Tissue segmentation by fuzzy clustering technique: case study on Alzheimer's disease

Abstract

Segmentation of brain images especially into three main tissue types: Gray Matter (GM), Cerebrospinal Fluid (CSF), and White Matter (WM) has important role in computer aided neurosurgery and diagnosis. In imaging, physical phenomena and the acquisition system are responsible for noise and the Partial Volume Effect (PVE) respectively, which affect the uncertainty and the imprecision. To reduce the effect of these different imperfections, we propose a clustering approach that is based on a fuzzy- possibilistic segmentation process for the assessment of WM, GM and CSF volumes from Alzheimer’s brain images. The brain segmentation scheme which is illustrated in the study of Alzheimer’s disease using Alzheimer’s disease Neuroimaging Initiative (ADNI) and real images take in consideration the PVE and it is less sensitive to noise.