Advanced Service Robots (ASR) Laboratory, Department of Mechatronics Engineering, School of Intelligent Systems Engineering, College of Interdisciplinary Science and Technology, University of Tehran, Tehran, Iran.
Department of Electrical Eng., California Polytechnic State University, San Luis Obispo, CA, 93407, USA.
Sci Rep. 2024 Oct 1;14(1):22696. doi: 10.1038/s41598-024-67729-9.
Pediatric Sleep Apnea-Hypopnea (SAH) presents a significant health challenge, particularly in diagnostic contexts, where conventional Polysomnography (PSG) testing, although effective, can be distressing for children. Addressing this, our research proposes a less invasive method to assess pediatric SAH severity by analyzing blood oxygen saturation (SpO2) signals. We adopted two advanced deep learning architectures, namely ResNet-based and attention-augmented hybrid CNN-BiGRU models, to process SpO2 signals in a one-dimensional (1D) format for Apnea-Hypopnea Index (AHI) estimation in pediatric subjects. Employing the CHAT dataset, which includes 844 SpO2 signals, the data was partitioned into training (60%), testing (30%), and validation (10%) sets. A predefined validation subset was randomly selected to ensure the models' robustness via a threefold cross-validation approach. Comparative analysis revealed that while the ResNet model attained an average accuracy of 72.9% across four SAH severity categories with a kappa score of 0.57, the CNN-BiGRU-Attention model demonstrated superior performance, achieving an average accuracy of 75.95% and a kappa score of 0.63. This distinction underscores our method's efficacy in both estimating AHI and categorizing SAH severity levels with notable precision. Further, to evaluate diagnostic capabilities, the models were benchmarked against common AHI thresholds (1, 5, and 10 events/hour) in each test fold, affirming their effectiveness in identifying pediatric SAH. This study marks a significant advance in the field, offering a non-invasive, child-friendly alternative for pediatric SAH diagnosis. Although challenges persist in accurately estimating AHI, particularly in severe cases, our findings represent a critical stride towards improving diagnostic processes in pediatric SAH.
儿科睡眠呼吸暂停低通气（SAH）是一个重大的健康挑战，特别是在诊断方面，尽管传统的多导睡眠图（PSG）测试有效，但对儿童来说可能会带来困扰。针对这一问题，我们的研究提出了一种通过分析血氧饱和度（SpO2）信号来评估儿童 SAH 严重程度的非侵入性方法。我们采用了两种先进的深度学习架构，即基于 ResNet 的和注意力增强混合 CNN-BiGRU 模型，以一维（1D）格式处理 SpO2 信号，用于估计儿科患者的呼吸暂停-低通气指数（AHI）。使用包含 844 个 SpO2 信号的 CHAT 数据集，数据被分为训练（60%）、测试（30%）和验证（10%）集。通过三折交叉验证方法，随机选择一个预定义的验证子集来确保模型的稳健性。比较分析表明，虽然 ResNet 模型在四个 SAH 严重程度类别中平均准确率为 72.9%，kappa 评分为 0.57，但 CNN-BiGRU-Attention 模型表现出更好的性能，平均准确率为 75.95%，kappa 评分为 0.63。这一区别突显了我们的方法在估计 AHI 和分类 SAH 严重程度方面的有效性，具有显著的精度。此外，为了评估诊断能力，将模型在每个测试折叠中与常见的 AHI 阈值（1、5 和 10 次/小时）进行了基准测试，证实了它们在识别儿科 SAH 方面的有效性。这项研究是该领域的一项重大进展，为儿科 SAH 诊断提供了一种非侵入性、适合儿童的替代方法。尽管在准确估计 AHI 方面仍然存在挑战，特别是在严重情况下，但我们的研究结果代表了在改善儿科 SAH 诊断过程方面的重要进展。
