A team of medical researchers and engineers at Google Research has developed a way to use a smartphone's front-facing camera to monitor heart rate. The advantage of the development is that the system does not require additional hardware and therefore could become accessible to billions of people around the world.
A team of medical researchers and engineers at Google Research has developed a way to use a smartphone's front-facing camera to monitor heart rate. The advantage of the development is that the system does not require additional hardware and therefore could become accessible to billions of people around the world.
Over the past few years, manufacturers of personal medical devices have developed external heart monitors that can be worn, for example, in the form of a necklace or a smartwatch. But these devices are expensive. Researchers have found a cheaper alternative — a deep learning system that analyzes video from a smartphone's front-facing camera. It's called PHRM, writes Tech Xplore.
The idea for the system comes from the work of John Poel, who discovered back in the 1940s that skin changes color slightly with each pulse of blood flow. He went further, inventing a device that could monitor heart rate using light emitted and reflected from the skin, thus inventing a new science: photoplethysmography.
Many devices are based on Paul's work. A team at Google Research has also used it, but in a new way: by analyzing the reflected light when a smartphone is pointed at a person's face. It measures changes in blood volume, which can be used to monitor normal heart rate (HR) and resting heart rate.
To ensure that the system works for people of all skin colors, the researchers tested and fine-tuned its parameters to account for any skin color.
They also conducted extensive testing of the system using 495 volunteers who provided 225,773 videos of their faces. This was followed by additional testing on 185,970 videos provided by 205 individuals under various lighting conditions. The results were compared to those obtained using traditional devices such as electrocardiographs.
You can read more about the PHRM system on the arXiv scientific publication server.
