International Journal of Advanced Research in Computer and Communication Engineering

Abstract: Brain tumors necessitate early and precise diagnosis for improved patient prognoses. This project introduces an advanced automated system for brain tumor detection, classification, and staging utilizing MRI imaging and Convolutional Neural Networks (CNNs). The system employs robust image preprocessing and tumor segmentation, followed by deep learning-based classification to identify tumor type (e.g., glioma, meningioma, pituitary), stage (early, intermediate, advanced), and precise spatial location. Rigorous evaluation on public datasets demonstrates high accuracy in detection and classification across key metrics, affirming its diagnostic efficacy. Evaluated on publicly available datasets, the system demonstrated high accuracy in detection and classification, evidenced by strong metrics like precision, recall, F1-score, and overall accuracy. A user-friendly graphical interface (GUI) is also integrated for easy visualization and interpretation by healthcare professionals. Coupled with an intuitive graphical user interface for clinical interpretability, this non-invasive and time-efficient solution significantly reduces diagnostic error and aids in early intervention. This pioneering framework holds substantial promise for revolutionizing clinical diagnostics and treatment planning, with future potential for 3D imaging integration and enhanced model robustness. This automated and reliable solution has significant potential for clinical diagnostics and treatment planning by reducing human error and facilitating early diagnosis.

Keywords: Brain Tumor, MRI, Deep Learning, Convolutional Neural Networks, Image Segmentation, Tumor Classification, Medical Imaging.

Downloads: | DOI: 10.17148/IJARCCE.2025.14802