Assessment of Mandibular Movement Monitoring With Machine Learning Analysis for the Diagnosis of Obstructive Sleep Apnea
Assessment of Mandibular Movement Monitoring With Machine Learning Analysis for the Diagnosis of Obstructive Sleep Apnea
复制标题
下颌运动监测与机器学习分析在阻塞性睡眠呼吸暂停诊断中的应用
DOI：
10.1001/jamanetworkopen.2019.19657
复制
发表时间：
2020-01-22
影响因子：
13.8
通讯作者：
Gozal, David
中科院分区：
文献类型：
--
作者：
Pepin, Jean-Louis;Letesson, Clement;Gozal, David
这项诊断研究比较了睡眠期间下颌运动监测的性能，结合机器学习与多导睡眠图的自动分析，用于诊断成人阻塞性睡眠呼吸暂停。重要性鉴于阻塞性睡眠呼吸暂停（OSA）的高患病率，需要更简单和自动化的诊断方法。目的评估睡眠期间下颌运动（MM）监测结合机器学习自动分析是否适用于OSA诊断。设计、设置和参与者：对在一个学术机构（睡眠实验室，Centre Hospitalier Universaire Universite Catholique de Louvain纳穆尔Site Sainte-Elisabeth，纳穆尔，比利时）的睡眠诊所接受过夜实验室多导睡眠描记（PSG）作为参考方法与同时MM监测的成人进行诊断研究。疑似OSA患者于2017年7月5日至2018年10月31日入组。阻塞性睡眠呼吸暂停的诊断需要诱发体征或症状或相关的医学或精神共病，加上PSG衍生的呼吸障碍指数（PSG-RDI）至少为5次/h。即使在没有相关症状或合并症的情况下，PSG-RDI至少15起事件/h也符合诊断标准。不符合这些标准的患者被归类为没有OSA。通过Bland-Altman图评估一致性分析和诊断性能，比较PSG-RDI和日出系统RDI（Sr-RDI），并通过受试者工作特征曲线优化诊断阈值，从而评价在5起事件/h和15起事件/h时检测OSA的器械灵敏度和特异性。结果376例疑似OSA患者中，男性207例（55.1%），平均年龄49.7（13.2）岁，平均体重指数31.0（7.1）。在无OSA患者（n = 46;平均差异，1.31; 95% CI，-1.05至3.66起事件/h）和PSG-RDI至少为5起事件/h且有症状的OSA患者（n = 107;平均差异，-0.69; 95% CI，-3.77至2.38起事件/h）中，PSG-RDI和Sr-RDI之间存在可靠的一致性。在PSG-RDI至少为15起事件/h的OSA患者中，检测到Sr-RDI低估-11.74（95% CI，-20.83至-2.67）起事件/h，并通过优化日出系统诊断阈值进行校正。Sr-RDI显示出诊断能力，对于相应的PSG-RDI为5起事件/h和15起事件/h，受试者工作特征曲线下面积分别为0.95（95%CI，0.92-0.96）和0.93（95%CI，0.90-0.93）。在7.63次/h和12.65次/h两个最佳截止值下，Sr-RDI的准确率为0.92（95% CI，0.90-0.94）和0.88（95% CI，0.86-0.90）以及后验概率0.99（95%CI，0.99-0.99）和0.89（95% CI，0.88-0.91）PSG-RDI分别为至少5起事件/h和至少15起事件/h，对应阳性似然比为14.86（95% CI，9.86-30.12）和5.63（95% CI，4.92-7.27）。结论和相关性MM模式的自动分析为RDI计算提供了可靠的性能。使用这个指数在阻塞性睡眠呼吸暂停的诊断似乎是有前途的。问题如何进行自动下颌运动分析的诊断阻塞性睡眠呼吸暂停的比较与多导睡眠图的性能？结果在这项诊断性研究中，376名成人疑似阻塞性睡眠呼吸暂停，下颌运动衍生的呼吸障碍指数确定患者与多导睡眠呼吸障碍指数至少5个事件/小时或至少15个事件/小时的准确性分别为0.92和0.88。自动分析下颌运动模式可靠地计算呼吸障碍指数，并使用这种方法在阻塞性睡眠呼吸暂停的诊断似乎是有希望的。
This diagnostic study compares the performance of mandibular movement monitoring during sleep coupled with an automated analysis by machine learning vs polysomnography for the diagnosis of obstructive sleep apnea in adults.Importance Given the high prevalence of obstructive sleep apnea (OSA), there is a need for simpler and automated diagnostic approaches. Objective To evaluate whether mandibular movement (MM) monitoring during sleep coupled with an automated analysis by machine learning is appropriate for OSA diagnosis. Design, Setting, and Participants Diagnostic study of adults undergoing overnight in-laboratory polysomnography (PSG) as the reference method compared with simultaneous MM monitoring at a sleep clinic in an academic institution (Sleep Laboratory, Centre Hospitalier Universitaire Universite Catholique de Louvain Namur Site Sainte-Elisabeth, Namur, Belgium). Patients with suspected OSA were enrolled from July 5, 2017, to October 31, 2018. Main Outcomes and Measures Obstructive sleep apnea diagnosis required either evoking signs or symptoms or related medical or psychiatric comorbidities coupled with a PSG-derived respiratory disturbance index (PSG-RDI) of at least 5 events/h. A PSG-RDI of at least 15 events/h satisfied the diagnosis criteria even in the absence of associated symptoms or comorbidities. Patients who did not meet these criteria were classified as not having OSA. Agreement analysis and diagnostic performance were assessed by Bland-Altman plot comparing PSG-RDI and the Sunrise system RDI (Sr-RDI) with diagnosis threshold optimization via receiver operating characteristic curves, allowing for evaluation of the device sensitivity and specificity in detecting OSA at 5 events/h and 15 events/h. Results Among 376 consecutive adults with suspected OSA, the mean (SD) age was 49.7 (13.2) years, the mean (SD) body mass index was 31.0 (7.1), and 207 (55.1%) were men. Reliable agreement was found between PSG-RDI and Sr-RDI in patients without OSA (n = 46; mean difference, 1.31; 95% CI, -1.05 to 3.66 events/h) and in patients with OSA with a PSG-RDI of at least 5 events/h with symptoms (n = 107; mean difference, -0.69; 95% CI, -3.77 to 2.38 events/h). An Sr-RDI underestimation of -11.74 (95% CI, -20.83 to -2.67) events/h in patients with OSA with a PSG-RDI of at least 15 events/h was detected and corrected by optimization of the Sunrise system diagnostic threshold. The Sr-RDI showed diagnostic capability, with areas under the receiver operating characteristic curve of 0.95 (95% CI, 0.92-0.96) and 0.93 (95% CI, 0.90-0.93) for corresponding PSG-RDIs of 5 events/h and 15 events/h, respectively. At the 2 optimal cutoffs of 7.63 events/h and 12.65 events/h, Sr-RDI had accuracy of 0.92 (95% CI, 0.90-0.94) and 0.88 (95% CI, 0.86-0.90) as well as posttest probabilities of 0.99 (95% CI, 0.99-0.99) and 0.89 (95% CI, 0.88-0.91) at PSG-RDIs of at least 5 events/h and at least 15 events/h, respectively, corresponding to positive likelihood ratios of 14.86 (95% CI, 9.86-30.12) and 5.63 (95% CI, 4.92-7.27), respectively. Conclusions and Relevance Automatic analysis of MM patterns provided reliable performance in RDI calculation. The use of this index in OSA diagnosis appears to be promising.Question How does the performance of an automated mandibular movement analysis for the diagnosis of obstructive sleep apnea compare with that of polysomnography? Findings In this diagnostic study of 376 adults with suspected obstructive sleep apnea, the mandibular movement-derived respiratory disturbance index identified patients with a polysomnography respiratory disturbance index of at least 5 events/h or at least 15 events/h with accuracy of 0.92 and 0.88, respectively. Meaning Automatic analysis of mandibular movement patterns reliably calculated respiratory disturbance index, and the use of this approach in obstructive sleep apnea diagnosis appears to be promising.