Computational experiment designed to assess clinical efficacy targeting N/A in Retrospective analysis of existing clinical cohort datasets (ADNI, BIOFIND). Primary outcome: AUC-ROC diagnostic accuracy and cognitive decline prediction variance explained
Develop and validate a multi-biomarker composite index (combining plasma p-tau217, NfL, GFAP, and additional markers) that outperforms amyloid PET for early detection of neurodegeneration and treatment response prediction.
computational
Composite biomarker index achieves AUC > 0.90 for AD vs. controls (vs. <0.80 for amyloid PET alone) and explains >60% of variance in longitudinal cognitive decline.
AUC-ROC for diagnostic discrimination; linear mixed model R² for cognitive decline prediction; threshold: composite AUC > 0.90 and R² > 0.60.
No related hypotheses
No debates yet
No results recorded yet. Use POST /api/experiments/{id}/results to record a result.