Stroke Tractography Analysis

[Eric Kirby]

Hello!

I’m planning on starting up an analysis and I wanted to run some ideas by the group. I have DWI data, T1s, lesion masks (already in T1 space), and myelin water imaging data of individuals who suffered a stroke. My plan is to focus on white matter tracts of these individuals, specifically looking at FA, as well as myelin water along these tracts.

1. How should I be dealing with registrations of these different scans? Normally with stroke participants, it is common to use FSL and the lesion masks to do lesion filling of the T1 image and then register DWI and myelin water scans to that. However, I know DSI Studio has its own registration software. Would it be capable of registering T1, DWI, and myelin water to the same space accurately, even with lesions present?

2. After successful registration, I’d like to perform tractography to pull white matter tracts from each individual. Because they have suffered strokes and have lesions potentially affecting their white matter tracts, I’m thinking of using an ROI based approach instead of the tract recognition software DSI studio has. Thoughts?

3. If the myelin water scan is properly registered to the same space as the DWI, I’d like to look at tract profiles of FA and the myelin water measurement. I know I can just pull these using DSI Studio. However, I’d also like to calculate lesion load (see if the lesion from the lesion mask is touching/interacting with the tract in anyway). How do you propose I can quantitatively do that?

Thank you for reading and I look forward to your advice.

Cheers,

Eric

[Frank Yeh]

>
> 1. How should I be dealing with registrations of these different scans? Normally with stroke participants, it is common to use FSL and the lesion masks to do lesion filling of the T1 image and then register DWI and myelin water scans to that. However, I know DSI Studio has its own registration software. Would it be capable of registering T1, DWI, and myelin water to the same space accurately, even with lesions present?

It should work most of the time. If not, there is manual adjustment available.
You may check out other dicussions, or tutorial videos at
practicum.labsolver.org

>
> 2. After successful registration, I’d like to perform tractography to pull white matter tracts from each individual. Because they have suffered strokes and have lesions potentially affecting their white matter tracts, I’m thinking of using an ROI based approach instead of the tract recognition software DSI studio has. Thoughts?

I would do the same. The tutorial videos demonstrate some similar
works using a connectometry database.

>
> 3. If the myelin water scan is properly registered to the same space as the DWI, I’d like to look at tract profiles of FA and the myelin water measurement. I know I can just pull these using DSI Studio. However, I’d also like to calculate lesion load (see if the lesion from the lesion mask is touching/interacting with the tract in anyway). How do you propose I can quantitatively do that?
>

Good point and idea.
What I would do is to convert both tract and lesion to ROI as NIFTI
files, and use python code to compute the intersected volume.

There could be other working approaches.

Best,
Frank

> Thank you for reading and I look forward to your advice.
>
> Cheers,
>
> Eric
>

> --
> 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 visit https://groups.google.com/d/msgid/dsi-studio/36f28f92-2798-4962-84a8-336235020253n%40googlegroups.com.