Machine learning-based identification of inflammatory biomarkers for predicting pulmonary consolidation in children with Chlamydia pneumoniae infection - Takeaways - MDSpire

Machine learning-based identification of inflammatory biomarkers for predicting pulmonary consolidation in children with Chlamydia pneumoniae infection

  • By

  • Qianqian Dai

  • Zhiyuan Wang

  • Junlin Zhao

  • Yanan Wang

  • Menghua Li

  • Aliya Maimaitiniyazi

  • Xueli Wang

  • Jianjiang Cui

  • Zhenzhen Guo

  • Shengmeng Qu

  • Wen Zhao

  • Liang Ru

  • May 4, 2026

  • 0 min

Share

  • 1

    This study identified LDH, CRP, and ESR as key inflammatory biomarkers for predicting pulmonary consolidation in children with C. pneumoniae infection.

  • 2

    Machine learning algorithms, including LASSO and Random Forest, were utilized to select core predictive factors from a cohort of 42 pediatric patients.

  • 3

    K-means clustering analysis stratified patients into high and low inflammation groups, revealing significant differences in consolidation rates.

  • 4

    An online risk assessment system was developed, demonstrating excellent predictive performance with an AUC of 0.993 and high sensitivity and specificity.

  • 5

    The findings support the clinical utility of routine laboratory parameters in early identification of high-risk pediatric patients for tailored treatment.

Original Source(s)

Related Content