A new optical imaging technique usable on smartphones was able to come close to reference measurements of cuff-based measurement, according to a new study published in Circulation: Cardiovascular Imaging.
“High blood pressure is a major contributor to cardiovascular disease, and a leading cause of death and disability,” lead author Kang Lee, PhD, professor and research chair in developmental neuroscience at the University of Toronto, said in a press release. “To manage and prevent it, regular monitoring of one’s blood pressure is essential.”
Dr. Kang added that traditional cuff-based measuring devices, while highly accurate, are inconvenient and uncomfortable, and that users tend to disregard American Heart Association guidelines and manufacturers’ suggestions to take multiple measurements.
To this end, the current study included 1,328 Canadian and Chinese normotensive adults. The process consisted of capturing two-minute videos of participants’ faces in a well-lit environment using the software installed on iPhones. The software an advanced machine learning algorithm for the creation of predictive computational models for reference systolic, diastolic, and pulse pressure taken from facial blood flow data. The researchers reported using 70% of their data set for the training of those models and 15% for model testing, with the remaining 15% used for model validation.
The models predicted blood pressure measurement within a bias (plus or minus standard deviation) of 0.39±7.30 mm Hg for systolic blood pressure, -0.20±6.00 mm Hg for diastolic blood pressure, and 0.52±6.42 mm Hg for pulse pressure.
“If future studies confirm our results and show this method can be used to measure blood pressures that are clinically high or low, we will have the option of a contactless and non-invasive method to monitor blood pressures conveniently – perhaps anytime and anywhere – for health management purposes,” Dr. Lee said of the results.
Accompanying editorial author Ramakrishna Mukkamala, PhD, a professor in the Department of Electrical and Computer Engineering at Michigan State University in East Lansing, agreed that the results are exciting if reproducible.
“This study shows that facial video can contain some information about systolic blood pressure,” Dr. Mukkamala wrote “If future studies could confirm this exciting result in hypertensive patients and with video camera measurements made during daily life, then obtaining blood pressure information with a click of a camera may become reality.”