Brain Tumour Detection using Convolutional Neural Network with Machine Learning Algorithm

Detecting brain tumour earlier is very important and critical for treating a patient. By using feature extraction combined with convolutional neural network (CNN), this paper proposes a different approach to detecting brain tumours earlier as possible. NMRs of both healthy and brain tumour patients are taken for the dataset with CNN and classifiers used to evaluate the performance. The results show that the Sigmoid and ReLu classifiers achieved a good accuracy of 97.55%. Deploying a feature extraction algorithm in the CNN has increased the accuracy to 99.05%. The method proposed helps to improve the accuracy and precision of brain tumour decisions.