The primary objective of this clinical trial is to investigate the efficacy of motor imagery-based brain-computer interface (MI-BCI) technology in improving motor function among patients with spinal cord injury (SCI), as well as its impact on cortical motor area function across varying states. To achieve this, the study will implement MI-BCI intervention in SCI patients, evaluate post-treatment motor function improvements, and assess changes in cortical motor area oxygen metabolism (via functional near-infrared spectroscopy, fNIRS) and neural activity (via electroencephalography, EEG). The ultimate goal is to establish a novel rehabilitation strategy for SCI.

Specifically, the trial aims to: (1) determine whether MI-BCI effectively enhances motor function in SCI patients; and (2) clarify the differential effects of MI-BCI on cortical motor area function under distinct states (e.g., resting vs. task-performing) in this population.

Participants will be randomly assigned to one of two groups: the experimental group will undergo MI-BCI training, while the control group will receive active cycling training (as a conventional rehabilitation control). Both interventions will be structured as 20-minute sessions, administered 5 days per week, over a total of 4 weeks.Pre- and post-treatment assessments will include: lower limb motor function (measured by the Lower Limb Motor Score), activities of daily living (evaluated via the Modified Barthel Index), walking capacity (quantified using the Spinal Cord Injury Walking Index), and cortical motor activity (captured through fNIRS and EEG measurements).

Spinal cord injury (SCI) is a severe neurological disorder caused by traumatic events (e.g., motor vehicle collisions or falls) or non-traumatic factors related to diseases (e.g., cancer or infections). Patients with SCI primarily exhibit sensory and motor dysfunction below the injury level, clinically presenting as complete injuries, though pathologically, most cases are incomplete. A foreign study analyzing early mortality in 67 patients with clinically complete SCI revealed that 50 cases had partial continuity of central nervous system tissue at the injured segment, indicating that clinical completeness does not necessarily equate to complete severance of descending motor neurons in the spinal cord. Histological analysis of SCI centers also showed that 62% of specimens maintained central nervous system continuity. Therefore, developing interventions that more effectively promote neural circuit reconstruction to further enhance functional recovery in SCI patients is crucial.

SCI has a high global incidence and is on the rise; in 2019, over 20 million people worldwide were living with SCI, with approximately 900,000 new cases. From 1990 to 2019, the prevalence of SCI increased by 81.5%, and the incidence rose by 52.7%. Another epidemiological study noted an annual incidence of approximately 25-29 cases per million people over the past decade, with traumatic SCI caused by road traffic accidents and falls being the most common, particularly among young males.

Post-SCI motor dysfunction is one of the most critical factors affecting patients' independence. For patients with complete injuries or complete motor deficits, current approaches lack effective means to improve motor function. In recent years, brain-computer interface (BCI) technology has emerged as a potentially effective rehabilitation tool. BCI establishes a direct, real-time connection between the brain and external devices, enabling human-machine interaction without relying on peripheral nerves or muscles.

Motor imagery (MI) refers to the conscious mental simulation of specific body movements without actual physical execution. MI has demonstrated significant potential in motor skill learning and rehabilitation. When patients imagine performing a movement, the corresponding brain motor function areas are activated, enhancing central nervous system plasticity and promoting functional reorganization. Through repeated practice, this imaginative process can generate signals from the central nervous system that stimulate cortical and peripheral nerves, ultimately eliciting actual peripheral muscle movement. Motor imagery-based BCI (MI-BCI) technology translates subjects' motor imagery into commands to control external devices (e.g., robots), enabling actual movement. Even when limb function is impaired, MI-BCI can establish a closed-loop linkage system between the brain and limbs. Additionally, MI-BCI provides feedback through touch, vision, and proprioception, forming an active central-peripheral-central closed-loop control system. This design not only maximizes patient engagement but also enhances neural plasticity and improves motor function.

The efficacy of MI-BCI is based on specific electroencephalographic (EEG) mechanisms. During MI or actual movement, the EEG rhythm energy in the contralateral sensorimotor cortex significantly decreases, while that in the ipsilateral sensorimotor cortex markedly increases-particularly in the α-band (8-12 Hz) and β-band (18-25 Hz). This phenomenon, known as event-related desynchronization (ERD) and event-related synchronization (ERS), is particularly prominent.

Current research on MI-BCI for motor function recovery after neurological injury primarily focuses on stroke patients with hemiplegia. In recent years, MI-BCI training has also shown great potential in SCI rehabilitation. Studies have explored the combined effects of assisted motor training and MI-BCI in complete SCI patients, finding that sustained assisted motor training can induce partial recovery of sensory and motor function in chronic SCI patients, with MI-BCI further promoting neural recovery. Other research has demonstrated the efficacy of MI-BCI in neural remodeling and motor function recovery in SCI animal models (e.g., rats and non-human primates). However, the role of BCI in promoting motor function recovery in SCI patients remains underexplored, and no studies have yet reported whether MI-BCI can enhance SCI patients' motor function by strengthening intracerebral neural networks.

Functional magnetic resonance imaging (fMRI), EEG, and functional near-infrared spectroscopy (fNIRS) are commonly used techniques for assessing brain function. fMRI has limitations, including low temporal resolution, slow data acquisition, the need for patients to remain motionless in a confined space for extended periods, numerous contraindications, and relatively difficult data collection. EEG, generated by the synchronous summation of postsynaptic potentials from large numbers of cortical neurons, reflects the activity of multiple neurons and is categorized into four rhythm bands (δ, θ, α, and β) based on frequency. Due to its non-invasive data acquisition and relatively simple signal interpretation, EEG is one of the most widely used neuroimaging techniques.

fNIRS monitors real-time changes in the concentrations of oxygenated and deoxygenated hemoglobin in the cerebral cortex under different stimulation tasks, indirectly reflecting neural activity. Compared to fMRI, fNIRS enables real-time monitoring in natural environments, is easy to use, offers high temporal resolution, and provides better spatial resolution with minimal impact from head movement. Thus, fNIRS holds significant potential for neuroregulation-based rehabilitation. MI-BCI training may increase cortical activation in the supplementary motor area (SMA) and primary motor cortex (M1). fNIRS studies have shown that MI-BCI improves functional connectivity between the motor cortex and prefrontal cortex and enhances ERD.

To further clarify whether MI-BCI technology can effectively improve motor function in SCI patients and determine its impact on cortical motor area function under different states, this study aims to apply MI-BCI technology to SCI patients, evaluate post-treatment improvements in motor function, and assess changes in oxygen metabolism and EEG signals in the resting and task states of the cortical motor area using fNIRS and EEG. The goal is to provide a novel therapeutic approach for motor function recovery in SCI patients.

Research process: According to the inclusion and exclusion criteria, if a patient meets the enrollment conditions of this trial and agrees to participate in this study, after signing the informed consent form, they will start to enter this study. The following outcomes will be evaluated at baseline, 2 weeks, and 4 weeks post-treatment:1. Lower limb motor score.2. Modified Barthel Index.3. Spinal Cord Injury Walking Index.4. fNIRS.5.EEG.

All baseline and post-treatment outcome assessments (at 2 and 4 weeks) will be performed by a dedicated rehabilitation physician blinded to the study protocol and group assignments. The therapists administering daily conventional rehabilitation training will also be unaware of group allocations. Eligible participants will be assigned to either the experimental or control group:Group A (MI-BCI Group) and Group B (Cycling Group).

1. Group A (MI-BCI): Participants will use an EEG-based rehabilitation training device (L-B300). After wearing an EEG cap and lying supine, their lower limbs will be positioned on a pedal with the hip, knee, and ankle joints of the unaffected (stronger) leg maintained at 90°. The leg support will be adjusted to the mid-calf and secured with straps. Participants will be instructed to focus on a screen and perform MI training. If they fail to activate the device within 30 seconds under guidance, the EEG difficulty level will be reduced until successful activation within 30 seconds. The EEG difficulty level will be adjusted throughout the trial based on the participant's MI ability.
2. Group B (Cycling): Participants will lie supine, with their lower limbs fixed to the L-B300 device using the same method as Group A. The computer screen will be turned off, and the machine will be set to drive passive or active cycling movements of the lower limbs.Both groups will undergo 20-minute sessions, 5 days/week, for 4 weeks (total of 20 sessions).

Inclusion Criteria:

1. The vital signs are stable and the spine is stable, making the subject suitable for exercise testing.
2. Patients with spinal cord injury (SCI) who meet the international diagnostic criteria for SCI neurology revised by the American SCI Society in 2019 and have been diagnosed by CT or MRI.
3. The injury level of SCI is C5-T12, and the ASIA grade is A-C.
4. The course of the disease is ≤12 months (but the spinal shock period must have passed).
5. Age: 18-75 years old, regardless of gender.
6. Good cognitive function, able to understand and actively participate in the training program, and willing to sign the informed consent form for this clinical study.

Exclusion Criteria:

1. Those with tumors, tuberculosis, hematologic diseases, or dysfunction of important organs such as the heart and liver;
2. Those with unstable fractures;
3. Those with severe abnormal limb muscle tone and joint contracture deformities;
4. Those with severe pain that cannot tolerate activities;
5. Those with severe emotional problems who cannot cooperate to complete the study.

Study data will be made available under reasonable request after publication. Data will include de-identified participant data and the data dictionary. Requests can be submitted to the corresponding author. Request will be analyzed and ethical and legal implications of data sharing will be considered. Data will be shared after consent of study participants

After publication under reasonable request.