Korean Journal of Radiology

The Korean Society of Radiology (대한영상의학회)
권호 표지
DOI QR코드

DOI QR Code

Deep Learning Algorithm for Simultaneous Noise Reduction and Edge Sharpening in Low-Dose CT Images: A Pilot Study Using Lumbar Spine CT

  • Hyunjung Yeoh (Department of Radiology, Seoul National University College of Medicine) ;
  • Sung Hwan Hong (Department of Radiology, Seoul National University College of Medicine) ;
  • Chulkyun Ahn (Department of Transdisciplinary Studies, Program in Biomedical Radiation Sciences, Graduate School of Convergence Science and Technology, Seoul National University) ;
  • Ja-Young Choi (Department of Radiology, Seoul National University College of Medicine) ;
  • Hee-Dong Chae (Department of Radiology, Seoul National University College of Medicine) ;
  • Hye Jin Yoo (Department of Radiology, Seoul National University College of Medicine) ;
  • Jong Hyo Kim (Department of Radiology, Seoul National University College of Medicine)
  • Received : 2021.02.17
  • Accepted : 2021.06.01
  • Published : 2021.11.01
Download PDF
⟨ Previous Next ⟩

Abstract

Objective: The purpose of this study was to assess whether a deep learning (DL) algorithm could enable simultaneous noise reduction and edge sharpening in low-dose lumbar spine CT. Materials and Methods: This retrospective study included 52 patients (26 male and 26 female; median age, 60.5 years) who had undergone CT-guided lumbar bone biopsy between October 2015 and April 2020. Initial 100-mAs survey images and 50-mAs intraprocedural images were reconstructed by filtered back projection. Denoising was performed using a vendor-agnostic DL model (ClariCT.AITM, ClariPI) for the 50-mAS images, and the 50-mAs, denoised 50-mAs, and 100-mAs CT images were compared. Noise, signal-to-noise ratio (SNR), and edge rise distance (ERD) for image sharpness were measured. The data were summarized as the mean ± standard deviation for these parameters. Two musculoskeletal radiologists assessed the visibility of the normal anatomical structures. Results: Noise was lower in the denoised 50-mAs images (36.38 ± 7.03 Hounsfield unit [HU]) than the 50-mAs (93.33 ± 25.36 HU) and 100-mAs (63.33 ± 16.09 HU) images (p < 0.001). The SNRs for the images in descending order were as follows: denoised 50-mAs (1.46 ± 0.54), 100-mAs (0.99 ± 0.34), and 50-mAs (0.58 ± 0.18) images (p < 0.001). The denoised 50-mAs images had better edge sharpness than the 100-mAs images at the vertebral body (ERD; 0.94 ± 0.2 mm vs. 1.05 ± 0.24 mm, p = 0.036) and the psoas (ERD; 0.42 ± 0.09 mm vs. 0.50 ± 0.12 mm, p = 0.002). The denoised 50-mAs images significantly improved the visualization of the normal anatomical structures (p < 0.001). Conclusion: DL-based reconstruction may enable simultaneous noise reduction and improvement in image quality with the preservation of edge sharpness on low-dose lumbar spine CT. Investigations on further radiation dose reduction and the clinical applicability of this technique are warranted.

Keywords

Acknowledgement

This work was supported by the Korea Medical Device Development Fund grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety) (Project Number: KMDF_PR_20200901_0017, 9991007051). This research was also supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI20C2092).

References

  1. Alshamari M, Geijer M, Norrman E, Liden M, Krauss W, Wilamowski F, et al. Low dose CT of the lumbar spine compared with radiography: a study on image quality with implications for clinical practice. Acta Radiol 2016;57:602-611
  2. Alshamari M, Geijer M, Norrman E, Liden M, Krauss W, Jendeberg J, et al. Impact of iterative reconstruction on image quality of low-dose CT of the lumbar spine. Acta Radiol 2017;58:702-709
  3. Lee SH, Yun SJ, Jo HH, Kim DH, Song JG, Park YS. Diagnostic accuracy of low-dose versus ultra-low-dose CT for lumbar disc disease and facet joint osteoarthritis in patients with low back pain with MRI correlation. Skeletal Radiol 2018;47:491-504
  4. Lee SH, Yun SJ, Kim DH, Jo HH, Song JG, Park YS. Diagnostic usefulness of low-dose lumbar multi-detector CT with iterative reconstruction in trauma patients: acomparison with standard-dose CT. Br J Radiol 2017;90:20170181
  5. Yang CH, Wu TH, Lin CJ, Chiou YY, Chen YC, Sheu MH, et al. Knowledge-based iterative model reconstruction technique in computed tomography of lumbar spine lowers radiation dose and improves tissue differentiation for patients with lower back pain. Eur J Radiol 2016;85:1757-1764
  6. Willemink MJ, Noel PB. The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence. Eur Radiol 2019;29:2185-2195
  7. Shah A, Rees M, Kar E, Bolton K, Lee V, Panigrahy A. Adaptive statistical iterative reconstruction use for radiation dose reduction in pediatric lower-extremity CT: impact on diagnostic image quality. Skeletal Radiol 2018;47:785-793
  8. Masamoto K, Fujibayashi S, Otsuki B, Hara K, Fukushima Y, Koizumi K, et al. Utility of thoracolumbar low-dose CT with model-based iterative reconstruction for measuring pedicle diameter using a radiation dose less than a one-time lumbar X-ray. Spine (Phila Pa 1976) 2020;45:38-47
  9. Mayo-Smith WW, Hara AK, Mahesh M, Sahani DV, Pavlicek W. How I do it: managing radiation dose in CT. Radiology 2014;273:657-672
  10. Hong JH, Park EA, Lee W, Ahn C, Kim JH. Incremental image noise reduction in coronary CT angiography using a deep learning-based technique with iterative reconstruction. Korean J Radiol 2020;21:1165-1177
  11. Nam JG, Ahn C, Choi H, Hong W, Park J, Kim JH, et al. Image quality of ultralow-dose chest CT using deep learning techniques: potential superiority of vendor-agnostic post-processing over vendor-specific techniques. Eur Radiol 2021;31:5139-5147
  12. U.S. Food and Drug Administration. 510(k) premarket notification. Accessdata.fda.gov Web site. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPMN/pmn.cfm?ID=K183460. Accessed May 13, 2020
  13. Ahn C, Heo C, Kim JH. Combined low-dose simulation and deep learning for CT denoising: application in ultra-low-dose chest CT. Proceedings of SPIE - The International Society for Optical Engineering; 2019 Jan 7-9; Singapore, Singapore: SPIE; 2019; p. 110500E.
  14. Kolb M, Storz C, Kim JH, Weiss J, Afat S, Nikolaou K, et al. Effect of a novel denoising technique on image quality and diagnostic accuracy in low-dose CT in patients with suspected appendicitis. Eur J Radiol 2019;116:198-204
  15. Lim WH, Choi YH, Park JE, Cho YJ, Lee S, Cheon JE, et al. Application of vendor-neutral iterative reconstruction technique to pediatric abdominal computed tomography. Korean J Radiol 2019;20:1358-1367
  16. Lee S, Choi YH, Cho YJ, Lee SB, Cheon JE, Kim WS, et al. Noise reduction approach in pediatric abdominal CT combining deep learning and dual-energy technique. Eur Radiol 2021;31:2218-2226
  17. Patro SN, Chakraborty S, Sheikh A. The use of adaptive statistical iterative reconstruction (ASiR) technique in evaluation of patients with cervical spine trauma: impact on radiation dose reduction and image quality. Br J Radiol 2016;89:20150082
  18. McHugh ML. Interrater reliability: the kappa statistic. Biochem Med (Zagreb) 2012;22:276-282
  19. Abul-Kasim K, Overgaard A, Maly P, Ohlin A, Gunnarsson M, Sundgren PC. Low-dose helical computed tomography (CT) in the perioperative workup of adolescent idiopathic scoliosis. Eur Radiol 2009;19:610-618
  20. Horger M, Claussen CD, Bross-Bach U, Vonthein R, Trabold T, Heuschmid M, et al. Whole-body low-dose multidetector row-CT in the diagnosis of multiple myeloma: an alternative to conventional radiography. Eur J Radiol 2005;54:289-297
  21. Mileto A, Guimaraes LS, McCollough CH, Fletcher JG, Yu L. State of the art in abdominal CT: the limits of iterative reconstruction algorithms. Radiology 2019;293:491-503
  22. Kim JN, Park HJ, Kim MS, Kook SH, Ham SY, Kim E, et al. Radiation dose reduction in extremity multi-detector CT: a comparison of image quality with a standard dose protocol. Eur J Radiol 2021;135:109405
  23. Suzuki S, Machida H, Tanaka I, Ueno E. Vascular diameter measurement in CT angiography: comparison of model-based iterative reconstruction and standard filtered back projection algorithms in vitro. AJR Am J Roentgenol 2013;200:652-657
  24. Tatsugami F, Higaki T, Sakane H, Fukumoto W, Kaichi Y, Iida M, et al. Coronary artery stent evaluation with model-based iterative reconstruction at coronary CT angiography. Acad Radiol 2017;24:975-981
  25. Tatsugami F, Higaki T, Nakamura Y, Yu Z, Zhou J, Lu Y, et al. Deep learning-based image restoration algorithm for coronary CT angiography. Eur Radiol 2019;29:5322-5329
  26. Alshamari M, Geijer M, Norrman E, Geijer H. Low-dose computed tomography of the lumbar spine: a phantom study on imaging parameters and image quality. Acta Radiol 2014;55:824-832
  27. Gervaise A, Osemont B, Lecocq S, Noel A, Micard E, Felblinger J, et al. CT image quality improvement using adaptive iterative dose reduction with wide-volume acquisition on 320-detector CT. Eur Radiol 2012;22:295-301
  28. Bohy P, de Maertelaer V, Roquigny A, Keyzer C, Tack D, Gevenois PA. Multidetector CT in patients suspected of having lumbar disk herniation: comparison of standard-dose and simulated low-dose techniques. Radiology 2007;244:524-531
  29. Yang CH, Wu TH, Chiou YY, Hung SC, Lin CJ, Chen YC, et al. Imaging quality and diagnostic reliability of low-dose computed tomography lumbar spine for evaluating patients with spinal disorders. Spine J 2014;14:2682-2690