Non-invasive Brain Tumor Detection using Magnetic Resonance Imaging based Fractal Texture Features and Shape Measures

This study presents a novel non-invasive quantitative feature set using magnetic resonance imaging (MRI) for diagnosis of brain tumor and their grade classification. Texture features using segmentation-based fractal texture analysis (SFTA), and selected shape measures were extracted from the segmented tumor volume to differentiate low-grade (LG) tumor and high-grade (LG) brain tumor. Classification of the tumor grade is performed with support vector machine (SVM) classifier and testing and training dataset are obtained using k-fold cross-validation method. The brain detection method proposed in this study gave an overall specificity and sensitivity of 86% and 88% respectively. Also, an accuracy of 87% was achieved while classifying LG and HG brain tumor.