Low-Dose CT Reconstruction Based on Fused State-Space Modelling

Yucong Liu, Zhe Zhao, Dandan Xu, Yusong Lin
Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, PMLR 278:87-93, 2025.

Abstract

Low-dose CT is widely used in medical imaging, but reducing the radiation dose introduces noise that affects image quality. To this end, we propose a low-dose CT reconstruction method based on fused state-space modelling, which uses the FuseSSM module to extract contextual information in the spatial and channel domains, balances short-range and long-range sensitivities, and introducesthe Axial Attention mechanism to reduce the computational complexity, while enhancing the remote-dependent modelling and global texture consistency. The experiments validate the model on the Mayo-2016 dataset, which outperforms the comparative methods in PSNR, SSIM and RMSE metrics, showing good potential for clinical applications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v278-liu25a, title = {Low-Dose CT Reconstruction Based on Fused State-Space Modelling}, author = {Liu, Yucong and Zhao, Zhe and Xu, Dandan and Lin, Yusong}, booktitle = {Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing}, pages = {87--93}, year = {2025}, editor = {Zeng, Nianyin and Pachori, Ram Bilas and Wang, Dongshu}, volume = {278}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v278/main/assets/liu25a/liu25a.pdf}, url = {https://proceedings.mlr.press/v278/liu25a.html}, abstract = {Low-dose CT is widely used in medical imaging, but reducing the radiation dose introduces noise that affects image quality. To this end, we propose a low-dose CT reconstruction method based on fused state-space modelling, which uses the FuseSSM module to extract contextual information in the spatial and channel domains, balances short-range and long-range sensitivities, and introducesthe Axial Attention mechanism to reduce the computational complexity, while enhancing the remote-dependent modelling and global texture consistency. The experiments validate the model on the Mayo-2016 dataset, which outperforms the comparative methods in PSNR, SSIM and RMSE metrics, showing good potential for clinical applications.} }
Endnote
%0 Conference Paper %T Low-Dose CT Reconstruction Based on Fused State-Space Modelling %A Yucong Liu %A Zhe Zhao %A Dandan Xu %A Yusong Lin %B Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing %C Proceedings of Machine Learning Research %D 2025 %E Nianyin Zeng %E Ram Bilas Pachori %E Dongshu Wang %F pmlr-v278-liu25a %I PMLR %P 87--93 %U https://proceedings.mlr.press/v278/liu25a.html %V 278 %X Low-dose CT is widely used in medical imaging, but reducing the radiation dose introduces noise that affects image quality. To this end, we propose a low-dose CT reconstruction method based on fused state-space modelling, which uses the FuseSSM module to extract contextual information in the spatial and channel domains, balances short-range and long-range sensitivities, and introducesthe Axial Attention mechanism to reduce the computational complexity, while enhancing the remote-dependent modelling and global texture consistency. The experiments validate the model on the Mayo-2016 dataset, which outperforms the comparative methods in PSNR, SSIM and RMSE metrics, showing good potential for clinical applications.
APA
Liu, Y., Zhao, Z., Xu, D. & Lin, Y.. (2025). Low-Dose CT Reconstruction Based on Fused State-Space Modelling. Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, in Proceedings of Machine Learning Research 278:87-93 Available from https://proceedings.mlr.press/v278/liu25a.html.

Related Material