Expert-level sleep scoring with deep neural networks.
Expert-level sleep scoring with deep neural networks.
复制标题
DOI：
10.1093/jamia/ocy131
复制
发表时间：
2018-12-01
期刊：
影响因子：
--
通讯作者：
Bianchi MT
中科院分区：
文献类型：
--
作者：
Biswal S;Sun H;Goparaju B;Westover MB;Sun J;Bianchi MT
对实验室多导睡眠图（PSG）数据进行评分仍然是一项手动任务，需要对3个主要类别进行视觉注释：睡眠阶段、睡眠呼吸障碍和肢体运动。由于PSG信号的复杂性和患者之间的生理异质性，使该过程自动化的尝试受到阻碍。深度神经网络最近在其他复杂的医疗任务中取得了专家级的性能，如果有足够的训练数据，它非常适合PSG评分。我们使用深度递归和卷积神经网络（RCNN）的组合来监督学习指定睡眠阶段，睡眠呼吸暂停事件和肢体运动的临床标签。测试和培训的数据来自10000个临床PSG和5804个研究PSG。当在临床数据集上训练时，RCNN再现了睡眠分期、睡眠呼吸暂停和肢体运动的PSG诊断评分，在保持的测试数据上的准确率分别为87.6%、88.2%和84.7%，这一性能水平与人类专家相当。RCNN模型在独立研究PSG数据库上进行测试时表现同样出色。当在有限的通道上训练以模拟家庭监测设备时，仅注意到准确性略有降低：额导联仅用于睡眠分期，胸带信号仅用于呼吸暂停低通气指数。通过为睡眠评分创建准确的深度学习模型，我们的工作为更广泛、更及时地获得睡眠诊断开辟了道路。准确的评分自动化可以提高实验室和家庭睡眠诊断方法的实用性和效率，从而有可能将睡眠专业知识的范围扩展到专业诊所之外。
Scoring laboratory polysomnography (PSG) data remains a manual task of visually annotating 3 primary categories: sleep stages, sleep disordered breathing, and limb movements. Attempts to automate this process have been hampered by the complexity of PSG signals and physiological heterogeneity between patients. Deep neural networks, which have recently achieved expert-level performance for other complex medical tasks, are ideally suited to PSG scoring, given sufficient training data. We used a combination of deep recurrent and convolutional neural networks (RCNN) for supervised learning of clinical labels designating sleep stages, sleep apnea events, and limb movements. The data for testing and training were derived from 10 000 clinical PSGs and 5804 research PSGs. When trained on the clinical dataset, the RCNN reproduces PSG diagnostic scoring for sleep staging, sleep apnea, and limb movements with accuracies of 87.6%, 88.2% and 84.7% on held-out test data, a level of performance comparable to human experts. The RCNN model performs equally well when tested on the independent research PSG database. Only small reductions in accuracy were noted when training on limited channels to mimic at-home monitoring devices: frontal leads only for sleep staging, and thoracic belt signals only for the apnea-hypopnea index. By creating accurate deep learning models for sleep scoring, our work opens the path toward broader and more timely access to sleep diagnostics. Accurate scoring automation can improve the utility and efficiency of in-lab and at-home approaches to sleep diagnostics, potentially extending the reach of sleep expertise beyond specialty clinics.
登录
查看更多内容
影响因子：
5.6
作者：
Malhotra, Atul;Younes, Magdy;Pien, Grace W.
通讯作者：
Pien, Grace W.
影响因子：
3
作者：
Hassan, Ahnaf Rashik;Bhuiyan, Mohammed Imamul Hassan
通讯作者：
Bhuiyan, Mohammed Imamul Hassan
影响因子：
4.4
作者：
Shambroom, John R.;Fabregas, Stephan E.;Johnstone, Jack
通讯作者：
Johnstone, Jack
影响因子：
64.8
作者：
Esteva A;Kuprel B;Novoa RA;Ko J;Swetter SM;Blau HM;Thrun S
通讯作者：
Thrun S
影响因子：
3.2
作者：
Anderer, P;Gruber, G;Dorffner, G
通讯作者：
Dorffner, G