Telehealth and distributed clinical trials are quickly becoming the norm for doctor-patient interaction and clinical research. Heart rate and variability are important measures of health and fitness. Gauging heart rate and other quantitative markers of health is challenging in an increasingly virtual world. Wearable technology offers continuous and accurate measures but is currently only utilized by 21% of adults in the U.S. In contrast, computers and smartphones are available in 92% and 84% of households. With emerging solutions from the video analytics space, there is potential to better equip physicians and clinical trial researchers, while reducing reliance on wearable technology.

We have developed and integrated a scalable video-based heart rate detection algorithm into our Vivo platform based on recent developments in computer vision. The general approach captures facial landmarks to isolate RGB channel values from specific facial regions. From these regions, we apply signal processing techniques to tease out a fluctuation pattern similar to that of an ECG. Finally, using peak detection and error correction algorithms, we are able to measure and monitor heart rate and heart rate variability.