EUVIMEDEUVIMED entdecken
Crossref

Non-contact acoustic screening for sleep apnea: a subject-aware deep learning approach

Melike Aygün Çakıroğlu iD, Emel Kızılkaya Aydoğan, Ömer Faruk Bolattürk, Serhat Aydoğan, Sevda İsmailoğulları, Yılmaz Delice

Sleep and Breathing · 2026-03 · 30 (1)

DOI: 10.1007/s11325-026-03594-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sleep and Breathing, online erschienen am 11.02.2026. Verfasst von Melike Aygün Çakıroğlu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Distance Education, Research and Application Centre, Abdullah Gül University, Kayseri, Turkey. melikeaygun@gmail.com. Hauptthemen laut MEDLINE: Polysomnography, Deep Learning, Sleep Apnea, Obstructive, Mass Screening. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 11 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: To explore the feasibility of using camera-derived, non-contact audio synchronized with PSG for clinically relevant sleep-apnea classification, and to benchmark compact deep models under a subject-aware design using a previously unstudied, real-world dataset. METHODS: Thirty-two adults underwent simultaneous polysomnography (PSG) and camera-based non-contact audio recording. The synchronized audio segments were used to train and compare three compact deep-learning architectures (convolutional, attention-augmented, and transformer-based) under a subject-aware evaluation design that prevented identity leakage. Model performance and calibration were assessed at both segment and subject levels using standard statistical tests. RESULTS: Subject-level evaluation was based on a very small, imbalanced test set of six subjects (one positive). Within this limited yet previously unstudied local dataset, the CNN_trans model achieved an apparent perfect ranking performance (AUC = 1.00; 95% CI 0.00–1.00), though this likely reflects the small, imbalanced test cohort, with recall = 1.00 and precision = 0.55. The wide confidence interval reflects substantial statistical uncertainty, and DeLong comparisons showed no significant AUC difference between CNN_trans and CNN_att (ΔAUC = − 0.042; p = 0.43). CONCLUSION: PSG-synchronized, non-contact audio supports accurate and well-calibrated sleep-apnea classification with compact deep models. This subject-aware evaluation suggests that contactless acoustic monitoring may have potential clinical relevance, motivating larger, multi-site validation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PolysomnographyDeep LearningSleep Apnea, ObstructiveMass ScreeningHumansFemaleAdultMaleFeasibility StudiesMiddle Aged

Stichwörter der Autor:innen

Contactless (non-contact) audioDeep learningModel calibration and explainabilityPolysomnographySleep apnea

Autor:innen und Institutionen

  • Distance Education, Research and Application Centre, Abdullah Gül University, Kayseri, Turkey. melikeaygun@gmail.com.
  • Industrial Engineering Department, Engineering Faculty, Erciyes University, Kayseri, Turkey.
  • Neurology Department, Medicine Faculty, Hatay Mustafa Kemal University, Hatay, Turkey.
  • Anesthesiology, and Reanimation Clinic, T.R. Ministry of Health Kayseri City Education, and Research Hospital, Kayseri, Turkey.
  • Department of Neurology, Erciyes University Medical School, Kayseri, Turkey.
  • Industrial Engineering, Faculty of Engineering, Architecture and Design, Kayseri University, Kayseri, Turkey.

ORCID-Profile: Melike Aygün Çakıroğlu

Zitieren

APA

Aygün Çakıroğlu, M., Kızılkaya Aydoğan, E., Bolattürk, Ö. F., Aydoğan, S., İsmailoğulları, S., & Delice, Y. (2026). Non-contact acoustic screening for sleep apnea: a subject-aware deep learning approach. Sleep and Breathing, 30(1). https://doi.org/10.1007/s11325-026-03594-2

Vancouver

Aygün Çakıroğlu M, Kızılkaya Aydoğan E, Bolattürk ÖF, Aydoğan S, İsmailoğulları S, Delice Y. Non-contact acoustic screening for sleep apnea: a subject-aware deep learning approach. Sleep and Breathing. 2026;30(1). doi:10.1007/s11325-026-03594-2

BibTeX
@article{Aygnakrolu2026,
  author = {Aygün Çakıroğlu, Melike and Kızılkaya Aydoğan, Emel and Bolattürk, Ömer Faruk and Aydoğan, Serhat and İsmailoğulları, Sevda and Delice, Yılmaz},
  title = {Non-contact acoustic screening for sleep apnea: a subject-aware deep learning approach},
  journal = {Sleep and Breathing},
  year = {2026},
  volume = {30},
  number = {1},
  doi = {10.1007/s11325-026-03594-2}
}
Kostenlos starten →