Artificial Intelligence–Driven Imaging Advances in Lung Fibrosis: A Comprehensive Review

Abstract

Lung fibrosis is a chronic and progressive illness where there is pathologic tissue scarring, which affects lung architecture and respiratory organ functioning. It is caused by several factors, including idiopathic pulmonary fibrosis (IPF), radiation-related injury, tumour-related fibrosis, and post-COVID-19 complications, and all of them are associated with similar pathophysiology. This review investigates and summarises studies published from 2015 to 2025 on biological processes, clinical symptoms, and technological innovations in the diagnosis and monitoring of lung fibrosis. It emphasises the way that artificial intelligence (AI) and deep learning (DL) models, such as quantitative computed tomography (CT) and convolutional neural networks (CNNs), have enhanced the process of early detection, disease classification, and medical progression forecasting. The computational models, such as the agent-based and Monte Carlo simulations, which are used to study fibrotic dynamics, are also discussed in the review. In general, the combination of molecular knowledge, imaging, and AI-based systems can be considered a major next step in the creation of personalized diagnoses and better treatment outcomes in chronic fibrotic lung diseases.