Final Report Summary - REHABNET (REHABNET: NEUROSCIENCE BASED INTERACTIVE SYSTEMS FOR MOTOR REHABILITATION)
The result of the RehabNet project is an integrative platform for neuroscientists, engineers and clinicians to further study stroke recovery and improve the impact of rehabilitation strategies. During the project, we have developed 4 novel rehabilitation scenarios: (1) a bimanual motor training, (2) a dual motor cognitive-motor training for attention and memory, (3) a simulated city for the training of Activities of Daily Living in an ecologically valid context, and (4) a Motor Imagery based brain computer interface (BCI) system that combines VR with EEG based neurofeedback for motor rehabilitation. All scenarios are implemented with state of the art game engines, are platform independent and most of them are freely accessible through a web browser or as an app.
Recent meta-analysis of VR studies in stroke rehabilitation comparing the impact of virtual reality indicate that current VR based interventions directly leave out patients exhibiting no active movement, and that most exclude patients with very low muscle strength, arm control, with spasticity or perceptual or cognitive dysfunction. The RehabNet project has broadened modern VR rehabilitation approaches to (1) include those patients with worse prognostic (motor and cognitive) through an accessible and interface independent architecture; (2) provide very low-cost at-home rehabilitation solutions by making available all developed technology for free to the community; and (3) we have brought new insights on the impact and use of VR technologies for motor and cognitive rehabilitation, including the use of neurofeedback BCI systems.
In collaboration with clinical centers of the region of Madeira and mainland Portugal, we realized pilot and longitudinal studies to evaluate RehabNet and its clinical impact. We studied how we can reduce cognitive-motor interference in rehabilitation via the appropriate selection of interfaces. We parametrized cognitive training through computational modelling to enable the precise manipulation of cognitive demands required in our rehabilitation systems. We showed that VR combined with BCI is able to recruit motor areas to a larger extent by means of a dual motor training and motor imagery paradigm, as well as through motor priming. Finally, clinical studies revealed increased benefits of longitudinal VR training in ecologically valid environments when compared to standard of care.
With respect to the state-of-the art and prospects of research career development, the fellow passed from a temporary research position, to invited assistant professor with teaching commitments in the educational programs coordinated by M-ITI, and recently earned a tenure track assistant professor position at the same institution. The position encompasses a mixed profile with research, educational and administrative responsibilities. Additionally, the fellow has successfully consolidated a research group, the NeuroRehabLab that currently has 15+ members (2 faculty, 6 PhD students, 2 technical assistants and several graduate and undergraduate students).
To conclude, the RehabNet project brought further insights through the above mentioned advancement of the state of the art; has contributed to founding the “Open Rehab Initiative”, which provides a free novel worldwide available toolset comprising multimodal sensing technologies and game training scenarios for at-home use; we are currently negotiating a technology transfer agreement to study the future exploitation of the project results; and the Marie Curie fellow successfully transitioned to a stable position at the hosting institution.