Table 2. Eligibility criteria.
| Category | Inclusion criteria | Exclusion criteria |
|---|---|---|
| Population | Adult patients (≥18 y) undergoing surgery | ―a |
| Predictive tool | ML-based models explicitly developed to predict intraoperative bleeding risk | Models predicting only postoperative bleeding or failing to distinguish intraoperative or postoperative outcomes |
| Outcome reporting | Reported at least one performance metric: area under the curve (AUC), sensitivity, or specificity | ― |
| Study design | Primary research: retrospective or prospective cohort studies, case-control studies | Conference abstracts, reviews, case reports, editorials, letters |
| Publication status | Full text in Chinese or English (including peer-reviewed preprints) | Non–peer-reviewed manuscripts, publications not in Chinese or English |
| Data source | ― | Nonclinical or invalid sources: animal experiments, simulated datasets, nonhospital data |
Not applicable.