TITLE:
Recent Advances in the Study of Chest CT Features and Clinically Relevant Factors in Paediatric Pneumonia Caused by Different Pathogenic Microorganisms
AUTHORS:
Jian Cao, Qiang Fu
KEYWORDS:
Paediatric Pneumonia, Pathogenic Microorganisms, Chest CT, Clinical Features, Radiomics
JOURNAL NAME:
Journal of Biosciences and Medicines,
Vol.14 No.8,
August
26,
2026
ABSTRACT: Community-acquired pneumonia in children is one of the leading causes of hospitalisation and mortality among children worldwide. Pneumonias induced by various pathogenic microorganisms exhibit distinct clinical manifestations and imaging characteristics. While specific diagnostic features may be present, overlapping symptoms are frequently observed. Chest computed tomography (CT) provides a detailed visualization of the subtle structures within pulmonary parenchymal lesions, and when integrated with clinical factors, it enhances the early differentiation of pathogens. This article provides a comprehensive summary of the chest CT characteristics associated with bacterial pneumonia, viral pneumonia, and Mycoplasma pneumoniae pneumonia. It further analyzes the correlation patterns between CT findings and clinical indicators, and investigates the supplementary value of integrating these indicators for early etiological differential diagnosis. Numerous studies have indicated that bacterial pneumonia is primarily characterized by lobar consolidation, often associated with significantly elevated levels of C-reactive protein (CRP) and procalcitonin (PCT). However, it is important to acknowledge that these biomarkers are not specific to particular pathogens, and their concentrations can be influenced by the disease state, timing of sample collection, and various other factors; viral pneumonia is characterised by diffuse ground-glass opacities and small airway changes, and is commonly seen in infants and young children; Mycoplasma pneumoniae pneumonia is characterised by bronchial wall thickening, the “tree bud” sign, and a dissociation between symptoms and physical signs, and is most common in school-aged children. Integrating CT features with clinical factors can improve the accuracy of pathogen differentiation. Radiomics and machine learning offer new approaches to pathogen prediction; however, most existing models are based on single-centre, small-sample datasets and lack sufficient generalisation ability. Further validation is required in multi-centre, standardised datasets.