Abstract
This paper presents a new fully automatic method for segmentation of brain images that possess multiple sclerosis (MS) lesions. Multichannel magnetic resonance images are used to delineate multiple sclerosis lesions while segmenting the brain into its major structures. The method is an atlas based segmentation technique employing a topological atlas as well as a statistical atlas. An advantage of this approach is that all segmented structures are topologically constrained, thereby allowing subsequent processing with cortical unfolding or diffeomorphic shape analysis techniques. Validation on data from two studies demonstrates that the method has an accuracy comparable with other MS lesion segmentation methods, while simultaneously segmenting the whole brain.
Keywords
Source Code and Data
No source code files available for this publication.
Reviews
Martin Styner
Tuesday 29 July 2008
This methods extends the existing frameworks for topologically constrained tissue segmentation by Bazin to incorporate an MS lesion model. WM and lesions are grouped together for the topological constraint. Within this joint MS+WM region, each voxel is then classified either as WM or MS based on the computed tissue classification model (unlike WM, lesions have no prior weight in the classification).
The method is quite appealing with good results. Good discussion.
The paper is overall well written and no major revisions are needed.
Simon Warfield
Friday 25 July 2008
This paper describes an algorithm for the segmentation of brain MRI of MS patients.
The primary contribution of this work is the introduction of a topological constraint, which ensures that the segmentation of patients is topologically equivalent to that of healthy subjects.
The evaluation of the method indicates it achieves good performance.