Trunk Acceleration for Neuroprosthetic Control of Standing – a Pilot Study

J Appl Biomech, 2012 · DOI: · Published: February 1, 2012

Spinal Cord Injury Neurology Biomechanics

Simple Explanation

This study explores using trunk acceleration to control balance in standing with a neuroprosthesis after paralysis. It uses artificial neural networks (ANNs) to predict changes in center of pressure (COP) based on trunk acceleration in able-bodied subjects. The ANN was trained to use trunk acceleration to predict COP changes. This ANN then drove feedback control of ankle muscles in a computer model of someone using a standing neuroprosthesis. The feedback control, driven by the ANN, reduced upper-body loading during postural disturbances compared to using constant muscle stimulation.

Study Duration

Not specified

Participants

Three able-bodied male volunteers (age 21 – 26)

Evidence Level

Pilot Study

Key Findings

1
ANNs can be trained to predict changes in COP from trunk acceleration with correlation coefficients ranging from 0.67 to 0.77.
2
Feedback control using an ANN reduced average upper-body loading during perturbation onset and recovery by 42% compared to optimal, constant excitation.
3
Feedback control also reduced peak upper-body loading by 29% compared to optimal, constant excitation.

Research Summary

This pilot study investigates the use of trunk acceleration feedback control of center of pressure (COP) against postural disturbances in standing with a neuroprosthesis following paralysis. Artificial neural networks (ANNs) were trained to predict changes in COP using three-dimensional trunk acceleration as input from able-bodied subjects during bipedal stance perturbations. The results suggest that 3D acceleration may be an effective parameter for deploying control action to maintain standing under perturbation, and the ANN mapping of trunk acceleration input to COP output was effectively exploited for feedback control.

Practical Implications

Potential for Improved Neuroprosthetic Control

Trunk acceleration can be used as a feedback signal to improve control of standing neuroprostheses.

Reduced Reliance on Upper-Body Support

Feedback control based on trunk acceleration can reduce the need for upper-body support during standing.

Future Research Directions

Future studies should focus on developing comprehensive, SCI user-specific control structures and implementing them in a laboratory setting.

Study Limitations

1
Reliance on only ΔCOP for control of standing balance.
2
Utilization of able-bodied volunteers to represent individuals with SCI.
3
The ANN models may need to be adjusted for individual SCI patients to avoid triggering spastic reflexes.

Your Feedback

Was this summary helpful?

Back to Spinal Cord Injury