Korean Journal of Artificial Intelligence (한국인공지능학회지)

Korea Artificial Intelligence Association (한국인공지능학회)
권호 표지
DOI QR코드

DOI QR Code

Effectiveness of Noise Reduction in LDCT Images Based on SRCNN

  • Doo Bin KIM (Department of Radiological Science, Cheju Halla University) ;
  • Hyun Mee PARK (Department of Radiology, National Medical Center) ;
  • Sang Hoon JOON (Department of Radiology, National Medical Center) ;
  • Joo Wan HONG (Department of Radiological Science, Eulji University)
  • Received : 2024.10.17
  • Accepted : 2024.12.05
  • Published : 2024.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

This study aims to evaluate the performance of noise reduction in LDCT images using an SRCNN based AI model. Using the Lungman phantom, images of effective mAs 72 recommended by AAPM and effective mAs 709 without using the AEC function were acquired. SRCNN model input image used a GT, label image and GT image was used as a low-resolution image. Image evaluation was conducted in the lung apex, middle level lung, and carina of trachea regions, and PSNR, SSIM, SSIM error map, SNR, MSE, and RMSE were used as evaluation indices are based on label image. Lung apex results showed increase of 19.52, 29.69 and 23.8%, and decreased of 71.37, 46.43% respectively. Middle level lung results showed increase of 20.99, 20.0 and 26.26%, and decreased of 72.67, 47.72% respectively. Carina of trachea results showed increase of 22.05, 32.31 and 28.18%, and decreased of 73.93, 48.93% respectively. Image evaluation results were improvement in image quality due to noise reduction was confirmed using the SRCNN based AI model. Therefore, confirmed that applying the SRCNN to LDCT images can improve image quality by reducing noise, and it is considered that AI based post processing will be useful for CT images without AI.

Keywords

Acknowledgement

This paper was supported by Eulji University in 2023(EJRG-23-15)

References

  1. Alshamari, M., Geijer, M., Norrman, E., Liden, M., Krauss, W., Wilamowski, F., & Geijer, H. (2016). Low dose CT of the lumbar spine compared with radiography: a study on image quality with implications for clinical practice. Acta Radiologica, 57(5), 602-611. https://doi.org/10.1177/0284185115595667
  2. American Association of Physicists in Medicine. (2019). Lung Cancer Screening CT Protocols, Version 5.1. Alexandria, VA.
  3. Aslan, S., Bekci, T., Cakir, I. M., Ekiz, M., Yavuz, I., & Sahin, A. M. (2021). Diagnostic performance of low-dose chest CT to detect COVID-19: A Turkish population study. Diagnostic and Interventional Radiology, 27(2), 181. https://doi.org/10.5152/dir.2020.20350
  4. Christianson, O., Winslow, J., Frush, D. P., & Samei, E. (2015). Automated technique to measure noise in clinical CT examinations. American Journal of Roentgenology, 205(1), W93-W99. https://doi.org/10.2214/AJR.14.13613
  5. Diwakar, M., & Kumar, M. (2018). A review on CT image noise and its denoising. Biomedical Signal Processing and Control, 42, 73-88. https://doi.org/10.1016/j.bspc.2018.01.010
  6. Dong, C., Loy, C. C., He, K., & Tang, X. (2015). Image super-resolution using deep convolutional networks. Preprint on arXiv. https://doi.org/10.48550/arXiv.1501.00092
  7. Eisentopf, J., Achenbach, S., Ulzheimer, S., Layritz, C., Wuest, W., May, M., ... & Pflederer, T. (2013). Low-dose dual-source CT angiography with iterative reconstruction for coronary artery stent evaluation. JACC: Cardiovascular Imaging, 6(4), 458-465. https://doi.org/10.1016/j.jcmg.2012.10.023
  8. Ellmann, S., Kammerer, F., Allmendinger, T., Hammon, M., Janka, R., Lell, M., ... & Kramer, M. (2018). Advanced modeled iterative reconstruction (ADMIRE) facilitates radiation dose reduction in abdominal CT. Academic Radiology, 25(10), 1277-1284. https://doi.org/10.1016/j.acra.2018.01.014
  9. Fan, J., Yue, M., & Melnyk, R. (2014). Benefits of ASiR-V reconstruction for reducing patient radiation dose and preserving diagnostic quality in CT exams. White paper, GE Healthcare.
  10. Fu, B., Zhang, X., Wang, L., Ren, Y., & Thanh, D. N. (2022). A blind medical image denoising method with noise generation network. Journal of X-Ray Science and Technology, 30(3), 531-547. https://doi.org/10.3233/XST-211098
  11. Joemai, R. M., Veldkamp, W. J., Kroft, L. J., Hernandez-Giron, I., & Geleijns, J. (2013). Adaptive iterative dose reduction 3D versus filtered back projection in CT: evaluation of image quality. American Journal of Roentgenology, 201(6), 1291-1297. https://doi.org/10.2214/AJR.12.9780
  12. Kubo, T., Ohno, Y., Nishino, M., Lin, P. J., Gautam, S., Kauczor, H. U., & iLEAD Study Group. (2016). Low dose chest CT protocol (50amAs) as a routine protocol for comprehensive assessment of intrathoracic abnormality. European Journal of Radiology Open, 3, 86-94. https://doi.org/10.1016/j.ejro.2016.04.001
  13. Li, Q., Li, S., Li, R., Wu, W., Dong, Y., Zhao, J., ... & Aftab, R. (2022). Low-dose computed tomography image reconstruction via a multistage convolutional neural network with autoencoder perceptual loss network. Quantitative Imaging in Medicine and Surgery, 12(3), 1929. https://doi.org/10.21037/qims-21-465
  14. Li, Z., Yu, L., Trzasko, J. D., Lake, D. S., Blezek, D. J., Fletcher, J. G., ... & Manduca, A. (2014). Adaptive nonlocal means filtering based on local noise level for CT denoising. Medical physics, 41(1), 011908. https://doi.org/10.1118/1.4851635
  15. McLaughlin, P. D., Murphy, K. P., Hayes, S. A., Carey, K., Sammon, J., Crush, L., ... & Maher, M. M. (2014). Non-contrast CT at comparable dose to an abdominal radiograph in patients with acute renal colic; impact of iterative reconstruction on image quality and diagnostic performance. Insights into imaging, 5, 217-230. https://doi.org/10.1007/s13244-014-0310-z
  16. Punwani, S., Zhang, J., Davies, W., Greenhalgh, R., & Humphries, P. (2008). Paediatric CT: the effects of increasing image noise on pulmonary nodule detection. Pediatric radiology, 38, 192-201. https://doi.org/10.1007/s00247-007-0694-8
  17. Sadia, R. T., Chen, J., & Zhang, J. (2024). CT image denoising methods for image quality improvement and radiation dose reduction. Journal of Applied Clinical Medical Physics, 25(2), e14270. https://doi.org/10.1002/acm2.14270
  18. Tang, Y., Du, Q., Wang, J., Wu, Z., Li, Y., Li, M., ... & Zheng, J. (2022). CCN-CL: A content-noise complementary network with contrastive learning for low-dose computed tomography denoising. Computers in Biology and Medicine, 147, 105759. https://doi.org/10.1016/j.compbiomed.2022.105759