HealthFit AI: An AI-Driven Web-Based Health and Fitness Monitoring System with Real-Time Risk Prediction and Analysis

Abstract

HealthFit AI is a comprehensive AI-driven health monitoring system that integrates real-time tracking, predictive analytics, and intelligent recommendations into a single web-based platform. The system uses a Random Forest classifier trained on biometric data to predict risks such as cardiovascular diseases, diabetes, and hypertension. Additionally, an LSTM model processes sequential sleep data to evaluate sleep quality. The platform also includes exercise form analysis, chatbot-based interaction, and goal tracking. The system achieved approximately 87% accuracy, demonstrating its effectiveness in predictive healthcare. This paper presents HealthFit AI, a comprehensive full-stack web-based platform that integrates multiple artificial intelligence models to provide continuous health monitoring, disease risk prediction, exercise form analysis, sleep quality assessment, and personalized recommendations. The system employs a Random Forest classifier trained on biometric data to predict cardiovascular, diabetes, hypertension, and obesity risk levels. A Long Short-Term Memory (LSTM) neural network processes sequential sleep data to compute sleep quality scores. Real-time health metrics are displayed through an interactive smart wrist watch interface that tracks eight key parameters: heart rate, blood pressure, steps, sleep, calories, oxygen level, water intake, and exercise duration. The backend is implemented using the Flask web framework with JWT authentication and a MySQL relational database, while the frontend is built with React, TypeScript, and Tailwind CSS. All AI components are served through a dedicated service with rule-based fallback logic, ensuring the system remains functional even when trained models are unavailable. The platform also features an intelligent Q&A chatbot for health-related queries, appointment scheduling with doctors, and comprehensive goal tracking with visual progress indicators.

Introduction

The text presents HealthFit AI, an intelligent healthcare system designed to address the growing global burden of non-communicable diseases (NCDs) such as diabetes, cardiovascular diseases, and obesity. While existing health apps like Google Fit and Apple Health mainly focus on tracking data, they lack predictive and decision-making capabilities. HealthFit AI overcomes this limitation by integrating real-time monitoring with AI-based predictions and personalized health recommendations.

The system combines multiple features, including real-time tracking of health metrics via a smart wearable interface, Random Forest-based risk classification, LSTM-based sleep analysis, MediaPipe-based exercise posture detection, an AI chatbot for health queries, goal tracking, and telemedicine appointment scheduling. It uses a three-tier architecture consisting of a React frontend, Flask backend, independent AI services, and a MySQL database for efficient data management.

The methodology involves training machine learning models on biometric and lifestyle data to classify users into risk categories and analyze sleep quality. The system also calculates BMI, monitors eight key health metrics, and provides rule-based chatbot responses for health guidance.

Results show that the system achieves around 87% prediction accuracy, low latency (<200 ms), and real-time updates every 5 seconds, making it suitable for continuous health monitoring. The user interface is designed for clarity and ease of use, supporting dashboards, wearable integration, appointment booking, and AI-driven insights.

Overall, HealthFit AI provides a comprehensive, predictive, and user-friendly digital health platform that supports preventive healthcare and improves lifestyle management through artificial intelligence.

Conclusion

This paper presented HealthFit AI, an open-source web-based platform that integrates a Random Forest health risk classifier, an LSTM sleep quality analyzer, a smart wrist watch interface, an AI chatbot, and real-time health metric tracking into a unified system. The platform addresses the fragmentation of existing consumer health tools by combining clinical-grade monitoring, AI inference, telemedicine scheduling, and fitness coaching in a single deployable application. The Random Forest classifier achieved 87.3% cross-validation accuracy on synthetic biometric data across four risk classes. The smart watch interface tracks eight health metrics with automatic updates every 5 seconds. The AI chatbot provides intelligent responses to health-related queries. The modular microservice architecture ensures that any AI component can be independently upgraded without service disruption. HealthFit AI demonstrates the technical feasibility of delivering a comprehensive AI-powered health monitoring platform as an open-source application. Future work will focus on clinical validation with real patient populations, wearable device integration, federated learning for privacy-preserving model improvement, and regulatory compliance for clinical deployment.

References

[1] World Health Organization, \"Noncommunicable diseases,\" WHO Fact Sheet, 2023. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases [2] A. L. Beam and I. S. Kohane, \"Big data and machine learning in health care,\" JAMA, vol. 319, no. 13, pp. 1317–1318, 2018. [3] E. J. Topol, \"High-performance medicine: the convergence of human and artificial intelligence,\" Nature Medicine, vol. 25, no. 1, pp. 44–56, 2019. [4] L. Breiman, \"Random forests,\" Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [5] S. Hochreiter and J. Schmidhuber, \"Long short-term memory,\" Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997. [6] Z. Cao, G. Hidalgo, T. Simon, S. Wei, and Y. Sheikh, \"OpenPose: Realtime multi-person 2D pose estimation using part affinity fields,\" IEEE Trans. Pattern Anal. Mach. Intell., vol. 43, no. 1, pp. 172–186, 2021. [7] C. Lugaresi et al., \"MediaPipe: A framework for building perception pipelines,\" arXiv preprint arXiv:1906.08172, 2019. [8] A. Supratak, H. Dong, C. Wu, and Y. Guo, \"DeepSleepNet: A model for automatic sleep stage scoring based on raw single-channel EEG,\" IEEE Trans. Neural Syst. Rehabil. Eng., vol. 25, no. 11, pp. 1998–2008, 2017. [9] P. Rajpurkar, E. Chen, O. Banerjee, and E. J. Topol, \"AI in health and medicine,\" Nature Medicine, vol. 28, pp. 31–38, 2022. [10] TensorFlow Documentation, \"Machine Learning Framework by Google,\" 2023. [Online]. Available: https://www.tensorflow.org [11] Scikit-learn Documentation, \"Machine Learning in Python,\" 2023. [Online]. Available: https://scikit-learn.org [12] React Documentation, \"Frontend JavaScript Library,\" 2023. [Online]. Available: https://react.dev [13] Flask Documentation, \"Python Web Framework,\" 2023. [Online]. Available: https://flask.palletsprojects.com [14] Tailwind CSS Documentation, \"Utility-First CSS Framework,\" 2023. [Online]. Available: https://tailwindcss.com [15] MySQL Documentation, \"Relational Database Management System,\" 2023. [Online]. Available: https://www.mysql.com

Copyright

Copyright © 2026 Monish Fodse , Devank Lohar, Raj More , Shubham Naikade, Vijyalaxmi Tadkal. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.