OSA diagnosis by respiratory movement analysis using a conventional smartphone 3D camera

<bold>Introduction:</bold> Sleepwise (SW) is a noninvasive technology that analyses patients’ video-recorded respiratory movements to transform them into a breathing signal that determines episodes of hypopnea and apnea. While SW first version processed and analysed images from an infrared external camera, the latest version uses a conventional smartphone 3D camera. <bold>Aims and objectives:</bold> The aim of this work is to compare the concordance of SW updated version in the diagnosis of OSA and its severity with a simultaneous in-laboratory polysomnography. <bold>Methods:</bold> This is a pilot diagnostic accuracy study that included 20 consecutive adults referred to the Sleep Unit of the Hospital Universitari Germans Trias i Pujol (HGTiP) for an in-laboratory polysomnography from July 2023 to January 2024. SW analysis is based on a beta version of a smartphone app installed in an iPhone 13 (Apple, Inc., Cupertino, California, USA) placed 50 cm beside the subject. For this study, we took SW automatic analysis but a manual version for analyses is available. <bold>Results:</bold> In-laboratory polysomnography and SW had a Lin’s concordance correlation coefficient of 0.884 for the AHI and a κ of 0.829 (95% CI 0.71-0.94) for the severity of OSA. The time of sleep had a Lin’s concordance correlation coefficient of 0.324. <bold>Conclusions:</bold> The latest version of SW, using video analysis of respiratory movements recorded with a conventional smartphone 3D camera, showed good correlation with simultaneous in-laboratory polysomnography. <table-wrap><object-id>erj;64/suppl_68/PA4463/T1</object-id><object-id>T1</object-id><object-id>t1</object-id><table><colgroup><col></col><col></col><col></col></colgroup><tbody><tr><td></td><td>AHI</td><td>OSA severity</td></tr><tr><td></td><td>CCC</td><td>kappa 95% CI</td></tr><tr><td>PSG-SW</td><td>0.884</td><td>0.829 (0.71-0.94)</td></tr></tbody></table></table-wrap>