Biomedical Decision Support and Symptom Analysis System

BDSAS set out to show that a thoughtfully engineered open-source healthcare AI tool — one that speaks plain language, runs on accessible technology, and respects basic security principles — can deliver genuinely useful triage support. The evidence from our testing suggests it does. Response times were consistently within acceptable bounds, the AI's clinical reasoning held up across all test scenarios, and — perhaps most tellingly — the general practitioner in our UAT session endorsed the PDF reports as suitable for clinical handover. The combination of structured prompting, schema validation, and dual-layer input sanitization proved to be a practical and effective way to constrain hallucination without sacrificing the flexibility that makes large language models useful. BDSAS will not replace a doctor. But in a world where millions of people lack timely access to one, a reliable, transparent, and freely deployable first-step tool has real value — and this work demonstrates that building one is well within reach for an academic team using modern open-source infrastructure. Furthermore, the project demonstrates the potential of integrating Artificial Intelligence with web technologies to improve healthcare accessibility and patient awareness. By providing instant symptom analysis, personalized recommendations, and downloadable medical reports, BDSAS empowers users to make informed decisions about their health while reducing the burden on healthcare professionals. The system's scalable architecture, user-friendly interface, and modular design make it suitable for future enhancements such as multilingual support, electronic health record integration, wearable device connectivity, and advanced predictive analytics. Overall, BDSAS serves as a strong foundation for the development of intelligent healthcare solutions and highlights the significant role that AI can play in supporting early diagnosis, improving healthcare delivery, and promoting preventive care in both urban and rural communities.

[1]     S. Sutton, D. Pincock, D. C. Baumgart, D. Sadowski, C. Fedorak, and A. Kroeker, "An overview of clinical decision support systems: benefits, risks, and strategies for success," npj Digital Medicine, vol. 3, no. 17, pp. 1–10, 2020. [2]     E. Shortliffe and J. Cimino, Biomedical Informatics: Computer Applications in Health Care and Biomedicine, 4th ed. London, UK: Springer, 2014. [3]     A. Khalfallah, M. Ben Yahia, and S. Jemni, "Decision Support Systems in Healthcare: A Systematic Review," Healthcare Analytics, vol. 3, pp. 100-115, 2023. [4]     World Health Organization (WHO), "Health workforce projections 2030," Geneva, Switzerland, 2023. [5]     National Health Profile India, Ministry of Health and Family Welfare, Government of India, "Doctor-Population Ratio Statistics," New Delhi, India, 2023. [6]     M. Khosravi, A. Rahmani, and S. Hosseinzadeh, "Artificial Intelligence and Clinical Decision-Making in Healthcare: A Systematic Review of Reviews," Artificial Intelligence in Medicine, vol. 148, pp. 102-118, 2024. [7]     Meta AI, "LLaMA 3.3: Open Foundation and Instruction Models," 2024. [8]     MongoDB, "MongoDB Atlas Documentation," 2025. [9]     Flask Documentation, Pallets Project, 2025. [10]  D. Jurafsky and J. H. Martin, Speech and Language Processing, 3rd ed., Pearson, 2024.