Communications of the ACM
'Blood Pressure Video Selfies' Could Be Used to Monitor Patients
By tracking the intensity and movement of light reflected from blood vessels in the face, researchers at the University of Toronto used artificial intelligence to determine an individual's blood pressure with 96% accuracy
Credit: Kang Lee/American Heart Association
A study by scientists at the University of Toronto in Canada found two-minute-long smartphone videos of patients' faces can be used to take blood-pressure readings of those individuals that are 96% accurate.
The iPhone technique was tested on 1,328 patients with normal blood pressure.
The technique detects light reflected from subdermal facial blood vessels to measure blood flow.
Smartphone cameras are sufficiently sensitive to perceive the color of this reflected light, and software can track blood flow around the face.
The British Heart Foundation's Metin Avkiran said, "This innovative research is a powerful example of how data from mobile technology can be harnessed and analyzed by modern machine learning methods to potentially transform cardiovascular care."
From The Daily Mail (U.K.)
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA
No entries found