ZEPU-AI3 Lower Limb Feedback Robot Training Study

Background:

Lower limb motor dysfunction is a major cause of disability following neurological and musculoskeletal disorders. Conventional rehabilitation improves function but is often limited by therapist workload and the difficulty of delivering high-intensity, repetitive, task-specific training. Robotic rehabilitation systems have the potential to overcome these limitations by providing standardized, intensive, and feedback-guided exercise.

Objective:

To evaluate the safety, efficacy, and feasibility of the ZEPU-AI3 Lower Limb Feedback Training and Evaluation Robot in patients with lower limb motor dysfunction.

Study Design:

This is a prospective, single-center, pilot randomized controlled trial conducted at the Department of Physical Medicine and Rehabilitation, Bangladesh Medical University. Eligible participants will be randomly allocated in a 1:1 ratio to receive either ZEPU-AI3-assisted rehabilitation plus conventional rehabilitation or conventional rehabilitation alone.

Intervention:

Participants in the intervention group will receive robot-assisted lower limb feedback training using the ZEPU-AI3 device in addition to conventional rehabilitation. The control group will receive conventional rehabilitation according to institutional standards. The intervention duration, frequency, and intensity will follow the approved study protocol.

Outcome Measures:

Primary and secondary outcomes include safety, feasibility, lower limb motor recovery, gait performance, balance, functional mobility, and adverse events. Participants will be assessed at baseline and after completion of the intervention.

Significance:

This pilot study will generate preliminary evidence regarding the clinical utility, safety, and feasibility of the ZEPU-AI3 robotic rehabilitation system and will inform the design of future multicenter randomized controlled trials.

Inclusion criteria

• • Lower-limb motor dysfunction from one of the following: (a) stroke (onset 2-24 months), (b) incomplete spinal-cord injury (neurological level T12 and below, ASIA C or D), (c) orthopedic surgery (e.g., knee/hip replacement) with persistent gait impairment > 3month post-surgery. Stroke onset between 2-24 months ensures inclusion of individuals in the subacute to chronic phase, where gait recovery is still achievable and measurable. Incomplete SCI targets individuals with partial motor preservation, who are capable of engaging in active gait training with robotic assistance.

• Age >18 and ≤ 60 years. This age limit is chosen because individuals with this age range generally have better cardiopulmonary reserve and musculoskeletal tolerance, allowing safer participation in intensive robotic-assisted training with reduced risk of adverse events. Excluding younger patients minimizes heterogeneity related to growth, neurodevelopmental factors and congenital disorders, which may influence gait mechanics and response to robotic therapy. Excluding older adults (>60 years) helps reduce confounding from age-related degenerative changes that may affect gait outcomes and safety.
• Able to provide informed consent and understand instructions which ensures participants can actively engage in therapy, follow safety instructions and report adverse events.
• Weight ≤ 100 kg (as per device spec) and height within device adjustable range (manufacturer spec). Exceeding weight or height limits may compromise mechanical support, safety, and accurate gait training.
• Medically stable and cleared by physician for exoskeleton use (no active infection, uncontrolled cardiac/respiratory disease, severe osteoporosis, uncontrolled epilepsy, untreated DVT) through pre-screening with necessary investigations.

Exclusion criteria

• • Complete spinal cord injury with inability to bear any weight as it requires the patient to actively support some body weight or participate in stepping movements.

• Severe cognitive impairment (e.g., MMSE < 24) preventing safe participation.
• Severe spasticity (Modified Ashworth Scale > 3) in lower limbs. High muscle tone or rigidity increases risk of joint strain, skin injury and falls during robotic-assisted training.
• Unstable fractures, severe hip/knee contractures (> 30° fixed flexion), severe osteoarthritis requiring imminent surgery. They can pose high risk for injury during weight-bearing or gait cycles.
• Uncontrolled cardiac arrhythmia, pacemaker incompatibility or implanted electronic device incompatible with robot sensors. Individuals with these conditions may face elevated risk of adverse cardiac events or device malfunction.
• Pregnancy as safety data for robotic exoskeleton use in pregnant individuals are lacking.