Visual Feedback of Pattern Separability Improves Myoelectric Decoding Performance of Upper Limb Prostheses

2025-05-14

AI Summary

Pattern recognition (PR) control of upper-limb electromyographic (EMG) prostheses is often limited by users’ difficulty in generating highly discriminative EMG signals; conventional static decoders rely on heuristic, trial-and-error boundary tuning, resulting in poor generalizability. This paper introduces Reviewer—a decoder-informed motor training paradigm. It projects EMG signals into the classifier’s decision space via real-time 3D visualization, enabling intuitive assessment of pattern separability. Coupled with an online feedback-driven dynamic boundary update mechanism and Fitts-law-based evaluation tasks, Reviewer enables bidirectional, data-driven co-optimization between user and decoder. In a 10-session training study, the Reviewer group achieved a 27% increase in task completion rate, a 31% improvement in path efficiency, a 39% reduction in overshoot, and a significantly higher information transfer rate compared to a conventional virtual-arm control group.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

State-of-the-art upper limb myoelectric prostheses often use pattern recognition (PR) control systems that translate electromyography (EMG) signals into desired movements. As prosthesis movement complexity increases, users often struggle to produce sufficiently distinct EMG patterns for reliable classification. Existing training typically involves heuristic, trial-and-error user adjustments to static decoder boundaries. Goal: We introduce the Reviewer, a 3D visual interface projecting EMG signals directly into the decoder's classification space, providing intuitive, real-time insight into PR algorithm behavior. This structured feedback reduces cognitive load and fosters mutual, data-driven adaptation between user-generated EMG patterns and decoder boundaries. Methods: A 10-session study with 12 able-bodied participants compared PR performance after motor-based training and updating using the Reviewer versus conventional virtual arm visualization. Performance was assessed using a Fitts law task that involved the aperture of the cursor and the control of orientation. Results: Participants trained with the Reviewer achieved higher completion rates, reduced overshoot, and improved path efficiency and throughput compared to the standard visualization group. Significance: The Reviewer introduces decoder-informed motor training, facilitating immediate and consistent PR-based myoelectric control improvements. By iteratively refining control through real-time feedback, this approach reduces reliance on trial-and-error recalibration, enabling a more adaptive, self-correcting training framework. Conclusion: The 3D visual feedback significantly improves PR control in novice operators through structured training, enabling feedback-driven adaptation and reducing reliance on extensive heuristic adjustments.

Problem

• Improving EMG pattern distinctness for reliable prosthesis control
• Reducing cognitive load in myoelectric prosthesis training
• Enhancing real-time feedback for adaptive PR algorithm adjustment

Innovation

• 3D visual interface projects EMG signals
• Real-time feedback reduces cognitive load
• Data-driven adaptation improves classification accuracy