A Multichannel Dosimeter Based on Scintillating Fibers for

Medical Applications
K.-H. Becks1, J. Drees1 , K. Goldmann1, I.M. Gregor1 , M. Heintz2, Arndt Roeser*
1 Fachbereich Physik, Bergische Universität-Gesamthochschule Wuppertal

D-42097 Wuppertal, Germany

2Klinische Strahlenphysik, Universitätsklinikum Essen
D-45122 Essen, Germany
*Klinikum Wuppertal GmbH, Klinik f. Strahlentherapie u. Radio-Onkologie
D-42283 Wuppertal, Germany

ABSTRACT

Dosimetry with plastic scintillators is an interesting alternative for the measurement of

the absorbed dose. A scintillator does barely disturb the radiation field due to its mass
absorption coefficients which is water-equivalent in a wide range of energies.
Furthermore, plastic scintillation dosimeters provide a fast and direct reading of the
measured value combined with a high spatial resolution. In the set-up described so far,
the light produced by the scintillators was transported via light-guides to
single-channel or multi-channel photomultipliers to be transformed into an electric
current read out by pico-amperemeters. The use of photomultipliers becomes
expensive and complicated for a dosimeter system with many parallel channels. For
such applications, an image intensifier coupled to a CCD is a simpler approach which
can read out some 80 fibers in parallel [1] [2]. In this note, the design and the first
successful measurements of a prototype set-up are described.

Keywords: dosimetry, multichannel, scintillating fibers, brachy therapy, teletherapy

1 Introduction
There exist two radiation methods for cancer treatment, brachytherapy and tele–
therapy [2]. The common method is the teletherapy, a treatment with radiation from the
outside (e.g. 60 Co–source, linear accelerator). Detailed mapping of the used radiation
fields in 2 or 3 dimensions is desirable for the dosimetry. These measurements are
usually done with a single ionisation chamber. In this case it is only possible to measure
one field position at a time. For a detailed field mapping too much time is needed. With
the developed multichannel dosimeter direct and simultaneous reading of the absorbed
doses in several field points becomes possible.
 Another radiation treatment is the brachytherapy. For example, in the eye plaque
treatment an applicator with a radioactive source is applied directly to the eye (see fig-
ure 1). By changing the form of the applicator the field of the emitted radiation can be
influenced. Therefore high field gradients have to be measured without disturbing the
physical conditions. A high spatial resolution of the detector is as well necessary.
Plastic scintillators can be used as dosimeters because they emit visible light propor-
tional to the absorbed electron and gamma dose rate [3][5]. A scintillator does not dis-
turb the radiation field due to its mass absorption coefficient and its mass stopping power
which are water-equivalent in a wide range of energies. Furthermore, plastic scintilla-
tion dosimeters provide a fast and direct reading of the measuring values combined with
a high spatial resolution. Depending on the spatial resolution needed, the scintillators
can be as small as 1 mm3 or less. The light produced by the scintillators is transported
through clear optical plastic fibers. In a set-up discussed by Flühs et al. [3] and by Bed-
dar et al. [5], these fibers are read out by single-channel or multi-channel photomultipli-
ers. For applications where a small number of channels is required, they are the ideal
light-to-current converters. For set-ups with more than 20-30 channels, however, the
electronic expenditure to measure the currents is large. The parallel multi-amperemeter
systems required for that purpose are commercially not available. An image intensifier
coupled to a CCD is an alternative approach. Such a system is easier to handle. Only
some supply electronics and a frame grabber card for the data acquisition by a PC are
needed. In the following, we describe a system which allows to read out up to 80 detec-
tors in parallel. Using such a multi-channel dosimeter, all necessary basic functionality
tests were performed as well as some studies in a clinical environment.

2 The Set-Up of the Multichannel-Dosimeter

The described dosimeter will consist altogether of 80 channels, 40 of these channels with
a scintillator (1 mm3) 1 and an optical fiber 2 . The remaining 40 channels are without
scintillator. They will be used for the suppression of the Čerenkov light background pro-
duced in the light guiding fibers. The read-out system (see figure 2) is set up in such a
way that the incoming light guides are kept at fixed positions by a mask to ensure a good
optical contact between fibers and the image intensifier. The image intensifier transforms
the light from all 80 fibers into electrons, amplifies them in parallel, transforms them
back into photons and re-emits an amplified image of the incoming light signals. The
image intensifier keeps both, spatial resolution, i.e. the ability to distinguish between
the individual fibers, and the relative intensity of each fiber’s light, i.e. the amplification
is linear [1] [2].
The image emitted by the image intensifier is transported via an optical fiber taper
to a CCD. This fiber taper is a conical optical light guide consisting of a large number of
thin glass fibers. The diameter of these fibers all decrease towards one end of the taper,
so an image transmitted by the taper will be reduced in size, but its relative spatial reso-
lution will be maintained. The CCD integrates the light signal of each of the 80 fibers in
1 BC–400 from Bicron Technology
2 CWKF–1001–22E from Cunz Toray

2
parallel over 20 millisecond. The image is transformed into standard TV format and read
out by a personal computer equipped with a frame grabber 3 , an extension card for a PC
converting a video signal into digital information. To increase the precision, the digital
signal can be summed up in the PC over several integration periods of the CCD - typi-
cally between a few hundred and a few thousand – corresponding to total measurement
times of one to five minutes.
The dynamic range of the multi-channel dosimeter is determined by two factors. For
an individual reading, it is given by the maximum signal the CCD can handle without
saturation, and the minimum signal that can be distinguished from the background noise.
These limitations give a dynamic range of 5103. Furthermore, from one measurement
to another, the amplification of the multi-channel dosimeter’s image intensifier can be
adjusted over a range of 5104. The smallest detectable signals are in the order of a
few hundred photons per second emerging from a fiber of 1 mm diameter. The amount
of radiation equivalent to this signal depends on the size of the scintillator coupled to
the fiber, the transmission characteristics of the fiber, and the background noise.

3 Measurements
To test the system, depth dose measurements performed with a ionisation chamber were
compared to measurements performed with a 10-channel prototype of the described
multichannel dosimeter. The used source of radiation was a Siemens linear accelerator at
the Clinic for Radiation Therapy, Radio–Oncology and Nuclear Medicine in Wuppertal.
The acceleration bias was adjustable from 6 MV to 18 MV. Measurements at different
acceleration biases were performed.
The depth dose distribution is the dose distribution in a body or in a phantom along
the central beam. Mostly relative or proportional depth doses are indicated, which refer
to the absorbed dose maximum or an indicated reference depth. The so called build up
effect describes an increase of the absorbed dose, the passing through a maximum and
the following dose decline. For photons this build up effect is due to secondary electrons,
which are generated by photoelectric effect, Compton-effect, or pair production depend-
ing upon the photon energy. Starting from a photon energy of 0,6 MeV the Compton-
effect outweighs in water. The secondary electrons move in the same direction as the
primary beam and increase the absorbed dose up to a depth, which corresponds to the
range of the secondary electrons. The dose decline is caused by the attenuation of the
photon beams. From a certain depth the curve follows approximately an exponential
function. The slope of the dose decline depends on the quality of the radiation, the field
size and the divergence of the jet bundle [4]. This progression is shown in figure 3 and
figure 4.
In the described measurements a field size of 5 cm  5 cm was used. In a plate made
of rigid water (RW3 4 ) two out of x10 fibers were placed in a drilling for this purpose.
The fibers were positioned in such way, that the tips with the scintillating material were
3 National Instruments PCI–1407
4 a material, which has the same absorption coefficient as water

3
situated in the iso center 5 . One of the three fibers was a reference fiber without scintilla-
tor. Therefore it was possible to measure the background and the signal simultaneously.
The background has been subtracted from the measured values afterwards. In order to
suppress the Čerenkov light background. Further RW3 plates were subsequently placed
onto the previous and measurements were repeated until a depth of 20 cm was achieved.
As comparison, the measurements were repeated under same conditions with an ioniza-
tion chamber, calibrated for the same energy region. The determined values are shown
in figure 3. Only two out of ten fibres are shown exemplarily. The build up effect is more
strongly pronounced with the fibers than with the ionization chamber, which is to due
to the characteristics of the fibers. Figure 3 depicts the depth dose distribution measured
at 6MV with an ionisation chamber and the prototype multichannel dosimeter. In larger
depths than the maximum of the build up effect, the gradient of the fiber data corresponds
to the reference data. In smaller depths the curves of the fibers are clearly steeper than
those of the reference data. Due to their physical characteristics the scintillators do not
influence the physical environment. However the photons are more influenced by the gas
volume of the ionization chamber, the gas and the material of the chamber cause a mod-
ification of the environment. The depth dose distribution at an accelerating voltage of
18 MV is shown in figure 4. Here a steeper rise of the measured values for smaller depths
is being detected for the same reasons as described above. However the gradient of the
fiber values does not correspond to those of the ionization chamber. The depth dose dis-
tribution of the fibers are substantially steeper, than those of the ionization chamber. The
assumption is that this is due to the coupling of the optical fibers to the opto-electronic
read-out. The losses may be explained by a different quality of the optical coupling be-
tween fiber ends and the image intensifier. Because of the different manufacturing qual-
ity of the fiber ends, the losses during the transition diverse slightly and therefore dif-
ferent Čerenkov light background is registered in the reference fiber as actually occur
in the other fibers. In higher energy regions this effect becomes more apparent. There-
fore one should take more care when preparing the fibers for the coupling to the image
intesifier. A reference fiber for each scintillating fiber would improve the background
reduction. Measuring the Čerenkov light background of each fiber before attaching the
scintillator would indicate the fluctuations within the manufacturing of the fiber ends. It
is possible to reduce the Čerenkov light background further more by using longer light
guides. The attenuation of the guides is wavelength depending. The Čerenkov light will
be more attenuated in some wavelenth ranges than the information signal. All this will
be done for a further prototype before taking it into clinical environment.

4 Conclusion and Future Work

In conclusion, we have developed a new parallel read-out system for dosimetry with
plastic scintillators, plastic light guides, an image intensifier coupled to a CCD and read
out by a PC. This system combines the advantages of the plastic scintillator dosime-
try, such as water-equivalence, direct reading and high spatial resolution, with a paral-
lel, computer-controlled read-out of the measured signals. Further advantages over com-
5 intersection of the central beam with the rotation axis of the irradiation device

4
monly used systems are its intrinsic simplicity, the high number of channels that can be
read out in parallel with negligible cross-talk and in its low cost per channel. Although
no long term test of the entire system has been performed under clinical conditions. Yet,
the first experiences are very encouraging. Depth dose measurements with a prototye of
this dosimeter were performed at different particle energies and compared to measur-
ments with an ionisation chamber. For lower energies the results agree with each other.
The uncertainty due to Čerenkov light can be reduced by using reference fibers. At the
prototype dosimeter problems with the coupling of the light guides to the read out sys-
tem appeared. For a further prototype this coupling has to be optimised by cutting the
fibers more carful and polishing the ends finer. Also it is possible to reduce the Čerenkov
light by using about 10 m long light guides. Due to the wavelength depending attenua-
tion of the light guides the Čerenkov light will be more attenuated in some wavelength
ranges than the information signal. When applying the dosimeter in the teletherapy it
is also necessary to operate the read out electronic in a sufficient large distance to the
radiation source. This can automatically be ensured by the proposed long light guides.
Further studies will include long time measurements in the teletherapy with a new pro-
totype, an individual calibration of each channel, and user friendly software tools for the
operating of this system.

References

[1] M. Garg, Konzeption, Aufbau und Test eines Vielkanal-Dosimeters mit szintillierenden
Fasern, WUB–DIS 96–17, Universität Wuppertal, 1996

[2] I.M. Gregor, Ein Vielkanaldosimeter mit szintillierenden Fasern – Klinische Messungen bei
Energien oberhalb 1 MeV, WU D 98 – 14 , Universität Wuppertal, 1998

[3] D. Flühs, M. Heintz, F. Indenkämpen, C. Wieczorek, H. Kolanoski and U. Quast, Direct

reading measurement of absorbed dose with plastic scintillators - The general concept and
applications to ophthalmic plaque dosimetry, Med. Phys. 23 (1996), pages 427-434

[4] H. Reich, Dosimetrie ionisierender Strahlung, B.G. Teubner, Stuttgart, 1990

[5] A.S. Beddar, T.R. Mackie, and F.H. Attix, Water-equivalent plastic scintillation detectors
for high-energy beam dosimetry: 1. Physical characteristics and theoretical considerations,
Phys. Med. Biol. 37 (1992), pages 1883-1900