Abstract
The healthcare sector in India faces a critical challenge of accessibility, with a doctor-to-patient ratio of approximately 1:1456, significantly below the World Health Organization’s recommended standard. This paper presents Arogya Mitra, an AI-powered mobile health assistant application designed to bridge the healthcare information gap through intelligent conversational interfaces. Built using React Native with Expo framework for the frontend, Node.js/Express.js for the backend, MongoDB Atlas for data persistence, and Google Gemini API for natural language understanding and response generation, Arogya Mitra provides a comprehensive suite of health management features including AI-driven health consultations, symptom assessment, location-based doctor discovery via Google Maps, prescription digitization through camera-based scanning with ML Kit barcode recognition, and medication adherence support through scheduled push notifications. The system architecture follows a three-tier client-server model with RESTful API communication, JWT-based authentication, and Firebase Cloud Messaging for notification delivery. The application was evaluated through multi-level testing including unit, integration, system, performance, and security testing, followed by User Acceptance Testing (UAT) with 30 participants from diverse demographics. UAT results demonstrated an overall satisfaction score of 4.17 out of 5.00 (83.4%), with the AI chat quality and doctor finder features receiving the highest ratings. Performance benchmarks showed an average Gemini API response time of 2.1 seconds, cold start time of 2.8 seconds, and API success rate of 98.7%. The paper discusses the system’s architecture, implementation methodology, testing outcomes, limitations, and future enhancements including multi-language support and telemedicine integration.