Atrophy analysis of corpus callosum in Alzheimer brain MR images using anisotropic diffusion filtering and level sets.

Abstract

In this work, an attempt has been made to analyze the atrophy of Corpus Callosum (CC) in Alzheimer brain magnetic resonance images using anisotropic diffusion filtering and modified distance regularized level set method. Anisotropic diffusion filtering is used as preprocessing to obtain the edge map. The modified distance regularized level set method is employed to segment CC using this edge map. Geometric features are extracted from the segmented CC and are analyzed. Results show that anisotropic diffusion filtering is able to extract the edge map with high contrast and continuous boundaries. Modified distance regularized level set method could perform the segmentation of CC in both normal and Alzheimer images. The extracted geometric features such as minor axis, Euler number and solidity are able to demarcate the Alzheimer subjects from the control normals. As atrophy of CC is closely associated with the pathology, this study seems to be clinically useful.