ABSTRACT
The advancement of web technologies and machine learning has revolutionized modern healthcare by enabling intelligent, real-time health monitoring systems. This project presents a smart health monitoring system using the Flask web framework, MySQL database, and machine learning models for predictive diagnostics. The system collects physiological parameters such as ECG, heart rate, SpO2, temperature, and sound from sensors, processes them through trained models, and provides health condition predictions. A secure user authentication mechanism using hashed passwords ensures user data privacy, and a role-based access system categorizes users into admin, doctor, and normal user types. Each user role has access to specific functionalities: admins manage users and view real-time data, doctors monitor patients and their prediction history, while users input or receive sensor data and view their own reports. The machine learning models are serialized using joblib and loaded into the Flask backend for real-time inference. Inputs are normalized using pre-fitted scalers and classified into categories such as normal, abnormal, or critical for each health parameter. The system includes a /realtime module that fetches the latest sensor reading from the database and performs automated prediction without manual input. Data sent via the /receive endpoint is automatically stored and used for detection, making it compatible with IoT sensor integration. The entire workflow, from data acquisition to prediction and storage, is seamless and automated. Results are stored in a structured MySQL table and displayed back to users through HTML templates rendered with Flask. Additionally, all user interactions and predictions are logged, allowing doctors to access patient history and assess trends. This application improves early detection of health risks, especially in remote or under-resourced environments. It reduces dependence on manual monitoring and supports proactive healthcare delivery. Its modular design allows for future enhancements such as integration with wearable devices, cloud storage, live alerts, and mobile applications. This smart health system offers a practical and scalable solution for remote patient monitoring using real-time data and machine learning.


OBJECTIVES
a. To develop a real-time health monitoring system that collects vital parameters such as ECG, heart rate, SpO₂, temperature, and ambient sound using IoT devices or manual input.
b. To implement machine learning models capable of accurately classifying the physiological data into diagnostic categories such as normal, abnormal, or critical, enabling early detection of health issues.
c. To build a secure, role-based web application using Flask, where users, doctors, and admins have differentiated access to functionalities such as data input, analysis, report viewing, and user management.
d. To enable real-time predictions by integrating an automatic /realtime route that fetches the latest sensor values and instantly provides health insights to the user and stores them in the database.
e. To ensure secure data handling and accessibility, using techniques such as password hashing, session management, and structured database storage to protect user data and maintain system integrity.

Block Diagram