help with peripheral nerve reconstruction

[Michel Audette]

Dear Frank, 

I would like to pick your brains on how to optimize tractographic reconstruction for peripheral nerves, such as the sciatic, which we are attempting here. I would like to find out if we can improve on the current results by using 2D seed contours at both ends of a tract to constrain the path of the reconstruction. Is there a tractography option on DSI for doing that? See existing images provided by my collaborator Rohan Dod. 

In the medium to long term, I would also like to find out if it's possible to set up DSI Studio to use as a cue a warped 3D contour tree of peripheral nerves, drawn by an anatomist in a generic manner, as well as registered to the target anatomy based on muscles and bones, and use this putative 3D contour to generate a series of seed points automatically (in the a manner similar to brain atlases). If this would require significant additions to DSI studio, we could work together on a proposal of some kind, if you have the appetite to take part. I have a top peripheral nerve neurosurgeon with an interest in this.

Thanks for your kind support.  

Warm wishes,

Michel

Michel Audette, Ph.D.

Associate Professor, Department of Computational Modeling and Simulation Engineering,

Graduate Program Director, Biomedical Engineering,

Old Dominion University,

Norfolk, VA.

maud...@odu.edu; office phone: 757-683-6940. 

 

---

From: Dod, Rohan D. <Do...@EVMS.EDU>
Sent: Friday, October 1, 2021 3:45 PM
To: Audette, Michel A. <maud...@odu.edu>
Cc: Shanahan, Jack <jsha...@odu.edu>; Foster, Stephanie <sfos...@odu.edu>; Niaz, Atif <Ni...@EVMS.EDU>
Subject: DSI Studio Screenshots

 

EXTERNAL to ODU: This email is not from an ODU account. Do not click links or open attachments unless you recognize the sender and know the content is safe.

Hello Dr. Audette,

I have attached screenshots from DSI as requested. Please let me know if you would like any additional images. 

Regarding Stephanie's question about using polygonal free-hand to draw a region, it turns out that this actually won't properly allocate the region, so fiber tracking won't work on the ROI. This is hit-or-miss, and, as Dr. Audette knows, this issue has come up in the past. I would still encourage you guys to try it. 

Please feel free to reach out if you have questions!

Best regards,

Rohan Dod
MD Class of 2024 
Eastern Virginia Medical School 
do...@evms.edu
www.evms.edu

[Frank Yeh]

Hi Michel,

You may try a combination of ROI/ROA regions at [Step T3a: Assign
Regions] (details:
http://dsi-studio.labsolver.org/Manual/Fiber-Tracking#TOC-Region-Types
) to get better results.

>
> In the medium to long term, I would also like to find out if it's possible to set up DSI Studio to use as a cue a warped 3D contour tree of peripheral nerves, drawn by an anatomist in a generic manner, as well as registered to the target anatomy based on muscles and bones, and use this putative 3D contour to generate a series of seed points automatically (in a manner similar to brain atlases). If this would require significant additions to DSI studio, we could work together on a proposal of some kind, if you have the appetite to take part. I have a top peripheral nerve neurosurgeon with an interest in this.

We have manually draw brain tumor contours (an example here:
https://www.youtube.com/watch?v=ZkWBU_qnaKg )
It seems that the anatomist can do the same here. The contour can
be used as a limiting region by first inverting it using [Regions
Misc][Modify Region][Negate] and change its region type to
terminative, which will terminate tracts when a fiber reaches it.

If your team encounter any problem, please feel free to let me know.

Best regards,
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/CAPxub5Y%3D4C%3DSTz28N_W4cBY%2B83spFwk9Ly3%3DUiO2hge9BMz0oQ%40mail.gmail.com.