Combining wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity

Shang-Yang Lin; Cheng-Yu Tsai; Arnab Majumdar; Yu-Hsuan Ho; Yu-Wen Huang; Chun-Kai Kao; Shang-Min Yeh; Wen-Hua Hsu; Yi-Chun Kuan; Kang-Yun Lee; Po-Hao Feng; Chien-Hua Tseng; Kuan-Yuan Chen; Jiunn-Horng Kang; Hsin-Chien Lee; Cheng-Jung Wu; Wen-Te Liu

doi:10.5664/jcsm.11136

Combining wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity

J Clin Sleep Med. 2024 Mar 28. doi: 10.5664/jcsm.11136. Online ahead of print.

Authors

Shang-Yang Lin¹, Cheng-Yu Tsai^{2

3}, Arnab Majumdar², Yu-Hsuan Ho⁴, Yu-Wen Huang⁴, Chun-Kai Kao⁵, Shang-Min Yeh⁴, Wen-Hua Hsu¹, Yi-Chun Kuan^{6

7

8

9}, Kang-Yun Lee³, Po-Hao Feng³, Chien-Hua Tseng³, Kuan-Yuan Chen³, Jiunn-Horng Kang^{10

11

12}, Hsin-Chien Lee¹³, Cheng-Jung Wu¹⁴, Wen-Te Liu^{1

3

6

10}

Affiliations

¹ School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei, Taiwan.
² Department of Civil and Environmental Engineering, Imperial College London, London, UK.
³ Division of Pulmonary Medicine, Department of Internal Medicine, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan.
⁴ Advanced Technology Lab, Wistron Corporation, Taipei, Taiwan.
⁵ Wireless Technology and Antenna Research and Development Department, Wistron Corporation, Taipei, Taiwan.
⁶ Sleep Center, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan.
⁷ Department of Neurology, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan.
⁸ Department of Neurology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan.
⁹ Taipei Neuroscience Institute, Taipei Medical University, Taipei, Taiwan.
¹⁰ Research Center of Artificial Intelligence in Medicine, Taipei Medical University, Taipei, Taiwan.
¹¹ Department of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei, Taiwan.
¹² Graduate Institute of Nanomedicine and Medical Engineering, College of Biomedical Engineering, Taipei Medical University, Taipei, Taiwan.
¹³ Department of Psychiatry, Taipei Medical University Hospital, Taipei, Taiwan.
¹⁴ Department of Otolaryngology, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan.

PMID: 38546033
DOI: 10.5664/jcsm.11136

Abstract

Study objectives: The gold standard for diagnosing obstructive sleep apnea (OSA) is polysomnography (PSG). However, PSG is a time-consuming method with clinical limitations. This study aimed to create a wireless radar framework to screen the likelihood of two levels of OSA severity (i.e., moderate-to-severe and severe OSA) in accordance with clinical practice standards.

Methods: We conducted a prospective, simultaneous study using the wireless radar system and PSG in a Northern Taiwan sleep center, involving 196 patients. The wireless radar sleep monitor, incorporating hybrid models such as deep neural decision trees, estimated the respiratory disturbance index relative to the total sleep time established by PSG (RDI_{PSG_TST}), by analyzing continuous-wave signals indicative of breathing patterns. Analyses were performed to examine the correlation and agreement between the RDI_{PSG_TST} and apnea-hypopnea index (AHI), results obtained through PSG. Cut-off thresholds for RDI_{PSG_TST} were determined using Youden's index, and multiclass classification was performed, after which the results were compared.

Results: A strong correlation (ρ = 0.91) and agreement (average difference of 0.59 events/h) between AHI and RDI_{PSG_TST} were identified. In terms of the agreement between the two devices, the average difference between PSG-based AHI and radar-based RDI_{PSG_TST} was 0.59 events/h, while 187 out of 196 cases (95.41%) fell within the 95% confidence interval of differences. A moderate-to-severe OSA model achieved an accuracy of 90.3% (cut-off threshold for RDI_{PSG_TST}: 19.2 events/h). A severe OSA model achieved an accuracy of 92.4% (cut-off threshold for RDI_{PSG_TST}: 28.86 events/h). The mean accuracy of multiclass classification performance using these cut-off thresholds was 83.7%.

Conclusions: The wireless-radar-based sleep monitoring device, with cut-off thresholds, can provide rapid OSA screening with acceptable accuracy, and also alleviate the burden on PSG capacity. However, to independently apply this framework, the function of determining the radar-based total sleep time requires further optimizations and verification in future work.

Keywords: apnea–hypopnea index (AHI); obstructive sleep apnea (OSA); polysomnography (PSG); respiratory disturbance index relative to the total sleep time established by PSG (RDI_{PSG_TST}); wireless radar sleep monitoring device.