Lee Kun-Ta, Liu Wen-Te, Lin Yi-Chih, Chen Zhihe, Ho Yu-Hsuan, Huang Yu-Wen, Tsai Zong-Lin, Hsu Chih-Wei, Yeh Shang-Min, Lin Hsiao Yi, Majumdar Arnab, Chen Yen-Ling, Kuan Yi-Chun, Lee Kang-Yun, Feng Po-Hao, Chen Kuan-Yuan, Kang Jiunn-Horng, Lee Hsin-Chien, Ho Shu-Chuan, Tsai Cheng-Yu
Respiratory Therapy Room, Division of Pulmonary Medicine, Taipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan.
School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei, Taiwan.
J Multidiscip Healthc. 2025 Jan 23;18:381-393. doi: 10.2147/JMDH.S486261. eCollection 2025.
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.
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 (bRDI). Additionally, Youden's index was used to establish cutoff thresholds for the bRDI, followed by multiclass classification and outcome comparisons.
A strong correlation ( = 0.87) and high agreement (93.88% within the 95% confidence interval; 138/147) between the AHI and bRDI were identified. The moderate-to-severe OSA model achieved 83.67% accuracy (with a bRDI cutoff of 21.19 events/h), and the severe OSA model demonstrated 93.21% accuracy (with a bRDI cutoff of 28.14 events/h). The average accuracy of multiclass classification using these thresholds was 78.23%.
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.
诊断阻塞性睡眠呼吸暂停（OSA）的常用检查方法是多导睡眠图（PSG）和家庭睡眠呼吸暂停检测（HSAT）。然而，PSG和HSAT都需要将传感器连接到受试者身上，这可能会干扰他们的睡眠并影响结果。因此，在本研究中，我们旨在验证一种结合深度学习技术的无线雷达框架，以在家庭环境中筛查OSA风险。
本研究前瞻性地使用HSAT和24GHz无线雷达框架，在147个晚上收集了80名参与者的家庭睡眠参数。所提出的框架使用混合模型（即深度神经决策树），通过分析指示呼吸模式的连续波信号来识别呼吸事件。进行分析以检查呼吸暂停低通气指数（AHI）与通过HSAT获得的结果以及基于HSAT卧床时间的基于雷达的呼吸紊乱指数（bRDI）之间的相关性和一致性。此外，使用约登指数建立bRDI的截断阈值，随后进行多类分类和结果比较。
AHI与bRDI之间存在强相关性（=0.87）和高度一致性（95%置信区间内为93.88%；138/147）。中度至重度OSA模型的准确率达到83.67%（bRDI截断值为21.19次/小时），重度OSA模型的准确率为93.21%（bRDI截断值为28.14次/小时）。使用这些阈值进行多类分类的平均准确率为78.23%。
所提出的框架及其截断阈值有可能在家庭环境中作为HSAT的替代方法应用，在筛查OSA时提供可接受的准确率，且不受连接传感器的干扰。然而，基于雷达的总睡眠时间功能需要进一步优化和验证才能独立应用。
