Postherpetic Neuralgia Risk Prediction Model with XGBoost Performance Metrics

627 patient records from a training cohort and 219 from a validation cohort were used to develop a machine learning model for predicting postherpetic neuralgia risk. The study, authored by Xiao Chen and uploaded to figshare in 2026, identified five key risk factors and evaluated ten models. The optimal XGBoost model achieved an AUC of 0.826 in training and 0.840 in external validation.

Use Cases

• Training a risk stratification model based on identified factors like age and pain severity.
• Benchmarking machine learning algorithms for clinical prediction tasks.
• Interpreting model predictions for postherpetic neuralgia using SHAP analysis.
• Validating a predictive model on an independent external cohort of patients.

Strengths

• Model performance is quantified with specific metrics, including AUC of 0.840 in external validation.
• The study uses two distinct cohorts (training and validation) totaling 846 patient records.
• Five clinically accessible risk factors are explicitly identified for model interpretation.

Limitations

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

Provenance

Source

Retrospective analysis of prospective cohorts from two Chinese hospitals.

Collection Method

Data likely gathered from patient medical records for a clinical study.

Time Range

null

Freshness

Last updated 2026-04-16 04:20:25; freshness should be verified.

Geography

Patients from hospitals in Putian and Zhangzhou, China.