Innovations in Stroke Identification A Machine Learning-based Diagnostic Model using Neuroimages

The proposed Stroke Classification System is an AI-driven diagnostic support tool developed to assist clinicians in the timely detection of stroke using brain MRI images. The framework adopts a hybrid architecture that combines the deep feature extraction capability of VGG16 with the robust classification performance of XGBoost to accurately distinguish between stroke and non-stroke cases. Experimental evaluation demonstrates a precision of 97.5% under controlled conditions, highlighting its reliability for clinical assistance.To ensure practical usability, the system is deployed as a web application using Flask, enabling healthcare professionals to upload MRI scans and receive near real-time diagnostic predictions. The platform supports rapid inference, maintains diagnostic history, and is designed for scalable deployment on cloud services such as AWS and Heroku. By integrating accuracy, speed, and accessibility, the system serves as an efficient and user-friendly AI solution for stroke diagnosis support.