Periodic Reporting for period 2 - REBoT (Re-Empower the BOdy after Tetraplegia.) Reporting period: 2020-02-01 to 2021-01-31 Summary of the context and overall objectives of the project Loss of arm and hand control is one of the most invalidating consequences of high-level injuries to the spinal cord. Limb paralysis and sensory deprivation, over time, lead to progressive and permanent structural and functional changes in the Central Nervous System that can hinder recovery of upper limb function. One way to counteract this phenomenon is to engage muscles in skilled activities and skill learning in order to promote use-dependent plasticity. However, the effectiveness and specificity of these techniques are still limited by our poor understanding of the neural mechanisms underlying the recovery of function.The REBoT project aims at harnessing neuroplasticity by means of therapeutic interventions to promote functional recovery of the upper limb after cervical spinal cord injury (SCI). The ambitious, overarching goal is to promote a “rewiring” of the nervous system to bypass pathways interrupted by SCI through a progressive adaptation of the rehabilitative intervention.The leading hypothesis is that engaging the upper body in the skilled control of personalized physical or virtual interfaces brings about plastic changes in the sensorimotor pathways that can be exploited to design more effective and tailored assistive devices and neuro-prostheses. These, in turn, will contribute increasing the independence and therefore the quality of life after SCI.The work in REBoT capitalizes on the concept of personalized body-machine interfaces as tools to leverage residual arms and shoulders mobility with the dual objective of (i) strengthening residual voluntary control of the upper limb, and (ii) promoting specific functional and structural changes in the nervous system. Results from our studies indicated that wearable personalized interfaces allow delivering a customized intervention that supports both assistive and rehabilitative goals through the integration of multimodal signals from the user’s body. Our preliminary investigations suggest that control through wearable interfaces is not limited to low dimensional systems but can easily extend to complex, multi-articulated machines. We also demonstrated that interface co-adaptation can be a tool to increase user-interface interaction efficiency and to intuitively tailor the interface to their users’ unique abilities and needs. Moreover, results suggest that human-machine integration has the potential to drive neural adaptation and advocates for more studies to help characterizing the time-course and functional relevance of the changes induced by prolonged interface use.Overall, the results of the project call for a multidisciplinary evaluation of technologies applied to individuals with disabilities that encompasses the specific user’s impairment and the behavioral and neural factors supporting its operation. Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far The work conducted in REBoT followed three main streams with the goal of i) training upper-limb movements to promote residual voluntary control of the muscles, ii) establishing a correspondence between training of specific goal-directed movements and the resulting reorganization in cortical, spinal and supra-spinal structures of the nervous system, and iii) enabling control of haptic assistants and complex robotic interfaces.1) We explored how to leverage residual motor abilities through tailoring of a body-machine interface (BMI): i) we developed and validated a novel way to integrate multiple body signals into the operation of the BMI with the goal of strengthening voluntary control over specific muscles; ii) we designed and tested a novel algorithm for online co-adaptation that exploits the interconnection between the BMI and the user to track or to drive functional improvements. The studies informed the development of a general mathematical framework to study and simulate learning in a variety of human-machine coadaptation scenarios.2) We studied the relationship between learning induced by the interaction with a BMI and brain plasticity in proximal arm muscles evaluated by non-invasive brain stimulation in individuals with and without cervical spinal cord injury. BMI-specific changes were identified in both uninjured and SCI volunteers, pointing at cortical involvement during the initial stages of BMI operation. We conducted preliminary investigations to assess the applicability and potential benefits of high-density EMG recording for increasing the sensitivity of recording motor-evoked potentials.3) We extended the state-of-the-art implementation of the BMI to operate robotic systems and virtual and physical assistive devices with multiple degrees of freedom: i) we developed and tested a BMI decoder to allow controlling a 7-link robotic arm in a simultaneous and continuous way; ii) we exploited the interaction between the interface and a wrist exoskeleton to study the efficacy of visual feedback versus proprioceptive feedback when learning to operate the interface. Results suggest that proprioceptive feedback can teach efficient control strategies more effectively than visual feedback.The results of REBoT’s investigations have been published in several journal papers and conference proceedings, disseminated through presentation at various international scientific conferences, and received a best paper award recognition. REBoT’s concept and results have been presented both to other scientific events and to a broader non-specialized audience through two publicly available interviews. Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far) Positive results in animal models point to pharmacological treatments designed to promote axonal growth and to the combination of pharmacological cocktails and electrical stimulation as the new avenue for the restoration of the damaged connections in the spinal cord. However, one of the key challenges that hinder the translation to human research is how to drive the regeneration of functional connections. Currently, training is regarded as the ultimate key for translating drug-induced plasticity into recovery, but the design of effective treatment combinations suffers from the lack of insight into the mechanisms of plasticity. REBoT aims at responding to this compelling need by leveraging residual motor abilities of the individuals through tailoring of the body-machine interface with the goal of both strengthening residual voluntary control and promoting CNS plasticity. The results of the project will serve to drive the design of novel adaptive neuroprostheses and human-machine interfaces to seamlessly integrate with and match the user’s specific level of disability, while also promoting functional recovery.Wearable body-machine interfaces provide a viable way for individuals to access the technology outside of the laboratory environment. These interfaces employ a technology that can be rather inexpensive and therefore financially accessible to a broad body of potential users. Hence, the extreme flexibility of design and application of body-machine interfaces together with their potential wide availability and affordability advocate for their competitive advantage in supporting personalized therapy and assistance for people with impaired mobility. In addition to contributing increasing the quality of life of individuals with disabilities, the results of this research can be exploited well beyond the medical field into the world of artificial intelligence for human-machine interfaces and contactless control of devices in our daily activities. A summary of REBoT's objectives
