BCIwiki (bciwiki.com) — According to a paper published by ITM Web of Conferences, Tingxi Zhu published a review in ITM Web of Conferences on 2026-10-15 examining the principles, technologies, applications, and challenges of motor imagery-based brain-computer interfaces (MI-BCIs) in neurorehabilitation. The article discusses rehabilitation applications for motor dysfunction associated with stroke, spinal cord injury, and other neurological diseases.
Motor imagery means imagining movements. MI-BCIs decode associated neural signals to enable direct communication between the brain and external devices. Beginning with the theoretical foundations of BCI and motor imagery, the review covers EEG-based signal acquisition, processing, and classification: recording brain activity, processing the signals, and distinguishing signal categories.
The article focuses on integration with EEG, neurofeedback, functional electrical stimulation, robotic devices, virtual reality, and artificial intelligence. It discusses applications in stroke rehabilitation, spinal cord injury treatment, and virtual reality rehabilitation. This overview of technologies and applications does not establish that all such combinations have undergone equivalent clinical validation.
The author identifies low signal-to-noise ratio, meaning weak signals relative to noise, substantial individual differences, lengthy training, and device capability limitations as continuing challenges. Deep learning, wearable devices, personalized rehabilitation programs, and continued clinical validation are presented as future directions that may advance the use of MI-BCIs in rehabilitation medicine. That expectation should not be treated as an established treatment effect.