Artificial intelligence in chronic obstructive pulmonary disease: recent advances in imaging and physiological monitoring.
Authors
Affiliations (3)
Affiliations (3)
- Division of Pulmonary, Critical Care, and Sleep Medicine.
- Department of Internal Medicine, University of Cincinnati.
- Cincinnati Veterans Affairs Medical Center, Cincinnati, Ohio, USA.
Abstract
Chronic obstructive pulmonary disease (COPD) is a leading cause of worldwide morbidity and mortality, yet significant barriers in its diagnosis and management persist. Artificial intelligence is rapidly emerging as a powerful tool to address these challenges. This review summarizes recent trends in its application to advance the care of patients with COPD, focusing on imaging and physiologic parameters. Recent literature demonstrates significant progress in artificial intelligence enhanced imaging, with deep learning models applied to chest radiographs and computed tomography showing high accuracy in detecting COPD, quantifying disease features, and predicting clinical outcomes including exacerbations and mortality. Machine learning algorithms are improving the interpretation of pulmonary function tests and leveraging novel data streams from cough sounds and wearable smart devices for noninvasive diagnosis, severity assessment, and the prediction of acute exacerbations. While artificial intelligence holds immense potential to shift COPD care toward a more proactive and personalized model, most applications remain in early developmental stages, with critical challenges including the need for rigorous clinical validation, addressing algorithmic bias, and establishing standardized evaluation metrics.