necib, N. Grotus, P. Tylski, N. Rehfeld, I. Buvat IMNC UMR 8165 IN2P3 CNRS, Orsay, France Objective: PET imaging is a promising approach for the early assessment of tumor response to therapy. However, there are currently no widely accepted rules to interpret the changes seen in successive PET scans. An objective evaluation of the performance of methods dedicated to detecting and characterizing the changes occurring between successive PET scans is needed. In this work, we propose a method to perform realistic Monte Carlo simulations of patient scans in the context of therapy monitoring. Material and Methods: Serial "tumor-free" sinograms of a patient with no tumor in the regions of interest (here the lungs) are first simulated with the GATE simulation toolkit, based on real PET/CT acquisitions. In addition, "tumor-only" sinograms are simulated, based on real tumor contours and tumor changes seen in patients. A specific strategy is used to properly 430

relocate tumors from scan to scan. After attenuation corrections, the "tumor-free" and "tumor-only" sinograms are combined and reconstructed to produce serial scans including realistic tumor changes. We illustrate the relevance of such data to investigate the sensitivity of the EORTC criteria to detect tumor changes between pairs of PET/CT scans. Results: In our configurations including SUV changes always greater than 25% with consistent tumor volume changes, the EORTC criteria led to a sensitivity of 100% for SUV change detection, for ideal tumor contour and PET scan registration. In more realistic conditions (realistic image registration and contour definition), the sensitivity remains high (94%). Conclusion: We showed that it is possible to simulate realistic serial PET scans to be used for assessing the performance of methods dedicated to the detection and quantification of tumor changes.
M13-360:

H. Ishii1, K. Ogawa1, S. Kabuki2, T. Tanimori2 1 Hosei University, Faculty of Science and Engineering, Tokyo, Japan 2 Kyoto University,Graduate School of Science, Kyoto, Japan

Image Restoration in the Dual Advanced Compton Camera System

We developed an image restoration method for an advanced Compton camera (ACC) making use of Compton scattering and recoiled electrons. In this paper, we simulated a system with two ACCs located perpendicular to each other with Electron Gamma Shower ver.5 (EGS5) code, and reconstructed images with a modified source space tree algorithm. Moreover, we deblurred the reconstructed images with a parametric Wiener filter. The results of simulations confirmed the validity of our proposed methods.
M13-363:

P.-C. Huang1, I.-T. Hsiao2, K. M. Lin3, C.-H. Hsu1 1 Department of Biomedcal Engineering and Enviromental Sciences, National Tsing Hua University, Hsinchu, Taiwan 2 Department of Medical Imaging and Radiological Sciences, Chang Gung University, Taoyuan, Taiwan 3 Division of Medical Engineering Research, National Health Research Institutes, Miaoli, Taiwan

A Voxel-Driven System Matrix Design for Multipinhole SPECT with Overlapping Projection

Single pinhole SPECT can provide high-resolution functional imaging of small animals, while multipinhole SPECT can further increase photon sampling and hence improve system sensitivity. In addition, overlapping projection of a multipinhole system can improve the overall usage of detector area, but such a multiplexing detection inevitably leads to image quality deterioration. Since iterative image reconstruction permits the incorporation of these factors in the form of system matrix, we propose a voxelbased design which explicitly accounts for multipinhole geometry, aperture diameter, and projection overlap. For voxel-based system matrix, we can determine no only the entire process of photon emission from its source and through a pinhole (including voxel depth and aperture size), but also identify the pinhole channel which the photon is passing through. Furthermore, the projection of the central pinhole is a circle, while the projections of an off-centered pinhole consist of two half ellipses. The detection probabilities of each voxel through every pinhole can be then derived from the areas of these composite ellipses. Note the areas also depend on the depth of voxel to pinhole and aperture diameter. We plan to investigate a voxel-driven system matrix for multipinhole SPECT with overlapping projections. We will also investigate how such a system design can improve the uniformity of spatial resolution and sensitivity as well as computation and storage requirements for iterative image reconstruction.
M13-366:

Extended Field of View in Gamma Camera, Using New Approach for Position Calculation

O. Amir, M. Kogan Molecular Imaging, GE Healthcare, Tirat Carmel, Israel The common Gamma camera, based on photomultiplier tubes (PMTs), utilizes the Anger algorithm to calculate the event positions. The Anger algorithm enables detection of events in a field of view (FOV) bounded by the center of the external PMTs. Events beyond this FOV cannot be distinguished. In common cameras, where 76mm diameter PMTs are used, a loss of over 76 mm in the FOV in each dimension occurs. In this work we present a new algorithm for event position calculation. This algorithm calculates a cost function based on the estimated PMTs distances from an event. The coordinate that minimizes this cost function is chosen as the event position. In the current work we perform simulation and use the proposed algorithm to calculate the event positions. The results demonstrate that the FOV can be extended by 20-40mm in each dimension (depending on the detector design), using the proposed position calculation. Alternatively to increasing the FOV, the current algorithm enables usages of less PMTs while maintaining the same FOV, achieved with the Anger algorithm. The usage of the suggested algorithm results in smaller margins. Thus enabling better proximity to the patient that may improve the image quality

431

M13-369:

Collimator Study of a -Camera System Using GATE

M. Mikeli, D. Thanasas, E. Stiliaris Dept. of Physics, National & Kapodistrian University of Athens, Athens, Greece A collimator study for a small field, high resolution -Camera system by means of GATE (GEANT4 Application for Tomographic Emission) simulations is presented in this paper. The aim of this study was the optimal definition and design of the geometrical characteristics of a parallel hole Pb-collimator, suitable for our -Camera system, which is based on the R2486 (HAMAMATSU) Position Sensitive Photomultiplier Tube, for different radio-tracers. The methodology followed two basic steps: (a) A validation phase with an existing parallel hole Pb-collimator of hexagonal structure, which preceded the main study. In this phase, experimentally obtained results for planar images are directly compared to simulated data. A simple phantom structure, consisting of four parallel capillaries filled with 99mTc water solution, was imaged by the -Camera system for several phantom-collimator distances and the measured and Monte Carlo calculated spatial projections were compared. (b) A GATE simulation setup for the main collimator study geometry was constructed and the -Camera detector is repeated 36 times (in steps of 10 deg) around a ring. This construction allows the simultaneous detection of data for further SPECT reconstruction studies. Simulation data are accumulated for three ellipsoidal sources placed at the center of the ring with different tracer energies, different relative intensities and for several collimator geometries. The collimator sensitivity is tabulated for each tracer energy according to the ratio D/T, where D represents the hole diameter and T the collimator thickness. Finally, the spatial resolution is defined for some basic collimator hole patterns (triangular, square, cylindrical and hexagonal). SPECT images are also reconstructed and the detected resolution is discussed.
This work is partially supported by the 03E287 research project, implemented within the framework of the "Reinforcement Program of Human Research Manpower" (PENED) and co-financed by National and Community Funds (25% from the Greek Ministry of Development-General Secretariat of Research and Technology and 75% from E.U.-European Social Fund). The financial support by the program KAPODISTRIAS (Special Account for Research Grants) of the National and Kapodistrian University of Athens is gratefully acknowledged. M13-372:

Optimal Design of a New Kinetic Strategy for Extracting FDG Transport and Uptake Information in Microfluidic Multi-Chamber Cell Culture Chip Coupled with PSAPD camera

W. Sha, Z. Yu, N. Vu, A. F. Chatziioannou, H.-R. Tseng, M. E. Phelps, S.-C. Huang Molecular and Medical Pharmacology, University of California Los Angeles, Los Angeles,CA, USA Objective: Information on the cellular transport and phosphorylation rate constants (k1,k2,k3,k4,Ki) of a tracer reflects the biological state of cells. A microfluidic chip with an array of cell culture wells, coupled with a PSAPD camera (MF-PSAPD) has been developed to give continuous measurements of positron radioactivity in individual wells. New strategies of controlling the infusion of tracer are needed to provide reliable estimates of the parameters. Methods: A switching strategy (SS) was conceived that consists of multiple media-infusion cycles, each of which has a tracer incubation (TI) period followed by a backgroundremoved (BR) period (tracer-free media). Equally switching strategy (ESS) with 12 cycles of TI and BR periods (5 min each) was evaluated by computer simulation and by experiments on MF-PSAPD. The SS was optimized by using a simulated annealing algorithm and D-optimal criterion to get optimal switching strategy (OSS). The OSS was examined for 5 to 12 cycles. Influence of Sensitivity of the OSS to kinetic parameter values was examined by using 3 sets of k values and volume ratio values (Vbratio of medium space to cell volume in a well). The performance of constant infusion (CI), ESS and OSS were evaluated by simulation. K values were estimated by regressing the simulated kinetics with the FDG model. Results for each condition were compared among schedules. Results: Simulations showed that the 12-cycle ESS did not perform as well (i.e., with larger estimated variability of the model parameters) as a 5-cycle OSS. OSS was found to be insensitive to variations in Vb and k values, and tended to have longer TI at beginning but longer BR at later times. Estimated k values with OSS have much smaller %CV compared to those of CI, with the largest reduction for %CV of Ki--from 762% to 26% under the same count rate conditions. Conclusion: The new optimized strategy of tracer incubation/measurement can provide reliable estimates of FDG ks in MF-PSAPD.
M13-375:

Regional Brain Uptake of Ketone Bodies and Glucose in Elderly Humans: a 11C-Acetoacetate and 18F-FDG PET Study

M. Bentourkia, S. Tremblay, M. Fortier, E. Croteau, O. Sarrhini, E. Turcotte, S. Cunnane Dept of Nuclear Medicine and Radiobiology, Universite de Sherbrooke, Sherbrooke, Quebec, Canada The brain relies on glucose as its primary energy substrate. However, ketone bodies, i.e. acetoacetate and hydroxybutyrate, are the main replacement fuels for brain activity during fasting or on a ketogenic (very high fat, low carbohydrate) diet. In some neurological diseases like Alzheimers disease, ketone bodies are potentially an alternate brain energy substrate due to deteriorating brain glucose uptake. Some brain tumors can also be treated by markedly increasing brain availability of ketones. We report here a study of ketone and glucose metabolism in human brain in young (mean 26 years) and aged (mean 73 years) healthy individuals as measured with positron emission tomography (PET), using the radiotracers 11C-acetoacetate and 18F432

fluorodeoxyglucose (FDG). The subjects fasted for six hours prior to the PET scans. In each subject, computed tomography (CT) images were obtained first followed by 11C-acetoacetate for 20 min, a 30 min rest/wash-out period, and finally the FDG scan. Three blood samples were withdrawn during the two scans and were analyzed for plasma radioactivity and for concentration of acetoacetate, hydroxybutyrate and glucose. 11C-acetoacetate and FDG images in each individual were co-registered and regions of interest were drawn on the FDG images and then copied onto the 11C-acetoacetate images. For both radiotracers, a 3compartment kinetic model was applied to time-activity curves, from which rate constants were calculated. In the sixteen selected brain regions for this study, although the regional metabolic rate of glucose (rCMRG) and the metabolic rate of ketones assessed as Kket = K1*k3/(k2+k3) were lower in the elderly subjects compared to the young subjects, the differences between rCMRG and Kket were almost similar in young and aged individuals. This protocol for brain fuel measurement by PET can be combined with treatment or on ketogenic diet to further study brain metabolism in neurodegenerative pathologies.
Acknowledgements. Technical support: C. Filteau and C. Langevin. Financial support: Canada Research Chairs (SC), CIHR, CFI, CRC-EL. M13-378:

K.-P. Wong1, S.-C. Huang1, V. Kepe1, G. W. Small2,3, J. R. Barrio1 1 Department of Molecular and Medical Pharmacology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA 2 Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA 3 Semel Institute for Neuroscience and Human Behavior, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Positron emission tomography (PET) with 2-(1-{6-[(2-[18F]fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malononitrile ([18F]FDDNP) has been used for in vivo imaging of amyloid plaques and neurofibrillary tangles, the neuropathological hallmarks of Alzheimer's disease (AD). Due to its simplicity as compared to other simplified reference tissue modeling methods, Logan graphical analysis with cerebellum as reference input has been primarily used to estimate the distribution volume ratio (DVR) of [18F]FDDNP as a measure of amyloid burden in human brain. However, it produces spurious parametric DVR images especially when the noise level is high and the total scan duration is shortened. The purpose of this study was to characterize a new graphical analysis method against the Logan analysis for the estimation of the DVR estimates at regional and pixel-wise levels, using cerebellum as reference input. Imaging studies were performed on 9 normal control subjects and 12 AD patients. Effects on regional (frontal and medial temporal) DVR estimates and parametric DVR images obtained by the Logan and the new approaches using data in different time intervals (35-125 min, 65-125 min, 85-125 min, and 45-65 min) were investigated and compared to the reference values derived by the Logan analysis for the time interval 35-125 min. For regional analyses, with the exception of DVR estimates of medial temporal in ADs obtained by the Logan analysis for the interval 85-125 min, no significant difference was observed for all paired comparisons with the reference. Logan DVR images generated using shortened time intervals (85-125 min and 45-65 min) were extremely noisy. In contrast, the new method produced DVR images comparable to the reference images even when shortened time intervals were used. The new method provides more reliable regional DVR estimates and less noisy parametric images, indicating that it may be a better alternative to the Logan analysis for analyzing [18F]FDDNP PET data.
This work was supported in part by the U.S. Department of Energy (DOE contract DE-FG02-02ER63420) and NIH grant P01-AG024831. M13-381:

Evaluation of Two Graphical Approaches for Regional Analysis and Parametric Mapping of Dynamic [18F]FDDNP PET

L. Ritschl1, F. Bergner1, C. Fleischmann2, M. Kachelrie1 1 Institute of Medical Physics (IMP), University of Erlangen-Nrnberg, Erlangen, Germany 2 Ziehm Imaging GmbH, Nrnberg, Germany

Empirical Cupping Correction for CT Scanners with Tube Voltage Modulation (ECCU)

X-ray CT measures the attenuation of polychromatic x-rays through an object of interest. The CT data aquired is the negative logarithm of the relative x-ray intensity behind the patient. These data must undergo water precorrection to linearize the measured data and convert them into line integrals through the patient that can be reconstructed to yield the final CT image. The function to linearize the measured projection data depends on the tube voltage U. In most circumstances, CT scans are carried out with a constant tube voltage. For those cases there are dozens of different techniques to carry out water precorrection. In our case the tube voltage is rather modulated as a function of the object. We propose an empirical cupping correction (ECCU) algorithm to correct for CT cupping artifacts that are induced by nonlinearities in the projection data. The method is rawdata-based, empirical and does neither require knowledge of the x-ray spectrum nor of the attenuation coefficients. It aims at linearizing the attenuation data using a precorrection function of polynomial form in the polychromatic attenuation data q and in the tube voltage U. The coefficients of the polynomial are determined once using a calibration scan of a homogeneous phantom. Computing the coefficients is done in image domain by fitting a series of basis images to a template image. The template image is obtained directly from the uncorrected phantom image and no assumptions on the phantom size or of its positioning are made. Rawdata are precorrected by passing them through the once-determined polynomial. Numerical examples are shown to demonstrate the quality of the precorrection. 433

ECCU achieves to remove the cupping artifacts and to obtain well-calibrated CT-values. A combination of ECCU with analytical techniques yielding a hybrid cupping correction method is possible and allows for channel-dependent correction functions.
This work was supported by Ziehm Imaging GmbH, Nrnberg, Germany. The high-speed image reconstruction software RayConStruct-IR was provided by RayConStruct GmbH, Nrnberg, Germany. We also thank Professor Gisela Anton Physikalisches Institut IV, University of Erlangen-Nrnberg for her support. M13-384:

Metal Artifact Correction Methods in CT

H. Yang, Y. Tsai, I. Hsiao Department of Medical Imaging and Radiological Sciences,, Chang Gung University, Tao-yuan, Taiwan Metal artifact is one of major problems in CT imaging, and the artifact comes from that the X-ray in the diagnostic energy range is heavily attenuated after passing through metal objects, and much fewer signals are detected by the detectors. Many research works have been done in solving the metal artifacts in CT imaging. The main idea is to fill in the incomplete projection data resulted from metal attenuation, and most of them are interpolation-based due to the easy implementation and effectiveness. In this study, we proposed two metal artifact reduction algorithms: post-processing interpolation and mask-based correction methods. The post-processing interpolation (PPI) method is similar to conventional interpolation method but without the need for raw projection data. The mask-based metal correction method generated metal artifact free images based on the assumption of accurate ROI segmentation, and non-overlapping projection from each ROI. The preliminary results showed that the PPI method resulted comparable image as compared to conventional interpolation methods while the mask-based method lead to much improved image free of metal artifact. Further study on validating the proposed methods using clinical data will be conducted.
M13-387:

Measurement of the MTF of a Cone-Beam Breast Computed Tomography Laboratory Scanner

G. Mettivier, M. C. Montesi, A. Lauria, P. Russo Department of Physics, INFN and Universit di Napoli Federico II, Napoli, Italy Since the landmark article by Boone et al. in 2001 where Cone-Beam Breast CT (CBBCT) with dedicated scanners has been proposed and where quantitative estimates have been provided of radiation dose in this type of exam the pendant-geometry breast Computed Tomography technique is being gaining consideration as a new methodology in breast tumor diagnosis. The scientific interest is related to practical advantages of 3D visualization of tissue lesions with respect to the 2D visualization of conventional mammography at the same Mean Glandular Dose (MGD) as two-view mammography. The basic detector in the CBBCT scanner is a Flat Panel Digital Detector coupled to cone-beam irradiation geometry. Many research groups are investigating on this type of exam. In this context we have assembled a non-clinical bench-top CBBCT scanner prototype. It uses a computer-controlled X-ray tube (SB-80-250, Source-Ray Inc., Bohemia, NY, continuous output, fixed anode) and a CMOS Flat Panel Detector with a 0.2 mm thick CsI:Tl scintillator layer as a digital X-ray imaging detector with a sensitive area of 12 cm 12 cm with 0.05 mm 0.05 mm pixels. In this work we report the first spatial resolution (MTF) measurements of this prototype. In our measurements we used the thin wire method and placed in the scanner FOV a home-made reticule of several parallel long wires arranged in the longitudinal (axial) and transverse directions. The 25 micron thick Au wires are fixed in a 2-mm thick PMMA frame and are directed horizontally (parallel to xy-coronal plane) and vertically (parallel to the z-axis of rotation) with 1 cm separation between them and with a slight tilting. Using this reticule we can make a single CT acquisition to assess the lateral and longitudinal PSFs across a 10 cm axial and 7 cm transverse FOV. The wires were tilted by an approximate angle of 2 degrees with respect to the rows and columns of the detector pixel array, in order to enable the determination of the presampled MTF.
M13-390:

An Efficient Iterative Reconstruction Algorithm for X-Ray CT

J. Hsieh, F. Dong, J. Fan GE Healthcare, Waukesha, WI, USA Purpose: X-ray radiation dose in CT has received increased attention in recent years. Recently, an iterative reconstruction algorithm was proposed and was shown to offer substantial spatial resolution improvement, noise suppression, and dose reduction. However, its computational complexity is still beyond the reach of routine applications. We present an efficient iterative reconstruction algorithm that offers the benefits of noise and dose reduction while providing substantial improvement in reconstruction speed. Methods and Materials: In conventional iterative algorithms, a system matrix is designed to accurately model the CT system optics. A statistical model is also generated to guide the optimization process. The image is continuously updated to minimize the cost function. To significantly improve the computational efficiency, the image update process in the proposed algorithm is performed on one of the two objective functions: a regularization step to minimize the local variation and a step to minimize the difference between measured and synthesized projections. The selection of a particular operation is based on a set of predetermined criteria and the characteristics of the update candidate. Results: Phantom and patient studies were conducted. Phantom studies show that the proposed algorithm offers a 50% dose 434

reduction while maintaining the low-contrast resolution as compared to a FBP. In the clinical experiments, double-blind tests show that similar image quality can be obtained at half the radiation dose as compared to FBP. The algorithm was implemented on a special hardware that generates images at multiple frames per second. Conclusion: We present an efficient iterative reconstruction algorithm that offers significant dose reduction while maintaining the same image quality as FBP. Compared to other iterative reconstruction algorithms, the proposed algorithm has the capability to reconstruct images at clinically acceptable speed.
M13-393:

M. C. Montesi1,2, G. Mettivier1,2, A. Lauria1,2, P. Russo1,2 1 Dipartimento di Scienze Fisiche, Universit Federico II and INFN, Napoli, Italy 2 Istituto Nazionale di Fisica Nucleare, INFN sez. Napoli, Napoli, Italy

X-Ray Cone-Beam Breast Computed Tomography: Phantom Studies on Microcalcifications Visibility

We evaluate the microcalcification (MC) detection capability of the cone-beam breast CT laboratory scanner developed by our group, based on a flat-panel detector with 50 micron pitch and X-ray tube with a 50 micron focal spot size (inherent filtration 1.8 mm Al and 0.2 mm Cu added filtration). We used a hemi-ellipsoid PMMA breast phantom of 14 cm diameter containing MC inclusions in the form of small fragments of egg-shell (size 350-450 micron) and CaC03 grains (size 250-350 micron). Six sets of five MCs in a cluster configuration have beam positioned in six different location at the midplane inside the phantom. Cone beam CT projections with an isotropic voxel of (78 micron)^3 have been acquired at 80 kVp and 60 kVp tube voltage and the MCs visibility/detectability has been investigated as a function of the detector pixel size (50, 100, 200 micron) and of the mean glandular dose (MGD) (6, 8, 10 mGy). Results indicate, e.g. that at 80 kVp MCs of 350 micron size can be clearly detected at a calculated MGD of 10 mGy.
M13-396:

J. Wu1, E. Lewis1, K. Wells1, G. Ferns2, J. Giles3 1 Centre for Vision Speech and Signal Processing, University of Surrey, Guildford, Surrey, UK 2 Post Graduate Medical School, University of Surrey, Guildford, Surrey, UK 3 Conquest Hospital, St Leonards-on-Sea, East Sussex, UK

Automated Coronary Calcium Scoring Using Predictive Active Contour Segmentation

Coronary artery disease(CAD) is the leading cause of death in many countries, but given early treatment, it is preventable. However, treated patients may never go on to have significant CAD, thus making their treatment unnecessary, while exposing them to potential side eects. Calcium deposits in the coronary arteries are the main indicators of coronary artery disease, and are visualized on multi-slice CT(MSCT) scans. A more targeted approach to diagnosis and management using image processing methods on such images is suggested. A system has been developed that automatically obtains a calcium score from MSCT coronary scans while disregarding clinically non-relevant lesions. In order to segment relevant lesions, we have implemented region growing and active contour methods using automated seed point generation, to identify irrelevant high attenuation objects. A predictive method of artery segmentation using active contour data has also been identified to compensate for object movement across axial slices. The low axial resolution of MSCT and difficulty in distinguishing lesions in close proximity of one another presents the greatest challenge and also prevents current vessel segmentation methods being used. However, our results have shown that such deficiencies can be overcome and accurate scoring performed without the need for user interaction.
M13-399:

V. Lee1,2, G. J. O'Keefe3, B. A. Sobott1, D. J. Peake1, G. Roe2, R. P. Rassool1 1 School of Physics, University of Melbourne, Parkville, VIC, Australia 2 CRCBID Cooperative Research Centre for Biomedical Imaging, Bundoora, VIC, Australia 3 Centre for PET, Austin Hospital, Heidelberg, VIC, Australia Traditionally, the contrast in medical images can be enhanced with the introduction of contrast agents into the body. Contrast agents contain elemental concentrations such as iodine which provide greater attenuation of x-rays in regions of the body where the agent is present. To fully exploit this, dual energy x-ray imaging methods can also be used which allow analysis of transmitted x-ray energy information through iodinated objects to determine energy-dependent attenuation coefficients. This information can lead to improved contrast in tissue specific imaging. By utilising the multi-spectral nature of an x-ray source and the energy discrimination properties of the Pilatus Hybrid Pixel Detectors, it is possible to investigate the optimisation of this process. The potential of this approach has been demonstrated in a proof-of-concept using a PILATUS 100K single photon counting pixel detector. Images obtained with a Hamamatsu u-focus spectral x-ray source with a tungsten anode and variable end-point energy up to 80 keV are compared with images obtained using monochromatic x-rays from a synchrotron source. The K-edge absorption of iodine has been studied to quantify the optimal tube operation endpoint energy for greatest contrast when using iodinated contrast agents. 435

K-Edge Imaging with a Hybrid Pixel Detector

M13-402:

H. Fayad1, J. F. Clement1, T. Pan2, C. Roux1,3, C. Cheze Le Rest1, O. Pradier1, D. Visvikis1 1 INSERM U650, LaTIM, Brest, France 2 Department of Imaging Physics, M.D. Anderson Cancer Center, Houston, Texas, USA 3 Institut Telecom - Telecom Bretagne, Brest, France

Towards a Generic Respiratory Motion Model for 4D CT Imaging of the Thorax

One of the most important parameters reducing the sensitivity and specificity in the thoracic and abdominal areas is respiratory motion and associated deformations which represent today an important challenge in medical imaging. The negative impact of respiratory motion is not only evident in the field of cancer diagnostic imaging but also in therapy applications, particularly in the field of radiation therapy. Modelling of the respiratory motion is therefore very important for the efficacy of both diagnostic and therapeutic imaging applications. All respiratory motion models presented to date are patient specific and require a 4D CT acquisition for every patient implying an associated increased dose which cannot be easily justified for all patients. The objective of this study is the description of a framework for the creation of a global generic model from 4D CT images of different patients. This global model is based on principal component analysis (PCA) and can be adapted to a given patient anatomy needing only one static CT image in combination with respiratory synchronized images of the patient surface. In this work 4D CT images of three patients were used to create the model which was tested by subsequently adapting it on the static CT images of two different patients. Our comparative results between the model derived and the corresponding acquired 4D CT images demonstrate the efficiency of the developed generic motion model.
M14: Simulation and Modelling of Medical Imaging Systems Saturday, Oct. 31 M14-1: 14:00-15:30 International Ballroom Center

J. Tabary1, S. Marache2,3, S. Valette2, P. W. Segars4, C. Lartizien2 1 CEA-LETI MINATEC, Grenoble, France 2 CREATIS-LRMN, Lyon, France 3 Medisys, Philips healthcare, Suresnes, France 4 Radiology Department, Duke University, North Carolina, USA

Realistic X-Ray CT Simulation of the XCAT Phantom with SINDBAD

Computer simulations are a valuable tool to help the design stage of radiographic systems or to evaluate the efficiency of image processing techniques in medical applications, such as x-ray CT. In order to generate simulated projection images of the human body comparable to those obtained by a real X-ray CT system, the simulation tool should incorporate a good modelling of all the physics involved in the radiographic chain and be able to efficiently handle a realistic and detailed anthropomorphic phantom. The X-ray radiographic simulation software SINDBAD can model any radiographic set-up, with a complete modelling of the source, the detector, and of the interaction of radiation within the object represented by its Computed Aided Design (CAD) model. Both direct and scattered contributions are calculated thanks to a combination of analytical and Monte Carlo approaches. The purpose of this study is to demonstrate that the combined use of SINDBAD with the extended NURBS-Based Cardiac-Torso (XCAT) phantom can provide realistic and high-resolution x-ray projections in an efficient computing time. We present in this paper first X-ray CT projection data which are simulated under an angiography configuration. Although the XCAT phantom is complex, the time of simulation is reasonable, even with the estimation of the scatter radiation. These simulated data are then reconstructed to study the disturbing effect of the scattered radiations. In future studies, similar simulations will be performed in real clinical CT scanner conditions, with an accurate modelling of detectors and a precise description of the XCAT phantom (including airway tree), and then compared to real X-rays CT images.
M14-2:

B. Marendic1, Y. Yang1, M. King2, M. N. Wernick1, J. G. Brankov1 1 ECE Dept, Illinois Institute of Technology, Chicago,IL, USA 2 Division of Nuclear Medicine, University of Massachusetts Medical Center, Worcester, MA, USA

Imaging in Sitting Position May Reduce Liver Artifact in Myocardium Perfusion Imaging

In this work we evaluated liver artifacts in SPECT myocardium perfusion imaging. Liver artifact occurs due to the liver-heart proximity and is manifested as an increase or decrease in apparent radioactive uptake on myocardium. The main motivation for the work is that there are a few SPECT imaging devices in the todays market that image patients in the sitting position, as opposed to the conventional supine (laying) position. The changes of the patient position causes slow sag of the liver downward and thus increase liver-heart distance. Therefore in this work we investigate if this changing the liver-heart distance may result in more accurate image reconstructions, and thus lower myocardium defect misdetection rates. To show this we use the realistic NCAT phantom and SIMIND the state-of-the-art computational environment that simulates the SPECT imaging modality.
This work was supported by NIH grants HL065425 and HL091017.

436

M14-3:

S. Rechka1, R. Fontaine1, R. Lecomte2, M. Rafecas3 1 Department of Electrical and Computer Engineering, University of Sherbrooke, Sherbrooke, Quebec, Canada 2 Department of Nuclear Medicine and Radiobiology, University of Sherbrooke, Sherbrooke, Quebec, Canada 3 Instituto de Fisica Corpuscular, Universidad de Valencia/CSIC, Valencia, Spain Using the simulator GATE, the inter-crystal scatter in the LabPET has been characterized with regard to the energy threshold, detection efficiency, and quantification, for different gamma ray incidence angles. For the energy threshold 350 keV, the number of the inter-crystal scatter events may exceed 19% of the total recorded events, 15% are positioned in the adjacent crystals to the irradiated one and 4% are distributed beyond the bordering area. By lowering the energy threshold from 350 keV to 0 keV, the detection efficiency of the primary events increases by nearly a factor of 3; however at 0 keV the number of the inter-crystal scatter events may exceed 38% of the total recorded events, 23% are positioned in the adjacent crystals to the irradiated one, and 15% are distributed beyond the bordering area all over the scanner. The proportion and the spatial distribution of the inter-crystal scatter events are an indicator of the degree of mispositioning of lines of response (LOR) that may alter the precision of quantification. As the final aim is to increase the scanner sensitivity by lowering the energy threshold, new advanced methods should be developed to filter the scattered events adequately.
M14-4:

LabPET Inter-Crystal Scatter Study using GATE

Simulation of HyperSPECT: a High Resolution Small Animal SPECT System Using X-Ray Optics

S. Tibbelin, P. Nillius, B. Cederstrm, M. Danielsson Physics Department, The Royal Institute of Technology, Stockholm, Sweden Functional imaging is becoming increasingly important in the medical science due to the large advantages provided by in-vivo visualization of biological processes. Instead of collecting indirect symptoms it is possible to label and trace the active substance in a drug to directly measure parameters such as uptake, kinetics and specicity. The alternative to pre-clinical research, which is still dominating, is to only use post mortem examinations with for example autoradiography. This alternative has obvious limitations in terms of cost and the fact that it is not possible to follow an organisms development over time. In vivo imaging of living organisms is today mainly accomplished by PET, SPECT and fMRI. The problem is that the spatial resolution for realistic image acquisition times is of the order 1-3 mm. Considering that this is a typical size of an internal organ in a mouse it is a serious limitation. We propose a new imaging modality, HyperSPECT, based on the same principles as SPECT but with drastically improved efficiency and spatial resolution. This is achieved by incorporating a large number of x-ray lenses between the detectors and the object. In current SPECT a pin-hole geometry is standard, that involves an unfortunate trade-off between efficiency and spatial resolution, our solution would change this. The efficiency of the system is still about a factor 5 lower compared to PET but at a much higher spatial resolution. We believe the efficiency may be improved by further optimizing the lens design. The detectors will be large area, photon counting and consist of a columnar CsI scintillator coupled to a CMOS integrated circuit for electronic read-out. It is necessary to perform some signal processing to estimate the position of interaction of the x-ray within the scintillator based on the pixel hit map. According to our simulations this works well and the resolution achieved in this way is sufficiently high for the HyperSPECT system.
M14-5: Investigation of Photon Depth of Interaction Issues of a PET Detector Design for Simultaneous PETMR Brain Imaging

V. C. Spanoudaki, H. Peng, P. D. Olcott, C. S. Levin Radiology, Stanford University, Stanford,CA, USA Information about the depth of interaction (DoI) of 511 keV annihilation photons within the crystal volume of a PET system may be exploited in order to achieve uniform spatial resolution within the field of view (FoV). We are currently developing a PET ring insert for simultaneous PET-MR brain imaging. In order to assess the need for DoI with regards to image quality and quantification, Monte Carlo simulations of various detector configurations were performed for radial crystal layers of different thicknesses (front/rear thickness: 20/0mm, 10/10mm, 6/14mm, 5/15mm, 2/18mm). The results show that a spatial resolution variation of 300% along the radial direction to the edge using a single crystal layer of 20 mm thickness and 2.5mmx2.5mm pixel size can be reduced to 50% by employing two crystal layers of thicknesses of 6mm (front) and 14mm (rear) (and the same pixel size). The variations on system photon sensitivity have also been studied and demonstrate a 150% increase in singles sensitivity and a 20% increase in coincidence photon sensitivity going from the center to the edges of the FoV as a result of the ring geometry. This sensitivity pattern was found to be similar for all crystal layer configurations considered in this study. The effect of DoI on PET quantification was evaluated by performing region of interest (RoI) analysis on reconstructed point sources extending to the edges of the FoV (1mm diameter, 3.7 MBq each placed at 0, 5, 10 and 15 cm offsets from the center of the 36 cm diameter FoV). The extracted RoI values as a function of radial offset show 20% under-estimation from expected RoI value for the 1 mm diameter sphere for detector configurations with no DoI scheme and only 6% deviation from expected RoI value for the 10mm/10mm DoI segmentation scheme.

437

M14-6:

An Information-Theoretical Assessment of PET System Design

C.-M. Kao Radiology, The University of Chicago, Chicago, USA We investigate an information-theoretical approach for characterizing the performance properties of a PET scanner. We obtain an analytic expression for the rate of information that a PET scanner, treated as an information channel, can transmit from the source to the receiver (disk files or reconstruction algorithms). Quite desirably, the resulting rate function increases linearly with the imaging time and radioactivity. For SKE/BKE detection task, it yields a result proportional to the SNR square of the ideal observer. For developing general-purpose scanners, the rate function indicates that higher sensitivity and higher resolution are desired. It also yields a sensitivity-resolution product that can be useful for guiding the tradeoffs between sensitivity and resolution in PET scanner design. The resulting rate function assumes a simple form and can be readily evaluated. We numerically evaluate rate functions for various PET scanner designs and investigate the usefulness of this rate function in assessing PET scanner design. Currently, we ignore subject attenuation, scatter and randoms in the derivation. These factors however can be readily incorporated in the derived rate function.
M15: Animal Imaging Instrumentation and Techniques Saturday, Oct. 31 M15-1: 16:00-18:00 International Ballroom Center

G. S. Mitchell1, C. Li1, R. Robertson2, M. D. Silva2, S. R. Cherry1 1 Biomedical Engineering, UC Davis, Davis, CA, USA 2 Department of Imaging Sciences, Millennium Pharmaceuticals, Inc., Cambridge, MA, USA

Small Animal Cerenkov Luminescence Imaging

Cerenkov radiation is a well-known phenomenon, in which optical photons are emitted when a charged particle moves faster than the speed of light in a medium; however, it has not to our knowledge been exploited for in vivo imaging studies. We propose and demonstrate a method, based on the detection of Cerenkov radiation, to noninvasively image - and +-emitting radionuclides inside small animals. This new imaging method is referred to as Cerenkov Luminescence Imaging (CLI). We have successfully detected Cerenkov light from -emitting radionuclides such as 18F and 13N, and we find that Cerenkov light levels in small animals are sufficient for detection using reasonable injected doses of radiopharmaceutical (100-200 Ci) and a sensitive CCD camera. For example, with our commercial optical imaging system, we find that: 100 Ci of 18F in a water solution in a well plate produces 2x105 photons/s/sr (400-800 nm); and 200 Ci of 18F-FDG injected into a mouse produces a typical surface emission photon density of 2000 photons/s/cm2/sr (575-650 nm). These results are consistent with calculations based on a GEANT simulation. The optical photon intensity is proportional to the radionuclide activity and the Cerenkov light levels increase as expected with the refractive index of the medium. We have also verified that the spectral characteristics of the signal follow a 1/2 relationship as expected for Cerenkov light. By measuring the emitted light from several views of the mouse, it is possible to reconstruct 3D images of the Cerenkov light distribution using analogous techniques to those used for bioluminescence tomography.
M15-2:

P. Vaska1, D. Kim2, S. Southekal3, S. Krishnamoorthy3, S. Stoll4, J. Fried5, D. Schulz1, J.-F. Pratte6, A. Dragone7, Y.-G. Cui8, G. DeGeronimo5, A. Bolotnikov8, C. L. Woody4, P. O'Connor5 1 Medical Department, Brookhaven National Laboratory, Upton, NY, U.S.A. 2 Dept. of Physics, Myongji University, Seoul, South Korea 3 Dept. of Biomedical Engineering, Stony Brook University, Stony Brook, NY, U.S.A. 4 Physics Department, Brookhaven National Laboratory, Upton, NY, U.S.A. 5 Instrumentation Division, Brookhaven National Laboratory, Upton, NY, U.S.A. 6 University of Sherbrooke, Sherbrooke, QC, Canada 7 Stanford Linear Accelerator Center, Palo Alto, CA, U.S.A. 8 Nonproliferation and National Security Department, Brookhaven National Laboratory, Upton, NY, U.S.A. A full-ring PET system consisting of a novel configuration of six CZT pixel detectors and custom front-end electronics has been built to explore the limits of spatial resolution with this strategy. Initial spatial resolution measures approach the image pixel size of 500 microns throughout the field of view, enabled by the excellent 3D spatial resolution of the detector (especially in the cathode-anode direction) and a maximum likelihood image reconstruction approach with a fully measured forward model. Recent results to be presented include the investigation of charge sharing and induced charge between nearby anodes in order to improve photopeak detection efficiency, a higher resolution image space of 0.25 mm voxels to determine the ultimate resolution limit, and the development of methods to measure and correct for drift time thereby improving coincidence time resolution and reducing randoms acceptance by a factor of 6. Initial phantom and mouse images are also presented. 438

Performance of a Prototype Ultra-High Resolution PET Scanner Using CZT Pixel Detectors

This work was supported by the U.S. Department of Energy under Prime Contract No. DE-AC02-98CH10886 and the Laboratory Directed Research and Development program at Brookhaven National Laboratory. M15-3:

C. Bontus1, J. Rahmer1, B. Gleich1, J. Weizenecker2, J. Borgert1 1 Philips Research, Hamburg, Germany 2 Univ. of Applied Sciences, Karlsruhe, Germany

Cardiac MPI of Mice over the Complete Heart Cycle

Magnetic Particle Imaging (MPI) allows rapid imaging of magnetic tracer distributions with high sensitivity. We show volumetric in vivo heart images of a mouse. The concentration of the commercially available MR contrast agent was in a range which is approved for humans. Images at various stages of the cardiac cycle were reconstructed proving the unprecedented temporal resolution of MPI. The results suggest that MPI can be scaled up to human sized scanners enabling, e.g., dynamic imaging of coronary blood flow.
M15-4:

M. Yamaguchi1,2,3, N. Kawachi1, H. Shimada3,4, S. Watanabe2,5, S. Takeda2,5, H. Aono2,5, S. Ishikawa2,5, H. Odaka2,5, T. Takahashi2,5, K. Arakawa1,3,4, T. Nakano3,4 1 Takasaki Advanced Radiation Research Institute, Japan Atomic Energy Agency, Takasaki, Gunma, JAPAN 2 Institute of Space and Astronautical ScienceJapan Aerospace Exploration Agency, Japan Aerospace Exploration Agency, Sagamihara, Gunma, JAPAN 3 21st Century COE Program, Graduate School of Medicine, Gunma University, Maebash, Gunma, JAPAN 4 Graduate School of Medicine, Gunma University, Maebash, Gunma, JAPAN 5 Depertment of Physics, University of Tokyo, Bunkyou-ku, Tokyo, JAPAN The next generation of Compton Camera based on Si and CdTe semiconductor has been developed in Japan for cosmic gammaray observation in space. Angular resolution and hence sensitivity of Compton Cameras are improved by employing the imaging semiconductors with good energy and spatial resolution. Moreover, for the energy range from several tens keV to a few MeV, silicon is suitable for the scattering part of the Compton camera since Compton cross-section is relatively large in Si, compared with the photo absorption cross section. Also effects of Doppler broadening is small. On the other hand, CdTe is suitable for the absorbers because of its high photo-absorption efficiency for gamma rays in this energy region. The high angular resolution and high energy resolution of the Si/CdTe Compton camera is very attractive features for medical applications. In the previous simulation study, we reported that the Compton camera has a low spatial resolution along the longitudinal direction, but it can be inproved by using two camera heads. In the present work, we measured the spatial resolution of the Compton camera, especially for the longitudinal direction. The spatial resolutions are compared with the simulation study. The longitudinal resolution is measured to be about 27 mm and the lateral one is about 5 mm at the distance of 6 cm from the surface of the top detector. The simulation result and the experimental one agree well with each other. Moreover we demonstrate the capability of threedimensional imaging by using multi-head system consisted of plural Compton cameras. The spatial resolutions are consistent with the simulation result. These results implies that clear 3D-image can be realized by using a multi-head system of Compton cameras.
M15-5: System Integration of FastSPECT III, a Dedicated SPECT Rodent-Brain Imager Based on BazookaSPECT Detector Technology

Spatial Resolution of Multi-Head Si/CdTe Compton Camera for Medical Application

B. W. Miller1, L. R. Furenlid1,2, H. H. Barrett1,2, S. K. Moore3, H. B. Barber1,2, V. V. Nagarkar4 1 Optical Sciences, University of Arizona, Tucson, Arizona, USA 2 Dept. of Radiology Research, University of Arizona, Tucson, Arizona, USA 3 Dept. of Biomedical Engineering, University of Arizona, Tucson, Arizona, USA 4 Radiation Monitoring Devices, Inc., Watertown, Massachusetts, USA

FastSPECT III is a dedicated rodent-brain, single-photon-emission computed tomography (SPECT) imager undergoing final system integration. Twenty independent BazookaSPECT gamma-ray detectors acquire projections of a spherical field of view with pinholes selected for desired resolution and sensitivity. Each BazookaSPECT detector comprises a columnar CsI(Tl) scintillator, image-intensifier, optical lens and fast-frame-rate CCD camera. Data stream back to processing computers via firewire interfaces, and heavy use of graphics processing units (GPUs) ensures that each frame of data can be processed in real time to extract the images of individual gamma-ray events. Details of the system design, fabrication methods, and processing algorithms are presented. The expected performance in terms of resolution and sensitivity are discussed in the context of in vivo amyloid plaque imaging.

439

The Center for Gamma-Ray Imaging is supported by NIBIB Grant P41-EB002035-5. The authors would like to thank Dr. Michael Gehm and Wei Ren Ng at the Laboratory for Engineering Non-Traditional Sensors (LENS), University of Arizona, for use of their Objet Eden350 rapid prototyping printer in aperture/pinhole development and fabrication. M15-6:

J. E. Ortuno1,2, P. Guerra1,2, G. Kontaxakis1,2, M. J. Ledesma-Carbayo1,2, A. Santos1,2 1 Electronic Engineering Department, ETSIT, Universidad Politecnica de Madrid, Madrid, Spain 2 Networking Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Zaragoza, Spain In this work, we present a 3D measurement system based on structured light projection, in order to reconstruct the surface of mice in fluorescent molecular tomography (FMT). A miniaturized digital light processing (DLP) projector is placed inside a FMT scanner prototype, so that a set of fringe patterns are projected onto the mouse. The distorted patterns are acquired with the high sensitive CCD camera employed to collect fluorescence emission, thus image registration is not required in our setup. A highaccuracy phase map is retrieved with phase-shifted sinusoidal fringes. Phase error due to the nonlinear gamma function of the projector is compensated after a calibration procedure. Robust phase unwrapping is performed with an additional Gray-code projection sequence. The phase of the reference plane below the mouse is extrapolated from the surrounding values using polynomial splines, in order to avoid repeating background acquisitions. An automatic phase-to-height non-linear calibration scheme has been applied using step pyramid-shaped artifacts located in the extremes of the field of view. Occluded and shadowed regions are reduced to less than 2% of upper mouse surface using two distinct projector positions. The accuracy of the proposed method has been tested with a realistic mouse model and ray-tracing software. The measured root-mean-square error (RMSE) was inferior to 1/1000 of the field of view. We will employ this 3D shape acquisition method to allow high resolution spatial sampling of photons in FMT reconstruction.
This work has been partly founded by EMIL Network of Excellence and by the Spanish Ministry of Science (TEC2008-06715-C02-02). The authors wish to thank M. Desco, J.J. Vaquero and J. Aguirre for fruitful comments and motivating discussions. M15-7: Performance Evaluation of PETBox: a Low Cost Bench Top PET Scanner Dedicated to High Throughput Preclinical Imaging

3D Surface Acquisition for FMT Using High-Accuracy Fringe Projection Profilometry

H. Zhang1, N. T. Vu1, Q. Bao1, R. W. Silverman1, D. B. Stout1, R. Taschereau1, B. N. Berry-Pusey1, D. A. Williams1, D. A. Williams1, A. Douraghy1, F. R. Rannou2, M. E. Phelps1, A. F. Chatziioannou1 1 Molecular and Medical Pharmacology, UCLA Crump Institute for Molecular Imaging, Los Angeles, CA, USA 2 Department de Ingenieria Informatica, Universidad de Santiago de Chile, Santiago, Chile

PETBox is designed to be a low cost easy to use bench top PET measurement platform dedicated for high throughput quantitative pharmacokinetic studies in small preclinical models. To achieve this goal, the scanner is integrated with a complete animal management system that provides life support including reproducible positioning, temperature control, anesthesia, real-time monitoring of animal respiration and a pathogen barrier. This approach minimizes the overall cost and complexity of preclinical PET imaging and should enable non-imaging scientists to embrace the technology. The system uses two opposing detector heads, each one consisting of a pixilated BGO array coupled to H8500 multi-channel photomultiplier tubes. The BGO arrays were segmented into crystals with a pixel pitch of 2.2 mm and a total active area of 47.5 x 100 mm2. A custom designed readout board was used to send amplified position signals to eight free-running ADC channels. The digitized samples were then processed using a field programmable gate array (FPGA). FPGA logics were used to develop algorithms for baseline restoration, digital constant fraction discrimination, energy integration, event positioning and coincidence detection. Basic system performance parameters were measured including: sensitivity, energy resolution, coincidence timing resolution and intrinsic spatial resolution. In coincidence mode the system has a sensitivity of 3.8% at the center of the gantry and an intrinsic spatial resolution of 1.53mm FWHM along the radial direction. The system also has a coincidence timing resolution of 5.88ns. Initial imaging studies were performed with live mice. Mice were imaged at 2 hour post injection (static) and for one hour following injection (dynamic). For both studies projection images were generated.

440

M. Bergeron1, J. Cadorette2,3, C. Bureau-Oxton4, J.-F. Beaudoin2, M.-A. Tetrault5, M. D. Lepage3, G. Robert3, R. Fontaine5, R. Lecomte1 1 Nuclear Medicine and Radiobiology, Universite de Sherbrooke, Sherbrooke, Quebec, Canada 2 CIMS, Universite de Sherbrooke, Sherbrooke, Quebec, Canada 3 Gamma Medica-Ideas, Sherbrooke, Quebec, Canada 4 Physics, Universite de Sherbrooke, Sherbrooke, Quebec, Canada 5 Electrical and Computer Engineering, Universite de Sherbrooke, Sherbrooke, Quebec, Canada The LabPET12 is the latest version of LabPETTM APD-based digital scanners, having an 11.4 cm axial field of view and 9216/4608 crystals/electronic readout channels to achieve high sensitivity and count rate performance, while preserving the same very high spatial resolution. Its performance characteristics have been thoroughly investigated largely based on the new NEMA NU4 2008 small animal PET standard. Average energy resolution is 20+/-3% for LYSO and 19+/-3% for LGSO. After timing alignment, the overall FWHM timing resolution is 7.1/8.3/9.2 ns for LYSO/LYSO, mixed and LGSO/LGSO coincidences. Imaging performance was assessed using various energy windows with lower threshold from 100 to 350 keV. Sensitivity varied from 8 to 2.8%, between 100 and 350 keV. Peak NEC count rates were 449 kcps at 66 MBq and 179 kcps at 81 MBq for the mouse and rat NEMA phantoms, respectively. Images of rats and mice obtained in dynamic, gated and whole-body static modes were obtained with a variety of 18F, 11C and 64Cu-labelled radiotracers, demonstrating the efficacy and versatility of the scanner for molecular imaging studies. In conclusion, the LabPET12 provides excellent dynamic performance with shorter imaging time or lower injected radioactivity while achieving the same exquisite spatial resolution as the LabPET4 and LabPET8 scanners.

M15-8:

Performance Evaluation of the LabPET12, a Large Axial FOV APD-Based Digital PET Scanner

441

Special Focus Workshops

High Performance Medical Imaging (HPMI) 2009
HP1 Platforms and Architectures Tuesday, Oct. 27 HP1-1: 08:00-10:00, Dogwood & Camelia

B. Keck1, H. G. Hofmann1, H. Scherl2, M. Kowarschik2, J. Hornegger1 1 Chair of Pattern Recognition, University of Erlangen-Nuremberg, Erlangen, Germany 2 Imaging & IT Division, Medical Electronics & Imaging Solutions, Siemens AG, Erlangen, Germany Particular applications of computed tomography require high slice resolutions. The fastest iterative implementations on graphics cards use 3-D textures to exploit hardware-accelerated trilinear interpolation. However, the size of 3-D textures is subject to technical limitations, which makes them inapplicable here. Alternatively a 2-D texture array can be used instead of the 3-D texture as in early graphics implementations. The additional memory synchronizations cause a significant loss of performance. We utilize new features of the recently released CUDA 2.2 framework to improve the performance of the Simultaneous Algebraic Reconstruction Technique (SART). In this paper we present an enhanced version of our efficient implementation of the most time-consuming parts of the iterative reconstruction algorithm: forward- and back-projection. We explain the required strategy to adapt the algorithm for the CUDA 2.2 features, in particular the usage of 2-D texture lookups from pitchlinear memory. Finally, we compare the result to our previous ones with respect to both reconstruction speed and technical limitations. The proposed strategy is a new balance between performance and limitations in resolution.
This work was supported by Siemens AG, Imaging & IT Division, Medical Electronics & Imaging Solutions. HP1-2:

High Resolution Iterative CT Reconstruction Using Graphics Hardware

CT Image Reconstruction Using Larrabee

M. Baer, C. Maass, M. Kachelriess Institute of Medical Physics, Erlangen, Germany CT-Image reconstruction is a computationally very demanding task. Recently a new many core platform codename Larrabee (LRB), which is currently developed by Intel, has been described by Seiler et al. In this work we want to present different approaches on how to use the new LRB platform in CT image reconstruction. Therefore we will give an example of a LRB optimized code section which can be used instead of standard code for CT image reconstruction. Additionally some major facts about the LRB platform will be given.
This work was supported by a grant of the Intel Corporation. HP1-3:

CT Image Reconstruction on Intel Larrabee Using Half Precision Floating Point Values

C. A. Maass, M. Baer, M. Kachelriess Institute of Medical Physics, University Erlangen-Nuremberg, Erlangen, Germany Computed tomography is a highly quantitative modality where the data values processed are of a high dynamic range and therefore computing CT images requires high accuracy and high precision during all processing steps. In the past unsuccessful attempts have been made to use integer values instead of floating point values. Nevertheless, memory bandwith is one of the major bottlenecks in todays reconstruction algorithms and it would speed up the reconstruction process to represent a datum with only 16 bit rather than with 32 bit. Intels new Larrabee x86 architecture offers a half precision floating point value that has the potential to be useful for CT image reconstruction. This study evaluates the image quality of CT image reconstruction based on such half values and compares to the standard case of using float values. Our findings are that it is possible to apply half precision data for rawdata and reconstruction volume if some conditions are considered.

442

HP2 APIs and Applications Tuesday, Oct. 27 HP2-1: 10:30-12:30, Dogwood & Camelia

PET 3D Iterative Reconstruction on Cell BE

D. E. Chevalier, E. Drapkin GE Healthcare, Waukesha, WI, USA We present an optimized implementation of PET 3D iterative ML-OSEM algorithm for the Cell BE processor. We describe some of the techniques to achieve high performance on the Cell BE processor. Finally, we report measured reconstruction performance for time-of-flight, conventional static and gated studies using the Cell BE processor in the GE Discovery 600-series PET-CT scanner.
HP2-2:

W. Zhang1, Y. Wang2 1 Department of Engineering Physics, Tsinghua University, Beijing, China 2 Nuctech Company Limited, Beijing, China

A Preliminary Study of OpenCL for Accelerating CT Reconstruction and Image Recognition

OpenCL is a new specification for heterogeneous parallel platforms used in CT imaging (GPU, CBE, Larrabee, DSP, etc). In this work, we implement the typical back projection step of FBP and the Haralicks texture features extraction algorithm from CT Image using the Nvidia and AMDs early release of OpenCL implementation. We also test our implementations in the two vendors cards using the same code and show several efficiently optimizations.
HP2-3:

J. L. Herraiz1, S. Espana1, S. Garcia2, R. Cabido2, A. S. Montemayor2, J. J. Vaquero3, J. M. Udias1 1 Dpto. Fisica Atomica, Molecular y Nuclear, UNIVERSIDAD COMPLUTENSE DE MADRID, MADRID, SPAIN 2 Dpto. Ciencias de la Computacion, UNIVERSIDAD REY JUAN CARLOS, MADRID, SPAIN 3 Unidad de Medicina y Cirugia Experimental, HOSPITAL GU Gregorio Maranon, MADRID, SPAIN A CUDA implementation of the existing software FIRST (Fast Iterative Reconstruction Software for (PET) Tomography) is presented. This implementation uses consumer graphics processing units (GPUs) to accelerate time consuming parts of the reconstruction. FIRST (originally developed in FORTRAN) was first recoded into C language. This allowed for a direct use of NVIDIA C for CUDA, as well as a straightforward implementation and performance comparison between the C versions of the code running on the CPU and on the GPU. Reduction of execution time, compared to the fastest available CPU was obtained. The CUDA implementation is then further tuned, including loop re-ordering and better memory allocation, what allowed even better performance in the GPUs.
HP2-4:

GPU Acceleration of a Fully 3D Iterative Reconstruction Software for PET Using CUDA

Fully GPU Based Real Time Corrections and Reconstruction for Cone Beam Micro CT

G. Jakab, A. Racz, G. Nemeth, T. Bukki R&D, Mediso Ltd, Budapest, Hungary We developed a complete GPU based data processing for cone-beam micro CT application which performs not only the reconstruction but also all the correction of the projection images on-the-fly. Test measurements were performed and processing times was compared on different hardware setups. The performance of the GPU together with our modified algorithm allow to perform all the calculations within the shortest exposure time of the system for the highest reconstructed volume size thus makes possible the complete data processing in real-time.
HP2-5:

L. Caucci1,2, L. R. Furenlid1,2, H. H. Barrett1,2 1 College of Optical Sciences, University of Arizona, Tucson, Arizona, USA 2 Center for Gamma-Ray Imaging, University of Arizona, Tucson, Arizona, USA

List-Mode Maximum-Likelihood Expectation-maximization Image Reconstruction on GPU Hardware

In medical imaging, maximum-likelihood (ML) expectation-maximization (EM) image reconstruction algorithms are frequently applied to binned data, as opposed to raw event data, such as the list of PMT values recorded or ML estimates of event parameters. The current trend towards measuring more and more attributes and building larger and larger 2-D and 3-D detectors causes memory requirements for binning grow geometrically, making binned-data approaches challenging. In this paper, the MLEM algorithm is applied to the list of attributes estimated for each event in a Compton camera. This 443

approach is well known as list-mode (LM) MLEM reconstruction in the medical literature. An expression for the probability density function of the list of events recorded with a Compton camera is presented. We also present some preliminary results we obtained by running a 3-D ML position estimation algorithm on a GPU supercomputer equipped with four NVIDIA GeForce 9800 GX2 computing processors. Our GPU implementation resulted in a 250x speedup with respect to an implementation of the same algorithm on a conventional cluster machine. These promising results show the enormous potential of commodity GPU hardware for medical image reconstruction.
This work was supported by NIH grants 5 R37 EB000803-18 and 5 P41 EB002035-10. HP2-6:

E. Morton1, K. Mann1, A. Berman2, M. Knaup3, M. Kachelriess3 1 Rapiscan Systems Ltd, Salfords, UK 2 Mercury Computer Systems, Chelmsford, USA 3 Institute of Medical Physics (IMP), Universitt Erlangen-Nrnberg, Erlangen, Germany

Ultrafast 3D Reconstruction for X-Ray Real-Time Tomography (RTT)

Imaging complete objects in a tomographic manner in nearly real-time is of interest for many applications. One example is imaging of patients with a very high frame rate to be able to observe and study variations in the patients anatomy (e.g. in the heart or lung) with high temporal resolution. Another example is baggage inspection where the object is transported at speeds as high as 0.5 m/s. Performing an x-ray tomographic inspection with a high spatial resolution in the order of 1 mm under these conditions corresponds to near real-time imaging. The scope of this paper is a real-time CT system that does not require physically moving parts. This is achieved by using multifocus x-ray sources and a number of detector rings surrounding the object. We briefly describe the scanner design and then concentrate on the signal processing pipeline of the scanner. Image reconstruction is driven by a cluster of cell blades and a highly performant backprojection algorithm that allows for full vectorization. Thereby, a continuous reconstruction of 240 to 480 images per second, each consisting of 768 by 768 pixels and receiving contributions from 768 focal spots, is achieved with a total of four cell processors. While the real-time tomograph is a dedicated luggage scanner the concept realized and validated is of highest interest for clinical CT, too. The possibility of defining arbitrary scanning sequences may yield further improvements in image quality regarding the scanning of moving objects or regarding dose-efficient scans with organ-dependent scan trajectories.
HP3 Physics and Algorithms Tuesday, Oct. 27 HP3-1: 13:30-15:30, Dogwood & Camelia

Monte Carlo Simulation of X-Ray Imaging Using a Graphics Processing Unit

A. Badal, A. Badano DIAM/OSEL/CDRH, U.S. Food and Drug Administration, Silver Spring, MD, USA A new code for Monte Carlo simulations of radiation transport using a Graphics Processing Unit (GPU) is introduced. The code has been developed using the CUDA programming model, an extension to the C language that allows the execution of general purpose computations on the new generation of GPUs from NVIDIA. The accurate Compton and Rayleigh interaction models and interaction mean free paths from the PENELOPE package, and a generic voxelized geometry model, have been implemented in the new code. The secondary particles generated by Compton, photoelectric and pair-production events are not transported. An ideal x-ray detector and a cone beam source can be defined to reproduce an imaging system and facilitate the simulations of medical imaging applications. A 24-fold speed up factor with the GPU compared to the CPU is reported for a radiographic projection of a detailed anthropomorphic female phantom. A description of the simulation algorithm and the technical implementation in the GPU are provided. This work shows that GPUs are already a good alternative to CPUs for Monte Carlo simulation of x-ray transport.
The mention of commercial products herein is not to be construed as either an actual or implied endorsement of such products by the U.S. Department of Health and Human Services.

444

HP3-2:

S. A. McAllister1, K. E. Schubert1, R. W. Schulte2, S. N. Penfold3 1 Department of Computer Science and Engineering, California State University, San Bernardino, San Bernardino, CA, USA 2 Department of Radiation Medicine, Loma Linda University Medical Center, Lomda Linda, CA, USA 3 Centre for Medical Radiation Physics, University of Wollongong, Wollongong, NSW, Australia

General Purpose Graphics Processing Unit Speedup of Integral Relative Electron Density Calculation for Proton Computed Tomography

Proton Computed Tomography (pCT) is an imaging modality that is based on the tracking of individual protons as they traverse an object. These paths deviate from a straight line due to the effects of multiple Coulomb scattering (MCS) and must be tracked using a formalism that models MCS. A sparse iterative solver, like the algebraic reconstruction technique (ART), is used to reconstruct the object using these non-linear paths. As the protons traverse an object, their paths deviate from a straight line due to the effects of multiple Coulomb scattering (MCS) and must be tracked using a formalism that models MCS. A sparse iterative solver, like the algebraic reconstruction technique (ART), is used to reconstruct the object using these non-linear paths. In order to perform the reconstruction, the path integral of relative electron density along the most likely proton path must be calculated. This is accomplished by calculating the integral given by the Bethe-Bloch equation of energy loss of a charged particle along its path. These integrals can be calculated in parallel using Gaussian quadrature. This paper will investigate the use of general purpose graphics processing units (GPGPUs) for execution of these parallel algorithms as well as exploiting the structure of the data being processed. From the results, recommendations for new research directions will be suggested.
HP3-3:

R. E. Ansorge1, S. J. Sawiak2, G. B. Williams2 1 Cavendish Laboaratory, University of Cambridge, Cambridge, UK 2 Wolfson Brain Imaging Centre, University of Cambridge, Cambridge, UK

Exceptionally Fast Non-Linear 3D Image Registration Using GPUs

3D non-linear image registration has many important applications in medical image processing. For example following disease progression or comparing genetically modified animal models. MRI images of the brain are often used for these investigations and individual data sets contain large numbers of voxels (256x256x256 is typical). Unfortunately the high resolution deformation maps needed, require very lengthy computations. For example a cubic B-Spline based deformation map with control point spacings of a few voxels in each dimension can require days to compute on a single processor. Typically PC clusters have been used to speed the process but these are expensive, require cooled server rooms and dedicated technical support. We present new results from a demonstration program, AIRWC (Accelerated Image Registration with CUDA), which uses an NVIDIA GTX285 Graphics processor card to speed computation of both 12-parameter affine and 3D cubic B-Spline based image registrations. The performance of this program is very encouraging; typically affine registrations require a few seconds and high resolution BSpline registrations having up to 131x131x131 control points require ~6 minutes. The latter case corresponds a speedup of ~750 over our conventional code running on a single processor. Results are presented for both human and mouse brain imaging. A high resolution average brain was computed from 14 human scans by iteratively registering to an average target. In all about 112 highresolution B-Spline registrations were performed in about 9 hours using a DELL XPS630i PC running Windows Vista. The GPU coding was carried out using NVIDIA's C++ like CUDA programming language and some details are presented. Overall we have found the code development to be rather straight forward and thus we expect that the exploitation of GPUs for image registration will grow rapidly. We acknowledge the help of colleagues at the Wolfson Brain Imaging Centre in providing our images.
HP3-4:

A. Andreyev1, A. Sitek2, A. Celler1 1 Radiology, University of British Columbia, Vancouver, BC, Canada 2 Radiology, Harvard Medical School and Brigham and Women's Hospital, Boston, MA, USA

Acceleration of Blob-Based Iterative Reconstruction Algorithm Using Tesla GPU

Blob-based iterative image reconstruction techniques provide high-quality denoised images in the photon starving emission tomography. However, the attractiveness of blob-based iterative algorithms is devalued by their high demands on the computation time. In this study we investigate the use of graphic processing units (GPU) to parallelize the ray-driven blob-based OSEM reconstruction algorithm for SPECT. We obtained a speed-up factor of 14.7 as compared to the blob reconstruction performed using CPU. The reconstruction times of the superior in terms of image quality blob-based images on GPU were comparable to the reconstruction times of voxel-based image on the CPU. The algorithm will be further accelerated by more effective utilization of the GPU register space and shared GPU memory, which we plan to implement in the future.

445

HP3-5:

Branchless Vectorized Median Filtering

M. Kachelriess Institute of Medical Physics (IMP), University of Erlangen-Nuernberg, Erlangen, Germany Median filtering is an important tool in medical imaging. Based on the vector capabilities of modern hardware, which allows for vectorized min, max and mask operations, we provide a median algorithm that is both branchless and vectorized. Its runtime is predictable and data-independent and it can simultaneously operate on several one-dimensional signals thereby making use of data-level parallelism. Some median filtering algorithms require to sort the data that lie in the filter window before selecting the central entry. Others are recursive, starting from a pivot element to determine the number of elemens smaller than, equal to or larger than the pivot and then start over with one of these three subsets. In most cases the fast median algorithms have a data-dependent run-time. Our branchless vectorized median (BVM) filter keeps track of a sorted array from which values are deleted and to which new values are inserted. As a spin-off effect we may also our work to provide a data sorting algorithm that is branchless and vectorized. Although it is of O(N^2) computational complexity while other sort algorithms are between O(N) and O(N ln(N)) computational complexity it may outperform other implementations. This is mainly due to the fact that we have no unexpected branches which would stall the instruction pipeline and that we can simultaneously operate on vectors of data. Here, however, we focus on median filtering. BVM is compared to the median implementation of the Intel Performance Primitives (IPP) library using both constant data, which is an unrealistic setting, and linear or random data which more realistic. For constant data IPP is up to five times faster than BVM whose run-time does not depend on the data themselves. However, regarding the realistic case of median filtering noisy data BVM outperforms IPP by about a factor of 2 in our CPU-based implementations, which is far more than we did expect initially.
This study was supported by RayConStruct GmbH, Nuernberg, Germany. We thank the Intel Corporation and Fujitsu Technology Solutions GmbH for providing the high performance hardware. HPP Poster Session Tuesday, Oct. 27 HPP-1: 15:30-16:30, Dogwood & Camelia

A. Wirth1, A. Cserkaszky2, B. Kari1, D. Legrady2, S. Feher2, S. Czifrus2, B. Domonkos3 1 Department of Diagnostic Radiology and Oncotherapy, Semmelweis University, Budapest, Hungary 2 Institute of Nuclear Techniques, Budapest University of Technology and Economics, Budapest, Hungary 3 Mediso Medical Imaging Systems Ltd., Budapest, Hungary The presentation summarizes our results achieved so far in the TeraTomo project, the aim of which is to develop a high-speed Monte Carlo algorithm optimized on graphics processors (GPU) which would provide more accurate PET reconstructions by the utilization CT data. The reconstruction is based on the conventional full 3D ML-EM iteration. The most important feature of the algorithm is that the forward projection is a realistic Monte Carlo simulation of the annihilation photon trajectories, i.e. it includes scatters and absorption. The system matrix is calculated on the fly in each iteration step with basic back projection of the LORs. Application of Monte Carlo based forward projection can potentially lead to enhanced image quality. Nevertheless, in this way even approaching the quality of conventional analytical and numerical algorithms requires significantly more computing power. On the other hand, Monte Carlo algorithms are generally easily parallelized, consequently they are ideal subjects of GPU based programming. Specifically in our case the graphics processors were expected to be a faster solution by at least an order of magnitude than the similarly priced CPUs. One of our goals was to implement and benchmark the reconstruction on GPU architecture using Nvidia CUDA. The other goal was to investigate whether including the photon scatter process in the reconstruction algorithm has beneficial effect on the image quality.
HPP-2: CPU-Based Parallelization and Acceleration of a Fully 3D Iterative Reconstruction in Cone-Beam Collimator SPECT

Implementation of 3D Monte Carlo PET Reconstruction Algorithm on GPU

H. Ye1, A. Krol2,1, D. H. Feiglin2 1 Dept. of Physics, Syracuse University, Syracuse, NY, USA 2 Dept. of Radiology, SUNY Upstate Medical University, Syracuse, NY, USA

In order to reconstruct high quality images in cone-beam collimator SPECT, a fully 3D reconstruction using maximum a posteriori - ordered subsets expectation maximization algorithm and the cone-beam volumetric system model (MAP-OSEMCVSM) is required. Our studies, as well as other researchers confirmed that the MAP-OSEM-CVSM algorithm allows obtaining much better reconstructed images than other algorithms, such as FBP or MAP-OSEM with a line-length system model, but at the 446

price of the much longer computational time. In order to apply it in clinical practice, the algorithm has been parallelized using message passing interface (MPICH2) software on a 32-processor-cluster. The parallelization was implemented with a viewdistribution method for the two most time-consuming reconstruction steps: (i) the generation of the volumetric system model, and (ii) forward- and back-projection (322 hours/model and 98 minutes/iteration on a single 2.4GHz CPU, respectively). Total reconstruction time (TRT) vs. number of nodes has been studied using the simulated cone-beam SPECT for a numerical chest phantom for those two steps. The computational time was inversely proportional to the number of nodes because very little data sharing between nodes was required during generation of CVSM. However, it was not the case for the forward- and backprojection calculations because a large amount of data sharing among the master- and slave-nodes was needed in each iteration. We observe that the doubling the number of nodes from 15 to 30 reduces TRT by less than 35%. We also found that TRT could be further significantly reduced by increase in the number of ordered subsets when keeping the product of the number of iterations and the number of subsets the same for all reconstruction. Considering the hardware cost and the reconstructed image quality, 12 subsets with 10 nodes for a 120-view-projection set is recommended in the parallelized MAP-OSEM-CVSM reconstruction.
HPP-3:

Strategies for Accelerating Forward and Backprojection in List-Mode OSEM PET Reconstruction Using GPUs

W. Dieckmann, S. Thada, W. C. Barker National Institutes of Health, Bethesda, MD, USA MOLAR (Motion-compensation OSEM List-mode Algorithm for Resolution-recovery reconstruction) was developed to perform accurate, high-resolution image reconstruction for the Siemens ECAT HRRT. Due to its computational demands, cluster-based computing is necessary to attain reasonable reconstruction times. In this study we have sought to accelerate the forward projection, backprojection, and combined forward-backprojection modules of MOLAR by writing CUDA modules to be run on an NVIDIA Tesla S1070 GPU system. Because forward projection operates primarily on neighborhoods of image voxels, 3D arrays with caching facilitated substantial speedup compared to CPU-only code. Backprojection requires write access to image memory in a more random fashion so we used two image arrays, one x-dominant and one y-dominant. Using these strategies, we then varied the GPU thread distribution for performance tuning. Overall, we were able to obtain speedup factors of 41, 20, and 9 for forward, backprojection, and forward-backprojection. This work has enabled us to improve dramatically the reconstruction times of the entire MOLAR package.
HPP-4:

A GPU-Accelerated Implementation of the MOLAR PET Reconstruction Package

W. C. Barker, S. Thada, W. Dieckmann National Institutes of Health, Bethesda, MD, USA MOLAR (Motion-compensation OSEM List-mode Algorithm for Resolution-recovery reconstruction) is a PET package developed for the Siemens ECAT HRRT that has substantial computational demands. In this study we have implemented a GPUassisted version of MOLAR to facilitate more rapid reconstruction, especially for high count and dynamic data sets. Using modules written in CUDA and executed on an NVIDIA Tesla S1070 GPU system we were able to accelerate the major computeintensive components of MOLAR. In particular, we wrote CUDA modules for boundary checking, forward projection, backprojection and scatter modeling to support CPU code that performed the remaining tasks. The GPU code ran 30, 60, 20 and 200 times faster, respectively, for the GPU modules as compared to our CPU versions. The entire GPU-assisted MOLAR package was 31-49 times faster (over a range of 500k-10M coincidence events reconstructed), the code now being dominated by data I/O and event list updates. We anticipate that overall performance will benefit further from faster hard drives, and from more and faster CPU RAM. We also found that the performance scales well with multi-GPU implementations. The MOLAR packaged is now well positioned for rapid reconstruction and perhaps even 4D parametric modeling.
HPP-5:

J. Gregor1, M. W. Lenox2, P. Bingham3, L. F. Arrowood4 1 University of Tennessee, Knoxville, TN, USA 2 Texas A&M, College Station, TX, USA 3 Oak Ridge National Laboratory, Oak Ridge, TN, USA 4 Babcock & Wilcox Y-12, Oak Ridge, TN, USA

Multi-Core Cluster Implementation of SIRT with Application to Cone Beam Micro-CT

Iterative x-ray CT algorithms produce high-quality images but potentially do so at a prohibitive computational cost when applied to large problems. We describe how to implement SIRT, which is a well-known weighted least squares algorithm, for parallel execution using a commodity cluster of networked multi-core PCs. Algorithmic strategies applied to achieve acceptable reconstruction times include relaxation which eliminates half of the iterations otherwise needed, orthogonalized ordered subsets which greatly increases the rate of convergence, and focus of attention which reduces the overall problem size in a data-driven manner. Implementation strategies include a workload distribution scheme which provides each core with mutex-free access to 447

its local shared memory, as well as a simple modification thereof that ensures a balanced workload for the entire cluster. We illustrate the efficacy of the above scalable approach by providing experimental results for a cone beam micro-CT mouse data set.
HPP-6:

Y. Wang1, T. Feng2, L. Shen2, Y. Xing2 1 Joint Research Institute of Tsinghua University & Nuctech, Beijing, China 2 Department of Engineering Physics, Tsinghua University, Beijing, China

Research on ATI-CAL for Accelerating FBP Reconstruction

In recent years, different high performance image reconstruction techniques have been researched intensively. The techniques explored in accelerating CT reconstructions include cell processors [1,2], FPGA[2], GPU [3, 4, 5,], and Larrabee. Among them, GPU has been developing fast into one of the most important platforms for high-efficiency reconstruction because of the fast development of hardware and software development environments. In September 2006, NVIDIA Corp released a new GPGPU platform CUDA which enables user to employing the power parallelization capability of GPU with little knowledge of graphics API. For example, Scherl[6], Yang[7], and Bi[8] implemented their reconstruction algorithms on CUDA and gained fairly good speedup. AMD, being another important graphics card vendor also released its computing platform for GPGPU ATI Stream which includes a high-level Brook+ coding environment and a low-level Computer Abstract Layer (CAL) development environment. We accelerated 3D CT backprojection based on ATI-Brook+ stream computing,[9] and gained good performance, but it is not fully satisfying. In this work, we exploit the capability of ATI-CAL in the acceleration of CT reconstruction algorithm. The conventional FBP reconstruction algorithm is used as a tester. The strategies and techniques to implement the algorithm are studied, as well as the tricks to optimize the speed. The method is examined on a simulation study and gained much higher speedup than using ATIBrook+.
HPP-7: A Novel Implementation of Total Variation Iterative Reconstruction Algorithm Suitable for Parallel Computation

D. Shi Toshiba medical research institute, USA, Inc., Vernon Hills, IL, USA In this work, a simplified version of the total variation minimization iterative reconstruction algorithm for the conventional x-ray CT application, which was originally developed by the University of Chicago group, is proposed. The simplifications make it possible for complete parallel computing that is suitable to be implemented with a GPU board. Reconstructed images from real phantom data are demonstrated to verify the validity of our simplified algorithm.
D. Shi is with Toshiba Medical Research Institute USA, Inc. 706 N. Deerpath Dr., Vernon Hills, IL 60061 (e-mail: [email protected]). HPP-8:

Efficient Convolution via FFT with Real-to-Complex Packing

A. A. Zamyatin Toshiba Medical Research Institute USA, Vernon Hills, IL, USA We improve convolution speed by a factor of two by packing real signals of size N into a complex signal of size N/2 and performing N-size FFT. Note that standard FFT-based convolution methods require FFT of size 2N. The proposed method works with traditional ramp filter, with recently proposed Hilbert filters, and also with non-symmetric shifted Hilbert filters. In general, the proposed method can be applied to any filter, provided it uses a real kernel.
HPP-9: Accelerated Image Reconstruction on a Cluster of Two AMD GPUs in CBCT with Non-Uniform Detector Geometry

A. Bonissent, C. Morel CPPM-CNRS-IN2P3, Marseille, France The imXgam group of CPPM has developed the XPAD3 hybrid pixel X-ray detector with 560x960 pixels (130 m each). A Cone Beam CT scanner (CBCT) has been built, based on XPAD3 detectors, which enables volume reconstruction of a complete mouse with 65 m resolution. However the basic unit is a chip whose width is about1cm. These are assembled into barrettes which are in turn assembled in a tiled structure in order to reduce the dead regions. This results in a geometry where the apparent position and size of the pixels are not uniform. The volume reconstruction is performed using the analytic method of Feldkamp, Davis and Kress. Reconstruction time was estimated to a few hours due to the large number of pixels and images and the added complexity of the interpolation resulting from the non-uniform geometry. The algorithm was ported to a GPU architecture based on the AMD Firestream9270 board driven via the Brook+ library. All the processing is performed in the GPU and the host CPU is used only for steering and I/O operations. For improved performance, a dual-GPU version has been developed. With this setup, the total 448

reconstruction time is of order 2 minutes, mostly limited by disk I/O. Compared to the usual CPU implementation the gain in speed is of the order of 100.
HPP-10:

P. Giannetti1, A. Annovi2, G. Volpi, M. Berretta2, P. Laurelli2 1 INFN - Pisa, Pisa, Italy 2 INFN-LNF, Frascati, Italy

A Fast FPGA-Based Clustering Algorithm for Real Time Image Processing

Real time image analysis has undergone a rapid development in the last few years, due to the increasing computational power available at low cost. However computing power is still a limit for some high quality applications. High-resolution medical image processing, for example, are strongly demanding for both memory (~250 MB) and computational capabilities allowing for 3D processing in affordable time. We propose the reduction of execution time of image processing exploiting the computing power of parallel arrays of Field Programmable Gate Arrays (FPGAs). We apply this idea to an algorithm that finds clusters of contiguous pixels above a certain programmable threshold and process them to produce measurements that characterize their shape. It is a fast general-purpose algorithm for high-throughput clustering of 2-D data. The algorithm is designed to be implemented with FPGAs but it can also profit of cheaper custom electronics. The key feature is a very short processing time that scales linearly with the amount of data to be processed. This means that clustering can be performed in pipeline with the image acquisition, without suffering from combinatorial delays due to looping multiple times through the whole amount of data. The algorithm can be extended at 3-D images. It is organized in two steps: the I one clusters the data, the II one analyzes each cluster of data to extract the desired information and should be tailored for each possible application. These technologies combine the high performances of rigid dedicated hardware (capability of high parallelization) with most of the distinctive flexibility (reconfigurable logic) of generalpurpose but lower-performance CPUs. The typical clustering related calculations (II step) are separated from the combinatorial complications of clustering, enabling sophisticated calculations for the II step. achieving high quality in online applications. It has been implemented into a XC5vlx155 [1] device and simulated.
[1]http://www.xilinx.com/ HPP-11:

Fast Median Filtering Using Graphics Hardware

W. Chen, M. Beister, Y. Kyriakou, M. Kachelriess Medical Physic, Institute of Medical Physics, Erlangen, Germany Median filtering is a commonly used technique in smoothing and denoising applications. Based on the vector programming model of modern graphics processing units (GPUs), which directly support for minmax operations and compare and select as fundamental instructions. We implemented the branchless vectorized median (BVM) filter using NVIDIA's compute unified device architecture (CUDA). The BVM filter keeps track of a sorted array from which values are deleted and to which new values are inserted. It is branchless, it makes use of data--level parallelism and thereby its runtime is data--independent and highly predictable. We provide performance benchmarks in comparsion to other implementations and describe some important criteria such as the memory layout for a fast accessing scheme and discuss the bottlenecks in the computation.
This study was supported by CT Imaging GmbH, Erlangen, Germany. HPP-12:

Interactive GPU IP Chain Simulator for Mobile X-Ray Imaging

T. Brown, A. Cheryauka, L. Anderton GE Healthcare-Surgery, Salt Lake City, USA Trustworthy evaluation of clinical imagery is done based on visual perception by experts and focus groups. It deals with assessment of the perceived image fidelity, i.e. comparison of image data acquired, processed, and displayed using various imaging chain settings. To streamline tuning of X-ray fluoroscopic IP chain, improve effectiveness of observer studies, and support research efforts, we discuss HW and SW implementation and review the early results associated with GE OEC Interactive Imaging Chain Simulator. Our methodology is to create a friendly framework utilizing a common workstation, build tools to streamline image analysis, and improve user interactions. Heterogeneous computing and visualization platform has been tested to simulate real-time fluoroscopy on a mobile C-arm. The examples of tuning the anatomical profiles to address clinical requests and challenging cases are discussed.

449

HPP-13:

S. S. Huh1, L. Han1, W. L. Rogers2, N. Clinthorne3 1 BioMedical Engineering Department, University of Michigan, Ann Arbor, Michigan, USA 2 Division of Internal Medicine, University of Michigan, Ann Arbor, Michigan, USA 3 Department of Radiology, University of Michigan, Ann Arbor, Michigan, USA

Real Time Image Reconstruction Using GPUs for a Surgical PET Imaging Probe System

We present an on-line image reconstruction system using GPUs for a surgical PET imaging probe system. The PET imaging probe system provides intermediately reconstructed 3-dimensional images on line in order to detect small tumors or remnants of surgically removed tumors either during surgery or preoperatively. The on line image reconstruction requires high speed computation. Even multi-processor computers may not satisfied the requirement. Fortunately, currently available GPUs provide high speed computation. By combining the high speed computation of GPUs and a variant of one-pass list-mode OSEM that we are currently studying, we are trying to implement the PET imaging probe system. In the variant of the one-pass list mode OSEM, we can present intermediately reconstructed images by consecutively applying a non-overlapping data window or an overlapping data window on data. The PET probe system consists of a segment of a conventional PET scanner and a small, highresolution PET detector equipped with a position-tracker and collects coincidence events between them. The imaging probes proximity to the lesions and the small size crystals in the probe contribute to high spatial resolution. The continuously updated 3D image is reprojected in real-time onto a plane whose orientation is driven by the tracking system. Based on the intermediately reconstructed images and the preoperative PET scans, surgeons or clinicians can survey suspicious regions in detail. Since the on line image display could be almost achievable, the PET imaging probe can be viewed as a handheld, clinician-guided camera capable of seeing the distribution of the radio-tracer. We are also building a prototype PET imaging system. The GPU based image reconstruction system would be integrated with the prototype PET imaging system.
HPP-14:

J. W. Moore1, H. H. Barrett1,2, L. R. Furenlid1,2 1 College of Optical Sciences, University of Arizona, Tucson, AZ 2 Department of Radiology, University of Arizona, Tucson, AZ

Adaptive CT for High-Resolution, Controlled-Dose, Region-of-Interest Imaging

We have developed a exible x-ray micro-CT system, named FaCT, capable of changing its geometric conguration and acquisition protocol in order to best suit an object being imaged for a particular diagnostic task. High-performance computing technologies have been a major enabling factor for this adaptive CT system in terms of system control, fast reconstruction, and data analysis. In this work, we demonstrate an adaptive procedure in which a quick, sparse-projection pre-scan is performed, the data are reconstructed, and a point of interest is identied. Next, a diagnostic-quality scan is performed where, given the point of interest, the control computer calculates an illumination window for on-line control of an x-ray source masking aperture to transmit radiation only to a small region around the point of interest throughout the scan trajectory. Finally, the diagnostic scan data are used for an estimation task. We use a combination of a multi-core CPU and a pair of NVIDIA Tesla GPUs to perform these tasks.
This work was supported by the National Institutes of Health under NIBIB Grant P41-EB002035-5 (Center for Gamma-ray Imaging) and NCI Grant R24- CA83148 (Southwest Animal Imaging Resource).

Nuclear Techniques Applied to Cultural Heritage

CH1 Nuclear Techniques Applied to Cultural Heritage I Wednesday, Oct. 28 CH1-1: 09:00-12:00, Dogwood & Camelia

Ion Beam Analysis in the Studies of Cultural Heritage

Z. Smit Department of Physics, University of Ljubljana, Faculty of Mathematics and Physics, Ljubljana, Slovenia The method of proton-induced X-rays and gamma rays (PIXE-PIGE) were applied for non-destructive investigations of archaeological objects and objects of art. The materials studied involve metals of Roman military equipment, Venetian-type and medieval glass, and metal pigments on frescos and oil paintings. It is shown that an intense use of brass by the Romans can be dated to the middle of the 1st c. BC. The manufacturing details of several metal objects are presented. A numismatic study on the composition of silver in early medieval coins from Slovenia points that the ore from the mines in Carinthia was mainly used by the eastern mints and was intended for commerce with the Hungarian kingdom. By the differential PIXE method, based on sequential measurements by different impact energies, the concentration profiles of tinned, silvered and gilded surfaces were determined, identifying the method of fire gilding. A comprehensive analysis was made on the Venetian-type glass from Ljubljana and compared with the glasses from the other parts of Europe. It was found out that the same type of halophytic plant 450

flux was used in different European regions, reflecting the intensity of Venetian commerce. For the transition period between the natron-type Roman and plant-type medieval glass, a study of the Carolingian-period glass confirmed that the Roman tradition prevailed in the area of Eastern Alps into the 9th c. Metals pigments were identified on the works of the 17th c. and impressionist painters, and the composition of white pigments at particular authors was sought. The review also includes brief description of other techniques and materials, such as confocal XRF and analysis of ceramics and textiles.
CH1-2:

High-Energy Ion Beam Analysis of Cultural Heritage Objects

D. Strivay, F. Mathis, H. Calvo del Castillo, G. Chene, T. Dupuis, F.-P. Hocquet, L. Samain, C. Defeyt, H.-P. Garnir Centre Europeen d'Archeometrie, University of Liege, Liege, Belgium Our laboratory has a long history in the application of ion beam techniques to material analysis. Since the end of the nineties, we have mainly focused our research on cultural heritage objects analysis. We have two beam lines installed on a AVF cyclotron and two beam lines on a 2.5 MV van de Graaff accelerator. In parallel to IBA techniques, we are active also in mobile analysis systems. We will present here the last developments of our new system of high-energy ion beam analysis and recent applications in archaeometry.
This work is supported by the Belgian Inter-university Attraction Pole Program P6/16 Belgian State Belgian Science Policy. CH1-3: The Potentialities of Accelerator Based Techniques in Archaeology and Cultural Heritage Diagnostics: the Contribution of AMS and IBA

G. Quarta, L. Calcagnile, M. D'Elia CEDAD, Department of Engineering of Innovation, University of Salento, Lecce, Italy Accelerator-based analytical methods have increased, in the last years, their importance in the field of cultural heritage diagnostics and archaeometry. This is in particular true for AMS (Accelerator Mass Spectrometry) radiocarbon dating and IBA (Ion Beam Analysis) methods whose features make them suitable for the analysis of objects relevant for the cultural heritage. Since 2001 a multidisciplinary research centre (CEDAD: Centre for Dating and Diagnostics) has been established at the University of Salento, Lecce, Italy. The facility is based on a 3 MV Tandetron accelerator equipped with five experimental beam lines for AMS 14C dating, RBS-Channeling, PIXE-PIGE analyses in external beam mode, ion implantation and nuclear microprobe. Furthermore another beam line for the analysis by AMS of rare isotopes other that 14C such as 10Be, 26Al and 129I and uranium isotopes is under design. The main features and figure of merit of the AMS and the PIXE-PIGE lines will be presented together with recent applications in the compositional analysis of different classes of materials (such as metallic artifacts, glasses, obsidians, pottery and baroque statues ) and in the study of different archaeometric problems. Furthermore the possibility to perform in the same facility both AMS radiocarbon dating and IBA analyses resulted in the development of combined AMS-IBA approaches for the study of the same archaeometric problem. Through this approach the chronological information obtained by 14C are integrated with the compositional results obtained by IBA methods. The analysis of obsidian samples from different sites in Anatolia and in Southern Italy from 14C dated archaeological levels, for instance, allowed to obtain information about the provenance of the raw materials and to reconstruct the exploitation of the different sources over the time.
CH1-4: The New External Ion-Beam Station at the Demokritos Tandem 5.5 MV Accelerator: a Unique Analytical Tool in the Fields of Cultural Heritage and Environmental Science

D. Sokaras1, S. Karabagia1, E. Bistekos2, L. Georgiou2, J. Salomon3, M. Bogovac4, A. Lagoyannis1, S. Harissopulos1, V. Kantarelou1, E. Aloupi5, I. Aslani5, V. Paschalis6, A. G. Karydas1 1 Institute of Nuclear Physics, N.C.S.R. "Demokritos", Athens, Greece 2 Philon Models, Piraeus, Greece 3 Laboratoire du C2RMF, Centre de Rechereche et de Restauration des Musees du France, Paris, France 4 Institute Ruder Boskovic, Zagreb, Croatia 5 Thetis Authentics LtD, Athens, Greece 6 Benaki Museum, Athens, Greece

At the 5.5MV Tandem VdG accelerator of the Institute of Nuclear Physics of N.C.S.R. Demokritos, Athens, an external ionbeam set-up has been recently installed. The aim of this development was to establish a complete experimental set-up integrating the analytical capabilities of the PIXE, RBS and PIGE techniques, so that a complete elemental and near surface structural characterization of samples to be attained in an almost non-destructive way and without any limitation concerning their size or conductive state. A careful 3D mechanical drawing optimized the set-up experimental parameters fulfilling the special requirements imposed for optimum performance of the aforementioned techniques. For the data acquisition, novel hardware and software tools were utilized to perform digital pulse processing. In particular, two Xilinx XtremeDSP Development kits for Virtex-4 SX cards were used with a custom FPGA configuration as a complete platform for development of an on-chip digital 451

pulse processor for X-ray, -ray and charged particle detectors. The first applications were focused in the quality control of tagged materials such as technologically authentic replicas of attic ceramics with black glazed decoration and of coatings used by conservators for long term protection of a painting. Analytical diagnostic studies were carried out for representative paintings of contemporary Greek painters in order to identify and document materials and techniques. The accuracy of the quantitative analysis performed by the external ion-beam set-up techniques (PIXE, PIGE and RBS) was validated and assessed by analyzing a set of standard reference glass and geological materials.
CH2 Nuclear Techniques Applied to Cultural Heritage II Wednesday, Oct. 28 CH2-1: 14:00-17:00, Dogwood & Camelia

E. H. Lehmann1, S. Hartmann1, E. Deschler-Erb2 1 Spallation Neutron Source, Paul Scherrer Institute, Villigen, Switzerland 2 Dept. of Pre-Protohistory, University Zurich, Zurich, Switzerland

Neutron Imaging Methods for the Non-Invasive Investigation of Cultural Heritage Objects

Beside other methods for non-destructive inspection (e.g. X-rays, MRI, micro-waves, ) neutrons play a particular role in several respect. Because neutrons interact with the atomic nuclei of the objects under investigation, a specific contrast and transmission mechanism occurs. In particular, light elements like hydrogen, boron and lithium deliver high contrast while heavy elements are relatively transparent. Therefore, the inspection with neutrons delivers another degree of freedom. They can favorably be used in complement and superposition to other methods. The investigations of objects with cultural heritage importance using neutrons have been focused therefore onto metallic structures, where X-ray inspections fail or are limited. In this manner, organic parts can be made visible within the metallic cover (e.g. embedded hidden contents glues, resin, corrosion). The Swiss base for neutron research is the spallation neutron source SINQ, where two neutron imaging beam lines are installed, one for thermal the other one for cold neutrons. We present several results of studies with objects of cultural heritage importance in order to demonstrate the options and challenges in the filed. Most of the samples were taken from the native museums in Switzerland, where problems for transport and insurance are easier to handle. While simple transmission radiography can deliver the first impression of object content as a two-dimensional distribution with a superposition of all layers in beam direction, tomography is more useful if the whole sample volume is to be inspected. Despite of the challenging performance of the modern imaging tools it has to accept that this kind of investigations can only provide several new aspects in the final evaluation of the cultural heritage community, where the archaeologists and museums experts conclude their knowledge.
CH2-2:

M. Cotte1,2, J. Susini2 1 C2RMF - UMR171, Paris, France 2 ESRF, Grenoble, France

Synchrotron-Based X-Ray Absorption Spectroscopy for Art Conservation: Looking Back and Looking Forwards

Within the last decades, synchrotron-based X-ray absorption spectroscopies (XAS) techniques have been used to an increasing extent in the field of Cultural Heritage. These methods offer a combination of attributes particularly well suited for the chemical analysis of works of art. Their non-invasive character, low detection limit, high lateral resolution and high chemical sensitivity are highly relevant properties for the chemical characterization of precious, heterogeneous and complex materials. In particular, the high spatial resolution chemical mapping capability, providing information about local composition and chemical states, even on trace elements, is a unique asset. The chemistry involved in both the past history of the objects, i.e. during their fabrication, as well as in their future life, i.e. during preservation and restoration treatments, can be addressed by XAS. On the one hand, many studies intend to explain optical effects occurring in glass or ceramics, by probing the molecular environment. Considering that information about the current chemical state of specific elements provides a reliable fingerprint of chemical reactions mastered years, decades or centuries ago, models of ancient (and lost) technologies can be derived from XAS data. On the other hand, XAS can also be useful to characterize on-going reactions, which are then considered as alteration phenomena. In such cases, the main elemental composition is usually unchanged. Hence, monitoring oxidation state -or more generally chemical- modifications can be of great importance. Examples of XANES analyses, in particular carried out at the ID21 X-ray microscopy beamline (ESRF), will be given. Besides, current trends and recent instrumental developments will be reported.

452

CH2-3:

A. Adriaens1, M. Dowsett2 1 Department of Analytical Chemistry, Ghent University, Ghent, Belgium 2 Department of Physics, University of Warwick, Coventry, UK

Time-Resolved Spectroelectrochemistry Studies for the Protection of Heritage Metals

Corrosion is the major problem in the degradation of heritage metal objects. Besides being of fundamental interest, study of the corrosion behaviour is essential to the development of adequate conservation and preservation processes. The information one wants to acquire at this stage relates to the nature and the type of the attack, the environmental conditions contributing to the corrosion or weathering, and the effectiveness or otherwise of any countermeasures. In view of this we have developed and tested equipment and methods for time-resolved in-situ monitoring of corrosion, passivation and coating using synchrotron X-rays and electrochemistry. The core development is an electrochemical cell (eCell) which can be used for electrochemistry (both analysis as well as treatment) while at the same time the metal surface is analyzed through a thin electrolyte layer using a spectroscopic technique (spectroelectrochemistry). However it can also be used for gaseous exposure and fast spatially localized experiments involving wetting of metal surfaces with single droplets of fluid (in this case the "e"of eCell stands for environmental). eCell has been proven in comparatively short-term (maximum days) and high time resolution (seconds to minutes depending on the detector) experiments. The methods have been studied in the context of cultural heritage metals with a view to both fundamental studies in corrosion and its prevention and to deriving practical monitoring techniques for use by conservators. In this paper we will discuss two projects: (1) the development and testing of a coating to protect lead objects which is stable, reversible or easy to apply and to remove, protective against corrosion and aesthetically justified and (2) the development of a simple monitoring method during the stabilization of cupreous objects recovered from marine environments.
CH2-4: Elemental Analysis of Pigments of a Manuscript Illumination Using SR-XRF with a High-Resistivity CCD at the ALS

M. Battaglia, R. S. Celestre, D. Doering, T. S. Kim, P. Denes LBNL, Berkeley, CA, USA The use of synchrotron radiation induced X ray fluorescence (SR-XRF) is well suited for the elemental analysis of pigments in art works since the analysis is fast and non-destructive and small beam spot sizes make possible to probe tiny regions and disentangle individual pigments. XRF has been already successfully applied to the study of pigments used in illuminations of Medieval and Renaissance manuscripts. In this talk we present the results of a study of the pigments of a miniature produced by the workshop of Fra Angelico around 1450, now in private collection, performed at the LBNL Advanced Light Source using a thick, high-resistivity CCD detector. The use of a CCD offer some advantages. The active surface is large (118 mm2 in our case) and the high pixellisation allows to operate with large X-ray fluxes without pile-up. This is particularly important when performing XRF at a light source. Cluster shape analysis helps in rejecting spurious signals and scattered photons striking the detector at large angles. At the same time fully depleted thick CCDs offer high detection efficiency and excellent energy resolution (155 eV FWHM at 5.9 keV for our detector). The pigment study supports the conclusion of the stylistic analysis that two independent artists are at work on this illumination by a comparison of the composition of the pigments used in the body of the letter with those used for the marginal decoration and by a study of the layers of pigments used for the rendering of the faces with the support of the results of IR photography and published results of analyses carried out on other manuscripts and paintings by Fra Angelico and his workshop.

Contrast in Neutron Imaging

CN1 Contrast in Neutron Imaging 1 Tuesday, Oct. 27 CN1-1: 08:00-10:30, Grand Ballroom 6

Contrast Mechanism in Neutron Imaging , State of the art

E. H. Lehmann, A. Kaestner, L. Josic ASQ, Paul Scherrer Institute, Villigen, Switzerland Neutron imaging has been established as a tool for non-destructive investigations comparable and complementary to the common X-ray methods. Although available only at a limited number of places on a competitive level, recent methodical developments have convinced that the similar techniques can be provided to scientific and industrial customers than X-ray does (tomography, phase enhancement, polarization, real-time imaging, quantification, ). Neutron imaging is performed mostly by using thermal neutrons. If cold neutrons are available, their image contrast can be increased while the transmission becomes little less for most of the materials. It becomes important if large objects have to be transmitted. Nowadays, modern neutron imaging is exclusively performed with digital detectors. In this way, neutron fields transmitted through a sample can be considered as a matrix of pixel elements. While the standard contrast in neutron imaging is given by the material composition and the attenuation properties, 453

respectively, there are some more options to enhance the image contrast in a suitable way. Many of the crystalline structural materials show abrupt changes in the scattering cross-section at neutron energies in the cold range. This is caused by Bragg reflections at the lattice planes of the crystals. Using selected energies in the initial neutron beam above and below these Bragg edges enables to raise the contrast by factors. Neutrons can also be considered as waves with amplitudes and phases. Recently, some imaging techniques came up to use wave propagation methods of partly coherent neutrons to distinguish explicitly between the absorption and the phase contribution in the sample interaction. Such techniques have favorably be used already to investigate magnetic domain structures. Finally, the refraction of neutrons can be used to enhance edges if the beam conditions are tuned in the right way.
CN1-2:

Using Fast Neutrons to Image Induced Fissions

M. A. Blackston, P. A. Hausladen, P. R. Bingham, M. N. Ericson, L. Fabris Oak Ridge National Laboratory, Oak Ridge, TN, USA In contrast to x-ray inspection, which is often used in a variety of industrial and homeland security settings, the source requirements of neutron radiography typically limit its use to a few large reactor and spallation sources. At Oak Ridge National Laboratory, we have concentrated on performing fast-neutron radiography with a portable associated-particle deuterium-tritium neutron generator, where the time and direction of each neutron is known by the time and location of the associated alpha particle detected in an alpha-particle detector embedded in the generator. The coincidence technique enables good contrast with a relatively weak (108 neutrons s-1) source close to the inspected object. A source of this size can be taken to storage or production facilities to image objects in situ. Knowledge of the neutron direction also enables imaging of induced reactions, such as fission, that occur along the neutron path. In the prototype system, neutrons are detected using an array of pixelated block detectors covering a field of view of approximately 40 x 80 cm. The major components of the imaging system will be described, and the first combined images of transmission and fission for laboratory objects containing uranium will be presented.
CN1-3:

NECTAR - Radiography and Tomography Using Fission Neutrons

T. J. Buecherl, C. Lierse Lehrstuhl fuer Radiochemie, Technische Universitaet Muenchen, Garching, Germany The NECTAR (NEutron Computer Tomography And Radiography) facility is used for the non-destructive assay of samples covering a broad range of scientific and technical questions by means of radiography and tomography, basically where other methods fail. Fields of application are for example the investigation of hydrogen containing materials in large volume metallic samples like oil distributions in gear boxes, moisture distributions in large wooden blocks or the distribution of glue in conglutinated timbers etc. Turbine blades, motors, gear and shaft assemblies and other large sized metallic samples can be investigated for cracks and their material and geometrical composition. Sculptures, archaeological and biological samples are well suited objects to be characterised for their composition, too. Especially the combination of light and dense materials has to be emphasised (e. g. soil in metallic containers for investigation of root growth, wooden sculptures with metallic supporting structures etc.). Samples can have a maximum size of 1 m in diameter and 1 m in height. The maximum burden is up to 400 kg. Examples of applications will be presented.
CN1-4:

A. S. Tremsin1, J. B. McPhate1, W. Kockelmann2, E. Lehmann3, J. V. Vallerga1, O. H. W. Siegmund1, W. B. Feller4 1 Space Sciences Laboratory, University of California at Berkeley, Berkeley, USA 2 ISIS Facility, Rutherford Appleton Laboratory, Chilton, UK 3 SINQ, Paul Scherrer Institute, Villigen, Switzerland 4 NOVA Scientific, Inc., Sturbridge, USA

From Energy-Resolved Neutron Radiography to Imaging Phase, Texture and Residual Strain

The contrast mechanism in neutron radiography most often is associated with the variation of neutron absorption and/or scattering coefficients. A number of new neutron imaging techniques such as phase contrast and dark field imaging, energy resolved imaging and imaging of magnetic properties with polarized neutrons have emerged recently. These techniques substantially enhance the capabilities of non-destructive testing with neutrons and reveal features not measurable with traditional transmission radiography. Among the particular applications of energy resolved imaging is the possibility to map texture, phase and even strain within variety of samples, providing the energy resolution is sufficient to accurately map the wavelength of Bragg edges across the samples. The most challenging among the latter applications is the imaging of strain, which for realistic samples has to be measured with ~100 accuracy, which in turn requires the energy resolution of sub-mA or E/E ~10-4. Combination of modern bright pulsed neutron beams with high resolution detectors may enable adequate resolution for strain mapping through time of flight transmission spectroscopy. Each pixel in the resulting image will have not only the number of transmitted neutrons, but also their spectral distribution. Although the integrated intensity of pulsed sources is typically lower than that of a continuous source, the possibility to time tag all detected neutrons in each pulse and thus to acquire all energies simultaneously may result in 454

a much faster acquisition of the spectral data. We present the results of energy resolved imaging with neutron counting detectors with ~55 m spatial and ~1 s temporal resolution performed at pulsed neutron beamlines at ISIS facility. The accuracy of our measurements reached few hundred strain values, approaching the desired resolution for quantitative analysis of engineering samples.
CN1-5:

Neutron Radiography for Characterizing Irradiated Nuclear Fuels

B. A. Hilton, S. L. Hayes Idaho National Laboratory, Idaho Falls, ID, USA Neutron radiography (NR) or neutron imaging is a non-destructive testing technique that is complementary to x-ray radiography. Since NR is sensitive to materials that interact strongly with neutrons but not photons, it has specialized applications in postirradiation examinations of nuclear fuels and in characterizing light elements that have high neutron interaction crosssections (such as hydrogen, boron and lithium) within a metal or higher-Z material. A brief overview of the history of neutron radiography applications to characterize the irradiation performance of nuclear fuel will be presented. The environmental challenges and technique of NR unique to postirradiation examination (PIE) samples, which have very high beta-ray and gamma ray radiation fields, are discussed. A description of the neutron radiography capability for characterizing irradiated fuels and materials at the Idaho National laboratory will be presented. This facility includes a shielded hot cell and TRIGA reactor and is one of a few in the world that performs NR on irradiated fuels and highly radioactive samples. For PIE applications, the indirect or transfer technique is used with an activation foil converter, typically dysprosium and indium, which converts the image by subsequent autoradiography exposure of x-ray film negative or a digital x-ray imaging detector. Recent results and process variables comparing images obtained from photographic film and imaging panels are discussed. Neutron radiography is used as a non-destructive interrogation method to characterize irradiation behavior such as, fuel relocation, fuel swelling and axial growth, and other internal features. For light water reactor fuel applications, neutron radiography is used to identify locations of pelletpellet interfaces, missing pellet surface and high hydrogen content, which are important in understanding fuel degradation mechanisms.
CN2 Contrast in Neutron Imaging 2 Tuesday, Oct. 27 CN2-1: 10:45-11:57, Grand Ballroom 6

Quantitative Neutron Radiography and Tomography

L. Z. Josic, E. H. Lehmann, A. Kaestner Spallation Neutron Source Division, Neutron Imaging & Activation Group, Paul Scherrer Institut, CH-5232 Villigen PSI, Switzerland In the quantitative neutron radiography and tomography, the image contrast is interpreted as a variation of an effective macroscopic cross section () averaged over the neutron energy range and sample volume. To validate neutron imaging for the quantification of crystalline materials, transmission experiments were firstly performed with polychromatic thermal and cold neutron beams. It provided a determination of material composition and possible thickness ranges in the beam direction. As a primary source of information about the material structure, sample scattering was further investigated in more details. Energy selective measurements were performed in the cold energy range where the elastic coherent scattering dominates the total cross section, showing characteristic Bragg peaks. We report about results obtained with currently the highest energy and spatial resolutions achieved simultaneously. For selecting narrow energy bands from the initially polychromatic neutron beam, a double crystal monochromator was used, and as a detection system, a CCD camera system. It is shown that a detailed structure of the welded steel sample can be visualized and quantified directly from radiographs. With the maps of crystallographic orientations over the sample area of 2 x 2 cm and thickness of 11.2 mm, the detailed structure is energy resolved to the pixel size of the CCD camera chip of 10 m with the energy resolution (/) below 5 %.
CN2-2:

K.-I. Mochiki1, M. Okazaki2, Y. Otsuka2, K. Sugimoto3, H. Murakawa3, H. Asano3, N. Takenaka3 1 Nuclear Safety Eng., Tokyo City University, Tokyo, Japan 2 Computer Science, Tokyo City University, Tokyo, Japan 3 Mechanical Eng., Kobe University, Kobe, Japan We have developed a dynamic neutron CT system using a high performance neutron image intensifier. The neutron image intensifier has 4.5, 6 and 9 imaging fields. It has a Terbium activated Gd2O2S as the input-screen phosphor and an Europium activated Y2O2S multi-color scintillator as an output-screen. Its very bright characteristics are very useful for dynamic CT measurements. By combining this image intensifier with high speed CMOS cameras, we have demonstrated practical applications of dynamic CT. Tsystem consists of the neutron image intensifier, high speed CMOS cameras, image acquisition 455

A Dynamic Neutron CT System Using a Neutron Image Intensifier

boards and a personal computer with high performance HDDs. The personal computer has 6 high performance HDDs ( four HDDs for a RAID array and two for an operating system ) to achieve a long time dynamic CT measurement over 20 minutes with high data transfer rate of 262 MB/s which means 200 frames of 1280 x 1024 pixels per second. Experiments were made at JRR3M neutron radiography irradiation port (flux:1.2108n/cm2 s) of JAEA (Japan Atomic Energy Agency). There were three experimental conditions (A, B and C). In the condition A the COMS high speed camera of SV642M (PRIX) was used with the frame rate of 200 fps. The camera has 640 x 480 pixels and the pixel resolution was 212 m. 3D tomograms of a three stack fuel cell with 25 cm2 active area were obtained. In the condition B a high frame rate (max 500 fps) and high definition (1280 x 1024 pixels) CMOS camera of A504k (BASLER) was used for dynamic imaging of a sandglass over 3 minutes with the time scheme of one second of data taking 200 frames and five seconds of interval period. 3D tomograms of a sandglass were obtained. In the condition C as the diameter of the view area was changed to 4.5, the pixel resolution was improved to 106 m. However, the signal level decreased and frame rate was changed to 50. The system was developed to visualize dynamic phenomena of water in fuel cells .
The system was financially supported by New Energy and Industrial Technology Development Organization (NEDO). CN2-3:

A. S. Tremsin1, M. J. Mhlbauer2, B. Schillinger2, J. B. McPhate1, J. V. Vallerga1, O. H. W. Siegmund1, W. B. Feller3 1 Space Sciences Laboratory, University of California at Berkeley, Berkeley, USA 2 Forschungs-Neutronenquelle FRM-II, Technische Universitt Mnchen, Garching, Germany 3 NOVA Scientific, Inc., Sturbridge, USA The rapidly developing field of high resolution neutron radiography primarily concentrates on the non-destructive studies of stationary objects due to relatively long exposure times required to achieve an adequate neutron statistics. The combination of a high intensity neutron beam and a high timing and spatial resolution detector enables for investigation of dynamic processes in a stroboscopic mode, when the image frames can be synchronized with the studied object or acquired continuously at high acquisition frame rates. Although neutron statistics in the acquisition frames as short as 10 s is very low (typically 1000 n/cm2 at the sample), periodic processes can still be studied with high resolution by integrating a large number of frames synchronized to the process. In this paper we demonstrate the stroboscopic imaging capabilities of the highly collimated thermal neutron beamline ANTARES and a high resolution detector with neutron sensitive microchannel plates and Medipix2 readout. The dynamics of water uptake due to capillary forces as well as two- phase flow of air and water mixture was investigated, and stroboscopic imaging of an operational beam chopper and a spinning fan was performed with sub-100 m spatial resolution and acquisition frames varying between 10 s and 100 ms. The results of these experiments demonstrate the future possibilities to perform high resolution neutron radiography of fast and/or repetitive processes such as water flow and uptake, operation of fuel injection nozzles and many others.
CN2-4:

High Resolution Stroboscopic Neutron Radiography at the FRM-II ANTARES Facility

Status and Progress in Cold Neutron Imaging

A. P. Kaestner, E. H. Lehmann Spallation Neutron Source Division, Paul Scherrer Institute, CH-5232 Villigen-PSI, Switzerland Imaging with cold neutrons offers different imaging possibilities than thermal neutrons. We report the current status and the ongoing development of the ICON (Imaging with Cold Neutrons) beamline at Paul Scherrer Institute, Switzerland. Finally a collection of applications from the user community of ICON is given.

456

Author Index
Contributions from Collaborations
AGIPD-consortia AGIS collaboration ArDM Collaboration ARGO-YBJ Collaboration ATLAS Collaboration ATLAS Beam Conditions Monitor ATLAS Collaboration ATLAS Liquid Argon Calorimeter Group ATLAS Liquid Argon Calorimeter HECPAS Collaboration ATLAS Luminosity and Forward Physics working group ATLAS Muon Collaboration ATLAS Pixel Detector Collaboration ATLAS Planar Pixel R&D collaboration ATLAS planar pixel upgrade group ATLAS SCT ATLAS T/DAQ Collaboration ATLAS Tile calorimeter system AX-PET Collaboration BESIII Collaboration CALICE collaboration CALICE-UK DAQ groups CBM collaboration CDF II Silicon group ClearPEM Collaboration CMS BRM collaboration CMS Collaboration CMS Collaboration / RPC group CMS collaboration board CMS HCAL collaboration CMS pixel collaboration CMS RPC Collaboration CMS Tracker DQM group COMPASS Collaboration (CERN) D0 Collaboration Daya Bay Reactor Neutrino Collaboration Dayabay Collaboration DEPFET collaboration DES Collaboration Detector Group and Data Acquisition Systems Group DMTPC collaboration ECORAD collaboration N25-238 N13-67, N37-5 N25-153 N13-117 N23-4 N30-7 N10-5, N10-6, N13-77, N13-78, N13-173, N13-189, N17-4, N23-2, N25-254 N21-6, N43-5 N13-169 N16-3 N13-111, N16-1, N16-5, N34-7 N12-3 N13-203 N25-210 N12-2 N10-7 N25-248 M09-74 N13-145, N32-1, N45-1 N03-2, N05-3, N13-182, N13-190, N45-3 N42-7 N15-4 N12-6, N13-165 M10-4 N34-6 N13-76, N13-84, N13-88, N13-115, N13-175, N13-177, N13-185, N13-191, N13-204, N23-5, N43-4, N43-6, N45-5 N13-178 N13-184 N40-4 N12-4 N13-147 N13-118 N42-2 N08-3 N17-3 N25-240 N34-2 N07-1, N37-1 N25-26 N16-2 M13-84

457

EUDET Collaboration EUDET Consortium EXOTIC Collaboration FACT collaboration FCAL Collaboration FERMI LAT collaboration Fermilab Pixel R@D Group Focal Plane Instrumentation WP of the CTA Consortium Geant4-Nano5 Team GEM-TPC collaboration GERDA Collaboration GlueX Collaboration INFN-DASiPM2 collaboration INO Collaboration JointGEM Collaboration KM3NeT Consortium LANL-NIST Microcalorimeter Collaboration LCFI Collaboration LCTPC Collaboration LHCb collaboration LUX Collaboration MAGIC collaboration MEG Collaboration Minerva Collaboration Mohd. Moktar Bin Mat Nudin@Mat Nor NA62 Collaboration NEXT Open Science Grid PAMELA Collaboration PANDA collaboration PANDA EMC group PANDA-DIRC Group PHENEX Collaboration PHENIX Collaboration PHENIX Forward Upgrade Collaboration PHENIX Forward Upgrade Group PHENIX TEC/TRD group PoGOLite Collaboration RD50 Collaboration RD51 Collaboration S.M.A.R.T. Laboratory, Kansas State University Silicon Tracker group SiPBa group Slim5 Collaboration SNO Collaboration

N13-188 N45-2 N36-7 N35-2 N08-6 N13-59 N22-3 N40-5 N09-2 N15-2 N25-212 N30-5, N40-7 N19-2 N16-6 N13-195 N13-200, N25-43 N25-59 N13-193 N44-5 N21-5 N33-1 N35-1 N10-3 N36-6 M13-231 N13-205, N15-6, N21-2, N40-1 N16-4 N17-2 N37-2 N10-4 N25-94 N25-39 N13-68 N13-120, N13-162, N42-3 N15-7 N13-186 N13-122 N13-58 N08-1, N25-201 J05-2 N31-6 N02-1 M09-335 N13-172 N16-7

LANL/NIST/Star Cryoelectronics Microcalorimeter CollaborationN13-238

458

SOIPIX Collaboration STAR Collaboration T2K/FGD Collaboration The ATLAS Liquid Argon Calorimeter Group the EEE Collaboration TOTEM Collaboration TOTEM T2 Group VIPIX Collaboration

N25-205 N44-2 N15-5 N13-168, N13-171, N21-1 N13-56 N34-1 N44-4 N02-3, N21-7

Contributions from Individuals

A A. ILLAN, I. AALSETH, C. E. ABBA, A. ABBASPOUR TEHRANIFARD, A. ABDOLI, M. ABDUL RAHMAN, A. T. ABDULLAH, R. ABE, N. ABE, T. ABELLA, M. ABLER, D. ABLES, E. AB-RAHNI, A. ACCORSI, R. ACKERLY, T. ACKLEY, K. ADAMSON, P. E. ADLER, S. ADLER, V. ADRIAENS, A. AEZINIA, F. AFFOLDER, A. AFZAL, F. AGASTHYA, G. A. AGBEKO, N. AGELOU, M. AGNELLO, M. AGOSTEO, S. AGUIAR, P. AGUIRRE, J. AHLE, L. AHLUWALIA, G. AHMAD, M. AHMADIAN, A. AHN, S. AHN, Y. AHN, Y. B. AIHARA, H. AKAGI, T. AKATSUKA, T. AKGUN, U. AKHTAR, M. ALBANI, D. AL-BASHEER, A. K. M09-158 N25-74 M09-329, N13-25, N13-101, N13-102, N13-103, N13-112 M13-111 M05-184 N25-5 M13-231 N25-113, N25-114 N40-3 M05-397, M09-98 N43-2 N30-4 M04-6 M05-349 N25-1 N13-47 N25-147 M10-3 N13-118 CH2-3 N25-171 N08-4 M05-337 J05-8 M05-304, M09-215 N13-234 N25-68 N25-4 M05-151 M05-283, M13-285 N01-4, N25-122 N13-29 M05-316 M05-184 M07-4 J05-23 M09-86 N40-3 N39-4 M05-370 N13-191 M05-16 M13-120 N39-5 459

ALBERTO, B. ALDENER, M. ALEKSEJEVA, L. ALERS, G. ALESSIO, A. ALESSIO, A. M. ALIROL, O. AL-JARALLAH, M. I. ALLARD, D. ALLARD, M. ALLEKOTTE, I. ALLMENDINGER, T. ALLONGUE, B. ALLPORT, P. ALMEIDA, P. ALNOWAMI, M. ALOISIO, A. ALOUPI, E. ALURI, S. ALVAREZ, C. ALVAREZ, I. AMANO, H. AMBERS, S. D. AMBROSI, G. AMELI, F. AMERIO, S. AMIR, O. AMMAN, M. AMORINI, F. AMOROS, C. AMRAM, N. AN, S. AN, S. J. ANDERSEN, F. ANDERSON, J. T. ANDERSON, T. ANDERSSON, H. ANDERTON, L. ANDREAZZA, A. ANDREINI, K. ANDREYEV, A. ANDRICEK, L. ANDRITSCHKE, R. ANFRE, P. ANGELI, S. ANGERER, H. ANGHINOLFI, F. ANKERHOLD, U. ANNOVI, A. ANOKHIN, I. E. ANSORGE, R. E. ANTILOGUS, P. ANTON, G. ANTONACCI, A. G. ANTONIOU, A. ANTONUK, L. E. ANTUNES NOBREGA, R. ANZALONE, A. ANZELLOTTI, D. AOKI, K. AOKI, M. AONO, H.

M04-8 N25-60 N25-193 N25-156 M06-6, M09-17 M03-7 J05-19 N25-73 N13-69 M13-186 N13-69 M12-4 N25-28 N08-4 M05-307 M04-6 N13-14, N13-107, N21-4, N25-246 CH1-4 M12-1 N13-69 M02-5, M05-157, M09-155 M05-271 N25-64 N25-141 N13-107 N21-3 M13-366 N13-57, N13-70, N13-217, N14-2, N18-2, N24-3, N37-3 N25-66 N13-53, N35-6 N13-183 M05-79, M05-106, M13-69, M13-108, M13-339 M05-364 M09-224, M13-225 N10-2 M13-171 N13-82 HPP-12 N13-86 N25-176, N25-177 HP3-4, M07-3 N13-35, N22-1, N25-199, N34-2, N35-4 N13-64, N33-5 N25-87, N25-88 M09-323 N25-249, N42-4 N13-39 N25-184 HPP-10, N21-3 J02-7 HP3-3, M09-167 N13-54 J05-3, M13-42, M13-345, N25-3, N25-184, N25-185, N25-191, N25-208 N03-6 M05-301 J05-17 N13-180 N25-66 N13-86 M09-401 M09-251 M15-4 460

APOSTOLAKIS, J. ARACENA, I. ARAI, Y. ARAKAWA, K. ARCE, P. ARJONA, J. ARLICOT, N. ARLT, R. ARMSTRONG, A. ARMSTRONG, B. L. ARNABOLDI, C. ARROWOOD, L. F. ARTUSO, M. ARUTINOV, D. ASAI, K. T. ASAMI, F. ASANO, H. ASANO, T. ASCHAUER, F. ASHUROV, M. ASK, S. ASLANI, I. ASMA, E. ASO, T. ASOREY, H. ASPINALL, M. ASSANE, A. -A. I. ASTNER, S. ASTON, D. ASZTALOS, S. ATHANASIADES, A. ATRE, A. ATTANASI, F. AUCOTT, T. AUFFRAY, E. AUFFRAY, E. E. AUGELLI, M. AUGUSTINE, F. L. AURICCHIO, N. AVDEICHIKOV, V. AWADALLA, S. AY, M. R. AYAN, A. S. AYOUB, M. AZEVEDO, C. D. AZMOUN, B. AZZARELLO, P. B BABA, J. S. BABAPOUR MOFRAD, F. BABLA, H. BACHARACH, S. BACIAK, J. E. BACRANIA, M. K. BADAL, A. BADANO, A. BADAWI, R. D. BAE, J. B. BAE, J.

N13-89 N10-6 N22-2 M15-4 M10-6 M09-173 M13-144 N14-6 N29-5 N25-106 N13-42, N25-117 HPP-5 N25-117 N12-5, N13-179, N34-4 N13-66 N11-5, N13-158 CN2-2 J02-5 N24-5, N35-4 N13-163 N25-244 CH1-4 M02-6 N39-4 N13-69 N13-129 N13-30 M13-33 N45-7 N13-119 N06-4, N20-3 M05-214 M09-332, M13-90 N13-137 N43-2 N24-7 N09-3, N13-79, N33-4 N04-6, N25-131 M13-12, N13-65 M09-119 N14-2, N25-178, N25-219, N31-4 M05-184, M05-361, M09-308 M05-349 N25-232, N25-233, N25-234 N13-160, N13-161 N15-1 N25-141

M05-13, M09-347 M13-111 M13-81 M09-338 M13-240, N25-188 N13-238, N25-42, N25-58, N25-89, N25-145 HP3-1 HP3-1 M04-2, M13-51, M13-72, M13-348 N25-202 N25-148 461

BAEK, C. -H. M05-364, M09-143, M13-117, N25-51 BAER, M. HP1-2, HP1-3 M05-277 BAESSO, P. BAGAGLI, F. N39-1 BAGHAEI, H. M05-79, M05-106, M13-69, M13-108 BAI, B. M05-193 BAI, C. M05-19, M13-81 BAILEY, A. M13-168 BAILEY, S. N13-54 BAIOCCHI, M. M05-331, M09-92 BAKER, J. L. N41-7 N34-5 BAKER, O. K. BAKSHI, A. K. N25-47 BAL, G. M13-321 M09-26 BALAKRISHNAN, K. N25-35 BALBUENA, J. BALDA, M. M05-43 BALDAZZI, G. M05-391 BALDINI, W. N13-176 BALDOCK, C. M09-53 J05-21 BALDONI, G. BALKAY, L. J05-12, M09-152 BALL, M. N16-4 N25-33 BALL, R. W. BALLABRIGA, R. N19-7 BALLIZANY, R. J04-1, N28-5 M07-8 BALTES, C. BALUKRISHNA, S. M13-237 J05-7 BALUTI, F. BAN, G. N25-107 BANCROFT, C. N25-22 N13-57, N13-70, N37-3 BANDSTRA, M. S. BANERJEE, S. N04-4, N30-3 BANI-HASHEMI, A. R. J03-4, M08-5 BANZUZI, K. N13-82 BAO, Q. M05-172, M09-299, M15-7 BAPTISTA BONIFACIO, D. A. N25-102 BARALDI, P. N13-128 N13-152 BARATA, J. A. S. BARBARINO, G. N28-6 BARBER, H. B. M15-5 M08-8, M13-45 BARBER, W. C. BARBERIS, D. N13-86 BARBERO, M. N12-5, N13-179, N34-4 N22-4, N25-123 BARBIER, R. BARD, P. N13-114 BARDIES, M. M09-344 BARKAN, S. N25-180 BARKER, W. C. HPP-3, HPP-4 BARNETT, D. N38-2 N13-207, N14-1 BARNETT, D. S. BARONCELLI, A. N13-153 N42-1 BARR, D. S. BARRET, D. N35-6 BARRETT, H. H. HPP-14, HP2-5, J03-3, M09-269, M09-350, M15-5 M13-186 BARRETT, O. BARRILLON, P. M05-91 BARRIO, J. R. M13-378 BARROS, H. N13-69 BARROSO, R. C. M09-371, M09-398 BARTL, P. J05-3, M13-42 BARTON, P. J. N25-79 462

BARTRAM, R. H. BARZILOV, A. BASAGLIA, T. BASHKIROV, V. A. BASILE, M. BASS, R. B. BASU, A. BATDORF, M. T. BATES, R. L. BATI, M. BATSCH, T. BATTACHARYA, P. BATTAGLIA, M. BATTISTON, R. BAUDOT, J. BAUMBAUGH, B. BAYDJANOV, M. BAZZACCO, D. BEALL, J. A. BEAU, J. O. BEAUDOIN, J. -F. BEAULIEU, D. BECCHERLE, R. BECHE, J. -F. BECK, M. BECK, T. J. BECKER, J. BEDNAREK, D. R. BEDOGNI, R. BEEKMAN, F. J. BEGALLI, M. BEGHAEI, H. BEHARI, S. BEILICKE, M. BEIMFORDE, M. BEISTER, M. BELAISH, I. BELCARI, N. BELL, A. J. BELL, P. BELL, Z. BELL, Z. W. BELLI, F. BELLI, G. BELLINGER, S. L. BELLM, E. C. BELLOTTI, E. BELLOTTI, R. BENCIVENNI, G. BENDRIEM, B. BENHAMMOU, Y. BENLLOCH, J. M. BENNATI, P. BENNER, T. BENNET, D. A. BENNETT, B. L. BENNETT, D. BENNETT, D. A. BENOIT, D. BENOIT, LOUNIS,DINU, M. BENSIMON, C. BENSON, T.

N25-160, N25-161 N13-215 N09-7 N13-144 N21-3 N33-7 N25-231, N25-232, N25-233, N25-234 N13-207, N14-1 N08-1, N38-2 M13-234 N13-220 N01-1 CH2-4, N22-7 N25-141 N22-6, N25-123, N25-203 N13-199, N25-134, N25-142 N13-163 N25-196, N25-247 N25-42 N25-215 M15-8 N29-2 N12-5, N13-179 N13-13 N13-94 M08-8 N25-239 M08-1 N33-6 J01-4, J03-1, M06-2, M06-3, M09-68, M13-264 N09-3, N13-79, N13-91, N13-95, N39-7 M13-339 N12-6 N25-204, N31-2 N25-199 HPP-11 N13-81 M05-91, M09-332, M13-90, N25-102 N13-198 N13-124 N20-4 N09-7 N13-170 M05-277 N14-4, N30-6 N13-57, N13-70, N37-3 N25-213 N39-1 N44-1 M09-89 N13-183 M05-10, M09-149 M13-279, N25-126, N26-2 M04-1 N25-58 N25-30, N25-89 M09-257 N25-42 M09-254 N25-210 M05-142 M12-1 463

BENTOURKIA, M. BENTOUTOU, Y. BENUSSI, L. BENZAID, D. BEN-ZVI, I. BEQUE, D. BERDERMANN, E. BERGAUER, T. BERGERON, M. BERGHOLZ, M. BERGLUND, E. BERGNER, F. BERKUS, T. BERMAN, A. BERNSTEIN, A. BERNY, R. BERRADJA, K. BERRETTA, M. BERRY, J. E. BERRY-PUSEY, B. N. BERTOLONE, G. BERTOU, X. BERTOZZI, W. BERTRAM, M. BETANCOURT, C. BETTINI, M. BETZEL, G. T. BEVERIDGE, T. E. BEVERINA, L. BHALEREO, V. BHATTACHARYA, M. BHEESETTE, S. BI, W. BIAGETTI, D. BIAGI, S. BIAN, J. BIANCHI, G. BIANCHINI, D. BILHEUX, H. Z. BILKI, B. BILLOINT, O. BILSKI, P. BINDA, M. BINDLEY, G. BINGHAM, P. R. BINGHAM, P. BINKLEY, M. BIONTA, R. M. BIRCH, J. BIRCHER, C. J. BIRUMACHI, A. BISELLO, D. BISOGNI, G. BISOGNI, M. G. BISTEKOS, E. BIZARRI, G. BIZARRI, G. A. BJEOUMIKHOV, A. BLACKADAR, J. BLACKBURN, B. BLACKSTON, M. A. BLAIR, M. W.

M05-379, M13-375 M09-179, N13-60 N13-147, N15-3 N25-186 N38-5 M13-321 N13-45 N25-159 J05-11, M15-8 N25-159 N12-2 M09-206, M13-210, M13-381 M13-207, M13-210 HP2-6 N04-1 M13-144 M05-379 HPP-10 N25-214 M15-7 N22-5, N22-6, N25-167, N25-203 N13-69 N04-2 M08-3 N08-2, N08-4, N25-156, N25-158, N25-181 N21-3 J05-7 M05-118, M09-137 N28-7 N31-1 M05-190 N16-6 N13-216, N13-240 N25-183 N11-4 M09-386, M12-5, M13-3 N25-243 M05-391 N25-25 N13-197 N13-22, N13-24 N04-5 N13-112, N28-7 N14-2, N25-176, N25-177, N25-178, N25-219 CN1-2, N25-25 HPP-5 N13-94 N30-4 M05-130 M05-31, M13-87 N25-19 N22-7 N25-139 J01-2, M05-91 CH1-4 N01-4, N25-140 N25-122, N25-125 J02-1 N14-6 N04-3 CN1-2, N26-5 N25-30 464

BLALOCK, B. J. BLANCHOT, G. BLASI, N. BLASY, B. D. BLINDER, S. BLISS, M. BLM, P. BLOCKX, I. BLOOMFIELD, P. M. BLOSER, P. F. BLUME, M. BLUMENTHAL, D. BOATNER, L. BOATNER, L. A. BOCA, G. BOCCACCIO, P. BOCCI, V. BOCCONE, V. BODEK, K. BOEHMER, F. V. BOEHNEL, M. BOELLAARD, R. BOER, W. D. BOERNERT, P. BOESE, J. BOGDAN, M. BOGGS, S. E. BOGOVAC, M. BOHM, C. BOIANO, C. BOLAND, M. J. BOLANOS, L. BOLCH, W. BOLOTNIKOV, A. BOLOTNIKOV, A. E. BOMBELLI, L. BONDIL-BLIN, S. BONISSENT, A. BONTUS, C. BOONE, J. BOOTE, J. BORADE, R. B. BORGERT, J. BORGES, F. I. G. M. BOROZDIN, K. BOROZDIN, K. N. BORRAS, K. BORSDORF, A. BORSHCHOV, V. BORTOLATO, D. BOSCARDIN, M. BOSCO, P. BOSIN, A. BOSISIO, L. BOSKO, A. BOSTON, A. J. BOSTON, H. C. BOTTA, E. BOUCHER, Y. BOUCHER, Y. A. BOUDJEMLINE, K. BOUDOU, C.

N03-6, N30-6 N25-28 N25-69, N25-70 N29-4 M05-133, M13-99 N20-4 N25-159 M13-159 M05-133 N25-22 M05-199 N14-6 N01-1 N25-106, N25-135 N15-3 M13-279, N25-126, N26-2 N13-180 N25-153 N25-107 N44-3 N25-184 M05-4 N25-159 M11-6 M03-1 N27-5 N13-57, N13-70, N18-1, N33-2, N37-3 CH1-4 M11-4, N13-121, N25-252, N25-253 N25-65, N25-66 N13-4, N13-5, N38-3 N39-1 N39-5 M15-2, N25-209, N25-221 N31-4 N19-4 M05-91 HPP-9 M15-3 M13-3 N13-94 N25-41 M15-3 N13-151, N13-157 N13-245 N13-239 N25-159 M02-2 J05-10 N25-71 N02-4 N39-1 N13-72 N02-4 N25-56 J05-16, N13-129 J05-16, N13-129 N25-68 N25-223 N25-166 N16-7 N25-211 465

BOUMAN, C. BOUNOUA, A. BOURLAND, J. D. BOURRET-COURCHESNE, E. D. BOUSSELHAM, A. BOUSSION, N. BOUTCHKO, R. BOUVIER, J. BOWDEN, N. BOWEN, S. L. BOWSHER, J. E. BOWYER, T. W. BOYARINTSEVA, Y. A. BOYCE, N. O. BOYD, E. BRADLEY, D. A. BRADLEY, E. C. BRADY, S. M. BRAESS, H. BRAEUNINGER, H. BRAGA, D. BRAGHIERI, A. BRAGINSKY, D. BRAMBILLA, A. BRAMBILLA, S. BRANCHINI, P. BRAND, J. I. BRANDYS, I. BRANKOV, J. G. BRASSE, D. BRAVAR, U. BRAVIN, A. BRAVIN, E. BRAZ, D. BRAZ, D. C. BRECHER, C. BREDESON, G. BREEDING, J. E. BRENDEL, B. J. BRENNAN, J. BRESKIN, A. BRESSANI, T. BREUS, D. BRIEL, U. BRITTON, C. L. BRITTON, J. L. BROCK, I. BROENNIMANN, C. BROGNA, A. BROOKS, M. BROWN, A. BROWN, B. BROWN, H. BROWN, J. M. C. BROWN, L. O. BROWN, T. BRUALLA, L. BRUBAKER, E. BRUER-KRISCH, E. BRUNAO, S. BRUNET, C. -A.

M13-201 M09-179 M13-204 N25-41, N25-140, N25-143, N25-150 J03-3 M13-186 J01-6, M09-377, N25-150 N13-24 N04-1, N13-212 M13-51 M09-101 N25-60 N25-91 N20-7 N13-94 N25-5 N06-3 M09-374 M05-238, N32-6 N36-3 N27-4 N15-3 N25-207 N25-211 N25-71, N36-5 N13-14, N13-153 N29-4 N13-226 M05-286, M09-173, M14-2 N13-114, N25-101 N25-22 N39-6 N13-13 M09-371 M09-398 J05-14, J05-21 N08-2, N08-4, N25-158 M09-104 M08-6 N13-231, N13-236, N25-20, N29-1 N25-38 N25-68 N13-119 N33-3 N03-6, N30-6 N03-6 N11-1 N25-9, N25-174 N22-6, N25-203 N03-7 N13-94 N13-94 N08-4 M05-118, M09-137 N25-30 HPP-12 J02-6 N13-231, N29-1 N39-6 M04-8 M13-354 466

BRUNETTI, A. BRUNNER, S. S. BRUSCHI, M. BRUYNDONCKX, P. BUCCIANTONIO, M. BUCHHOLZ, G. BCHLER, A. BUCKLEY, S. BUDDEN, B. BUDINGER, T. F. BUDTZ-JRGENSEN, C. BUDZANOWSKI, M. BUECHERL, T. J. BUERGER, C. BUETHER, F. BUFALINO, S. BUGALHO, R. BUKALA, D. BUKKI, T. BULYCHEVA, A. BUNCH, S. C. BUNDSCHUH, R. A. BUONOMO, B. BURATO, G. BURBAR, Z. BURCHILL, W. E. BURDETTE, D. BUREAU-OXTON, C. BURGER, A. BURGER, J. BURGGRAF, L. W. BURGHARD, B. J. BUSBY, R. BUSCA, P. BUSO, S. BUTHER, F. BUTLER, A. P. BUTLER, J. BUTLER, P. H. BUTSYK, S. A. BUVAT, I. BUZUG, T. M. BYARS, L. G. BYNUM, K. D. BYRD, J. BYRUM, K. C CABELLO, J. CABIDO, R. CABRERA, A. CADINI, F. CADORETTE, J. CAI, J. CAI, W. CAIAFA, A. CALABRESE, R. CALAMINNUS, C. CALCAGNILE, L. CAL-GONZALEZ, J.

M05-370 N24-7 N25-68 M05-67, M05-85, N25-40 N21-3 N25-184 N02-1 M10-5 N13-132 J01-6 N13-62 N04-5 CN1-3 M09-41 M09-38, M13-324 N25-68 M05-307 M11-5 HP2-4, M05-274, M09-152 N31-7 N03-6 M09-197 N33-6 N25-87, N25-88 M05-217, M13-198 NP-1 J05-10, M09-83, M13-120 M15-8 N01-1, N25-204, N31-2 M13-234 N25-147 N13-207, N14-1 N38-5 N26-3 N25-28 M04-7 N25-12 N13-197 N25-12 N13-120 M05-76, M09-344, M13-357 M05-160 M07-2 N06-2 N13-13 N10-2

M05-346, M13-168 HP2-3 N13-123 N13-128 M15-8 N13-10 M05-166 M12-6 N13-176 M11-5 CH1-3 M05-226 467

CALICE, CALICE, C. CALLIER, S. CALVI, M. CALVO DEL CASTILLO, H. CAMARDA, G. CAMARDA, G. S. CAMARLINGHI, N. CAMERA, F. CAMPANA, S. CAMPBELL, M. CANADAS, M. CANADAS CASTRO, M. CANDELARIA, G. CANESSA, E. CANEVARI, C. CANNING, A. CANTWELL, B. J. CAO, X. CAO, Y. CAPASSO, L. CAPEANS, M. CAPPUZZELLO, F. CARDELLA, G. CARDENAS, E. S. CARDOSO, B. CARDUNER, H. CAREY, D. CARINI, G. A. CARINI, G. CARITA, L. P. M. M. CARLES, M. CARLIER, T. CARLINO, G. CAROLAN, M. G. CAROLI, E. CARPENTIERI, C. CARPINELLI, M. CARR, C. CARRAMATE, L. F. CARREL, F. CARSON, R. E. CARTER, C. CARTY, M. CASE, G. L. CASEY, M. E. CASSE, G. CASSOLA, V. F. CASSOL-BRUNNER, F. CASTALDI, R. CASTIGLIONI, I. CASTILLO, M. CASTILLO, V. CASTOLDI, A. CASTRO, E. CATALDO, R. CATANA, C. CATANZARITI, E. CATASTINI, P. CATES, J. W. CATTADORI, C. CAUCCI, L.

N13-194 N45-4 N03-1 N25-117 CH1-2 N25-209, N25-221 N31-4 N39-1 N13-128, N25-69, N25-70, N25-71, N25-247, N36-5 N13-86 N19-7 M10-6 M05-103 M13-171 N13-92 M09-110 N25-143, N25-150 N25-231 M13-228 M12-6 N13-107, N25-246 N11-6 N25-65 N25-66 N13-218, N13-225 N25-34 M13-144, N25-38 N25-128 N25-224, N38-4 N19-3, N25-218 N25-124 M09-149 M09-344 N13-86 M09-236 N13-65 N39-1 N13-72 N13-227 J04-7 N13-234 M03-3, M05-154, M09-161 N25-235 N13-53 N13-132 M05-292, M09-89, M13-123 N08-4 N39-3 M09-254 N25-243 M05-181, M09-110 N13-69 N13-113 J02-1, N02-2, N13-74, N25-2, N25-66 N13-198 N39-1 M04-1, M11-3 N39-1 N21-3 N25-120 N13-37, N13-43, N25-213 HP2-5 468

CAVALLARO, M. CAVALLARO, S. CAVALLI, A. CAZZANTI, L. CEBULA, A. T. CECCHI, C. CECCUCCI, A. CEDERSTRM, B. CEDERWALL, B. CELESTRE, R. S. CELLER, A. CENCELLI, V. O. CENNI, J. CERELLO, P. CERIONI, S. CERVENA, J. CESAREO, R. CEVENINI, F. CHA, B. K. CHA, B. CHABANAT, E. CHABRE, A. CHAGANI, H. CHAKOUMAKOS, B. C. CHALON, S. CHAMORRO, J. CHAN, C. CHAN, Y. D. CHAN, Y. -D. CHANDY, K. M. CHANG, C. CHANG, H. -K. CHANG, W. CHANG, X. CHANG, Y. -H. CHANG, Y. CHAPMAN, E. CHAPMAN, E. C. CHAPMAN, J. W. CHARLES-EDWARDS, G. CHARON, Y. CHASSAING, C. CHATTERJEE, S. CHATTERJI, S. CHATZIIOANNOU, A. F. CHATZIIOANNOU, A. CHAUDHARI, A. J. CHAUDHRY, A. CHAVES, R. CHAVEZ, R. CHEKHTMAN, A. CHEN, C. -H. CHEN, C. -T. CHEN, F. CHEN, H. CHEN, J. -C. CHEN, K. -H. CHEN, M. CHEN, S. M. CHEN, S. CHEN, W. CHEN, Y. P.

N25-65 N25-66 N23-6 N13-125 M13-54 N43-3 N03-4, N13-41 M14-4 M11-4 CH2-4 HP3-4, M05-295, M05-313, M07-3, M09-164, M09-209, M13-192 N13-131 N21-3 N39-1 N44-1 N25-75 M05-370 N25-246 N25-108 N25-148 N25-123 N25-76 N25-118, N25-136 N25-135 M13-144 M05-283 M05-166 N24-3 N24-1 N06-7 M05-373 N13-57, N13-70, N37-3 M05-58, M09-71, M13-93 N38-5 N13-57, N13-70, N37-3 N13-181 N04-6, N13-237 N25-131 N25-33 M09-248 M05-25, M09-242 N25-215 N25-47 N25-187, N25-197 M05-172, M09-50, M13-372, M15-7 M09-299 M04-2 N25-143, N25-150 M02-5, M05-157, M05-175, M09-155 M09-158 N13-59 N25-6 M05-7, M05-100, M09-80 M09-260 N14-2, N25-176, N25-177, N25-178, N25-219, N31-4, N34-5 M13-300 N13-51 M13-123, M13-132 N25-245 J01-1, M11-2, M13-312, N07-2 HPP-11, N19-3, N25-218, N25-224, N38-4 N05-1 469

CHEN, Y. CHEN, Z. CHENE, G. CHENG, J. CHENG, J. -C. CHENG, J. -C. (. CHENG, K. CHENG, M. -H. CHENG, X. CHENG-LIAO, J. CHEON, G. J. CHEON, G. -J. CHERAN, S. C. CHEREPI, N. CHEREPY, N. CHEREPY, N. J. CHERIYADAT, A. CHERLIN, A. CHERRY, M. L. CHERRY, S. CHERRY, S. R. CHERYAUKA, A. CHESI, E. CHEUNG, C. L. CHEVALIER, D. E. CHEZE LE REST, C. CHI, C. -Y. CHIANG, S. -H. W. CHICHESTER, D. L. CHILDRES, I. CHINCARINI, A. CHIPAUX, R. CHIRINOS, J. M. CHIU, J. -L. CHIVERS, D. H. CHIVERS, D. CHO, G. CHO, H. CHO, I. -C. CHO, M. K. CHO, S. CHO, S. Y. CHO, Y. H. CHOI, C. I. CHOI, C. W. CHOI, G. S. CHOI, I. W. CHOI, J. -H. CHOI, Y. CHOONG, W. -S. CHOU, H. P. CHOU, H. -P. CHOURASIYA, G. CHOW, K. CHOWDHURY, P. CHOY, S. CHRISTE, S. CHRISTIAN, J. CHRISTIAN, J. F. CHRISTOPHERSEN, M. CHU, Y. CHUN, R.

N07-7, N13-143 J04-3, J05-20, M05-394, M09-221, M13-222, N13-211, N13-216, N13-240 CH1-2 N13-140, N13-211, N25-48 M09-215 M05-304 M09-257 N13-181 N25-242 M03-5 M05-97, M05-115, N25-14 M13-180 N39-1 N25-122 N01-1, N01-4, N01-6 N25-149 N06-3 N25-207 N13-132 M05-268, M13-129 M04-2, M05-61, M05-328, M09-125, M09-134, M13-60, M13-72, M15-1 HPP-12 J05-10, M09-83, M13-120 N02-6 HP2-1 M05-46, M13-402 N15-1 N13-50, N25-11 N13-150, N13-214, N13-232, N13-242 N05-1 N39-1 N33-3 N13-69 N13-57, N13-70, N37-3 N13-136, N18-5, N25-225, N25-226 N13-135, N13-137, N18-2 N25-108, N25-148 N25-36 N25-6 M05-232 M05-22, M07-4, M09-185, M11-3, M12-5 N25-162 N25-227, N25-228 N13-8 N25-14 N13-148 N13-8 N25-127 J05-23, M05-334, M09-8, M09-86, M13-27, M13-117, M13-258, M13-261, M13-336, N25-51 M05-7, M09-80, M09-275, M10-5, N01-4, N25-122, N25-129 N13-20, N25-45 N13-19 N25-47 N13-13 N36-2 M05-373 N13-61 M05-268, M13-60, N25-86, N25-137 N04-6, N13-237, N25-131, N36-1 N05-5, N25-175 N13-73 N03-6 470

CHUNG, Y. H. CHUNG, Y. CHURILOV, A. CIAMPI, G. CIBINETTO, G. CINDRO, V. CINTI, M. N. CINTI, M. CIOBANU, M. CIOCCA, C. CIODARO XAVIER, T. CIPOLLA, V. CIRIGNANO, L. CISBANI, E. CITRON, Z. CLACKDOYLE, R. CLARK, G. CLARKE, S. D. CLARYSSE, P. CLASIE, B. J. CLAUS, G. CLAUSSEN, C. D. CLECKLER, J. CLEMENT, J. F. CLINTHORNE, N. CLINTHORNE, N. H. CLUZEL, R. CLYNE, M. N. COATH, R. COCHRAN, E. COCO, B. J. COFFER, A. COHEN, D. COHEN, Y. COHEN-ZADA, I. COLBY, B. COLILLI, S. COLLAZUOL, G. COLLEDANI, C. COLLINS, J. COLOM, R. J. COMBET, M. P. COMTAT, C. CONDE, C. A. N. CONFORTI, S. CONLIN, K. E. CONTARATO, D. CONTI, M. CONWAY, A. M. CONWAY, M. R. CONWELL, R. L. COOK, E. COOK, J. M. COOK, N. J. COOK, W. R. COOL, S. COOPER, G. M. COOPER, M. W. COOPER, R. J. COOPER, R. G. COPPOLA, T. CORRIN, E. P.

M05-364, M09-8, M09-143, M13-117, N25-51 M05-334, M13-336 N01-5, N25-230 N25-230 N13-176 J05-10, M09-83, M13-234, N08-4, N25-206 N25-126, N26-2 M13-279 N13-44, N13-45 N13-86 N25-248 N13-53 N25-230 M05-331, M09-92 N15-1 M09-317 N06-7 N04-4, N13-244, N25-57, N30-3 M05-235 J02-7 N22-6, N25-203 M05-259 N25-134 M05-46, M13-402 HPP-13, M05-367, M13-120 J05-10, M09-83 N22-4 N25-12 N25-31 J05-10, M09-83, M13-120 M13-147 N13-137 N13-247 J04-4, N05-6, N13-81, N13-226, N13-230, N25-133 N25-133 N08-4, N25-158 M09-92 M05-91, N10-1, N25-139 N22-5, N22-6, N25-203 N13-27 M05-10 N13-141, N25-67 M05-211, M09-122 N13-151, N13-152, N13-157, N25-124 N19-1 N06-6, N13-249 N22-7 M05-70, M09-89, M13-156 N02-6, N25-229 M09-236 M05-19, M13-81 N13-206 N25-30, N25-89 N25-12 N14-5, N31-1 J05-14, M09-14 N42-1 N25-60 N13-129 N25-130 M09-392 N13-188 471

CORTIANA, G. CORTINA, E. COSTA, F. COSTANZA, S. COSTIN, M. COTTA RAMUSINO, A. COTTE, M. COUCEIRO, M. COUTURE, A. J. COWEE, M. COWIE, E. N. CRESCIOLI, F. CRESPI, F. C. L. CRESPO, P. CRESSLER, J. D. CRESSWELL, J. R. CRESTANI, R. CRIPPS, N. A. CROCE, M. P. CROOKS, J. CROTEAU, E. CROW, L. CROWELL, A. S. CROWLEY, E. M. CSERKASZKY, A. CUDDY, S. CUELLAR, L. CUEVAS, C. CUI, J. CUI, Y. CUI, Y. -G. CUNNANE, S. CUNNINGHAM, M. F. CUNNINGHAM, N. CUNSOLO, A. CUPLOV, V. CURADO DA SILVA, R. M. CUSANNO, F. CUSSONNEAU, J. -P. CUSTELCEAN, R. CUTAJAR, D. CUTLER, P. A. CUTLER, S. J. CZARNACKI, W. CZIFRUS, S. D DA SILVA, A. M. DA VIA, C. D'ABRAMO, E. DABROWSKI, W. DAHLBOM, M. DAI, Q. DAI, T. DALLA BETTA, G. -F. DALPRA, S. DALY, F. D'AMICO, A. DAMRON, E. D'ANDRAGORA, A.

N23-4 N25-192 N13-116 N15-3 M13-309 N03-4 CH2-2 M09-353 N25-115 N26-4 N05-2 N21-3 N25-71, N36-5 M09-353 N13-48 J05-16, N13-129 N25-178 N13-204 N25-89, N25-145 N25-31 M13-375 N03-6, N25-130 J01-3 M05-229 HPP-1 M05-280 N13-239, N13-245 N25-241 N25-176, N25-177 N25-209, N25-221, N31-4 M15-2 M13-375 N06-3 N04-4, N30-3 N25-65 N13-177 N13-65 M05-331, M09-92 N25-38 N25-135 J05-6, M09-236 N25-128 M13-297 N25-92, N25-103 HPP-1

J04-7 N13-202, N25-182 N13-18, N13-33, N13-131, N26-2 N13-39 M05-49 M13-150 M05-136, N13-183 N02-4 N23-6 N13-53 N13-107 N25-180 N13-37, N13-43, N25-213 472

DANG, J. DANIELSSON, H. DANIELSSON, M. DANON, Y. DARAMBARA, D. G. DARBO, G. DARDENNES, C. DAS, A. DATTA, A. DAVID, C. DAVID, E. DAVID, S. DAWOOD, M. DAWSON, N. DAY, A. R. DAZELEY, S. DE ASMUNDIS, R. DE BEENHOUWER, J. DE CARLO, F. DE CASTRO, A. DE CASTRO, I. F. DE FILIPPO, E. DE GERONIMO, G. DE GRUYTER, R. DE LA TAILLE, C. DE LEDESMA, S. DE LENTDECKER, G. W. P. DE LIMA, D. E. F. DE LUCIA, E. DE MAN, B. DE MASI, R. DE MATTIA, M. DE MESQUITA, C. H. DE MITRI, I. DE NAPOLI, M. DE NOLFO, G. A. DE NOTARISTEFANI, F. DE NUNZIO, G. DE OLIVEIRA, R. DE ROBERTIS, G. DE ROSA, G. DE ROUFFIGNAC, P. DE SALVO, A. DE SEIXAS, J. M. DE SOUZA, W. P. DE SOUZA SANTOS, D. DE VINCENTIS, G. DE VITA, G. DEERLI, Y. DECKER, K. DECLERCK, J. DEFEYT, C. DEFRISE, M. DEGENHARDT, C. DEGERLI, Y. DEGERONIMO, G. DEL GUERRA, A. DEL SESTO, R. DEL SESTO, R. E. DEL SORDO, S. DELAGNES, E. DELATAILLE, C.

M09-59 N13-205 M14-4 N13-11, N30-1 J03-7 N12-5, N13-179 N14-3 M05-190 N25-220 M08-2 N11-1 M13-186 M04-7, M09-38, M13-324 N08-2, N08-4, N25-158 N25-74 N04-1 N28-6 M13-252 N39-1 N13-69 N25-152 N25-66 M05-109, N13-47, N19-3, N25-218, N25-224 J04-1, N28-5 M05-91, N03-1, N19-1 N13-94 N44-5 N13-90 N44-1 M12-6, M13-201 N22-6, N25-203 N13-84 N25-146 N39-1 N25-68 N20-5 M13-279, N13-18, N13-33, N13-131, N25-126, N26-2 N39-1 N11-1, N13-141, N25-67 N05-4, N44-1 N28-6 N29-2 N13-86 N13-90 N13-93 N09-3, N39-7 M05-331, M09-92, N25-126 N19-4, N24-5 N25-203 N25-235 M05-292, M13-177, M13-183 CH1-2 M09-317 J04-1, N28-5 N22-5, N22-6, N25-167 M15-2 J01-2, M05-91, M09-332, M13-90, N25-102, N25-139 N29-3 N25-30, N25-115 N13-65 N13-21 N13-54 473

DELEYE, S. D'ELIA, M. DELLA PIETRA, M. DELLACASA, G. DELL'AGNELLO, L. DELL'ORSO, M. DELSO, G. DEMARCHI, D. DENDOOVEN, P. DENDOOVEN, P. G. DENES, P. DENG, J. DENG, X. DENG, Z. DENNERLEIN, F. DENNEY, T. S. DEO, N. DEPASSE, P. DEPPE, H. DEPPNER, I. DEPTUCH, G. W. DERENZO, S. E. DERZON, M. S. DESCHLER-ERB, E. DESCO, M. DESCOURT, P. DESPEISSE, M. M. DEVADHAS, D. DEVITA, G. DEVITO, D. M. DEVOL, T. A. DEWARAJA, Y. K. DEWONCK, S. DEY, J. DHAH, H. K. DHAWAN, S. K. DHILLON, I. DI DOMENICO, G. DI GIROLAMO, A. DI LUISE, S. DI VACRI, A. DIAS, T. H. V. T. DIAZ, D. G. DIAZ, T. DIECKMANN, W. DIEHL, E. DIERLAMM, A. DIERRE, F. DIEZB, S. DIFILIPPO, F. P. DIKAIOS, N. DILORENZO, G. DIMITROV, D. DIMMOCK, M. R. DINAPOLI, R. DINARDO, M. E. DINELLE, K. DINU, N. DIOSZEGI, I. DISSERTORI, G. DISTEL, J. DIXIT, N.

M05-148 CH1-3 N25-246 N03-4, N13-41 N13-86, N17-1, N23-6 N21-3 M09-197 N24-4 J03-1, M06-2 J01-4 CH2-4, N13-13, N22-7 M13-132 M05-112, M05-121 N13-31 M03-1, M03-6 M05-286 N02-6 N25-123 N13-15 N13-140 N27-1 M05-64, N25-41, N25-125, N25-129, N25-140, N25-143, N25-150 N25-16 CH2-1 M05-226, M05-283, M05-397, M09-98, M13-285 M09-254, M09-344 N24-7 M13-237 N35-4 N25-106 N25-55 M04-3 N25-87, N25-88 M05-55 M13-57 N34-5 M03-2 M13-12 N13-86 N13-153, N13-159, N16-1 N13-43, N25-213 N13-157 N13-140 N13-3 HPP-4, HPP-3 N13-183 N25-159 N25-232, N25-233, N25-234 N13-48 M09-113 M05-253 N25-141 N38-5 M05-118, M09-137, N13-129 N13-17 N12-4 M13-99 M05-91 N13-233 N32-3 N38-5 M13-213 474

DO CARMO, S. J. C. DOERING, D. DOERING, W. M. DOESBURG, R. M. N. DOKE, T. DOKHALE, P. DOKHALE, P. A. DOLAN, J. L. DOLBNY, I. DOLENEC, R. DOLEV, E. DOLEZAL, B. DOMENICI, D. DOMIGAN, P. DOMINJON, A. DOMONKOS, B. DONATI, A. DONEPUDI, R. V. DONG, F. DONG, H. T. DONG, M. DONG, Y. DONGIOVANNI, D. DONNARD, J. DOOLITTLE, L. DORE, D. DORENBOS, P. DORHEIM, S. DORIA, A. DORIESE, W. B. DOROGOV, P. DOROKHOV, A. DORVAUX, O. DOS SANTOS, J. M. F. DOS SANTOS, J. M. DOSHI, N. DOURAGHY, A. DOWBEN, P. A. DOWDELL, S. J. DOWKONTT, P. DOWSETT, M. DOZIERE, G. DRAGICEVIC, M. DRAGONE, A. DRAKE, G. DRAPKIN, E. DRESS, A. DRESSENDORFER, P. V. DRESSLER, P. DRESSNANDT, N. DREYER, J. G. DROZDOWSKI, W. DRURY, O. DRURY, O. B. DU, H. DU, Q. DU, Y. DUARTE, O. DUARTE PINTO, S. DUBBERT, J. DUBE, S. DUBOULOZ, A.

N13-151 CH2-4 N32-2 N25-12 N13-149 M05-268, M13-60, N25-86, N25-131, N25-137 J01-5, M09-134 N13-232 N38-2 N25-118, N25-136 J04-4 N25-134 N44-1 M10-3 N25-123 HPP-1 N13-65 M05-370 M13-390 N25-241 N13-145 M05-100 N17-1 M13-144 N13-13 N13-234 J05-9, M13-264 N44-3 N13-86 N25-42, N25-58 N25-193 N22-5, N22-6, N25-123, N25-203 J05-1 N11-4 J04-7, N13-160, N13-161 N13-46 M15-7 N29-4 J02-7 N31-2 CH2-3 N22-6 N25-159 M15-2, N38-4 N10-2, N13-197 HP2-1 N13-13 N09-7 N25-215 N13-39, N13-46, N13-48 N25-78 J05-9 N01-1, N01-6 N25-78, N25-149, N30-4 M05-61, M09-134 N13-6 M09-296, N13-6 N21-5 N11-1 N34-7 N12-5, N13-179 J03-4 475

DUBRAWSKI, A. DULINSKI, W. DULUCQ, F. DUMONT, J. -L. DUNN, P. DUNN, W. L. DUPONT, E. DUPUIS, T. DURHAM, M. DURST, J. DURST, J. R. DUSCH, E. DUVAL, M. -A. DUVAL, S. DUVAUCHELLE, P. DWUZNIK, M. E EBERL, S. EBINE, M. EBRAHIMI, A. EBRAHIMI, B. ECKERLIN, G. ECKSTEIN, D. EDENBRANDT, L. EDMUNDS, T. EL FAKHRI, G. ELBS, J. ELDER, M. S. EL-HANANY, U. ELLEDGE, D. ELMGREN, K. EL-MOHRI, Y. ELY, J. H. EMERSON, V. EMRI, M. ENDO, T. ENGELS, R. ENGH, D. ENGHARDT, W. ENOMOTO, S. ENQVIST, A. ENTINE, G. ERANEN, S. ERDEI, G. ERFLE, J. ERICSON, M. N. ERIKSSON, L. ERIKSSON, L. A. ERIKSSON, M. ESCADA, J. ESCH, E. ESCH, E. I. ESPANA, S. ESPOSITO, A. ESPOSITO, B. ESPY, M. A. ETZION, E. EVANS, T. M. EVERMAN, J. L.

N13-247 N22-5, N22-6, N25-123, N25-167, N25-203 N03-1, N19-1 N14-3 M05-109 N30-6 N13-234 CH1-2 N15-1 J05-3, M05-43, N25-3, N25-185, N25-191 M13-345, N25-208 M05-211 M05-25, M09-242 N25-38 M13-309 N13-39

M05-187 N25-19 M13-111 M13-111 N25-159 N25-159 M09-302 N06-7 M05-229, M11-3 N13-64 N33-7 N25-207 N12-5, N13-179 N25-60 J05-17 N25-144 N13-48 J05-12, M09-152 N25-98, N28-2 N25-99 N13-94 M05-238, M06-5, N32-6 M09-281 N13-243 M09-14 N25-200 M05-274 N25-159 CN1-2, N03-6 J04-6, N01-7 M09-89 M09-89 N13-157 N29-3 N25-115 HP2-3, M05-226 N33-6 N13-170 N13-227 N13-183, N25-33 N09-4, N17-6 M09-104 476

EYRING, A. F F. -W.SADROZINSKI, H. F.-W. SADROZISNKI, H. FAANHOF, A. FABBRI, A. FABBRI, R. FABRIS, L. FACCIO, F. FADEYEV, V. FADEYEV, V. A. FADLER, F. FAILOR, B. H. FAIVRE-CHAUVET, A. FALCHIERI, D. FALEEV, N. I. FAN, J. FAN, R. FANCHINI, E. FANTACCI, M. E. FANTECHI, R. FANTI, V. FANUCCI, L. FARELL, R. FARIA, L. O. D. FARNCOMBE, T. H. FARRELL, R. FARSONI, A. FAULER, A. FAYAD, H. FAZZI, A. FEHER, S. FEIGLIN, D. H. FELD, L. FELICI, G. FELICIELLO, A. FELLA, A. FELLER, W. B. FENG, D. FENG, D. D. FENG, L. FENG, T. FENG, Y. FENG, Z. FENIMORE, E. FENU, G. FERNANDAZ, P. FERNANDEZ, M. FERNANDEZ, P. B. FERNANDEZ, P. FERNANDEZ, Y. FERNANDO, P. FERNS, G. FERRANDO, N. FERRARI, T. FERREIRA, A. L. FERREIRA, C. S. FERREIRA, N. FERREIRA, N. C.

N34-4

N25-156 N08-4, N25-158 N25-60 N13-18, N13-33, N13-131, N26-2 N03-1, N13-190 CN1-2, N06-3 N25-28 N05-5, N08-4, N25-158 N13-104, N25-181 M08-5 N13-218, N13-225 M13-144 M05-391, N07-4, N13-116 N25-161 M13-390 N13-143, N25-242 N25-117 N39-1 N21-2 M09-74, N25-8, N39-6 N25-243 N36-2 N25-44 M05-220, M05-337 J01-5, M05-328, M09-134 N25-85 N25-9, N25-191, N41-2 M05-46, M13-402 N25-4 HPP-1 HPP-2 N25-159 N44-1 N25-68 N13-87, N23-1 CN1-4, CN2-3 M05-187 M05-166 N25-180 HPP-6 M13-240 N07-2 N26-4 N13-72 N25-237 N25-159 N13-98 M13-186 M05-151 M05-142 M13-396 M05-10 N17-1 N11-4 M05-307 M09-353 M05-307 477

FERREIRA MARQUES, R. FERRER, A. FERRER, M. L. FERRERO, A. FERRETTI, C. FERRIS, K. F. FESQUET, M. FESSLER, J. A. FIANDRINI, E. FIDLER, E. FIEDERLE, M. FIEDLER, F. FIESELER, M. FIGUEIRAS, F. P. FIGUEIREDO, A. M. FIGUEROA, R. G. FINK, D. FIORETTI, V. FIORINA, E. FIORINI, C. FIORINI, M. FIRMINO, S. F. FISCHER, D. A. FISCHER, H. FISCHER, P. FISCHL, B. FISHER, S. FISHMAN, E. FITZGERALD, N. D. FLANDRE, D. FLASKA, M. FLEISCHMANN, C. FLEMMING, H. FLEMMING, R. M. FLETA, C. FLOHR, T. FLOHR, T. G. FLORES, S. FLUMERFELT, E. L. FOEHL, K. FOLGER, G. FONG, K. W. FONSECA-RODRIGUES, S. S. D. O. FONTAINE, R. FONTANA, A. FONTE, P. FORMAN, L. FORRESTER, J. FORTIER, M. FOTI, A. FOUGERON, D. FOURGUETTE, D. FOXE, M. FR&OUMLJDH, A. FR&OUMLJDH, C. FR&OUMLJDH, E. FRACH, T. FRAME, K. FRANCHI, G. FRANCOIS, E. FRANDES, M.

M09-353 N13-113 N13-86 M13-348 N13-183 N25-217 N13-53 M03-7, M04-3, N13-126 M13-12 N13-199 N25-9, N25-191, N41-2 M05-238, N32-6 M04-7, M13-324 M05-163, M09-182 N25-10 M05-403 N25-75 N13-63, N33-3 N39-1 M09-329, N13-101, N13-128, N19-4, N26-3, N28-7 N03-4 N13-93 N11-7 N36-3 M05-28, M05-244, N13-196, N28-4 M04-1 N01-6 M09-389 M13-282 N25-192 N13-232, N13-235, N13-243, N13-244, N25-64 M13-381 N13-15 N29-5 N25-7, N38-2 M12-4 M09-389 N13-69 N13-249 N05-2 N13-89 N30-4 N13-95 J05-11, M13-354, M14-3, M15-8, N13-30, N13-108 N25-68 M09-353 N13-233 N25-60 M13-375 N25-65 N12-5, N13-179 N25-22 N05-1 N25-195 N25-195 N25-195 J04-1, N28-5 N14-6 M09-332 M04-8 M13-219 478

FRANK, J. FRANKLIN, D. FRANKLIN, D. R. FRANSEN, M. FRASER, A. M. FRATONI, R. FRETWURST, E. FREY, E. FREY, E. C. FREY, M. FRIED, J. FRIEDL, M. FRIEDMAN, P. S. FRIEDRICH, J. FRIEDRICH, S. FRISCH, H. FRISCH, H. J. FROST, C. FRUTSCHY, K. FRYER, T. D. FU, G. FU, L. FUCHS, C. FUENTES, C. FUERST, J. FUERST, S. FUERSTNER, M. FUJII, K. FUJII, M. FUJII, Y. FUJIMAKI, S. FUJIMOTO, Y. FUJITA, K. FUJITA, Y. FUJIWARA, T. FUKASAWA, R. FUKAZAWA, Y. FUKUCHI, T. FUKUDA, K. FUKUDA, M. FUKUMITSU, K. FUKUYAMA, T. FULHAM, M. FULLEM, T. Z. FULTON, R. FULTON, R. R. FUNG, E. K. FUNG, G. S. K. FUNK, L. L. FURENLID, L. R. FURFARO, E. FURLETOV, S. FURUMIYA, T. FURUTA, M. FURUTAKA, K. FURUYA, Y. FUSAYASU, T. G G. PUNTONET, C.

N13-52 M09-236 M05-109 N11-3 N13-239 M09-92 N25-159 M09-383 M05-298, M09-35, M09-293, M09-296, M13-18 N25-159 M15-2, N13-47, N19-3, N25-218, N38-4 N25-159 N25-33 N42-4 N25-78, N30-4 M05-7, M09-80 N27-5 N25-22 M12-6 M05-253 M13-141 M03-4, M13-51 N13-114 N25-28 M08-5 M09-263 N04-5 N35-3 N05-7 J05-13 N25-13 N25-97 M09-56, N13-156 N22-2 N13-49, N13-156, N27-2 M07-6 N35-7, N37-4 M09-281 N25-96, N25-110, N35-7 M09-56 M13-30 N13-61, N41-3 M05-187 N13-11 M05-166, M09-107 M05-16, M09-53 M05-154 M05-40, M09-389 N25-26 HPP-14, HP2-5, M09-269, M15-5 N13-180 N13-196 M05-1, N27-2 M05-1 N07-3, N24-2, N42-5 N25-113 N19-5

M09-158 479

GABOS, P. GABRIELLI, A. GADEA, R. GAIONI, L. GAITANIS, A. GALASHOV, E. GALASSI, M. GALASSI, M. C. GALKIN, S. M. GALLAS, B. GALLEZOT, J. -D. GALLI, L. GALLIVANONE, F. GALLOWAY, M. GALUNOV, N. GALUNOV, N. Z. GAMBARDELLA, L. M. GAN, K. K. GANDHI, T. GANDOLFI, E. GANEM, A. GANGADHARAN, B. GARBOLINO, S. GARCIA, C. GARCIA, F. GARCIA, S. GARCIA DE ACILU, P. GARCIA-HERNANDEZ, J. -C. GARCIA-SCIVERES, M. GARDNER, C. S. GARGANO, G. GARGIONI, E. GARIBALDI, F. GARNIR, H. -P. GARRETT, P. GARSON, A. I. GARSON III, A. GARSON III, A. B. GASANOV, E. GASKIN, J. A. GASKIN, J. GATTA, M. GAUBAS, E. GAUME, R. GAUTIER, D. C. GAYSINSKIY, V. GEBHARDT, P. GEHM, M. E. GEITHNER, P. GEKTIN, A. V. GEKTIN, A. F. GELIN, M. GEMME, G. GENAT, J. -F. GENAT, J. -F. C. GENNERT, M. GENNERT, M. A. GENOVA, P. GENT, C. GEORGE, L. GEORGE, R. T. GEORGIOU, E.

N02-4 N21-7, N24-4 M05-10 N03-5, N25-198 M09-212 N25-138 N26-4 N06-1, N13-229 N29-6 M13-18 M03-3 N10-3 M09-110 N13-217, N14-2 N25-90 N25-116, N29-7 M09-371 N03-3 M08-8 N13-116 N06-7 J03-4 N03-4 N13-166 N13-7, N13-82, N25-35 HP2-3 M05-103 M09-239 N12-5, N13-179 N30-4 N39-1 N09-3, N39-7 M05-331, M09-92 CH1-2 N25-72 N31-2 J05-4, M05-127 N25-204 N13-163 N19-3, N25-224 N25-218 N44-1 N25-159 N01-6 N25-89 J04-2, M09-14, N20-4 M05-28 M09-269 M13-345 N25-91, N25-103 N13-241 N22-6, N25-203 N39-1 M05-7, M09-80 N27-5 M05-178 M09-200 N15-3 N13-206 N13-206 M13-288 M05-208 480

GEORGIOU, L. GERACI, A. GERACI, E. GERASYMOV, Y. V. GERBER, W. H. GERGANOV, G. V. GERLING, M. GERMAN, U. GERMANI, S. GERONIMO, G. D. GERSHMAN, B. GESO, M. GESSLER, P. GEVIN, O. GHELMAN, M. GHITA, M. GHOSH ROY, D. GIACOBBE, B. GIACOMELLI, L. GIANNETTI, P. GIANOTTI, P. GIERLIK, M. GIFFORD, H. GIFFORD, H. C. GIGANTE, G. E. GIL, E. C. GILARDI, M. C. GILBERTSON, R. D. GILES, J. GILES, N. C. GILLAM, J. E. GILLAND, D. R. GILLICH, D. J. GILLISPIE, S. B. GINESTAR, D. GINZBURG, D. GIOKARIS, N. GIORDANO, M. GIORDANO, R. GIOT, M. GIOVANETTI, G. GISPERT, J. D. GIUBILATO, P. GIULIANI, E. GIULIANI, F. GIVAUDAN, A. GLASER, M. GLASSER, F. GLAZIER, D. GLEICH, B. GLESENER, L. GLODO, J. GLUSHKOV, I. GMAR, M. GNANI, D. GNANVO, K. GODA, J. GODA, J. M. GODBEER, A. GODDARD, J. S. GODET, O. GODINEZ, F.

CH1-4 N13-25, N13-101, N13-102, N13-103, N13-112 N25-66 N29-7 M05-403 M05-343 N08-2, N08-4, N25-156, N25-158, N25-181 N13-81 N43-3 N31-2 M13-171 N25-1 N25-252, N25-253 N13-21, N35-6 N13-230, N25-133 N39-5 M09-317 N25-68 N13-170 HPP-10, N21-3 N15-3 N01-7 M09-173 M02-4, M09-191, M09-200 M05-370 N25-159 M09-110 N25-30, N25-115 M13-396 N25-106 M05-118, M09-137 M13-54 N13-11, N30-1 M13-342 N25-77 J04-4, N05-6, N13-230, N25-133 M05-208 M08-3 N13-14, N13-107, N21-4, N25-246 N25-76 N13-107 M05-151, M05-163, M09-182 N22-7 N21-3 M09-92 N13-53 N04-5 J05-19 N05-2 M15-3 N13-61 J05-21, N01-5, N25-81 N11-6 N13-234 N12-5, N13-179, N38-1 J03-2, N13-246 N13-106 N25-62, N25-63 N25-31 M05-13, M09-347, N06-3 N35-6 M13-51 481

GOERTZEN, A. L. M05-352, M09-65, M09-356 N25-238, N25-239 GOETTLICHER, P. GOETTMANN, W. N30-2 N22-6, N25-203 GOFFE, M. GOKHALE, M. N06-7 N07-3, N24-2 GOKO, S. GOLA, A. M09-329, N26-3 GOLBY, A. M13-243 M09-119 GOLUBEV, P. GOMEZ, F. J02-6 M09-182 GOMEZ, V. GOMEZ BERISSO, M. N13-69 GONELLA, L. N34-4 N13-230 GONEN, E. GONG, G. H. N25-245 N07-2 GONG, G. GONG, H. N13-10, N25-245 N25-1 GONG, S. GONON, G. N25-211 N25-159 GONZALEZ, F. J. GONZALEZ, F. N13-53 N13-69 GONZALEZ, J. GONZALEZ, V. N13-113 GONZALEZ, X. J. M05-163, M09-182 GONZALEZ MALINE, D. N23-7 GONZALEZ-CASTANO, D. M. J02-6 GONZALEZ-DIAZ, D. N15-4 GOORDEN, M. C. M06-3, M09-68, M13-264 GORELIKOV, D. N29-2 N39-1 GORI, I. GORKE, H. N36-3 GORRIZ, J. M. M02-5, M05-157, M09-155, M09-158, M09-335 M05-175 GRRIZ, J. M. GOSTILO, V. V. N31-7 GOSWAMI, B. M04-6 GOTHE, D. A. N36-4 N36-4 GOURISHETTY, A. K. GOUVEA, A. L. N13-160, N13-161 GOVINDARAJAN, N. M09-230 GRAAFSMA, H. N25-239 N27-5 GRABAS, H. GRABIEC, P. N25-159 N25-212 GRABMAYR, P. GRABOSCH, H. -J. N30-2 N02-6, N25-229 GRAFF, R. T. GRAHN, K. -J. N13-168 N13-69 GRAJALES, J. GRASS, M. M09-203 N25-66 GRASSI, L. GRASSO, L. N13-246 GRAVEL, P. M07-6, M13-21 N13-153 GRAZIANI, E. GRAZIOSO, R. M13-9, N28-1 GREEN, A. N13-245 N13-239 GREEN, J. A. GREEN, J. M09-236 GREENBERG, C. N13-236, N25-20 GREENWELL, R. N03-6 GREER, K. L. M09-101 GREGOR, I. M. N45-2 GREGOR, J. HPP-5 N23-6 GREGORI, D. 482

GREGORY, J. GREIFFENBERG, D. GREZES-BESSET, L. GRICHINE, V. GRICIA, M. GRIFFITHS, J. GRILLO, A. A. GRIMMER, R. GRINT, A. N. GRINYOV, B. V. GROBSHTEIN, Y. GRODNER, M. GRODZICKA, M. GROENEVELD, A. B. J. GROHMAN, M. A. GROISELLE, C. J. GROMOV, V. GRONBECH-JENSEN, N. GRONEWALD, M. GROS, S. GROS DAILLON, E. GROSICAR, B. GROSSWENDT, B. GROTUS, N. GROZA, M. GRUBE, B. GRUBER, E. A. GRUDBERG, P. GRYBO, P. GRYBOS, P. GU, S. GU, Y. GU, Z. GUADA, C. GUADAGNO, G. GUARDALA, N. A. GUARDIOLA, C. GUATELLI, S. GUAZZONI, C. GUAZZONI, P. GUENTHER, R. W. GUEORGUIEV, A. GUERGUEIEV, A. GUERIN, B. GUERRA, P. GUERREIRO, C. GUEVARA DAY, W. R. GUEYE, P. GUIDA, R. GUIGNANDON, A. GUILLOUX, F. GUL, R. GULLBERG, G. GULLBERG, G. T. GUNDIAH, G. GUNI, E. GUNJI, S. GUO, L. GUO, T. GUO, Z. GUSTAVINO, C. GUY, M.

N13-183 N41-2 M05-235 N13-89 M09-92 N13-206 N13-48 M09-380, M12-3, M13-207 J05-16, N13-129 N29-6, N29-7 M09-350 N25-159 N25-103 M05-4 N25-16 N14-3 N12-5, N13-179 N25-150 N13-29 N36-2 J05-19 J05-10, M09-83 N09-3, N39-7 M13-357 N01-1, N25-204, N31-2 N42-4 J02-2 N13-27 N13-38 N19-6 M05-178, M09-200 M05-145, M09-131 M05-88 N13-69 N06-2 N20-5 N25-7 J05-6 J02-1, N02-2, N13-74, N25-2, N25-66 N25-65, N25-66 M11-6 N13-224 N13-244 M11-3 M13-285, M15-6 M05-307 N13-69 M09-230 N11-6 M09-290 N25-167, N25-203 N31-4 N13-135, N18-2 J01-6, M05-40, M05-64, M05-223, M05-310, M09-338, M09-377, N18-5 N25-140, N25-143, N25-150 N25-191, N25-208 N13-68 J04-2 M11-6 N25-23 N13-56 M04-6, M13-36 483

H HA, J. H. HA, S. -H. HAAS, D. A. HABA, H. HAEFER, A. HAELKER, O. HAHN, C. HAHN, F. HAIDER, S. HAJIMA, R. HALL, C. J. HALL, N. C. HALLEZ, H. HALSTED, P. HAMADA, M. M. HAMAGAKI, H. HAMAMURA, M. J. HAMARNEH, G. HAMBY, D. HAMILL, J. HAMM, M. HAMMIG, M. D. HAN, J. H. HAN, L. HAN, X. HANAGAKI, K. HANDS, A. HANEY, S. K. HANLEN, D. R. HANNAN, B. W. HANRAHAN, S. M. HANSEN, S. B. HANSEN, T. -E. HANSEN, T. HANSON, C. J. HANS-WERNER, S. HARA, K. HARADA, H. HARADA, Y. HARDING, K. G. HARDWICK, S. HARISSOPULOS, S. HARKNESS, L. J. HARLIN, K. HARRIS, B. HARRISON, D. HARRISON, F. HARRISON, F. A. HARRISON, R. L. HARTMANN, F. HARTMANN, J. HARTMANN, R. HARTMANN, S. HARTMANN, U. HARTSOUGH, N. E. HARZ, M. T. HASEBE, N. HASEGAWA, K. HASEGAWA, T. HASEGAWA, Y. N13-148, N25-162, N25-190, N25-227, N25-228 J01-1 N25-60 M09-281 N04-2 N13-64 N45-6 N11-6 N11-6 N25-109 M05-118, M09-137 M13-306 M05-148 M09-248 N25-146 N11-5, N19-5 J01-1 M05-313 N25-85 M13-156 M04-1 N13-130, N13-133, N25-32, N25-179 M05-52 HPP-13, M05-367 M09-386, M12-5, M13-3, N25-61 N22-2 J02-4 N08-5 N13-26 N25-130 M05-64, N25-140 M05-133 N25-164 N25-164 N13-227 M04-8 N13-174, N22-2 N07-3, N24-2, N25-109 M09-188 N25-177 N13-206 CH1-4 J05-16, N13-129 M13-171 N06-1, N13-229 M12-6 N14-5 N31-1 M13-342 N25-159 J03-4 J02-1, N02-2, N13-64, N33-5, N36-3 CH2-1 N13-17 M08-8 M09-176 N05-7 N39-2 M13-105 M09-251 484

HASELMAN, M. D. HASHIZUME, N. HASI, J. HASTY, R. D. HATAZAWA, J. HATT, M. HATTORI, K. HATZISTRATIS, D. S. HAUCK, S. HAUCK, S. A. HAUF, S. HAUSLADEN, P. HAUSLADEN, P. A. HAUTEFEUILLE, B. HAVENS, T. HAWARI, A. I. HAWLEY, A. HAWRAMI, R. HAYAKAWA, K. HAYAKAWA, T. HAYASHI, M. HAYASHI, S. R. HAYATO, A. HAYES, J. C. HAYES, S. L. HAYWARD, J. P. HAZEN, E. HAZEN, E. S. HE, B. HE, J. HE, X. HE, Z. HEEMSKERK, J. W. T. HEERING, A. H. HEFFNER, M. D. HEGYESI, G. HEIJNE, E. H. M. HEINI, S. HEINTZ, M. K. HEINZINGER, K. HEISMANN, B. J. HEISTER, A. HEMMICK, T. HEMPEREK, T. HENDERSON, C. M. HENG, Y. HENGARTNER, N. W. HENNIG, W. HENRICH, B. HENRY, J. J. HENSELER, D. HENSLEY, W. K. HERBACH, C. -M. HERHOLZ, K. HERMAN, C. HERNANDEZ, L. HERRAIZ, J. L. HERRMAN, S. HERRMANN, F. HERRMANN, N. HERRMANN, S.

M06-1 M05-1 N13-202, N25-182, N38-1 N04-2 M05-262, M09-251, M10-1 M13-186 N18-7, N20-1 N13-34 M06-1 M13-126 N33-3 N04-3 CN1-2 N25-87, N25-88 M11-6 N25-24, N25-25 N38-3 N01-5 N25-172 N25-109 N25-50 N25-176 N11-5, N13-158 N25-144 CN1-5 N25-120 N13-197 N13-167 M05-298, M09-35, M09-293 N25-1 M13-18 N13-47, N13-126, N13-134, N18-3, N18-4, N25-165, N25-166, N25-169, N25-214, N25-223, N26-7, N31-3 M06-3, M09-68, M13-264 N40-4 N42-6 J05-12, M09-152 N19-7 N25-215 N27-5 N35-4 M05-43 N13-167 N15-1 N12-5, N13-179 N30-6 N25-95 N13-239, N13-245 N13-27, N13-119 N25-9, N25-239 N25-106 M13-9, N28-1 N06-5, N13-207, N14-1 N01-3, N13-224, N13-244 M13-183 N13-47, N25-223 N25-134 HP2-3, M05-226 N19-4 N42-2 N13-44, N13-45, N13-140 N13-35, N13-64, N24-5, N35-4 485

HERTZ, K. L. HERZOG, H. HESELIUS, S. -J. HESTERMAN, J. Y. HESZ, G. HEUSER, J. M. HIDVEGI, A. HIGASHI, N. HIGGINS, W. HIGGINS, W. M. HIGON, E. HIGUCHI, I. HILCHENBACH, M. HILDENBRAND, K. D. HILL, G. HILL, T. R. HILLEMANNS, H. H. HILLER, L. HILTON, B. A. HILTON, G. C. HIMMI, A. HINDES, J. HINO, M. HIRAGA, J. S. HIRANO, Y. HIRSEMANN, H. HITOMI, K. HNATOWICZ, V. HNSEL, S. HO, E. Y. T. HOBART, K. D. HOCH, R. HOCQUET, F. -P. HOEFERKAMP, M. HOEJGAARD, L. HOEK, M. HOEPPNER, C. HOERAUF, G. HOFF, G. HOFF, J. HOFF, J. R. HOFFMANN, D. H. H. HOFFMANN, K. H. HOFMANN, H. G. HOFMANN, M. HOFMANN, T. HOFSTAETTER, A. HOHBERG, M. HOHLMANN, M. HOLLAND, C. E. HOLLINGSWORTH, M. S. HOLLOWAY, P. H. HOLM, S. HONDA, M. HONG, I. HONG, K. J. HONG, K. HONG, S. J. HONG, X. HONKIMKI, V. HONSCHEID, K. HOOVER, A. S.

N13-214 M04-4, M07-2, M09-287 N13-139 M13-171 M09-146 N25-187, N25-197 N25-252, N25-253 N18-7 N01-1, N25-230 N01-5, N25-81 N13-113 M09-251 N24-5 N13-44, N13-45 N05-2 N13-106, N25-62 N24-7 N06-7, N18-6 CN1-5 N25-42 N22-6, N25-203 N13-183 N25-163 N35-3 M09-305, M10-1, M13-78 N25-239 N25-172 N25-75 N25-159 M09-320 N08-5 J03-2 CH1-2 N25-158 M09-224, M13-225 N05-2 N44-3 J03-4 N13-93, N39-3 N13-197, N27-1 N03-7 N36-3 N25-159 HP1-1 M05-259 N25-168 N32-2 M09-263 J03-2, N13-246 N13-214 N34-6 N25-127 M05-133, M09-224, M13-225 M13-63 M05-217, M13-198 M09-8, M09-86, M13-27, M13-261 M05-334, M13-258, M13-336, N45-6 M09-62, M09-128, M09-143 M09-257 N13-65 J05-10, M09-83, M13-120, N07-1, N37-1 N06-1, N13-229, N25-42, N25-58 486

HOPPE, E. W. HOPPIN, J. HORANSKY, R. D. HORN, B. K. P. HORNBACK, D. E. HORNEGGER, J. HORROCKS, J. HOSSAIN, A. HOSSAIN, M. HOSSBACH, T. W. HOSTICKA, B. HOTELING, N. HOTELING, N. J. HOU, F. HOU, L. HOWARD, D. HOWE, J. Y. HOWELL, C. R. HOWRAMI, R. HRUBEC, J. HSIAO, I. -T. HSIAO, I. HSIEH, H. HSIEH, J. HSIEH, S. C. HSIUNG, Y. B. HSU, C. HSU, C. -H. HU, C. HU, H. HU, H. -T. HU, T. HU, W. HU, Z. HUANG, L. -T. HUANG, M. HUANG, M. A. HUANG, P. -C. HUANG, Q. HUANG, S. -C. HUANG, W. HUANG, Z. HUBER, J. S. HUBER, S. HUEGGING, F. HUESMAN, R. H. HUGHES, H. L. HUGHES, P. HUGHES, T. HU-GUO, C. HUH, S. S. HUH, W. HUH, Y. S. HUH, Y. HUISMAN, M. C. HULLINGER, D. HUMBERT, B. HUMPHRIES, T. HUNG, W. -C. HUNT, A. W. HUNTER, S. D. HUNTER, W. C. J.

N25-74 M13-171 N25-42 N13-229 N06-3, N26-5 HP1-1, M02-2, M13-291 N13-206 N25-209, N25-221, N31-4 M13-246 N25-74 N25-60 N25-58 N25-42 N13-143 N13-10 N25-134 N25-106 J01-3 N01-1 N25-159 M09-227, M13-363 M13-216, M13-384 M13-216 M12-1, M13-390 N25-45 N13-181 M13-216 M09-227, M13-363, N25-6 N22-5, N22-6 M09-221 N13-181 N32-1, N45-1 M05-334, M09-8, M09-86, M13-27, M13-258, M13-261, M13-336 M09-245, M11-6 M13-276 N39-5 N13-57, N13-70, N37-3 M13-363 M05-223, M05-310 M13-372, M13-378 N13-6, N13-50 J02-3, J04-3, M13-222 M05-64 N42-4 N02-7, N13-109, N34-4 M09-377 N08-5 M09-8 M05-295, M09-164 N25-203 HPP-13, M13-120 M03-7 M09-8, M13-27, M13-261 M05-334, M09-266, M13-258, M13-336, N25-36, N25-100 M05-4 N25-235 N13-114, N25-101 M05-313 N13-57, N13-70, N37-3 N13-218, N13-225 N20-5 M05-94, M13-342 487

HUOVELIN, J. HUPE, O. HURLBUT, C. HURLEY, F. HURLEY, R. F. HURST, T. A. HUTCHESON, A. HUTCHESON, A. L. HUTTON, B. HWANG, C. -S. HWANG, J. Y. HWANG, Y. G. HYNES, M. V. HYRONIMUS, B. J. HYUN, H. J. I IANAKIEV, K. D. IASELLI, G. IATROU, M. IBRAGIMOV, J. ICHIHARA, T. ICHIKAWA, G. ICHIMIYA, R. IDA, C. IGASHIRA, M. IGUCHI, T. IIDA, H. IIJIMA, K. IKEDA, H. IKEGAMI, Y. IKEMOTO, Y. ILYIN, A. M. ILYINA, I. A. IM, G. H. IMAIZUMI, M. IMHOFF, M. IMREK, J. INADAMA, N. INANIWA, T. INIEWSKI, K. INOUE, A. INOUE, Y. INTROINI, V. M. INUZUKA, M. INZINNA, L. IONICA, M. IORDANOVA, A. IRASTORZA, I. IRVING, T. IRWIN, K. D. ISAB, A. H. ISHIBASHI, H. ISHIDA, K. ISHIHARA, N. ISHII, H. ISHII, K. ISHII, S. ISHIKAWA, A. ISHIKAWA, S.

N13-82, N35-5 N25-184 N13-220, N25-104, N25-134 N25-181 N08-4, N13-144, N25-156, N25-158 N25-149 N13-213 N06-2, N20-6 N26-3 N13-51 M05-364, M09-143, M13-117, N25-51 M09-266, N25-36, N25-100 N06-1, N13-229 N25-74 N25-202

N13-106, N25-62, N25-63, N29-5 N05-4 M13-330 N13-163 M13-288 N25-163 N22-2 N18-7, N20-1 N07-3, N24-2 N25-49, N25-50 M09-305, M10-1, M13-78 N13-146 M13-15, N22-2, N25-194, N41-6 N13-174, N22-2 N22-2, N25-205 N13-55 N13-55 M13-261 M09-251 N25-101 J05-12, M09-152 J04-5, M05-82, M09-32, M13-102, M13-267 J04-5 N25-171, N25-178, N25-216 N25-27 M13-96 N25-4 N19-5 M12-6 M13-12, N25-141 N15-1 N36-3 J04-2 N25-42, N25-58 N25-73 M13-15, N25-80 M10-1 N13-154 M13-360 M09-95 N25-13 M05-37 M15-4 488

ISHIKAWA, S. -N. ISHIKAWA, T. ISHIKAWA, Y. ISHIOKA, N. S. ISHITSU, T. ISHITSUKA, J. ISHITSUKA, M. ISHIZU, S. ISHIZUKA, T. ISLAM, A. U. ISLAMOV, A. K. ISOBE, M. ISOCRATE, R. ISOLA, A. A. ITALIANO, A. ITO, M. ITO, R. ITO, S. IVANCHENKO, V. IVANOV, A. I. IVANOV, V. IWAHASHI, T. IWAI, G. IWAI, H. IWAKI, S. IWAMOTO, Y. IWANCZYK, J. S. IWANOWSKA, J. IWASAKI, M. IWASE, H. IWATA, K. IYOMOTO, N. IZAGUIRRE, E. W. IZZO, V. J JAASKELAINEN, K. JACEWICZ, M. JACK, S. JACKSON, C. JACKSON, C. J. JAIN, A. JAKAB, G. JAKOBSSON, B. JAKUBEK, J. JALALI, M. S. JALILIAN, R. JAMES, M. R. JAMES, R. JAMES, R. B. JAN, M. -L. JANG, D. Y. JANG, D. JANG, G. -W. JANG, N. JANG, S. S. JANSEN, F. P. JANVIER, B. JARAMILLO, R. JARMAN, K. D.

N13-61, N41-3 M13-75, N13-154 M13-15 N25-13 M09-5 N13-69 N13-123 N25-96, N35-7 N13-154 N25-73 N25-160, N25-161 N25-49 N25-247 M09-203 N13-86 M09-62, M09-128, M09-143 N13-154 M09-188, N25-13 N13-89 M09-368, N13-105 N25-193, N31-7 N11-5, N13-158 N13-154 N25-49 J05-13, N18-7, N20-1 N25-83 M08-8 N01-7, N13-220, N25-90, N25-103, N25-138 N13-9 N13-154 M13-147 N13-156 M05-130 N13-14, N13-107, N21-4, N25-246

N22-6, N25-203 N44-1 N25-239 M09-8, M10-5 N13-241 M08-1 HP2-4 M09-119 J05-6, M05-241, M09-236, N19-7, N20-2, N25-17, N25-21, N25-189 N25-216 N05-1 N09-1 N25-209 N25-221, N31-4 M13-276 N13-148, N25-54 N13-76 J05-18 M09-266 N13-219 M13-321 M05-25, M09-242 N25-159 N13-208 489

JARRON, P. JARRON, P. P. JASTRZEMBSKI, E. JASZCZAK, R. J. JAWORSKI, J. M. JEAN, P. JEGLOT, J. JEMIAN, P. R. JENKINS, D. G. JENSEN, G. U. JENSEN, J. JEON, I. JEON, S. JEONG, K. Y. JEONG, M. JERNIGAN, J. G. JIANG, H. JIN, H. B. JIN, L. JIN, X. JIN, Y. JING, F. JOERG, M. JOHANNESSON, G. JOHNSON, B. B. JOHNSON, E. JOHNSON, E. B. JOHNSON, K. JOHNSON, M. E. JOHNSON, M. L. JOHNSON, O. JOHNSON, W. N. JOHNSON-WILLIAMS, N. G. JONES, D. M. JONES, J. JONES, J. P. JONES, L. L. JONES, M. JONES, M. P. JONES, T. JONES, W. F. JOO, S. -K. JOSEPH, M. JOSHI, A. A. JOSHI, P. JOSHI KAYE, S. JOSIC, L. JOSIC, L. Z. JOUNG, J. JOVANOVIC, I. JOYCE, M. J. JUDENHOFER, M. S. JUDSON, D. S. JUELS, P. JULYAN, P. J. JUNG, I. JUNG, J. H. JUNG, J. JUNG, S. -H. JUNNARKAR, S. JUNNARKAR, S. S. JUNWEI, D.

N03-4, N13-41 N24-7 N25-241 M09-101 N18-3 N13-57, N37-3 N13-54 N13-98 J05-1 N25-164 M13-225 M09-395 M09-266, N25-36, N25-100 M09-404 N25-179 N06-7 N25-176 M13-27 N02-5 M05-394, M09-161 M05-88, M05-136 M05-223 M04-8 N06-7 N13-214 M13-60, N25-86, N25-137 N04-6, N13-237, N25-131, N36-1 M09-200 N13-187 N33-7 M10-3 N06-2, N13-213 M13-126 N25-217 M04-6, M13-36, N04-3 M05-292 N27-7 J05-16 N13-129 M05-319 M09-104 M13-180 N13-133 M04-2 J05-1 N26-7 CN1-1 CN2-1 J05-23, M09-86, M13-180 N05-1 N13-129 M04-5, M11-5, M13-246 J05-16, N13-129 M10-3 M05-319 N25-204, N31-2 M05-115, M09-8, M09-86, M13-27, M13-261 M05-334, M13-258, M13-336 J05-18, M09-272 M05-169 M05-247, M05-376, M09-257 N13-99 490

K KABUKI, S. KACHEL, M. KACHELRIE, M. KACHELRIESS, M. KADMON, Y. KADRMAS, D. J. KAEMMERLING, P. KAESTNER, A. KAESTNER, A. P. KAFTANDJIAN, V. KAGAN, H. KAH, D. H. KAHLAOUI, N. KAISER, R. KALENDER, W. A. KALINKA, G. KALLIOKOSKI, M. KALLIOPUSKA, J. J. KAMADA, K. KAMIN, J. KAMIYA, Y. KAMIYAMA, T. KAMMERING, R. KANAI, Y. KANAYAMA, Y. KANG, B. H. KANG, J. H. KANG, J. KANG, J. U. KANG, K. KANG, S. M. KANG, S. Y. KANG, X. KANTARELOU, V. KAO, C. -M. KAPADIA, A. J. KAPLON, J. KAPPLER, S. KAPUSTA, M. KAPUSTINSKY, J. KARABAGIA, S. KARABARBOUNIS, A. KARAFASOULIS, K. KARAGOUNIS, M. KARAKATSANIS, N. A. KARAVAEVA, N. KARAVAEVA, N. L. KARGAR, A. KARI, B. KARKAR, S. KARMGARD, D. KARNOWSKI, T. P. KARP, J. S. KARPINSKI, W. KARPIUS, P. J. KARYDAS, A. G. KASHINO, G. KASSEBAUM, B. L. KASTE, J. J05-13, M05-271, M13-360, N18-7, N20-1 N13-38 M13-381 HPP-11, HP1-2, HP1-3, HP2-6, HP3-5, M08-4, M09-206, M09-380, M12-3, M12-4, M13-207, M13-210 J04-4, N05-6, N13-226, N13-230, N25-133 M02-1, M13-165 N13-109 CN1-1, CN2-1 CN2-4 M13-309 J05-10, M09-83, M13-120 N25-202 J05-2 N05-2 M08-4 J05-12 N13-7 N25-200 N25-98, N25-110, N25-121, N28-2 N15-1 N25-163 N07-3, N24-2, N25-18 N13-121 M09-251 M09-281 N13-8, N13-148, N25-54, N25-116 M09-8, M09-86, M13-27, M13-261 M05-334, M13-258, M13-336 M05-52 J02-3, J04-3, J05-20, M09-221, M13-222 N25-54, N25-190 N13-148, N25-54 M05-88 CH1-4 M05-7, M05-100, M09-80, M14-6 J05-8 N03-4, N13-39, N13-41 M02-2, M09-365 N01-7 N03-7 CH1-4 M05-208 N13-221 N12-5, N13-29, N13-179, N25-239, N34-4 M05-358 N25-90 N29-7 N25-170, N31-6 HPP-1 M09-254 N25-142 N06-3 M05-349, M13-66 N25-159 N25-42, N25-58, N25-59 CH1-4 J02-5 M05-130 M09-245, M11-6 491

KATABUCHI, T. KATADA, Y. KATAGIRI, M. KATAKURA, H. KATAOKA, J. KATO, Y. KATOH, C. KATRIN, W. KAUSSEN, G. KAVATSYUK, M. KAWABATA, N. KAWACHI, N. KAWAGUCHI, N. KAWAI, H. KAWAI, Y. KAWAMOTO, S. KAWANO, H. KAWARABAYASHI, J. KAWASAKI, S. KAWASHIMA, A. KAWASHIMA, H. KAWAZOE, Y. KAWRAKOW, I. KAYE, S. KAYE, W. R. KAYE, W. KAZKAZ, K. KAZUKAUSKAS, V. KECK, B. KEEREMAN, V. KEESING, D. B. KEEVIL, S. F. KEHAYIAS, C. E. KEHAYIAS, J. J. KEIL, A. KEILLOR, M. E. KEISTER, J. KEISTER, J. W. KELLER, S. H. KELLY, M. KEMMERLING, G. KEMMLER, N. KENDZIORRA, E. KENNEY, C. J. KEPE, V. KEPPEL, C. KEREKES, R. A. KERI, T. KERNAN, W. J. KETZER, B. KHAIDUKOV, N. N. M. KHALID, F. F. KHANNA, R. KHAPLANOV, A. KHIARI, F. Z. KHOMENKOV, V. KIBILKO, M. J. KIERSTEAD, J. KII, T. KIKUCHI, K. KIKUZAWA, N. KIM, B.

N07-3, N24-2 N35-3 N25-19 N35-3 M13-15, N28-2, N37-4 N13-154 M09-5 M04-8 N13-185 N10-4 M13-15 M15-4, N25-13 N25-96, N25-110, N25-113, N25-114, N35-7 M13-267, M13-270 N40-3 M09-389 N35-3 N25-27, N25-49, N25-50 N25-163 M05-1 M05-271 N25-114 M05-343 N14-5 N25-214, N26-7, N31-3 N25-223 N18-6, N33-1 N25-159 HP1-1 M05-28, M09-41, M13-252, M13-255 M09-362, M13-48 M09-248 J02-2 J02-2 M05-199 N25-74 N19-3, N25-218, N25-224 N38-5 M05-133, M09-224, M13-225 M09-248 N25-99 M11-5 N33-3 N13-202, N25-182, N38-1 M13-378 M09-230 N06-3 N05-2 N06-6, N13-249 N44-3 N25-44 N27-7, N35-5 N34-5 M11-4 N25-73 N25-159 N07-5 N34-5 N25-109 N25-13 N25-109 M09-266, N25-36, N25-100 492

KIM, C. H. KIM, C. L. KIM, C. KIM, D. S. KIM, D. H. KIM, D. K. KIM, D. KIM, G. KIM, G. D. KIM, H. J. KIM, H. O. KIM, H. KIM, H. S. KIM, H. K. KIM, J. C. KIM, J. S. KIM, J. S. K. KIM, J. KIM, J. G. KIM, J. E. KIM, J. K. KIM, J. Y. KIM, J. -S. KIM, K. H. KIM, K. KIM, K. M. KIM, K. -M. KIM, M. W. KIM, M. -S. KIM, M. KIM, M. -W. KIM, S. KIM, S. H. KIM, T. S. KIM, W. KIM, Y. K. KIM, Y. KIM, Y. -K. KIM, Y. S. KIM, Y. -S. KIM, Y. -B. KIMMEL, N. KIMURA, A. KIMURA, H. KIN, T. KINAHAN, P. KINAHAN, P. E. KINASHI, Y. KINDEM, J. KING, A. P. KING, M. KING, M. A. KINLAW, M. T. KINO, K. KIPPEN, R. M. KIRBY, N. KIRIHARA, L. J. KIRKHAM, R. KIROV, A. S. KIRZEDER, M. KIS, M. KIS, S. A.

N09-3, N13-83 M05-73 N25-148 M13-27 N25-162 N25-108 J05-13, M03-2, M15-2 N25-32 N13-8, N25-116 M05-97, N25-202 N25-202 M05-7, M09-80, N25-148, N25-230 N13-148, N25-162, N25-190, N25-227, N25-228 M05-232 N25-169 M05-97, M05-115, M05-364 N25-14 J05-13, N25-148 M05-97, M05-115, N25-14 M09-233 N25-116 N25-108 N13-219 N25-209, N25-221 M05-97, M09-395, N31-4 M05-115, N25-14 M13-180 M09-233 M09-272 M05-400 J05-18 N45-6 N25-162 CH2-4 N45-6 N13-8, N13-148, N25-54, N25-100, N25-116 J05-23, M05-400, M13-180 N25-228 M09-233 M09-272 J05-18 N33-5 N07-3, N24-2, N39-2, N42-5 M05-271 N07-3, N24-2, N42-5 M06-6, M09-17 M03-7, M09-47 J02-5 M13-81 M09-41 M05-178, M14-2 M02-4, M09-191, M09-200 N13-218 N07-3, N24-2 N13-229, N18-1 N25-174, N38-3 N13-207, N14-1 M05-109 M05-343 N13-199 N13-44, N13-45 J05-12, M09-152 493

KISHIMOTO, S. KISHISHITA, T. KISSEL, E. KISTENEV, E. P. KITAGUCHI, M. KITAMURA, K. KITAMURA, S. KITATANI, F. KITCHEN, I. KIYANAGI, Y. KIYUNA, T. KLAMRA, W. KLANNER, R. KLEIN, K. KLEINES, H. KLEINFELDER, S. KLIMENKO, A. V. KLOTZSCH, H. KLUGE, A. A. KLUGE, A. KLUIT, R. KNAUP, M. KNICKERBOCKER, K. L. KNIEST, F. KNOPP, M. KNOPP, M. V. KO, D. K. KOBASHI, K. KOBAYASHI, H. KOBAYASHI, M. KOCH, K. KOCH, M. KOCH, M. T. KOCHANOWSKA, D. KOCHIYAMA, M. KOCKELMANN, W. KOEHLER, M. KOENIGSMANN, K. KOERFER, M. KOESTERS, T. KOGAN, M. KOHARA, R. KOHRIKI, T. KOI, T. KOIKE, T. KOIZUMI, M. KOK, A. KOKUBUN, M. KOLAGANI, R. M. KOLB, A. KOLBASIN, V. A. KOLOPUS, J. A. KOLTHAMMER, J. A. KOLYA, S. KOMAMIYA, S. KOMAROV, S. KOMAROV, S. A. KOMIYAMA, T. KONAMI, S. KONG, Y. KONICZEK, M. KONIG, W.

M13-15 N25-194, N41-6 N13-94 N15-7 N25-163 M05-1, M05-37, M13-96 N13-154 N07-3, N24-2, N25-109 M13-168 N07-3, N24-2 N25-194 N25-138 N25-159, N25-239 N25-159 N13-109 N02-5, N13-50, N25-11 N04-2 N29-2 N24-7 N13-41 N12-5, N13-179 HP2-6 N42-1 N25-104 M13-120 M13-306 N13-8 M09-5 M13-75 N05-7 N13-23 N13-29 N13-196 N25-221 N22-2 CN1-4 N02-4 N36-3 N30-2 M13-324 M13-366 J05-13 N13-174, N22-2 N13-66 M09-29, N44-6 M13-15, N07-3, N24-2, N42-5 N25-164 N13-61, N37-4, N41-3 N30-4 M05-259, M13-9 N13-105 N25-135 M13-66 N13-202 N25-163 J05-4, M05-127, M13-48 M05-355, M09-278, M09-362 N13-149 N11-5, N13-158 N01-3, N13-224 J05-17 N13-45 494

KONIK, A. KONNO, C. KONNO, T. KONOROV, I. KONOROV, I. V. KONTAXAKIS, G. KOPANS, D. B. KORAL, K. F. KORBLY, S. E. KORDOS, M. KOREVAAR, M. A. N. KORJIK, M. KORMOLL, T. KORPAR, S. KORTELAINEN, M. J. KOSHIMUTA, S. KOSHINO, K. KOSOV, M. KOSTERS, T. KOTOV, A. KOTOV, I. V. KOTSUBO, V. KOTTHAUS, R. KOUA, K. M. C. KOUTALONIS, M. KOUZES, R. T. KOVANEN, A. M. KOWARSCHIK, M. KOWASH, B. R. KOZAKI, T. KOZIEL, M. KOZOREZOV, A. KOZOREZOV, A. G. KRAEMER, M. KRAMBERGER, G. KRAMMER, M. KRATZ, B. KRAUS, R. H. KRAWCZYNKSII, H. KRAWCZYNSKI, H. KREIDL, C. KREISLER, B. KRENNRICH, F. KREPS, A. KREUGER, R. KRISHNAMOORTHY, S. KRIZAN, P. KROL, A. KROMBACH, G. A. KRON, T. KROSS, B. KRUCKER, S. KRUEGER, H. KRUMREY, M. KRUTH, A. KUB, F. KUB, F. J. KUBO, H. KUBO, N. KUBOTA, Y. KUCHARSKI, K. KUCZEWSKI, A.

M09-359 N25-49, N25-83 N13-123 N42-4, N44-3 N25-249 M09-212, M15-6 M09-2 M04-3 N04-2 M05-142 M06-3, M09-68, M13-264 N32-2 M06-5 N25-118, N25-136 N13-192 N19-5 M09-305 N13-89 M13-294 N13-52 N13-52 N25-42 N36-3 N13-30 N13-206 N06-6, N13-249 N13-11, N30-1 HP1-1 N20-7, N26-6 N25-18 N22-6 N31-7 N41-4 N42-4 M13-234, N08-4, N25-206 N25-159 M05-160 N13-227 N25-204 J05-4, M05-127, N31-2 N13-196 M05-43, N25-3, N25-185 N10-2 N10-2 M09-68, M13-264 M05-265, M15-2 N25-118, N25-136 HPP-2 M11-6 M09-236 J01-3 N13-61 N13-29, N13-196, N25-239, N34-4 N24-5 N12-5, N13-179 N25-29 N08-5 J05-13, M05-271, M13-15, N18-7, N20-1 M09-5 M09-401 N25-159 N38-4 495

KUDO, H. KUDROLLI, H. KUDROLLI, H. A. KUEHL, P. KUHLS-GILCRIST, A. KULINICH, P. KUMAR, S. KUMAZAWA, Y. KUNATH, D. KUNTZ, J. KUNZ, P. KUNZE, H. KUPINSKI, M. A. KURATA, Y. KUROSAWA, S. KUROZUMI, A. KUSAKAWA, T. KUSAYANAGI, T. KUSKE, T. KUSNER, M. R. KUSTER, M. KUVVETLI, I. KUWERT, T. KUZE, M. KWAK, S. W. KWAN, S. KWON, C. KWON, J. KWON, S. I. KYME, A. KYME, A. Z. KYRIAKOU, Y. KYUSHIMA, H. L L&OUMLHNER, H. LA GUIDARA, E. LA MARRA, D. LA RIVIERE, P. J. LA RUE, G. S. LABOV, S. LABOV, S. E. LACASTA, C. LACKAS, C. LACOMBE, K. LACROIX, J. -S. LACY, J. L. LAFAGE-PROUST, M. -H. LAFOREST, R. LAGE, E. LAGOYANNIS, A. LAINE, F. LAJTOS, I. LAKSHMI, S. LAKTINEH, I. LAMANNA, G. LAMBLIN, J. LAMBROPOULOS, C. LAMBROPOULOS, C. P. LAMI, S.

M10-1, M13-45 M10-3 M09-314, M13-315, M13-318 M13-171 M08-1 M09-314, M10-3, M13-315, M13-318 N25-47 M05-1, N27-2 M05-238, M06-5 N01-6 M13-207, M13-210 M03-1 M09-350 N25-80 J05-13, M05-271, N18-7, N20-1 N05-7 N25-13 M09-341 N32-2 N25-104 N33-3, N36-3 N13-62 M13-291 N13-123 N13-219 N24-6 M09-233 M09-185 M09-128 M09-107 M05-16, M09-53 HPP-11, M08-4 N40-3

J03-1 N25-66 N13-39 M08-7, M09-284 N13-26 N13-247 N06-7 J05-10, M09-83, M13-120, N13-166 M13-171 N13-53, N35-6 N14-3 N06-4, N20-3 M09-290 M05-304, M09-215 M05-397, M09-98 CH1-4 N13-234 M09-152 N36-2 N05-3 N10-1 N25-38 N13-221 N13-34 N44-4 496

LAMMERTSMA, A. A. LAMOTTE, T. LAMPEN, T. LAMPERT, M. O. LAMPL, W. LAN, K. LAN, K. A. LANAS, I. LANCONELLI, N. LANDE, J. LANE, J. L. LANG, P. LANGE, W. LANGER, M. LANKES, K. LANNI, F. LANSLEY, S. P. LANZA, R. LANZA, R. C. LARDO, A. C. LARKIN, A. LAROCQUE, L. LARTIZIEN, C. LARUE, H. LASAYGUES, P. LASHLEY, J. R. LAU, F. W. Y. LAUBE, K. LAUBENSTEIN, M. LAUF, T. LAURELLI, P. LAURIA, A. LAVEZZI, L. LAVRIJSEN, W. T. L. P. LAWSON, P. LAYMAN, R. LAZARO PONTHUS, D. LAZARO-PONTHUS, D. LAZARUS, I. H. LAZUKA, D. LE POGAM, A. LE PROVOST, H. LE TOURNEUR, P. LEAHY, R. LEAHY, R. M. LEAO, C. R. LEAS, B. E. LEBBOLO, H. LEBEDEV, A. N. LEBLANC, G. S. LECHNER, P. LECOMTE, E. LECOMTE, P. LECOMTE, R. LECOQ, P. LECOQ, P. P. LECOQ, P. R. LEDESMA, M. J. LEDESMA-CARBAYO, M. J. LEDOUX, R. J. LEE, B. LEE, C.

M05-4 M09-239 N13-192 N25-215 N23-2 M13-87 J01-5, M05-31, N07-5 N13-97 M05-385, M05-391, M13-279, N25-126, N26-2 N35-6 N25-244 N36-3 N25-159 M09-290 M13-246 N34-5 J05-7 N06-1 N13-229 M13-288 N09-7 N25-22 M09-170, M14-1 N25-251 M09-59 N29-5 M13-135 M05-238 N25-213 N19-4, N24-5, N35-4 HPP-10 M09-392, M13-387, M13-393 N15-3, N25-68 N17-7 N13-167 M13-306 M13-309 M09-239, M09-311 J05-16 M10-3 M09-170 N13-53 N14-3 M09-44, M13-198 M04-2, M05-193, M07-4 N25-222 N06-2 N13-54 N13-155 N13-4, N13-5 N24-5, N26-3, N35-4 N35-6 N32-3 J05-11, J05-15, M13-354, M14-3, M15-8, N13-30, N13-108 N43-2 N24-7 N26-1 M13-285 M15-6 N04-2 N25-237 N25-148 497

LEE, C. M. LEE, C. H. LEE, C. S. LEE, D. LEE, D. H. LEE, H. LEE, H. -L. LEE, J. H. LEE, J. S. LEE, J. LEE, J. -S. LEE, J. M. LEE, K. LEE, M. N. LEE, N. H. LEE, N. -Y. LEE, S. H. LEE, S. J. LEE, S. K. LEE, S. -J. LEE, T. S. LEE, T. -S. LEE, T. LEE, V. LEE, W. LEE, W. H. LEE, W. G. LEE, Y. J. LEE, Y. S. LEFEBVRE, F. LEFESVRE, I. LEFORT, T. LEGERE, J. LEGERE, J. S. LEGOU, P. LEGOUPIL, S. LEGRADY, D. LEHMANN, E. LEHMANN, E. H. LEITE, M. A. L. LEITH, D. LEMAITRE, V. LEMATRE, C. LENOX, M. W. LENTERING, R. LENTI, M. LENZ, W. LEPAGE, M. D. LERAY, P. LERCH, M. L. F. LERCH, M. L. LERCHE, C. W. LESKO, K. T. LEUNG, W. C. LEVIN, C. LEVIN, C. S. LEVIN, D. S. LEVINE, M. LEVINSON, S. LEWELLEN, T. LEWELLEN, T. K. LEWIS, E.

M09-128 N25-54 N13-83 N26-4 M09-266, N25-36, N25-100 M13-180 N25-237 N13-83, N25-227, N25-228 M09-62, M09-128, M09-143, N13-83, N13-217, N14-2 J05-23, M13-135, N13-173 M13-300 N13-8 J05-23, M09-86, M13-180, N25-204, N31-2 M13-24 N25-190 M09-86 N13-83 M05-13, M09-347, N25-108 N25-116 M05-364, M09-143, M13-24, M13-117, N25-51 M05-115 M05-40, M05-205 N13-127 M05-124, M13-399, N25-174, N38-3 N13-127, N13-148 M05-97, M05-115 N25-116 M05-115 M05-97 M05-25, M09-242 N14-3 N25-107 N29-2 N25-22 N13-141, N25-67 M09-239, M09-311, M13-309 HPP-1 CN1-4 CH2-1, CN1-1, CN2-1, CN2-4 N13-122 N13-201, N45-7 N25-159 M05-67 HPP-5 N01-3, N13-224 N40-1 M05-247 M15-8 M13-144, N25-38 J05-6, M05-109, M09-236, N13-241 J02-7 M05-10, M09-149 N24-3 M09-107 M06-4, M13-351 M05-145, M09-131, M10-2, M13-6, M13-135, M14-5 N13-183, N16-5, N25-33 M09-2 N13-81, N13-226 M06-6 J03-6, M05-94, M06-1, M09-77, M13-114, M13-126, M13-342 M04-6, M13-36, M13-396 498

LEWIS, P. S. LEWIS, R. A. LI, A. LI, A. O. LI, B. LI, C. LI, H. LI, J. LI, L. LI, N. LI, Q. LI, S. LI, T. LI, W. LI, X. LI, Y. LI, Z. LIANG, H. LIANG, H. -C. LIANG, J. -S. LIAO, J. LIDEY, L. LIEMANN, G. LIERSE, C. LIETAER, N. LILL, J. -O. LIM, G. LIM, H. K. LIM, H. LIM, S. M. LIMA, J. A. C. LIMOUSIN, O. LIN, C. -H. LIN, J. R. LIN, K. LIN, K. M. LIN, W. F. LIN, W. -C. LIN, Y. C. LIN, Y. LINDGREN, S. LINGENFELTER, D. J. LINGERTAT, H. LINHAR, V. LINHART, V. LINK, J. T. LINSAY, P. LINTEREUR, A. T. LIPP, J. LISTER, C. J. LISTRATENKO, O. LITKE, K. E. LITMANOVICH, G. LITTLEWOOD, P. B. LIU, A. LIU, C. -C. LIU, G. LIU, L. LIU, M. LIU, R. -S. LIU, S. LIU, X.

N42-1 M05-118, M09-137 N25-30 N25-89 M12-1 M05-187, M15-1 M05-79, M05-106, M13-69, M13-108, M13-339 N13-6 M05-394, M09-221 M05-325, M13-39 J05-4, M05-127, M05-193, M07-4, M09-44, N25-204, N25-240, N31-2 N02-5 N13-209, N13-210, N25-236 N25-177 M09-77, M13-114, M13-126 M05-58, M09-71, M09-260, M13-93, N13-10, N13-31, N13-140, N13-142, N13-209 M05-67, M05-85, M13-240, N19-3, N25-155, N25-218, N25-224, N38-4 M05-58, M13-93 M13-276 N13-57, N13-70, N37-3 M05-34 N25-60 N22-1 CN1-3 N25-164 N13-139 M13-258 M09-8, M13-27, M13-261 M05-334, M13-258, M13-336 N25-14 M13-288 N13-21, N35-6 N13-57, N13-70, N37-3 N13-19 M13-216 M13-363 N13-20 M13-276 N25-245 M09-44 N08-2, N08-4, N25-156, N25-158 N13-126 J05-14 M09-83 J05-10 N20-5 N13-237 N25-188 N27-7 N06-2, N36-2 J05-10 N25-74 N25-207 N29-5 N06-7 M09-227 M05-34 N25-48, N25-154 M13-150 M13-300 M05-79, M05-106, M05-220, M13-69, M13-108, M13-339 N13-42, N25-40 499

LIU, Y. LIU, Z. -K. LJUNGBERG, M. LJUNGGREN, K. LLOP, J. LLOPART, X. LLOSA, G. LO MEO, S. LOCKE, J. B. LOCKHART, C. LOCKWOOD, J. LODDO, F. LOEF, E. V. LOEHNER, H. LOEHR, A. B. LOFFREDO, S. LOHSTROH, A. LOIS, C. LOIZEAU, P. LOKITZ, S. LOMBARDO, I. LONGONI, A. LONGTIN, R. LOOKER, Q. LOPEZ, G. LOPEZ, J. A. LOPEZ, M. LPEZ, M. LOPEZ TORRES, E. LOPEZ-JIMENEZ, M. -J. LORDI, V. LORENZ, E. LORINCZ, E. LOSACKER, M. LOUCHET, J. LOUKAS, D. LOUTCHANSKI, A. LOW, D. LOZANO, M. LU, N. LU, P. LU, Y. LUCCHESI, D. LUCENTINI, M. LUCHERINI, V. LUCIGNANI, G. LUCKEY, D. LUGERT, S. LUGIEZ, F. LUKE, P. N. LUMINARI, L. LUO, P. LUO, W. LUPPI, E. LUTTON, E. LUTZ, B. LUTZ, G. LYMANETS, A. LYNN, D. LYONS, T. LYOUSSI, A. LYRA, M. E.

J02-5, M05-88, M05-136, N13-31, N13-211 N13-57, N13-70, N37-3 M09-302 M09-119, M13-138 M09-182 N19-7 J01-2, J05-10, M05-91, M09-83, N25-139 M05-385, M13-279 N13-246 M09-77, M13-114 M05-142 N05-4, N44-1 J05-21 J01-4, M06-2 M13-345 N13-14 N31-5 M05-151 N13-140 M13-147 N25-66 N26-3 M12-6 N26-4 N05-1 N13-69 M02-5, M05-157, M09-155, M09-158 M05-175 N39-1 N14-3 N25-222 M13-9 M05-274 M13-159 M09-311 N13-221 N25-193 M05-130 N25-7, N25-35, N38-2 N25-15 N25-219 N25-23 N21-3 M09-92 N15-3 N26-3 N32-3 N32-2 N13-21, N35-6 N13-57, N13-70, N13-217, N14-2, N24-1, N24-3, N37-3 N13-86 N25-55 M05-49, M09-176 N13-87, N23-1 M09-311 N03-1 N22-1, N24-5, N35-4, N36-3 N25-187, N25-197 N34-5 N20-3 N25-76 M05-301 500

LYSETSKA, O. K. M MA, D. MA, H. MA, T. MAASS, C. MAASS, C. A. MACCHIOLO, A. MACDONALD, K. L. MACDONALD, L. R. MACE, E. K. MACHANN, J. MACKEWN, J. E. MACKIN, R. MACRAIGHNE, A. MADDEN, T. MADDEN, T. J. MADDOCK, P. MADEJSKI, G. MADHAV, P. MADI FILHO, T. MADSEN, M. MADZHUNKOV, Y. Y. MAEDA, S. MAEHLUM, G. E. MAENPAA, T. MAEO, S. MAESTAS, B. A. MAGALHAES, C. M. S. MAGAZZU, G. MAGAZZU', G. MAGLIOZZI, M. L. MAGNONI, L. MAIER, F. C. MAIGNE, L. MAINTAS, D. MAITY, T. K. MAJ, P. MAJEWSKI, P. MAJEWSKI, S. MAJOR, P. MAKISHIMA, K. MALAGUTI, R. MALAKHOV, N. MALI, W. MALTZ, J. MALTZ, J. S. MANDELLI, E. MANDI, I. MANDIC, I. MANDROU, P. MANENTI, A. MANEUSKI, D. MANFREDI, P. F. MANGEARD, M. MANGHISONI, M. MANIAWSKI, P. MANJESHWAR, R. MANJESHWAR, R. M.

N29-6

N13-26 N21-6, N43-5 M05-88, M05-112, M05-121, M05-136 HP1-2, M12-3 HP1-3, M09-380 N25-199 N13-136 J03-6, M05-94 N06-6, N13-249 M02-3 M09-248 N13-129 N38-2 N25-237 N13-98 N08-4, N25-158, N25-181 N37-4 M13-297 N25-146 M09-359 M05-343 N25-50 J01-1, M09-383 N13-192 N25-96, N25-97 N13-244 N25-10 J01-2 N25-243 M05-331, M09-92 N13-86 M11-5 M09-344 M05-208 N25-52 N19-6 N24-5, N35-4 J01-3, J01-7, M05-331, M09-92, M09-230, M13-249 M09-146 N11-5, N13-158, N37-4 M13-12 M08-8 M08-3 N13-135, N18-2 J03-4, M08-5, N18-5 N38-1 M13-234, N25-206 N08-4, N13-48 N35-6 N13-25, N13-101, N13-102, N13-103, N13-112 N38-2 N13-13 N14-3 N03-5, N25-198 M11-6 M13-330 M02-6 501

MANN, A. MANN, A. B. MANN, K. MANNHEIM, J. G. MANNING, P. MANOR, A. MANUEL PEREZ, J. MAO, R. MARACHE, S. MARACHE-FRANCISCO, S. MARAMRAJU, S. H. MARANO, G. MARCATILI, S. MARCHAL, J. MARCHANT, J. MARCHETTO, F. MARCINKOWSKI, R. MARCO-HERNANDEZ, R. MARCUS, E. MARCZEWSKI, J. MARENDIC, B. MARGOTTI, A. MARI APARICI, C. MARIAM, F. G. MARINARO, D. G. MARINELLI, M. MARIQUITO, J. C. R. MARK, E. MARKIEWICZ, P. MARKS, T. MARLEAU, P. MARLEAU, P. A. MAROCCO, D. MARONE, A. MARONE, F. MARSDEN, P. MARSDEN, P. K. MARTELLI, B. MARTELLOTTI, G. MARTHANDAM, P. MARTI I GARCIA, S. MARTIN, C. S. MARTIN, D. MARTIN, E. MARTIN, J. P. MARTIN, J. MARTIN ALBARRAN, E. MARTIN ALBARRAN, M. E. MARTIN-CHASSARD, G. MARTINEZ, N. P. MARTINEZ, O. MARTINEZ-MCKINNEY, F. MARTINEZ-MOELLER, A. MARTINEZ-MOLLER, A. MARTINI, A. MARTINS, M. C. MARTINS, M. V. MARTOIU, S. MARUHASHI, A. MARZEDDU, R. MARZOCCA, C. MASCARENHAS, N.

N25-249 M13-333, N42-4 HP2-6 M04-5, M11-5 N13-104 J04-4, N05-6, N25-133 M05-67 N32-4, N32-5, N40-2 M14-1 M09-170 M05-169, M05-247, M05-376, M09-257 M05-331, M09-92 J01-2, M05-91, N25-139 N38-2 N13-199, N25-134, N25-142 N03-4, N13-41 N25-92 N13-166 J04-4, N05-6, N13-81, N13-226, N13-230, N25-133 N25-159 M14-2 M05-391 M09-338 N13-239 J05-6 N13-18 N13-151 N14-6 M13-183 N13-106 N13-231, N13-236, N25-20 N13-212, N29-1 N13-170 N13-128 M08-2 M05-28, M13-255 M05-256, M09-41, M09-248 N13-86, N23-6 N13-180 N14-2 N13-166 N06-4, N20-3 N13-54 N25-192 N25-72 M05-127, N25-204, N31-2 N03-4 N13-41 N03-1, N19-1 M09-98 N13-69 N13-48 M05-199 M09-197 N13-86 N13-91 M05-307 N03-4, N13-41 J02-5 N25-8 N19-2 N13-236, N25-20 502

MASETTI, S. MASIK, J. MASNERI, S. MASONI, A. MASSAFRA, A. MASTROIANNI, S. MASUDA, K. MASUNAGA, S. MATELA, N. MATHEWS, S. MATHIS, F. MATHIS, J. M. MATHIS, M. J. MATHY, F. MATIS, H. MATLASHOV, A. N. MATSUDA, H. MATSUMOTO, H. MATSUNAGA, Y. MATSUOKA, M. MATSUYAMA, S. MATSUZAKI, K. MATTEUZZI, C. MATTHEWS, C. G. MATTHEWS, J. MATTHEWS, J. C. MATTIAZZO, S. MATTINGLY, J. K. MATYI, R. MAVROMANOLAKIS, G. MAXIM, V. MAY, E. MAYERS, G. MAYOROV, M. MAZUMDAR, I. MAZZA, G. MAZZITELLI, G. MCALLISTER, S. A. MCBRIDE, S. MCCLESKEY, T. M. MCCLISH, M. MCCLORY, J. W. MCCLUSKEY, C. W. MCCONCHIE, S. MCCURLEY, J. W. MCDONALD, B. S. MCDONALD, J. MCDONALD, L. R. MCDONALD, S. A. MCDOUGALD, W. MCELROY, J. MCELROY, J. E. MCGRATH, J. MCGREGOR, D. S. MCINTOSH, B. MCINTYRE, J. I. MCKENNA, M. MCKENZIE, J. MCKERNAN, M. A. MCKIGNEY, E. A. MCKINNEY, G. W. MCKINNON, A. D.

M05-391 N25-244 M05-181 N13-72 N39-1 N13-117 N25-109 J02-5 M05-307 N13-199 CH1-2 M13-147 N13-165 M09-122 N13-13 N13-227 M13-15 N13-149 M13-15 N35-7 M09-95 M09-5 N25-117 M05-49, M09-176 M13-183 M05-319 N22-7 N25-46, N25-57 N25-177 N13-197, N43-2 M13-219 N13-197 N13-48 N14-6 N36-4 N03-4, N13-41 N33-6 HP3-2 N13-61 N25-115 M13-60, N25-86, N25-131, N36-2 N25-157, N29-4 N25-62 N04-3 M13-57 M05-196, M09-116 M13-171 M13-342 M08-2 J03-6 N06-1 N13-229 N13-129 N03-6, N14-4, N25-170, N30-6 M09-356 N25-60, N25-144 N25-142 N31-4 N30-4 N25-30, N25-89, N25-115, N25-145 N09-1 N33-7 503

MCKISSON, J. MCKISSON, J. E. MCLAUGHLIN, J. J. MCLEAN, R. MCLENNAN, A. MCLEOD, I. MCMARR, P. J. MCNAMARA, J. E. MCNEIL, W. V. MCNEIL, W. J. MCPHATE, J. B. MEAD, J. MEDASANI, S. MEDATHATI, N. V. K. MEDDELER, G. MEEHAN, S. MEEKS, K. MEGNA, R. MEHRANIAN, A. MEIDINGER, N. MEIER, D. MEIKLE, S. MEIKLE, S. R. MEISSANI, M. MEKKAOUI, A. MELCHER, C. MELCHER, C. L. MELCHIORRI, M. MELFO, A. MENAA, N. MENARD, L. MENDES, L. MENESES, A. A. M. MENG, L. J. MENG, L. -J. MENNESSIER, C. MENOUNI, M. MENSAH, S. MERCIER, C. MERHOF, D. MEROLA, L. MEROLI, S. MERTENS, M. MESSA, C. METAXAS, A. METCALFE, J. METTIVIER, G. METZLER, S. D. MEURIS, A. MEYER, A. MEYER, E. MEYER, J. MEYER, T. T. MEZA, E. MHLBAUER, M. J. MICHAEL, B. MICHAUD, J. -B. MICHEL, C. MICHEL, C. J. MICHEL, T. MICHELIS, S. MICOU, C.

M05-13, M09-347, M13-249 J01-3, J01-7 N13-125 N06-7, N14-5, N31-1 M09-374 N08-4 N08-5 M09-200 N26-4 N14-4, N30-6 CN1-4, CN2-3 N38-4 M05-370 M13-213 N38-1 N27-5 N25-120 N39-1 M05-361 N13-64, N33-5 J01-1, M09-383, M11-2, M13-312 M05-166 M05-16, M09-53, M09-107 M10-6 N12-5, N13-179 N01-7, N25-128 J04-6, M01-2, M09-89, N01-2, N25-160, N25-161 M05-28, M05-91, N28-4 N13-69 N25-112 M05-25, M09-242 M05-307 M09-371 M05-325 M05-127, M05-250, M08-7, M09-11, M13-39, M13-141, N41-5 M09-317 N12-5, N13-179 M09-59 N25-72 M13-183 N13-86 N25-183 N13-109 M09-110 M13-168 N25-158 M05-385, M09-392, M13-387, M13-393 M05-196, M05-349, M09-101 N13-21 N23-3 M09-206 J05-7 N24-7 N13-69 CN2-3 M04-8 M13-354 M04-1, M05-217, M07-2, M13-198 M05-133 J05-3, M13-42, M13-345, N25-3, N25-184, N25-185, N25-191, N25-208 N25-28 N05-6 504

MIEDERER, I. MIELNIK, M. MIHAILESCU, L. MIHALCZO, J. MIKELI, M. MIKU, M. MIKU, M. MIKUZ, M. MILANO, L. MILBRATH, B. D. MILITARU, O. MILLER, B. W. MILLER, E. C. MILLER, L. MILLER, S. MILLER, T. M. MILLION, B. MILOV, A. MILOVIDOV, M. MIN, B. J. MIN, B. MINANO, M. MINATO, K. MINCER, A. MINOHARA, S. -I. MINTZER, E. E. MIO, R. MIRABBASI, S. MIRANDA LOZA, P. MIRZOYAN, R. MISHRA, K. K. MISKIMEN, R. MISSEVITCH, O. MITCHELL, G. S. MITCHELL, L. J. MITEV, K. K. MITO, Y. MITRA, D. MITSUHASHI, T. MITSUMOTO, T. MIUCHI, K. MIURA, T. MIYACHI, T. MIYAKE, H. MIYAMOTO, K. MIYAMOTO, M. MIYAMOTO, N. MIYAOKA, R. MIYAOKA, R. S. MIYASAKA, H. MIYAZAKI, O. MIYOSHI, T. MIZUMOTO, M. MIZUNO, T. MIZUTA, T. MLLER, T. MLOTOK, V. MLYNARCZYK, J. MOCCHIUTTI, E. MOCHIKI, K. -I. MOCKO, M. MOEHRS, S.

M13-33 N25-164 N13-135, N13-217, N14-2, N18-2, N18-5, N25-225, N25-226 N04-3 M05-208, M13-369 N25-206 M13-234 J05-10, M09-83, M13-120, N08-4, N30-7 M13-12 N13-208 N25-159 M09-269, M13-138, M15-5 N25-57 N25-132 J04-2, J05-14, N20-4 N09-4 N25-71, N36-5 N15-1 N14-6 M09-8, M09-86, M13-27, M13-261 M05-334, M13-258, M13-336 N13-166 M10-1 N34-5 M09-401 N25-74 M07-6 N25-216 N13-69 N40-5 N25-24, N25-25 N36-1 N32-2 M15-1 N06-2 M05-343, N25-53 N25-194 J03-2, N13-246 M13-270 J02-5 J05-13, M05-271, N13-149, N18-7, N20-1 M13-15 N05-7 N22-2 N35-3 N25-97 N25-18 M06-6 J03-6, M05-94, M06-1, M09-77, M13-114, M13-126 N14-5, N31-1 J05-13 N22-2 N07-3, N24-2 N35-7, N37-4 M05-37, M13-96 N25-159 M05-28, M05-244 N25-37 N37-2 CN2-2 N06-1, N13-229 M09-332, N25-102 505

MOELLER, T. MOLINA-PEREZ, J. MOLITSKY, M. MOLLS, M. MOLNAR, J. MOMOSE, T. MONETA, L. MONIEZ, M. MONNET, O. MONROY, M. MONTAGNA, P. MONTEIRO, C. M. B. MONTEMAYOR, A. S. MONTEMONT, G. MONTESI, M. C. MONTEVERDE, P. MONTMONT, G. MONZO, J. M. MOON, K. MOORE, J. W. MOORE, R. H. MOORE, S. C. MOORE, S. K. MOORE, T. MOORHEAD, G. MORALES, J. MORALLES, M. MORAN, J. T. MOREL, C. MOREL, F. MOREL, M. M. MORELLO, G. MORI, H. MORICH, M. MORIMOTO, Y. MORINAGA, S. MORIYA, T. MORRIS, C. MORRIS, S. J. MORRISSEY, Q. R. MORSANI, F. MORTEAU, E. MORTON, E. MOSCHINI, G. MOSER, H. G. MOSER, H. -G. MOSER, S. MOSES, W. MOSES, W. W. MOSHER, J. C. MOSHER, S. W. MOSS, C. E. MOSZYSKI, M. MOSZYNSKI, M. MOTOMURA, S. MOUGEL, F. MOUTROUSTEGUY, F. MOYA, D. MOZZANICA, A. MROWKA, S. MUEHLLEHNER, G. MUELLER, K.

M05-313 N13-75 N25-237 M13-33 J05-12, M09-152 M09-56 N23-7 N13-54 M09-122, N13-22, N13-24 N13-13 N15-3 N11-4 HP2-3 M05-76, M09-122, N13-22 M09-392, M13-387, M13-393 M10-3 N13-24 M05-10 M13-180 HPP-14, M09-269 M09-2 M13-243 M15-5 N13-94 M05-109 N36-3 N25-102 N25-157 HPP-9, M09-254 N22-5, N22-6, N25-167, N25-203 N24-7 N13-153, N44-1 M09-56 M11-6 M09-5 N05-7 M13-30, M13-267 N13-239 N13-207, N14-1 N27-7, N35-5 J01-2 M13-144, N25-38 HP2-6 M13-279, N13-33, N13-131, N25-126, N26-2 N22-1 N25-199 N25-65, N25-66 M05-268, N01-4 J01-6, M05-7, M05-64, M09-80, M10-5, N25-122, N25-143 N13-227 N09-4 N13-106, N25-62, N25-63 N01-3, N25-103 N01-7, N13-220, N25-90, N25-92, N25-104, N25-138 M09-281 N25-211 N14-3 N25-159 N25-239 N13-212, N13-236, N25-20 M01-1 M09-23 506

MUENCHAUSEN, R. MUENCHAUSEN, R. E. MUENSTER, I. MUENSTERMANN, D. MUKHERJEE, J. M. MUKHOPADHYAY, S. MULLENS, J. A. MULLER, E. MULLINS, J. T. MULNIX, T. MURA, D. MURAKAMI, T. MURAKAWA, H. MURALI, S. MURALIDHARAN, P. MURAYAMA, H. MURRAY, M. MURRUGARRA, C. MUSICO, P. MUSITELLI, G. MUSSGILLER, A. MUSSO, C. MUSTO, E. MUSUMECI, M. MUSWICK, G. MYCIELSKI, A. MYJAK, M. J. MYRONAKIS, M. E. N NACHTRAB, F. NAGAHASHI, H. NAGAI, T. NAGARKAR, V. V. NAGASAKA, Y. NAGATA, S. NAGLIS, M. NAGORNAY, L. L. NAKA, S. NAKAHARA, T. NAKAMOTO, Y. NAKAMURA, K. NAKAMURA, S. NAKAMURA, T. NAKANO, T. NAKAYAMA, H. NAKAZAWA, K. NAKAZAWA, M. NALCIOGLU, O. NAM, K. T. NAM, S. H. NAM, S. NAM, S. -H. NAQVI, A. A. NARAYANAN, S. NARBONNE, J. NARD, R. NATALI, D. NAVAB, N. NAVARRIA, F.

N29-3 N25-30, N25-89, N25-115 N25-3 N13-203 M09-191 N04-6, N13-237, N25-131, N36-1 N26-5 N38-5 N25-231 M05-154, M09-161 N13-72 M09-29, N13-146, N44-6 CN2-2 M05-121 N25-108 J04-5, M05-82, M09-32, M13-102, M13-105, M13-267 N05-2 N13-69 M05-331 M05-277 N25-159 N34-5 N13-86 M05-331, M09-92 M11-6 N25-221 N13-26, N13-207, N14-1 J03-7

N25-168 M09-401 N13-66 J04-2, J05-14, M09-14, M15-5, N20-4 N13-123 N25-84 N15-1 N29-6 N25-27 M13-75 M05-271 N18-7 N07-3, N24-2, N35-3, N42-5 N13-138, N25-19, N25-83, N25-84 M15-4 M13-270 M09-95, N37-4 M05-1 J01-1 M09-233 M09-233, N25-36 M05-400 J05-18, M09-272 N25-73 N13-98 N35-6 M05-277 N28-7 M05-199 M13-279, N13-33, N25-126, N26-2 507

NAVARRIA, F. L. NAVILIAT-CUNCIC, O. NAZARYAN, V. NEAL, J. S. NECIB, H. NECULAES, V. B. NEKOLLA, S. NEKOLLA, S. G. NELSON, A. T. NELSON, A. J. NELSON, K. NEMETH, G. NESSI-TEDALDI, F. NETTER, E. NEUBAUER, M. NEUBERT, S. NEUKIRCHEN, C. NEVES, P. N. B. NEVIN, J. NEVIN, J. D. NEWCOMER, F. M. NEWCOMER, M. NEWMAN, G. NG, T. NGO, H. NGUYEN, H. NGUYEN, T. NGUYEN, V. -G. NGUYEN-HU, X. NICOL, S. NICOLETTO, M. NIEDERLOEHNER, D. NIELSEN, J. NIESSEN, W. J. NIKITA, K. S. NIKL, M. NIKOLIC, R. J. NIKOLOGIANNIS, A. NILLIUS, P. NINKOVIC, J. NINO, J. C. NISBET, A. NISHIKAWA, R. M. NISHIKIDO, F. NISHIMORI, N. NISHIMURA, H. NISHIMURA, K. NISIUS, R. NITIN, G. NIU, H. NIU, M. NIZERY, F. NOELDGEN, H. NOGUEIRA, L. P. NOH, J. NOLAN, P. J. NOO, F. NORBECK, E. NORDT, A. NORENBERG, J. P. NORGREN, M. NORLIN, B.

N13-131 N25-107 M09-230 N25-106, N25-135 M13-357 M12-6 M13-177 M09-197 N13-3 N25-229 N06-7, N13-247, N18-6 HP2-4, M09-146 N32-3 M05-25, M09-242 N25-254 N44-3 M08-3 N13-152 M10-3 M09-314, M13-315, M13-318 N13-48 N13-39, N13-46 M13-282 M05-328 M13-99 N13-106, N25-62 M09-395 M13-24 N02-5 M09-254 N21-3 M05-43, M09-365 N13-104 M09-203 M05-358 N25-113, N25-114, N25-121 N02-6, N25-229 N13-34 M14-4 N22-1 N25-188 N25-5 M09-2 J04-5, M05-82, M09-32, M13-102, M13-267 N25-109 N18-7, N20-1 N13-201, N45-7 N25-199 N25-176 N25-6 N25-151 N13-141, N25-67 N25-99 M09-398 N13-130, N13-133 J05-16, N13-129 M02-2, M03-6 N13-197 N36-3 M13-171 N13-104 N25-195 508

NOURBAKHSH, S. NOVAK, D. NOVIKOV, I. NOVIKOVA, E. I. NOVOTNY, R. W. NOY, M. NUESSLIN, F. NUNEZ, L. A. NURITDINOV, I. NUYTS, J. NYGARD, E. O OKEEFE, G. OBERLA, E. OBERLACK, H. G. OBRYK, B. OCHOA, H. J. O'CONNOR, P. ODA, K. ODAKA, H. OELHAFEN, M. OGAWA, K. OGDEN, R. T. OGER, T. OH, K. M. OH, K. OH, K. -M. OHAMA, T. OHGAKI, H. OHI, J. OHNO, M. OHSHITA, H. OHTA, M. OHTANI, A. OJHA, N. OKADA, H. OKADA, T. OKAMOTO, T. OKAZAKI, M. O'KEEFE, G. J. OKERLUND, D. OKUDAIRA, O. OLCOTT, P. OLCOTT, P. D. OLDENDICK, J. OLESEN, O. V. OLIVEIRA, C. A. B. OLIVEIRA, C. A. OLIVEIRA, L. F. OLIVER, K. OLKO, P. OLSCHNER, F. O'NEAL, J. P. O'NILS, M. ONISHCHENKO, G. M. ONO, K. ONODERA, C. OPEIL, C. P. OPPOSITS, G.

M13-279 J05-12, M09-152 N13-215 N13-213 N32-2 N03-4, N13-41 M13-33 N13-69 N13-163 M05-214, M09-140, M13-153 M08-8

N25-1 M05-7, M09-80, N27-5 N13-169 N04-5 M13-294 M05-265, M15-2, N13-52, N38-4 M13-105 M15-4, N13-61, N41-3 M13-207, M13-210 M05-271, M05-382, M09-341, M13-75, M13-360 M05-373 N25-38 M09-233 M05-400 M09-272 N13-154 N25-109 M05-1, M13-96, N27-2 M09-56 N44-6 N07-3, N24-2 M05-37, M13-96 M09-245, M11-6 N25-27 M05-271 M13-30 CN2-2 M05-124, M13-399, N13-4, N13-5, N25-173 M12-1 N05-7 M06-4 M13-6, M14-5 M05-58, M09-71, M13-93 M09-224, M13-225 N11-4 J04-7 M09-371, M09-398 M04-8 N04-5 N25-230 M05-64 N25-195 N29-6 J02-5 N25-172 N29-5 J05-12, M09-152 509

ORAVA, R. N25-35 ORBOM, A. M05-139, M09-119 M05-58, M13-93 ORDONEZ, C. ORDONEZ, C. E. M09-71 ORECCHINI, D. N15-3 N35-1 ORITO, R. ORLANDI, S. N25-28 N25-74 ORRELL, J. L. ORRIGO, S. E. A. N25-65 N22-6, N25-167, N25-203 ORSINI, F. ORSOLINI CENCELLI, V. M13-279, N13-18, N13-33, N25-126, N26-2 ORTEGA MYNEZ, L. M13-294 M13-285, M15-6 ORTUNO, J. E. OSADA, H. J04-5, M09-32, M13-267 O'SHEA, V. N38-2 N07-3, N24-2, N42-5 OSHIMA, M. OSOVIZKY, A. J04-4, N05-6, N13-81, N13-226, N13-230, N25-133 OSTLUND, J. N13-247 OSUKA, S. M13-30 O'SULLIVAN, F. M13-282 M05-304, M09-215 O'SULLIVAN, J. O'SULLIVAN, J. A. M09-362, M13-48 J05-6 OTHMAN, M. A. R. OTINIANO ORMACHEA, L. J. N13-69 OTIS, B. P. J03-6 OTOMO, T. N44-6 O'TOOLE, J. A. N13-1, N13-2, N13-3 OTSUKA, Y. CN2-2 OTTE, A. N. N13-67, N37-5 N04-1 OUEDRAOGO, S. OULD MOHAMED, M. S. M09-317 OUVRIER BUFFET, P. N25-211 OUYANG, J. M11-3 OUYANG, Q. N13-143 OVECHKINA, O. J04-2 OWENS, A. N31-7, N41-4 N25-82 OWENS, I. J. OWENS, T. N13-227 N13-125 OWSLEY, L. M. D. OXLEY, D. C. J05-16, N13-129 OYA, P. M09-383 J02-1, N13-74, N25-2 OZKAN, C. P PACKARD, N. PAGANI, G. PAGANO, A. PAGANO, D. PAIDI, A. PAL, D. PALMER, D. PALMER, D. M. PALMONARI, F. PAN, T. PAN, X. -B. PAN, X. PANAYIOTAKIS, G. PANDOLA, L. PANI, P. PANI, R. M13-3 N28-7 N25-66 M05-277 J03-4, M08-5 M09-245, M11-6 N06-1 N13-229 N13-115 M05-46, M13-402 M05-292, M13-177 M05-22, M09-2, M09-386, M12-5, M13-3 M09-212 N13-43, N25-213 M13-279 M13-279, N13-33, N13-131, N25-126, N26-2 510

PANI, S. PANIN, V. Y. PANTANO, D. PAOLUCCI, P. PAPALEO, R. PAPANICOLAS, C. PARAGES, F. M. PARETO, D. PARK, B. H. PARK, H. PARK, J. A. PARK, J. H. PARK, J. P. PARK, K. PARK, M. -A. PARK, M. -J. PARK, S. K. PARK, S. H. PARK, S. -H. PARK, S. PARK, S. Y. PARK, W. H. PARK, Y. PARKER, J. D. PARKER, R. F. PARKER, S. I. PARKES, C. PARL, C. PARODI, K. PARSEY, R. V. PARZEFALL, U. PASCHALIS, V. PASSERI, D. PATEL, S. PATLOLLA, D. R. PATT, B. E. PATTON, S. PAUL, D. PAUL, P. PAUL, S. PAUSCH, G. PAUSS, F. PAUWELS, B. PAVIA, J. PAVLOV, N. PAYNE, R. PAYNE, S. A. PAYNE, S. PAYTON, A. PEAKE, D. PEAKE, D. J. PEARCE, M. PEARSON, E. PECZELI, I. PEDASH, V. Y. PEDEMONTE, S. PEDRETTI, A. PEIZERAT, A. PELC, N. PELIZZARI, C. A. PELLEGRINI, G. PELLEGRINI, R.

N13-206 M05-202, M05-289 N22-7 N13-184 N13-200 M05-208 M05-286 M05-151, M05-163, M09-182 N13-148 N25-202 M05-97, M05-115, N25-14 N13-83 N25-162 J05-23 M13-243 M09-62 M09-128 N13-148, N25-162, N25-190 N25-227, N25-228 N45-6 M05-232 N04-2 M09-266 N18-7, N20-1 N13-106, N25-62 N13-202, N25-182, N38-1 N38-2 N25-251 M05-238 M05-373 N02-4, N25-201 CH1-4 N25-183 M09-113 M09-104 J01-1, M09-383, M13-312 N17-3 J05-4 N14-3 M13-333, N25-249, N42-4, N44-3 N01-3, N13-220, N13-224, N13-244, N25-92, N25-104 N32-3 N25-40 M05-151 M10-5 N25-60 N25-122, N25-149, N25-229 N01-1, N01-4, N01-6 N13-129 N38-3 M13-399, N13-4, N13-5, N25-174 N13-58 M12-5 M05-274 M09-368, N13-105 M09-329 N26-3 N13-22, N13-24 M12-6 M05-22, M12-5 N25-7, N25-35, N38-2 M13-279, N25-126 511

PELOSO, R. PELTONEN, J. PELTONEN, S. PENA, E. G. PENATZER, J. PENFOLD, S. N. PENG, H. PENG, Q. PENG, S. PENNICARD, D. PENSO, G. PENUMADU, D. PEPIN, C. PEPIN, C. M. PEREGO, D. L. PERES, M. PEREVERTAYLO, V. PEREVOZTCHIKOV, V. PEREZ, G. PEREZ, J. M. PEREZ, K. L. PEREZ, Y. PEREZ-BECKER, D. PERIC, I. PERINI, L. PERNECKER, S. PERREY, H. PERROTTA, A. PERSSON, A. PESSINA, G. PESTOTNIK, R. PETASECCA, M. PETER, Z. PETERSON, M. PETERSON, M. I. PETERSON, T. E. PETROSKY, J. C. PETRUCCI, F. PETULLA', F. PETYT, D. A. PEYRIN, F. PFUND, D. M. PHELPS, M. E. PHILIPPE, L. PHILLIPS, S. PHLIPS, B. PHLIPS, B. F. PI, H. PIA, M. G. PIAZZA, A. PICHLER, B. J. PIEMONTE, C. PIENDIBENE, M. PIERCE, L. A. PIERLUIGI, D. PIETRAS, B. PIGNATEL, G. U. PIKNA, M. PILI, C. PINCI, D. PINELLI, D. PINELLI, D. A.

N26-3, N28-7 N35-5 M09-218 M05-403 M11-6 HP3-2 M06-4, M13-6, M13-351, M14-5 M10-5 N25-23 N38-2 N13-180 N25-132 N13-30 J05-11, J05-15 N25-117 J04-7 J02-7 N13-52 N13-69 M05-103 M13-297 N13-69 N13-57, N13-70, N37-3 N13-196 N13-86 N13-39 N25-239 M13-84, M13-279 M11-4 N13-42, N25-117 N25-118 J02-7, J05-6, M05-109, M09-236, N13-241 M09-290 M05-139 M13-138 M05-196, M09-116 N25-157, N29-4 N13-153 N13-18, N13-33, N13-131, N26-2 N43-4 M09-290 N13-208 M09-50, M13-372, M15-7 N33-3 N13-48 N25-29 N05-5, N06-2, N13-213, N20-6, N25-175 N13-85, N17-2 N09-2, N09-3, N09-5, N09-7, N13-79, N13-80, N13-91, N33-2, N33-3, N33-4, N39-7 N28-4 M04-5, M05-259, M11-5, M13-9, M13-246 M05-28, M05-91, M05-244, N02-4, N25-139, N28-4 N21-3 M06-6 N15-3 N13-129 N25-141 N13-7 N25-8 N13-180 N25-218 N19-3 512

PINHEIRO, C. J. G. PINHEIRO, C. J. PINOT, L. PINTILIE, J. PIRARD, B. PIRRONE, S. PISANI, R. PISTILLI, M. PISTOLESE, M. PIVEN, L. A. PIVOVAROFF, M. PLACIDI, M. PLANETA-WILSON, B. PLATSCH, G. PLETTNER, C. PLIMLEY, B. C. PLONKA, C. POCHET, T. POIRRIER, E. POITRASSON-RIVIRE, A. POLA, A. POLI, G. POLITI, G. POLL, A. POLTORAK, K. POMMES, K. POMORSKI, M. PONCE, E. PONCET, J. -F. PONS, R. POOK, D. POON, J. K. POPKOW, A. POPOTA, F. D. POPOVIC, K. PORCEDDU, I. PORRO, M. PORTER, L. J. PORTO, F. POSPISIL, S. POTDEVIN, G. POTIRIADIS, C. POULADIAN, M. POUMAREDE, B. POURHAMZEH, B. POURMOGHADDAS, A. POVEDA, J. POWOLONY, F. F. POZZI, S. A. POZZI, S. POZZOBON, N. PRADIER, O. PRASAD, R. PRATTE, J. F. PRATTE, J. -F. PREKAS, G. PRELL, D. PRESCHER, G. PRESSYANOV, D. S. PRETORIUS, H. PRETORIUS, P. H. PRETTYMAN, T.

M09-371 M09-398 M05-25, M09-242 N25-239 N25-215 N25-66 N15-1 N44-1 N13-86 N29-6 N13-247 N13-13 M03-3, M05-154, M09-161 M13-183 N01-3, N13-220, N13-224 N13-137 N25-107 N29-6 N14-3 N13-244 N25-4 N26-3 N25-66 M13-159 N13-39 N13-75 N13-45 N13-69 N36-3 N13-53, N35-6 M05-124 M13-72, M13-348 N13-137 M05-151 M05-16 N13-72 N19-4, N24-5, N35-4 NP-2 N25-66 J05-6, M09-236, N19-7 N25-239 N13-221 M13-111 N13-234 N13-137 M09-20, M13-162 N13-113 N24-7 N04-4, N13-232, N13-235, N13-243, N13-244, N25-57, N25-64, N30-3 N13-96 N22-7 M05-46, M13-402 M09-308 M05-247, M05-376 M05-169, M09-257, M15-2, N13-30, N38-4 N31-5 M08-4 J04-1, N28-5 N25-53 M09-200 M09-173 N13-237 513

PREZADO, Y. PRIEGNITZ, M. PRINJA, A. K. PRINO, F. PRISBY, R. PROFFITT, J. PROISSL, M. PROKOPOVICH, D. PROKOPOVICH, D. A. PROKSA, R. PROSPERINI, A. PROSSER, A. PROST, R. PROTSENKO, M. PRUESSMANN, D. F. PRUSSIN, S. G. PRYDDERCH, M. L. PUGLIESE, G. PULKO, J. PULLIA, A. PURNIMA, S. PURSCHKE, M. PURSCHKE, M. L. PUSHKARSKY, V. PYM, A. T. G. Q QI, H. QI, J. QI, Y. QIAN, H. QIU, W. QUARATI, F. QUARTA, G. QUARTA, M. QUEIROZ, P. QUEIROZ, P. P. QUINTERO, C. QUINTERO-SEGOVIA, A. QUINTIERI, L. QUISPE, J. R RA, J. B. RAASHID, M. RABIN, M. W. RACHINHAS, P. J. B. M. RACITI, G. RACZ, A. RADEKA, V. RADERMACHER, E. RADICS, B. RADLEY, I. RAFANELLI, N. RAFECAS, M. RAGKOUSIS, L. RAHMER, J. RAHMIM, A.

N39-6 M05-238 N09-6 N07-4, N13-116 M09-290 M13-249 N15-1 J05-6 J02-4, J02-7 M08-6 N23-6 N24-6 M13-219 J05-10 N13-3 N25-225, N25-226 N27-7, N35-5 N05-4 M09-263 N13-32, N13-37, N13-43, N25-196, N25-247, N27-3 M13-237 M05-169, M05-247, M05-265 J03-5, M05-376, M09-257, N42-3 J04-4, N05-6, N13-81, N25-133 N25-231, N25-233, N25-234

N13-142 M03-4, M03-5, M05-34, M07-1, M09-125, M09-134, M13-51, M13-129, N07-6 M09-260, M13-150 M13-321 N25-188 N41-4 CH1-3 N39-1 N09-3, N39-7 N13-91, N13-95 N13-69 N13-246 N09-3, N13-79, N33-4, N33-6 N13-69

M09-404 N25-73 N25-42, N25-58, N25-82 N13-157 N25-68 HP2-4 N27-6 N34-1 N13-78 N25-232, N25-233, N25-234 N21-3 M05-199, M14-3 M05-208 M15-3 M03-8, M13-195 514

RAIMUND, E. M04-8 N25-160, N25-161 RAKOV, A. F. RALPH, W. N01-6 N13-133, N25-179 RAMADOSS, S. RAMBAUD, D. N13-53, N35-6 N01-1 RAMEY, J. RAMEY, J. O. N25-135 RAMIREZ, J. M02-5, M05-157, M05-175, M09-155, M09-158 RAMIREZ, R. M05-79, M13-108 RAMIREZ, R. A. M05-106 RAMIREZ JARAMILLO, R. A. M13-69 N13-109 RAMM, M. RAMON, P. N35-6 N19-3, N25-218, N25-224 RAMSEY, B. D. RANA, V. N31-1 N25-8, N39-6 RANDACCIO, P. RANIERI, A. N44-1 RANNOU, F. M09-299 M05-172, M15-7, N13-97 RANNOU, F. R. N38-5 RAO, T. RAPISARDA, E. N25-68 N23-3 RAPSEVICIUS, V. RARBI, F. N03-2 RASSOOL, R. P. M05-124, M13-399, N13-4, N13-5, N25-173, N25-174, N38-3 RATCLIFF, B. N13-201, N45-7 RATERING, D. M07-8 RATIB, O. M09-308, M13-177 M05-103 RATO MENDES, P. M05-67 RATO MENDEZ, P. RATTI, A. N13-13 RATTI, L. N03-5, N22-5, N25-198 RATTI, S. M05-277 M02-2, M09-206, M12-4 RAUPACH, R. RAUX, L. N03-1 RAVINDRAN, B. P. M13-237 RAVINDRANATH, B. J03-5, M05-169, M05-247, M05-376, M09-257 N15-1 RAVINOVICH, I. RAY, S. N13-247 RAYLMAN, R. M13-249 RE, V. N02-3, N03-5, N22-5, N25-198 READER, A. J. M07-6, M07-7, M13-21 N13-86 REBATTO, D. REBUFFEL, V. M05-76 RECCHIA, F. N25-71 RECHKA, S. M13-354, M14-3 REEDY, E. T. E. N13-218, N13-225 REESE, T. M11-3 M05-229 REESE, T. G. N19-3, N25-218, N25-224, N38-4 REHAK, P. REHFELD, N. M13-357 REHLICH, K. N13-121, N25-252, N25-253 REHMAN, K. U. N25-73 N13-214 REICHENBACH, B. REID, C. D. J01-3 REIFARTH, R. N25-115 REILHAC, A. M05-214 REINECKE, M. N13-182 REINHARD, M. J05-6 REINHARD, M. I. J02-4, J02-7 REINHARDT, C. E. N02-6 REINISCH, L. J05-7 REIS, J. M09-353 515

REISCHEL, G. REISER, I. S. REMIREZ, R. REMOUE, N. RENAUX, C. RENISCH, S. RENKER, D. RENZI, R. F. REPOND, J. RESCIA, S. RESCONI, S. RESNICK, P. J. RETICO, A. RETIERE, F. REUTTER, B. W. REYNOLDS, J. REYNOLDS, P. D. REZNIK, A. RHODES, W. H. RIAHI ALAM, N. RIBOLDI, S. RIBON, A. RICCARDI, A. RICCI, P. RICCIO, F. RICHER, J. -P. RICHTER, J. RICHTER, R. RICHTER, R. H. RIDOLFI, S. RIEDEL, R. A. RIEDLER, P. RIFFAUD, B. RILEY, K. J. RINALDI, L. RINGBOM, A. RINKEL, J. RIOT, V. J. RIPAMONTI, G. RIPOLL, J. RISSI, M. RITMAN, J. RITSCHL, L. RITT, S. RITTER, G. RITZERT, M. RIVA, M. RIVERA, H. RIVERA, R. RIVERA, R. A. RIVETTI, A. RIZZO, F. RO, Y. ROBERT, C. ROBERT, G. ROBERTS, J. ROBERTS, O. J. ROBERTSON, R. ROBINSON, D. J. ROBSON, C. C. W. ROCCHISANI, J. -M. RODGERS, D. J.

M11-5 M09-2 M13-339 N13-53, N35-6 N25-192 M05-28, M11-6 M13-9 N25-16 N13-197 M05-247, N13-48, N21-1, N27-6, N34-5 N13-86 N13-214 N39-1 N40-6 J01-6, M09-338, M09-377 M12-6 M13-135 M05-280 J05-21 M05-361 N13-25, N13-37, N13-43, N13-103, N25-247, N36-5 N13-89 M05-181 N23-6 N25-65, N25-66 N13-24 M09-302 N25-199, N33-5 N13-35, N22-1, N24-5, N35-4 N25-126 N25-130 N03-4 N36-3 N20-4 N13-86 N25-60 N25-211 N42-6 N13-102 M05-283 N35-2 N13-109, N15-3 M13-381 N13-17 N25-22 M05-28, M05-244 N13-170 N13-69 N24-6 N13-110 N03-4, N07-4, N13-41 N25-66 M09-185 M05-76 M15-8 N01-6 J05-1 M15-1, N25-86 N13-26 N13-121 M09-311 N13-63 516

RODRIGO, T. RODRIGUESB, M. R. D. ROE, G. ROECK, W. W. ROEDER, M. ROEMER, K. ROESSL, E. ROGEL, G. ROGERS, W. L. ROHEL, J. ROHLF, J. ROLNICK, S. ROMA, L. ROMANOV, L. ROMANOV, L. V. ROMOLI, M. RONCALI, E. RONCHIERI, E. RONCHIN, S. ROOS, M. ROPELEWSKI, L. ROS, A. ROS, D. ROSALES, M. ROSA-NETO, P. ROSELLO, J. ROSEN, B. R. ROSENFELD, A. B. ROSNER, G. ROSS, M. ROSS, S. ROSSETTO, O. ROSSI, P. L. ROSSO, V. ROSTAING, J. -P. ROTA KOPS, E. ROTH, J. ROTHFUSS, H. ROTHFUSS, H. E. ROTONDI, A. ROUAIX, G. ROUSSEAU, M. ROUX, C. ROVERO, A. C. ROWLANDS, J. A. ROY, J. ROYLE, G. RUAN, D. RUBINOV, P. RUCHTI, R. RUCKMAN, L. RUCKMAN, L. L. RUDDY, F. H. RUDDY, T. D. RUDIN, M. RUDIN, S. RUDOLF JANIK, R. RUDY, C. R. RUGGIERO, G. RUOTSALAINEN, U. RUSS, W. RUSSO, A.

N25-159 N25-65 M13-399 J01-1 N15-3 N01-3, N25-92 M08-6 N25-107 HPP-13 N13-135, N18-2, N18-5 N13-167 N15-1 M05-391 M10-3 M09-314, M13-315 N36-7 M05-61, M13-60 N17-1, N23-6 N02-4 N13-82 N11-1 M09-149 M05-151, M09-98 N13-69 M07-6 J02-6 M04-1 J02-4, J02-7, J05-6, M05-109, M09-236, N13-241 N05-2 N25-147 M13-330 N13-24 M05-391 M09-332, M13-90, N25-102 N13-22, N13-24 M04-4, M07-2, M09-287 N25-184 M05-70 M05-202 N15-3, N25-68 N35-6 J05-1 M05-46, M13-402 N13-69 M05-280 M12-6 J02-1 M05-388 N36-6 N13-199, N25-134, N25-142 N13-201, N45-7 N27-5 N41-1 M05-142 M07-8 M08-1 N13-7 N25-42 N10-1 M09-218 N25-56 N15-3 517

RUSSO, P. RUSSO, S. RUSSOTTO, P. RUZ, J. RVACHOV, T. RYAN, J. RYAN, J. M. RYZHIKOV, V. D. S SAAD, A. SAADATMAND, K. SAAVEDRA, O. SABET, H. SACCO, D. SACHFERS, K. SADROZINSKI, H. F. W. SADROZINSKI, H. F. -W. SAEY, P. SAFAVI-NAEINI, M. SAINT JAMES, S. SAITO, K. SAITO, S. SAITO, Y. SAJI, H. SAKAE, T. SAKAGUCHI, T. SAKAI, T. SAKALAUSKAS, S. SAKAMOTO, Y. SAKASAI, K. SAKATA, M. SAKIMOTO, T. SAKUMURA, T. SAKURAI, Y. SAKWE, S. A. SALACKA, J. S. SALAMANNA, G. SALAS-GONZALEZ, D. SALAYMEH, S. SALAZAR, H. SALENO, M. R. SALIKHBAEV, U. S. SALIKHBAYEV, U. SALOMON, J. SALOMONI, D. SALVACHUA, B. SALVADOR, S. SALWEN, C. SAMAIN, L. SAMARIN, A. SAMMET, J. SAMPIETRO, M. SANAMI, T. SANCHEZ, F. SANCHIS, E. SANDIN, H. SANDNESS, G. A. SANDY, A. R. SANNER, R.

M05-385, M09-392, M13-387, M13-393 N28-6 N25-66 N36-3 M05-280 N29-2 N25-22 N29-6

M05-313 N29-2 N13-69 M09-95 N13-153, N25-126 M13-294 N13-48 N05-5, N25-181 N25-60 M05-109 M05-61 M13-105, N13-146, N13-149, N25-119 N13-61, N41-3 M13-270 M05-271 J05-13 N15-1 M13-30 N25-159 N13-123 N13-138, N25-19, N25-83, N25-84 M05-382 M10-1 N41-6 J02-5 N25-231, N25-233 N25-82, N25-145 N10-5 M02-5, M05-157, M05-175, M09-155, M09-158 N13-125 N13-69 N41-7 N25-160 N13-163 CH1-4 N13-86 N13-113 N25-101 N13-233 CH1-2 M13-54 N25-159 N28-7 N13-146, N25-119 M09-149 N13-113 N13-227 N06-5 N13-98 N01-6 518

SANTAVENERE, F. SANTIN, G. SANTOCCHIA, A. SANTOS, A. SANTOS, C. A. SANTOS, D. D. S. SANTOS, D. S. SANTOS, F. P. SANTOS, L. A. P. SANUKI, T. SAPONARA, S. SAPUNENKO, V. SARACCO, P. SARASOLA MARTIN, I. SARDARI, D. SARIN, V. SARRHINI, O. SARRUT, D. SARTORI, L. SARUSI, B. SASAKI, S. SASAKI, T. SASOV, A. SASSI, M. SATO, G. SATO, R. SATO, T. SATO, Y. SATOH, M. SATOH, S. SATTARI, S. SAUER, K. SAUNDERS, A. SAUTER, A. SAUVE, A. C. SAVELIEV, V. D. SAVINE, A. SAWANO, T. SAWHNEY, K. SAWIAK, S. J. SBARRA, C. SCAFE', R. SCHAART, D. R. SCHAEFER, D. SCHAEFER, W. SCHAEFERS, K. P. SCHAEFFTER, T. SCHAFERS, K. P. SCHAFERS, M. SCHAKNOWSKI, N. A. SCHALLER, G. SCHEINS, J. J. SCHELTEN, J. SCHERL, H. SCHEUERMANN, J. S. SCHIAVONE, F. SCHICK, F. SCHILLINGER, B. SCHINDLER, T. H. SCHIOPPA, M. SCHIPPER, J. D. SCHIRMER, J.

M09-92 N13-63 N45-5 M09-332, M13-90, M13-285, M15-6 J04-7, N13-161 N13-95 N13-91 N13-157, N25-124 N25-10 N25-163 N25-243 N23-6 N09-3, N09-5, N13-79, N33-4 M05-103 M13-111 J05-14 M13-375 M05-235 N25-250 N13-81, N13-226 N13-146, N13-149, N25-119 N39-2, N39-4 N25-40 N36-5 N25-194, N41-3, N41-6 N37-4 J02-5 M13-105, N25-50 M05-1 J04-5 N25-156 M13-201 N13-239 M05-259, M11-5 M09-377 N25-180 N13-171 N18-7 N38-2 HP3-3, M09-167 N25-68 M13-279, N25-126 J01-4, J03-1, J05-9, M06-2, M06-3 M09-203 M11-6 M09-38, M13-324 M05-28, M05-256, M09-41, M09-248 M04-7 M04-7 N11-7 N35-4 M04-4, M07-2 N25-99 HP1-1 M13-66 N13-65 M02-3 CN2-3 M13-177 N13-153, N44-1 N12-5, N13-179 M09-326 519

SCHIRRU, R. SCHLEMMER, H. P. SCHLEPER, P. SCHLEYER, P. J. SCHLOMKA, J. -P. SCHLYER, D. SCHLYER, D. J. SCHMALBROCK, P. SCHMAND, M. SCHMID, A. SCHMIDT, A. SCHMIDT, B. SCHMITT, B. SCHMITZ, A. SCHNECKE, M. SCHNEEBELI, M. SCHOBER, O. SCHOELKOPF, B. SCHOLZE, F. SCHOPPER, F. SCHOTANUS, P. SCHOUKENS, J. SCHROEDTER, M. SCHROM, B. T. SCHUBERT, A. SCHUBERT, K. E. SCHUHMACHER, H. SCHULCZ, F. SCHULMAN, T. SCHULTE, R. SCHULTE, R. W. SCHULTZ, D. SCHULTZ, L. SCHULTZ, L. J. SCHULTZ-COULON, H. -C. SCHULZ, D. SCHULZ, V. SCHUTTAUF, A. SCHUWALOW, S. SCHWANDT, J. SCHWEINERT, G. SCHWEPPE, J. E. SCHWIENING, J. SCHWOEBEL, P. R. SCIOLLA, G. SCOTT, A. SCOTT, C. D. SCOTT, K. SCOTT, P. D. SCRAGGS, D. P. SEABURY, E. H. SEBASTIA, A. SEFRI, R. SEGA, T. SEGARS, P. W. SEGARS, W. P. SEGHOUR, A. SEGONI, I. SEGOVIA, F. SEIDEL, J. G. SEIDEL, S. SEIDEN, A.

M09-371 M05-259 N25-159 M05-256 M08-6 M05-169, M05-265 M05-247, M05-376, M09-257 M13-120 M07-2, N28-1 M05-259 M11-5 M12-4 N25-239 J04-1, N28-5 N35-4 N25-9 M09-38 M03-2 N24-5 N35-4 N13-220, N25-104 M05-85 N10-2 N25-144 M05-124, N25-173, N25-174, N38-3 HP3-2 N13-170 N05-6 N41-5 N09-3, N39-7 HP3-2, J02-7, N13-144 N29-4 N13-245 N06-1, N13-227, N13-229, N13-239 N13-36 M05-169, M05-247, M15-2 M05-28, M05-244, M05-256, M09-41, M11-6, M13-252, M13-255, N28-4 N13-44, N13-45 N08-6 N25-159, N25-239 N25-176 N06-5 N25-39 N13-214 N16-2 M05-124 N13-126 N25-144 N25-232 J05-16 N13-150, N13-242 M05-10, M09-149 N13-54 N22-2 M14-1 M05-40 N25-186 N23-3 M02-5, M05-157, M05-175, M09-155, M09-158 N41-1 N25-158 N13-48 520

SEIF, R. SEIFERT, A. SEIFERT, C. E. SEIFERT, S. SEIPEL, H. A. SEITZ, B. SEKIMOTO, M. SEKIWA, H. SELLER, P. SELLERS, C. SELLIN, P. J. SEN, I. SENDA, M. SENZIG, B. SENZIG, R. SEO, C. SEO, H. SEO, J. H. SERIEF, C. SERVAGENT, N. SERVOLI, L. SEVILLA, S. G. SFIENTI, C. SHA, W. SHAH, K. SHAH, K. S. SHAKIRIN, G. SHAO, B. B. SHAO, B. SHAO, L. SHAO, Y. SHAPIRO, A. H. SHARMA, A. SHARMA, D. SHARMA, S. L. SHCHERBININ, S. SHEARER, E. SHEETS, S. SHEINFELD, M. SHELDON, P. SHEN, L. SHEN, W. SHENAI, A. SHEVIAKOV, I. SHI, B. SHI, C. SHI, D. SHI, K. SHI, X. SHIBAMURA, E. SHIBUYA, K. SHIGA, T. SHIGEMATSU, N. SHIH, Y. -C. SHIKAMA, T. SHIMADA, H. SHIMAZOE, K. SHIMIZU, H. M. SHIMIZU, K. SHIMIZU, S. SHIMOSEGAWA, E. SHIMURA, N.

N05-6, N13-230 N25-74 N13-207, N14-1 J01-4, J03-1, J05-9, M06-2, N25-237 N13-218, N13-225 N05-2 M09-29, N44-6 N25-97 N35-5 N13-94 N31-5, N41-1 N25-132 M13-63 M12-1 M12-6 M09-266 N13-83 M05-52 M09-179 M13-144 N25-183 N13-39 N25-68 M13-372 J03-3, M05-268, M05-328, N01-1, N25-137, N25-230 J01-5, J05-21, M09-134, M13-60, N01-5, N25-81, N25-86, N36-2 M05-238 N25-245 N13-10, N25-61 M09-26, M09-191, M11-6 J01-5, M05-31, M13-87, N07-5 N42-1 N11-2 N15-1 N25-52 M05-295, M09-164, M09-209, M13-192 N13-199 N01-4, N25-122 N13-226 N13-94 HPP-6 N03-1, N13-36 N27-1 N25-239 N13-49, N13-156 M13-189 HPP-7 M13-33 N25-239 N25-119 J04-5, M05-82, M09-32, M13-267 M09-5 M13-75 J03-6 N25-84 M15-4 M09-56, N13-49, N27-2 N25-163 M13-63 J05-11, J05-15, N25-80 M09-251 N25-80 521

SHIN, J. -W. SHIN, S. H. SHIN, S. SHINDE, Y. SHIRAHATA, T. SHIRAN, N. V. SHIREY, R. SHIZUMA, T. SHLEGEL, V. SHOJI, T. SHOKAIR, I. SHOKOUHI, S. SHORT, J. SHUBIN, V. SHUSHAKOV, D. SHUTO, K. SIBCZYSKI, P. SIBCZYNSKI, P. SIBOMANA, M. SICILIANO, E. R. SIDDHANTA, S. SIDDONS, D. P. SIDKY, E. SIDKY, E. Y. SIDLETSKIY, O. T. SIEBERT, R. SIEGEL, S. B. SIEGMUND, O. H. W. SIEVERS, P. SIFFERT, P. SIGEMORI, A. SIGNORONI, A. SIKORSKI, M. SILVA, A. L. SILVA, E. C. SILVA, M. D. SILVA, R. S. SILVERMAN, R. W. SIM, K. S. SIM, K. -S. SIMAS FILHO, E. F. SIMBURGER, G. SIMON, F. SIMONIS, H. J. SIMPSON, J. SINGH, B. SIPILA, H. T. SIPILA, H. J. SISNIEGA, A. SITAR, B. SITARSKY, K. SITEK, A. SITTA, M. SIVASWAMY, J. SJODEN, G. E. SJOSTRAND, K. SKLYAR, A. SKOUROLIAKOU, K. SLATER, J. M. SLAUGHTER, R. C. SLONE, D. SMALL, G. W.

M09-272 M09-8, M13-27, M13-261 M05-334, M13-258, M13-336 N05-4 J05-13 N25-91 N25-82 N25-109 N25-138 N25-172 N13-247 M05-196, M09-116 M12-6 N25-111, N28-3 N25-111, N28-3 M13-75 N25-103 N25-90 M05-133, M09-224, M13-225 N06-6, N13-249 N25-8 N13-12, N19-3, N25-218, N25-224, N38-4 M09-386, M13-3 M09-2, M12-5 N29-7 M05-25, M09-242 M09-278 CN1-4, CN2-3, N25-82 N25-184 N41-2 N13-123 M05-181 N13-98 J04-7, N13-160 N25-44 M15-1 N13-91 M09-50, M15-7 M09-128 M09-62 N13-90 J05-4 N45-3 N25-159 J05-16 J04-2, M09-14 N13-139 N13-139 M05-397, M09-98 N13-7 N25-111, N28-3 HP3-4, M05-340, M07-3, M09-338 N13-116 M13-213 N39-5 M09-302 M05-178, M09-200 M05-301 N13-144 N25-147 N06-7 M13-378 522

SMEDLEY, J. SMIT, Z. SMITH, A. W. SMITH, C. F. SMITH, G. C. SMITH, H. SMITH, J. R. SMITH, J. SMITH, L. L. SMITH, M. F. SMITH, N. A. SMITH, P. H. SMITH, S. D. SOARES, A. J. SOBERING, T. J. SOBOTT, B. SOBOTT, B. A. SOBRINHO, M. C. SODINI, N. SOKARAS, D. SOLANO, I. SOLANS, C. SOLANS SANCHEZ, C. SOLDEVILA, U. SOLF, T. SOLLIMA, C. SOLTAU, H. SOMAYAJULA, S. SOMER, E. J. SON, S. SONG, C. G. SONG, K. S. SONG, N. SONG, T. Y. SONG, T. -Y. SONG, X. SOPCZAK, A. SORENSEN, A. G. SORIANO, A. SOSSI, V. SOUCY, J. -P. SOUDANI, K. SOUNG YEE, L. SOUTHEKAL, S. SOUTHEKAL, S. S. SOUVATZOGLOU, M. SOUZA, D. N. SOWARDS-EMMERD, D. SOYAMA, K. SOZZI, M. SPANOUDAKI, V. C. SPARTIOTIS, K. SPECHT, M. SPELLER, R. SPENCER, E. SPIAZZI, G. SPIELER, H. SPIGA, J. SPIJKSTRA, J. J. SPINELLA, F. SPIRITI, E. SPORTELLI, G.

N38-5 CH1-1 N10-2 N29-6 N11-7 N34-5 N45-6 N25-235 N29-5 J01-3, M05-13, M09-347 N25-30 N09-6 M05-247 N25-152 N30-6 N38-3 M05-124, M13-399, N25-173, N25-174 N25-10 M09-398 CH1-4 N13-214 N13-113 N13-189 N13-166 M05-28, M05-244, M13-252, N28-4 N25-43 N35-4 M05-193 M09-248 N20-5 M05-52 M09-266, N25-36, N25-100 M05-298, M09-293 M09-278, M13-48 N25-190 M09-191 N13-193 M04-1 N25-34 M04-5, M05-259, M13-99 M07-6 N14-3 N25-192 M05-169, M15-2 J03-5, M05-247, M05-376, M09-257 M13-33 N25-10 M09-26 N13-138, N25-19, N25-84 N10-1 M10-2, M13-135, M13-333, M14-5 N41-5 N22-6, N25-203 N13-206 N13-48 N25-28 N13-48 N25-8, N39-6 M05-4 M09-332, M13-90 N25-37 M09-332, M13-90 523

SPRENGER, F. SPRUNG, M. SPURRIER, M. SPYROU, G. SQUILLANTE, M. R. SRA, S. SRIVASTAVA, A. K. ST. JAMES, S. STACK, B. C. STAELENS, S. STAELENS, S. G. STAHLBUSH, R. E. STAMPANONI, M. STANFIELD, P. R. STANGE, S. STANITZKI, M. STANKOVA, V. STANOVNIK, A. STAPELS, C. STAPELS, C. J. STAPLES, C. STAPNES, S. STARINSKY, N. STARSJA, S. STAUFFER, A. D. STEARNS, C. STEELE, J. STEFAN, F. STEFAN, M. STEFANO, A. STEIN, J. STEINBACH, C. STEINBRCK, G. STEINBRUECK, G. STELZER, J. STENNER, P. STENTON, K. STEPHEN, J. B. STEPHENSON, R. STEVENSON, R. C. STEWART, J. STEZELBERGER, T. STIERSTORFER, K. STILIARIS, E. STILLER, J. STJOHN, J. STOCKMANNS, T. STOLIN, A. STOLIN, A. V. STOLIN, S. STOLL, S. STOLL, S. P. STONEHILL, L. C. STORAS, P. STORASTA, J. STOUT, D. B. STRAND, S. -E. STRDER, L. STREUN, M. STRIVAY, D. STRMEN, P. STRUEDER, L.

M08-5 N13-98 N25-128 M09-212 N04-6 M03-2 N25-159, N25-239 M09-125, M09-134, M13-72, M13-129 M09-14 M13-252, M13-255 M05-148 N08-5 M08-2 N13-244 N25-115, N29-3 N25-31 J05-10, M09-83 N25-118, N25-136 M13-60, N25-86, N25-137 N04-6, N13-237, N25-131, N36-1 M05-268 N12-1 N25-72 M10-3 N13-157 M13-330 N13-212, N13-231, N29-1 M04-8 M04-8 M09-110 N01-3, N13-220, N13-224, N13-244, N25-104 M05-274 N25-159 N13-175 N13-124 M12-4 N29-2 N13-65 N27-7 N13-130, N13-133 N30-5 N13-13 M09-365, M09-389 M05-208, M09-323, M13-369 N13-13 N13-167 N13-109 J01-3, M05-13, M13-249 J01-7 M09-347 M05-169, M05-247, M05-265, M15-2, N15-1 M05-376, M09-257 N25-82, N26-4 N25-164 N25-159 M05-172, M09-65, M15-7 M05-139, M09-119, M13-138 N35-4 N25-251 CH1-2 N13-7 J02-1, N02-2, N13-64, N24-5, N33-3, N33-5, N36-3 524

STRYDHORST, J. STUDEN, A. STUDENSKI, M. T. STURM, B. W. STURM, B. STUTE, S. STUTZMANN, J. -S. SU, G. SU, I. -T. SU, J. -L. SU, K. -H. SUARDI, A. SUAREZ, R. SUBRAMANIAN, M. SUDHAKAR, M. SUGA, M. SUGIMOTO, K. SUGIMOTO, S. SUGIYAMA, E. SUKHANOV, A. Y. SULLIVAN, N. SUMIYOSHI, T. SUN, C. SUN, F. W. SUN, L. SUN, X. SUNDAL, B. M. SUNDERLAND, J. SURROW, B. SURTI, S. SUSINI, J. SUTTON, A. K. SUYAMA, M. SUYAMA, T. SUZUI, N. SUZUKI, A. SUZUKI, M. SVARER, C. SVOBODA, R. SWAIN, S. K. SWANY, M. SWEANY, M. SWIDERSKI, L. SWIENTEK, K. SWINHOE, M. R. SWINHOE, M. T. SYNTFELD-KAUCH, A. SYNTFELD-KAZUCH, A. SYRJSUO, M. SYRZYCKI, M. J. SZABELSKI, J. SZARKA, I. SZAWOWSKI, M. SZAWLOWSKI, M. SZCZNIAK, T. SZCZESNIAK, T. SZCZYGIE, R. SZCZYGIEL, R. SZLAVECZ, A. SZOSTAK, A. SZUPRYCZYNSKI, P.

M05-142 J05-10, M09-83, M13-120, M13-234 M13-54 N25-229 N01-1 M13-357 N25-38 N25-48 M05-130 M13-276 M13-300 N13-25, N13-101, N13-102, N13-103, N13-112 N25-144 N25-29 N09-3, N09-5, N13-79, N13-80, N33-4 M13-267 CN2-2 N13-61, N41-3 M09-251 N13-186 N29-2 N13-154 N13-71 J03-6 N06-4, N20-3 J01-5, M05-31, M13-87, N07-5 J01-1, M09-383 M09-359 N44-2 M13-66 CH2-2 N13-48 N40-3 N25-110 N25-13 M13-270 J02-5 M09-224, M13-225 N04-1 J05-5 N13-94 N04-1 N01-3, N01-7, N13-220, N25-90, N25-92, N25-104, N25-138 N13-39 N29-5 N25-62 N25-103 N01-7, N25-90, N25-92, N25-104, N25-138 N35-5 N25-171 N13-220 N13-7 N25-103 N01-7 N01-3, N25-103 N01-7, N13-220, N25-90, N25-92, N25-104, N25-138 N13-38 N19-6 M09-146 N25-8 N25-128

525

T TABARELLI DE FATIS, T. TABARY, J. TABATA, M. TABATABEI, G. TACKETT, A. TADJER, M. J. TAGGART, J. TAGHIBAKHSH, F. TAGUCHI, K. TAHERION, S. TAI, Y. -C. TAINTER, K. H. TAJIMA, H. TAKACS, P. TAKADA, A. TAKADA, E. TAKAHASHI, H. TAKAHASHI, K. TAKAHASHI, M. TAKAHASHI, T. TAKAKURA, K. TAKASAWA, M. TAKASHIMA, T. TAKAYANAGI, T. TAKECHI, S. TAKEDA, S. TAKEDA, T. TAKENAKA, N. TAKEUCHI, W. TAKOUKAM TALLA, P. TALBOT, P. S. TALEB, N. TAM, W. -P. S. TAMAGAWA, T. TAMAKI, N. TAN, H. TAN, J. W. TAN, J. -W. TANAKA, H. TANAKA, K. TANAKA, M. TANAKA, S. TANAKA, T. TANAKA, Y. TANDEL, S. K. TANG, C. TANG, F. TANG, J. TANGARO, S. TANIMORI, T. TANIUE, K. TAO, X. TAPFER, A. TAPIAS, G. TARASENKO, O. A. TAROLLI, A. TARTONI, N. TASCHEREAU, R. TASHIRO, E. TASHIRO, M. N43-6 J05-19, M14-1 M13-270 M11-5 N13-94 N08-5 M10-3 M05-280 M08-8, M09-389, M13-45 N14-2 J05-4, M05-127, M05-355, M09-278, M09-362, M13-48 M13-147 N13-61, N37-4 N13-52 N18-7 N25-27 M09-56, M13-30, N13-49, N13-156, N27-2, N28-2, N35-7 N13-154 M05-271, N18-7, N20-1, N25-180 M15-4, N13-61, N25-194, N37-4, N41-3, N41-6 N25-83 M09-251 N13-66 J05-13 N05-7 M15-4 M05-370, M13-105 CN2-2 M09-5 J05-3 M05-319 M09-179 M13-300 N11-5, N13-158 M09-5 N13-27 M05-127, M09-11, N41-5 M05-250 J02-5, N25-114 M05-37, M13-96, N13-154 N40-3 M13-15, N39-2 N13-61, N37-4 N19-5 N36-2 N13-6 M05-7, M09-80, N27-5 M03-8, M13-195, N29-4 N39-1 J05-13, M05-271, M13-360, N18-7, N20-1 N18-7 N07-2 M13-333 M09-98 N29-7 N28-4 N38-2 M05-172, M09-299, M15-7 N13-154 N37-4 526

TAUCHI, K. TAUCHI, T. TAUREG, H. TAVERNIER, S. TAVORA, L. M. N. TAW, F. TAWARA, H. TAYLOR, G. C. TAZAWA, Y. TEIXIDOR, F. TELLMANN, L. TELLO, J. C. TENNEY, C. R. TENZE, L. TENZER, C. TERADA, S. TERADA, Y. TERAMOTO, N. TER-ANTONYAN, R. TERASAWA, K. TERUNUMA, T. TETRAULT, M. -A. THACKER, S. THADA, S. THANASAS, D. THEBAULT, F. THEODORATOS, G. THERS, D. THIBAULT, J. -B. THIELEMANS, K. THOMAS, H. M. T. THOMAS, R. THOMAS, S. THOMPSON, A. C. THOMPSON, C. J. THOMPSON, R. THOMPSON, R. L. THOMPSON, S. J. THON, A. THRALL, C. L. THRAN, A. THUN, R. THUNGSTR&OUMLM, G. TIAN, G. TIAN, X. TIAN, Y. TIBBELIN, S. TICKNER, J. TICONA PERALDA, R. TIEDEMANN, L. TIEMAN, B. J. TILLEMENT, O. TILLOTSON, T. TINDALL, C. S. TINTORI, C. TIPTON, P. L. TIURANIEMI, S. TKACZYK, J. E. TLUSTOS, L. TOBIN, K. W. TOCUT, V. TODD-POKROPEK, A.

N22-2 N13-9 N11-1 M05-67, M05-85, M09-59 N13-152 N29-3 N13-146, N13-149, N13-158, N25-119, N35-3 J02-4 N13-48 N25-7 M04-4, M07-2, M09-287 N13-69 M13-57 N13-92 N33-3 N13-174, N22-2 N13-61, N37-4 M10-1 M09-101 N13-149 J05-13 M15-8 M09-14 HPP-3, HPP-4 M05-208, M13-369 N14-3 N13-34 M13-144, N25-38 M13-201 M02-6 M13-237 N25-5 N27-4 N38-1 M07-6, M09-65 N13-202, N25-60 N25-234 N13-218 M05-244 N13-134 M08-6 N13-183 N25-195 M05-166 N25-242 N13-142 M14-4 N25-189 N13-69 N13-64 N13-98 N25-87, N25-88 N01-6 N24-1 N07-5 N34-5 N03-4, N13-41 N25-177 N19-7 N25-25 N13-54 M05-316 527

TODRI, A. TOGA, A. W. TOGASHI, K. TOH, K. TOH, Y. TOHME, M. S. TOLEDO, F. J. TOMASI, D. TOMASSETTI, L. TOMASZEWSKI, D. TOMEI, S. TOMIOKA, K. TOMIOKA, S. TOMITA, H. TONAMI, H. TONG, S. TONGIANI, C. TOOLE, T. S. TOOLE, T. TOOLIN, M. TOOLIN, M. J. TOPPING, G. TORIGOE, Y. TORNAI, M. P. TORNGA, S. R. TORNGA, S. C. TORRES, I. TORRES, J. TORRES, R. C. TORRIOLI, S. TOSCANO, L. TOURRETTE, T. TOVESSON, F. TOWELL, R. TOWNSEND, D. W. TOYOKAWA, H. TRAVERSI, G. TREBOSSEN, R. TREIS, J. TREMBLAY, S. TREMSIN, A. TREMSIN, A. S. TRENTADUE, R. TRIMPL, M. TRINDADE, A. M. F. TRINDADE, A. TROCINO, D. TROMBA, G. TRON, L. TRONCON, C. TROUW, F. R. TRUMMER, M. TRUNK, U. TSAI, Y. TSAI, Y. -J. TSAO, H. -W. TSENG, H. -R. TSENG, T. -K. TSERRUYA, I. TSOUMPAS, C. TSUBOYAMA, T. TSUCHIYA, B.

N24-6 M04-2 M05-271 N13-138, N25-19, N25-83, N25-84 N07-3, N24-2, N42-5 M07-1 M05-10 M09-257 N13-87, N13-176, N23-1 N25-159 M09-170 M13-270 N25-18 N25-49, N25-50 M05-1, M13-96 M09-17 N25-243 M09-314, M13-315, M13-318 M10-3 N06-1 N13-229 M13-99 N35-3 M13-297 N06-1, N13-229 N25-30 M11-6 N13-113 N13-90 M05-331 N07-4, N13-116 N13-53 N29-3 N13-162 M09-89 N25-109 N03-5, N25-198 M05-211, M09-122 N19-4, N24-5, N35-4 M13-375 N29-2 CN1-4, CN2-3, N25-82 N13-178 N27-1 N13-65, N13-151, N13-152 M05-307 N13-88 M09-398 J05-12 N12-3 N42-1 M05-313 N25-239 M13-384 M09-227 N13-51 M13-372 N25-127 N15-1 M05-256, M09-41 N13-9, N22-2 N25-84 528

TSUDA, T. TSUI, B. TSUI, B. M. W. TSUJI, H. TSUKAHARA, Y. TSUKAMOTO, T. TSUNG, J. -W. TSUNO, S. TSUTSUI, H. TSUTSUMI, K. TULL, C. R. TUNA, U. TUNG, C. H. TUNG, C. -H. TUOMINEN, E. TUPPUTI, S. A. TURCHETTA, R. TURCO, A. TURCOTTE, E. TURKINGTON, T. G. TURNER, W. TURQUETI, M. A. TYLSKI, P. TYMCHUK, I. TYNDEL, M. TZANAKOS, G. U UCHIDA, T. UCHIHORI, Y. UDIAS, J. M. UEDA, M. UENO, K. UENO, Y. UGOROWSKI, P. B. UHER, J. UHLMANN, N. UJHELYI, F. ULLAN, M. ULLOM, J. N. UMSTADTER, D. UNDERWOOD, D. UNNO, Y. UNO, S. UR, C. URBAITIS, A. V. URIBE, J. USUI, T. USUKI, Y. UTES, M. J. UZHINSKIY, V. V VACCAROSSA, L. VACIK, J. VAFAI, N. VAINIO, R. VAITKUS, J.

M05-1 M05-331 J01-1, M05-40, M05-205, M09-383, M09-389, M11-2, M13-312 N25-50 J05-13 N25-13 N02-7 N10-7 J02-5 N25-98, N28-2 N25-180 M09-218 M09-245 M11-6 N13-7 N05-4 N25-31 M05-391 M13-375 M07-5, M13-174 N13-13 N24-6 M13-357 J05-10 N25-31 M09-212

M09-29, N44-6 N13-149 HP2-3, M05-226 M05-271 J05-13, N18-7, N20-1 M09-5 N25-170 N20-2, N25-21, N25-189 N25-168 M05-274 N13-48, N25-35 N25-42, N25-58 N04-4, N30-3 N13-197 N13-174, N22-2 M09-29, N44-6 N13-43, N25-247 N13-227 M12-6 N25-80 N25-98, N25-121, N28-2 N13-187 N13-89

N13-86 N25-75 M13-99 N35-5 N25-159 529

VALASTYAN, I. VALE, L. R. VALENTE, P. VALERO, A. VALERY, D. VALETTE, S. VALICENTI, R. A. VALIN, I. VALLERGA, J. V. VALLS, J. VALTONEN, E. VAN DAM, H. T. VAN DEN BERG, L. VAN DER KOLK, E. VAN DER KOUWE, A. J. VAN DER LINDEN, A. VAN HOLEN, R. VAN LEEUWEN, D. A. VAN LOEF, E. VAN LOEF, E. V. VAN LOEF, E. V. D. VAN MULLEKOM, P. VAN NIEUWENHUYSE, B. VAN ROOIJ, L. F. VAN SILFHOUT, R. VAN STENIS, M. VAN STEVENDAAL, U. VAN VELDEN, F. H. P. VANDEMEULEBROUCKE, J. VANDENBERGHE, S. VANDENBROUCKE, A. VANDENBROUCKE, M. VANDERSTEEN, G. VANDONE, V. VANIER, P. E. VAQUERO, J. J. VARELA, E. VARELA, J. VARNER, G. VARNER, G. S. VAROLI, V. VASCOE, C. VASILIEV, Y. VASKA, P. VASOURA, E. VASYUKOV, S. VA'VRA, J. VAVRIK, D. VAX, E. VAZQUEZ-FLORES, G. J. VECCHI, S. VECCHIO, S. VEENHOF, R. VEERAMANI, P. VEGA-ACEVEDO, N. VELAN, S. VELARDE, A. VELAZQUEZ, M. VELOSO, J. F. C. A. VELOSO, J. F. VELOSO, J. J. F. C. A. VENKATARAMAN, R.

J05-12, M09-152 N25-42 N15-6, N33-6 N13-113 N32-2 M14-1 N13-1, N13-2, N13-3 N22-6, N25-203 CN1-4, CN2-3, N25-82 N13-113 N35-5 J01-4, J03-1, J05-9, M06-2 N41-7 M13-264 M04-1 M13-159 M13-252 N25-12 N01-1 N25-81 N01-5 N26-3 M05-148 M05-4 N13-129 N11-1 M09-203 M05-4 M05-235 M05-28, M05-148, M09-41, M13-252, M13-255 M13-135 N15-2, N44-3 M05-85 N25-71, N36-5 N13-233 HP2-3, M05-226, M05-283, M05-397, M09-98, M13-285 N13-69 M05-307, M10-4 N13-201, N45-7 N27-5 N25-4 M13-147 N25-138 M05-169, M05-247, M05-265, M05-376, M09-257, M15-2 M05-301 N25-91 N13-201, N45-7 N25-17 N13-226, N13-230 N06-4 N13-96 N25-102 N11-4 N31-5 N13-97 M13-249 N13-69 N25-78 M05-403 J04-7, N13-160, N13-161, N25-152 N11-4 N25-56, N25-112 530

VERBEKE, J. M. VERDINI, P. G. VERDU, G. VERDUCCI, M. VERESS, A. I. VERGER, L. VERHAEGHE, J. VERHEYDEN, R. VERNET, B. VERNON, E. VERONA RINATI, G. VETTER, K. VICINI, A. VIDAL, F. P. VIDAL-MIGALLON, I. VIERGEVER, M. A. VIGIL, R. D. VIGNEAULT, M. VIJA, A. H. VILA, I. VILELA, E. VILIM, R. B. VILLA, M. VILLANI, E. G. VILLASENOR, L. VILLENA, J. L. VILUCCHI, E. VINAS, C. VINKE, R. VINOGRADOV, S. VINOGRADOVA, T. VIROLAINEN, T. VIRTO, A. L. VISCOGLIOSI, N. VISVIKIS, D. VITELLI, L. VITULO, P. VIVIANI, C. VO, D. T. VOGEL, J. VOLEGOV, P. L. VOLOKH, L. VOLPI, G. VON WILPERT, J. VONCK, K. VONKEN, E. -J. VORONKIN, E. F. VOSS, B. VOSS, L. F. VOUTSINAS, G. VU, C. VU, C. Q. VU, N. VU, N. T. VUNCKX, K. VYKYDAL, Z. W WADA, Y. WAGENAAR, D.

N13-248 N25-243 N25-77 N13-111 M05-40 M05-76, N25-211 M07-6, M07-7, M13-21 N25-118, N25-136 N14-3 N13-47, N19-3, N25-218 N13-18 N13-135, N13-137, N13-217, N14-2, N18-2, N18-5, N24-1, N25-225, N25-226 M05-277 M09-311 M05-397, M09-98 M08-3 N41-7 N13-199, N25-142 M13-291 N25-159 N25-44 N14-7 N13-172, N16-3 N24-4 N13-69 M09-98 N13-86 N25-7 J01-4, J03-1, M06-2 N25-111, N28-3 N25-111, N28-3 N25-200 N25-159 J05-11, N13-30 M05-46, M09-170, M09-344, M13-186, M13-402 M05-331, M09-92 M05-277 M05-277 N25-42, N25-58 N36-3 N13-227 M09-350 HPP-10, N21-3 N08-4, N25-158 M05-148 M08-3 N29-6 N13-195 N02-6, N25-229 N22-6, N25-203 M10-5 N22-7 M13-372 M09-50, M15-7 M05-214, M09-140, M13-153 M05-241, N19-7

M13-105 M11-2, M13-312 531

WAGENAAR, D. J. J01-1, M09-383, M13-147 WAGENKNECHT, G. M04-4, M09-287, M13-159 N09-4 WAGNER, J. C. N13-134, N18-3, N18-4 WAHL, C. G. WAHL, D. N13-69 N06-1, N13-229 WAKEFORD, D. WALBY, M. N13-27 WALENTA, A. H. M04-8 M05-319 WALKER, M. D. WALKER, S. M10-3 WALLACE, M. S. N06-1, N13-229, N26-4 N04-4 WALSH, W. J. WALSTROM, P. L. N13-239 N13-1 WALTHERS-ELLIS, G. G. WAN, L. M13-228 WANG, C. M05-79, M05-106, M13-69, M13-108, M13-339 N13-181 WANG, C. -Y. N13-181 WANG, C. -H. WANG, E. N38-5 WANG, G. M13-132, N07-6 M09-191, N13-140, N25-117 WANG, J. WANG, S. M05-88, M05-136 M13-300 WANG, S. -J. WANG, T. F. N02-6 WANG, W. N18-3 WANG, X. M05-187, M09-383, N07-7, N25-95, N25-245 HPP-6, HP2-2, J05-20, J05-21, M05-34, N13-140, N13-216, N25-95 WANG, Y. WANG, Z. J04-3, M13-222, N29-4 WARBURTON, W. K. N13-27, N13-119 N04-2, N25-74 WARREN, G. A. WASSATSCH, A. N13-35, N19-4 WATABE, H. M09-305, M10-1, M13-78 M13-30, M13-267 WATANABE, M. WATANABE, S. M15-4, N13-61, N37-4, N41-3 M09-281 WATANABE, Y. WATERS, W. M12-6 M13-327 WATSON, C. C. WATSON, S. N04-3 N13-202, N25-182 WATTS, S. WAWRZYNIAK, G. M08-8 WEAVER, M. M09-236 N13-183, N25-33 WEAVERDYCK, C. N41-4 WEBB, A. WEBB-ROBERTSON, B. -J. M. N25-217 N34-3 WEBER, M. M. WEBER, M. N23-5 N25-41 WEBER, M. J. WEDROWSKI, M. M05-67, M05-85 N06-1, N20-7 WEHE, D. WEHE, D. K. N13-222, N13-229, N25-79, N26-6 WEHRL, H. F. M05-259, M11-5 M09-260 WEI, Q. WEI, W. N19-1 WEI, X. N22-5, N25-167 N09-3, N13-79, N13-80, N33-2, N33-3, N33-4, N33-5 WEIDENSPOINTNER, G. WEILHAMMER, P. J05-10, M09-83, M13-120 WEIRICH, C. M07-2 WEISENBERGER, A. M09-230, M13-249 WEISENBERGER, A. G. J01-3, J01-7, M05-13, M09-347 WEISSHAAR, D. N25-196 WEISSLER, B. M05-28 M15-3 WEIZENECKER, J. 532

WEIZEORICK, J. WELLS, B. T. WELLS, K. WELLS, R. G. WENGROWICZ, U. WERMES, N. WERNICK, M. N. WESSEL, J. C. WESTBROOK, E. WHEELER, R. WHITE, A. P. WHITE, S. WIANT, D. B. WICEK, F. WIELAND, O. WIENER, J. WIIK, L. A. M. WILDER, M. WILDON, D. W. WILKENS, J. WILLIAMS, C. S. WILLIAMS, D. A. WILLIAMS, G. B. WILLIAMS, M. C. S. C. WILLIAMS, R. WILLIAMS, R. M. WILLIAMS, Y. Z. WILLIAMSON, T. WILMS, A. WILSON, A. WILSON, J. WILSON, J. M. WILSON, K. WILSON, M. WILSON, N. WINANT, C. D. WING, M. WINTER, M. WINTER, M. J. WINTZ, P. WIRTH, A. WIRTH, S. WIRTZ, D. WISNIEWSKA, M. WISNIEWSKI, D. WITKOWSKA-BARAN, M. WOLLENWEBER, S. WOLNIEWICZ, K. WOLOSCHEK, S. WOLOSHUN, K. A. WOLSKI, D. WONG, A. WONG, K. -P. WONG, W. -H. WOO, S. K. WOOD, J. WOODRING, M. L. WOODY, C. WOODY, C. L. WOOLF, R. S. WORSTELL, B. WORSTELL, W. A.

N13-98, N25-237 N13-133 M04-6, M05-346, M13-36, M13-168, M13-396 M05-142, M09-20, M13-162 N13-230 N02-7, N12-5, N13-29, N13-179, N13-196, N34-4 M05-286, M14-2 M08-8 N13-202, N25-182, N38-1 N06-7 N45-6 J03-2 M13-204 N13-54 N25-71, N36-5 M09-26 N02-4 N13-48 M05-196 M13-33 N25-147 M15-7 HP3-3, M09-167 N24-7 N25-60 N25-144 N25-177 N13-199 N25-94 N13-183 N25-241 M07-5, M13-174 M11-1 N25-31 N25-60 M09-338 N42-7 N22-5, N22-6, N25-203 N25-233 N15-3 HPP-1 M09-365 M05-28 N25-135 N25-135 N25-221 M13-330 N38-4 M12-1 N13-1, N13-2, N13-3 N13-220, N25-104 M05-373 M13-378 M05-79, M05-106, M13-69, M13-108, M13-339 M05-97, M05-115, N25-14 N25-22 N06-6, N13-249 M05-169, M09-257, N15-1 M05-247, M05-265, M05-376, M15-2 N25-22 M10-3 M09-314, M13-315, M13-318 533

WORTH, C. WRIGHT, D. WRIGHT, D. M. WRIGHT, J. WRIGHT, V. WROBLESKI, D. A. WROE, A. WROE, A. J. WROSCH, M. WU, B. WU, H. WU, J. WU, M. WU, Q. WU, S. WU, W. WU, X. WU, Y. WU, Y. K. WU, Z. WUBBELING, F. WUERSLIN, C. WUESTNER, P. WULF, E. A. WULF, F. WUNDERER, C. B. WURTZ, J. WYCKHUYS, T. X XIA, D. XIA, H. XIA, L. XIA, T. XIA, Y. XIE, Q. XIE, Z. XING, Y. XU, F. XU, J. XU, T. XU, Z. XUE, H. XUE, T. Y YAHIL, A. YAJIMA, S. YAMADA, M. YAMADA, T. YAMADA, Y. YAMAGISHI, H. YAMAGUCHI, M. YAMAGUCHI, Y. L. YAMAKAWA, Y. YAMAMOTO, E. YAMAMOTO, H. YAMAMOTO, S.

M09-314, M10-3, M13-315, M13-318 N13-89, N18-6 N13-248 N08-2, N25-181 N38-2 N25-30 N09-3, N39-7 J02-7 N25-34 N13-209 J05-4, M05-127, M05-355, M09-278, M09-362, M13-48 J05-22, M13-396, N13-16 N06-7 N38-5 N13-167 M13-243 M12-1 M05-61, M05-328, M09-134 N13-71 M05-88, M05-136 M13-294 M02-3 N13-109 N06-2, N13-213 N30-2 N13-57, N13-70, N13-217, N14-2, N33-2, N37-3 N25-101 M05-148

M05-22, M09-386 N25-95 N13-197 M09-47 M05-88, M05-121 M13-228, N07-7, N25-151 M09-284 HPP-6, M05-394, M13-189, N13-211, N13-216, N13-240 M09-23 J01-1, M09-389, M11-2, M13-312 N25-151 N25-95 N13-223 N25-61

M13-291 J02-5 N25-176 M13-105 M05-1, M09-188, N13-154 N13-138 M15-4 N11-5 M05-1 J05-13 N25-80 M05-262, M09-251, M13-63, M13-273 534

YAMAOKA, K. YAMASHITA, T. YAMAYA, T. YAN, J. YAN, L. YAN, S. YANAGIDA, T. YANG, B. YANG, G. YANG, H. YANG, K. YANG, P. YANG, X. YANG, Y. YANKAM NJIWA, J. YAO, R. YAO, Y. YAREMA, R. YASUDA, R. YASUOKA, K. YAVER, H. YAZAKI, Y. YE, H. YE, J. YE, M. YE, Z. YEE, R. M. YI, F. YIN, Q. YIN, Y. YOKOTA, Y. YONG, D. YONG, G. J. YONGGANG, W. YOO, H. -S. YOO, M. G. B. G. B. N. L. YOON, H. S. YOON, K. YOON, P. S. YOON, W. -K. YORK, R. YOSHIDA, E. YOSHIKAWA, A. YOSHINO, M. YOU, J. YOUNGMAN, C. YOUSEFZADEH, H. C. YU, A. R. YU, B. YU, H. YU, J. YU, Z. YUASA, T. YUN, M. S. YUN, M. YUN, M. -S. Z ZACHARIADOU, K.

N37-4 M13-30, M13-267, N39-4 J04-5, M05-82, M09-32, M13-102, M13-105, M13-267 M03-3 N13-6 M13-132 N25-96, N25-97, N25-98, N25-110, N25-113, N25-114, N25-121, N28-2, N35-7 M02-3 N25-209, N25-221, N31-4 M13-384, N13-222, N25-56, N25-112 M13-3, N01-2 N25-236 N25-106 M05-61, M05-286, M05-328, M09-134, M13-60, M14-2, N13-100, N13-209, N13-210, N25-15, N25-236 M07-8 M05-112, M05-121 N06-7, N13-77, N13-247, N17-4 N03-7, N27-1 N25-19 J05-13 N13-13 J04-5, M09-32, M13-267 HPP-2 M09-26 N13-100 N08-3, N22-3 N22-7 N13-143 N07-7 J05-4, M05-127, M05-355 N25-96, N25-97, N25-110, N25-113, N25-114, N25-121, N35-7 M09-293 N30-4 N13-99 N13-219 N25-162 M09-128 J05-23 N13-12 N13-219 N14-6 J04-5, M05-82, M09-32, M13-102, M13-105, M13-267 N25-96, N25-97, N25-98, N25-110, N25-113, N25-114, N25-121, N28-2, N35-7 M13-15 M05-223 N25-239 N13-108 M05-97 N11-7 M13-132, M13-171 M11-2, M13-312, N45-6 M13-201, M13-372 M05-370 M09-233 M05-400 J05-18

N13-221 535

ZAIDER, M. ZAIDI, H. ZALIPSKA, J. ZALUZEC, N. J. ZAMALLOA, M. A. ZAMYATIN, A. A. ZANDERIGO, F. ZANETTI, G. ZANNOLI, S. ZAPPI, R. ZASINAS, E. ZAVATTINI, G. ZAVRTANIK, M. ZEINTL, J. ZEKAUSKAS, M. ZENG, G. L. ZENG, K. ZENG, M. ZENG, S. ZENG, Z. ZENIYA, T. ZENNARO, M. ZERVAKIS, E. G. ZETTA, L. ZHANG, B. ZHANG, F. ZHANG, G. ZHANG, H. F. ZHANG, H. ZHANG, J. ZHANG, L. ZHANG, N. ZHANG, Q. ZHANG, W. ZHANG, X. ZHANG, Y. ZHANG, Z. ZHAO, C. ZHAO, Q. ZHAO, T. ZHAO, Y. ZHAO, Z. ZHIVKOVA, V. N. ZHONG, Z. ZHOU, B. ZHOU, J. ZHOU, L. ZHOU, V. W. ZHOU, Y. ZHU, J. ZHU, K. ZHU, R. -Y. ZHU, X. ZHU, Y. F. ZHU, Y. ZHURAVLEVA, M. ZIAO, Z. ZIEBELL, A. L. ZIEGLER, S. ZIEGLER, S. I. ZIEGLER, S. L. ZIEMONS, K.

M09-236 M05-184, M05-361, M09-308 N15-5 N25-180 N13-69 HPP-8, M05-322 M05-373 N13-72 N07-4, N13-116 N13-86 N25-159 M13-12 M13-234, N25-206 M13-291 N13-94 M05-223, M09-194 M13-201 N13-10, N25-61, N25-242 M13-228, N07-7 N25-48 M10-1, M13-78 N13-92 N13-34 N25-65, N25-66 N25-151 N13-47, N25-214, N25-223, N31-3 N13-129 M09-284 M09-260, M13-150, M15-7 M13-147, M13-306 J02-3, J04-3, J05-20, M09-221, M13-222, N13-216, N13-240, N32-4, N32-5, N40-2 M13-9, N28-1 N13-197 HP2-2, N25-176 M09-260, M12-6, N25-151, N44-3 M05-79, M05-106, M13-69, M13-108, M13-228, M13-339, N13-140, N25-95, N25-236 N25-31, N25-151 M13-150 J05-17 N13-143, N25-95 N13-28 J05-20 N25-53 M05-370 N13-183, N25-33 M09-125, M13-129 M13-153 M05-16, M09-107 M03-8 N25-151 N25-23 N32-4, N32-5, N40-2, N43-1 M05-229, N13-140 N25-223 N25-165 J04-6, N25-128 N25-25 J02-7 M09-326, M13-33 M09-197, M09-263, M13-333 N25-217 M05-67, N25-251 536

ZIMBAL, A. ZIMMER, M. ZIMMERMAN, T. ZIO, E. ZIOCK, K. P. ZIOCK, K. -P. ZOBOLI, A. ZOCCA, F. ZOCCOLI, A. ZOGLAUER, A. ZONTAR, D. ZORN, C. ZORZI, N. ZOTEV, V. S. ZOU, Y. ZUFFA, M. ZURAWSKI, J. ZUVIC, M. ZWERGER, A.

N13-170 N25-239 N03-7, N27-1 N13-128 N06-3 N26-5 N02-4 N13-32, N13-37, N13-43, N25-196, N25-247, N27-3 N25-68 N09-3, N13-57, N13-70, N13-79, N13-80, N13-217, N14-2, N18-1, N33-2, N33-3, N33-4, N33-5, N37-3 J05-10, M09-83 J01-3, N40-7 M05-28, M05-244, N02-4, N28-4 N13-227 M12-2, M12-6, N25-23 M05-391 N13-94 N25-215 N25-191, N41-2

537