Alarcón Ángel Serrano, Madrid Natividad Martínez, Seepold Ralf, Ortega Juan Antonio
School of Informatics, Reutlingen University, Reutlingen, Germany.
Computer Languages and Systems, University of Seville, Sevilla, Spain.
Front Neurosci. 2023 Jul 14;17:1155900. doi: 10.3389/fnins.2023.1155900. eCollection 2023.
Polysomnography (PSG) is the gold standard for detecting obstructive sleep apnea (OSA). However, this technique has many disadvantages when using it outside the hospital or for daily use. Portable monitors (PMs) aim to streamline the OSA detection process through deep learning (DL).
We studied how to detect OSA events and calculate the apnea-hypopnea index (AHI) by using deep learning models that aim to be implemented on PMs. Several deep learning models are presented after being trained on polysomnography data from the National Sleep Research Resource (NSRR) repository. The best hyperparameters for the DL architecture are presented. In addition, emphasis is focused on model explainability techniques, concretely on Gradient-weighted Class Activation Mapping (Grad-CAM).
The results for the best DL model are presented and analyzed. The interpretability of the DL model is also analyzed by studying the regions of the signals that are most relevant for the model to make the decision. The model that yields the best result is a one-dimensional convolutional neural network (1D-CNN) with 84.3% accuracy.
The use of PMs using machine learning techniques for detecting OSA events still has a long way to go. However, our method for developing explainable DL models demonstrates that PMs appear to be a promising alternative to PSG in the future for the detection of obstructive apnea events and the automatic calculation of AHI.
多导睡眠图（PSG）是检测阻塞性睡眠呼吸暂停（OSA）的金标准。然而，在医院外使用或日常使用该技术时存在许多缺点。便携式监测仪（PMs）旨在通过深度学习（DL）简化OSA检测过程。
我们研究了如何使用旨在在PMs上实现的深度学习模型来检测OSA事件并计算呼吸暂停低通气指数（AHI）。在对来自国家睡眠研究资源（NSRR）库的多导睡眠图数据进行训练后，展示了几种深度学习模型。给出了DL架构的最佳超参数。此外，重点关注模型可解释性技术，具体而言是梯度加权类激活映射（Grad-CAM）。
展示并分析了最佳DL模型的结果。还通过研究与模型决策最相关的信号区域来分析DL模型的可解释性。产生最佳结果的模型是一个一维卷积神经网络（1D-CNN），准确率为84.3%。
使用机器学习技术的PMs检测OSA事件仍有很长的路要走。然而，我们开发可解释DL模型的方法表明，PMs在未来检测阻塞性呼吸暂停事件和自动计算AHI方面似乎是PSG的一个有前景的替代方案。
