Privacy Preserving AI Tool for Early Detection of Endometriosis

Endometriosis is a common yet highly debilitating condition often plagued by a significant diagnostic delay, severely impacting patient well-being. Addressing this gap, this paper presents Endo-Insight-Buddy, a novel, privacy-first AI-powered web application designed to provide users with an instant, personalized assessment of their potential risk for endometriosis. Unlike clinical diagnostic tools, Endo-Insight-Buddy functions as an interactive and educational self-assessment platform that leverages a machine learning model to evaluate user input against known clinical, symptomatic, and demographic risk factors. The application is built using a modern stack (Vite, TypeScript, React, Tailwind CSS) and, crucially, operates under a zero data retention privacy model, ensuring all user-provided health data (age, symptom severity, medical history, optional biomarkers) is analysed entirely within the user's browser. The system culminates in visualized risk percentage results, highlighting key contributing factors, and providing evidence-based recommendations for seeking professional medical advice and self-care. By empowering individuals with actionable, statistically derived risk insights in a completely private environment, Endo-Insight-Buddy aims to reduce the time to diagnosis by encouraging earlier specialist consultation, ultimately strengthening patient autonomy and improving long-term health outcomes.