Brain Tumor Classification Based On MRI Image Processing With Alexnet Architecture

Abstract

The high mortality rate due to brain tumors causes this disease needs to be detected early. Several studies have developed an automatic brain tumor classification to detect brain tumors. However, most proposed solution methods rely on 2 class classification and manual extraction. Therefore, we created a system to classify brain tumors into four classes: no tumor, glioma, meningioma, and pituitary. The system will be using a Convolutional Neural Network (CNN) with the AlexNet architecture with integrated feature extraction. The data for this study was obtained from Kaggle and consisted of 3,264 data. In the training process, we applied cross-validations to get the best model used in testing. The results show that the highest accuracy is obtained when using the Adamax optimizer and a learning rate of 0.001 is 93 in categorizing brain tumor type datasets. Which indicates the system can be effectively implemented in brain tumor classification.