Beyond SOFA and APACHE II, Novel Risk Stratification Models Using Readily Available Biomarkers in Critical Care.
Source check: Source checked. Not independent scientific confirmation.
Finding: A 19,720-patient ICU cohort produced a logistic risk model using biomarker ratios, ventilation and CRRT that internally discriminated in-hospital mortality (AUC 0.929), comparable to SOFA (0.931); the paper does not establish superiority or external generalization. The reviewed source supports this bounded result; external generalization, causal effects, and sustained operational impact remain unestablished where the source reports retrospective or internal evaluation.
Technology: XGBoost/SHAP ranks predictors before logistic mortality models; the enabled outcome is ICU mortality risk stratification.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- XGBoost/SHAP ranks predictors before logistic mortality models; the enabled outcome is ICU mortality risk stratification.
- Test and result
- The primary Results report 666 deaths and Model 3 (LAR, NPAR, mechanical ventilation, CRRT) AUC 0.929 (95% CI 0.921–0.937); SOFA AUC 0.931 and DeLong p=0.052. This supports a bounded internally tested clinical prediction result, not claimed improvement over SOFA.
- Limitation / still unknown
- Primary Methods describe repeated 70/30 train/test splits; Results also report 10-fold internal cross-validation. No external cohort is reported, and the split descriptions are not fully consistent; conclusions are limited to internal retrospective performance.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://europepmc.org/article/MED/40361939 · https://pmc.ncbi.nlm.nih.gov/articles/PMC12071242/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.