From Pattern Recognition to Decision Guidance: Artificial Intelligence in Interstitial Lung Disease Diagnosis and Surveillance
Deliang Yang, Xue Li, David Chi Leung LamKey Points
Current ILD guidelines centre diagnosis on clinical context, exposure and autoimmune assessment, high‐resolution CT, pulmonary function testing, and multidisciplinary discussion, while explicit endorsement of artificial intelligence remains limited. Some artificial intelligence tools have already entered ILD‐related clinical workflows by flagging interstitial abnormalities on routine CT, supporting non‐invasive assessment of suspected fibrotic ILD, and quantifying fibrotic burden and interval progression on serial imaging. Research‐stage AI models are beginning to address unresolved gaps in clinical guidelines by reducing diagnostic uncertainty in borderline cases, enabling subgroup‐specific risk stratification in connective tissue disease‐associated ILD, and forecasting progression or follow‐up needs from longitudinal multimodal data. The remaining need is to develop a validated personalised multimodal decision‐support system that can integrate imaging, clinical, serological, pathological, and longitudinal data to guide biopsy, surveillance, and management decisions for uncertain ILD cases.