An Automated Brain Image Analysis System for Brain Cancer using Shearlets

In this paper, an Automated Brain Image Analysis (ABIA) system that classifies the Magnetic Resonance Imaging (MRI) of human brain is presented. The classification MRI images into normal or low grade high plays a vital role for early diagnosis. Non-Subsampled Shearlet Transform (NSST) captures more visual information than conventional wavelet transforms employed feature extraction. As space NSST very high, statistical t-test applied to select dominant directional sub-bands at each level decomposition based on sub-band energies. A combination features includes Gray Level Co-occurrence Matrix (GLCM) features, Histograms Positive Coefficients (HPSC), and Negative (HNSC) are estimated. combined set utilized in phase where hybrid approach designed with three classifiers; k-Nearest Neighbor (kNN), Naive Bayes (NB) Support Vector Machine (SVM) classifiers. output individual trained classifiers testing input hybridized take final decision. quantitative results ABIA Repository Molecular Neoplasia Data (REMBRANDT) database show overall improved performance comparison single classifier model accuracy 99% normal/abnormal 98% risk classification.