Assisting post-stroke grasp & release.

Rationale:

The majority of stroke patients have to cope with impaired arm and hand function after a stroke. Post stroke rehabilitation training aims to regain (partly) lost functions by stimulation of restoration or promoting compensational strategies, in order to increase the level of independence. During rehabilitation training movements are practiced preferably with high intensity, in a task-oriented way, with an active contribution of the stroke survivor in a motivating environment. An effective training modality that is commonly applied in post stroke upper extremity rehabilitation training is arm support by means of gravity compensation. In order to ensure active participation of the patient, movement is best supported only then when needed. In addition, to increase functional abilities of the affected arm, hand function should also be trained. A promising technique to train hand function after stroke is electrical stimulation of wrist and finger extensors and thumb muscles. By the application of electrode arrays in combination with self learning algorithms, the possibility of more selective stimulation becomes available. This raises the question whether it is possible to assist the fine motor control of the hand with such electrodes. In addition, array electrodes provide the possibility of quick donning and doffing, by using self calibrating algorithms, automatically selecting the best electrode for the task which should be assisted.



Objective:

The primary objective of the present study is to evaluate whether the applied self-learning control algorithms for robotics and electrical stimulation result in improved grasping of selected objects of the ARA test. Improvement will be measured by the ARAT score (0-3). These selected objects are: 10 cm wooden cube, 2.5 cm wooden cube, 6 mm ball bearing and 1.5 cm marble.



Study population:

10 healthy elderly and 10 stroke patients for each session, not necessarily the same subjects for each session, so a maximum of 30 healthy elderly and 40 stroke patients will be included.



Intervention:

In this study subjects receive electrical stimulation of thumb and finger muscles. This electrical stimulation is applied by different control strategies. Each time a movement is performed, movement errors are evaluated and based on that, the stimuation parameters (position, amplitude, duration) are altered in order to minimize these errors. Thus the algorithms are self-learning and learn how to stimulate based on the performance during the repeated movements. In addition, during T3 and T4 arm movement is supported by a robotic manipulator. This manipulator applies forces to the arm in order to move the arm in towards the target objects.



Main study parameters/endpoints:

The main study parameter will be ARAT scores for the selected ARAT objects (10cm cube, 2,5 cm cube, 6mm ball bearing and 1,5 cm marble). The endpoint of the study is an effective control algorithm, which is able to assist stroke patients to successfully grasp these selected ARAT objects.

Functional grasping of selected objects from the ARA test (10cm cube, 2.5 cm cube, 6 mm ball bearing and 1,5cm marble) can be improved by combining robotics and electrical stimulation by means of effective self-learning control algorithms.

The experiment consist of four measurements, spaced approximately three months apart.

Patients will receive different types of single and multichannel electrical stimulation with and without robotic assistance.



The electrical stimulation is applied by array electrodes. The stimulation parameters are based on the performance in the previous trial, which is evaluated based on the measured movement errors. Stimulation parameters (location, amplitude and duration) are updated in an iterative process.



The instantaneous influence on functional hand opening and reaching of these types of electrical stimulation is addressed by the ability to grasp selected objects from the ARA test.