http://mipav.cit.nih.

gov

1
 Medical Image Processing,
Analysis & Visualization
in Clinical Research
Justin Senseney
[email protected]
dcb.cit.nih.gov/~senseneyj

Biomedical Image Processing Research Services Section

Center for Information Technology
mipav.cit.nih.gov

2
 MIPAV Team
Employees
Ruida Cheng
William Gandler
Matthew McAuliffe
Evan McCreedy
Justin Senseney

Fellows
Sara Shen (Maryland)

Contractors
Alexandra Bokinsky, Geometric Tools Inc. (Visualization)
Olga Vovk, SRA International Inc. (Technical Writing)

Alumni
Paul Hemler (Hampden-Sydney), Agatha Monzon, Nishith Pandya (FITBIR),
3
Beth Tyriee (Kentucky), Hailong Wang (Heidelberg)
Requirements for an Image Quantification and
Visualization Application
• Portability
– cross-platform or platform-independent execution
• Data format independence
– access to images: DICOM, Analyze, TIFF, Raw, …
• Extensibility
– plugins and/or scripts
• Scalability
– foundation to support the growth to larger and more intricate data structures
• Usability
– coherent graphical user interface (GUI)

4
 Portability
Java Primer PC

Java Interpreter
Source Code Byte Code Machine
for the PC
(class files) code

UNIX
Java applications can be "written once and
run anywhere", significantly reducing
cross-platform development and
maintenance.
Java Interpreter Machine
for a code
Unix Workstation

5
 Data Independence

• DICOM file reader/writer

• DICOM Query/Retrieve and “Catcher”
• Comprehensive file format support/conversion
– http://mipav.cit.nih.gov/fileformat.html

• MIPAV XML file format

6
 Extensibility
Plugins and Scripts
• Plugins
– Functions written in Java using the MIPAV API.
• Scripts
– Use MIPAV to record and save function(s) applied to
image dataset(s).
– Apply the script to any number of image datasets
using the script wizard.

7
 Scalability
• Model Image is an n-dimensional structure.

• Algorithms can support up to 4D datasets.

• Viewers support 4D dataset with fusion.

8
 Usability
• GUI elements

• Scripting system

• Command-line tools

9
10
 Functional Overview
GUI
Views – with data fusion
2D planar, VOIs
“Lightbox”, Algorithms
Cine (movie), 32K Filtering
Multi-planar, Manual and Segmentation/classification
3D tri-planar, automated S
Measurement/quantification
Surface render, (supports 3D texture contouring Registration/fusion c
mapped volume rendering Utilities r
Volume render Plugins i
p
Data (Image) types: n-dimensional structure t
(boolean, byte, unsigned byte, short, i
unsigned short, int, long, float, double, Complex, ARGB) n
g
File types
PACS (Raw, Analyze, DICOM 3.0, GE, Siemens, Bruker, Interfile,
DICOM 3.0:
Micro cat, MINC, MRC, FITS, Cheshire, AFNI, TIFF, JPEG, GIF,
Query/Retrieve, Catcher
BMP, AVI, QuickTime, Biorad, Ziess LSM510, XML, and more)

11
Opening Images

12
Opening Images

13
Image Browser

14
Opening Images

15
Saving Image As (use suffix)

16
 Code Snapshot
int destExtents[] = new int[2];
destExtents[0] = image.getExtents()[0]; // X dim
destExtents[1] = image.getExtents()[1]; // Y dim

// Make a result image of Unsigned byte type

resultImage = new ModelImage(ModelStorageBase.UBYTE, destExtents, “Result Image”, null);

int length = destExtents[0] * destExtents[1];

for (int i = 0; i < length; i++){
destImage.set(i, i%256);
}

ViewJFrameImage imageFrame;
ModelLUT LUTa = new ModelLUT(ModelLUT.COOLHOT, 256, dimExtentsLUT);
imageFrame = new ViewJFrameImage(resultImage, LUTa, new Dimension(610,200), userInterface);

17
 Algorithms
• Filters

• Calculation

• Registration

• Transformation

• Surface extraction

• Classification/Segmentation
18
 Download and Setup
1. http://mipav.cit.nih.gov/download
2. Fill in form
3. Install (e.g. installMIPAV.exe)

** Nightly download - lastest

changes but might have bugs.
** Archived releases also available.

19
Memory Allocation

General Rules
• Do not exceed the computer’s physical
RAM. For example if the computer has 1GB
do not exceed approx 800MB.

• For 32-bit Windows systems do not exceed

1,400MB

20
Memory Usage

Press to recover memory

21
MIPAV Program Options

22
 Digital Image Communication in Medicine
(DICOM).
American College of Radiology (ACR) and the National Electrical
Manufacturers Association (NEMA) formed a joint committee in 1983 to develop a
standard in Digital Image Communication in Medicine (DICOM).

1. Promote communication of digital image information, regardless of device manufacturer

2. Facilitate the development and expansion of picture archiving and communication

systems (PACS) that can also interface with other systems of hospital information

3. Allow the creation of diagnostic information databases that can be interrogated by a wide
variety of devices distributed geographically.

23
 DICOM Model
Imaging Workstation (MIPAV)
(PC, MAC, UNIX workstation)

Image
Query
Processing
Images
&
Receiver
Visualization

Internet

DICOM Server

Images Patient
PACS
Database
Imaging Device

24
 DICOM
communication
interface

25
DICOM

26
 DICOM

Access to image header

information

27
DICOM Anonymization

28
DICOM File Browser

29
XML Schema File Format

30
XML Schema File Format

31
XML Schema File Format

32
Image Attributes

33
Volume of Interest (VOI)

34
VOI

35
Lookup Table (LUT)

36
Multi-planar and Lightbox

37
 Image Fusion
The loading of two images into the same frame

Controls blending between the two images

38
Structural MRI and Functional MRI

39
Animation Tool

40
Masks and Surfaces

41
Scripting - Record

42
Scripting - Run

43
Help

44
Bug Report

45
 MIPAV

Visualization

Processing Macros/Plugins

Quantification Ubiquitous File writer

file reader

46
http://mipav.cit.nih.gov

[email protected]
47