基于触觉诱发表面肌电的触觉再现检测方法研究

As for study of the tactile representation (TR) in the field of human-computer interaction (HCI), it is always an interesting research program of the domestic and foreign research institutions that detecting the real tactile (RT) and the virtual tactile (VT) in the firing state of antennal nerve, and objective analysing the electrophysiological effect of tactile stimulation (TS). At present, the primary method in the the field of HCI is adjustment of tactile stimulation close VT to RT according to the user's feeling to. However, there is the temporal difference and individual difference of tactile sense. Adjusting TS based on subjective feeling always makes TR lack of universality and real-time. Even using microelectrode or probe to penetrate into neural, the tradional detection method can cause pain interference with tactile sensation. Therefore, this project will base on the principle of the tactile evoked surface electromyography (eSEMG) to follow these steps: (1)use skin surface electrode array to get weak eSEMG; (2)extract eSEMG characteristics associated with tactile stimulation and neural excitation to indirectly obtain tactile neural signal; (3)use time-frequency analysis, statistical pattern recognition, ensemble learning method to explore the associated mechanism between the surface TS and subcortical neural signals; (4)develop a method for detection of TR based on tactile eSEMG. The expecting outcome of this project: (1) provide new detection techniques and methods of electrophysiological detection for TR; (2) provides a new approach of research for the use of electrophysiological theory to analyse the TR in the field of HCI.

人机交互触觉再现研究中，检测真实触觉与虚拟触觉的触觉神经兴奋状态，客观分析触觉刺激的电生理功效，一直是国内外研究机构感兴趣的课题。目前，人机交互领域的主要方法是根据使用者主观感觉调整触觉刺激，使虚拟触觉接近真实触觉。然而主观调整触觉刺激，不同时间和不同人体，主观感觉存在差异，这导致触觉再现缺乏普适性和实时性。即使采用微电极或探针刺入神经的检测方法，又会导致痛觉干扰触觉感觉。因此，本项目根据触觉诱发肌电变化的原理，采用皮肤表面阵列电极获取微弱的表面肌电，通过提取与触觉刺激和神经兴奋关联的表面肌电特征，间接获取触觉神经信号。再运用时频分析、统计模式识别、集成学习等方法，探索体表触觉刺激与触觉诱发表面肌电之间的的关联机理，构建基于触觉诱发表面肌电的触觉再现检测方法。本项目的预期成果，将为触觉再现的电生理检测提供体表检测技术和理论方法，为运用电生理学分析人机交互领域的触觉再现提供新的研究途径。

项目背景：分析触觉刺激的电生理信号，能够定量分析人机交互虚拟触觉功效。目前，人机交互领域主观调整触觉刺激导致触觉再现缺乏普适性和实时性，即使采用探针刺入神经的检测方法，又会导致痛觉干扰触觉感觉。因此根据触觉诱发肌电变化的原理，通过提取与触觉刺激力和神经兴奋关联的表面肌电特征，间接获取触觉神经信号，运用电生理学分析人机交互领域的触觉刺激功效。首先研究动物模型（大鼠）指尖刺激与正中传入神经的关联性，进一步分析人体模型的指尖刺激与表面肌电，最后研究膝关节手术病人大腿力刺激与表面肌电的关联机理。重要结果：（1）研发出两套检测触觉刺激装置，一套用于检测动物模型的触觉刺激与正中神经放电，一套用于检测人体指尖触觉刺激与表面肌电。（2）获得动物模型（大鼠）指尖刺激与正中传入神经的关联性结果；（3）获得人体模型的指尖触觉刺激与表面肌电关联性结果；（4）临床应用，主要为：外力刺激膝关节镜手术病人大腿肌肉，采集关联表面肌电，运用表面肌电特征值提取，研究获得表面肌电-肌肉功能检测检测方法，实现对膝关节镜手术病人的肌肉触压功能检测。 关键数据：当指尖刺激力在2.4-4.4N，刺激力频率为1-5Hz时，关联肌肉的表面肌电信号强度呈相应的线性变化，上述研究及成果为定量同步检测触觉刺激力与表面肌电信号，提供了一种新的科学研究途径。