Using machine learning algorithms to predict MACE in peritoneal dialysis patients.
Source check: Source checked. Not independent scientific confirmation.
Finding: Machine-learning models predicted major adverse cardiac events in a retrospective peritoneal-dialysis cohort and a separate hospital cohort. The study reports development AUCs of 0.80 overall, 0.86 at one year, and 0.75 at five years, with external-cohort AUC 0.768; its broad composite outcome and retrospective design do not establish intervention benefit.
Technology: XGBoost, random forest, and AdaBoost rank baseline clinical variables and predict overall, one-year, and five-year MACE risk.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Machine-learning models estimate MACE risk from baseline clinical factors.
- Test and result
- Among 1,006 patients, 606 formed the development cohort and 400 an external hospital cohort; reported external-validation AUC was 0.768.
- Limitation / still unknown
- Retrospective data and a composite including cardiovascular hospitalization and all-cause mortality; prospective implementation and patient benefit were not tested.
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/PMC13039696/ · https://www.nature.com/articles/s41598-026-45362-y
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 and reported validation scope checked..
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.