Granulomatosis with polyangiitis (GPA) is a rare systemic disease characterized by two parallel processes: necrotizing granulomatous inflammation and low-immune vasculitis predominantly affecting small vessels. Differential diagnosis of lung lesions on CT in patients with an established diagnosis of granulomatosis with polyangiitis can be very difficult.Purpose. Developing computed tomography criteria for the differential diagnosis of infiltrative changes in the lungs in patients with GPA and community-acquired bacterial pneumonia.Materials and methods. 67 CT examinations of the chest by 24 patients with verified GPA with infiltrative lung lesions and 36 CT examinations by 30 patients with bacterial pneumonia without concomitant pulmonary pathology, a comparative analysis of the following characteristics of ground glass opacity symptom was performed: “location", “craniocaudal distribution”, “uniformity”, “localization”, “quantity”, “association with consolidation”, “association with pleural effusion”. CT was performed natively on a Toshiba Aquilion Prime CT scanner according to a standard examination protocol with a slice thickness of 1 mm. Statistical processing of the obtained results was carried outusing the software application RStudio, version 1.3.1093 for mac OS (RStudio, PBC). To study the relationships between two categorical variables, the χ2 independence test and the principal component method for categoricalvariables were used.Results. GPA is characterized by multiple bilateral areas of ground glass opacity compaction, often with a central location, without a statistically significant craniocaudal dependence in the lung regions. In community-acquired pneumonia, this sign is more likely to be unilateral with a peripheral location in the lower lobes of the lung. Statistically significant differences in the degree of homogeneity, combination with consolidation, pleural effusion have not been established.Conclusion. CT reveals the characteristic features of the ground glass opacity CT sign in GPA and pneumonia, which, together with clinical and laboratory data, increase the accuracy of radiodiagnosis of these diseases.