Decoding Sensorimotor Rhythms during Robotic-Assisted Treadmill Walking for Brain Computer Interface (BCI) Applications

PLoS ONE, 2015 · DOI: 10.1371/journal.pone.0137910 · Published: December 16, 2015

Assistive Technology Neurology Neurorehabilitation

Simple Explanation

This study explores using brain signals to control robotic devices that help people walk, particularly those recovering from strokes. The research investigates if it's possible to decode a person's intention to walk from their brain activity while using a robot-assisted treadmill. EEG was used to measure brain activity in healthy volunteers and stroke patients during both active and passive robot-assisted walking. The goal was to see if a computer could accurately distinguish between these states and baseline (rest). The results showed that it is possible to decode walking intention from brain signals, which suggests that BCI systems could be used to control robotic gait-training devices. These findings support the development of new rehabilitation strategies for individuals with walking disorders.

Study Duration

Not specified

Participants

10 healthy volunteers and three acute stroke patients

Evidence Level

Not specified

Key Findings

1
Classification accuracies when distinguishing walking from baseline were high, both in healthy participants (above 93%) and stroke patients (above 89%), indicating the feasibility of BCI-based robotic-assisted training devices.
2
Modulation of high beta and low gamma activity during the gait cycle was observed in healthy volunteers, but not in the stroke patients, suggesting sensorimotor integration deficits in stroke patients may affect these oscillations.
3
The classifier's weights indicated that event-related desynchronization (ERD) in the mu rhythm (bilaterally) and ERD in the beta and low gamma bands (centro-medially) contributed to the high classification performance.

Research Summary

This study aimed to demonstrate the feasibility of using EEG-based BCI to control robot-assisted gait devices by decoding walking intention during robot-assisted gait training in healthy volunteers and stroke patients. The results showed high classification accuracies in distinguishing walking from baseline in both groups, suggesting the potential for BCI-based robotic rehabilitation. Differences in brain activity and muscle activation patterns between active and passive walking were also explored, revealing insights into cortical involvement during gait and potential limitations for BCI control.

Practical Implications

BCI-Controlled Gait Training

The findings support the development of BCI systems for controlling robot-assisted gait training, potentially enhancing rehabilitation outcomes for stroke patients.

Biomarker for Motor Recovery

Modulation of low gamma frequencies during the gait cycle could be explored as a potential biomarker for assessing motor recovery in future studies.

Personalized Rehabilitation Strategies

Understanding the cortical involvement during active and passive walking can contribute to personalized rehabilitation strategies tailored to individual patient needs.

Study Limitations

1
The study included a small sample size of stroke patients, limiting the generalizability of the findings.
2
Despite applying CCA, remaining sources of unwanted artifacts cannot be entirely excluded and therefore they could have potentially contributed to the classification between passive and active walking
3
The accuracy to detect intention or movement by noninvasive brain signals can be limited.

Your Feedback

Was this summary helpful?

Back to Assistive Technology