Convolutional Neural Network Based Multi Class Classification Model for Brain Tumor Diagnosis

Classification of brain tumors is a key task for evaluating tumors and treating them according to their classes. There are a variety of ways for detecting brain tumors, but MRI is the most often utilised technique which gives more accurate result than other techniques. The Deep Learning (DL) models performed exceptionally well in classification and segmentation tasks. This research proposes a Convolutional Neural Network (CNN) based multiclass classification model that is capable of classifying brain tumor in to different classes. Dataset for this research has been obtained from Kaggle, which has roughly 2870 images. The tumors are divided into four groups in this dataset (meningioma tumor, glioma tumor, pituitary tumor and no-tumor). The acquired result demonstrates the modal's ability to classify brain tumors.