Alzheimer Disease Detection Using Edge Enhanced K Means Clustering Algorithm

Abstract

The wide usage of Segmentation of images is in medical application. In medical imaging the most significant problems that we come across is misclassification and discontinuous boundaries. Further low contrast, complex shape and unclear boundaries also deteriorate the Magnetic Resonance Images segmentation performance. In order to extract brain tissues we need to register either with deformation model or any atlases. In this paper we have addressed all these problems and extracted most significant brain tissue hippocampus by merging clustering approach and region growing techniques. After analyzing the structure of hippocampus we can easily diagnose many cognitive disorders. An edge enhance K Mean Clustering Algorithm has been proposed which successfully extracts the hippocampus and gives comparatively better result than existing approaches.

Keywords: Misclassification, Hippocampus, Skull Stripping, Edge Enhancement, Magnetic Resonance Imaging (MRI), Alzheimer’s disease (AD)