MODERN ASSESSMENT OF PROGRESSION RISK IN INTERSTITIAL LUNG DISEASES (ILD): QUANTITATIVE COMPUTED TOMOGRAPHY (qCT), BIOMARKERS, AND ARTIFICIAL INTELLIGENCE IN CLINICAL PRACTICE

Interstitial lung diseases (ILD) constitute a heterogeneous group of conditions with a complex clinical course, in which accurate prediction of the rate of fibrosis is crucial for optimizing therapy. However, traditional diagnostic methods, which are based on visual assessment of high-resolution computed tomography (HRCT) and pulmonary function tests (PFT), are burdened by significant inter-observer variability and limited sensitivity in detecting early structural changes. This review of the literature from 2010-2026 analyzes the potential of innovative technologies, such as quantitative computed tomography (qCT), radiomics, and artificial intelligence (AI), in the process of risk assessment and monitoring of ILD progression. Scientific evidence indicates that qCT parameters, including normal lung volume percentage (NLV%) and fibrosis volume percentage (FLV%), offer an objective and reproducible measure of disease severity, showing a strong correlation with spirometric parameters (FVC, DLCO) and the risk of death in patients with idiopathic pulmonary fibrosis (IPF) or connective tissue diseases (CTD-ILD). Modern diagnostics increasingly utilize hybrid artificial intelligence models, which achieve an accuracy of 94-95% in differentiating morphological patterns. These methods are complemented by radiomics, which, by extracting hundreds of numerical features related to parenchymal texture heterogeneity, allows for the differentiation of fibrotic from inflammatory patterns with a precision unattainable by the human eye. Despite these promising prospects, the process of implementing these tools into clinical practice faces systemic barriers, such as technical variability among scanners, the lack of standardized image reconstruction protocols, and IT challenges related to integrating algorithms with hospital systems. A key prerequisite for the full implementation of AI and qCT in the care of patients with ILD is the conduct of multicenter prospective studies based on hard endpoints and the development of an international ethical and legal framework regulating responsibility for algorithm-assisted decisions.