Artificial intelligence for interstitial lung disease assessment on chest CT

In the past few years, many tools using artificial intelligence (AI) have been developed for medical image analysis, especially in thoracic imaging. In this chapter, we review the main potential applications of AI for interstitial lung disease (ILD) analysis on chest computed tomography (CT). Most of the published works have focused on severity assessment which relies on ILD quantification. Textural analysis and more recently deep learning tools have been developed to quantify ILD. These tools have been used to show the relationship between ILD extent and functional impairment or outcome. AI has also been used to diagnose ILD on chest CT and for pattern recognition. Finally, some work has evaluated the contribution of AI for ILD monitoring, either by studying changes in ILD extent or by studying fibrosis-related lung shrinking. Although much research has been conducted on AI for ILD, the use of AI remains limited in clinical practice; however, the transition from research to practice is expected. Cite as: Chassagnon G, Marini R, Canniff E, et al. Artificial intelligence for interstitial lung disease assessment on chest CT. In: Pinnock H, Poberezhets V, Drummond D, eds. Digital Respiratory Healthcare (ERS Monograph). Sheffield, European Respiratory Society, 2023; pp. 276–286 [https://doi.org/10.1183/2312508X.10002523].