detection and severity scoring of chronic obstructive pulmonary disease using volumetric analysis of lung ct images

background chronic obstructive pulmonary disease (copd) is a devastating disease.while there is no cure for copd and the lung damage associated with this disease cannot be reversed, it is still very important to diagnose it as early as possible. objectives in this paper, we propose a novel method based on the measurement of air trapping in the lungs from ct images to detect copd and to evaluate its severity. patients and methods twenty-five patients and twelve normal adults were included in this study. the proposed method found volumetric changes of the lungs from inspiration to expiration. to this end, trachea ct images at full inspiration and expiration were compared and changes in the areas and volumes of the lungs between inspiration and expiration were used to define quantitative measures (features). using these features, the subjects were classified into two groups of normal and copd patients using a bayesian classifier. in addition, t-tests were applied to evaluate discrimination powers of the features for this classification. results for the cases studied, the proposed method estimated air trapping in the lungs from ct images without human intervention. based on the results, a mathematical model was developed to relate variations of lung volumes to the severity of the disease. conclusions as a computer aided diagnosis (cad) system, the proposed method may assist radiologists in the detection of copd. it quantifies air trapping in the lungs and thus may assist them with the scoring of the disease by quantifying the severity of the disease.