The article presents a method for obtaining quantitative signs of pulmonary pathology, based on the analysis of medical imaging computed tomography. These signs may be suitable for the differential diagnosis of spherical lung formations. In particular, the possibility of diagnosing cancer and pulmonary tuberculosis was investigated by quantitative statistical characteristics of images of spherical lung formations. Two of these characteristics (the densitometric value and its standard deviation) can be called classical, since modern basic software of computer tomographs can easily calculate them. In addition, characteristics were considered, for which it is necessary to carry out additional calculations using fractal analysis. The article presents the results of a study of the relationship of statistical and fractal characteristics of a tomographic image with the type of pathology in the lungs. The results of a statistical analysis of the fractal dimension of the region of interest and its related parameters are presented, and signs are found suitable for the diagnosis of pulmonary pathologies.