• spatiotemporal stimulation
• recovery of reaching and grasping
• cervical epidural electrical stimulation
• spinal cord injury
• stimulation protocols

Recovery of arm and hand movements is the highest priority for people with tetraplegia. Epidural electrical stimulation (EES) and in particular spatiotemporal strategies are showing promising results for the restoration of lower-limb movements in animals and humans after spinal cord injury. Hence, spatiotemporal EES should be exploited even for the recovery of upper-limb functions. On the path towards a clinical translation, our goal is to develop a system for the recovery of 3D reaching and grasping in non-human primates (NHPs). It is a brain-spine interface where a motor predictor is applied to the cortical signals to online decode the intended muscles activity and EES is delivered accordingly through a customized spinal interface. This thesis aimed to design an algorithm for the real-time selection of the parameters of the EES bursts to deliver for engaging the muscles as similarly as possible to the decoded activity. Electromyographic signals acquired during behavioural experiments in one NHP were analysed revealing that natural 3D reaching requires a complex and variable kinetic strategy. Moreover, single-pulse EES experiments during anaesthesia showed the possibility of selectively engaging functionally relevant muscles groups. Given this experimental prove of the effect of different stimulation parameters on the individual muscles, algorithms for the selection of the EES bursts were proposed. Offline tests revealed that with the activation of only one or two electrodes a time it is possible to reproduce the muscular activation pattern characterizing the task. These algorithms were then integrated into a real-time platform that will be used to test the brain-spine interface on injured NHPs.