MRI image categorization and identification of brain tumours based on bandlet transforms utilising neural networks as opposed to support vector machine classifier

In order to diagnose brain tumours more effectively, the research compares and contrasts two algorithms: a neural network classifier and a support vector machine. The data set used for this experiment is taken from the Kaggle open access dataset.A total of 20 number of MRI images used for Neural Network (group 1) it is compared with support vector machine (group 2).The maximum allowed error is fixed at 0.5 and the sample size is assumed to be 20 in each group based on prior findings with an 80 percent G power.80 percent of the photos are utilised for training, 10 percent are used for testing, and 10 percent are used for validation. Performance analysis is completed to verify the more effective methods in SPSS software. The research using Support Vector Machine (SVM) in detection of brain disorder has attained accuracy of 92% and the proposed system using neural network has attained Accuracy (96%) in diagnosis of brain tumor with the significance level
$\mathrm{p}_{\unicode{x00A1}{0.05}}$
which gives better sensing performance. The diagnosis of brain tumour disease utilising neural networks appears to have superior outcomes than support vector machines in this approach of brain tumour detection (SVM)