Promoting Neuroplastic Mechanisms With Modeling-based Optimization of a Novel Robot-asisted Walking Rehabilitation Therapy

Patients with spinal cord injuries referred to the robotics service for gait rehabilitation will complete 40 sessions with robotic devices. An initial assessment will determine the most appropriate device. Patients assigned to the Lokomat will undergo 20 sessions, after which they will be reassessed to determine whether to continue with an additional 20 sessions or switch to a different robotic device, based on medical criteria. This study evaluates patient progress during the 40 Lokomat sessions at pre-, mid-, and post-intervention time points. A follow-up assessment will be conducted one month after completion of the Lokomat sessions to evaluate the patient progress without robotic assistance. A descriptive analysis will be performed to characterize baseline demographic data. Descriptive statistics will also be applied to the following outcome measure groups: functional assessment, robotic gait trainer variables, pain and fatigue scales, and the therapy acceptance questionnaire. The relationship between motor evoked potential (MEP) responses, lower extremity motor scores (LEMS), and force production capacity estimated using a musculoskeletal model will be investigated. MEP responses will be classified into three groups based on responsiveness (normal, delayed, and absent), and LEMS scores will be classified into four groups. A nonparametric Spearman correlation coefficient will be used to assess the correlation between MEP responses and motor scores. Nonparametric Kruskal-Wallis tests will evaluate whether the mean response differs across the four LEMS groups. For variables that do not meet the assumptions of normality for one-way analysis of variance (ANOVA), Mann-Whitney tests with Bonferroni correction will be used as a post hoc test to determine differences in responsiveness between groups. P values \<0.05 will be considered statistically significant.