AI-Driven Brain Tumor Detection and Segmentation Using Computer Vision: A Solution for Accessible Healthcare in Sub-Saharan Africa

Recent strides in artificial intelligence (AI) and deep learning techniques have propelled the development of an AI-powered brain tumour detection model. This study Uses powerful YOLO(You Only Look Once)Algorithm, neural network optimisation, and image preprocessing to craft a robust AI model capable of accurately detecting and segmenting diverse brain tumour types and normal cases. In this study, we aim to classify brain tumors such as glioma, meningioma, and pituitary tumor from a brain comprehensive dataset of 3000 MRI (Magnetic Resonance Imaging) scans, the model achieves an accuracy of 92% on detection and 96% on segmentation. Its integration into a user-friendly Web app, BScan, enhances accessibility and practicality. The app provides detection and area segmentation of where the tumor is located to support medical professionals in making supporting decisions with a specific focus on healthcare challenges in Sub-saharan Africa. The model prioritizes interpretability enhancement and has the potential to cultivate collaboration between AI experts and medical practitioners, thus advancing brain tumor detection and diagnosis. While promising, the model demands computational resources and diverse datasets. This research also highlights AI’s potential to transform healthcare diagnostics, ensuring precise and efficient brain tumor detection.