Tsai Cheng-Yu, Liu Wen-Te, Hsu Wen-Hua, Majumdar Arnab, Stettler Marc, Lee Kang-Yun, Cheng Wun-Hao, Wu Dean, Lee Hsin-Chien, Kuan Yi-Chun, Wu Cheng-Jung, Lin Yi-Chih, Ho Shu-Chuan
Department of Civil and Environmental Engineering, Imperial College London, London, UK.
School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei, Taiwan.
Digit Health. 2023 Mar 6;9:20552076231152751. doi: 10.1177/20552076231152751. eCollection 2023 Jan-Dec.
Obstructive sleep apnea (OSA) is typically diagnosed by polysomnography (PSG). However, PSG is time-consuming and has some clinical limitations. This study thus aimed to establish machine learning models to screen for the risk of having moderate-to-severe and severe OSA based on easily acquired features.
We collected PSG data on 3529 patients from Taiwan and further derived the number of snoring events. Their baseline characteristics and anthropometric measures were obtained, and correlations among the collected variables were investigated. Next, six common supervised machine learning techniques were utilized, including random forest (RF), extreme gradient boosting (XGBoost), k-nearest neighbor (kNN), support vector machine (SVM), logistic regression (LR), and naïve Bayes (NB). First, data were independently separated into a training and validation dataset (80%) and a test dataset (20%). The approach with the highest accuracy in the training and validation phase was employed to classify the test dataset. Next, feature importance was investigated by calculating the Shapley value of every factor, which represented the impact on OSA risk screening.
The RF produced the highest accuracy (of >70%) in the training and validation phase in screening for both OSA severities. Hence, we employed the RF to classify the test dataset, and results showed a 79.32% accuracy for moderate-to-severe OSA and 74.37% accuracy for severe OSA. Snoring events and the visceral fat level were the most and second most essential features of screening for OSA risk.
The established model can be considered for screening for the risk of having moderate-to-severe or severe OSA.
阻塞性睡眠呼吸暂停（OSA）通常通过多导睡眠图（PSG）进行诊断。然而，PSG耗时且存在一些临床局限性。因此，本研究旨在基于易于获取的特征建立机器学习模型，以筛查中重度和重度OSA的风险。
我们收集了来自台湾的3529例患者的PSG数据，并进一步得出打鼾事件的数量。获取了他们的基线特征和人体测量指标，并研究了所收集变量之间的相关性。接下来，使用了六种常见的监督机器学习技术，包括随机森林（RF）、极端梯度提升（XGBoost）、k近邻（kNN）、支持向量机（SVM）、逻辑回归（LR）和朴素贝叶斯（NB）。首先，将数据独立分为训练和验证数据集（80%）以及测试数据集（20%）。采用在训练和验证阶段准确率最高的方法对测试数据集进行分类。接下来，通过计算每个因素的Shapley值来研究特征重要性，该值代表对OSA风险筛查的影响。
RF在训练和验证阶段对两种OSA严重程度的筛查中均产生了最高准确率（>70%）。因此，我们使用RF对测试数据集进行分类，结果显示中重度OSA的准确率为79.32%，重度OSA的准确率为74.37%。打鼾事件和内脏脂肪水平是筛查OSA风险最重要和第二重要的特征。
所建立的模型可用于筛查中重度或重度OSA的风险。
