REMAKE

Objectives

The REMAKE project has three General Objectives (GO): Each GO has a number of specific objectives (SO). The GOs are:

GO1. To translate body-derived signals onto BMI commands, encoding subjects’ state, impairment and residual abilities.

We will focus on the upper side of the body and will record electromyographic activities (EMGs), movements and forces. These signals will be used to characterize the stroke survivors’ strategies, as compared to those observed in healthy individuals (SO 1.1), and will be translated into input variables for the BMI (SO 1.2). The expected outcome of SO 1.1 will consist of an experimental protocol and apparatus to extract information about the motion-force strategies adopted by individual stroke survivors. The experimental apparatus will include 2D/3D motion capture, EMG device and force sensor(s). This identification of subjects’ features and potential capabilities will incorporate knowledge about redundancy structure (synergies) as well as information about ranges of motion, force exertion, and active workspace. Comparison with a normative model will allow identifying intact, impaired (silent) and abnormal (compensatory) features. The expected outcome of SO 1.2 will be a large set of BMI input signals encoding the subjects’ status and abilities, to be used for the control of the rehabilitation devices.

GO 2. To design and implement adaptive BMIs for rehabilitation devices, based on the individual characteristic of each subject. We will devise techniques for the automatic definition of a BMI for rehabilitation (rBMI) that translates biological signals (see GO1) into control commands for three types of devices, namely (i) a cursor on the computer screen, (ii) an endpoint robot, and (iii) an powered exoskeleton. In all cases, the rBMIs will have the ability to:

• -  (SO 2.1) Emphasize the synergies that are silent or under-activated during normal non- assisted movements, and strengthening a subject’s weaker muscle combinations by exercising and giving a functional role to the residual abilities that are not normally used in ecological tasks.
• -  (SO 2.2) Induce subjects to explore new strategies by manipulating the map between the body signals and the device control commands.
• -  (SO 2.3) Discourage the compensatory strategies in favour of new and more independent strategies and coordination patterns. 
The expected outcomes will consist of (i) a ‘motor rBMI’ i.e a rBMI suitable for controlling the above mentioned devices and (ii) a set of experimental protocols and pilot trials that demonstrate the effectiveness of the proposed training methods


GO 3. To encode information of the subject’s state of motion and interaction with the environment into appropriate sensory feedback. The purpose of feedback signals is to close the loop, thus making the interaction between the subject and the world bidirectional. The specific objectives will be: (SO 3.1) designing an experimental protocol for investigating which is the most appropriate way to provide feedback information during a rehabilitation treatment. For example, is it better to use redundant or reduced feedback information? Which is the best way to train subjects in the use of different sensory modalities, with particular emphasis on proprioception?; (SO 3.2) designing an experimental protocol for investigating the possibility of substituting or enhancing the loss of perception with different stimuli such as the ones generated by vibration and tactile devices; (SO 3.3) developing a sensory rBMI i.e an interface that encodes information about the subjects’ state and their interaction with the rehabilitation device, maximizing the feedback information that is understandable by the nervous system.