Photoplethysmogram (PPG) signals are easily contaminated by motion artifacts in real-world settings, despite their widespread use in Internet-of-Things (IoT) based wearable and smart health devices for cardiovascular health monitoring. This study proposed a lightweight deep neural network, called Tiny-PPG, for accurate and real-time PPG artifact segmentation on IoT edge devices. The model was trained and tested on a public dataset, PPG DaLiA, which featured complex artifacts with diverse lengths and morphologies during various daily activities of 15 subjects using a watch-type device (Empatica E4). The model structure, training method and loss function were specifically designed to balance detection accuracy and speed for real-time PPG artifact detection in resource-constrained embedded devices. To optimize the model size and capability in multi-scale feature representation, the model employed deep separable convolution and atrous spatial pyramid pooling modules, respectively. Additionally, the contrastive loss was also utilized to further optimize the feature embeddings. With additional model pruning, Tiny-PPG achieved state-of-the-art detection accuracy of 87.8% while only having 19,726 model parameters (0.15 megabytes), and was successfully deployed on an STM32 embedded system for real-time PPG artifact detection. Therefore, this study provides an effective solution for resource-constraint IoT smart health devices in PPG artifact detection.

翻译：光电容积描记（PPG）信号虽在基于物联网（IoT）的可穿戴及智能健康设备的心血管健康监测中广泛应用，但在实际场景中易受运动伪迹干扰。本研究提出名为Tiny-PPG的轻量深度神经网络，用于在IoT边缘设备上实现高精度实时的PPG伪迹分割。模型基于公开数据集PPG DaLiA进行训练与测试，该数据集包含15名受试者佩戴手表式设备（Empatica E4）进行多种日常活动时产生的形态与长度各异的复杂伪迹。为在资源受限的嵌入式设备上实现实时PPG伪迹检测，模型结构、训练方法与损失函数均针对检测精度与速度的平衡进行专门设计。通过采用深度可分离卷积与空洞空间金字塔池化模块，分别优化模型参数量与多尺度特征表征能力；同时引入对比损失进一步优化特征嵌入。经模型剪枝后，Tiny-PPG以仅19,726个模型参数（0.15兆字节）达到87.8%的检测精度，成功部署于STM32嵌入式系统实现实时PPG伪迹检测。本研究为资源受限的IoT智能健康设备在PPG伪迹检测领域提供了有效解决方案。