Clinical predictors, species distribution, and machine learning-based risk stratification of invasive candidiasis and in-hospital mortality: a retrospective episode-level cohort analysis
Source check: Source checked. Not independent scientific confirmation.
Finding: In this 206-episode cohort, invasive candidiasis occurred in 37 episodes (18%); the Random Forest achieved AUROC 0.987 for invasive disease and 0.776 for mortality. The near-perfect invasive-disease discrimination was driven mainly by age and is explicitly described as exploratory. Five-fold stratified cross-validation with imputation applied within folds; no external cohort, patient-level independence or prospective validation. This supports the reported retrospective result only and not clinical deployment.
Technology: Four supervised classifiers (Random Forest, Gradient Boosting, Extra Trees and Logistic Regression) model invasive candidiasis and in-hospital mortality from 24 clinical and microbiological predictors; permutation importance identifies their contribution.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Four supervised classifiers (Random Forest, Gradient Boosting, Extra Trees and Logistic Regression) model invasive candidiasis and in-hospital mortality from 24 clinical and microbiological predictors; permutation importance identifies their contribution.
- Test and result
- In this 206-episode cohort, invasive candidiasis occurred in 37 episodes (18%); the Random Forest achieved AUROC 0.987 for invasive disease and 0.776 for mortality. The near-perfect invasive-disease discrimination was driven mainly by age and is explicitly described as exploratory.
- Limitation / still unknown
- Five-fold stratified cross-validation with imputation applied within folds; no external cohort, patient-level independence or prospective validation. This supports the reported retrospective result only and not clinical deployment.
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/PMC13721703/
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 source full text Methods and Results; version/date checked against bibliographic record..
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.