Skip to main content
SpringerLink
Log in

Accelerating susceptibility-weighted imaging with deep learning by complex-valued convolutional neural network (ComplexNet): validation in clinical brain imaging

  • Imaging Informatics and Artificial Intelligence
  • Published:
European Radiology Aims and scope Submit manuscript

Abstract

Objectives

Susceptibility-weighted imaging (SWI) is crucial for the characterization of intracranial hemorrhage and mineralization, but has the drawback of long acquisition times. We aimed to propose a deep learning model to accelerate SWI, and evaluate the clinical feasibility of this approach.

Methods

A complex-valued convolutional neural network (ComplexNet) was developed to reconstruct high-quality SWI from highly accelerated k-space data. ComplexNet can leverage the inherently complex-valued nature of SWI data and learn richer representations by using complex-valued network. SWI data were acquired from 117 participants who underwent clinical brain MRI examination between 2019 and 2021, including patients with tumor, stroke, hemorrhage, traumatic brain injury, etc. Reconstruction quality was evaluated using quantitative image metrics and image quality scores, including overall image quality, signal-to-noise ratio, sharpness, and artifacts.

Results

The average reconstruction time of ComplexNet was 19 ms per section (1.33 s per participant). ComplexNet achieved significantly improved quantitative image metrics compared to a conventional compressed sensing method and a real-valued network with acceleration rates of 5 and 8 (p < 0.001). Meanwhile, there was no significant difference between fully sampled and ComplexNet approaches in terms of overall image quality and artifacts (p > 0.05) at both acceleration rates. Furthermore, ComplexNet showed comparable diagnostic performance to the fully sampled SWI for visualizing a wide range of pathology, including hemorrhage, cerebral microbleeds, and brain tumor.

Conclusions

ComplexNet can effectively accelerate SWI while providing superior performance in terms of overall image quality and visualization of pathology for routine clinical brain imaging.

Key Points

• The complex-valued convolutional neural network (ComplexNet) allowed fast and high-quality reconstruction of highly accelerated SWI data, with an average reconstruction time of 19 ms per section.

• ComplexNet achieved significantly improved quantitative image metrics compared to a conventional compressed sensing method and a real-valued network with acceleration rates of 5 and 8 (p < 0.001).

• ComplexNet showed comparable diagnostic performance to the fully sampled SWI for visualizing a wide range of pathology, including hemorrhage, cerebral microbleeds, and brain tumor.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Abbreviations

CMBs:

Cerebral microbleeds

CNN:

Convolutional neural network

ComplexNet:

Complex-valued convolutional neural network

CS:

Compressed sensing

GRE:

Gradient echo

MARS:

Microbleed Anatomical Rating Scale

PSNR:

Peak signal-to-noise ratio

R :

Acceleration rate

RealNet:

real-valued convolutional neural network

SSIM:

Structural similarity

SWI:

Susceptibility-weighted imaging

References

  1. Liu C, Li W, Tong KA, Yeom KW, Kuzminski S (2015) Susceptibility-weighted imaging and quantitative susceptibility mapping in the brain. J Magn Reson Imaging 42:23–41

    Article  Google Scholar 

  2. Conklin J, Longo MGF, Cauley SF et al (2019) Validation of highly accelerated wave–CAIPI SWI compared with conventional SWI and T2*-weighted gradient recalled-echo for routine clinical brain MRI at 3T. AJNR Am J Neuroradiol 40:2073–2080

    CAS  PubMed  PubMed Central  Google Scholar 

  3. Haller S, Haacke EM, Thurnher MM, Barkhof F (2021) Susceptibility-weighted imaging: technical essentials and clinical neurologic applications. Radiology 299:3–26

    Article  Google Scholar 

  4. Bilgic B, Ye H, Wald LL, Setsompopa K (2017) Simultaneous time interleaved multislice (STIMS) for rapid susceptibility weighted acquisition. Neuroimage 155:577–586

    Article  Google Scholar 

  5. Yaman B, Hosseini SAH, Moeller S, Ellermann J, Uğurbil K, Akçakaya M (2020) Self-supervised learning of physics-guided reconstruction neural networks without fully-sampled reference data. Magn Reson Med 84:3172–3191

    Article  Google Scholar 

  6. Cole E, Cheng J, Pauly J, Vasanawala S (2021) Analysis of deep complex-valued convolutional neural networks for MRI reconstruction and phase-focused applications. Magn Reson Med 86:1093–1109

    Article  Google Scholar 

  7. Chung MS, Lee EJ, Kim S, Kim SO, Byun JS (2020) Wave-CAIPI susceptibility-weighted imaging achieves diagnostic performance comparable to conventional susceptibility-weighted imaging in half the scan time. Eur Radiol 30:2182–2190

    Article  Google Scholar 

  8. Chen F, Taviani V, Malkiel I et al (2018) Variable-density single-shot fast spin-echo MRI with deep learning reconstruction by using variational networks. Radiology 289:366–373

    Article  Google Scholar 

  9. Duan C, Deng H, Xiao S et al (2019) Fast and accurate reconstruction of human lung gas MRI with deep learning. Magn Reson Med 82:2273–2285

    Article  Google Scholar 

  10. Shanbhogue K, Tong A, Smereka P et al (2021) Accelerated single-shot T2-weighted fat-suppressed (FS) MRI of the liver with deep learning–based image reconstruction: qualitative and quantitative comparison of image quality with conventional T2-weighted FS sequence. Eur Radiol. https://doi.org/10.1007/s00330-021-08008-3

  11. Dedmari MA, Conjeti S, Estrada S, Ehses P, Stöcker T, Reuter M (2018) Complex fully convolutional neural networks for MR image reconstruction. Proceedings of first international workshop, machine learning for medical image reconstruction (MLMIR), at MICCAI 30–38

  12. Wang S, Cheng H, Ying L et al (2020) DeepcomplexMRI: Exploiting deep residual network for fast parallel MR imaging with complex convolution. Magn Reson Imaging 68:136–147

    Article  Google Scholar 

  13. El-Rewaidy H, Neisius U, Mancio J et al (2020) Deep complex convolutional network for fast reconstruction of 3D late gadolinium enhancement cardiac MRI. NMR Biomed 33:e4312

    Article  CAS  Google Scholar 

  14. Gao Y, Cloos M, Liu F, Crozier S, Pike GB, Sun H (2021) Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction. Neuroimage 240:118404

    Article  CAS  Google Scholar 

  15. Sun H, Cleary JO, Glarin R et al (2020) Extracting more for less: multi-echo MP2RAGE for simultaneous T1-weighted imaging, T1 mapping, R2* mapping, SWI, and QSM from a single acquisition. Magn Reson Med 83:1178–1191

    Article  Google Scholar 

  16. Liu F, Samsonov A, Chen L, Kijowski R, Feng L (2019) SANTIS: Sampling-augmented neural network with incoherent structure for MR image reconstruction. Magn Reson Med 82:1890–1904

    Article  Google Scholar 

  17. Schlemper J, Caballero J, Hajnal JV, Price AN, Rueckert D (2018) A deep cascade of convolutional neural networks for dynamic MR image reconstruction. IEEE Trans Med Imaging 37:491–503

    Article  Google Scholar 

  18. Duan C, Deng H, Xiao S et al (2021) Accelerate gas diffusion-weighted MRI for lung morphometry with deep learning. Eur Radiol. https://doi.org/10.1007/s00330-021-08126-y

  19. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. https://arxiv.org/abs/1412.6980. Accessed 22 Dec 2014

  20. Chan KS, Marques JP (2021) SEPIA-susceptibility mapping pipeline tool for phase images. Neuroimage 227:117611

    Article  Google Scholar 

  21. Lustig M, Donoho D, Pauly JM (2007) Sparse MRI: the application of compressed sensing for rapid MR imaging. Magn Reson Med 58:1182–1195

    Article  Google Scholar 

  22. Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13:600–612

    Article  Google Scholar 

  23. Chung MS, Lee EJ, Kim S, Kim SO, Byun JS (2020) Wave-CAIPI susceptibility-weighted imaging achieves diagnostic performance comparable to conventional susceptibility-weighted imaging in half the scan time. Eur Radiol 30:2182–2190

    Article  Google Scholar 

  24. Gregoire S, Chaudhary U, Brown M et al (2009) The microbleed anatomical rating scale (MARS) reliability of a tool to map brain microbleeds. Neurology 73:1759–1766

    Article  CAS  Google Scholar 

  25. Ahn S, Park SH, Lee KH (2013) How to demonstrate similarity by using noninferiority and equivalence statistical testing in radiology research. Radiology 267:328–338

    Article  Google Scholar 

  26. Tanenbaum LN, Tsiouris AJ, Johnson AN et al (2017) Synthetic MRI for clinical neuroimaging: results of the magnetic resonance image compilation (MAGiC) prospective, multicenter, multireader trial. AJNR Am J Neuroradiol 38:1103–1110

    Article  CAS  Google Scholar 

  27. Antun V, Renna F, Poon C, Adcock B, Hansen AC (2020) On instabilities of deep learning in image reconstruction and the potential costs of AI. Proc Natl Acad Sci U S A 117:30088–30095

    Article  CAS  Google Scholar 

  28. Sandino CM, Cheng JY, Chen F, Mardani M, Pauly JM, Vasanawala SS (2020) Compressed sensing: from research to clinical practice with deep neural networks. IEEE Signal Process Mag 37:111–127

    Article  Google Scholar 

  29. Liu S, Buch S, Chen Y et al (2017) Susceptibility-weighted imaging: current status and future directions. NMR Biomed 30:e3552

    Article  Google Scholar 

  30. Zhang Z, Romero A, Muckley MJ, Vincent P, Yang L, Drozdzal M (2019) Reducing uncertainty in undersampled MRI reconstruction with active acquisition. Proceedings of the IEEE conference on computer vision and pattern recognition 2049–2058

  31. Edupuganti V, Mardani M, Vasanawala S, Pauly J (2020) Uncertainty quantification in deep MRI reconstruction. IEEE Trans Med Imaging 40:239–250

    Article  Google Scholar 

  32. Mardani M, Gong E, Cheng JY et al (2018) Deep generative adversarial neural networks for compressive sensing MRI. IEEE Trans Med Imaging 38:167–179

    Article  Google Scholar 

  33. Ledig C, Theis L, Huszár F, Caballero J, Cunningham A (2017) Photorealistic single image super-resolution using a generative adversarial network. Proceedings of the IEEE conference on computer vision and pattern recognition 4681–4690

  34. Quan TM, Nguyen-Duc T, Jeong WK (2018) Compressed sensing MRI reconstruction using a generative adversarial network with a cyclic loss. IEEE Trans Med Imaging 37:1488–1497

    Article  Google Scholar 

  35. Yang Q, Yan P, Zhang Y et al (2018) Low-dose CT image denoising using a generative adversarial network with Wasserstein distance and perceptual loss. IEEE Trans Med Imaging 37:1348–1357

    Article  Google Scholar 

Download references

Funding

This study has received funding by the National Natural Science Foundation of China (81825012, 81730048, 81625011).

Author information

Authors and Affiliations

  1. Department of Radiology, Chinese PLA General Hospital, Beijing, 100853, People’s Republic of China

    Caohui Duan, Yongqin Xiong, Kun Cheng, Jinhao Lyu, Xiangbing Bian, Jing Zhang, Dekang Zhang & Xin Lou

  2. Department of Neurosurgery, Chinese PLA General Hospital, 28 Fuxing Road, Beijing, 100853, People’s Republic of China

    Sa Xiao, Cheng Wang & Ling Chen

  3. Key Laboratory of Magnetic Resonance in Biological Systems, State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, National Center for Magnetic Resonance in Wuhan, Wuhan Institute of Physics and Mathematics, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences–Wuhan National Laboratory for Optoelectronics, Wuhan, 430071, People’s Republic of China

    Xin Zhou

Authors
  1. Caohui Duan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yongqin Xiong
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kun Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sa Xiao
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jinhao Lyu
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Cheng Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Xiangbing Bian
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Jing Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Dekang Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Ling Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Xin Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Xin Lou
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Xin Lou.

Ethics declarations

Ethics approval

Institutional Review Board approval was obtained.

Informed Consent

Written informed consent was obtained from all subjects (patients) in this study.

Conflict of Interest

The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.

Guarantor

The scientific guarantor of this publication is Xin Lou, MD.

Statistics and Biometry

One of the authors has significant statistical expertise.

Methodology

• Prospective

• Observational

• Performed at one institution

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

ESM 1

(DOCX 748 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Duan, C., Xiong, Y., Cheng, K. et al. Accelerating susceptibility-weighted imaging with deep learning by complex-valued convolutional neural network (ComplexNet): validation in clinical brain imaging. Eur Radiol 32, 5679–5687 (2022). https://doi.org/10.1007/s00330-022-08638-1

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-022-08638-1

Keywords

Search

Navigation