Heart rate estimation using a webcam

Medicine and telecommunications have been working back to back since long ago for granting assistance to people who is in a remote place, by means of telephone or radio communications in the beginnings, or videoconference nowadays. In the last years, there is a new field of application for telemedicine: those patients who need to be frequently supervised by their doctors. The highly frequent visits to the doctor have an impact in Social Security costs, and also in the routine of the patient, whose daily routine is largely conditioned by these visits. For these reasons, one of the objectives of modern telemedicine is to increase the freedom and convenience for the patients, by allowing them to carry out these periodic consultations at home.

One of the ways of increasing the accessibility to these services is by means of tools that can make the process as transparent as possible for the user. One novel example of this is heart rate monitoring via webcam. Each heart beat produces a variation in the amount of blood present in the face skin, causing small illumination changes. These fluctuations cannot be noticed by the human eye, but they can be detected by means of a standard webcam and appropriate image and signal processing techniques, allowing an accurate estimation of the heart rate in a seamless manner for the user.

Following the steps given by the researchers of the Massachusetts Institute of Technology (who applied this technique for the first time) and Philips (which released an IPad2 app in 2011), in Gradiant we are working in the analysis and improvement of this technology towards its future application in other projects which are currently active in the centre, like the “Smart Mirror”, telemonitoring, automatic emotion analysis, or even to detect impostors in face recognition systems.