Artificial intelligence applications in radiology: the new frontier in interstitial lung diseases.
Authors
Abstract
Interstitial lung disease (ILD) comprises a heterogeneous group of disorders with diverse clinical behaviors, for which early diagnosis, accurate risk stratification, and timely intervention remain challenging. High-resolution computed tomography (HRCT) plays a pivotal role in ILD evaluation; however, conventional visual interpretation is limited by subjectivity and inter-observer variability. Recent advances in artificial intelligence (AI), particularly deep learning, have created new opportunities to improve imaging-based assessment throughout the ILD care pathway. This review summarizes current applications of AI in ILD, with a focus on early detection, diagnostic classification, prognostic assessment, and longitudinal monitoring. AI-driven imaging analysis can enhance the identification of subtle interstitial abnormalities, improve classification of radiologic patterns, and generate quantitative biomarkers associated with disease severity and progression. Emerging multimodal models integrating imaging, clinical, and functional data may further refine risk stratification and support individualized management. Despite these advances, important barriers to widespread clinical implementation remain, including limited external validation, poor interpretability, data heterogeneity, and uncertain impact on patient-centered outcomes. AI is reshaping the role of thoracic imaging in ILD from descriptive interpretation toward quantitative and decision-supportive analysis. It also has the potential to optimize multidisciplinary discussion and improve efficiency in routine clinical workflows. To facilitate translation into clinical practice, future efforts should prioritize prospective validation, multimodal integration, and clinically meaningful implementation.