Abstract
Medical image superresolution models are typically trained on synthetic low-resolution/high-resolution image pairs. This is easier and does not reflect the real-world challenge of learning superresolution for clinical CT scans from micro-CT. Furthermore, most medical image superresolution models are evaluated using standard image similarity metrics like Peak-Signal-to-Noise-Ratio (PSNR) and Structural Similarity Index Measure (SSIM), even though it is well known, that they favor image characteristics that are not suitable in a medical image setting. In this paper, we investigate how training on real or synthetic data affects model performance. We also evaluate how models perform, when using measures that are relevant for clinical bone analysis. We train two well-established superresolution models, SRGAN and ESRGAN on two real-world multiscale human bone CT datasets. We compare this approach with training the models on synthetic data, where low-resolution images are produced by subsampling and blurring high-resolution images, as is the common approach when assessing superresolution architectures for medical images. When evaluating performance, we calculate both clinically relevant bone measures, Bone Volume Fraction, Thickness and Degree of Anisotropy as well as PSNR and SSIM. We find that training on real data generally makes the model perform worse than when training on synthetic data. This suggests that more can be learned from using real-world data when exploring medical image superresolution. It also seems that the similarity metrics do not always align with the bone measures in terms of lowest error, indicating that including the bone measures in our evaluation gives a fuller picture of how our models perform.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Domander, R., Felder, A.A., Doube, M.: BoneJ2 - refactoring established research software. Wellcome Open Res. 6, 37 (2021). https://doi.org/10.12688/wellcomeopenres.16619.2
Du, W., Tian, S.: Transformer and GAN-based super-resolution reconstruction network for medical images. Tsinghua Sci. Technol. 29(1), 197–206 (2024). https://doi.org/10.26599/TST.2022.9010071
Frazer, L., Louis, N., Zbijewski, W., Vaishnav, J., Nicolella, D.: Super-resolution of clinical computed tomography to improve fracture risk prediction. J. Bone Mineral Res. 37, S235 (2022). https://doi.org/10.1002/jbmr.4515
Georgescu, M.I., Ionescu, R.T., Verga, N.: Convolutional neural networks with intermediate loss for 3D super-resolution of CT and MRI scans. IEEE Access 8, 49112–49124 (2020). https://doi.org/10.1109/ACCESS.2020.2980266
Jhuboo, R., Redko, I., Guignandon, A., Peyrin, F., Sebban, M.: Why do state-of-the-art super-resolution methods not work well for bone microstructure CT imaging? In: 2022 30th European Signal Processing Conference (EUSIPCO), pp. 1283–1287 (2022). https://doi.org/10.23919/EUSIPCO55093.2022.9909945
Kalluvila, A.: Super-resolution of brain MRI via U-Net architecture. Int. J. Adv. Comput. Sci. Appl. 14(5) (2023). https://doi.org/10.14569/IJACSA.2023.0140503
Ledig, C., et al.: Photo-realistic single image super-resolution using a generative adversarial network. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 105–114. IEEE Computer Society, Los Alamitos, CA, USA, July 2017. https://doi.org/10.1109/CVPR.2017.19
Li, G., et al.: Transformer-empowered multi-scale contextual matching and aggregation for multi-contrast MRI super-resolution. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 20604–20613. IEEE Computer Society, Los Alamitos, CA, USA, June 2022. https://doi.org/10.1109/CVPR52688.2022.01998
Li, Y., Sixou, B., Peyrin, F.: A review of the deep learning methods for medical images super resolution problems. IRBM 42(2), 120–133 (2021)
Liu, X., Su, S., Gu, W., Yao, T., Shen, J., Mo, Y.: Super-resolution reconstruction of CT images based on multi-scale information fused generative adversarial networks. Ann. Biomed. Eng. 52, 57–70 (2023). https://doi.org/10.1007/s10439-023-03412-w
Lu, W., et al.: Asymmetric convolution Swin transformer for medical image super-resolution. Alexandria Eng. J. 85, 177–184 (2023). https://doi.org/10.1016/j.aej.2023.11.044
Molino, G., Montalbano, G., Pontremoli, C., Fiorilli, S., Vitale-Brovarone, C.: Imaging techniques for the assessment of the bone osteoporosis-induced variations with particular focus on micro-CT potential. Appl. Sci. (Switzerland) 10, 1–27 (2020). https://doi.org/10.3390/app10248939
Qiu, D., Zheng, L., Zhu, J., Huang, D.: Multiple improved residual networks for medical image super-resolution. Future Gener. Comput. Syst. 116, 200–208 (2021). https://doi.org/10.1016/j.future.2020.11.001
Rajendran, K., et al.: First clinical photon-counting detector CT system: technical evaluation. Radiology 303(1), 130–138 (2022). https://doi.org/10.1148/radiol.212579
Sadoughi, S., Subramanian, A., Ramil, G., Burghardt, A.J., Kazakia, G.J.: A Laplace-Hamming binarization approach for second-generation HR-PQCT rescues fine feature segmentation. J. Bone Mineral Res. 38, 1006–1014 (2023). https://doi.org/10.1002/jbmr.4819
Umirzakova, S., Mardieva, S., Muksimova, S., Ahmad, S., Whangbo, T.: Enhancing the super-resolution of medical images: introducing the deep residual feature distillation channel attention network for optimized performance and efficiency. Bioengineering 10, 1332 (2023). https://doi.org/10.3390/bioengineering10111332
Wang, C., Lv, X., Shao, M., Qian, Y., Zhang, Y.: A novel fuzzy hierarchical fusion attention convolution neural network for medical image super-resolution reconstruction. Inf. Sci. 622, 424–436 (2023). https://doi.org/10.1016/j.ins.2022.11.140. No info about how they produce their LR images. I assume simple interpolation
Wang, X., et al.: ESRGAN: enhanced super-resolution generative adversarial networks. In: Leal-Taixé, L., Roth, S. (eds.) ECCV 2018. LNCS, vol. 11133, pp. 63–79. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-11021-5_5
Yang, K., et al.: Residual feature attentional fusion network for lightweight chest CT image super-resolution. Comput. Mater. Continua 75, 5159–5176 (2023). https://doi.org/10.32604/cmc.2023.036401
Zhao, D., Wang, W., Xiao, Z., Zhang, F.: Super-resolution reconstruction of medical images based on deep residual attention network. Multimedia Tools Appl. (2023). https://doi.org/10.1007/s11042-023-16478-4. No information about how they construct their LR images. I assume simple interpolation
Zhu, J., Tan, C., Yang, J., Yang, G., Lio, P.: Arbitrary scale super-resolution for medical images. Int. J. Neural Syst. 31, 2150037 (2021). https://doi.org/10.1142/S0129065721500374
Acknowledgements
The ESS lighthouse on hard materials in 3D, SOLID, funded by the Danish Agency for Science and Higher Education supported the project.
Laboratory of biological anthropology, ArchaeoScience, Globe Institute, University of Copenhagen and the National Museum of Denmark provided the archaeological bones used to produce the FACTS dataset.
The Bone Quality Research Lab (BQRL) at University of California, San Francisco provided access to the XtremeCT dataset
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Bardenfleth, S.W., Dahl, V.A., Villa, C., Kazakia, G., Dahl, A.B. (2024). Superresolution of Real-World Multiscale Bone CT Verified with Clinical Bone Measures. In: Yap, M.H., Kendrick, C., Behera, A., Cootes, T., Zwiggelaar, R. (eds) Medical Image Understanding and Analysis. MIUA 2024. Lecture Notes in Computer Science, vol 14860. Springer, Cham. https://doi.org/10.1007/978-3-031-66958-3_12
Download citation
DOI: https://doi.org/10.1007/978-3-031-66958-3_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-66957-6
Online ISBN: 978-3-031-66958-3
eBook Packages: Computer ScienceComputer Science (R0)