Fiber count of SAF.tck does not match the number of rejected fibers

[cmendosanchez]

Dear Dmitri,

I used the following command to perform the surface-based tracking for short association fibres.

[SD, PS] = SAF_pipeline('fs_dir', fs_dir, 'fod', fod,'noregister', 'tckgen', '-cutoff 0.05 -select 50000 -maxlen 40', 'surfseed','remesh_seeds');

The script generated the following print

Converting brain.mgz to brain.nii.gz
Assuming T1 and DWI are co-registered
Processing surfaces
Edge length cut off for white surface remeshing (seed input): 1.68
Number of coordinates added for seeding: 16802
Running mrtrix tckgen
tckgen string:
tckgen -seed_coordinates_global /home/cris/Escritorio/fod/103818/SAF_output/remeshed_seeds.txt -output_seeds /home/cris/Escritorio/fod/103818/SAF_output/DWI_seeds_out.txt -cutoff 0.05 -select 50000 -maxlen 40 /home/cris/Escritorio/fod/103818/wmfod_norm.mif.gz /home/cris/Escritorio/fod/103818/SAF_output/surf_tracks.tck -force
Loading streamlines
Preparing for streamline-cortex filters. Current time: 19:49
Starting streamline-cortex filters. Current time: 19:49
Number of streamlines before filtering: 50000
Starting GG & HH filtering
Time spent on GG & HH filtering: 00:00:00
Number of streamlines rejected due to extremely short length: 0
Number of streamlines rejected during GG filtering: 30975
Number of streamlines rejected during HH filtering: 125
Filtering lengths
Number of streamlines rejected due to maxlen: 0
Starting GWG filtering
Time spent on GWG filtering: 00:00:07
Number of streamlines rejected during GWG filtering: 5878
Final SAF tractogram: /home/cris/Escritorio/fod/103818/SAF_output/SAF.tck
Total time taken: 00:00:40
Saving final matlab files

The SAF.tck file had a fiber count of 13147. However, it should be less taking into account the GG, HH and GWG filtering (50000-30975-125-5878 = 13022 fibers).

I suspect that the fibers rejected by the HH filter (125 fibers rejected) were not removed from the SAF.tck file due to 50000-30975-5878 =13147 fibers.

Do you have any advice on how to solve this issue? Also, I was wondering if it is possible to retrieve the indices of the intersected triangles from the SAF.tck fibers.

Many thanks,

Cristóbal

[dmitrishastin]

Hi Cristóbal, I am afraid I am not working on this project for the rest of the year due to other work commitments. I won't be able to look into this for you, but in short, from memory I think the filtered out streamlines should be appropriately removed from the final tck file but the stepwise numbers generated in the report were a bit off - so it was a reporting issue not a filtering issue (again, this is from memory!). The indices of the intersected triangles I believe should be stored in SD structure but admittedly I haven't made it easy to work out what is what! They do not get sored in the .tck as that would increase file size unnecessarily. D

…

On Sun, 10 Sept 2023, 00:33 cmendosanchez, ***@***.***> wrote: Dear Dmitri, I used the following command to perform the surface-based tracking for short association fibres. [SD, PS] = SAF_pipeline('fs_dir', fs_dir, 'fod', fod,'noregister', 'tckgen', '-cutoff 0.05 -select 50000 -maxlen 40', 'surfseed','remesh_seeds'); The script generated the following print Converting brain.mgz to brain.nii.gz Assuming T1 and DWI are co-registered Processing surfaces Edge length cut off for white surface remeshing (seed input): 1.68 Number of coordinates added for seeding: 16802 Running mrtrix tckgen tckgen string: tckgen -seed_coordinates_global /home/cris/Escritorio/fod/103818/SAF_output/remeshed_seeds.txt -output_seeds /home/cris/Escritorio/fod/103818/SAF_output/DWI_seeds_out.txt -cutoff 0.05 -select 50000 -maxlen 40 /home/cris/Escritorio/fod/103818/wmfod_norm.mif.gz /home/cris/Escritorio/fod/103818/SAF_output/surf_tracks.tck -force Loading streamlines Preparing for streamline-cortex filters. Current time: 19:49 Starting streamline-cortex filters. Current time: 19:49 Number of streamlines before filtering: 50000 Starting GG & HH filtering Time spent on GG & HH filtering: 00:00:00 Number of streamlines rejected due to extremely short length: 0 Number of streamlines rejected during GG filtering: 30975 Number of streamlines rejected during HH filtering: 125 Filtering lengths Number of streamlines rejected due to maxlen: 0 Starting GWG filtering Time spent on GWG filtering: 00:00:07 Number of streamlines rejected during GWG filtering: 5878 Final SAF tractogram: /home/cris/Escritorio/fod/103818/SAF_output/SAF.tck Total time taken: 00:00:40 Saving final matlab files The SAF.tck file had a fiber count of 13147. However, it should be less taking into account the GG, HH and GWG filtering (50000-30975-125-5878 = 13022 fibers). I suspect that the fibers rejected by the HH filter (125 fibers rejected) were not removed from the SAF.tck file due to 50000-30975-5878 =13147 fibers. Do you have any advice on how to solve this issue? Also, I was wondering if it is possible to retrieve the indices of the intersected triangles from the SAF.tck fibers. Many thanks, Cristóbal — Reply to this email directly, view it on GitHub <#4>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANC6TATVJAUAZDGCWLXDYYLXZT4ENANCNFSM6AAAAAA4RX3UBM> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>