Neural Network-Based Muscle Torque Estimation Using Mechanomyography During Electrically-Evoked Knee Extension and Standing in Spinal Cord Injury

Frontiers in Neurorobotics, 2018 · DOI: 10.3389/fnbot.2018.00050 · Published: August 10, 2018

Spinal Cord Injury Neurorehabilitation Bioinformatics

Simple Explanation

This study explores using artificial neural networks (ANN) and mechanomyography (MMG) to monitor muscle torque, especially when direct measurement is difficult. MMG signals from the quadriceps muscles were used to estimate knee torque during functional electrical stimulation (FES)-assisted exercises in people with spinal cord injuries (SCI). The ANN models developed could estimate muscle torque in real-time, potentially improving the safety of automated FES control for standing in SCI individuals.

Study Duration

Not specified

Participants

3 SCI individuals for ANN design, 5 SCI individuals for standing protocol

Evidence Level

Not specified

Key Findings

1
ANN models using MMG data (RMS and RMS-ZC) can predict knee extension torque during FES with reasonable accuracy.
2
The average correlation between predicted and actual torque during knee extension was 0.87 ± 0.11 for the RMS model and 0.84 ± 0.13 for the RMS-ZC model.
3
Both models identified a critical point around a 50% torque drop, indicating significant changes in MMG patterns.

Research Summary

This study designed an artificial neural network (ANN) to predict torque exerted around the knee joint by the quadriceps muscle using mechanomyography (MMG) parameters during FES isometric knee extension and standing in SCI individuals. Two ANN models were developed based on different inputs: Root mean square (RMS) MMG and RMS-Zero crossing (ZC). The performance of the ANN was evaluated by comparing model predicted torque against the actual torque derived from the dynamometer. The developed ANN models could be used to estimate muscle torque in real-time, thereby providing safer automated FES control of standing in persons with motor-complete SCI.

Practical Implications

Real-time Torque Monitoring

The ANN models offer a way to monitor muscle torque in real-time, which is valuable when direct measurement is not feasible.

Safer FES Control

The ability to estimate torque can lead to safer and more effective automated FES control, particularly for standing.

Personalized Rehabilitation

By understanding individual muscle performance through torque estimation, rehabilitation programs can be tailored for better outcomes.

Study Limitations

1
The ANN model was analyzed only during quiet standing and isometric knee extension.
2
Future studies should include a wider movement pattern such as sit-to-stand movement, which is another nonmeasurable knee torque movement.
3
This research also focused on a speciﬁc set of parameters for the FES.

Your Feedback

Was this summary helpful?

Back to Spinal Cord Injury