IEEE Trans Biomed Circuits Syst. 2018 Feb;12(1):68-79. doi: 10.1109/TBCAS.2017.2757400.
Wearable technology has gained increasing popularity in the applications of healthcare, sports science, and biomedical engineering in recent years. Because of its convenient nature, the wearable technology is particularly useful in the acquisition of the physiological signals. Specifically, the (surface electromyography) sEMG systems, which measure the muscle activation potentials, greatly benefit from this technology in both clinical and industrial applications. However, the current wearable sEMG systems have several drawbacks including inefficient noise cancellation, insufficient measurement quality, and difficult integration to customized applications. Additionally, none of these sEMG data acquisition systems can detect sEMG signals (i.e., contractions), which provides a valuable environment for further studies such as human machine interaction, gesture recognition, and fatigue tracking. To this end, we introduce an embedded, eight channel, noise canceling, wireless, wearable sEMG data acquisition system with adaptive muscle contraction detection. Our design consists of two stages, which are the sEMG sensors and the multichannel data acquisition unit. For the first stage, we propose a low cost, dry, and active sEMG sensor that captures the muscle activation potentials, a data acquisition unit that evaluates these captured multichannel sEMG signals and transmits them to a user interface. In the data acquisition unit, the sEMG signals are processed through embedded, adaptive methods in order to reject the power line noise and detect the muscle contractions. Through extensive experiments, we demonstrate that our sEMG sensor outperforms a widely used commercially available product and our data acquisition system achieves 4.583 dB SNR gain with accuracy in the detection of the contractions.
可穿戴技术近年来在医疗保健、运动科学和生物医学工程等领域的应用越来越受到关注。由于其方便的性质，可穿戴技术在生理信号的获取中特别有用。具体来说，测量肌肉激活电位的表面肌电图 (sEMG) 系统在临床和工业应用中非常受益于这项技术。然而，目前的可穿戴 sEMG 系统存在一些缺点，包括噪声消除效率低、测量质量不足以及难以集成到定制应用中。此外，这些 sEMG 数据采集系统都无法检测 sEMG 信号（即收缩），这为进一步的研究提供了宝贵的环境，例如人机交互、手势识别和疲劳跟踪。为此，我们引入了一种嵌入式、八通道、噪声消除、无线、可穿戴的 sEMG 数据采集系统，具有自适应肌肉收缩检测功能。我们的设计由两个阶段组成，即 sEMG 传感器和多通道数据采集单元。在第一阶段，我们提出了一种低成本、干式、主动的 sEMG 传感器，用于捕获肌肉激活电位，以及一个数据采集单元，用于评估这些捕获的多通道 sEMG 信号并将其传输到用户界面。在数据采集单元中，sEMG 信号通过嵌入式自适应方法进行处理，以消除电源线噪声并检测肌肉收缩。通过广泛的实验，我们证明我们的 sEMG 传感器优于广泛使用的商业产品，并且我们的数据采集系统在收缩检测方面实现了 4.583dB SNR 增益和准确性。
