A Short Literature Review of Brain Tumour Segmentation and Classification Framework: Challenges Issues and Future Scope

Brain tumor is mainly caused by the deformed or uncontrolled tissue or cells in the human brain. In the medical industry, the brain tumor is effectively diagnosed by using Magnetic Resonance Imaging (MRI) images. The former imaging methods are more time-consuming, dangerous and invasive. Hence, the tumor segmentation and classification becomes the challenging task. Rather than other imaging types, MRI is as effective as containing the more information for diagnosing the tumor disorder. Yet, the manual process is futile to detect the brain tumors. Hence, Computer Aided Diagnosis (CAD) is a pre-requisite in hospitals, diagnostic centers, etc. However, such models are also facing the troubles like misclassification issue; maintaining the large-scale data, some complexities, etc. Therefore, this survey concentrates on exploring the literature survey of traditional brain tumor segmentation and classification approaches using MRI images. It also demonstrates the various learning methods for analyzing the performance, in which the certain challenges are noted. Further, the introduction section is followed by review works and a chronological analysis of conventional approaches. It accounts for dataset utilization, deployed techniques, performance measures, and advantages and disadvantages. At last, the research gaps illustrate and direct future research trends of developing the effective segmentation and classification model.