Analysis of MRI-based brain tumor detection using RFCM clustering and SVM classifier

Abstract

The infected tumor area from a magnetic resonance image can be segmented, detected, and extracted accurately by the radiologist experts only through the experience. The complexities and limitations involved in this process are investigated/overcome through distributed rough fuzzy C-means (DRFCM). The support vector machine (SVM)-based classifier improves the accuracy and quality of the segmented tissue. Typically, the best clustering process makes the index values of XB, DB, and RAND as minimum as possible. The performance and quality analysis of the proposed method have been evaluated based on the accuracy, specificity, sensitivity, and also the similarity index of dice coefficient.