This technology simultaneously acquires surface electromyography and three ultrasound imaging signals to capture detailed muscle activity. Real-time processed ultrasound images extract subject-specific muscle variables which are integrated into a hill-type muscle model. An optimization process computes an allocation coefficient that effectively fuses the ultrasound and sEMG data, enabling accurate prediction of joint torque. These features allow the system to measure and model muscle dynamics with high resolution, promising improved performance in human movement analysis for applications such as robotic neurorehabilitation.
Description
What sets this approach apart is its dual-modality integration that significantly enhances movement intent detection over traditional single-sensor methods. By combining the rapid responsiveness of sEMG with the detailed, subject-specific insights provided by ultrasound imaging, the technology addresses limitations inherent in isolated measurement systems. Experimental validations confirm that blending these signals leads to more reliable torque prediction, thus offering a more robust solution for dynamic, person-specific rehabilitation therapy. This innovative fusion of data sources positions the technology as a cutting-edge tool in enhancing the efficacy of human-machine interfaces.
Applications
- Neurorehabilitation robot control
- Exoskeleton assist systems
- Prosthetic movement detection
- Remote therapy monitoring
Advantages
- Enhanced accuracy in movement intent detection by fusing sEMG and ultrasound data into a personalized muscle model.
- Real-time, subject-specific muscle variable measurements improve joint torque predictions.
- Robust integration of multiple sensor modalities overcomes the limitations of single-sensor approaches.
- Enables more effective robotic neurorehabilitation support for individuals recovering from stroke or spinal cord injury.
IP Status
Copyright
