Automated MR brain image segmentation and classification using neuro fuzzy

Abstract

Segmentation is an important step in many applications, being also important in those that deal with medical images. Thresholding is one of the most important and used techniques for image segmentation. Some segmentation techniques based on thresholding is performed in order to segment the tumors. The conventional method in medicine for brain MR images classification and tumors detection is human inspection. The use of artificial intelligent techniques for instant, neural networks, fuzzy logic, neuro fuzzy has shown great potential in this field. Hence, in this project the neuro fuzzy system or ANFlS was applied for classification purposes. The performance of the ANFlS classifier was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed ANFlS classifier has potential in classifying the tumors.