Design and Implementation of a Personalized FES Protocol Toward Dexterous Hand Movements Rehabilitation

A significant number of pathologies, such as stroke or spinal cord injury, can lead to severe motor impairments and paralysis, drastically affecting a person's quality of life and autonomy. One of the most widely used rehabilitation techniques is Functional Electrical Stimulation (FES), a treatment that induces muscle contraction by delivering low-energy electrical pulses to the muscles through non-invasive electrodes. This approach, which has also been proven effective in improving muscle strength, alleviating pain, and reducing spasticity, enables patients to restore physiological muscle activity by leveraging the neuroplasticity of the central nervous system. Despite extensive research over the past decades, applying FES to recover dexterous hand movements remains challenging. While surface electrodes allow for a simple and non-invasive approach, they limit the muscle selectivity of stimulation, which is crucial given the complexity of the muscles involved in hand movements. Moreover, the high inter-subject variability in response to electrical stimulation highlights the need for customized solutions. In this thesis project, an experimental protocol to assess the feasibility of stimulating dexterous hand gestures and software to efficiently control the electrical stimulator were developed. A first experimental campaign was conducted on thirteen participants to evaluate the possibility of stimulating individual finger movements selectively. Specifically, eight movements were examined: thumb opposition and extension, flexion and extension of the index and middle fingers, and combined flexion and extension of the ring and little fingers. In this phase, particular attention was given to the customization of the stimulation, in terms of electrode placement and stimulation parameters, to enhance both movement effectiveness and subject comfort. Subsequently, in a separate session, a second test was carried out on five of the initial subjects to stimulate dexterous hand gestures. Seven specific hand movements were targeted: One, Two, Three, Four, Hand Open, Hand Close, and Thumb Up. This phase explored the feasibility of eliciting more complex hand movements by combining the finger motions identified in the first phase and allowed to assess the consistency of the stimulation parameters across sessions. The results demonstrated that the proposed method allows for selective finger stimulation. Indeed, 69.23% of the participants exhibited at least five of the eight targeted motions. Considering also movements that were not entirely isolated or very limited, 76.92% of the participants achieved at least six out of eight motions, with three participants displaying all the targeted finger movements. The extension movements were more frequently elicited and better isolated than flexion movements, during which wrist flexion was often inadvertently elicited. Regarding the thumb, both opposition and extension were observed in all subjects, though extension was frequently limited. The final part of the protocol confirmed that achieving complex gestures by combining finger movements is feasible. However, the quality of the final gesture strongly depends on the precision of the individual finger movements and the use of multiple stimulation channels could result in unwanted wrist motions, resulting in suboptimal movements. With further optimization, the proposed approach could contribute to a more effective and personalized therapy for hand motion recovery, ultimately improving patients' quality of life.