Algorithms for image analysis of corpus callosum degeneration for multiple sclerosis

Multiple sclerosis has been known to cause atrophy and deformation in the corpus callosum. In longitudinal studies, these changes have been typically quantified by using medical image analysis techniques for measuring and analyzing the size and shape of a corpus callosum cross-section embedded in a specially selected measurement plane. Our contributions are three fold: (i) We introduce and quantify the effects of the error in measurement plane selection and image interpolation on the cross-sectional area of the corpus callosum; (ii) We present a novel and clinically meaningful criterion for the measurement plane selection which addresses significant drawbacks with previous methods of plane selection. We present a nested optimization based framework for accurate identification and extraction of this plane with simultaneous segmentation of its corpus callosum cross-section; (iii) We present a medial shape representation based technique for longitudinal, regional and deformation-specific shape analysis of the corpus callosum.