Related Experiment Video
Updated: May 11, 2026
09:42
Motor Imagery Brain-Computer Interface in Rehabilitation of Upper Limb Motor Dysfunction After Stroke
Published on: September 1, 2023
Neuromuscular electrical stimulation induced brain patterns to decode motor imagery
C Vidaurre1, J Pascual, A Ramos-Murguialday
1Machine Learning Group, Berlin Institute of Technology, Berlin, Germany. carmen.vidaurre@tu-berlin.de
Summary
Neuromuscular electrical stimulation (NMES) patterns can improve brain-computer interface (BCI) efficiency by decoding motor imagery. This method offers a new way to train BCI systems for users with and without motor control difficulties.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Rehabilitation Technology
Background:
- Brain-computer interfaces (BCIs) are crucial for restoring function but often suffer from inefficiency, particularly in patient populations.
- Existing solutions to BCI inefficiency have shown limited success.
- Motor imagery (MI) is a common BCI control paradigm, but its effectiveness can be challenging for some users.
Purpose of the Study:
- To investigate the potential of using afferent patterns induced by neuromuscular electrical stimulation (NMES) to improve BCI performance.
- To explore whether NMES-induced afferent patterns can support the calibration and decoding of motor imagery (MI) in BCI systems.
- To assess the feasibility of this approach for both healthy users and patients with motor impairments.
Main Methods:
- Electroencephalography (EEG) data were recorded from 10 healthy participants during NMES of hands and feet and during MI of the same limbs.
- Features and classifiers were extracted from the EEG data.
- The ability to decode MI using classifiers trained on NMES-induced afferent patterns was evaluated.
Main Results:
- Offline analysis demonstrated successful decoding of MI using a classifier trained on afferent patterns evoked by NMES.
- The NMES-based classifier model was found to be superior to models trained solely on MI data.
- This indicates that afferent patterns from NMES can effectively represent and decode motor intentions.
Conclusions:
- Afferent patterns generated by NMES can effectively support BCI system calibration and facilitate the decoding of motor imagery.
- This novel approach may offer a new method for training sensorimotor rhythm (SMR)-based BCIs, especially for healthy individuals struggling with BCI control.
- It presents a promising alternative for training MI-based BCIs in individuals unable to perform voluntary movements but possessing residual afferent pathways, such as stroke or ALS patients.
