Machine learning-driven risk assessment of severe Mycoplasma pneumoniae in children: analysis based on core…

AI / informaticsFrontiers in medicine· 2026

Machine learning-driven risk assessment of severe Mycoplasma pneumoniae in children: analysis based on core clinical and immunological features.

Duoduo Li, Li Wang, Xiaolu Zhao … Fenglian Zhu

Pulmonology & Critical CareAI / informaticsPrognosisQ1IF 3.4

Machine learning model predicts severe Mycoplasma pneumoniae pneumonia in children using admission data.

Machine learning-driven risk assessment of severe Mycoplasma pneumoniae in children: analysis based on core clinical and immunological features.

Duoduo Li … Fenglian Zhu

Frontiers in medicine · 2026 · Q1 in General Medicine

Purpose

To construct and validate an interpretable machine learning model for early prediction of severe Mycoplasma pneumoniae pneumonia (severe Mycoplasma pneumoniae pneumonia) in children.

Methods

We systematically evaluated 113 algorithms based on data collected within 24 h of admission.

n = 402 pediatric MPP patients

Results

higher scores mean better prediction, 1.0 would be perfect

AUC 0.81 (95% CI 0.72 to 0.89)

null 0

0.72

0.89

CI excludes the null - significant

More results

The model identified 16 core predictors: dyspnea, C-reactive protein (c-reactive protein), total T lymphocytes (CD3+), sputum plug formation (PB), CD3+CD4+CD8- T cells, CD3+CD56+NKT cells, LDH, IL-6, creatine kinase (creatine kinase), PCT, ALT, IgA, abnormal coagulation function (D-dimer), creatine kinase-MB, erythrocyte sedimentation rate (erythrocyte sedimentation rate), and MP-DNA load.

More results

SHapley Additive exPlanations (SHAP) analysis revealed that dyspnea, elevated c-reactive protein, and decreased T lymphocytes synergistically drive severe illness risk.

“

Conclusion · 1 of 2

The Gradient Boosting Machine-based model effectively predicts severe Mycoplasma pneumoniae pneumonia risk.

Conclusion · 2 of 2

Pending prospective multi-center validation, we plan to embed this transparent, data-driven tool into electronic medical record (electronic medical record) systems to provide real-time early warning and individualized risk assessment.

Machine learning-driven risk assessment of severe Mycoplasma pneumoniae in children: analysis based on core clinical and immunological features.

Machine learning-driven risk assessment of severe Mycoplasma pneumoniae in children: analysis based on core clinical and immunological features.

Related papers