Even with increasing popularity of MRI imaging techniques, the assessment of lesions in brain area is still performed manually or semi-manually. The major drawbacks to manual image segmentation are time consuming and subjectivity of human decision. Manual assessment of pathological changes is too cumbersome and it is not devoid of errors. Therefore, development of tools for an automatic assessment of lesions in a brain area is one of the most challenging tasks of present day medical image processing. The crucial problem in an automatic assessment of brain tumors is image segmentation. The tumors differ in shape, size and location, and they may appear at different places with different intensities. Therefore, it is very difficult to find the precise tumor in the brain. The proposed method helps in the automatic detection of brain tumor through the help of level set method and the classification of tumor as benign or malignant using the Artificial Neural Network.