We have all sorts of sensors that we can embed in wearable Health Tech devices that can measure all sorts of important biometric data, including standard patient vital signs such as heart rate, heart rate variability, blood pressure, and blood oxygen levels. However, there are many situations where it would be helpful to be able to gather this data without having to strap some device onto the patient. Wouldn’t it be great if a video camera could capture this information from a distance?
That’s exactly what researchers at Rice University have been working on. The basic principle is pretty simple; as your heart beats, the pressure increases in your blood vessels, including the capillaries near the skin’s surface. The increased pressure expands the vessels, filling them with more blood. This change in blood volume results in very subtle changes in skin color, as the blood has a different color than the surrounding tissue. Using a camera to Monitor vital signs based on these changes has the unwieldy name of photoplethysmography, which is why it is often referred to simply as PPG.
The problem with most PPG systems is that they do not work well with patients who have dark colored skin, or in low-light level conditions, or when the patient’s face is moving (such as when speaking). The Rice researchers have developed a new algorithm to make PPG far more accurate under more circumstances. Using face tracking, they divide the face image into different regions and then use data from multiple regions in a weighted average computation. This technique results in far more accurate readings across a variety of skin colors and lighting situations.
This development has far-reaching implications. It could result in low-cost accurate vital sign monitoring, which could be used by consumers with smartphones or as a means of providing better healthcare services in under-developed countries around the world.