Proposed CNN Model for Classification of Brain Tumor Disease
Rahul Singh, Neha Sharma, Rupesh Gupta
Abstract
A brain tumor is a group of abnormal cells within the brain or surrounding tissues. Several variables, including family history, radiation exposure, and some genetic disorders, might increase the likelihood of developing a brain tumor. The typical method for detecting brain tumors is to perform MRI scans, which a medical specialist then examines for diagnosis. While time-consuming, this process is fraught with the possibility of human error, especially when the tumor is in its early stages. As a result, brain tumor diagnosis must be made properly and as soon as possible. With quick and accurate brain tumor identification, this work aims to prevent premature death, provide health in resource-constrained conditions, and promote patients' healthy lifestyles. A CNN model is created in this study to detect brain cancers, and the dataset contains 251 scans. Because datasets are limited in availability, data augmentation is employed to expand the dataset's coverage. The suggested CNN model's outputs were evaluated using the metrics Accuracy, F1-Score, Precision, and Recall. In aggregate, the model has an accuracy of 85%. As a result, deep-learning CNN models have been demonstrated to detect brain tumors while spending no time or resources effectively.