Decisive tissue segmentation in mr images: Classification analysis of alzheimer’s disease using patch differential clustering

Abstract

Alzheimer’s disease usually occurs in the elderly, and over a course period of time, it contributes to dementia. Substantiation of cerebrovascular disease progressively develops to cognitive performance in the very earlier stage of Alzheimer’s disease. Structural brain imaging plays a vital role in recognition of changes that appear in brain relevant to Alzheimer’s disease and various brain diseases. Magnetic resonance imaging takes place a major task in identification and analyzing of Alzheimer’s disease diagnosing process before the surgical planning. An eminent segmentation process of disease-related tissues leads the physician in making the decision on diagnosing. In this work, segmentation task has been performed with patch image differential clustering (PIDC) based on the labeling over the patches. According to the segmentation operation, the various brain matters like gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) are grouped individually. Finally, the volume estimation of above-mentioned brain matters is adopted for initializing the nearest particle interconnect (NPI) classifier for classifying the abnormality of AD.