Chronic Nonhealing Wound Infection Prediction Model with External Validation

A predictive model for infection risk in chronic nonhealing wounds developed using electronic medical record data. The model was built on a primary cohort of 500 patients from two tertiary hospitals in China and validated on an external cohort of 300 patients. It was authored by Yang Jiang and last updated in May 2026.

Use Cases

• Predict infection risk in chronic wounds based on six clinical predictors identified in the study
• Validate a machine learning model's performance using AUROC and calibration metrics described in the results
• Interpret model predictions using SHAP analysis to understand feature importance as mentioned in the methodology
• Implement a clinical decision support tool using the deployed web calculator referenced in the conclusion

Strengths

• Model was developed on a primary cohort of 500 patients and validated on an external cohort of 300 patients
• Performance metrics are reported, including AUROC of 0.884 in testing and 0.855 in external validation
• Six consensus predictors were identified, enhancing model interpretability

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count for the underlying data is unknown, which may limit suitability assessment
• Data may reflect geographic bias inherent to the two tertiary hospitals in China

Provenance

Source

Two tertiary hospitals in China

Collection Method

Two-center retrospective study using routine electronic medical record data

Freshness

Last updated 2026-05-19 05:44:09; freshness should be verified

Geography

China