Blood Biomarker Signature for Distinguishing Pediatric HRSV and Mycoplasma Pneumonia

A five-biomarker signature developed using machine learning to distinguish human respiratory syncytial virus from mycoplasma pneumoniae infections in pediatric community-acquired pneumonia. The model, based on a retrospective cohort study, achieved an AUC-ROC of 0.89. The dataset was published by Xiandan Chen on figshare under a CC-BY-4.0 license.

Use Cases

• Training a diagnostic classifier based on the five identified biomarkers (eosinophilic granulocyte, immunoglobulin A, LDH, β2-microglobulin, AGR).
• Validating machine learning models for distinguishing viral from bacterial pneumonia in pediatric patients.
• Conducting feature importance analysis on clinical blood test results for infectious disease differentiation.

Strengths

• The diagnostic model achieved an AUC-ROC of 0.89 with a 95% CI of 0.85–0.90.
• A specific five-biomarker signature is identified and validated.
• The dataset is openly licensed under CC-BY-4.0.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is small in size (70.9 KB), indicating limited scope.

Provenance

Source

figshare, author Xiandan Chen.

Collection Method

Retrospective cohort study analyzing clinical data and blood samples from pediatric CAP patients.

Freshness

Last updated 2026-06-02 05:40:24; freshness should be verified.