Exploring the impact of super-resolution deep learning on MR angiography image quality.

Deep learning Magnetic resonance imaging Retrospective studies

Journal

Neuroradiology
ISSN: 1432-1920
Titre abrégé: Neuroradiology
Pays: Germany
ID NLM: 1302751

Informations de publication

Date de publication:
27 Dec 2023
Historique:
received: 20 09 2023
accepted: 14 12 2023
medline: 27 12 2023
pubmed: 27 12 2023
entrez: 26 12 2023
Statut: aheadofprint

Résumé

The aim of this study is to assess the effect of super-resolution deep learning-based reconstruction (SR-DLR), which uses k-space properties, on image quality of intracranial time-of-flight (TOF) magnetic resonance angiography (MRA) at 3 T. This retrospective study involved 35 patients who underwent intracranial TOF-MRA using a 3-T MRI system with SR-DLR based on k-space properties in October and November 2022. We reconstructed MRA with SR-DLR (matrix = 1008 × 1008) and MRA without SR-DLR (matrix = 336 × 336). We measured the signal-to-noise ratio (SNR), contrast, and contrast-to-noise ratio (CNR) in the basilar artery (BA) and the anterior cerebral artery (ACA) and the sharpness of the posterior cerebral artery (PCA) using the slope of the signal intensity profile curve at the half-peak points. Two radiologists evaluated image noise, artifacts, contrast, sharpness, and overall image quality of the two image types using a 4-point scale. We compared quantitative and qualitative scores between images with and without SR-DLR using the Wilcoxon signed-rank test. The SNRs, contrasts, and CNRs were all significantly higher in images with SR-DLR than those without SR-DLR (p < 0.001). The slope was significantly greater in images with SR-DLR than those without SR-DLR (p < 0.001). The qualitative scores in MRAs with SR-DLR were all significantly higher than MRAs without SR-DLR (p < 0.001). SR-DLR with k-space properties can offer the benefits of increased spatial resolution without the associated drawbacks of longer scan times and reduced SNR and CNR in intracranial MRA.

Identifiants

DOI: 10.1007/s00234-023-03271-1 PMID: 38148334
pubmed: 38148334
doi: 10.1007/s00234-023-03271-1
pii: 10.1007/s00234-023-03271-1
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Informations de copyright

© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

Références

Mandell DM, Mossa-Basha M, Qiao Y et al (2017) Intracranial vessel wall MRI: principles and expert consensus recommendations of the American Society of Neuroradiology. AJNR Am J Neuroradiol 38:218–229. https://doi.org/10.3174/ajnr.A4893
doi: 10.3174/ajnr.A4893 pubmed: 27469212 pmcid: 7963837
Park JE, Jung SC, Lee SH et al (2017) Comparison of 3D magnetic resonance imaging and digital subtraction angiography for intracranial artery stenosis. Eur Radiol 27:4737–4746. https://doi.org/10.1007/s00330-017-4860-6
doi: 10.1007/s00330-017-4860-6 pubmed: 28500366
Willinek WA, Born M, Simon B et al (2003) Time-of-flight MR angiography: comparison of 3.0-T imaging and 1.5-T imaging--initial experience. Radiology 229:913–920. https://doi.org/10.1148/radiol.2293020782
doi: 10.1148/radiol.2293020782 pubmed: 14657322
Meixner CR, Liebig P, Speier P et al (2019) High resolution time-of-flight MR-angiography at 7 T exploiting VERSE saturation, compressed sensing and segmentation. Magn Reson Imaging 63:193–204. https://doi.org/10.1016/j.mri.2019.08.014
doi: 10.1016/j.mri.2019.08.014 pubmed: 31434005
Stalder AF, Schmidt M, Quick HH et al (2015) Highly undersampled contrast-enhanced MRA with iterative reconstruction: integration in a clinical setting. Magn Reson Med 74:1652–1660. https://doi.org/10.1002/mrm.25565
doi: 10.1002/mrm.25565 pubmed: 25522299
Kidoh M, Shinoda K, Kitajima M et al (2020) Deep learning based noise reduction for brain MR imaging: tests on phantoms and healthy volunteers. Magn Reson Med Sci 19:195–206. https://doi.org/10.2463/mrms.mp.2019-0018
doi: 10.2463/mrms.mp.2019-0018 pubmed: 31484849
Uetani H, Nakaura T, Kitajima 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 63:63–71. https://doi.org/10.1007/s00234-020-02513-w
doi: 10.1007/s00234-020-02513-w pubmed: 32794075
Uetani H, Nakaura T, Kitajima M 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 32:4527–4536. https://doi.org/10.1007/s00330-022-08552-6
doi: 10.1007/s00330-022-08552-6 pubmed: 35169896
Higaki T, Nakamura Y, Tatsugami F et al (2019) Improvement of image quality at CT and MRI using deep learning. Jpn J Radiol 37:73–80. https://doi.org/10.1007/s11604-018-0796-2
doi: 10.1007/s11604-018-0796-2 pubmed: 30498876
Yasaka K, Akai H, Sugawara H et al (2022) Impact of deep learning reconstruction on intracranial 1.5 T magnetic resonance angiography. Jpn J Radiol 40:476–483. https://doi.org/10.1007/s11604-021-01225-2
doi: 10.1007/s11604-021-01225-2 pubmed: 34851499
Akai H, Yasaka K, Sugawara H et al (2023) Commercially available deep-learning-reconstruction of MR imaging of the knee at 1.5T has higher image quality than conventionally-reconstructed imaging at 3T: a normal volunteer study. Magn Reson Med Sci 22:353–360. https://doi.org/10.2463/mrms.mp.2022-0020
doi: 10.2463/mrms.mp.2022-0020 pubmed: 35811127
Chaudhari AS, Fang Z, Kogan F et al (2018) Super-resolution musculoskeletal MRI using deep learning. Magn Reson Med 80:2139–2154. https://doi.org/10.1002/mrm.27178
doi: 10.1002/mrm.27178 pubmed: 29582464 pmcid: 6107420
Chaudhari AS, Stevens KJ, Wood JP et al (2020) Utility of deep learning super-resolution in the context of osteoarthritis MRI biomarkers. J Magn Reson Imaging 51:768–779. https://doi.org/10.1002/jmri.26872
doi: 10.1002/jmri.26872 pubmed: 31313397
Pham CH, Tor-Diez C, Meunier H et al (2019) Multiscale brain MRI super-resolution using deep 3D convolutional networks. Comput Med Imaging Graph 77:101647. https://doi.org/10.1016/j.compmedimag.2019.101647
doi: 10.1016/j.compmedimag.2019.101647 pubmed: 31493703
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 4:733–741. https://doi.org/10.1002/jmri.1880040517
doi: 10.1002/jmri.1880040517 pubmed: 7981519
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 14:270–280. https://doi.org/10.1002/jmri.1183
doi: 10.1002/jmri.1183 pubmed: 11536404
Nakaura T, Higaki T, Awai K et al (2020) A primer for understanding radiology articles about machine learning and deep learning. Diagn Interv Imaging 101:765–770. https://doi.org/10.1016/j.diii.2020.10.001
doi: 10.1016/j.diii.2020.10.001 pubmed: 33121910
Mazurowski MA, Buda M, Saha A, Bashir MR (2019) Deep learning in radiology: an overview of the concepts and a survey of the state of the art with focus on MRI. J Magn Reson Imaging 49:939–954. https://doi.org/10.1002/jmri.26534
doi: 10.1002/jmri.26534 pubmed: 30575178
Wicaksono KP, Fujimoto K, Fushimi Y et al (2022) Super-resolution application of generative adversarial network on brain time-of-flight MR angiography: image quality and diagnostic utility evaluation. Eur Radiol. https://doi.org/10.1007/s00330-022-09103-9
Zhang K, Haoji H, Philbrick K et al (2021) SOUP-GAN: super-resolution MRI using generative adversarial networks. Tomography. https://doi.org/10.3390/tomography8020073
Miyazawa H, Natori T, Kameda H et al (2019) Detecting lenticulostriate artery lesions in patients with acute ischemic stroke using high-resolution MRA at 7 T. Int J Stroke 14:290–297. https://doi.org/10.1177/1747493018806163
doi: 10.1177/1747493018806163 pubmed: 30299228
Radojewski P, Slotboom J, Joseph A et al (2021) Clinical implementation of 7T MRI for the identification of incidental intracranial aneurysms versus anatomic variants. Am J Neuroradiol. https://doi.org/10.3174/ajnr.a7331
Koktzoglou I, Huang R, Huang R et al (2021) Super-resolution head and neck MRA using deep machine learning. Magn Reson Med 86:335–345. https://doi.org/10.1002/mrm.28738
doi: 10.1002/mrm.28738 pubmed: 33619802 pmcid: 8034362
Sharma SD, Fong CL, Tzung BS et al (2013) Clinical image quality assessment of accelerated magnetic resonance neuroimaging using compressed sensing. Invest Radiol 48:638–645. https://doi.org/10.1097/RLI.0b013e31828a012d
doi: 10.1097/RLI.0b013e31828a012d pubmed: 23538890

Auteurs

Masamichi Hokamura (M)

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

Hiroyuki Uetani (H)

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

Takeshi Nakaura (T)

Department of Diagnostic Radiology, Graduate School of Medical Sciences, Kumamoto University, Honjo 1-1-1, Chuo-ku, Kumamoto-shi, Kumamoto, 860-8556, Japan. kff00712@nifty.com.
ORCID: 0000-0002-9010-0341

Kensei Matsuo (K)

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

Kosuke Morita (K)

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

Yasunori Nagayama (Y)

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

Masafumi Kidoh (M)

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

Yuichi Yamashita (Y)

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

Mitsuharu Ueda (M)

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

Akitake Mukasa (A)

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

Toshinori Hirai (T)

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

Classifications MeSH