Brain Computer Interface based robotic rehabilitation with online modification of task speed

Area of Science:

• Rehabilitation Robotics
• Neuroscience
• Human-Computer Interaction

Background:

• Robot-assisted therapy offers rehabilitation benefits but often lacks patient engagement.
• Continuous monitoring of patient intention is crucial for adaptive therapeutic interventions.

Purpose of the Study:

• To develop a systematic approach for online adaptation of robot-assisted rehabilitation exercises.
• To enable patients to control exercise speed using their motor imagery intentions via a BCI.

Main Methods:

• Utilized an electroencephalogram (EEG) based Brain-Computer Interface (BCI) for continuous intention monitoring.
• Employed Linear Discriminant Analysis (LDA) to classify motor imagery patterns (ERS/ERD) and used posterior probabilities as continuous control signals.
• Implemented a Passive Velocity Field Control (PVFC) strategy to map motor imagery levels to robot-assisted exercise speed.

Main Results:

• Demonstrated a framework for online adaptation of rehabilitation exercises based on patient intention.
• Feasibility studies with healthy volunteers confirmed the effectiveness of the proposed approach.
• The system allowed patients to actively control the speed of contour following tasks.

Conclusions:

• The proposed BCI-controlled robot-assisted rehabilitation system enhances patient involvement by providing online control over exercise speed.
• This approach has the potential to significantly increase the efficacy of robot-assisted therapies.
• Continuous adaptation of rehabilitation exercises based on real-time intention monitoring is feasible and beneficial.