Emphysema quantification using low-dose computed tomography with deep learning–based kernel conversion comparison

Abstract

Objective

This study determined the effect of dose reduction and kernel selection on quantifying emphysema using low-dose computed tomography (LDCT) and evaluated the efficiency of a deep learning–based kernel conversion technique in normalizing kernels for emphysema quantification.

Methods

A sample of 131 participants underwent LDCT and standard-dose computed tomography (SDCT) at 1- to 2-year intervals. LDCT images were reconstructed with B31f and B50f kernels, and SDCT images were reconstructed with B30f kernels. A deep learning model was used to convert the LDCT image from a B50f kernel to a B31f kernel. Emphysema indices (EIs), lung attenuation at 15th percentile (perc15), and mean lung density (MLD) were calculated. Comparisons among the different kernel types for both LDCT and SDCT were performed using Friedman’s test and Bland-Altman plots.

Results

All values of LDCT B50f were significantly different compared with the values of LDCT B31f and SDCT B30f (p < 0.05). Although there was a statistical difference, the variation of the values of LDCT B50f significantly decreased after kernel normalization. The 95% limits of agreement between the SDCT and LDCT kernels (B31f and converted B50f) ranged from − 2.9 to 4.3% and from − 3.2 to 4.4%, respectively. However, there were no significant differences in EIs and perc15 between SDCT and LDCT converted B50f in the non-chronic obstructive pulmonary disease (COPD) participants (p > 0.05).

Conclusion

The deep learning–based CT kernel conversion of sharp kernel in LDCT significantly reduced variation in emphysema quantification, and could be used for emphysema quantification.

Key Points

• Low-dose computed tomography with smooth kernel showed adequate performance in quantifying emphysema compared with standard-dose CT.

• Emphysema quantification is affected by kernel selection and the application of a sharp kernel resulted in a significant overestimation of emphysema.

• Deep learning–based kernel normalization of sharp kernel significantly reduced variation in emphysema quantification.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Abbreviations

CNN:

Conversion neural network

COPD:

Chronic obstructive pulmonary disease

CTDIvol :

Volumetric computed tomography dose index

DLP:

Dose-length product

EI:

Emphysema index

HU:

Hounsfield unit

LDCT:

Low-dose computed tomography

MLD:

Mean lung density

Perc15:

Lung attenuation at 15th percentile

SDCT:

Standard-dose computed tomography

References

  1. 1.

    Snider GL, Kleinerman J, Thurlbeck WM, Bengali ZH (1985) The definition of emphysema. Report of a National Heart, Lung, and Blood Institute, Division of Lung Diseases workshop. Am Rev Respir Dis 132(1):182–185

  2. 2.

    Soejima K, Yamaguchi K, Kohda E et al (2000) Longitudinal follow-up study of smoking-induced lung density changes by high-resolution computed tomography. Am J Respir Crit Care Med 161:1264–1273

    CAS  Article  Google Scholar 

  3. 3.

    Gietema HA, Schilham AM, van Ginneken B, van Klaveren RJ, Lammers JW, Prokop M (2007) Monitoring of smoking-induced emphysema with CT in a lung cancer screening setting: detection of real increase in extent of emphysema. Radiology 244:890–897

    Article  Google Scholar 

  4. 4.

    Gevenois PA, de Maertelaer V, De Vuyst P, Zanen J, Yernault JC (1995) Comparison of computed density and macroscopic morphometry in pulmonary emphysema. Am J Respir Crit Care Med 152:653–657

    CAS  Article  Google Scholar 

  5. 5.

    Madani A, De Maertelaer V, Zanen J, Gevenois PA (2007) Pulmonary emphysema: radiation dose and section thickness at multidetector CT quantification--comparison with macroscopic and microscopic morphometry. Radiology 243:250–257

    Article  Google Scholar 

  6. 6.

    Boedeker KL, McNitt-Gray MF, Rogers SR et al (2004) Emphysema: effect of reconstruction algorithm on CT imaging measures. Radiology 232:295–301

    Article  Google Scholar 

  7. 7.

    Yuan R, Mayo JR, Hogg JC et al (2007) The effects of radiation dose and CT manufacturer on measurements of lung densitometry. Chest 132:617–623

    Article  Google Scholar 

  8. 8.

    Lee SM, Lee JG, Lee G et al (2019) CT image conversion among different reconstruction kernels without a sinogram by using a convolutional neural network. Korean J Radiol 20:295–303

    Article  Google Scholar 

  9. 9.

    Gierada DS, Bierhals AJ, Choong CK et al (2010) Effects of CT section thickness and reconstruction kernel on emphysema quantification relationship to the magnitude of the CT emphysema index. Acad Radiol 17:146–156

    Article  Google Scholar 

  10. 10.

    Jin H, Heo C, Kim JH (2019) Deep learning-enabled accurate normalization of reconstruction kernel effects on emphysema quantification in low-dose CT. Phys Med Biol 64:135010

    Article  Google Scholar 

  11. 11.

    Hong Y, Kwon J, Lee S et al (2014) Methodology of an observational cohort study for subjects with chronic obstructive pulmonary disease in dusty areas near cement plants. J Pulm Respir Med 4:169

    Google Scholar 

  12. 12.

    Bhatt SP, Washko GR, Hoffman EA et al (2019) Imaging advances in chronic obstructive pulmonary disease. Insights from the Genetic Epidemiology of Chronic Obstructive Pulmonary Disease (COPDGene) study. Am J Respir Crit Care Med 199:286–301

    Article  Google Scholar 

  13. 13.

    Deak PD, Smal Y, Kalender WA (2010) Multisection CT protocols: sex- and age-specific conversion factors used to determine effective dose from dose-length product. Radiology 257:158–166

    Article  Google Scholar 

  14. 14.

    Wang R, Sui X, Schoepf UJ et al (2015) Ultralow-radiation-dose chest CT: accuracy for lung densitometry and emphysema detection. AJR Am J Roentgenol 204:743–749

    Article  Google Scholar 

  15. 15.

    Gierada DS, Pilgram TK, Whiting BR et al (2007) Comparison of standard- and low-radiation-dose CT for quantification of emphysema. AJR Am J Roentgenol 188:42–47

    Article  Google Scholar 

  16. 16.

    O’Brien C, Kok HK, Kelly B et al (2019) To investigate dose reduction and comparability of standard dose CT vs ultra low dose CT in evaluating pulmonary emphysema. Clin Imaging 53:115–119

    Article  Google Scholar 

  17. 17.

    Shaker SB, Stavngaard T, Laursen LC, Stoel BC, Dirksen A (2011) Rapid fall in lung density following smoking cessation in COPD. COPD 8:2–7

    Article  Google Scholar 

  18. 18.

    Ashraf H, Lo P, Shaker SB et al (2011) Short-term effect of changes in smoking behaviour on emphysema quantification by CT. Thorax 66:55–60

    Article  Google Scholar 

  19. 19.

    Jobst BJ, Weinheimer O, Trauth M et al (2018) Effect of smoking cessation on quantitative computed tomography in smokers at risk in a lung cancer screening population. Eur Radiol 28:807–815

    Article  Google Scholar 

  20. 20.

    Mohamed Hoesein FA, Zanen P, de Jong PA et al (2013) Rate of progression of CT-quantified emphysema in male current and ex-smokers: a follow-up study. Respir Res 14:55

    Article  Google Scholar 

  21. 21.

    Zach JA, Williams A, Jou SS et al (2016) Current smoking status is associated with lower quantitative CT measures of emphysema and gas trapping. J Thorac Imaging 31:29–36

    Article  Google Scholar 

  22. 22.

    Gallardo-Estrella L, Lynch DA, Prokop M et al (2016) Normalizing computed tomography data reconstructed with different filter kernels: effect on emphysema quantification. Eur Radiol 26:478–486

    Article  Google Scholar 

  23. 23.

    Kim H, Goo JM, Ohno Y et al (2019) Effect of reconstruction parameters on the quantitative analysis of chest computed tomography. J Thorac Imaging 34:92–102

    Article  Google Scholar 

  24. 24.

    Bartel ST, Bierhals AJ, Pilgram TK et al (2011) Equating quantitative emphysema measurements on different CT image reconstructions. Med Phys 38:4894–4902

    Article  Google Scholar 

  25. 25.

    Ceresa M, Bastarrika G, de Torres JP et al (2011) Robust, standardized quantification of pulmonary emphysema in low dose CT exams. Acad Radiol 18:1382–1390

    Article  Google Scholar 

  26. 26.

    Gallardo-Estrella L, Pompe E, de Jong PA et al (2017) Normalized emphysema scores on low dose CT: validation as an imaging biomarker for mortality. PLoS One 12:e0188902

    Article  Google Scholar 

  27. 27.

    Ohkubo M, Wada S, Kayugawa A, Matsumoto T, Murao K (2011) Image filtering as an alternative to the application of a different reconstruction kernel in CT imaging: feasibility study in lung cancer screening. Med Phys 38:3915–3923

    Article  Google Scholar 

  28. 28.

    Jin H, Heo C, Kim JH (2018) Impact of deep learning of deep learning on the normalization of reconstruction kernel effects in imaging biomarker quantification: a pilot study in CT emphysema. Medical Imaging 2018: Computer-Aided Diagnosis: International Society for Optics and Photonics 2018:105753L

  29. 29.

    Madani A, Van Muylem A, Gevenois PA (2010) Pulmonary emphysema: effect of lung volume on objective quantification at thin-section CT. Radiology 257:260–268

    Article  Google Scholar 

Download references

Funding

The scientific grantor of this research is So Hyeon Bak. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF 2018R1D1A1B07049670).

Author information

Affiliations

Authors

Corresponding authors

Correspondence to Jong Hyo Kim or Woo Jin Kim.

Ethics declarations

Guarantor

This scientific guarantor of this research is Woo Jin Kim.

Conflict of interest

The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.

Statistics and biometry

Two of the authors (Sung Ok Kwon and Bom Kim) have significant statistical expertise.

Informed consent

Written informed consent was obtained from all subjects (patients) in this study.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• retrospective

• cross sectional study

• performed at one institution

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

ESM 1

(DOCX 19 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bak, S.H., Kim, J.H., Jin, H. et al. Emphysema quantification using low-dose computed tomography with deep learning–based kernel conversion comparison. Eur Radiol (2020). https://doi.org/10.1007/s00330-020-07020-3

Download citation

Keywords

  • Emphysema
  • Deep learning
  • Densitometry
  • Tomography