Research progress on quantitative magnetic susceptibility imaging reconstruction method based on improved U-network model_Journal of Biomedical Engineering

Authors：

 YANG Wenyang ¹ , ZHANG Ruijie ¹ , KEUNG Steven ²

1. College of Computer, Xi'an Shiyou University, Xi’an 710065, P. R. China;
2. Department of Medical Imaging and Radiation Sciences, Monash University, Clayton, VIC 3800, Australian;

Corresponding?author：

YANG Wenyang, Email: ywy80910@163.com

Keywords：

Quantitative magnetic susceptibility image; improved U-network model; Auxiliary diagnosis; Magnetic susceptibility

DOI：

10.7507/1001-5515.202412074

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Quantitative magnetic susceptibility imaging (QSM) is an imaging method based on magnetic resonance imaging (MRI) phase signal processing and inversion to obtain tissue magnetic susceptibility distribution, which can generate images reflecting the magnetic characteristics of tissues. QSM reconstruction process is complex, in which dipole inversion stage is the most challenging and decisive link, and traditional methods are easily affected by pathological conditions at this stage, resulting in artifacts and deviations. With the development of deep learning and machine vision technology, using U-network (U-Net) model to improve dipole inversion process can effectively avoid the shortcomings of traditional algorithms. In this paper, the application of the improved model based on U-Net architecture in dipole inversion from 2020 to now is summarized. Firstly, the theoretical concept of QSM is introduced. Secondly, the existing improved models based on U-Net architecture are divided into three categories: improved U-Net based on structural optimization, improved U-Net based on physical constraints and improved U-Net based on improving generalization ability, and their main characteristics and design starting points are sorted out. Finally, the development trend of the future model is prospected and summarized. To sum up, it is expected that the difficulties and challenges of dipole inversion will be solved, the accuracy of QSM images will be improved, and support for disease-aided diagnosis will be provided by summarizing and comparing different improved U-Net models in this paper.

1.	Ahmed M, Chen J, Arvin A, et al. The diamagnetic component map from quantitative susceptibility mapping (QSM) source separation reveals pathological alteration in Alzheimer's disease-driven neurodegeneration. NeuroImage, 2023, 280: 120357.
2.	Padmapriya T, Sriramakrishnan P, Kalaiselvi T, et al. Advancements of MRI-based brain tumor segmentation from traditional to recent trends: a review. Curr Med Imaging, 2022, 18(12): 1261-1275.
3.	Bian B, Hou L, Chai Y, et al. Visualizing the habenula using 3T high-resolution MP2RAGE and QSM: a preliminary study. Am J Neuroradiol, 2024, 45(4): 504-510.
4.	Huang C, Li J, Liu C, et al. Investigation of brain iron levels in Chinese patients with Alzheimer's disease. Front Aging Neurosci, 2023, 15: 1168845.
5.	Chen M, Wang Y, Zhang C, et al. Free water and iron content in the substantia nigra at different stages of Parkinson's disease. Eur J Radiol, 2023, 167: 111030.
6.	王瑞奇, 詹逸珺, 裴建. 定量磁化率成像在阿爾茨海默病診斷及病程進展跟蹤中的應用研究. 磁共振成像, 2024, 15(5): 187-191.
7.	Neupane B, Aryal J, Rajabifard A. Dual skip connections in U-Net, ResUnet and U-Net3+ for remote extraction of buildings. arXiv preprint, 2023, arXiv: 2303.09064.
8.	Jebin B M, Shyla S I, Bel K N S, et al. Brain tumor detection and classification using U-Net and CNN with brain texture pattern analysis. Biomedical Signal Processing and Control, 2025, 110: 108156.
9.	Gkotsoulias D G, J?ger C, Müller R, et al. Chaos and COSMOS-considerations on QSM methods with multiple and single orientations and effects from local anisotropy. Magn Reson Imaging, 2024, 110: 104-111.
10.	Beljaards L, Pezzotti N, Rao C, et al. AI-based motion artifact severity estimation in undersampled MRI allowing for selection of appropriate reconstruction models. Med Phys, 2024, 51(5): 3555-3565.
11.	Liu J, Koch K M. Meta-QSM: an image-resolution-arbitrary network for QSM reconstruction. arXiv preprint, 2019, arXiv: 1908.00206.
12.	Gao Y, Zhu X, Moffat B A, et al. xQSM: quantitative susceptibility mapping with octave convolutional and noise‐regularized neural networks. NMR Biomed, 2021, 34(3): e4461.
13.	Zhu X, Gao Y, Liu F, et al. BFRnet: a deep learning-based MR background field removal method for QSM of the brain containing significant pathological susceptibility sources. Z Med Phys, 2023, 33(4): 578-590.
14.	Graf S, Wohlgemuth W A, Deistung A. Incorporating a-priori information in deep learning models for quantitative susceptibility mapping via adaptive convolution. Front Neurosci, 2024, 18: 1366165.
15.	Gao Y, Xiong Z, Fazlollahi A, et al. Instant tissue field and magnetic susceptibility mapping from MRI raw phase using Laplacian enhanced deep neural networks. NeuroImage, 2022, 259: 119410.
16.	Gao Y, Cloos M, Liu F, et al. Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction. NeuroImage, 2021, 240: 118404.
17.	Si W, Guo Y, Zhang Q, et al. Quantitative susceptibility mapping using multi-channel convolutional neural networks with dipole-adaptive multi-frequency inputs. Front Neurosci, 2023, 17: 1165446.
18.	馬靚怡, 喬志偉. 基于 Chambolle-Pock 框架的核 TV 多通道圖像重建算法. CT 理論與應用研究, 2022, 31(6): 731-747.
19.	Zhang M, Li M, Zhou J, et al. High-dimensional embedding network derived prior for compressive sensing MRI reconstruction. Med Image Anal, 2020, 64: 101717.
20.	Xiong Z, Gao Y, Liu Y, et al. Quantitative susceptibility mapping through model-based deep image prior (MoDIP). NeuroImage, 2024, 291: 120583.
21.	Liu J, Koch K M. Model-based learning for quantitative susceptibility mapping. arXiv preprint, 2020, arXiv: 2004.06259.
22.	Lai K W, Aggarwal M, van Zijl P, et al. Learned proximal networks for quantitative susceptibility mapping. Med Image Comput Comput Assist Interv, 2020, 12262: 125-135.
23.	Venkatesh V, Mathew R S, Yalavarthy P K. Spinet-QSM: model-based deep learning with schatten p-norm regularization for improved quantitative susceptibility mapping. Magnetic Resonance Materials in Physics, Biology and Medicine, 2024, 37(3): 411-427.
24.	Feng R, Zhao J, Wang H, et al. MoDL-QSM: model-based deep learning for quantitative susceptibility mapping. NeuroImage, 2021, 240: 118376.
25.	Jung W, Yoon J, Ji S, et al. Exploring linearity of deep neural network trained QSM: QSMnet+. NeuroImage, 2020, 211: 116619.
26.	Kan H, Uchida Y, Kawaguchi S, et al. Quantitative susceptibility mapping for susceptibility source separation with adaptive relaxometric constant estimation (QSM-ARCS) from solely gradient-echo data. NeuroImage, 2024, 296: 120676.
27.	Gao Y, Xiong Z, Shan S, et al. Plug-and-play latent feature editing for orientation-adaptive quantitative susceptibility mapping neural networks. Medical Image Analysis, 2024, 94: 103160.
28.	Xiong Z, Gao Y, Liu F, et al. Affine transformation edited and refined deep neural network for quantitative susceptibility mapping. NeuroImage, 2023, 267: 119842.
29.	Li J, Wang W, Chen C, et al. TransBTSV2: towards better and more efficient volumetric segmentation of medical images. arXiv preprint, 2022, arXiv: 2201.12785.
30.	Zhang X, Yang S, Jiang Y, et al. FAFS-UNet: redesigning skip connections in UNet with feature aggregation and feature selection. Computers in Biology and Medicine, 2024, 170: 108009.
31.	Guan X, Zhao Y, Nyatega C O, et al. Brain tumor segmentation network with multi-view ensemble discrimination and kernel-sharing dilated convolution. Brain Sciences, 2023, 13(4): 650.
32.	張奕涵, 柏正堯, 尤逸琳, 等. 自適應模態融合雙編碼器MRI腦腫瘤分割網絡. 中國圖象圖形學報, 2024, 29(3): 768-781.
33.	Shi Y, Feng R, Li Z, et al. Towards in vivo ground truth susceptibility for single-orientation deep learning QSM: a multi-orientation gradient-echo MRI dataset. NeuroImage, 2022, 261: 119522.
34.	趙艷紅, 張曉文, 張濤, 等. MRI定量磁化率成像(QSM)在帕金森病診斷中的應用研究. 影像研究與醫學應用, 2024, 8(10): 16-19.
35.	Hanspach J, Bollmann S, Grigo J, et al. Deep learning-based quantitative susceptibility mapping (QSM) in the presence of fat using synthetically generated multi-echo phase training data. Magnetic Resonance in Medicine, 2022, 88(4): 1548-1560.
36.	Bechler E, Stabinska J, Thiel T, et al. Feasibility of quantitative susceptibility mapping (QSM) of the human kidney. Magnetic Resonance Materials in Physics, Biology and Medicine, 2021, 34(3): 389-397.

1. Ahmed M, Chen J, Arvin A, et al. The diamagnetic component map from quantitative susceptibility mapping (QSM) source separation reveals pathological alteration in Alzheimer's disease-driven neurodegeneration. NeuroImage, 2023, 280: 120357.
2. Padmapriya T, Sriramakrishnan P, Kalaiselvi T, et al. Advancements of MRI-based brain tumor segmentation from traditional to recent trends: a review. Curr Med Imaging, 2022, 18(12): 1261-1275.
3. Bian B, Hou L, Chai Y, et al. Visualizing the habenula using 3T high-resolution MP2RAGE and QSM: a preliminary study. Am J Neuroradiol, 2024, 45(4): 504-510.
4. Huang C, Li J, Liu C, et al. Investigation of brain iron levels in Chinese patients with Alzheimer's disease. Front Aging Neurosci, 2023, 15: 1168845.
5. Chen M, Wang Y, Zhang C, et al. Free water and iron content in the substantia nigra at different stages of Parkinson's disease. Eur J Radiol, 2023, 167: 111030.
6. 王瑞奇, 詹逸珺, 裴建. 定量磁化率成像在阿爾茨海默病診斷及病程進展跟蹤中的應用研究. 磁共振成像, 2024, 15(5): 187-191.
7. Neupane B, Aryal J, Rajabifard A. Dual skip connections in U-Net, ResUnet and U-Net3+ for remote extraction of buildings. arXiv preprint, 2023, arXiv: 2303.09064.
8. Jebin B M, Shyla S I, Bel K N S, et al. Brain tumor detection and classification using U-Net and CNN with brain texture pattern analysis. Biomedical Signal Processing and Control, 2025, 110: 108156.
9. Gkotsoulias D G, J?ger C, Müller R, et al. Chaos and COSMOS-considerations on QSM methods with multiple and single orientations and effects from local anisotropy. Magn Reson Imaging, 2024, 110: 104-111.
10. Beljaards L, Pezzotti N, Rao C, et al. AI-based motion artifact severity estimation in undersampled MRI allowing for selection of appropriate reconstruction models. Med Phys, 2024, 51(5): 3555-3565.
11. Liu J, Koch K M. Meta-QSM: an image-resolution-arbitrary network for QSM reconstruction. arXiv preprint, 2019, arXiv: 1908.00206.
12. Gao Y, Zhu X, Moffat B A, et al. xQSM: quantitative susceptibility mapping with octave convolutional and noise‐regularized neural networks. NMR Biomed, 2021, 34(3): e4461.
13. Zhu X, Gao Y, Liu F, et al. BFRnet: a deep learning-based MR background field removal method for QSM of the brain containing significant pathological susceptibility sources. Z Med Phys, 2023, 33(4): 578-590.
14. Graf S, Wohlgemuth W A, Deistung A. Incorporating a-priori information in deep learning models for quantitative susceptibility mapping via adaptive convolution. Front Neurosci, 2024, 18: 1366165.
15. Gao Y, Xiong Z, Fazlollahi A, et al. Instant tissue field and magnetic susceptibility mapping from MRI raw phase using Laplacian enhanced deep neural networks. NeuroImage, 2022, 259: 119410.
16. Gao Y, Cloos M, Liu F, et al. Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction. NeuroImage, 2021, 240: 118404.
17. Si W, Guo Y, Zhang Q, et al. Quantitative susceptibility mapping using multi-channel convolutional neural networks with dipole-adaptive multi-frequency inputs. Front Neurosci, 2023, 17: 1165446.
18. 馬靚怡, 喬志偉. 基于 Chambolle-Pock 框架的核 TV 多通道圖像重建算法. CT 理論與應用研究, 2022, 31(6): 731-747.
19. Zhang M, Li M, Zhou J, et al. High-dimensional embedding network derived prior for compressive sensing MRI reconstruction. Med Image Anal, 2020, 64: 101717.
20. Xiong Z, Gao Y, Liu Y, et al. Quantitative susceptibility mapping through model-based deep image prior (MoDIP). NeuroImage, 2024, 291: 120583.
21. Liu J, Koch K M. Model-based learning for quantitative susceptibility mapping. arXiv preprint, 2020, arXiv: 2004.06259.
22. Lai K W, Aggarwal M, van Zijl P, et al. Learned proximal networks for quantitative susceptibility mapping. Med Image Comput Comput Assist Interv, 2020, 12262: 125-135.
23. Venkatesh V, Mathew R S, Yalavarthy P K. Spinet-QSM: model-based deep learning with schatten p-norm regularization for improved quantitative susceptibility mapping. Magnetic Resonance Materials in Physics, Biology and Medicine, 2024, 37(3): 411-427.
24. Feng R, Zhao J, Wang H, et al. MoDL-QSM: model-based deep learning for quantitative susceptibility mapping. NeuroImage, 2021, 240: 118376.
25. Jung W, Yoon J, Ji S, et al. Exploring linearity of deep neural network trained QSM: QSMnet+. NeuroImage, 2020, 211: 116619.
26. Kan H, Uchida Y, Kawaguchi S, et al. Quantitative susceptibility mapping for susceptibility source separation with adaptive relaxometric constant estimation (QSM-ARCS) from solely gradient-echo data. NeuroImage, 2024, 296: 120676.
27. Gao Y, Xiong Z, Shan S, et al. Plug-and-play latent feature editing for orientation-adaptive quantitative susceptibility mapping neural networks. Medical Image Analysis, 2024, 94: 103160.
28. Xiong Z, Gao Y, Liu F, et al. Affine transformation edited and refined deep neural network for quantitative susceptibility mapping. NeuroImage, 2023, 267: 119842.
29. Li J, Wang W, Chen C, et al. TransBTSV2: towards better and more efficient volumetric segmentation of medical images. arXiv preprint, 2022, arXiv: 2201.12785.
30. Zhang X, Yang S, Jiang Y, et al. FAFS-UNet: redesigning skip connections in UNet with feature aggregation and feature selection. Computers in Biology and Medicine, 2024, 170: 108009.
31. Guan X, Zhao Y, Nyatega C O, et al. Brain tumor segmentation network with multi-view ensemble discrimination and kernel-sharing dilated convolution. Brain Sciences, 2023, 13(4): 650.
32. 張奕涵, 柏正堯, 尤逸琳, 等. 自適應模態融合雙編碼器MRI腦腫瘤分割網絡. 中國圖象圖形學報, 2024, 29(3): 768-781.
33. Shi Y, Feng R, Li Z, et al. Towards in vivo ground truth susceptibility for single-orientation deep learning QSM: a multi-orientation gradient-echo MRI dataset. NeuroImage, 2022, 261: 119522.
34. 趙艷紅, 張曉文, 張濤, 等. MRI定量磁化率成像(QSM)在帕金森病診斷中的應用研究. 影像研究與醫學應用, 2024, 8(10): 16-19.
35. Hanspach J, Bollmann S, Grigo J, et al. Deep learning-based quantitative susceptibility mapping (QSM) in the presence of fat using synthetically generated multi-echo phase training data. Magnetic Resonance in Medicine, 2022, 88(4): 1548-1560.
36. Bechler E, Stabinska J, Thiel T, et al. Feasibility of quantitative susceptibility mapping (QSM) of the human kidney. Magnetic Resonance Materials in Physics, Biology and Medicine, 2021, 34(3): 389-397.

Journal of Biomedical Engineering

Latest ArticlesResearch progress on quantitative magnetic susceptibility imaging reconstruction method based on improved U-network model

Abstract Full text Figures/Tables Video References Cited by

Format

Content