Researchers at the University of Western
Australia have demonstrated that facial features captured from 3D photographs
may be useful as a screening tool for sleep apnea. 400 individuals participated
in the study, and the researchers found that they could predict which patients
would have sleep apnea with up to 91% accuracy.
300 individuals with sleep apnea and 100 individuals without sleep apnea were enrolled in the study, which involved overnight sleep studies and 3D photography with a craniofacial scanner. The 3D photos were analyzed to resolve 24 anatomical landmarks and linear measurements, such as the shortest distance between two points, and geodesic measurements including the shortest distance between two points on a curved surface, were calculated.
The researchers found that geodesic measurements of
craniofacial anatomy resulted in higher accuracy than linear measurements; 89%
with geodesic measurements, compared to 86% with linear measurements. When
linear and geodesic measurements were combined into one predictive unit, the researchers
found 91% accuracy.
This data shows there may be promise in using 3D facial features as a screening tool for sleep apnea in the future using inexpensive cameras, potentially even those already inside many existing smartphones.
“Obstructive Sleep Apnea (OSA) is a huge public health problem, and despite effective treatments being available, many with OSA are currently undiagnosed,” said Dr. Peter Eastwood, lead investigator of the study. “Therefore, simple, accurate screening tools are needed to predict those who have OSA.”
The paper in the Journal of Clinical Sleep Medicine: Predicting sleep apnea from three-dimensional face photography