tracking thereshold

42 views
Skip to first unread message

najme soleimani

unread,
Apr 10, 2021, 2:54:44 AM4/10/21
to DSI Studio
Dear Frank,
I want to set a tracking threshold that not only removes the false tracts but also best estimates the tracts while using command lines but I face a problem. The tracking threshold varies subject by subject in my data set, therefore I can not set a value for all subjects. Actually I set the diffusion sampling length ratio to 1.25 that best estimates the tracts for my dataset.
I would be grateful if you could help me with it.

Kind regards,
Najme

Frank Yeh

unread,
Apr 10, 2021, 9:45:36 AM4/10/21
to dsi-s...@googlegroups.com
Hi Najme,

    You may use the default setting (not specify it in the command line), and DSI Studio will get an estimate based on each subject's scan condition.

Best,
Frank

--
You received this message because you are subscribed to the Google Groups "DSI Studio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/2ef42458-ca13-4fce-8f7f-ebf32464998an%40googlegroups.com.

najme soleimani

unread,
Apr 10, 2021, 12:30:46 PM4/10/21
to DSI Studio
Actually when I use the default setting for tracking threshold as you mentioned, It estimates a lot of fibers in CSF by mistake and I need to remove them. also this threshold as I mentioned varies subject by subject, therefore I can not set a value for all subjects in command line.
I wonder if there is another solution.
Thanks a lot...


Frank Yeh

unread,
Apr 10, 2021, 12:51:50 PM4/10/21
to dsi-s...@googlegroups.com
I see. Please try --otsu_threshold=0.7 or even higher (e.g., 0.8 0.9...) and do not specify --fa_threshold
Best,
Frank

najme soleimani

unread,
Apr 10, 2021, 3:34:55 PM4/10/21
to DSI Studio
Thank you dear Frank, I try what you said. I changed the --otsu_threshold and use the default tracking thereshold(not specify in command line) but it seems changing this parameter makes no difference in results and still those false tracts in CSF are seen...

best,
Najme

Frank Yeh

unread,
Apr 10, 2021, 5:23:47 PM4/10/21
to dsi-s...@googlegroups.com
Did you use DTI or GQI reconstruction?
Frank

najme soleimani

unread,
Apr 11, 2021, 1:08:10 AM4/11/21
to DSI Studio
GQI reconstruction with diffusion sampling length ratio of 1.25...I also run the steps you have mentioned with GUI interface but the results are the same.
Thanks a million

Frank Yeh

unread,
Apr 11, 2021, 8:26:44 AM4/11/21
to dsi-s...@googlegroups.com
If possible, you can send me a sample SRC file, and I will check if there are other issues.
Frank

najme soleimani

unread,
Apr 12, 2021, 2:38:12 AM4/12/21
to DSI Studio
I drop boxed you the SRC file of my data...Thank you for your consideration.
Best,
Najme

Frank Yeh

unread,
Apr 12, 2021, 11:24:26 AM4/12/21
to dsi-s...@googlegroups.com
Thanks. The data are HCP data, and if you concern about the CSF, you can include CSF regions from Freesurfer segmentation as a terminative region to handle this issue.

Best regards,
Frank


Reply all
Reply to author
Forward
0 new messages