A Deep Learning Approach for Brain Tumor Detection using Magnetic Resonance Imaging

Abstract

Brain Tumors are brought on by the proliferation of aberrant cells in the tissue of the brain. One of the most serious conditions that can affect both adults and children is brain tumours. It progresses rapidly, and if the patient is not given the right care, there is little chance of survival. Improving a patient's life expectancy requires accurate diagnosis and well-planned treatment. Magnetic resonance imaging (MRI) is the primary method used to diagnose brain tumours. The suggested model has an activation function, a modified hidden layer architecture, and an automatic feature extractor. After running a number of test cases, the suggested model had a low cross-entropy rate and scored 97.8% precision and 98.6% accuracy. The suggested model has demonstrated superior tumour detection performance when compared to alternative methods including YOLOv5, mask region-based CNN (mask RCNN), adjacent feature propagation network (AFPNet), and Fourier CNN (FCNN).