Abnormal Brain Tumors Classification

The classification of abnormal brain tumors is vital for accurate diagnosis and effective treatment planning in neurology. This study investigates the application of advanced deep learning models, including ResNet50, MobileNet, InceptionNet, and a traditional Convolutional Neural Network (CNN), for brain tumor classification. Using a comprehensive dataset categorized into glioma, meningioma, no tumor, and pituitary tumors, the research evaluates these models' performance. ResNet50’s deep residual framework, MobileNet’s lightweight design, and InceptionNet’s multi-path architecture are assessed alongside a standard CNN to identify their strengths and limitations in processing medical imaging data. Evaluation metrics such as accuracy, sensitivity, specificity, and computational efficiency provide a comparative analysis of each model. The findings offer critical insights into developing automated diagnostic tools, advancing the precision and efficiency of brain tumor classification systems in clinical applications.

Keywords: Brain Tumor Classification, ResNet50, MobileNet, InceptionNet, Convolutional Neural Network, Medical Imaging, Deep Learning.