Analysis of the reliability of a mobile phone application in craniofacial measurement for obstructive sleep apnea

Obstructive sleep apnea (OSA) disrupts breathing due to upper airway obstruction during sleep. This study presents the development of a mobile application for 2D craniofacial measurements associated with OSA, aiming to provide an accessible approach.

This single-center, cross-sectional, retrospective analysis utilized data from 128 patients who underwent polysomnography and facial photography. Craniofacial measurements were obtained using the mobile app and standard desktop software (Studio 3^®). Measurement reliability and agreement were assessed using the Intraclass Correlation Coefficient (ICC) and Bland-Altman analysis, with F-tests applied to ICCs.

The application showed good reliability for key measurements, including profile angle (ICC = 0.86, p < 0.01), nasolabial angle (ICC = 0.84, p < 0.01), neck angle (ICC = 0.88, p < 0.01), facial height (ICC = 0.73, p < 0.01), and facial width (ICC = 0.75, p < 0.01), with significant F-tests. Bland-Altman analysis corroborated these results. However, upper lip length (ICC = 0.42, p > 0.05) and lower lip length (ICC = 0.54, p > 0.05) showed lower reliability and non-significant F-tests. The mentolabial sulcus (ICC = 0.51, p = 0.04) exhibited moderate reliability but higher variability.

This mobile application shows promise as a reliable, accessible tool for OSA-related craniofacial assessment, especially useful in resource-limited settings. ICC and F-test values for primary measurements suggest the app's consistency with traditional methods, though the reduced reliability of lip length and mentolabial sulcus measurements may limit their clinical utility in individual assessments. Therefore, further refinements and validation in diverse populations are recommended.