Development and validation of venous thrombosis risk prediction model and scale for lung cancer patients in intensive care unit based on interpretable machine learning.
Source check: Source checked. Not independent scientific confirmation.
Finding: An interpretable model predicted venous-thrombosis risk among lung-cancer patients in ICU.
Technology: Boruta, variance-inflation analysis and LASSO screened predictors; nine models, including CatBoost, estimated ICU venous-thromboembolism risk, with SHAP interpretation and a score derived from the best model.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Boruta, variance-inflation analysis and LASSO screened predictors; nine models, including CatBoost, estimated ICU venous-thromboembolism risk, with SHAP interpretation and a score derived from the best model.
- Test and result
- In 3,694 ICU records across MIMIC-IV, eICU and NWICU, CatBoost AUCs were 0.7648 in training, 0.6437 in internal validation, 0.6822 in eICU and 0.7153 in NWICU. This retrospective prediction does not establish clinical benefit.
- Limitation / still unknown
- The cohort is drawn from existing ICU databases; reported validation AUCs are modest (0.64–0.72), and no prospective clinical-impact evaluation is reported.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC13559455/ · https://europepmc.org/article/MED/42724662
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.