2009 Summer Project Week 4D Imaging discussion

Meeting: 4D image for 3D Slicer

Time & Place

• 8:30am - , June 23, 2009
• MIT. Grier Rooms A & B: 34-401A & 34-401B.

Agenda

• Demo of the prototype of 4D Image Module in 3D Slicer
• Possible applications, requirements,
• Review the current implementation
  • 4D Bundle: vtkMRML4DBundleNode class
    • Child class of vtkMRMLLinearTransformNode
    • pointers to volume frames
    • pointers to display0 and display1 volume nodes
    • Copies image arrays from volume frames to display0 / display1 to scroll in time-dimension
  • 4D Analysis
    • Python interface
    • Plotting based on vtkXYPlot

• GUI design

Minutes

(Based on Jayender's minutes. Thanks, Jay!)

• Junichi introduced 4D imaging and analysis for lung perfusion.
• Ron suggested to develop another module for preparing the nrrhd file from dicom images since this is input to the 4D module. The user should find it easy to input the images to the module.
• Control GUI for interacting with the module could be added outside the main Slicer GUI.
• For the analysis, a color coding scheme / label name should be introduced to identify the curves being analyzed.
• It was common consensus that this implementation of the 4D module is the best since it is the only one available in Slicer!
• DCE is an emerging capability. Therefore it is important that the module should be capable of handing this scenario.
• fMRI is another possible application for the module
• Danielle’s Cardiac application.
• In ITK, time series database has been implemented. We should evaluate how these two strategies compare with each other.
• Discuss memory usage issue. It would be nice to have pipeline mechanism so that Slicer reads files as required.
• Caching may be good.
• Check volume rendering with 4D images. It is a good application to use CUDA.
• 4D image registration is very useful and should be worked on. Batchmake is also quite useful.

• Evaluate how the 4D module integrates with the change tracker.
• Irregular time stamps for the data should also be included in the 4D data. This would be useful to evaluate the growth or recession of a tumor over time. Explicit time stamping is required.
• It should also be possible to perform parameter calculation based on the actual time stamp. This will enhance the analysis aspect of the module.
• It will also be useful to develop a bidirectional map between the image and parameter space. How does a parameter influence the image as well as parameter evaluation from the images.
• Scalar color maps should be integrated with the 4D module and should be corresponded to look up tables. This is especially useful for nuclear medicine.
• Is it possible to load and manipulate two different fourD bundles simultaneously?
• Hide volume from Data module?

Go to 2009_Summer_Project_Week_4D_Imaging Go to 2009_Summer_Project_Week