Brain Tumor Classification Using Hybrid Model of PSO and SVM Classifier

Abstract

Medical image processing is the very significant process for any disease diagnosis now a days. MRI is usually used to detect the presence of tumor and its type.The process which is followed in classification of brain tumor is very complicated. There are various steps for the medical image processing like image segmentation, image extraction and image classification. Different types of features are extracted from the segmented MRI images like intensity, shapes and texture based features. The feature selection approach is used to select the small subset of features from MRI image which minimize redundancy and maximize relevance to the target. In this paper, online database of MRI images containing brain tumor is taken then a machine learning model is developed by using the Particle Swarm Optimization(PSO) algorithm for feature selection and then Support Vector Machine(SVM) classifier is used to classify the type of tumor in present brain MRI images. © 2018 IEEE.