Please cite this article in press as: Owrangi AM, et al. Semi-automated scoring of pulmonary emphysema from X-ray CT: Trainee reproducibility and accuracy. Eur J Radiol (2013), http://dx.doi.org/10.1016/j.ejrad.2013.07.013 ARTICLE IN PRESS G Model EURR-6457; No. of Pages 8 European Journal of Radiology xxx (2013) xxx–xxx Contents lists available at ScienceDirect European Journal of Radiology jo ur nal ho me page: www.elsevier.com/locate/ejrad Editorial musings Semi-automated scoring of pulmonary emphysema from X-ray CT: Trainee reproducibility and accuracy Amir M. Owrangi a,b,1 , Brandon Entwistle c,2 , Andrew Lu c,3 , Jack Chiu c,4 , Nabil Hussain c,5 , Roya Etemad-Rezai c,6 , Grace Parraga a,b,c,d,∗ a Imaging Research Laboratories, Robarts Research Institute, London, Canada b Graduate Program in Biomedical Engineering c Department of Medical Imaging d Department of Medical Biophysics, The University of Western Ontario, London, Canada a r t i c l e i n f o Article history: Received 26 November 2012 Received in revised form 14 May 2013 Accepted 16 July 2013 Keywords: Computed tomography Subjective quantification Emphysema COPD a b s t r a c t Objective: We 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 Methods: Thoracic 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 (HU 15 ), relative area at -950 HU (RA 950 ), low attenuation clusters at -950 HU (LAC 950 ), -856 HU (LAC 856 ) and the diffusing capacity for carbon monoxide (DL CO%pred ). Intra- and inter-observer reproducibility was evaluated using coefficients-of-variation (COV), intra-class (ICC) and Pearson correlations. Results: Trainee–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 RA 950 (r = 0.88, p < 0.0001), HU 15 (r = -0.77, p < 0.0001), LAC 950 (r = 0.76, p < 0.0001), LAC 856 (r = 0.74, p = 0.0001) and DL CO%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. Conclusion: We developed a GUI for rapid and interactive emphysema scoring that allows for comparison of multiple readers with clinical and radiological standards. © 2013 Elsevier Ireland Ltd. All rights reserved. ∗ Correspondence at: Imaging Research Laboratories Robarts Research Institute 100 Perth Drive London, Canada, N6A 5K8. Tel.: +1 519 913 5265; fax: +1 519 913 5238/519 913 5260. E-mail addresses: aowrangi@robarts.ca (A.M. Owrangi), Brandon.Entwistle@londonhospitals.ca (B. Entwistle), Andrew.Lu@londonhospitals.ca (A. Lu), Jack.Chiu@londonhospitals.ca (J. Chiu), Nabil.Hussain@londonhospitals.ca (N. Hussain), Roya.EtemadRezai@lhsc.on.ca (R. Etemad-Rezai), gparraga@robarts.ca, gep@imaging.robarts.ca (G. Parraga). 1 1Imaging Research Laboratories, Robarts Research Institute 100 Perth Drive, London, Canada, N6A 5K8; Tel.: +1 519 913 5265; fax: +1 519 913 5260. 2 2Department of Medical Imaging, University Hospital, 339 Windermere Road, London, Canada, N6A 5A5; Tel.: +1 519 685 8500 ext 33648; fax: +1 519 663 8803. 3 3Department of Medical Imaging, University Hospital, 339 Windermere Road, London, Canada, N6A 5A5; Tel.: +1 519 685 8500 ext 33648; fax: +1 519 663 8803 4 4Department of Medical Imaging, University Hospital, 339 Windermere Road, London, Canada, N6A 5A5; Tel.: +1 519 685 8500 ext 33648; fax: +1 519 663 8803. 5 5Department of Medical Imaging, University Hospital, 339 Windermere Road, London, Canada, N6A 5A5; Tel.: +1 519 685 8500 ext 33648; fax: +1 519 663 8803. 6 6Department of Medical Imaging, University Hospital, 339 Windermere Road, London, Canada, N6A 5A5; Tel.: +1 519 685 8500x33648; fax: +1 519 663 8803. 1. Introduction Concomitant with airway abnormalities and other morpholog- ical consequences of chronic pulmonary inflammation, the lungs of smokers and many ex-smokers also typically show evidence of emphysema, defined as lung tissue destruction resulting in reduced gas exchange in the respiratory system [1,2] – a major component of chronic obstructive pulmonary disease (COPD) [3]. Early detection of pulmonary emphysema in at-risk patients, even in the absence of symptoms, remains a diagnostic challenge but may help prevent obstructive ventilatory impairment later in life [4]. Although magnetic resonance imaging (MRI) and nuclear medicine methods can be used to quantify the extent of pul- monary emphysema [5,6], thoracic X-ray computed tomography (CT) is the imaging modality of choice for clinical detection and for research studies that aim to monitor emphysema longitudi- nally [7,8], mainly because of short acquisition times, high spatial resolution and the rich tissue information content based on the differential attenuation of X-ray in the lung tissue and airspaces. 0720-048X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2013.07.013