Recent work on large cohorts of chronic stroke (>6 months post-stroke) have shown that intensive training of the upper limb in the chronic stroke patients can lead to substantial motor and functional gains that are maintained at 6 months post intervention.

A very prolonged (12 weeks) and very intensive (5 hours daily) training applied to chronic patients after stroke brings a substantial gain both motor and functional which is maintained at 3 months post intervention.

Robotic rehabilitation have been shown to be as effective as any other treatment used in rehabilitation. But the methods of implementation remain widely debated. At that time, most robotic therapies have tried to reproduce functional movement mainly pointing objects.

We want to demonstrate that analytic movements of the elbow and the shoulder performed with the Luna-EMG robot can replace part of usual physiotherapy treatment with at least the same effectiveness on the recovery of fluid movements of the upper limb after a stroke.

This is a pilot study following a single case experimental design with multiple baselines across subjects (MBD).

People included in the protocol will benefit from 2 hours of daily treatment during 6 to 7.5 weeks. The first 2 weeks will be dedicated to baseline measurement of their upper limb (UL) status with daily treatment for balance and gait control without any treatment of the upper limb. The rest of the time will be dedicated to upper limb treatment exclusively. It will be divided into 2 UL training periods, one of them including a 30mn treatment with the luna EMG robot. The duration of second period will be randomly defined between 2 to 3.5 weeks and the last period will last 2 weeks. The treatment during the first UL treatment period will be randomly allocated to Luna EMG treatment or not. The next period will thus follow the inverse scheme (without if the first contains Luna EMG treatment). The fluidity will be assessed through pointing movements assessed every two days. Evaluation of motor deficiency (Fugl Meyer) and functional status (ARAT score) of the UL will be weekly assessed after the inclusion visit. 3D motion analysis of the UL will also be weekly performed.

All patients will benefit from a regular follow-up and from an organized care protocol. The sessions will be intensive, all subjects will be able to benefit from the new treatment (LUNA EMG). We can expect a direct individual benefit from the proposed intensive treatment, at least for precision of UL movements.

Inclusion Criteria:

• Age >= 18 years
• Hemiparesis after a first stroke with more than 1 year time interval
• Functional upper limb: ability to do a pointing task at 90% of the hand-acromion distance (free trunk).
• Spastic hypertonia of the elbow flexors 0-2/4 on the Modified Ashworth scale
• Patient affiliated with social security
• Patient having signed a consent to participate in the research

Exclusion Criteria:

• Inability to sign written consent
• Flexed elbow > 30° during passive mobilization of the upper limb
• EVA>3 in the upper limb at rest or during mobilization
• Complete loss of upper limb proprioception
• Neuro-orthopedic surgery of the upper limb for spasticity: neurotomy, tenotomy, transfer
• Pregnant or breastfeeding woman

All collected data that underlie results in a publication

Data will be available beginning five years and ending fifteen years following the final study report completion

Data access requests will be reviewed by the internal committee of Brest UH. Requestors will be required to sign and complete a data access agreement.

Assessors will not know what phase the patient is in at the time of assessment or what experimental design the patient was randomized to.