Analysis of AI based Brain Tumor Detection and Diagnosis

Abstract

Cancer is caused by uncontrolled cell development in any part of the body. Nowadays cancer is one of the fast growing diseases. Tumors appear in various forms, each with its own set of characteristics and treatment options. Primary and metastatic brain tumors are the two major forms of brain tumors. Detecting and diagnosing the tumor at the early stage can save the lives drastically. Now the fast growth of tumor, presently need research. As a result, scientists and researchers have been trying to develop advanced tools and methods for identifying tumor types and their stages. MRI (Magnetic Resonance Image) and CT (Computed Tomography) are the most commonly preferred modalities for detecting the tumors by re-sectioning and analyzing anomalies in brain tissue or location. Due to the benefits of Magnetic Resonance Image over Computer Tomography scan, the doctors are preferred to use the MRI modality. MRI is non-invasive imaging which is one among the deeply considered modality in the field of medical area network which explain how to distinguish the tumor from the MRI identification of a brain cerebrum. This analysis aims at presenting an overview about Brain Tumor Segmentation, Detection and Diagnosis using Deep Learning (DL), Machine Learning (ML), and Transfer Learning (TL) techniques. This analysis gives an outline of partially and fully automated segmentation techniques that are referred from the previous studies. The collection of the databases used for segmenting and classifying are tabulated in this article. The study blends the presentation of best-in-class approaches from standard segmentation and classification techniques with a quantitative analysis. The comparison mentioned in the articles would greatly provide a detailed summary about the brain tumor detection and diagnosis techniques.