Lin Chuang, Xiao Qiong, Zhao Penghui
The School of Information Science and Technology, Dalian Maritime University, Dalian, 116026, China.
J Neuroeng Rehabil. 2024 Dec 30;21(1):231. doi: 10.1186/s12984-024-01525-4.
Simultaneous and proportional control (SPC) based on surface electromyographic (sEMG) signals has emerged as a research hotspot in the field of human-machine interaction (HMI). However, the existing continuous motion estimation methods mostly have an average Pearson coefficient (CC) of less than 0.85, while high-precision methods suffer from the problem of long inference time (> 200 ms) and can only estimate SPC of less than 15 hand movements, which limits their applications in HMI. To overcome these problems, we propose a smooth Multi-scale Attention Patching Encoder Network (sMAPEN).
The sMAPEN consists of three modules, the Multi-scale Attention Fusion (MAF) module, the Patching Encoder (PE) module, and a smoothing layer. The MAF module adaptively captures the local spatiotemporal features at multiple scales, the PE module acquires the global spatiotemporal features of sEMG, and the smoothing layer further improves prediction stability.
To evaluate the performance of the model, we conducted continuous estimation of 40 subjects performing over 40 different hand movements on the Ninapro DB2. The results show that the average Pearson correlation coefficient (CC), normalized root mean square error (NRMSE), coefficient of determination (R), and smoothness (SMOOTH) of the sMAPEN model are 0.9082, 0.0646°, 0.8163, and - 0.0017, respectively, which significantly outperforms that of the state-of-the-art methods in all metrics (p < 0.01). Furthermore, we tested the deployment performance of sMAPEN on the portable device, with a delay of only 97.93 ms.
Our model can predict up to 40 hand movements while achieving the highest predicting accuracy compared with other methods. Besides, the lightweight design strategy brings an improvement in inference speed, which enables the model to be deployed on wearable devices. All these promotions imply that sMAPEN holds great potential in HMI.
基于表面肌电（sEMG）信号的同步比例控制（SPC）已成为人机交互（HMI）领域的研究热点。然而，现有的连续运动估计方法大多平均皮尔逊系数（CC）小于0.85，而高精度方法存在推理时间长（>200毫秒）的问题，并且只能估计少于15种手部动作的SPC，这限制了它们在HMI中的应用。为克服这些问题，我们提出了一种平滑多尺度注意力补丁编码器网络（sMAPEN）。
sMAPEN由三个模块组成，即多尺度注意力融合（MAF）模块、补丁编码器（PE）模块和平滑层。MAF模块自适应地捕获多个尺度的局部时空特征，PE模块获取sEMG的全局时空特征，平滑层进一步提高预测稳定性。
为评估模型性能，我们在Ninapro DB2上对40名受试者进行40多种不同手部动作的连续估计。结果表明，sMAPEN模型的平均皮尔逊相关系数（CC）、归一化均方根误差（NRMSE）、决定系数（R）和平滑度（SMOOTH）分别为0.9082、0.0646°、0.8163和 -0.0017，在所有指标上均显著优于现有方法（p <0.01）。此外，我们测试了sMAPEN在便携式设备上的部署性能，延迟仅为97.93毫秒。
与其他方法相比，我们的模型在预测多达40种手部动作的同时实现了最高的预测精度。此外，轻量级设计策略提高了推理速度，使模型能够部署在可穿戴设备上。所有这些优势表明sMAPEN在HMI中具有巨大潜力。
