Semi-automated scoring of pulmonary emphysema from X-ray CT: Trainee reproducibility and accuracy

ObjectiveWe developed a semi-automated tool to quantify emphysema from thoracic X-ray multi-detector (64-slice) computed tomography (CT) for training purposes and multi-reader studies.Materials and MethodsThoracic X-ray CT was acquired in 93 ex-smokers, who were evaluated by six trainees with little or no expertise (trainees) and a single experienced thoracic radiologist (expert). A graphic user interface (GUI) was developed for emphysema quantification based on the percentile of lung where a score of 0 = no abnormalities, 1 = 1–25%, 2 = 26–50%, 3 = 51–75% and 4 = 76–100% for each lung side/slice. Trainees blinded to subject characteristics scored randomized images twice; accuracy was determined by comparison to expert scores, density histogram 15th percentile (HU15), relative area at −950 HU (RA950), low attenuation clusters at −950 HU (LAC950), −856 HU (LAC856) and the diffusing capacity for carbon monoxide (DLCO%pred). Intra- and inter-observer reproducibility was evaluated using coefficients-of-variation (COV), intra-class (ICC) and Pearson correlations.ResultsTrainee–expert correlations were significant (r = 0.85–0.97, p < 0.0001) and a significant trainee bias (0.15 ± 0.22) was observed. Emphysema score was correlated with RA950 (r = 0.88, p < 0.0001), HU15 (r = −0.77, p < 0.0001), LAC950 (r = 0.76, p < 0.0001), LAC856 (r = 0.74, p = 0.0001) and DLCO%pred (r = −0.71, p < 0.0001). Intra-observer reproducibility (COV = 4–27%; ICC = 0.75–0.94) was moderate to high for trainees; intra- and inter-observer COV were negatively and non-linearly correlated with emphysema score.ConclusionWe developed a GUI for rapid and interactive emphysema scoring that allows for comparison of multiple readers with clinical and radiological standards.