Two-Dimensional Stockwell Transform and Deep Convolutional Neural Network for Multi-Class Diagnosis of Pathological Brain

Since the brain lesion detection and classification is a vital diagnosis task, in this paper, problem of magnetic resonance imaging (MRI) investigated. Recent advantages machine learning deep allows researchers to develop robust computer-aided (CAD) tools for lesions. Feature extraction an essential step any scheme. Time-frequency analysis methods provide localized information that makes them more attractive image applications. Owing two-dimensional discrete orthonormal Stockwell transform (2D DOST), we propose use it extract efficient features from MRIs obtain feature matrix. there are some irrelevant features, two-directional principal component ((2D)2PCA) used reduce dimension Finally, convolution neural networks (CNNs) designed trained MRI classification. Simulation results indicate proposed CAD tool outperforms recently introduced ones can efficiently diagnose scans.