Non-Invasive Heart Rate Prediction Using Phase Video Analysis

Monitoring heart rate accurately is crucial for diagnosing and managing cardiovascular diseases. Conventional methods usually require direct skin contact, which can be uncomfortable or impractical in some cases. This project investigates a non-invasive technique for predicting heart rate through phase video analysis, utilizing advancements in computer vision and signal processing. The objective of this study to develop a non-invasive heart rate prediction system using phase video analysis. By leveraging computer vision techniques and machine learrning algorithms, the system captures facial videos to detect subtle changes in the forehead region caused by blood flow. This data is then processed to estimate the user’s heart rate in real-time. The study focuses on addressing challenges such as noise reduction, real-time processing, and accurate heart rate estimation under varying lighting and user movements. The results demonstrate the system’s robustness and accuracy, highlighting its potential applications in healthcare, fitness monitoring, and human-computer interaction.