Computed tomography in ARDS, from morphological insights to AI-powered multi-modal analysis: a narrative review.
Authors
Affiliations (3)
Affiliations (3)
- Department of Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1277 Jiefang Avenue, Wuhan, 430022, Hubei, China.
- Department of Critical Care Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1095 Jiefang Avenue, Wuhan, 430030, Hubei, China.
- Department of Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1277 Jiefang Avenue, Wuhan, 430022, Hubei, China. [email protected].
Abstract
Acute respiratory distress syndrome (ARDS) is a critical clinical condition characterized by acute respiratory failure and high mortality. It poses considerable challenges in both diagnosis and management. Imaging constitutes a central element of the conceptual framework for ARDS, with computed tomography (CT) being an essential technical tool for studying the morphological and pathological mechanisms of lung tissue in ARDS. CT imaging has provided profound insights into the respiratory mechanics in ARDS and has informed the optimization of ventilation strategies. It is widely used to characterize the typical pathophysiological manifestations of ARDS in the lungs and can quantify the distribution of ventilation, perfusion, and pulmonary edema. Moreover, CT-based morphological classification of ARDS constitutes a significant component of ARDS subphenotypes research. However, given the heterogeneity in both its diagnosis and response to treatment, a single assessment model is insufficient to meet the management needs of patients with ARDS. The widespread application of artificial intelligence (AI) has greatly facilitated the quantitative analysis of CT imaging, enabling the integration of multidimensional data, such as CT imaging, pulmonary functional data, and laboratory tests. This narrative review adopts a CT-centric viewpoint, delineating the progressive shift in the diagnosis, phenotyping, and management of ARDS from qualitative to quantitative analysis and from unimodal to multimodal evaluation, propelled by ongoing advances in AI. Looking forward, CT-based multimodal fusion analysis holds promise for identifying more precise therapeutic biomarkers and advancing the development of individualized treatment strategies for ARDS.