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

Amir M. Owrangi, Brandon Entwistle, Andrew Lu, Jack Chiu, Nabil Hussain, Roya Etemad-Rezai, Grace Parraga

Research output: Contribution to journalReview article

2 Citations (Scopus)

Abstract

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 (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. 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 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. Conclusion We developed a GUI for rapid and interactive emphysema scoring that allows for comparison of multiple readers with clinical and radiological standards.

Original languageEnglish (US)
JournalEuropean Journal of Radiology
Volume82
Issue number11
DOIs
StatePublished - Nov 1 2013

Fingerprint

Pulmonary Emphysema
X Ray Computed Tomography
Emphysema
Thorax
Lung
Observer Variation
Carbon Monoxide
Tomography
X-Rays
Radiologists

Keywords

  • Computed tomography
  • COPD
  • Emphysema
  • Subjective quantification

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Semi-automated scoring of pulmonary emphysema from X-ray CT : Trainee reproducibility and accuracy. / Owrangi, Amir M.; Entwistle, Brandon; Lu, Andrew; Chiu, Jack; Hussain, Nabil; Etemad-Rezai, Roya; Parraga, Grace.

In: European Journal of Radiology, Vol. 82, No. 11, 01.11.2013.

Research output: Contribution to journalReview article

Owrangi, Amir M. ; Entwistle, Brandon ; Lu, Andrew ; Chiu, Jack ; Hussain, Nabil ; Etemad-Rezai, Roya ; Parraga, Grace. / Semi-automated scoring of pulmonary emphysema from X-ray CT : Trainee reproducibility and accuracy. In: European Journal of Radiology. 2013 ; Vol. 82, No. 11.
@article{6a747f3ce8be4766adc039558ef1adff,
title = "Semi-automated scoring of pulmonary emphysema from X-ray CT: Trainee reproducibility and accuracy",
abstract = "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 (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. 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 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. Conclusion We developed a GUI for rapid and interactive emphysema scoring that allows for comparison of multiple readers with clinical and radiological standards.",
keywords = "Computed tomography, COPD, Emphysema, Subjective quantification",
author = "Owrangi, {Amir M.} and Brandon Entwistle and Andrew Lu and Jack Chiu and Nabil Hussain and Roya Etemad-Rezai and Grace Parraga",
year = "2013",
month = "11",
day = "1",
doi = "10.1016/j.ejrad.2013.07.013",
language = "English (US)",
volume = "82",
journal = "European Journal of Radiology",
issn = "0720-048X",
publisher = "Elsevier Ireland Ltd",
number = "11",

}

TY - JOUR

T1 - Semi-automated scoring of pulmonary emphysema from X-ray CT

T2 - Trainee reproducibility and accuracy

AU - Owrangi, Amir M.

AU - Entwistle, Brandon

AU - Lu, Andrew

AU - Chiu, Jack

AU - Hussain, Nabil

AU - Etemad-Rezai, Roya

AU - Parraga, Grace

PY - 2013/11/1

Y1 - 2013/11/1

N2 - 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 (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. 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 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. Conclusion We developed a GUI for rapid and interactive emphysema scoring that allows for comparison of multiple readers with clinical and radiological standards.

AB - 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 (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. 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 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. Conclusion We developed a GUI for rapid and interactive emphysema scoring that allows for comparison of multiple readers with clinical and radiological standards.

KW - Computed tomography

KW - COPD

KW - Emphysema

KW - Subjective quantification

UR - http://www.scopus.com/inward/record.url?scp=84885184355&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885184355&partnerID=8YFLogxK

U2 - 10.1016/j.ejrad.2013.07.013

DO - 10.1016/j.ejrad.2013.07.013

M3 - Review article

C2 - 23957938

AN - SCOPUS:84885184355

VL - 82

JO - European Journal of Radiology

JF - European Journal of Radiology

SN - 0720-048X

IS - 11

ER -