Automated Brain Tumor Classification in MRI Scans Using Machine Learning Techniques

Brain tumor MRI image classification is crucial for early diagnosis and efficient treatment planning, minimizing the risk of misdiagnosis and enhancing patient outcomes. Conventional approaches, such as manual diagnosis and simple machine learning models, tend to be challenged by feature dependency, lack of interpretability, and inability to capture intricate patterns. The proposed work overcomes these limitations using ensemble models such as Extra Trees and Random Forest that improve prediction precision by learning complex data patterns. Furthermore, interpretability techniques such as SHapley Additive exPlanations and Local Interpretable Model-agnostic Explanations offer explainability, facilitating accurate decision-making and rendering AI-augmented diagnosis more credible in medical imaging scenarios.