ArmAssist is low cost simply robotic devise for arm training and assessment. It is suitable for extended home rehabilitation and prolonged hospital rehabilitation without therapist supervision. The device contains several games which are structured as exercise for disabled patients. The primary aim of this study is to correlate efficacy of the arm training with ArmAssist robot system to conventional care modified by duration and structure of exercise. The secondary aim of this study is to measure the correlation between standard clinical assessment scales and ArmAssist-based assessment metrics.

Inclusion Criteria:

• unilateral paresis;
• ability to understand and follow simple instructions;
• minimum ability to perform active movements, even though trunk compensation, using the shoulder and/or the elbow joints.

Exclusion Criteria:

• bilateral impairment;
• severe sensory deficits in the paretic upper limb;
• cognitive impairment or behavioral dysfunction that would influence the ability to understand or perform the experiment;
• inability to provide informed consent;
• other current severe medical problems.