Journal of radiological science and technology (대한방사선기술학회지:방사선기술과학)

Korean Society of Radiological Science (대한방사선과학회)
권호 표지
DOI QR코드

DOI QR Code

Optimization of Non-Local Means Algorithm in Low-Dose Computed Tomographic Image Based on Noise Level and Similarity Evaluations

노이즈 레벨 및 유사도 평가 기반 저선량 조건의 전산화 단층 검사 영상에서의 비지역적 평균 알고리즘의 최적화

  • Ha-Seon Jeong (Department of Radiological Science, Gachon University) ;
  • Ie-Jun Kim (Department of Radiological Science, Gachon University ) ;
  • Su-Bin Park (Department of Radiological Science, Gachon University ) ;
  • Suyeon Park (Department of Radiological Science, Gachon University ) ;
  • Yunji Oh (Department of Radiological Science, Gachon University ) ;
  • Woo-Seok Lee (Department of Radiological Science, Gachon University ) ;
  • Kang-Hyeon Seo (Department of Radiology, Hallym Hospital) ;
  • Youngjin Lee (Department of Radiological Science, Gachon University )
  • 정하선 (가천대학교 방사선학과) ;
  • 김이준 (가천대학교 방사선학과) ;
  • 박수빈 (가천대학교 방사선학과) ;
  • 박수연 (가천대학교 방사선학과) ;
  • 오윤지 (가천대학교 방사선학과) ;
  • 이우석 (가천대학교 방사선학과) ;
  • 서강현 (인천한림병원 영상의학과) ;
  • 이영진 (가천대학교 방사선학과)
  • Received : 2024.01.09
  • Accepted : 2024.01.22
  • Published : 2024.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we optimized the FNLM algorithm through a simulation study and applied it to a phantom scanned by low-dose CT to evaluate whether the FNLM algorithm can be used to obtain improved image quality images. We optimized the FNLM algorithm with MASH phantom and FASH phantom, which the algorithm was applied with MATLAB, increasing the smoothing factor from 0.01 to 0.05 with increments of 0.001 and measuring COV, RMSE, and PSNR values of the phantoms. For both phantom, COV and RMSE decreased, and PSNR increased as the smoothing factor increased. Based on the above results, we optimized a smoothing factor value of 0.043 for the FNLM algorithm. Then we applied the optimized FNLM algorithm to low dose lung CT and lung CT under normal conditions. In both images, the COV decreased by 55.33 times and 5.08 times respectively, and we confirmed that the quality of the image of low dose CT applying the optimized FNLM algorithm was 5.08 times better than the image of lung CT under normal conditions. In conclusion, we found that the smoothing factor of 0.043 among the factors of the FNLM algorithm showed the best results and validated the performance by reducing the noise in the low-quality CT images due to low dose with the optimized FNLM algorithm.

Keywords

References

  1. Singh D, Kumar V, Vaishali, Kaur M. Classification of COVID-19 patients from chest CT images using multi-objective differential evolution-based convolutional neural networks. European Journal of Clinical Microbiology & Infectious Diseases. 2020; 39(7):1379-89. DOI: https://doi.org/10.1007/s10096-020-03901-z
  2. Kim JS, Jeon MC, Han MS. Clinical application and future of the latest CT. Journal of the Korean Magnet ics Society. 2020;30(6):233-8. DOI: https://doi.org/10.1109/trpms.2020.3020212
  3. Ma J, Huang J, Feng Q, Zhang H, Lu H, Liang Z, Chen W. Low-dose computed tomography image restoration using previous normal-dose scan. Medical Physics. 2011;38(10):5713-31. DOI: https://doi.org/10.1118/1.3638125
  4. Raman SP, Mahesh M, Blasko RV, Fishman EK. CT scan parameters and radiation dose: Practical advice for radiologists. Journal of the American College of Radiology. 2013;10(11):840-6. DOI: https://doi.org/10.1016/j.jacr.2013.05.032
  5. Hamberg LM, Rhea JT, Hunter GJ, Thrall JH. Multi-detector row CT: Radiation dose characteristics. Radiology. 2003;226(3):762-72. DOI: https://doi.org/10.1148/radiol.2263020205
  6. Sarhan HG, Noor NMN, Saini SM, Bahari N. Recent advances in computed tomography radiation dosimetry. Asian Journal of Medical Technology. 2023;3(1):65-77. DOI: https://doi.org/10.32896/ajmedtech.v3n1.65-77
  7. Xiong T, Ye W. Improved adaptive kalman-median filter for line-scan x-ray transmission image. Sensors. 2022;22(13). DOI: https://doi.org/10.3390/s22134993
  8. Moghbela M, Mashohora S, Mahmud R, Saripan MIB. Automatic liver segmentation on computed tomography using random walkers for treatment planning. EXCLI Journal. 2016;15:500-17. DOI: https://doi.org/10.17179/excli2016-473
  9. Anam C, Fujibuchi T, Toyoda T, Sato N, Haryanto F, Widita R, et al. An investigation of a CT noise reduction using a modified of wiener filtering-edge detection. Journal of Physics: Conference Series. 2019; 1217(1):012022. DOI: https://doi.org/10.1088/1742-6596/1217/1/012022
  10. Froment J. Parameter-free fast pixelwise non-l ocal means denoising. Image Processing on Line. 2014;4:300-26. DOI: https://doi.org/10.5201/ipol.2014.120
  11. Ghane B, Karimian A, Mostafapour S, Gholamiankhak F, Shojaerazavi S, Arabi H. Quantitative analysis of image quality in low-dose computed tomography imaging for COVID-19 patients. Journal of Medical Signals & Sensors. 2023;13(2):118-28. DOI:https://doi.org/10.4103/jmss.jmss_173_21
  12. Buades A, Coll B, Morel JM. Non-local means denoising. Image Processing on Line. 2011;1:208-12. DOI: https://doi.org/10.5201/ipol.2011.bcm_nlm
  13. Darbon J, Cunha A, Chan TF, Osher S, Jensen GJ. Fast nonlocal filtering applied to electron cryomic roscopy. 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro. 2008;1331-4. DOI: https://doi.org/10.1109/ISBI.2008.4541250
  14. Kim DH, Keum BJ, Ahn HC, Lee HS. Empirical non-local algorithm for image and video denoising. 2013 IEEE International Conference on Consumer Electronics(ICCE). 2013:498-9. DOI: https://doi.org/10.1109/ICCE.2013.6486993
  15. Zimmer A, Ghuman P. CUDA optimization of non-local means extended to wrapped gaussian distributions for interferometric phase denoising. Procedia Computer Science. 2016;80:166-77. DOI:https://doi.org/10.1016/j.procs.2016.05.307
  16. Ahmed AS, EI-Behaidy WH, Youssif AAA. Medical denoising system based on stacked convolutional autoencoder for enhancing 2-dimensional gel electrophoresis noise reduction. Biomedical Signal Processing. 2021;36:102842. DOI: https://doi.org/10.1016/j.bspc.2021.102842
  17. Oyama A, Kumagai S, Arai N, Takata T, Saikawa Y, Shiraishi K, et al. Image quality improvement in cone-beam CT using the super-resolution technique. Journal of Radiation Research. 2018;59(4):501-10. DOI: https://doi.org/10.1093/jrr/rry019
  18. Kim BG, Kang SH, Park CR, Jeong HW, Lee YJ. Noise level and similarity analysis for computed tomographic thoracic image with fast non-local means denoising algorithm. Applied Sciences. 2020;10(21). DOI: https://doi.org/10.3390/app10217455
  19. Kang SH, Kim JY. Application of fast non-local means algorithm for noise reduction using separable color channels in light microscopy images. International Journal of Environmental Research and Public Health. 2021;18(6):2903. DOI: https://doi.org/10.3390/ijerph18062903
  20. Van De Ville D, Kocher M. Non-local means with dimensionality reduction and SURE-Based parameter selection. IEEE Transactions on Image Processing. 2011;20(9):2683-90. DOI: https://doi.org/10.1109/TIP.2011.2121083
  21. Choi DH, Kim JH, Choi JH, Kang SH, Lee YJ. Image optimization of fast non local means noise reduction algorithm using various filtering factors with human anthropomorphic phantom: A simulation study. Journal of the Korean Society of Radiology. 2019; 13(3):453-8. DOI: https://doi.org/10.7742/jksr.2019.13.3.453
  22. Seo KH, Kang SH, Shim JN, Lee YJ. Optimization of smoothing factor for fast non-local means algorithm in high pitch based low-dose computed tomography images with tin-filter. Radiation Physics and Chemistry. 2023;206:110762. DOI: https://doi.org/10.1016/j.radphyschem.2023.110762