A deep learning–based algorithm was developed to stage Chronic Obstructive Pulmonary Disease (COPD) severity by quantifying emphysema and air trapping on CT images, assessing its prognostic value for disease progression and mortality. In a study of 8,951 patients, the algorithm-defined CT stages correlated with increased odds of disease progression and mortality, particularly in severe and very severe cases. When combined with GOLD staging, CT-based severity staging enhanced prognostic accuracy, demonstrating its potential as a complementary tool for COPD assessment.