A Machine Learning-Based Web Application for Real-Time Patient Risk Level Monitoring

In today’s healthcare environment, continuous monitoring of patient vitals is crucial for preventing medical emergencies and improving treatment outcomes. However, due to the increasing patient load and shortage of medical staff, timely identification of high-risk patients remains a major challenge. This project introduces a Machine Learning–based web application that predicts patient risk levels in real time using vital parameters such as temperature, heart rate, blood pressure, oxygen saturation (SpO₂), and blood sugar. The proposed system employs the Random Forest algorithm implemented using Scikit-learn, ensuring accurate and efficient classification of patient conditions into four categories — Normal, Medium, Severe, and Critical. The web application integrates a FastAPI backend for model deployment and a frontend interface built with HTML, CSS, and JavaScript, supported by visualization tools like Chart.js and Plotly for dynamic data representation. Testing with sample healthcare datasets demonstrated an accuracy above 90%, proving the reliability of the model. The system allows healthcare professionals to monitor patients in real time, receive visual alerts, and prioritize treatment efficiently. This intelligent solution transforms patient monitoring into a proactive, data-driven process that enhances safety, responsiveness, and decision-making in clinical environments.