"Stroke affects 1.1 million Europeans every year, and a majority of stroke survivors show deficits in arm function 6 more after stroke. However, current motor therapy treatments are both expensive and relatively ineffective.The objectives of this IIF proposal are: 1. To develop and validate a multiple-time scale computational model of stroke recovery based on a scientific understanding of neural plasticity and motor learning after brain injury, and 2. To develop a novel adaptive, personalized, and low cost motor trainer to improve function of the upper extremity after stroke based on predictions of long-term recovery from the predictive model.The uniqueness of this proposal lies in the care that will be taken to ensure that the models of upper limb recovery are valid, relevant, and practically useful.The predictive model and the adaptive trainer will be developed in close collaboration with the RE-ARM research group of the M2H laboratory at the University of Montpellier, whose members have the rare opportunity to collect a very large amount of repeated arm movement data, as well as functional and neurological data from 50 patients in the acute phase after stroke using a novel serious game for stroke rehabilitation via an ongoing clinical trial conducted at the Physical and Rehabilitation Medicine departments from Montpellier’s and Nîmes university hospitals. Additional collaborations withe a top French Informatics and Robotics lab, the LIRMM, and the recently created Laboratory of Excellence NUMEV will ensure the success of the proposed project."