Related Experiment Video
Updated: May 29, 2026
09:14
Surface Electromyographic Biofeedback as a Rehabilitation Tool for Patients with Global Brachial Plexus Injury Receiving Bionic Reconstruction
Published on: September 28, 2019
Target Achievement Control Test: evaluating real-time myoelectric pattern-recognition control of multifunctional
Ann M Simon1, Levi J Hargrove, Blair A Lock
1Center for Bionic Medicine, Rehabilitation Institute of Chicago, Chicago, IL 60611, USA. asimon@ric.org
Journal of Rehabilitation Research and Development
|September 23, 2011
Summary
Myoelectric control systems show high accuracy, but real-time performance is unclear. A new Target Achievement Control Test (TAC Test) reveals 3-DOF classifiers are slower and less efficient for prosthetic arm control.
Area of Science:
- Biomedical Engineering
- Rehabilitation Technology
- Human-Computer Interaction
Background:
- Myoelectric control systems achieve high classification accuracy (~95%) using pattern recognition.
- Translating offline accuracy to real-time closed-loop control performance remains a challenge.
- Existing real-time virtual tests often use oversimplified tasks, ignoring factors like motion speed and unintended movements.
Purpose of the Study:
- To evaluate the real-time closed-loop performance of myoelectric pattern recognition systems.
- To introduce and validate a more challenging virtual test, the Target Achievement Control Test (TAC Test).
- To compare the performance of 1-degree of freedom (DOF) versus 3-DOF classifiers under varying task complexities.
Main Methods:
- Five subjects with transradial amputation participated in the TAC Test.
- Subjects controlled a virtual arm to reach target postures using myoelectric pattern recognition.
- The test involved variations in classifier complexity (1-DOF vs. 3-DOF) and task complexity (one vs. three required motions per posture).
Main Results:
- No significant difference in classification accuracy was found between 1-DOF (97.2% ± 2.0%) and 3-DOF (94.1% ± 3.1%) classifiers (p = 0.14).
- Subjects using the 3-DOF classifier completed 31% fewer trials and took significantly more time.
- Reaching a three-motion posture took 3.6 ± 0.8 times longer than a one-motion posture.
Conclusions:
- Real-time closed-loop performance measures are crucial for evaluating myoelectric control systems.
- The TAC Test provides a more challenging and realistic assessment of real-time pattern recognition performance.
- Task complexity significantly impacts control system efficiency, highlighting the need for optimized control strategies.
