REVIEW PAPER ON DISEASE PREDICTION AND DOCTOR CONNECTION AI SYSTEM

This work presents a practical, low-cost health system that turns a simple symptom entry into a full prediction and doctor chat flow. Using Flask for the web server, scikit-learn for the ML model, MySQL for storing profiles and records, and Razorpay for secure payment, the framework runs on any phone browser with modular Python code and separate components like Prediction Engine and Chat LinK Generator. This design makes it easy to maintain and extend to other areas or features, helping people in rural spots or busy cities check symptoms fast without internet for the main parts. This system demonstrates how intelligent technologies can support early diagnosis and connect patients with healthcare professionals, contributing to faster and more effective healthcare delivery. Future enhancements may include improved prediction accuracy, telemedicine features, and integration with wearable health devices for real-time monitoring.  The integration of AI-based prediction, doctor recommendation, secure payment, and patient record management improves healthcare accessibility, efficiency, and user convenience.  The system enables quick disease prediction based on user symptoms, helping patients gain preliminary health insights within seconds. By recommending suitable doctors and facilitating direct consultation through a single platform, it reduces the time and effort required to access medical assistance.

[1]       A. Sharma et al., "Symptom Checker using Naive Bayes on Indian Patient Data," IEEE India Conference (INDICON), pp. 210–215, 2023. [2]       S. Kumar and P. Reddy, "Teleconsultation via WhatsApp in Rural Tamil Nadu," Journal of Rural Health, vol. 40, no. 2, pp. 112–119,2024. [3]       R. Rao et al., "Razorpay Integration for MicroPayments in Healthcare Apps," Journal of Startup Engineering, vol. 5, no. 1, pp. 45–52, 2023. [4]       N. Gupta and A. Mehta, "Disease Prediction Web App using Flask and Scikit-learn," B.Tech Final Year Project Report, NIT Surathkal, 2023. [5]    World Health Organization, "Home Remedies for Common Illnesses," WHO Guidelines, 2022.[Online]. Available: https://www.who.int/publications [6]    J. Semigrad et al., "Confidence Scoring in Medical ML Models," Nature Medicine, vol. 27, pp. 1456–1463, 2021. [7]       Scikit-learn Developers, "Multinomial Naive Bayes Documentation," 2023. [Online]. Available: https://scikitlearn.org/stable/modules/naive_b ayes.html [8]       Flask Documentation, "Flask Web Development," Pallets Projects, 2024. [Online]. Available: https://flask.palletsprojects.com [9]       Razorpay Documentation, "Payment Gateway Integration  inPython,"  2024.   [Online]. Available: https://razorpay.com/docs [10]    Kaggle,     "Disease     Symptom     Prediction Dataset," 2022.[Online]. Available: https://www.kaggle.com/datasets/itachi9604/ diseasesymptom-prediction