Performance analysis of multistage brain tumor detection and diagnosis methods using machine and deep learning approaches

Abstract

The fast development of health care technology in the field of newlineinformation and communication engineering is used to give the fast recovering newlinesolutions for many health care problems of the patients around the world. newlineAccording to the report received from National Cancer Institute Statistics newline(NCIS) and World Health Organization (WHO), every year 12,764 persons are newlineaffected by brain tumors. In present, the detection of tumors in brain, lung and newlineliver are screened using various modern health care techniques (Kong et al. newline2015). In this paper, the brain tumor screening using digital health care system is newlineproposed. Brain tumor is the abnormal development of cells in brain regions, newlinewhich can spread the tumor regions to the nearby or its surrounding cells. The newlinedetection of brain tumors in brain Magnetic Resonance Imaging (MRI) image is newlinean important process for preventing earlier death. In this research work, the brain newlinetumors are detected and diagnosed using three different proposed approaches. newlineIn proposed Approach-I, the brain tumors are detected and segmented newlineusing Hybrid Classification (HC) approach. The HC approach stated in this newlineresearch work detects the tumor affected brain MRI image and then newlinesegmentation approach is applied on the detected tumor affected brain MRI newlineimage in order to segment the tumor regions. Finally, the performance of the newlineproposed tumor detection method is compared with the other conventional newlinemethods. newlineThe proposed Approach-II states an automated computer aided newlinemethod for detecting and locating the brain tumors in brain MRI images using newlinemodified deep learning algorithms. The proposed method has three sub modules newlineas preprocessing, classifications and segmentation. In this Approach-II, newlinepreprocessing is used to convert the image resolution into fixed format. The newlinepreprocessed brain MRI images are classified into either tumor case or nontumor newlinecase using classification approach. newline