CLASSIFICATION OF SPECIFIC PATTERNS ALLUDING CANCER PRESENCE IN PROSTATIC MRI IMAGES

Several attempts in medical image processing have been made to detect physiological anomalies using various theoretical models .A standard imaginary of suspected regions of the body may have several types of anomalies corresponding to different causes. Detecting most specific ones bearing possibilities of cancer presence has continued to be a challenge. Neural network has been used for classification of images into cancerous and non cancerous types. In this paper we attempt to implement and classify Gabor filtered prostate MRI images using back propagation neural network for identifying tumorous and non tumorous images