Skip to main content

Advertisement

SpringerLink
Log in

Feasibility study of super-resolution deep learning-based reconstruction using k-space data in brain diffusion-weighted images

  • Diagnostic Neuroradiology
  • Published:
Neuroradiology Aims and scope Submit manuscript

Abstract

Purpose

The purpose of this study is to evaluate the influence of super-resolution deep learning-based reconstruction (SR-DLR), which utilizes k-space data, on the quality of images and the quantitation of the apparent diffusion coefficient (ADC) for diffusion-weighted images (DWI) in brain magnetic resonance imaging (MRI).

Methods

A retrospective analysis was performed on 34 patients who had undergone DWI using a 3 T MRI system with SR-DLR reconstruction based on k-space data in August 2022. DWI was reconstructed with SR-DLR (Matrix = 684 × 684) and without SR-DLR (Matrix = 228 × 228). Measurements were made of the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) in white matter (WM) and grey matter (GM), and the full width at half maximum (FWHM) of the septum pellucidum. Two radiologists assessed image noise, contrast, artifacts, blur, and the overall quality of three image types using a four-point scale. Quantitative and qualitative scores between images with and without SR-DLR were compared using the Wilcoxon signed-rank test.

Results

Images with SR-DLR showed significantly higher SNRs and CNRs than those without SR-DLR (p < 0.001). No statistically significant variances were found in the apparent diffusion coefficients (ADCs) in WM and GM between images with and without SR-DLR (ADC in WM, p = 0.945; ADC in GM, p = 0.235). Moreover, the FWHM without SR-DLR was notably lower compared to that with SR-DLR (p < 0.001).

Conclusion

SR-DLR has the potential to augment the quality of DWI in DL MRI scans without significantly impacting ADC quantitation.

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
Fig. 6

Similar content being viewed by others

Data availability

The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.

Abbreviations

ADC:

Apparent diffusion coefficient

CNR:

Contrast-to-noise ratio

DLR:

Deep learning-based reconstruction

DWI:

Diffusion-weighted imaging

EPI:

Echo-planar imaging

FWHM:

Full width at half maximum

GM:

Grey matter

SD:

Signal intensity

SI:

Standard deviation

SNR:

Signal-to-noise ratio

SR-DLR:

Super-resolution deep learning reconstruction

WM:

White matter

References

  1. Kono K, Inoue Y, Nakayama K, Shakudo M, Morino M, Ohata K et al (2001) The role of diffusion-weighted imaging in patients with brain tumors. AJNR Am J Neuroradiol 22(6):1081–1088

    CAS  PubMed  PubMed Central  Google Scholar 

  2. Okorie CK, Ogbole GI, Owolabi MO, Ogun O, Adeyinka A, Ogunniyi A (2015) Role of diffusion-weighted imaging in acute stroke management using low-field magnetic resonance imaging in resource-limited settings. West Afr J Radiol 22(2):61–66

    Article  PubMed  PubMed Central  Google Scholar 

  3. Butts K, Riederer SJ, Ehman RL, Thompson RM, Jack CR (1994) Interleaved echo planar imaging on a standard MRI system. Magn Reson Med 31(1):67–72

    Article  CAS  PubMed  Google Scholar 

  4. Porter DA, Heidemann RM (2009) High resolution diffusion-weighted imaging using readout-segmented echo-planar imaging, parallel imaging and a two-dimensional navigator-based reacquisition. Magn Reson Med 62(2):468–475

    Article  PubMed  Google Scholar 

  5. Morelli J, Porter D, Ai F, Gerdes C, Saettele M, Feiweier T et al (2013) Clinical evaluation of single-shot and readout-segmented diffusion-weighted imaging in stroke patients at 3 T. Acta Radiol 54(3):299–306

    Article  PubMed  Google Scholar 

  6. Wang Y, Ma X, Zhang Z, Dai E, Jeong HK, Xie B et al (2018) A comparison of readout segmented EPI and interleaved EPI in high-resolution diffusion weighted imaging. Magn Reson Imaging 1(47):39–47

    Article  Google Scholar 

  7. Kidoh M, Shinoda K, Kitajima M, Isogawa K, Nambu M, Uetani H et al (2020) Deep learning based noise reduction for brain MR imaging: tests on phantoms and healthy volunteers. Magn Reson Med Sci 19(3):195–206

    Article  PubMed  Google Scholar 

  8. Uetani H, Nakaura T, Nakaura T, Kitajima M, Morita K, Haraoka Kentaro et al (2022) Hybrid deep-learning-based denoising method for compressed sensing in pituitary MRI: comparison with the conventional wavelet-based denoising method. Eur Radiol 32(7):4527–453

    Article  PubMed  Google Scholar 

  9. Uetani H, Nakaura T, Kitajima M, Yuichi Y, Hamasaki T, Tateishi M et al (2021) A preliminary study of deep learning-based reconstruction specialized for denoising in high-frequency domain: usefulness in high-resolution three-dimensional magnetic resonance cisternography of the cerebellopontine angle. Neuroradiology 63(1):63–71

    Article  PubMed  Google Scholar 

  10. Higaki T, Nakamura Y, Tatsugami F, Nakaura T, Awai K (2019) Improvement of image quality at CT and MRI using deep learning. Jpn J Radiol 37(1):73–80

    Article  PubMed  Google Scholar 

  11. Yasaka K, Akai H, Sugawara H, Tajima T, Akahane M, Yoshioka N et al (2022) Impact of deep learning reconstruction on intracranial 1.5 T magnetic resonance angiography. Jpn J Radiol 40(5):476–83

    Article  PubMed  Google Scholar 

  12. Chaudhari AS, Fang Z, Kogan F, Wood JP, Stevens KJ, Gibbons EK et al (2018) Super-resolution musculoskeletal MRI using deep learning. Magn Reson Med 80(5):2139–2154

    Article  PubMed  PubMed Central  Google Scholar 

  13. Chaudhari AS, Stevens KJ, Wood JP, Chakraborty Amit K, Chakraborty A, Chakraborty Amit et al (2020) Utility of deep learning super-resolution in the context of osteoarthritis MRI biomarkers. J Magn Reson Imaging. 51(3):768–79

    Article  PubMed  Google Scholar 

  14. Pham CH, Tor-Díez C, Meunier H, Bednarek N, Fablet R et al (2019) Multiscale brain MRI super-resolution using deep 3D convolutional networks. Comput Med Imaging Graph 77:101647

    Article  PubMed  Google Scholar 

  15. Du YP, Parker DL, Davis WL, Cao G (1994) Reduction of partial-volume artifacts with zero-filled interpolation in three-dimensional MR angiography. J Magn Reson Imaging 4(5):733–741

    Article  CAS  PubMed  Google Scholar 

  16. Bernstein MA, Fain SB, Riederer SJ (2001) Effect of windowing and zero-filled reconstruction of MRI data on spatial resolution and acquisition strategy. J Magn Reson Imaging 14(3):270–280

    Article  CAS  PubMed  Google Scholar 

  17. Kutsuna H, Uematsu S, Shinoda K (2023) High resolution MR reconstruction with functionally separate neural networks. ISMRM. No. 2922

  18. Basser PJ, Mattiello J, LeBihan D (1994) Estimation of the effective self-diffusion tensor from the NMR spin echo. J Magn Reson B 103(3):247–254

    Article  CAS  PubMed  Google Scholar 

  19. Yoshida M, Nakaura T, Inoue T, Tanoue S, Takada S, Utsunomiya D et al (2018) Magnetic resonance cholangiopancreatography with GRASE sequence at 3.0T: does it improve image quality and acquisition time as compared with 3D TSE? Eur Radiol 28(6):2436–43

    Article  PubMed  Google Scholar 

  20. Chen Z, Pawar K, Ekanayake M, Pain C, Zhong S, Egan GF (2022) Deep learning for image enhancement and correction in magnetic resonance imaging—state-of-the-art and challenges. J Digit Imaging 36(1):204–230

    Article  PubMed  PubMed Central  Google Scholar 

  21. Zhao M, Wei Y, Wong KKL (2022) A generative adversarial network technique for high-quality super-resolution reconstruction of cardiac magnetic resonance images. Magn Reson Imaging 1(85):153–160

    Article  Google Scholar 

  22. Chun J, Chun J, Zhang H, Gach HM, Olberg S, Mazur TR et al (2019) MRI super-resolution reconstruction for MRI-guided adaptive radiotherapy using cascaded deep learning: in the presence of limited training data and unknown translation model. Med Phys 46(9):4148–4164

    Article  PubMed  Google Scholar 

  23. Gibbs JW (1898) Fourier’s Series. Nature 59(1522):200–200

    Article  Google Scholar 

Download references

Acknowledgements

We thank Ms. Tae Hamakawa from Department of Diagnostic Radiology, Kumamoto University, Japan, for her help with the measuring in the quantitative analysis. We thank Mr. Takumi Saito from Canon Medical systems for the adjustment of SR-DLR reconstruction parameters.

Author information

Authors and Affiliations

  1. Department of Central Radiology, Kumamoto University Hospital, Honjo 1-1-1, Kumamoto, 860-8556, Japan

    Kensei Matsuo & Kosuke Morita

  2. Department of Diagnostic Radiology, Graduate School of Medical Sciences, Kumamoto University, Honjo 1-1-1, Chuo, Kumamoto, 860-8556, Japan

    Takeshi Nakaura, Hiroyuki Uetani, Yasunori Nagayama, Masafumi Kidoh, Masamichi Hokamura & Toshinori Hirai

  3. Canon Medical Systems Corporation, 70-1, Yanagi, Saiwai, Kawasaki, Kanagawa, 212-0015, Japan

    Yuichi Yamashita

  4. MRI Systems Division, Canon Medical Systems Corporation, 1385 Shimoishigami, Otawara, Tochigi, 324-8550, Japan

    Kensuke Shinoda

  5. Department of Neurology, Graduate School of Medical Sciences, Kumamoto University, Honjo 1-1-1, Chuo, Kumamoto, 860-8556, Japan

    Mitsuharu Ueda

  6. Department of Neurosurgery, Graduate School of Medical Sciences, Kumamoto University, Honjo 1-1-1, Chuo, Kumamoto, Japan

    Akitake Mukasa

Authors
  1. Kensei Matsuo
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Takeshi Nakaura
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kosuke Morita
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Hiroyuki Uetani
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yasunori Nagayama
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Masafumi Kidoh
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Masamichi Hokamura
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Yuichi Yamashita
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Kensuke Shinoda
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Mitsuharu Ueda
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Akitake Mukasa
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Toshinori Hirai
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Takeshi Nakaura.

Ethics declarations

Conflicts of interest statement

Toshinori Hirai has received research support from Canon Medical Systems. Yuichi Yamahita (MRI Clinical Strategy Group Manager) and Kensuke Shinoda (Senior Engineer) are employees of Canon Medical Systems. They don’t have a fiduciary responsibility there. Canon Medical Systems had no control over the interpretation, writing, or publication of this work. Takeshi Nakaura (Corresponding author) controlled the data.

Ethics approval

This retrospective study was approved by the institutional review board (Kumamoto University, No. 1865).

Informed consent

Informed consent for this retrospective study was waived by the institutional ethics committee.

Additional information

Publisher's note

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

Summary

The findings from this feasibility study indicate that SR-DLR, when utilizing k-space data, may improve the quality of DWI in brain MRI while not significantly affecting the quantitation of ADCs.

Key results

1. Super-resolution deep learning-based reconstruction (SR-DLR) using k-space data significantly improved signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in diffusion-weighted images (DWI) in brain MRI.

2. Apparent diffusion coefficient (ADC) quantitation in the white and grey matter was not significantly affected by the use of SR-DLR in DWI scans.

3. The full width at half maximum (FWHM) of the septum pellucidum was significantly reduced with the use of SR-DLR, indicating a reduction in image blur.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Matsuo, K., Nakaura, T., Morita, K. et al. Feasibility study of super-resolution deep learning-based reconstruction using k-space data in brain diffusion-weighted images. Neuroradiology 65, 1619–1629 (2023). https://doi.org/10.1007/s00234-023-03212-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00234-023-03212-y

Keywords

Search

Navigation