Evaluation of Deep Learning Approaches for Detection of Brain Tumours using MRI

A significant health problem that can be fatal is a brain tumour, if it not detected and cured at the right time. Therefore, early tumour detection is essential for arranging therapy as soon as possible. One of the most important factors in neurosurgery is the identification of brain tumour boundaries. Among the most serious reasons for death in humans is a brain tumour, which is an abnormal development of brain cells. A technique for detecting brain tumours can identify early-stage tumours. Magnetic Resonance Imaging (MRI) segmentation of brain tumours is the field's dominant research topic these days. Finding the precise dimensions and position of brain tumour monitoring is a very helpful procedure. These Content-based Image Retrieval (CBIR) techniques are now widely used in the automatic diagnosis of disease using MR imaging, mammography, and other sources. This gap can be addressed utilising the deep learning feature extraction technique and the innovative edge detection method, bringing accuracy noticeably closer to the manual results of a human evaluator as part of the goal of sustainable development through innovation. This paper provides the in-depth survey of the several techniques used by many researchers and concludes that the best strategy to identify the region of interest is Fuzzy C-Mean Algorithm among possible automated segmented techniques.