10.22376/ijpbs.2019.10.1.p1-12 Volume 7 Issue 4 2016 (October - December) The classification system is generally categorized into Neural Network(NN) Classification or Data Mining classification and it comprises the tasks of pre-processing, feature extraction, classification and evaluation. The choice of classification method is related to the classes/groups, patterns/features, feature extraction, feature selection, the selection of training, testing samples and its time complexity. Medical image classification using Neural Networks(NN) is a supervised learning method and it is one of the significant research areas assist to examine the patient's images and is an important task of medical image analysis for computer aided diagnosis. The objective of this work is to segment and classify the Regions Of Interest (ROI) from MRI brain images using semi supervised approach referred as RKFCM-RBF-QPSO. The experimental section of this paper shows that the proposed approach produces accuracy of 98% with Root Mean Square Error(RMSE) of 0.1897 which is found to be better than other learning methods. ANUSUYAS.VENKATESAN RKFCM, RBF, QPSO, MRI images, Clustering and Classification. 655-661