Pediatric Mycoplasma Pneumoniae Patient Data for Predicting Plastic Bronchitis

307 pediatric patients with Mycoplasma pneumoniae pneumonia (MPP) who underwent bronchoscopy intervention from April 2023 to June 2025. The data was used to develop and evaluate machine learning models for the early prediction of plastic bronchitis (PB). The study, authored by Pei Wang, resulted in a web-based risk predictor deployed online.

Use Cases

• Training binary classification models to predict plastic bronchitis risk based on clinical indices like retinol-binding protein 4 and D-dimer levels.
• Benchmarking model performance of algorithms like XGBoost and logistic regression on a clinical cohort.
• Developing feature importance explanations using SHAP analysis for clinical interpretability.
• Creating clinical decision support tools, such as the provided online risk calculator, for early intervention.

Strengths

• Dataset includes 307 patient records with a defined temporal range from April 2023 to June 2025.
• Model performance is quantitatively reported, with the best model achieving an AUC of 0.905 on the test set.
• Key predictive features are identified and listed, including retinol-binding protein 4 level and fever duration.
• Includes a deployed, freely accessible online calculator for clinical application.

Limitations

• Row count for the underlying feature data is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 28.3 KB, indicating limited scope and likely a feature matrix rather than raw patient records.

Provenance

Source

figshare, author Pei Wang.

Collection Method

Retrospective cohort study of pediatric patients who underwent bronchoscopy.

Time Range

April 2023 to June 2025

Freshness

Last updated 2026-04-23 05:34:38; freshness should be verified.