Feasibility of AI-driven multichannel FES-assisted gait and cycling training in chronic neurological disorders: a case series

Abstract Background Gait impairments following chronic neurological conditions, such as spinal cord injury (SCI), represent a major challenge for rehabilitation, with limited options to sustain functional recovery and patient engagement. Functional electrical stimulation (FES) has shown potential to enhance gait performance, but its integration into routine rehabilitation requires clinically feasible and acceptable solutions. Objective To assess the feasibility, safety, and acceptability of integrating NeuroSkin®, a wearable AI-powered multichannel FES system, into walking rehabilitation programs for individuals with chronic neurological gait impairments, and to explore its potential to improve functional outcomes. Methods This retrospective case series included four individuals (three males, one female; aged 54–78 years) with chronic neurological gait impairments of diverse etiologies (acute disseminated encephalomyelitis, idiopathic transverse myelitis, incomplete spinal cord injury, and polyradiculoneuritis). Participants underwent 20 rehabilitation sessions over 6 weeks, which combined FES-assisted cycling and ambulation with the NeuroSkin device. The system integrates inertial measurement units and plantar pressure sensors to deliver phase-specific stimulation across multiple lower limb muscle groups. Feasibility was assessed through completion rates and adherence to the study protocol. Gait, balance, and spasticity outcomes were recorded before (PRE) and after the intervention (POST) with standardized assessments (10-Meter Walk Test, 2-Minute Walk Test, Berg Balance Scale, Stand Chair Test, Modified Ashworth Scale, SCATS, and Neuropathic Pain Diagnostic Questionnaire). Results All participants completed the intervention, demonstrating good feasibility, tolerance, and adherence. Improvements were observed in walking speed and distance (10mWT and 2MWT), with concomitant improvement in lower limb strength and reductions in spasticity for three patients. These findings support the potential integration of AI-driven multichannel FES into long-term gait rehabilitation programs for individuals with chronic neurological conditions. Conclusions The integration of the NeuroSkin AI-driven FES system into rehabilitation programs for individuals with chronic neurological gait impairments was feasible, well accepted by patients, and safe. Furthermore, while improvements in gait performance were observed, these findings should be interpreted as exploratory and require confirmation in controlled studies.