Task-oriented repetitive movements can improve motor performance in patients with neurological or orthopaedic lesions. The application of robotics and automation technology can serve to assist, enhance, evaluate, and document neurological and orthopedic rehabilitation. This paper deals with the application of "patient-cooperative" techniques to robot-aided gait rehabilitation of neurological disorders. We define patient-cooperative to mean that, during movement, the technical system takes into account the patient's intention and voluntary efforts rather than imposing any predefined movements or inflexible strategies. It is hypothesized that such cooperative robotic approaches can improve the therapeutic outcome compared to classical rehabilitation strategies. New cooperative strategies are presented that detect the patient's voluntary efforts. First, this enables the patient increased freedom of movement by a certain amount of robot compliance. Second, the robot behavior adapts to the existing voluntary motor abilities. And third, the robotic system displays and improves the patient contribution by visual biofeedback. Initial experimental results are presented to evaluate the basic principle and technical function of proposed approaches. Further improvements of the technical design and additional clinical testing is required to prove whether the therapeutic outcome can be enhanced by such cooperative strategies.

In clinical practice the dominant view is that the signs of exaggerated tendon tap reflexes associated with muscle hypertonia are responsible for the spastic movement disorder. Consequently, most anti-spastic treatments are directed at reducing reflex activity. During the last years an increasing body of evidence suggests a discrepancy between clinical spasticity and spastic movement disorder.This is primarily due to the different role reflexes play in the passive and active condition, respectively. Today we know that a central motor lesion is associated with a loss of supraspinal drive and a defective utilization of afferent input with an impaired behaviour of short-and longlatency reflexes. This leads to a paresis and a mal-adaptation of the movement pattern.Secondary changes in mechanical muscle fibre, collagen tissue and tendon properties (e.g. loss of sarcomers; sub-clinical contractures) result in spastic muscle tone, which at part compensates for paresis. This allows functional movements on a simpler level of organisation. Anti-spastic drugs can accentuate paresis and therefore should be applied with caution in mobile subjects.

The aim of this study was to differentiate the effects of body load and joint movements on the leg muscle activation pattern during assisted locomotion in spinal man. Stepping movements were induced by a driven gait orthosis (DGO) on a treadmill in patients with complete para-/tetraplegia and, for comparison, in healthy subjects. All subjects were unloaded by 70% of their body weight. EMG of upper and lower leg muscles and joint movements of the DGO of both legs were recorded. In the patients, normal stepping movements and those mainly restricted to the hips (blocked knees) were associated with a pattern of leg muscle EMG activity that corresponded to that of the healthy subjects, but the amplitude was smaller. Locomotor movements restricted to imposed ankle joint movements were followed by no, or only focal EMG responses in the stretched muscles. Unilateral locomotion in the patients was associated with a normal pattern of leg muscle EMG activity restricted to the moving side, while in the healthy subjects a bilateral activation occurred. This indicates that interlimb coordination depends on a supraspinal input. During locomotion with 100% body unloading in healthy subjects and patients, no EMG activity was present. Thus, it can be concluded that afferent input from hip joints, in combination with that from load receptors, plays a crucial role in the generation of locomotor activity in the isolated human spinal cord. This is in line with observations from infant stepping experiments and experiments in cats. Afferent feedback from knee and ankle joints may be involved largely in the control of focal movements.

No abstract

1Background: Arm hemiparesis, secondary to stroke, is common and disabling. The robot ARMin, 2