This study investigated the value of chest CT-derived body composition parameters and clinical risk factors in predicting viral pneumonia severity. Patients were categorized into severe and non-severe groups, and a subgroup analysis of 30-day adverse outcomes was performed in the severe group. Although both the clinical and imaging models demonstrated good predictive performance, the combined model showed improved accuracy, as evidenced by an increase in the AUC from 0.796 and 0.735 to 0.843.
