DTI-based thalamus segmentation - A group study

Abstract

Fast and accurate segmentation of deep gray matter structures is important for clinical applications such as for example surgery planning for the placement of deep brain stimulation implants. We implemented a segmentation method that is solely based on local diffusion information and did evaluate it on a group of 53 healthy volunteers to investigate the reproducibility and thereby the usability of this method. We were able to show that the segmentation produces subdivisions which are reproducible for a large population. We could also show that not only the three previously reported large clusters were reproducible but also four smaller clusters could be found in the same position in more than 60% of the subjects.