Department of Convergence Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, 05505, Republic of Korea.
Comput Biol Med. 2022 Jun;145:105430. doi: 10.1016/j.compbiomed.2022.105430. Epub 2022 Mar 22.
Quality assessment of bio-signals is important to prevent clinical misdiagnosis. With the introduction of mobile and wearable health care, it is becoming increasingly important to distinguish available signals from noise. The goal of this study was to develop a signal quality assessment technology for photoplethysmogram (PPG) widely used in wearable healthcare. In this study, we developed and verified a deep neural network (DNN)-based signal quality assessment model using about 1.6 million 5-s segment length PPG big data of about 29 GB from the MIMIC III PPG waveform database. The DNN model was implemented through a 1D convolutional neural network (CNN). The number of CNN layers, number of fully connected nodes, dropout rate, batch size, and learning rate of the model were optimized through Bayesian optimization. As a result, 6 CNN layers, 1,546 fully connected layer nodes, 825 batch size, 0.2 dropout rate, and 0.002 learning rate were needed for an optimal model. Performance metrics of the result of classifying waveform quality into 'Good' and 'Bad', the accuracy, specificity, sensitivity, area under the receiver operating curve, and area under the precision-recall curve were 0.978, 0.948, 0.993, 0.985, 0.980, and 0.969, respectively. Additionally, in the case of simulated real-time application, it was confirmed that the proposed signal quality score tracked the decrease in pulse quality well.
生物信号质量评估对于防止临床误诊非常重要。随着移动和可穿戴式医疗保健的引入，区分可用信号和噪声变得越来越重要。本研究旨在为广泛应用于可穿戴式医疗保健的光电容积脉搏波（PPG）开发一种信号质量评估技术。在这项研究中，我们使用来自 MIMIC III PPG 波形数据库的大约 29GB、长度约为 1.6 百万个 5 秒段的 PPG 大数据，开发并验证了一种基于深度神经网络（DNN）的信号质量评估模型。DNN 模型是通过一维卷积神经网络（CNN）实现的。通过贝叶斯优化对模型的 CNN 层数、全连接节点数、辍学率、批量大小和学习率进行了优化。结果，最优模型需要 6 个 CNN 层、1546 个全连接层节点、825 个批量大小、0.2 辍学率和 0.002 学习率。将波形质量分类为“良好”和“不良”的结果的性能指标，准确率、特异性、敏感性、接收者操作曲线下面积和精度召回曲线下面积分别为 0.978、0.948、0.993、0.985、0.980 和 0.969。此外，在模拟实时应用的情况下，证实了所提出的信号质量评分能够很好地跟踪脉搏质量的下降。
