Lung ultrasound in interstitial lung disease: from bedside screening to prognostic integration and AI-assisted...

Interstitial lung diseases (ILDs) require early recognition and longitudinal assessment, yet repeated high-resolution computed tomography (HRCT) is often limited by access, cost, and cumulative radiation exposure, particularly in connective tissue disease-associated ILD (CTD-ILD). Lung ultrasound (LUS) is a bedside, radiation-free, repeatable adjunct that primarily evaluates B-line burden and pleural line abnormalities. In this review, we summarize core sonographic signs and key diagnostic pitfalls, and synthesize the evidence for three clinical domains: screening, case finding and triage in high-risk populations, phenotype-relevant cues, and interval monitoring between HRCT examinations, ideally alongside pulmonary function tests. We further compare scanning protocols and scoring approaches, including B-line-dominant, pleural line-dominant, and composite frameworks, and outline practical strategies for harmonized acquisition, reproducible scoring, and structured reporting. We also review the emerging role of artificial intelligence (AI) in acquisition guidance, quality control, and feature quantification, while emphasizing important boundaries related to device variability, disease-spectrum shift, and limited specificity. Current evidence most strongly supports LUS as a complementary tool for screening, triage and bedside trend monitoring, particularly in CTD-ILD, whereas phenotype-oriented interpretation, progression-sensitive markers, and AI-enabled deployment remain developmental. Future priorities include multicentre harmonization of protocols and scoring, prospective validation of clinically meaningful change, and transparent external evaluation to support safe integration into multidisciplinary care pathways.