A Personalized Multimodal BCI-Soft Robotics System for Rehabilitating Upper Limb Function in Chronic Stroke Patients.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Medicine

Background:

• Multimodal brain-computer interfaces (BCIs) combining electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) offer enhanced performance potential.
• Personalized rehabilitation strategies are crucial for optimizing patient outcomes in stroke recovery.
• Respiration can confound functional near-infrared spectroscopy (fNIRS) signals, impacting brain-computer interface (BCI) data accuracy.

Purpose of the Study:

• To develop and evaluate a personalized multimodal EEG- and fNIRS-based BCI system with soft robotic (BCI-SR) components for stroke rehabilitation.
• To investigate the impact of synchronizing motor imagery (MI) cues with the respiratory cycle on BCI performance.
• To assess the efficacy of the personalized BCI-SR system in improving upper-extremity motor function in chronic stroke patients.

Main Methods:

• Collected multimodal EEG and fNIRS data during motor imagery tasks from healthy participants.
• Incorporated a breathing sensor for respiration synchronization (RS) of motor imagery cues.
• Conducted a 6-week clinical trial with four chronic stroke patients using the personalized BCI-SR system, measuring upper-extremity motor function with Fugl-Meyer Assessment (FMA) and Action Research Arm Test (ARAT).

Main Results:

• Respiration synchronization (RS) reduced variability in oxyhemoglobin (HbO) readings.
• Demonstrated significant improvements in FMA and ARAT scores post-rehabilitation compared to baseline.
• Observed striking coherence in EEG and fNIRS activation patterns across all stroke patients.

Conclusions:

• Personalized rehabilitation treatment using the multimodal BCI-SR system enhances BCI performance and motor recovery in stroke patients.
• Synchronizing motor imagery cues to respiration improves the consistency of hemodynamic signals, leading to better motor imagery performance.
• The proposed multimodal BCI-SR system shows promise for promoting neuroplasticity and improving motor function in stroke survivors.