EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Cross-Modality Transfer Learning from PSG to FMCW Radar for Event-Level Apnea–Hypopnea Segmentation

Saihu Lu, Peng Wang iD, Zhenfeng Li, Pang Wu, Xianxiang Chen iD, Lidong Du iD, Libin Jiang, Zhen Fang

Bioengineering · 27.02.2026 · 13 (3) · 283

DOI: 10.3390/bioengineering13030283 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 27.02.2026. Verfasst von Saihu Lu und 7 weiteren Autor:innen. Beteiligte Einrichtung: Aerospace Information Research Institute, Chinese Academy of Sciences (AIRCAS), Beijing 100094, China. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Beijing Nova Programme Interdisciplinary Cooperation Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sleep apnea–hypopnea syndrome (SAHS) is a common sleep-related breathing disorder associated with substantial cardiovascular and neurocognitive risks. Although polysomnography (PSG) remains the clinical gold standard for diagnosis, its cost, operational burden, and limited accessibility hinder scalable and longitudinal home monitoring. Frequency-modulated continuous-wave (FMCW) radar provides unobtrusive, non-contact respiration sensing, yet radar-based event detection is often constrained by scarce annotations and pronounced domain shifts relative to PSG signals. In this work, we propose a deep learning framework for apnea–hypopnea event detection from FMCW radar that combines a 1D U-Net segmentation backbone with multi-head self-attention (MHSA) and cross-modality transfer learning. The model was first pre-trained on a large public PSG dataset to learn transferable respiratory-event representations and then fine-tuned on a smaller clinically annotated radar respiration dataset using synchronized PSG labels. It produced per-sample event probabilities, which were further refined via temporal post-processing to generate event-level detections and apnea–hypopnea index (AHI) estimates. Experimental results demonstrate strong performance in the radar domain, achieving precision of 0.8137±0.0332, recall of 0.8369±0.0470, and an F1-score of 0.8167±0.0052. Overall, these results indicate that PSG-to-radar transfer learning enables accurate, low-cost, and non-contact sleep apnea screening, supporting scalable longitudinal monitoring in home-like settings.

Schlagwörter

Stichwörter der Autor:innen

AHI estimationFMCW radarapnea–hypopneahealth caretransfer learning

Autor:innen und Institutionen

  • Aerospace Information Research Institute, Chinese Academy of Sciences (AIRCAS), Beijing 100094, China.
  • School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China.
  • Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing 100730, China.

ORCID-Profile: Peng Wang, Xianxiang Chen, Lidong Du

Förderung

  • National Natural Science Foundation of China
  • Beijing Nova Programme Interdisciplinary Cooperation Project

Zitieren

APA

Lu, S., Wang, P., Li, Z., Wu, P., Chen, X., Du, L., Jiang, L., & Fang, Z. (2026). Cross-Modality Transfer Learning from PSG to FMCW Radar for Event-Level Apnea–Hypopnea Segmentation. Bioengineering, 13(3), 283. https://doi.org/10.3390/bioengineering13030283

Vancouver

Lu S, Wang P, Li Z, Wu P, Chen X, Du L, et al. Cross-Modality Transfer Learning from PSG to FMCW Radar for Event-Level Apnea–Hypopnea Segmentation. Bioengineering. 2026;13(3):283. doi:10.3390/bioengineering13030283

BibTeX
@article{Lu2026,
  author = {Lu, Saihu and Wang, Peng and Li, Zhenfeng and Wu, Pang and Chen, Xianxiang and Du, Lidong and Jiang, Libin and Fang, Zhen},
  title = {Cross-Modality Transfer Learning from PSG to FMCW Radar for Event-Level Apnea–Hypopnea Segmentation},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {3},
  pages = {283},
  doi = {10.3390/bioengineering13030283}
}
Kostenlos starten →