Abstract
A supervised-learning strategy has showed outstanding performance in various fields. However, the application of the supervised learning for restoring digital tomosynthesis (DT) image is limited due to the necessity of a large amount of training data, the high computational cost and the occurrence of structural distortion in an output image. In this study, a multi-agent reinforcement learning (RL) network was developed to provide a denoising model for the DT image, and the effect of the network architecture was evaluated in terms of noise property, quantitative accuracy and training time. The multi-agent RL network consisted of shared, value and policy sub-networks, and the each sub-network was connected with the others in the form of a fully convolutional network (FCN). Five types of the shared sub-network were designed using general convolution layers, dilated convolution layers and residual blocks to modify the network architecture. The network training was implemented to extract image features, output a self-calculated reward and determine an optimal policy for the shared, value and policy sub-networks, respectively. The results showed that the proposed denoising models not only preserved the high quantitative accuracy but also improved the noise property of the output DT image by an average factor of 6.73. The applications of the dilated convolution layer and residual block enhanced the noise reduction capability of the proposed denoising models, and the training efficiency was dependent on the architecture type of the shared sub-network. In conclusion, the proposed model based on the multi-agent RL enables the DT image denoising, and the performance of the denoising model can be optimized by modifying the architecture of the shared sub-network.
Similar content being viewed by others
References
A. Zhong, B. Li, N. Luo, Y. Xu, L. Zhou et al., IEEE Access 8, 112078 (2020)
K. Zhang, W. Zuo, Y. Chen, D. Meng, L. Zhang, IEEE Trans. Image Process. 26, 3142 (2017)
K. Zhang, W. Zuo, L. Zhang, IEEE Trans. Image Process. 27, 4608 (2018)
P. Thanapol, K. Lavangnananda, P. Bouvry, F. Pinel, F. Leprevost, in the International Conference on Information Technology (Chonburi, Thailand, 2020) p. 300
W. Zhao, W. Zhang, X. Pan, P. Zhuang, X. Xie et al., Inf. Sci. 615, 475 (2022)
L. Gao, K. Xie, X. Wu, Z. Lu, C. Li et al., Radiat. Oncol. 16, 1 (2021)
T. Pang, H. Zheng, Y. Quan, H. Ji, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (Nashville, 2021) p. 2043
M. Meng, S. Li, L. Yao, D. Li, M. Zhu, et al., in Proceedings of SPIE Medical Imaging (Houston, 2020) p. 67
E. Levin, R. Pieraccini, W. Echkert, in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (Seattle, 1998) p. 201
R. Furuta, N. Inoue, T. Yamasaki, IEEE Trans. Multimed. 22, 1704 (2019)
V. Mnih, K. Kavukcuoglu, D. Silver, A. Graves, I. Antonoglou, et al., arXiv:1312.5602 (2013). Accessed 2 May 2023
J.D. Williams, K. Asadi, G. Zweig, arXiv:1702.03274 (2017). Accessed 2 May 2023
K. Vassilo, C. Heatwole, T. Taha, A. Mehmood, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (Seattle, 2020) p. 512
J.T. Dobbins III., H.P. McAdams, D.J. Godfrey, C.M. Li, J. Thorac. Imaging 23, 86 (2008)
A. Smith, Fundamentals of breast tomosynthesis, White Paper, Hologic Inc., WP-00007 8 (2008)
T. Nguyen, G. Levy, E. Poncelet, T. Le Thanh, J.F. Prolongeau et al., Diagn. Interv. Imaging 96, 843 (2015)
J.S. Kirby, S.G. Armato, K. Drukker, F. Li, L. Hadjiiski et al., J. Med. Imaging 3, 004506 (2016)
R.L. Siddon, Med. Phys. 12, 252 (1985)
M.A.C. Vieira, P.R. Bakic, A.D.A. Maidment, H. Schiabel, N.D.A. Mascarenhas, in the International Workshop on Digital Mammography (Philadelphia, 2012) p. 268
L.R. Borges, P.R. Bakic, A. Foi, A.D.A. Maidment, M.A.C. Vieira, in Proceedings of SPIE Medical Imaging (Orlando, 2017) p. 36
K. Nam, D. Lee, S. Lee, Phys. Med. Biol. 68, 125006 (2023)
K. He, X. Zhang, S. Ren, J. Sun, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (Las Vegas, 2016) p. 770
K. Zhang, W. Zuo, S. Gu, L. Zhang, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (Honolulu, 2017) p. 3929
V. Mnih, A.P. Badia, M. Mirza, A. Graves, T. Harley, et al. in the International Conference on Machine Learning (2016) p. 1928
D.P. Kingma, J.L. Ba, arXiv:1412.6980 (2014). Accessed 2 May 2023
J.T. Dobbins III., E. Samei, N.T. Ranger, Y. Chen, Med. Phys. 33, 1466 (2006)
Z. Wang, A.C. Bovik, H.R. Sheikh, E.P. Simoncelli, IEEE Trans. Image Process. 13, 600 (2004)
A. Hore, D. Ziou, in the International Conference on Pattern Recognition (Istanbul, Turkey, 2010) p. 2366
Y. Lee, S. Lee, J. Instrum. 10, P10035 (2015)
Y. Bengio, P. Simard, P. Frasconi, IEEE Trans. Neural Netw. 5, 157 (1994)
X. Glorot, Y. Bengio, in Proceedings of the International Conference on Artificial Intelligence and Statistics (Sardinia, Italy, 2010), p. 249
Acknowledgements
This paper was supported by the Konyang University Research Fund in the second half of 2021. And, this work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2023-00211810).
Funding
This paper was supported by the Konyang University Research Fund in the second half of 2021. And, this work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2023–00211810).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The author declares that I have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Nam, K., Lee, S. Effect of network architectures in multi-agent reinforcement learning for denoising digital tomosynthesis images. J. Korean Phys. Soc. 84, 479–487 (2024). https://doi.org/10.1007/s40042-024-01009-7
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40042-024-01009-7