An automated MRI segmentation by using fuzzy C mean and volumetric analysis

Automated MRI segmentation techniques are helpful for a physician for early diagnosis of degenerating diseases in individual patients. Here we are using the T1weighted axial MR images of neuro degenerative diseases. The assessment of the accuracy of the result is done by an expert. FCM an unsupervised clustering technique is implemented in order to classify the brain voxel. The brain voxels are classified into three main tissue types: Gray Matter (GM), White Matter (WM) and Cerebro-Spinal Fluid (CSF). We hypothesized that extracting volumetric data from patient´s MR brain images, relating them to reference data would aid diagnostic readers in classifying neurodegenerative diseases. Volumetric anatomical information extracted from brain images using automatic segmentation can support diagnostic decision making.