Predicting sleep apnea from 3-dimensional face photography. P Eastwood, SZ Gilani, et al.

Date: August 2019.
Source: Journal of clinical sleep medicine (JCSM) 16(4).
Objective: Craniofacial anatomy is recognised as an important predisposing factor in the pathogenesis of obstructive sleep apnea (OSA). Two-dimensional (2D) photography has shown that craniofacial features are related to the presence and severity of OSA. Three-dimensional (3D) has potential advantages over 2D imaging as it overcomes the limitation of representing a 3D structure (the face) in two dimensions. This study aimed to use 3D facial surface analysis of linear and geodesic (shortest line between points over a curved surface) distances to determine the combination of measurements that best predicts severity of OSA.
Methods: 400 adults underwent overnight polysomnography and 3D face photography. 100 did not have OSA (apnea-hypopnea index, AHI <5 events/hr), 100 had mild OSA (5≤AHI<15 events/hr), 100 had moderate OSA (15≤AHI<30 events/hr) and 100 had severe OSA (AHI≥ 30 events/hr). Measurements of linear distances and angles, and geodesic distances were obtained between anatomical landmarks from the 3D photographs and their relationship to the presence and severity of OSA determined.
Results: Relative to linear measurements, geodesic measurements of craniofacial anatomy improved the ability to identify individuals with and without OSA (classification accuracy 86% and 89% respectively, p<0.01). A maximum classification accuracy of 91% was achieved when linear and geodesic craniofacial measurements were combined into a single predictive algorithm.
Conclusions: This study showed that geodesic measurements add value to the capacity to identify patients with OSA from 3D photographs of the face. These preliminary results also suggest that it might also be possible to predict the severity of an individual’s OSA from such photographs.

Article: Predicting sleep apnea from 3-dimensional face photography.
Authors: Peter Eastwood, Syed Zulqarnain Gilani, Nigel McArdle, David Hillman, Jennifer Walsh, Kathleen Maddison, Mithran Goonewardene, Ajmal Mian. University of Western Australia and Sir Charles Gairdner Hospital, Western Australia.