Benchmark RadarMS-MLB
MS-MLB evaluates machine learning models for classifying multiple sclerosis versus healthy controls from whole blood RNA expression data (GSE17048). It uses a shared pipeline with nested cross-validation and a holdout set, and reports the MS Research Score composite metric.
- Released
- 2026-08-04
- Readiness
- Runnable
- Primary field
- Health & Life Sciences
Why it matters
Existing MS transcriptomic studies lack reproducible and standardized evaluation. This benchmark provides a leakage-controlled pipeline and an external model submission pathway for comparable research comparison.
Motivation
Multiple sclerosis (MS) is diagnosed through clinical assessment, magnetic resonance imaging, laboratory evidence when appropriate, and exclusion of better explanations.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.