An Improved System for Brain Pathology Classification using Hybrid Deep Learning Algorithm

Speech, recognizing, learning, programming, and problem-solving are all meant to be included in artificial intelligence computing activities. Artificial intelligence (AI), a subfield of computer engineering, is concerned with developing sophisticated software or technologies that act and function like human. Deep learning, a category of algorithms used in deep learning, is based on learning data representations and is one of many machine learning techniques. Deep learning is used to develop brain tumour screening and classification models for rapid and uncomplicated tumour identification using Magnetic Resonance Images (MRI) imaging. This paper shows how the pretrained model Alexnet with Transfer Learning (TL) may be trained on tiny and innovative fresh data for classification problems using the stochastic-gradient descent with momentum-SGDM optimizer. Numerous hybrid models were presented with the aim of classifying a brain tumour MR image as a benign as well as malignant tumour. The accuracy, error rate, and confusion matrix parameter of the suggested models are evaluated in order to show how the works have improve.