Enabling Low-Dose In Vivo Benchtop X-ray Fluorescence Computed Tomography through Deep-Learning-Based Denoising
Abstract
:1. Introduction
2. Materials and Methods
2.1. Experimental Setup
2.2. Deep Blind Image Denoising Model
2.2.1. Dataset
2.2.2. Proposed Model
3. Results
4. Discussion and Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Abbreviations
XFCT | X-ray fluorescence computed tomography |
XRF | X-ray fluorescence |
AI | Artificial Intelligence |
DL | Deep Learning |
SCUNet | Swin-Conv-UNet |
CT | Computed Tomography |
PSNR | Peak signal-to-noise ratio |
SSIM | Structural Similarity Index |
BM3D | Block-Matching and 3D filtering |
BM4D | Block-Matching and 3D filtering |
MSE | Mean Squared Error |
NLM | Non-local means |
DnCNN | Denoising Convolutional Neural Network |
CNN | Convolutional Neural Network |
GAN | Generative Adversarial Networks |
Gd | Gadolinium |
wt | Weight |
Swin | Shifted window Transformer |
keV | Kilo electron volt |
mA | milliampere |
CBCT | Cone-Beam Computed Tomography |
MAP | Maximum A Posteriori |
DRUNet | Dilated-Residual U-Net |
SwinIR | Image Restoration Using Swin Transformer |
SConv | Strided Convolution |
TConv | Transposed Convolution |
RConv | Residual Convolutional |
SwinT | Swin Transformer |
BW | Bin width |
AWGN | Additive White Gaussian Noise |
References
- Staufer, T.; Grüner, F. Review of development and recent advances in biomedical X-ray fluorescence imaging. Int. J. Mol. Sci. 2023, 24, 10990. [Google Scholar] [CrossRef]
- Lim, B.; Son, S.; Kim, H.; Nah, S.; Mu Lee, K. Enhanced deep residual networks for single image super-resolution. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Honolulu, HI, USA, 21–26 July 2017; pp. 136–144. [Google Scholar] [CrossRef]
- Kingma, D.P.; Ba, J. Adam: A method for stochastic optimization. ar**v 2014, ar**v:1412.6980. [Google Scholar]
- Wang, Z.; Cun, X.; Bao, J.; Zhou, W.; Liu, J.; Li, H.U. A General U-Shaped Transformer for Image Restoration. ar**v 2021, ar**v:2106.03106. [Google Scholar]
- Cao, H.; Wang, Y.; Chen, J.; Jiang, D.; Zhang, X.; Tian, Q.; Wang, M. Swin-unet: Unet-like pure transformer for medical image segmentation. In Proceedings of the European Conference on Computer Vision, Tel Aviv, Israel, 23–27 October 2022; Springer: Berlin/Heidelberg, Germany, 2022; pp. 205–218. [Google Scholar]
- Li, Y.; Zhang, K.; Cao, J.; Timofte, R.; Van Gool, L.L. Bringing locality to vision transformers. ar**v 2021, ar**v:2104.05707. [Google Scholar]
- Rehman, A.; Wang, Z. SSIM-based non-local means image denoising. In Proceedings of the 2011 18th IEEE International Conference on Image Processing, Brussels, Belgium, 11–14 September 2011; IEEE: Piscataway, NJ, USA, 2011; pp. 217–220. [Google Scholar] [CrossRef]
- Dabov, K.; Foi, A.; Katkovnik, V.; Egiazarian, K. Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans. Image Process. 2007, 16, 2080–2095. [Google Scholar] [CrossRef] [PubMed]
- Heo, Y.C.; Kim, K.; Lee, Y. Image denoising using non-local means (NLM) approach in magnetic resonance (MR) imaging: A systematic review. Appl. Sci. 2020, 10, 7028. [Google Scholar] [CrossRef]
- Babu, D.; K Jose, S. Review on CNN based image denoising. In Proceedings of the International Conference on Systems, Energy & Environment (ICSEE), Singapore, 3–5 February 2021. [Google Scholar] [CrossRef]
- Zhang, K.; Li, Y.; Liang, J.; Cao, J.; Zhang, Y.; Tang, H.; Fan, D.P.; Timofte, R.; Gool, L.V. Practical blind image denoising via Swin-Conv-UNet and data synthesis. Mach. Intell. Res. 2023, 20, 822–836. [Google Scholar] [CrossRef]
- Liang, T.; **, Y.; Li, Y.; Wang, T. Edcnn: Edge enhancement-based densely connected network with compound loss for low-dose ct denoising. In Proceedings of the 2020 15th IEEE International Conference on Signal Processing (ICSP), Bei**g, China, 6–9 December 2020; IEEE: Piscataway, NJ, USA, 2020; Volume 1, pp. 193–198. [Google Scholar] [CrossRef]
Bin Widths | Noise Level | BM3D | BM4D | NLM | DnCNN | SCUNet | Proposed Model |
---|---|---|---|---|---|---|---|
BW-0.05 | 25% | 13.93 | 31.44 | 27.92 | 36.82 | 22.77 | 29.68 |
50% | 13.91 | 28.61 | 25.16 | 31.07 | 24.51 | 26.97 | |
75% | 13.89 | 29.18 | 24.58 | 24.75 | 25.29 | 31.44 | |
BW-0.1 | 25% | 13.70 | 29.78 | 32.42 | 38.88 | 22.87 | 35.48 |
50% | 13.70 | 31.20 | 28.19 | 34.50 | 27.82 | 29.06 | |
75% | 13.67 | 27.02 | 27.71 | 27.75 | 26.77 | 29.33 | |
BW-0.5 | 25% | 11.76 | 28.18 | 39.94 | 49.35 | 22.54 | 31.81 |
50% | 11.75 | 33.27 | 38.77 | 43.66 | 22.74 | 32.67 | |
75% | 11.75 | 31.22 | 36.75 | 38.40 | 25.58 | 39.05 |
Bin Widths | Noise Level | BM3D | BM4D | NLM | DnCNN | SCUNet | Proposed Model |
---|---|---|---|---|---|---|---|
BW-0.05 | 25% | 0.073 | 0.7902 | 0.7435 | 0.9430 | 0.4661 | 0.7868 |
50% | 0.0720 | 0.7710 | 0.739 | 0.872 | 0.5177 | 0.8369 | |
75% | 0.071 | 0.7917 | 0.7156 | 0.7431 | 0.6472 | 0.8654 | |
BW-0.1 | 25% | 0.0716 | 0.7425 | 0.7985 | 0.8594 | 0.6153 | 0.8284 |
50% | 0.0715 | 0.7154 | 0.7984 | 0.8023 | 0.6206 | 0.8658 | |
75% | 0.0712 | 0.7556 | 0.7801 | 0.7867 | 0.6453 | 0.8218 | |
BW-0.5 | 25% | 0.0617 | 0.8349 | 0.8029 | 0.9139 | 0.5773 | 0.8786 |
50% | 0.0616 | 0.8314 | 0.7626 | 0.9014 | 0.4972 | 0.8466 | |
75% | 0.0615 | 0.8424 | 0.7611 | 0.7582 | 0.5290 | 0.8638 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Mahmoodian, N.; Rezapourian, M.; Inamdar, A.A.; Kumar, K.; Fachet, M.; Hoeschen, C. Enabling Low-Dose In Vivo Benchtop X-ray Fluorescence Computed Tomography through Deep-Learning-Based Denoising. J. Imaging 2024, 10, 127. https://doi.org/10.3390/jimaging10060127
Mahmoodian N, Rezapourian M, Inamdar AA, Kumar K, Fachet M, Hoeschen C. Enabling Low-Dose In Vivo Benchtop X-ray Fluorescence Computed Tomography through Deep-Learning-Based Denoising. Journal of Imaging. 2024; 10(6):127. https://doi.org/10.3390/jimaging10060127
Chicago/Turabian StyleMahmoodian, Naghmeh, Mohammad Rezapourian, Asim Abdulsamad Inamdar, Kunal Kumar, Melanie Fachet, and Christoph Hoeschen. 2024. "Enabling Low-Dose In Vivo Benchtop X-ray Fluorescence Computed Tomography through Deep-Learning-Based Denoising" Journal of Imaging 10, no. 6: 127. https://doi.org/10.3390/jimaging10060127