No abstract

In this paper, we present an automated SVM segmentation scheme for the MR images for the early diagnosis of neurodegenerative diseases. This method consists of three steps. In the first method we undergo the preprocessing part which removes the skull and the unwanted areas, and then the features are extracted. The third step, the multiple SVM is used to segment the MR images into Gray Matter (GM), White Matter (WM), and Cerebro Spinal Fluid (CSF). The SVM technique is a powerful discriminator is able to handle nonlinear classification problems. The proposed method is used to segment GM, WM and CSF from real magnetic resonance imaging (MRI) in south Indian population. The automated segmented brain tissues are then evaluated by comparing it with the corresponding ground truth set by the radiologist.

Brain tissue segmentation of MRI helps in the possibility of improved clinical decision making and diagnosis, and it also gives a new insight into the mechanism of the disease. Manual interaction is time consuming and it may be bias and variable. We have developed an automatic segmentation algorithm for brain MRI using Artificial Neural Network (ANN) technique. The contribution of this work is an approach for automatically segmenting the brain tissues into White Matter (WM), Gray Matter (GM), Cerebro-Spinal Fluid (CSF). The algorithm developed achieves better performance comparable to the expert segmentation. However, in the clinical analysis accurate segmentation of MR image is very important and crucial for the early diagnosis. Segmentation procedure was done with real time MR images

No abstract