Abstract

The proposed work is to present an effective approach to diagnose of dementia using MRI images and classify into different stages. There are many manual segmentation algorithms on detection and classification or very simple and specific segmentation algorithms to segment each region of interest exclusively. Thus, the proposed system shall use one of the most effective automatic segmentation techniques on MRI images at once. The regions of interest to segment are CSF (Cerebralspinal fluid), gray matter, and white matter and ventricles using the effective segmentation method called level set segmentation. The features are extracted from these four regions of interest and classification of the dementia is performed using K-nearest neighbor.

Authors and Affiliations