A machine learning-based predictive model for multilobar pulmonary consolidation induced by macrolide-resistant Mycoplasma pneumoniae pneumonia caused by the 23S rRNA A2063G mutation
Machine learning model predicts multilobar pulmonary consolidation in children with macrolide-resistant Mycoplasma pneumoniae, indicating improved treatment strategies.
Key Points
XG-Boost model achieved a ROC curve area of 0.976, indicating high predictive power for pulmonary consolidation.
Key predictors included C-reactive protein and lactate dehydrogenase, enhancing risk assessment in children with pneumonia.
The study analyzed 404 cases between October 2024 and February 2025, focusing on antibiotic resistance in pneumonia.
Data-driven model interpretability is achieved through Sharpley Additive Explanations, supporting clinical decisions.