AI Driven Brain Tumor Classification and Localization Using Faster R-CNN

The increasing prevalence of brain tumors necessitates the development of accurate, automated methods for early diagnosis and localization. Traditional approaches, which rely on manual MRI interpretation by radiologists, are prone to errors due to varying expertise levels and limited resources. To overcome these limitations, this study proposes an AI-driven solution for brain tumor classification and localization using Faster R-CNN, a deep learning architecture known for its robust object detection capabilities. The proposed model utilizes public MRI datasets to train a Faster R-CNN model, which effectively identifies and locates brain tumors with high precision. The model's performance is evaluated using standard metrics such as accuracy, precision, recall, F1-score, and Intersection over Union (IoU), demonstrating superior results compared to conventional methods. A user-friendly web-based interface developed using Streamlit allows radiologists to upload MRI images and receive real-time diagnostic results, enhancing clinical workflows. Despite its high performance, the study acknowledges the need for improvements in image resolution and model interpretability. Future work will explore the integration of 3D MRI data and explainable AI techniques to further enhance the model's transparency and clinical applicability.