A New Method of Brain Tissues Segmentation from MRI with Accuracy Estimation

In this paper, a new concept has been incorporated using level set methodology for the specific segmentation of brain tissues in magnetic resonance imaging (MRI) brain images. In this segmentation, the normal tissues such as WM (White Matter), GM (Gray Matter) and CSF (Cerebrospinal Fluid) with other part of human head such as skull, marrow, and muscular skin are segmented. The segmentation has been done by using repeated level set method based on the condition sharp peak greater than three. The each segmented component is generating a hierarchical structure to make correct tissue segmentation. The performance of the segmentation method is estimated by different accuracy, sensitivity and error correction metric. The performance of segmentation process is analyzed using a defined set of MRI brain. From visualization and mathematical both point of measurement proposed gives very superior results on brain MR images.