Evaluation of different stages of Alzheimer's disease using unsupervised clustering techniques and voxel based morphometry

Alzheimer's disease (AD In) is a complex and progressive neuro degenerative disorder. The concept of Mild Cognitive Impairment (MCI) is considered as a well established precursor stage of AD. Early recognition of AD allows time to plan for the future and to treat patients before marked deterioration occurs. The main objective of this longitudinal study is to identify the structural characteristic of Grey Matter (GM), White Matter (WM) and Cerebrospinal fluid (CSF) at baseline and structural changes after one year that could serve as accurate predictors of future development of MCI in the normal aging controls and AD in the MCI patients. This paper we have explored the efficacy of different approaches such as k-means, FCM and VBM for the segmentation and the comparative analysis of longitudinal MR images of AD patients and controls. The Longitudinal structural changes were evaluated using Magnetic Resonance imaging (MRI) over a period of 2 years in 22 MCI patients and 20 normal aging controls. Due to the progressive nature AD, the serial imaging studies tracking the course of progression might improve the diagnostic accuracy of AD. In these results we report a statistically significant trend towards accelerated volume loss in the MCI follow up group compared to the MCI baseline group was observed. The observed changes at an individual level were small, and within the accuracy range of the measurements.