sEMG-based Motion Recognition for Robotic Surgery Training - A Preliminary Study.

Area of Science:

• Biomedical Engineering
• Robotics
• Neuroscience

Background:

• Robotic surgery demands efficient surgeon training and skill assessment.
• Conventional methods may not adequately address the learning curve for robotic procedures.
• Surface electromyography (sEMG) offers a novel approach to analyze muscle activity during training.

Purpose of the Study:

• To explore the feasibility of using sEMG signals for motion primitive recognition in robotic surgery training.
• To develop interpretable information from sEMG data to enhance trainee performance.
• To lay the groundwork for new skill assessment criteria based on muscle activity.

Main Methods:

• Utilized machine learning (ML) techniques to analyze sEMG signals.
• Focused on recognizing hand motions along 3 Cartesian axes.
• Employed a commercial robotic surgery training platform with a virtual reality (VR) environment.

Main Results:

• Demonstrated the feasibility and promise of sEMG-based motion recognition for hand movements.
• Identified specific motion patterns with varying recognition accuracy.
• Established sEMG as a potential basis for skill assessment and training guidance.

Conclusions:

• sEMG-based motion recognition is a viable tool for robotic surgery training.
• Further research involving deep learning is needed to improve recognition accuracy for all motion patterns.
• This approach can lead to more effective and objective surgical skill development.