Rhesus EM Segmentation

Objective:

• Use the EM segmentation algorithm in slicer to segment Rhesus images.

Progress:

• Results from 2007 Project Week
• Followingup on a recommendation by Martin Syner, we have extended the algorithm from the project week to use a lobular parcellation map to automate the splitting of regions. We are calling this Lobulated EM Segmentation (LEMS). We are testing this method on a N=7 subject data set currently and a paper is in preparation.

• LEMS was used to segment 7 subjects imaged in alcohol naive and post induction states. No significant changes in the GM-WM volume ratio was obseved.

• A tutorial for EMSegment (in Slicer2.X) was developed for non-human primate data. The tutorial and the images are available here

Key Investigators:

• Virginia Tech: Chris Wyatt
• NAMIC: Kilian Pohl

Links: