Using itk-snap for automatic/semi automatic segmentation of Low Grade Glioma

125 views
Skip to first unread message

youri.ko...@telecomnancy.net

unread,
May 3, 2016, 12:15:17 AM5/3/16
to itksnap-users
Hello,
First i'd like to thank you for sharing and bringing forth such a reliable software.
I'm currently working on the automatic/semi automatic segmentation of Low Grade Glioma for a project and we are looking for a software to do it, itk snap seems to be a good choice.
However i'm experiencing some difficulties on Low grade glioma since the contrast and contour difference isn't so easy to see, do you have any hints on how i could clearly and easily split it (by using any means of segmentation in the software)
Lasty is a fully automatic segmentation possible in your opinion with itk snap ?

You'll find a picture that I may need to segmentate for instance attached.
(If you need the Dicom file ask me i'll share it with you by google drive, it is too big to be sent here)

Feel free to ask any question,
Thank you for your attention and your time.
picture.PNG

Paul Yushkevich

unread,
May 3, 2016, 11:07:02 AM5/3/16
to itksna...@googlegroups.com
Hello Youri,

We have been using the semi-automated mode of ITK-SNAP for some high-grade GBMs, and here is a video that shows how this can be done step by step. It really helps to have multiple modalities (T1, T1-contrast enhanced, T2, FLAIR). 


I suggest using a similar approach.

Thanks!
Paul

--
You received this message because you are subscribed to the Google Groups "itksnap-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to itksnap-user...@googlegroups.com.
To post to this group, send email to itksna...@googlegroups.com.
Visit this group at https://groups.google.com/group/itksnap-users.
For more options, visit https://groups.google.com/d/optout.



--
Paul A. Yushkevich, Ph.D.
Associate Professor
Penn Image Computing and Science Laboratory
Department of Radiology
University of Pennsylvania

Alina _

unread,
Jun 21, 2016, 5:38:39 AM6/21/16
to itksna...@googlegroups.com
Hi,

Thanks for the video, I find the segmentation by tissue classes a great tool. Can the program learn from a training dataset and apply that knowledge on another dataset without the need to draw the samples for the new dataset?

Thanks,
Alina

Paul Yushkevich

unread,
Jul 1, 2016, 7:30:37 AM7/1/16
to itksna...@googlegroups.com
Hi Alina

This functionality is currently supported in C3D. Please see the documentation for the commands -rf-train and -rf-apply. This is not currently available in ITK-SNAP.

Paul
Reply all
Reply to author
Forward
0 new messages