The effectiveness of robotic over conventional therapy is arguable and the best therapy strategy is still not clear. Basing on the existing evidence on motor learning and practice-induced neuroplasticity, the investigators plan to use an exoskeleton robotic hand to enhance the rehabilitation volume in the subacute stage of stroke patients. The robotic system allows intensive and customized training of hand and finger movements. The aim of this study is (1) to establish whether robot assisted therapy provides any additional motor recovery for the hand when administered during the subacute stage in a Chinese adult population diagnosed with stroke; (2) to evaluate the feasibility and efficacy of robot-assisted hand rehabilitation in improving arm function abilities in subacute hemiplegic patients.

Introduction The number of incident strokes, prevalent stroke survivors, disability-adjusted life-years (DALYs) lost due to stroke, and stroke-related deaths is increasing in the word (1). Stroke is the most common cause of complex disability in Taiwan (2). After the acute phase of stroke, many patients are left with impairment of upper-limb movement (hemiparesis) because of the hand weakness and abnormal contractions. Motor recovery of the hand is the slowest and most difficult, leading to limited hand activities and occupational disability. Therefore, facilitating motor recovery of hand after stroke is crucial in stroke rehabilitation.

The mechanisms of recovery after stroke are multifactorial and the effect of rehabilitation programs is complex (3). Activity-dependent neural plasticity of the cortical maps adjacent to the lesion probably occurs, particularly during the acute period after stroke (4). In order to stimulate such plasticity, many new rehabilitation methods, including rehabilitation robots, have been developed according to the principles of motor learning (5). Robotic systems can provide repetitive, reproducible, interactive forms of physical therapy that can be quantified (6). The advantages of using robots in neuromotor rehabilitation includes favoring attention and reducing the effort of the patient during training (7), boosting motivation and adherence to treatment (8), as well as help in multi-sensory and sensorimotor integration (9). Some results are very promising, showing that robot-assisted therapy is safe and well tolerated and that it has a positive impact on muscle strength and function in the paretic arm (10-12). However, the quality of these evidence is still controversial and inconclusive. The effectiveness of robotic over conventional therapy is arguable and the best therapy strategy is still not clear. Furthermore, there is little understanding of the neurological mechanisms involved in functional recovery of the hand (13). In this study, the investigators hypothesize that the robot based assistance would outperform conventional therapy during the subacute stage of stroke.

Study objective

1. To establish whether robot assisted therapy provides any additional motor recovery for the hand when administered during the subacute stage in a Chinese adult population diagnosed with stroke.
2. To evaluate the feasibility and efficacy of robot-assisted hand rehabilitation in improving arm function abilities in subacute hemiplegic patients.

Inclusion Criteria:

• Adult patients (>20 years old) with a diagnosis of hemorrhagic or ischemic stroke and who experience severe upper extremity hemiparesis.

Exclusion Criteria:

• severe pain and instability in the wrist of the affected arm, severe cognitive impairment, aphasia, hemispatial neglect, apraxia and joint contractures greater than 20 degrees in the affected hand.