Classification of Brain Tumor MRI Images using Convolutional Neural Network

Abstract

Brain tumors occur as a result of abnormal growth of cells in the brain. There are several different kinds of brain tumors depending on the cells they grow from. The most common brain tumors are Glioma, Meningioma and Pituitary. Over the Years Deep Learning algorithms have shown a significant improvement in determining tumors of the brain. The availability of MRI images has also contributed much to this study since the performance of deep learning algorithms increases in accordance with an increment of data. The different types of tumors of the brain have their own effect on the human body, therefore it is necessary that our focus should not be determining tumor but rather the kind of tumor. In this study, MRI images were pre-processed and trained on a Convolutional Neural Network (CNN) model. After several training, the best model with an F-score of 93.3%, macro-average of 93% and an accuracy of 93.4% with the least prediction error was chosen. © The Institution of Engineering & Technology 2023.