Predictive Performance of Machine Learning with Evoked Potentials for SCI and MS Prognosis: A Meta-Analysis

Evoked potentials (EPs), including somatosensory evoked potentials (SSEPs) and motor evoked potentials (MEPs), assess neural conduction in spinal cord injury (SCI) and multiple sclerosis (MS), conditions marked by demyelination, inflammation, and axonal damage. Machine learning (ML) enhances EPs’ prognostic utility, but evidence synthesis is limited. This meta-analysis evaluated the predictive performance of EP-based ML models for SCI recovery (ASIA scale) and MS progression (EDSS) using a random-effects model. Six studies (n=560) were included, extracting accuracy and AUC. Pooled results showed high predictive accuracy (79.2%, 95% CI 76.8–81.6%) and AUC (0.82, 95% CI 0.79–0.85). Sensitivity analysis excluding an animal study (n=528) yielded similar accuracy (78.5%, 95% CI 75.9–81.1%) and AUC (0.81, 95% CI 0.78–0.84). SSEP latency and MEP time series were key predictors, with amplitude critical in SCI and multimodal approaches enhancing performance. Moderate heterogeneity (I²=56–62%) and limited studies constrain generalizability. This meta-analysis highlights EPs’ prognostic potential in ML-driven precision neurology, advocating for further human studies to validate multimodal approaches.