Revolutionary AI-Powered Multiple Sclerosis Progression Prediction

Background: A Critical Healthcare Challenge

Multiple sclerosis affects approximately 2.8 million individuals worldwide. MS typically begins as Relapsing-Remitting MS (RRMS), characterized by episodes of neurological symptoms followed by periods of recovery. Over time, approximately 50% of RRMS patients transition to Secondary Progressive MS (SPMS), marked by gradual worsening without distinct relapses—often diagnosed 3+ years after progression begins.

Disease severity is measured using the Expanded Disability Status Scale (EDSS), a 0-10 scale where higher scores indicate greater disability. EDSS 4.0 marks a clinically meaningful threshold—the transition from fully ambulatory to mobility-limited status, when intervention decisions become critical. Current clinical practice relies primarily on EDSS and MRI-based assessments, yet these measures often detect pathological changes only after irreversible neurodegeneration has occurred.

The need for accurate, early prognostic tools has never been more urgent. Motor Evoked Potentials (MEPs) offer a promising solution: they provide quantitative measures of corticospinal tract integrity and can detect subclinical damage before symptoms manifest. However, extracting meaningful predictive patterns from longitudinal MEP data has historically required analysis beyond conventional clinical methods.

Our Solution: Ensemble Machine Learning for Clinical-Grade Progression Prediction

We've developed MS-PROGRESS, an advanced machine learning system that analyzes longitudinal MEP recordings to predict which patients will progress to EDSS ≥4.0 with unprecedented accuracy. Our technology achieves AUC 0.80—outperforming both the MSBase benchmark (0.71) and prior MEP-based approaches (0.75)—while identifying a 6-fold difference in progression rates between low-risk and high-risk patients.

This isn't incremental improvement. Our ensemble approach processes MEP data across multiple clinic visits (up to 16 per patient), capturing disease trajectory rather than isolated snapshots. The system analyzes bilateral upper and lower extremity recordings jointly, preserving topographic relationships in corticospinal function. A key discovery: amplitude features reflecting axonal loss proved more predictive than latency features reflecting demyelination—suggesting irreversible axonal damage drives the critical EDSS 4.0 transition.

Why This Matters: Real-World Impact

For Healthcare Providers:

  • Early risk stratification identifying high-risk patients (52.8% progression rate) versus low-risk patients (8.5%) before clinical decline

  • Non-invasive assessment using standard MEP equipment already available in most MS centers

  • Objective measurements that complement clinical judgment and support treatment escalation decisions

  • Deployable anywhere via portable MEP equipment with zero cloud dependency—solar-powered operation enables assessment in remote clinics without grid access

For Patients and Families:

  • Earlier intervention when disease-modifying therapies are most effective at preserving function

  • Informed decision-making with objective data to guide treatment intensity discussions

  • Reduced uncertainty through quantifiable risk assessment rather than "wait and see" approaches

  • Proactive planning enabling lifestyle and care adjustments before mobility limitations develop