Survey of Brain Tumour Detection and Prediction Using Machine Learning, Deep Learning and Metaheuristic Techniques

The intricate anatomy of the brain makes it difficult for researchers to identify and predict brain Tumours in their early stages. Manually analyzing many Magnetic Resonance Imaging (MRI) images can occasionally be exhausting and biased. This survey aims at many methods for early brain Tumour diagnosis and prediction using Machine Learning, Deep Learning and Metaheuristics. The survey depicts that Convolutional Neural Network (CNN) is the most preferred algorithm for detection and prediction of Brain Tumour as compared to K-Nearest Neighbor (KNN), Deep Neural Netwok (DNN), Support Vector Machine (SVM), and Artificial Neural Network (ANN). Compared to previous metaheuristic algorithms such as Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), and Whale Optimization have demonstrated superior outcomes in Brain Tumour Detection.