Bilateral and monolateral hand amputated subjects su ffer from strong functional de ficits due to their impairment. Surface Electromyography (sEMG) currently gives some control capabilities but these are limited, often not natural and usually require long training times. The application of modern machine learning techniques to analyse sEMG activity related to natural movements seems promising but it is far from practice due to two main aspects. First, the effects of the amputation on the nervous system of the subjects are not fully clear; second, there is a strong lack of accuracy in the movement classi cation accuracy and a few wrong movements can have important negative effects. In recent work, we started to improve this situation through the establishment of a benchmark database of sEMG data for hand movements (htttp://ninapro.hevs.ch/) , which was welcomed with enthusiasm by the scienti c community. Other recent scienti c papers show that the combination of visual data and electromyography can strongly extend the capabilities of dexterous prostheses. With MEGANE PRO, we aim to bring the research in this fi eld to its next step, i.e. to better understand the neurologic and neurocognitive effects of amputation on the persons and to strongly improve robotic prosthesis control possibilities by hand amputated subjects. Thus, this project could improve the state of the art in hand prosthetics and also improve the clinical outcome of the patients (e.g., by respecting the individual phantom limb phenomenology). We aim to pursue this objective along four collaborating approaches that are the sub-projects.