Differentiating Round Pneumonia from Malignant Lung Masses Using Histogram-Based Image Analysis

INTRODUCTION: Round pneumonia is characterized by round, well-circumscribed infiltrative lesions on computed tomography (CT) that can mimic malignant lung masses. This study aimed to evaluate distinguishing parameters between round pneumonia and primary malignant lung masses using histogram-based image analysis, and to investigate the diagnostic contribution of this method.
METHODS: In this retrospective study, 60 patients (30 with round pneumonia, 30 with primary malignant lung mass) diagnosed between 2020 and 2024 were included. Unenhanced thoracic CT images were analyzed; lesions were assessed by two radiologists, and a circular region of interest (ROI) was selected on an appropriate slice for each lesion.Histogram and texture analysis parameters obtained from these ROIs were compared between the two groups. The Mann-Whitney U test was used for two-group comparisons and Spearman’s rank test for correlation (p < 0.05).
RESULTS: Histogram-based analysis revealed many parameters with statistically significant differences between round pneumonia and malignant lung masses. In particular, features such as standard deviation, variance, entropy, range, interquartile range, and contrast provided significant separation between the two groups (p < 0.001). On receiver operating characteristic (ROC) analysis, histogram variance achieved the highest diagnostic performance, distinguishing pneumonia vs. tumor with 100% sensitivity and 100% specificity (AUC = 1.00, threshold ≈ 1.91 × 10³ HU²). No significant correlation was found between the standardized uptake value (SUV) and any histogram parameter in the tumor group.
DISCUSSION AND CONCLUSION: CT histogram analysis is a useful non-invasive tool for distinguishing round pneumonia from malignant lung lesions with similar imaging features, potentially reducing unnecessary invasive procedures.

Keywords: CT histogram analysis, Lung cancer, Round pneumonia