Real-Time HRV Monitoring using Pupil Measurements Via FastAPI-YOLO Detection

The study presents an innovative non-contact technique for assessing heart rate variability (HRV) by examining pupil dynamics. It taps into the fascinating link between pupil dilation and heart rate, which is controlled by the autonomic nervous system. To make this work, researchers developed a deep learning model that predicts HRV based on various pupil metrics, including width, height, and the average size of both pupils. The implementation involved using FastAPI to deploy the model by setting up OTP-based email authentication for storing data in a database, and employing YOLO and CNN for extracting pupil dimensions from images. This method offers a scalable and reliable tool for precise pupil-based HRV analysis. The experimental results were impressive, showing a 93.15% accuracy in correlating pupil movements with HRV parameters, with R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> values of 0.7644 for MEAN_RR and 0.7627 for heart rate (HR).