At the University of South Australia, researchers designed a system that allows them to measure a patient’s blood pressure with a camera. The camera visualizes the patient’s forehead and focuses on two regions in particular to optically determine photoplethysmographic signals that AI algorithms then convert to blood pressure data. The researchers tested their system in 25 volunteers, and so far it has proven to be approximately 90% as accurate as traditional pressure cuff measurements. The system could provide contact-free measurements in as little as ten seconds, and could be very useful for large health screens or in situations where minimal contact is preferred for safety, such as during a pandemic.
Blood pressure is a key health indicator, but our method of measuring it can be a little cumbersome. Inflatable cuffs can be uncomfortable and a bit of a faff to use. If we could develop a non-invasive and rapid alternative, it could save trouble all around, particularly in cases where reduced contact between patients and healthcare staff is desirable, such as during the COVID-19 pandemic.
“Monitoring blood pressure is essential to detect and manage cardiovascular diseases, the leading cause of global mortality, responsible for almost 18 million deaths in 2019,” said Javaan Chahl, a researcher involved in the study. “Furthermore, in the past 30 years, the number of adults with hypertension has risen from 650 million to 1.28 billion worldwide. The health sector needs a system that can accurately measure blood pressure and assess cardiovascular risks when physical contact with patients is unsafe or difficult, such as during the recent COVID outbreak. If we can perfect this technique, it will help manage one of the most serious health challenges facing the world today,”
The team behind the technology has developed similar systems in the past, including image processing algorithms that can determine someone’s heart rate from drone video footage, and other algorithms that can measure a variety of important parameters in newborns, including breathing rate, jaundice, oxygen saturation, and temperature.
This latest technology has been tested in 25 volunteers with different skin tones and in different ambient light conditions. The researchers compared the readings with those obtained using a commercial sphygmomanometer and found that the contact-free camera-based technology was approximately 90% accurate, suggesting that is has significant promise as a diagnostic technology.
Study in journal Inventions: Contactless Blood Pressure Estimation System Using a Computer Vision System