Rough K-means and support vector machine based brain tumor detection

In this paper, we present a proposed algorithm to classify brain MRI as tumor-free or tumor present. For computing difference between normal and abnormal MR images, a set of features is calculated. The number of features of the original feature set is reduced by rough set based K-means algorithm and classification of the dataset into tumor-free or tumor-present category has been done by support vector machine (SVM). The proposed algorithm gives better classification result for a smaller set of training dataset.