Liu Ruhan, Li Chenyang, Xu Huajun, Wu Kejia, Li Xinyi, Liu Yupu, Yuan Jie, Meng Lili, Zou Jianyin, Huang Weijun, Yi Hongliang, Sheng Bin, Guan Jian, Yin Shankai
Department of Otolaryngology Head and Neck Surgery and Shanghai Key Laboratory of Sleep Disordered Breathing & Otolaryngological Institute of Shanghai Jiao Tong University, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Shanghai, People's Republic of China.
Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai, People's Republic of China.
Nat Sci Sleep. 2022 May 17;14:927-940. doi: 10.2147/NSS.S355369. eCollection 2022.
Misdiagnosis and missed diagnosis of sleep-disordered breathing (SDB) is common because polysomnography (PSG) is time-consuming, expensive, and uncomfortable. The use of recording methods based on the oxygen saturation (SpO2) signals detected by wearable devices is impractical and inaccurate for extracting signal features and detecting apnoeic events. We propose a method to automatically detect the apnoea-based SpO signal segments and compute the apnoea-hypopnea index (AHI) for SDB screening and grading.
First, apnoea-related desaturation segments in raw SpO signals were detected; global features were extracted from whole night signals. Then, the SpO signal segments and global features were fed into a bi-directional long short-term memory convolutional neural network model to identify apnoea-related and non-apnoea-related events. The apnoea-related segments were used to assess the AHI.
The model was trained on 500 individuals and tested on 8131 individuals from two public hospitals and one private centre. In the testing data, the classification accuracy for apnoea-related segments was 84.3%. Individuals with SDB (AHI 15) were identified with a mean accuracy of 88.95%.
Using automatic SDB detection based on SpO signals can accurately screen for SDB.
睡眠呼吸障碍（SDB）的误诊和漏诊很常见，因为多导睡眠图（PSG）耗时、昂贵且让人不适。基于可穿戴设备检测到的血氧饱和度（SpO2）信号的记录方法在提取信号特征和检测呼吸暂停事件方面不实用且不准确。我们提出一种方法，用于自动检测基于呼吸暂停的SpO信号段，并计算呼吸暂停低通气指数（AHI）以进行SDB筛查和分级。
首先，检测原始SpO信号中与呼吸暂停相关的血氧饱和度下降段；从整夜信号中提取全局特征。然后，将SpO信号段和全局特征输入双向长短期记忆卷积神经网络模型，以识别与呼吸暂停相关和不相关的事件。与呼吸暂停相关的段用于评估AHI。
该模型在500名个体上进行训练，并在来自两家公立医院和一个私立中心的8131名个体上进行测试。在测试数据中，与呼吸暂停相关段的分类准确率为84.3%。识别出SDB（AHI≥15）个体的平均准确率为88.95%。
使用基于SpO信号的SDB自动检测方法可以准确筛查SDB。
