Utilizing a Wireless Radar Framework in Combination With Deep Learning Approaches to Evaluate Obstructive Sleep Apnea Severity in Home-Setting Environments

Abstract

Kun-Ta Lee,1 Wen-Te Liu,2– 6 Yi-Chih Lin,3,7 Zhihe Chen,8 Yu-Hsuan Ho,9 Yu-Wen Huang,9 Zong-Lin Tsai,9 Chih-Wei Hsu,9 Shang-Min Yeh,9 Hsiao Yi Lin,9 Arnab Majumdar,8 Yen-Ling Chen,10 Yi-Chun Kuan,3,11– 13 Kang-Yun Lee,4,6 Po-Hao Feng,4,6 Kuan-Yuan Chen,4,6 Jiunn-Horng Kang,5,6,14– 16 Hsin-Chien Lee,17 Shu-Chuan Ho,2,6,* Cheng-Yu Tsai2– 6,16,18,* 1Respiratory Therapy Room, Division of Pulmonary Medicine, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan; 2School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei, Taiwan; 3Sleep Center, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan; 4Division of Pulmonary Medicine, Department of Internal Medicine, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan; 5Research Center of Artificial Intelligence in Medicine, Taipei Medical University, Taipei, Taiwan; 6Research Center of Thoracic Medicine, Taipei Medical University, Taipei, Taiwan; 7Department of Otolaryngology, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan; 8Department of Civil and Environmental Engineering, Imperial College London, London, UK; 9Advanced Technology Lab, Wistron Corporation, Taipei, Taiwan; 10Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan; 11Department of Neurology, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan; 12Department of Neurology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan; 13Taipei Neuroscience Institute, Taipei Medical University, Taipei, Taiwan; 14Department of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei, Taiwan; 15Graduate Institute of Nanomedicine and Medical Engineering, College of Biomedical Engineering, Taipei Medical University, Taipei, Taiwan; 16Professional Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan; 17Department of Psychiatry, Taipei Medical University Hospital, Taipei, Taiwan; 18School of Biomedical Engineering, College of Biomedical Engineering, Taipei Medical University, Taipei, Taiwan*These authors contributed equally to this workCorrespondence: Cheng-Yu Tsai, School of Biomedical Engineering, College of Biomedical Engineering, Taipei Medical University, 250 Wuxing Street, Taipei, 11031, Taiwan, Tel [phone], Fax [phone], Email [email protected] Shu-Chuan Ho, School of Respiratory Therapy, College of Medicine, Taipei Medical University, 250 Wuxing Street, Taipei, 11031, Taiwan, Tel [phone], Fax [phone], Email [email protected]: Common examinations for diagnosing obstructive sleep apnea (OSA) are polysomnography (PSG) and home sleep apnea testing (HSAT). However, both PSG and HSAT require that sensors be attached to a subject, which may disturb their sleep and affect the results. Hence, in this study, we aimed to verify a wireless radar framework combined with deep learning techniques to screen for the risk of OSA in home-based environments.Methods: This study prospectively collected home-based sleep parameters from 80 participants over 147 nights using both HSAT and a 24-GHz wireless radar framework. The proposed framework, using hybrid models (ie, deep neural decision trees), identified respiratory events by analyzing continuous-wave signals indicative of breathing patterns. Analyses were performed to examine correlations and agreement of the apnea-hypopnea index (AHI) with results obtained through HSAT and the radar-based respiratory disturbance index based on the time in bed from HSAT (bRDITIB). Additionally, Youden’s index was used to establish cutoff thresholds for the bRDITIB, followed by multiclass classification and outcome comparisons.Results: A strong correlation (ρ = 0.87) and high agreement (93.88% within the 95% confidence interval; 138/147) between the AHI and bRDITIB were identified. The moderate-to-severe OSA model achieved 83.67% accuracy (with a bRDITIB cutoff of 21.19 events/h), and the severe OSA model demonstrated 93.21% accuracy (with a bRDITIB cutoff of 28.14 events/h). The average accuracy of multiclass classification using these thresholds was 78.23%.Conclusion: The proposed framework, with its cutoff thresholds, has the potential to be applied in home settings as a surrogate for HSAT, offering acceptable accuracy in screening for OSA without the interference of attached sensors. However, further optimization and verification of the radar-based total sleep time function are necessary for independent application.Keywords: obstructive sleep apnea, OSA, home sleep apnea testing, HSAT, wireless radar framework, apnea-hypopnea index, AHI, respiratory disturbance index based on the time in bed from HSAT, bRDITIB

Keywords

• obstructive sleep apnea
• osa
• home sleep apnea testing
• hsat
• wireless radar framework
• apnea-hypopnea index
• ahi
• respiratory disturbance index based on the time in bed from hsat
• brditib