2008 Winter Project Week Tractography Meeting Notes

Brief Summary

"We know when the track is bad (or ugly) but we don't know when it is good." CF Westin

1. All sites could analyze data
2. All sites could generate all requested tracks
3. ROIs need work
4. Ugly R fornix

Action Plans

• Upload to our SRB workspace the DTI registered (using FSL's FLIRT) Freesurfer generated white matter mask in nrrd format at same resolution as DTI data for each subject (Anastasia, Sylvain, bug-reporter as needed).
• Upload to our SRB workspace DTI registered additional neuroanatomical label maps for tracks (e.g. ending structures) adding arcuate fasiculus as agreed upon by clinical break-out group at 3-4 PM (Randy/Marek/Sonia).
• Upload to our SRB workspace helical phantom binary code and other materials with documentation (Nathan Hageman) and download to play with (all others)
• Regular bi-weekly T-cons, Tuesdays at 11 AM EST/ 9 AM MST/ 8 AM PST
• First T-con will be Tuesday January 15 or 22
• Agenda for first T-con will be review of Nathan Hageman's helical phantom and choice of metrics for all to generate and share.
• All sites analyze full n = 9 test/retest data set and sent agreed upon metrics to Sonia in time for Summer Programmer's week.
• Intermediate goals to be set by group in our regular T-cons

Discussion Topics

1) ROI definitions and impact on different methods

• Force all methods to use the same ROIs or branch and have two or more classes of optimal ROIs?
• For streamline methods larger ROIs than the tract itself good
• For shortest path methods, connections stop at ROI
• Role of DTI Atlas
• Missing arcuate fasiculus

Summary:

• We should all start with the same ROIs, and then allow them to be adapted to specific method, as long as that process is well-documented. Report back the modified/new ROI.
• Need to add end-points
• Randy and others will explore finding personnel and resources to generate expert neuroanatomical segmentation of the tracks.

2) Debugging

Create/use existing synthetic tensor data set (Gordon's helix, crossing fiber situation). Good for debugging algorithm method comparison. Nathan Hageman w/input from other algorithm teams will generate helical synthetic data set. What outcome metrics from it will each team generate?

3) Review and define metrics to compare results, which methods to compare?

Current metrics: tensor invariants (FA, trace, mode, norm), volume, length.

1. Binary mask is needed for volume calculation. How do we generate binary masks and choice of threshold is a part of the method.
2. Possibility to include weights to voxels inside the mask.
3. Overlap metrics can be used to compare methods with different spatial extent. Need to send the binary masks to Sonia to do such calculations.
4. Masks allow us to look at distribution of measures
5. Concentrate on two tracts?

• Parameter settings within a method may change the results comparable to the difference between methods.
• Bundle thickness: mix of radius of tract and uncertainty
• How to define volume from connectivity map (shortest path, or stochastic)
• Quantification of registration error and its propagation forward and impact on final results

1. Agree on units for volume measures. Suggest cubic centimeters (ml) vs mm3

How to approach Test- Retest data

Start discussion

Milestones and dates for next steps in project

Registration &/or resampling to support visit 1- visit 2 analysis and group analysis

Should we make plans for a NAMIC sponsored MICCAI event: Tractography Grand Challenge e.g. the one done last year for segmentation? See http://mbi.dkfz-heidelberg.de/grand-challenge2007/)

4:00- 5:00 PM NAMIC DTI module discussion

The NAMIC DTI module should include a comprehensive set of modules that allow users to efficiently explore, process visualize and analyze diffusion imaging data. This breakout first discusses the status of each Core-1 site w.r.t. integration of tools to Slicer-3, then reviews the current status of the Slicer-3 DTI module, and finally contrasts the two to find gaps and open issues.

• Each Core-1 DTI development site: Status Slicer-3 Integration
  • Utah-I (Tom Fletcher/ Ross Whitaker):
    • Volumetric connectivity ready as command-line procedure, only needs a wrapper

for Slicer-3

• • • Current activity: EPI and Eddy-current correction based on field maps (will

collaborate with C-F Westin)** Georgia Tech (John Melonakos/ Alan Tannenbaum):

• • • Geodesic Tractography Segmentation ready as command-line procedure, only

needs a wrapper for Slicer-3

• • Utah-II (Casey Goodlett/ Guido Gerig):
    • library of tensor estimation modules
    • modules for mapping/resampling tensor fields (activity for ITK modules to be

synchronized with Francois Budin/Sylvain Bouix)

• • • group-wise population analysis available as batch process but needs

NAMIC-compatible framework for group-wise registration of populations of images. Collaboration with MIT (Serdar Balci, Polina Golland)

• • • Fiber bundle editing/postprocessing/parametrization: FiberViewer might be

operated as command-line tool with wrapper. Collaboration with engineering core (Alex Yarmarkovich, Steve Pieper) w.r.t. full integration of editing into Slicer-3.

• • Harvard (Tri Ngo, C-F Westin):
    • filtering modules and stochastic tracking available
    • work towards full integration into slicer, e.g. by pull-down menue to select

different tractography procedures

• • UCLA (Nathan Hageman):
    • helix 3-D tensor phantom to test correctness of different DTI processing

methods

• • • development of fluid-mechanics-based tractography as standalone ITK module,

requires parallel processing and interactive visualization

• • • development of processing for diffusion spectrum imaging (DSI) / analysis of

ODFs

• Engineering Core-2: Status, next steps (Alex Yarmarkovich, Steve Pieper)

• Contrasting plans of Core-1 with future development plans as identified by the

Engineering Core (Alex Yarmarkovich)Slicer3 DTI status

• • nifty to nrrd: in progress by engineering core
  • The new mBIRN compliant DTI data from MIND used for the tractography comparison

used the following preprocessing:

• • • Registration (we used FLIRT for DTI to structural MR)
    • EPI distortion correction (we used FUGUE), ITK?

It will be a priority to develop these preprocessing functions within Slicer-3:

* Registration: see update by Stephen Aylward in AHM Registration breakout 

[1], Demonstration of a registration module in Slicer [2]

• Eddy-current and EPI correction see above plan by Westin/Fletcher/Whitaker

The DTI workshop group concluded that the list provided by Alex matched deliverables proposed by developers very well.

• Action plan Core-1 and Core-2 interaction: Core-2 teams will visit Core-1

developers for face-to-face discussion of transfer and integration of tools. It was decided in a separate meeting that first, the Utah groups will host the engineering Core-2 people for primarily discussing Slicer-3 DTI module developments.

Notes

From Casey's presentation: Could do tractography for all tracks. Made suggestion to . His Atlas presentation brought up idea of using one set of ROI specifications for an entire study in atlas space and map them back to the individual.

From John's presentation: His method required that he had to shrink the provided ROIs for uncinate fasciulus and corpus callosum to make his method work. Then could generate fiber volumes for each tract. His tracts all STOP at ROIs (end points). If given ROIs farther apart could generate longer tracts. Needed a white matter mask- created his own with Tri by thresholding FA (>0.15). Discussion regarding whether the ROIs based on Susumu Mori's criteria would actually be optimal or even work for every method.

From Tri's presentation: Able to generate all tracks with given ROIs. Needed white matter mask (see above). 30 minutes/tract on 63GB computer. Seed with source, filter with sink. Had to cut uncinate fasciulus. For later statistical analysis, considered only non-zero connectivity value voxels. Reminds us that bundle thickness is related to both uncertainty and physical radius Internal capsule, why so much blue?

From Vince Magnotta's presentation: GTRACT fine for forceps major and minor, internal capusule and Fast Marching algorithm better for fornix, hippocampus-cingulum and cingulum. For the Fast Marching algorithm, modified the algorithm to use cost image generations based on tensor and anisotropy, gradient descent. His algorithm would perform "better" for core of cingulum bundle, needs a smaller, more focal ROI. For uncinate, couldn't get it to bend around, would need intermediate ROIs.

Sonia's preliminary cross-algorithm study:

1. rank order of qualitative accuracy (strongly underscores need to clarify the clinical goal of the tractography results)

From Carlo:

1. Commended us for making effort against this tough problem and for the decision to control pre-processing steps
2. What is the impact of adjusting parameters within an algorithm on the outcome metrics? May need to explore this in addition to the cross algorithm assessments.
3. What is the effect of the tissue properties that surround the track on results within and across algorithm.
4. Differences due to the topology, how to visualize? Binary masks in/out of track volume.

Back to 2008_Winter_Project_Week_Tractography, AHM_2008, Events