Abstract : Treatment is complicated by the laborious and prone to error nature of modern MRI analysis for brain tumour detection. Using several MRI datasets, we suggest a technique that combines deep learning, conventional classifiers, and CNNs. Algorithms for activation and SVM validation are used. Our CNN surpassed prior records with an accuracy of 99.74% when it was implemented in Python using TensorFlow and Keras. With this automated method, brain tumour diagnosis should be greatly improved, allowing patients' treatment choices to be made more quickly.

Keywords: Alzheimer’s, CNN, Keras, TensorFlow, Max Pooling, Batch Normalization.

| DOI: 10.17148/IARJSET.2024.11642