HealthDay News — Transdermal optical imaging can be used to predict systolic, diastolic, and pulse pressure from facial blood flow data, according to a study published online Aug. 6 in Circulation: Cardiovascular Imaging.

Hong Luo, M.D., Ph.D., from The Affiliated Hospital of Hangzhou Normal University in China, and colleagues used an advanced machine learning algorithm to create computational models that predicted reference systolic, diastolic, and pulse pressure from facial blood flow data for 1,328 normotensive adults. Seventy percent of the data set was used to train the models and 15 percent was used to test them. Model performance was validated using the remaining 15 percent of the sample.

The researchers found that the models could predict blood pressure with a measurement bias of 0.39 ± 7.30 mm Hg, −0.20 ± 6.00 mm Hg, and 0.52 ± 6.42 mm Hg for systolic pressure, diastolic pressure, and pulse pressure, respectively.

“This technology has shown great promise in this initial phase of the validation process, and we think that the likelihood of successful validation is high,” the authors write. “The realization of this technology would enable the creation of convenient and widely available blood pressure measurement tools in smartphones.”

Two authors hold a patent for Transdermal Optical Imaging Technology.

Abstract/Full Text (subscription or payment may be required)

Editorial (subscription or payment may be required)

Related Articles