基于多模态信号的手腕意向性震颤抑制系统研究

As a common disease of neurological disorders, intention tremor of wrist seriously affects the patient's daily life and social interaction, such as fine movements, writing and catering. However, many factors and complex characteristics result in wrist intentional tremor. The control theory and method of suppressing tremor are not perfect. Until now, there is still no complete method to cure it in medical science. The aim of this project is to study the wrist intentional tremor suppression system based on multi-modal signals. Through characteristics of mult-imodal signals for intention tremors of two degrees of freedom, wrist musculoskeletal model is exactly established through the optimization of Hammerstein model structure. An identification method with high precision is proposed, which is suitable for patients with different clinical needs. With muscle fatigue, changes of tremor signal and musculoskeletal model parameters, a robust control system of functional electrical stimulation is designed to suppress the wrist intention tremor and doesn't affect the patient's voluntary motion. Through simulation and preliminary clinical trials, the proposed control theory and methods of tremor suppression are further improved. The project will provide theoretical and technical supports for effective tremor suppressions.

手腕意向性震颤作为常见的运动型障碍神经疾病，严重影响患者精细动作、书写和餐饮等日常生活及社会交往。但引发手腕意向性震颤的因素众多、特征复杂，抑制震颤的控制理论及方法尚不完善，医学上尚无完全治愈方法。本项目拟基于多模态信号对手腕意向性震颤抑制系统进行研究，通过深入分析手腕内外翻及上下切两个自由度的意向性震颤信号及相应肌肉块肌音信号等多模态信号的特征，优化Hammerstein模型结构，建立准确的手腕肌骨模型，提出适合于不同患者临床需求的快速高精度参数辨识方法；考虑肌肉疲劳、震颤信号及肌骨模型参数等的不同变化特征，对重复控制框架进行优化，设计手腕意向性震颤抑制鲁棒控制策略，在提高手腕意向性震颤抑制效果的同时，不影响患者主观意向运动；搭建基于多模态信号的手腕意向性震颤抑制控制系统，开展实验和临床验证，为有效抑制震颤提供理论与技术支持。

手腕意向性震颤是一种运动型障碍神经疾病，严重影响患者日常生活。但引发因素众多、特征复杂，通过功能性电刺激抑制手腕意向性震颤的控制理论及方法尚不完善。申请人通过本项目研究，基于肌电信号、肌音信号与手腕的加速度等多模态信息采集，深入分析了手腕内外翻及上下切两个自由度的意向性震颤信号及相应肌肉生理信号等信号特征，基于Hammerstein模型结构建立优化的手腕肌骨模型，提出了适合于不同患者的肌骨模型参数辨识方法；充分考虑不同人的震颤信号及肌骨模型参数变化，设计并研究了基于梯度的重复控制、多周期重复控制、插入式重复控制、多模切换重复控制等功能性电刺激的控制策略，实现了手腕意向性震颤抑制，在提高手腕意向性震颤抑制效果的同时，不影响患者主观意向运动；通过仿真及初步临床试验，进一步证明了控制方法的适用性与有效性。搭建了基于多模态信号的手腕意向性震颤抑制平台，为功能性电刺激实现手腕意向性震颤控制的临床推广提供了理论基础与技术支持。在项目执行期间，发表学术论文6篇，其中SCI收录2篇，EI收录4篇；申请国家发明专利6项；培养了2名博士研究生和3名硕士研究生。