Identification of the Ground-Glass Opacity on Computed Tomography Images of the Lungs Based on the Analysis of Groups of Textural Features

We propose a technology for detecting the ground-glass opacity on computed tomography images of the lungs based on the analysis of groups of textural features. The urgency of the problem is determined by the large number of lung diseases that can be diagnosed using the ground-glass symptom, as well as the ability to speed up the work of a specialist doctor. The method is based on the selection of lung areas on the original images, followed by the formation of an informative space of textural features. The study made it possible to identify the most informative of 5 groups of textural features and to achieve an accuracy of classification of this symptom of 94.7%.