Skip to main content
SpringerLink
Log in

A deep-learning-based scatter correction with water equivalent path length map for digital radiography

  • Research Article
  • Published:
Radiological Physics and Technology Aims and scope Submit manuscript

Abstract

We proposed a new deep learning (DL) model for accurate scatter correction in digital radiography. The proposed network featured a pixel-wise water equivalent path length (WEPL) map of subjects with diverse sizes and 3D inner structures. The proposed U-Net model comprises two concatenated modules: one for generating a WEPL map and the other for predicting scatter using the WEPL map as auxiliary information. First, 3D CT images were used as numerical phantoms for training and validation, generating observed and scattered images by Monte Carlo simulation, and WEPL maps using Siddon’s algorithm. Then, we optimised the model without overfitting. Next, we validated the proposed model's performance by comparing it with other DL models. The proposed model obtained scatter-corrected images with a peak signal-to-noise ratio of 44.24 ± 2.89 dB and a structural similarity index measure of 0.9987 ± 0.0004, which were higher than other DL models. Finally, scatter fractions (SFs) were compared with other DL models using an actual phantom to confirm practicality. Among DL models, the proposed model showed the smallest deviation from measured SF values. Furthermore, using an actual radiograph containing an acrylic object, the contrast-to-noise ratio (CNR) of the proposed model and the anti-scatter grid were compared. The CNR of the images corrected using the proposed model are 16% and 82% higher than those of the raw and grid-applied images, respectively. The advantage of the proposed method is that no actual radiography system is required for collecting training dataset, as the dataset is created from CT images using Monte Carlo simulation.

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

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

Similar content being viewed by others

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Kunitomo H, Ichikawa K. Signal-to-noise ratio improvements using anti-scatter grids with different object thicknesses and tube voltages. Physica Med. 2020;73:105–10. https://doi.org/10.1016/j.ejmp.2020.04.014.

    Article  Google Scholar 

  2. Barnes GT. Contrast and scatter in x-ray imaging. Radiographics. 1991;11(2):307–23. https://doi.org/10.1148/radiographics.11.2.2028065.

    Article  CAS  PubMed  Google Scholar 

  3. Ning R, Chen B, Yu R, Conover D, Tang X, Ning Y. Flat panel detector-based cone-beam volume CT angiography imaging: system evaluation. IEEE Trans Med Imaging. 2000;19(9):949–63. https://doi.org/10.1109/42.887842.

    Article  CAS  PubMed  Google Scholar 

  4. Siewerdsen JH, Jaffray DA. Cone-beam computed tomography with a flat-panel imager: magnitude and effects of x-ray scatter. Med Phys. 2001;28(2):220–31. https://doi.org/10.1118/1.1339879.

    Article  CAS  PubMed  Google Scholar 

  5. Chan H-P, Doi K. Investigation of the performance of antiscatter grids: Monte Carlo simulation studies. Phys Med Biol. 1982;27(6):785. https://doi.org/10.1088/0031-9155/27/6/002.

    Article  CAS  PubMed  Google Scholar 

  6. Kalender WA. Calculation of x-ray grid characteristics by Monte Carlo methods. Phys Med Biol. 1982;27(3):353. https://doi.org/10.1088/0031-9155/27/3/002.

    Article  CAS  PubMed  Google Scholar 

  7. Love LA, Kruger RA. Scatter estimation for a digital radiographic system using convolution filtering. Med Phys. 1987;14(2):178–85. https://doi.org/10.1118/1.596126.

    Article  CAS  PubMed  Google Scholar 

  8. Naimuddin S, Hasegawa B, Mistretta CA. Scatter-glare correction using a convolution algorithm with variable weighting. Med Phys. 1987;14(3):330–4. https://doi.org/10.1118/1.596088.

    Article  CAS  PubMed  Google Scholar 

  9. Seibert JA, Boone JM. X-ray scatter removal by deconvolution. Med Phys. 1988;15(4):567–75. https://doi.org/10.1118/1.596208.

    Article  CAS  PubMed  Google Scholar 

  10. Wagner FC, Macovski A, Nishimura DG. A characterization of the scatter point-spread-function in terms of air gaps. IEEE Trans Med Imaging. 1988;7(4):337–44. https://doi.org/10.1109/42.14517.

    Article  CAS  PubMed  Google Scholar 

  11. Siewerdsen JH, Moseley DJ, Bakhtiar B, Richard S, Jaffray DA. The influence of antiscatter grids on soft-tissue detectability in cone-beam computed tomography with flat-panel detectors: antiscatter grids in cone-beam CT. Med Phys. 2004;31(12):3506–20. https://doi.org/10.1118/1.1819789.

    Article  CAS  PubMed  Google Scholar 

  12. Nykänen K, Siltanen S. X-ray scattering in full-field digital mammography. Med Phys. 2003;30(7):1864–73. https://doi.org/10.1118/1.1584160.

    Article  PubMed  Google Scholar 

  13. Ducote JL, Molloi S. Scatter correction in digital mammography based on image deconvolution. Phys Med Biol. 2010;55(5):1295. https://doi.org/10.1088/0031-9155/55/5/003.

    Article  CAS  PubMed  Google Scholar 

  14. Ersahin A, Molloi S, Qian Y-J. A digital filtration technique for scatter-glare correction based on thickness estimation. IEEE Trans Med Imaging. 1995;14(3):587–95. https://doi.org/10.1109/42.414624.

    Article  CAS  PubMed  Google Scholar 

  15. Díaz O, Dance DR, Young KC, Elangovan P, Bakic PR, Wells K. A fast scatter field estimator for digital breast tomosynthesis. Phys Med Imaging SPIE. 2012;8313:63–71. https://doi.org/10.1117/12.911494.

    Article  CAS  Google Scholar 

  16. Maier J, Sawall S, Knaup M, Kachelrieß M. Deep scatter estimation (DSE): accurate real-time scatter estimation for X-ray CT using a deep convolutional neural network. J Nondestr Eval. 2018;37:1–9. https://doi.org/10.1007/s10921-018-0507-z.

    Article  Google Scholar 

  17. Jiang Y, Yang C, Yang P, Hu X, Luo C, Xue Y, Xu L, Hu X, Zhang L, Wang J. Scatter correction of cone-beam CT using a deep residual convolution neural network (DRCNN). Phys Med Biol. 2019;64(14): 145003. https://doi.org/10.1088/1361-6560/ab23a6.

    Article  PubMed  Google Scholar 

  18. Lee H, Lee J. A deep learning-based scatter correction of simulated x-ray images. Electronics. 2019;8(9):944. https://doi.org/10.3390/electronics8090944.

    Article  Google Scholar 

  19. Nomura Y, Xu Q, Shirato H, Shimizu S, Xing L. Projection-domain scatter correction for cone beam computed tomography using a residual convolutional neural network. Med Phys. 2019;46(7):3142–55. https://doi.org/10.1002/mp.13583.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Pautasso JJ, Caballo M, Mikerov M, Boone JM, Michielsen K, Sechopoulos I. Deep learning for x-ray scatter correction in dedicated breast CT. Med Phys. 2022;50:2022–36. https://doi.org/10.1002/mp.16185.

    Article  Google Scholar 

  21. Wang J, Duan X, Christner JA, Leng S, Yu L, McCollough CH. Attenuation-based estimation of patient size for the purpose of size specific dose estimation in CT. Part I. Development and validation of methods using the CT image. Med Phys. 2012;39(11):6764–71. https://doi.org/10.1118/1.4757586.

    Article  PubMed  Google Scholar 

  22. AAPM, Size-specific dose estimates (ssde) in pediatric and adult bodyct examinations, aapm task group 204, (2011) 204. https://www.aapm.org/pubs/reports/rpt_204.pdf.Accessed on 27 Sep 2023

  23. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2016), 770–778. https://doi.org/10.1109/CVPR.2016.90.

  24. Siddon RL. Fast calculation of the exact radiological path for a three-dimensional CT array. Med Phys. 1985;12(2):252–5. https://doi.org/10.1118/1.595715.

    Article  CAS  PubMed  Google Scholar 

  25. Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation, International Conference on Medical Image Computing and Computer-Assisted Intervention (Berlin: Springer) (2015) 234–241. https://doi.org/10.1007/978-3-319-24574-4_28.

  26. Osman AFI, Tamam NM. Deep learning-based convolutional neural network for intramodality brain MRI synthesis. J Appl Clin Med Phys. 2022;23(4): e13530. https://doi.org/10.1002/acm2.13530.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Clark K, Vendt B, Smith K, Freymann J, Kirby J, Koppel P, Moore S, Phillips S, Maffitt D, Pringle M. The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository. J Digit Imaging. 2013;26:1045–57. https://doi.org/10.1007/s10278-013-9622-7.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Zhu Y-M, Cochoff SM, Sukalac R. Automatic patient table removal in CT images. J Digit Imaging. 2012;25(4):480–5. https://doi.org/10.1007/s10278-012-9454-x.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Schneider W, Bortfeld T, Schlegel W. Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions. Phys Med Biol. 2000;45(2):459. https://doi.org/10.1088/0031-9155/45/2/314.

    Article  CAS  PubMed  Google Scholar 

  30. Tucker DM, Barnes GT, Chakraborty DP. Semiempirical model for generating tungsten target x-ray spectra. Med Phys. 1991;18(2):211–8. https://doi.org/10.1118/1.596709.

    Article  CAS  PubMed  Google Scholar 

  31. Bert J, Lemaréchal Y, Benoit D, Garcia M-P, Visvikis D. GGEMS: GPU GEant4-based Monte Carlo simulation platform. Contributionsa la simulation Monte-Carlo pour l’optimisation du traitement en radiothérapie. 2016;69:4987–5006.

    Google Scholar 

  32. Allison J, Amako K, Apostolakis JEA, Araujo H, Dubois PA, Asai M, Barrand G, Capra R, Chauvie S, Chytracek R. Geant4 developments and applications. IEEE Trans Nucl Sci. 2006;53(1):270–8. https://doi.org/10.1109/TNS.2006.869826.

    Article  Google Scholar 

  33. Stepanek J. Electron and positron atomic elastic scattering cross sections. Radiat Phys Chem. 2003;66(2):99–116. https://doi.org/10.1016/S0969-806X(02)00386-9.

    Article  CAS  Google Scholar 

  34. Wang Z, Bovik AC, Sheikh HR, Simoncelli EP. Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process. 2004;13(4):600–12. https://doi.org/10.1109/TIP.2003.819861.

    Article  PubMed  Google Scholar 

  35. Zhou Y, Mathur T, Molloi S. Scatter and veiling glare estimation based on sampled primary intensity. Med Phys. 1999;26(11):2301–10. https://doi.org/10.1118/1.598744.

    Article  CAS  PubMed  Google Scholar 

  36. Zhu L, Wang J, Xing L. Noise suppression in scatter correction for cone-beam CT. Med Phys. 2009;36(3):741–52. https://doi.org/10.1118/1.3063001.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Ning R, Tang X, Conover D. X-ray scatter correction algorithm for cone beam CT imaging. Med Phys. 2004;31(5):1195–202. https://doi.org/10.1118/1.1711475.

    Article  PubMed  Google Scholar 

  38. Zhu L, Strobel N, Fahrig R. X-ray scatter correction for cone-beam CT using moving blocker array. SPIE. 2005;5745:251–8. https://doi.org/10.1117/12.594699.

    Article  Google Scholar 

  39. Zhao, Hang, et al. "Loss functions for image restoration with neural networks." IEEE Transactions on computational imaging 3.1 (2016): 47-57.　https://doi.org/10.1109/TCI.2016.2644865

Download references

Author information

Authors and Affiliations

  1. Graduate School of Science and Engineering, Yamagata University, Yonezawa, 992-8510, Japan

    Masayuki Hattori, Hisato Tsubakiya & Tetsuya Yuasa

  2. Department of Radiology, Yamagata University Hospital, Yamagata, 990-9585, Japan

    Masayuki Hattori & Koji Suzuki

  3. Department of Heavy Particle Medical Science, Graduate School of Medicine, Yamagata University, Yamagata, 990-9585, Japan

    Sung-Hyun Lee & Takayuki Kanai

  4. Department of Radiation Oncology, Tokyo Women’s Medical University, Shinjuku, Tokyo, Japan

    Takayuki Kanai

Authors
  1. Masayuki Hattori
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Hisato Tsubakiya
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sung-Hyun Lee
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Takayuki Kanai
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Koji Suzuki
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Tetsuya Yuasa
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Masayuki Hattori.

Ethics declarations

Conflict of interest

The authors declare that they have no declarations of interest that may have influenced the work reported in this paper.

Ethical approval

Ethical approval was not required because our study only used the data from The Cancer Imaging Archive, which publishes large-scale anonymized datasets of medical images and phantoms.

Additional information

Publisher's Note

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

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hattori, M., Tsubakiya, H., Lee, SH. et al. A deep-learning-based scatter correction with water equivalent path length map for digital radiography. Radiol Phys Technol 17, 488–503 (2024). https://doi.org/10.1007/s12194-024-00807-9

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12194-024-00807-9

Keywords

Search

Navigation