[Mristudio-users] Roi extraction from template and back projection in native space

[Susumu Mori]

The image transformation by LDDMM is reversible. So, A->B and B->A are
determined simultaneously and generate two transformation, Ki and H maps
for both ways.

If you choose an atlas as a template and transform a patient data to it,
patient->atlas is Ki map and atlas->patient is H map.

If you define one ROI by yourself in image A, then you can apply this ROI
to B. There are several important points;

> For the transformation, you want to use "nearest neighbor". Your ROI is
1/0 masking. So, after the transformation, you still want 1/0 masking file
without any interpolated values like 0.5.

> If you define your ROI in only one slice, your ROI may disappear after
the nearest neighbor transformation because the voxels with "1" is
surrounded by a overwhelming number of surronding "0". So, you need to
define your ROI in at least 3 slices.

On Tue, Nov 10, 2015 at 7:29 PM, Shaimaa Abdelsattar <shaimaa96 at hotmail.com>
wrote:

> Thanks Dr.Mori for your reply. I want to ask what is the difference
> between this process and the reverse ( I mean registration of the native
> space into template) regarding the accuracy of Roi data.
> Second question:
> If I applied hybrid approach for Roi drawing in a single slice, will this
> give accurate results as back projection of ROI?
> Sorry for elongation, but I got confused regarding the most accurate and
> precise method for ROI selection.
>
> Thanks
> Shaimaa
> Sent from my Windows Phone
> ------------------------------
> From: Susumu Mori <smoriw at gmail.com>
> Sent: ‎11/‎11/‎2015 02:22 AM
> To: DTI Studio, ROI Editor, DiffeoMap Questions/Support
> <mristudio-users at mristudio.org>; Yajing Zhang <zhangyajing9996 at gmail.com>
> Subject: Re: [Mristudio-users] Roi extraction from template and back
> projection in native space
>
> You need to warp our atlas image to your subject data using DiffeoMap and
> apply the transformation to the atlases ROI file, such that the the subject
> brain is automatically segmented into various regions.
>
> Then you need to pick regions of interest using RoiEditor and feed it into
> DtiStudio for fiber tracking.
>
> I know that this is quite a process to master.
>
> I would recommend to consult with Dr. Zhang about how to streamline the
> process.
>
> On Tue, Nov 10, 2015 at 6:18 PM, Shaimaa Abdelsattar <
> shaimaa96 at hotmail.com> wrote:
>
> Dear experts,
> I'd like to ask how I can extract a roi of certain structure  in the
> template and back project it in subject's native space for more accurate
> measurements using DTI studio, Roi editor or  DiffeoMap.
>
> Thank you in advance
> Shaimaa Mohammad, MD
>
> Sent from my Windows Phone
>
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
>
>
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mristudio.org/pipermail/mristudio-users/attachments/20151114/35d95e80/attachment.html>