RT Journal Article
SR Electronic
T1 VALIDITY OF A DIAGNOSTIC METHOD BY “NO CONTACT” TECHNOLOGY IN DETECTION OF SLEEP APNEA THROUGH A THERMOGRAPHIC CAMERA AND ARTIFICIAL INTELLIGENCE
JF European Respiratory Journal
JO Eur Respir J
FD European Respiratory Society
SP PA2482
DO 10.1183/13993003.congress-2021.PA2482
VO 58
IS suppl 65
A1 Rodriguez, Paula
A1 De Leceta, Aitor Moreno Fernández
A1 Martínez García, Alexeiv
A1 Pía Martínez, Carla
A1 Gómez, Salvador Delis
A1 Álvarez Ruiz De Larrinaga, Ainhoa
A1 Durán-Cantolla, Joaquin
YR 2021
UL http://erj.ersjournals.com/content/58/suppl_65/PA2482.abstract
AB Introduction: Diagnosis method of sleep apnea is polysomnography (PSG) and respiratory polygraphy (RP). Both systems are costly, time consuming and uncomfortable. The objective of our study was to validate an infrared thermographic camera with an artificial intelligence system, in order to identify apneic respiratory events in adults with clinical suspicion of OSA.Methods: Total number of respiratory events (apneas and hypopneas) detected by facial temperature micro-changes were analyzed and compared with events detected by nasal flow quantified by the PSG nasal cannula and thermistor. The result of both methods was classified by binary form as normal study if it presented an AHI≤10 or pathological if the AHI was>10.Results: 30 records were full valuable (age 48±11 years, 83% men, body mass index 27± 3.7 kg/m2, and Epworth of 9±4 points). 16 studies were normal on PSG and 14 were pathological. Thermographic camera was able to detect all normal studies (100% specificity) and 12 of the 14 pathological studies, classifying only 2 pathological studies as normal. Table I presents the validity results.Conclusions: Use of this system is highly valid, especially for diagnosing the presence of OSA (AHI >10). If these results are confirmed with a greater number of patients, the potential implantation of thermography treated with artificial intelligence could result in a new “no contact” diagnostic system for patients with suspected OSA. ParameterValueCI 95%Sensitivity85.7%67.4%-100%Specifity100%100%Predictive Value (+)100%100%Predective Value (-)88.9%74.4%-100%LR(+)Maximun-LR(-)0.140.04-0.52Accuracy93.3%84.4-100%FootnotesCite this article as: European Respiratory Journal 2021; 58: Suppl. 65, PA2482.This abstract was presented at the 2021 ERS International Congress, in session “Prediction of exacerbations in patients with COPD”.This is an ERS International Congress abstract. No full-text version is available. Further material to accompany this abstract may be available at www.ers-education.org (ERS member access only).