Department of Otorhinolaryngology-Head and Neck Surgery, Konkuk University School of Medicine, 120-1, Neungdong-ro, Gwangjin-gu, Seoul 05030, Korea.
Department of Otorhinolaryngology-Head and Neck Surgery, Soonchunhyang University College of Medicine, Bucheon Hospital, 170, Jomaru-ro, Bucheon 14584, Korea.
Medicina (Kaunas). 2022 Jun 9;58(6):779. doi: 10.3390/medicina58060779.
Polysomnography is manually scored by sleep experts. However, manual scoring is a time-consuming and labor-intensive task. The goal of this study was to verify the accuracy of automated sleep-stage scoring based on a deep learning algorithm compared to manual sleep-stage scoring. A total of 602 polysomnography datasets from subjects (Male:Female = 397:205) aged 19 to 65 years (mean age, 43.8, standard deviation = 12.2) were included in the study. The performance of the proposed model was evaluated based on kappa value and bootstrapped point-estimate of median percent agreement with a 95% bootstrap confidence interval and R = 1000. The proposed model was trained using 482 datasets and validated using 48 datasets. For testing, 72 datasets were selected randomly. The proposed model exhibited good concordance rates with manual scoring for stages W (94%), N1 (83.9%), N2 (89%), N3 (92%), and R (93%). The average kappa value was 0.84. For the bootstrap method, high overall agreement between the automated deep learning algorithm and manual scoring was observed in stages W (98%), N1 (94%), N2 (92%), N3 (99%), and R (98%) and total (96%). Automated sleep-stage scoring using the proposed model may be a reliable method for sleep-stage classification.
多导睡眠图由睡眠专家手动评分。然而，手动评分是一项耗时且劳动密集型的任务。本研究的目的是验证基于深度学习算法的自动睡眠分期评分与手动睡眠分期评分相比的准确性。共纳入了 602 名年龄在 19 至 65 岁（平均年龄 43.8，标准差=12.2）的受试者的多导睡眠图数据集（男性：女性=397:205）。该模型的性能通过kappa 值和中位数百分比一致性的bootstrap 点估计（95%bootstrap 置信区间和 R=1000）进行评估。该模型使用 482 个数据集进行训练，使用 48 个数据集进行验证，使用 72 个随机数据集进行测试。该模型与手动评分在 W 期（94%）、N1 期（83.9%）、N2 期（89%）、N3 期（92%）和 R 期（93%）的一致性较好。平均 kappa 值为 0.84。对于bootstrap 方法，在 W 期（98%）、N1 期（94%）、N2 期（92%）、N3 期（99%）和 R 期（98%）和总分期（96%）中，自动睡眠分期评分与手动评分之间存在高度总体一致性。使用该模型进行自动睡眠分期评分可能是一种可靠的睡眠分期分类方法。
