Early after stroke, patients often have significant motor impairment and sensory deficit. Evidence has demonstrated heightened plasticity and significant recovery in the acute phase (first months) post stroke but there has been a lack of effective and practical protocols and devices for early intensive sensorimotor therapy.This research study will conduct a randomized clinical trial of an intensive motor-sensory rehabilitation on patients with acute stroke using a wearable rehabilitation robot. The primary aims are to facilitate sensorimotor recovery, reduce ankle impairments, and improve balance and gait functions. This clinical trial will be conducted on the Study and Control groups of acute stroke survivors.

The study will investigate an early intensive rehabilitation in acute stroke for motor relearning, reducing ankle impairments and improving balance and mobility/locomotion functions.

The acute stroke survivor will be randomly placed into two groups. Subjects in the Study group will receive robot-aided motor relearning under real-time feedback, stretching under intelligent control, sensory stimulation, and active movement training with interactive games. Subjects in the Control group will receive passive movement in the middle ROM without intelligent stretching and active movement training without robotic guidance.

For both groups, the therapeutic training will be conducted during 5 hourly sessions (including breaks/transitions between tasks) each week over about 3-week hospital stay. Both groups will also receive the standard of care in the hospital and rehabilitation service. Treatment outcome measures will be obtained through blinded assessments and evaluated before and after training involving biomechanical, neuromuscular and clinical outcome measures. Carry-over effects will be further evaluated 1 month after the treatment ends.

Aim 1: To evaluate biomechanical and neuromuscular changes as defined by the passive and active range of motion (ROM), flexor-extensor muscle strength, joint stiffness, proprioception and reflex excitability, and compare these measures between the two groups. The biomechanical and neuromuscular outcome measures will be obtained through blinded assessments and evaluated before and after training using the wearable rehabilitation robot.

Hypothesis 1: Robot-guided motor relearning, stretching and active movement training (Study group) will improve the biomechanical and neuromuscular outcome measures more than those of the Control group.

Aim 2: To evaluate the clinical outcome measures as defined by Fugl-Meyer score (lower extremity), modified Ashworth scale, Berg balance scale, 10 meter walk test, and to compare between the Study and Control groups.

Hypothesis 2: The Study group will improve the clinical outcome measures more than the Control group.

Inclusion Criteria:

1. First time unilateral acute stroke, hemorrhagic or ischemic, 24 hours after admission in hospital to 1 year post stroke.
2. Hemiplegia or hemiparesis
3. Age 18-80
4. Ankle impairments

Exclusion Criteria:

1. No impairment or very mild ankle impairment of ankle.
2. Unstable medical conditions that interferes with ability to training and exercise.
3. Severe cardiovascular disorders that interfere with ability to perform moderate movement exercises.
4. Cognitive impairment or aphasia with inability to follow instructions
5. Pressure ulcer, recent surgical incision or active skin disease with open wounds present below knee of treated limb
6. Severe pain in legs