Tuberculosis and Respiratory Diseases

The Korean Academy of Tuberculosis and Respiratory Diseases (대한결핵및호흡기학회)
권호 표지
DOI QR코드

DOI QR Code

Kernel Conversion Improves the Correlation between the Extent of Emphysema and Clinical Parameters in Chronic Obstructive Pulmonary Disease: A Multicenter Cohort Study

  • Tai Joon An (Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Yeouido St. Mary's Hospital, College of Medicine, The Catholic University of Korea) ;
  • Youlim Kim (Division of Pulmonary and Allergy, Department of Internal Medicine, Konkuk University Medical Center, Konkuk University School of Medicine) ;
  • Hyun Lee (Department of Internal Medicine, Hanyang University College of Medicine) ;
  • Hyeon-Kyoung Koo (Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Inje University Ilsan Paik Hospital, Inje University College of Medicine) ;
  • Naoya Tanabe (Department of Respiratory Medicine, Graduate School of Medicine, Kyoto University) ;
  • Kum Ju Chae (Department of Radiology, Research Institute of Clinical Medicine of Jeonbuk National University-Biomedical Research Institute of Jeonbuk National University Hospital) ;
  • Kwang Ha Yoo (Division of Pulmonary and Allergy, Department of Internal Medicine, Konkuk University Medical Center, Konkuk University School of Medicine)
  • Received : 2024.11.03
  • Accepted : 2025.01.23
  • Published : 2025.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

Background: Computed tomography (CT) scans are utilized to assess emphysema, a prominent phenotype of chronic obstructive pulmonary disease (COPD). Variability in CT protocols and equipment across hospitals can impact accuracy. This study aims to implement kernel conversion across different CT settings and evaluate changes in the correlation between the emphysema index pre- and post-kernel conversion, along with clinical measures in COPD patients. Methods: Data were extracted from the Korea COPD Subgroup Study database, which included CT scan images from 484 COPD patients. These images underwent kernel conversion. Emphysema extent was quantified using the percentage of low-attenuation areas (%LAA-950) determined by a deep learning-based program. The correlation between %LAA-950 and clinical parameters, including lung function tests, the modified Medical Research Council (mMRC), 6-minute walking distance (6MWD), COPD assessment test (CAT), and the St. George's Respiratory Questionnaire for COPD (SGRQ-c), was analyzed. Subsequently, these values were compared across various CT settings. Results: A total of 484 participants were included. Kernel conversion significantly reduced the variance in %LAA-950 values (before vs. after: 12.6±11.0 vs. 8.8±11.9). Post-kernel conversion, %LAA-950 demonstrated moderate correlations with forced expiratory volume in 1 second (r=-0.41), residual volume/total lung capacity (r=0.42), mMRC (r=0.25), CAT score (r=0.12), SGRQ-c (r=0.21), and 6MWD (r=0.15), all of which were improved compared to the unconverted dataset (all p<0.01). Conclusion: CT images processed through kernel conversion enhance the correlation between the extent of emphysema and clinical parameters in COPD.

Keywords

Acknowledgement

This work was supported by the Research Program funded Korea National Institute of Health (Fund CODE 2016ER670100, 2016ER670101, 2016ER670102, 2018ER67100, 2018ER67101, 2018ER67102, 2021ER120500, 2021ER120501, 2021ER120502, 2024ER120100, and 2024ER120101).

References

  1. Global Initiative for Chronic Obstructive Lung Disease. 2022 GOLD reports: 2022 global strategy for prevention, diagnosis and management of COPD [Internet]. Deer Park: GOLD; 2022 [cited 2025 Feb 24]. https://goldcopd.org/2022-gold-reports.
  2. Chai CS, Liam CK, Pang YK, Ng DL, Tan SB, Wong TS, et al. Clinical phenotypes of COPD and health-related quality of life: a cross-sectional study. Int J Chron Obstruct Pulmon Dis 2019;14:565-73. https://doi.org/10.2147/COPD.S196109
  3. Han MK, Agusti A, Calverley PM, Celli BR, Criner G, Curtis JL, et al. Chronic obstructive pulmonary disease phenotypes: the future of COPD. Am J Respir Crit Care Med 2010;182:598-604. https://doi.org/10.1164/rccm.200912-1843CC
  4. Johannessen A, Skorge TD, Bottai M, Grydeland TB, Nilsen RM, Coxson H, et al. Mortality by level of emphysema and airway wall thickness. Am J Respir Crit Care Med 2013;187:602-8. https://doi.org/10.1164/rccm.201209-1722OC
  5. Lin CW, Huang HY, Chung FT, Lo CY, Huang YC, Wang TW, et al. Emphysema-predominant COPD had a greater 5-year mortality and a worse annual decline in lung function than airway obstruction-predominant COPD or asthma at initial same degree of airflow obstruction. Medicina (Kaunas) 2021;57:1261. https://doi.org/10.3390/medicina57111261
  6. Gevenois PA, De Vuyst P, de Maertelaer V, Zanen J, Jacobovitz D, Cosio MG, et al. Comparison of computed density and microscopic morphometry in pulmonary emphysema. Am J Respir Crit Care Med 1996;154:187-92. https://doi.org/10.1164/ajrccm.154.1.8680679
  7. Ostridge K, Wilkinson TM. Present and future utility of computed tomography scanning in the assessment and management of COPD. Eur Respir J 2016;48:216-28. https://doi.org/10.1183/13993003.00041-2016
  8. Lee YK, Oh YM, Lee JH, Kim EK, Lee JH, Kim N, et al. Quantitative assessment of emphysema, air trapping, and airway thickening on computed tomography. Lung 2008;186:157-65. https://doi.org/10.1007/s00408-008-9071-0
  9. Han MK, Kazerooni EA, Lynch DA, Liu LX, Murray S, Curtis JL, et al. Chronic obstructive pulmonary disease exacerbations in the COPDGene study: associated radiologic phenotypes. Radiology 2011;261:274-82. https://doi.org/10.1148/radiol.11110173
  10. Jairam PM, van der Graaf Y, Lammers JW, Mali WP, de Jong PA; PROVIDI Study group. Incidental findings on chest CT imaging are associated with increased COPD exacerbations and mortality. Thorax 2015;70:725-31. https://doi.org/10.1136/thoraxjnl-2014-206160
  11. Lee JY, Chon GR, Rhee CK, Kim DK, Yoon HK, Lee JH, et al. Characteristics of patients with chronic obstructive pulmonary disease at the first visit to a pulmonary medical center in Korea: the Korea COPD Subgroup Study team cohort. J Korean Med Sci 2016;31:553-60. https://doi.org/10.3346/jkms.2016.31.4.553
  12. Park HJ, Byun MK, Kim HJ, Ahn CM, Lee JH, Shin KC, et al. Asthma-COPD overlap shows favorable clinical outcomes compared to pure COPD in a Korean COPD cohort. Allergy Asthma Immunol Res 2017;9:431-7. https://doi.org/10.4168/aair.2017.9.5.431
  13. Jo YS, Hwang YI, Yoo KH, Kim TH, Lee MG, Lee SH, et al. Effect of inhaled corticosteroids on exacerbation of asthma-COPD overlap according to different diagnostic criteria. J Allergy Clin Immunol Pract 2020;8:1625-33. https://doi.org/10.1016/j.jaip.2020.01.004
  14. Kim Y, Kim SH, Rhee CK, Lee JS, Lee CY, Kim DK, et al. Air trapping and the risk of COPD exacerbation: analysis from prospective KOCOSS cohort. Front Med (Lausanne) 2022;9:835069. https://doi.org/10.3389/fmed.2022.835069
  15. Joo H, Park SY, Park SY, Park SY, Kim SH, Cho YS, et al. Phenotype of asthma-COPD overlap in COPD and severe asthma cohorts. J Korean Med Sci 2022;37:e236. https://doi.org/10.3346/jkms.2022.37.e236
  16. Choi JY, Yoon HK, Lee SY, Kim JW, Choi HS, Kim YI, et al. Comparison of clinical characteristics between chronic bronchitis and non-chronic bronchitis in patients with chronic obstructive pulmonary disease. BMC Pulm Med 2022;22:69. https://doi.org/10.1186/s12890-022-01854-x
  17. Shin SH, Kwon SO, Kim V, Silverman EK, Kim TH, Kim DK, et al. Association of body mass index and COPD exacerbation among patients with chronic bronchitis. Respir Res 2022;23:52. https://doi.org/10.1186/s12931-022-01957-3
  18. Choi H, Kim H, Jin KN, Jeong YJ, Chae KJ, Lee KH, et al. A challenge for emphysema quantification using a deep learning algorithm with low-dose chest computed tomography. J Thorac Imaging 2022;37:253-61. https://doi.org/10.1097/RTI.0000000000000647
  19. Park H, Hwang EJ, Goo JM. Deep learning-based kernel adaptation enhances quantification of emphysema on low-dose chest CT for predicting long-term mortality. Invest Radiol 2024;59:278-86. https://doi.org/10.1097/RLI.0000000000001003
  20. Lee SM, Lee JG, Lee G, Choe J, Do KH, Kim N, et al. CT image conversion among different reconstruction kernels without a sinogram by using a convolutional neural network. Korean J Radiol 2019;20:295-303. https://doi.org/10.3348/kjr.2018.0249
  21. Soliman MA, Kelahan LC, Magnetta M, Savas H, Agrawal R, Avery RJ, et al. A framework for harmonization of radiomics data for multicenter studies and clinical trials. JCO Clin Cancer Inform 2022;6:e2200023. https://doi.org/10.1200/CCI.22.00023
  22. Kim H, Park CM, Lee M, Park SJ, Song YS, Lee JH, et al. Impact of reconstruction algorithms on CT radiomic features of pulmonary tumors: analysis of intra- and inter-reader variability and inter-reconstruction algorithm variability. PLoS One 2016;11:e0164924. https://doi.org/10.1371/journal.pone.0164924
  23. Eldevik K, Nordhoy W, Skretting A. Relationship between sharpness and noise in CT images reconstructed with different kernels. Radiat Prot Dosimetry 2010;139:430-3. https://doi.org/10.1093/rpd/ncq063
  24. Choe J, Lee SM, Do KH, Lee G, Lee JG, Lee SM, et al. Deep learning-based image conversion of CT reconstruction kernels improves radiomics reproducibility for pulmonary nodules or masses. Radiology 2019;292:365-73. https://doi.org/10.1148/radiol.2019181960
  25. Bakker JT, Klooster K, Wisselink HJ, Pelgrim GJ, Vliegenthart R, Slebos DJ. Effect of chest computed tomography kernel use on emphysema score in severe chronic obstructive pulmonary disease patients evaluated for lung volume reduction. Respiration 2023;102:164-72. https://doi.org/10.1159/000528628
  26. Lynch DA, Austin JH, Hogg JC, Grenier PA, Kauczor HU, Bankier AA, et al. CT-definable subtypes of chronic obstructive pulmonary disease: a statement of the Fleischner Society. Radiology 2015;277:192-205. https://doi.org/10.1148/radiol.2015141579
  27. Gierada DS, Bierhals AJ, Choong CK, Bartel ST, Ritter JH, Das NA, et al. Effects of CT section thickness and reconstruction kernel on emphysema quantification relationship to the magnitude of the CT emphysema index. Acad Radiol 2010;17:146-56.
  28. Bak SH, Kim JH, Jin H, Kwon SO, Kim B, Cha YK, et al. Emphysema quantification using low-dose computed tomography with deep learning-based kernel conversion comparison. Eur Radiol 2020;30:6779-87. https://doi.org/10.1007/s00330-020-07020-3