💻 Research Focus

Our lab develops non-invasive BCI systems that translate brain activity into real-time control commands for assistive technologies and rehabilitation devices. We focus on motor imagery (MI)-based BCIs, EEG-EMG hybrid interfaces, and closed-loop neurofeedback systems to enable intention-driven neurorehabilitation for stroke and other neurological conditions.

We aim to build clinically viable BCIs that are accurate, robust, and usable even by patients with severe motor impairment.

Background

A Brain-Computer Interface (BCI) is a system that enables direct communication between the brain and an external device, bypassing the need for muscular output. BCIs are especially promising for restoring motor function in patients with paralysis due to stroke, spinal cord injury, or neurodegenerative disease.

Most BCI systems use electroencephalography (EEG) to decode neural signals related to motor intention, such as motor imagery. However, current challenges include:

• Low signal-to-noise ratio in EEG

• Inter-subject variability

• Difficulty achieving continuous, smooth control

To address these, we develop:

• Hybrid BCI systems combining EEG with electromyography (EMG) to enhance decoding reliability

• Adaptive machine learning algorithms to personalize control systems

• Real-time feedback paradigms that engage users in motor learning

We also integrate BCIs with rehabilitation robots and neuromodulation systems for closed-loop therapy.

Key Research Directions

• Real-time classification of motor imagery and hand gestures using EEG and EMG

• Transfer learning and adaptive models for robust cross-user BCI systems

• EEG-EMG coherence analysis to assess motor intent and brain-muscle coupling

• Integration of BCI with robotic assistive devices and soft wearable exoskeletons

• Development of BCI-based rehabilitation games and GUIs for user engagement