Share：
Share this content in WeChat
X
[Chinese][PDF]6342
• Technical Article •
Application value of convolutional neural network single-shot technique in brain magnetic resonance imaging in poorly coordinated patients
LIU Kai  SUN Haitao  CHEN Caizhong  WANG Jian  WEN Xixi  ZENG Mengsu 

Cite this article as: LIU K, SUN H T, CHEN C Z, et al. Application value of convolutional neural network single-shot technique in brain magnetic resonance imaging in poorly coordinated patients[J]. Chin J Magn Reson Imaging, 2023, 14(1): 111-115. DOI:10.12015/issn.1674-8034.2023.01.020.


[Abstract] Objective To investigate the feasibility of using convolutional neural network single excitation technique (CNN-SS) in magnetic resonance imaging of brain in poorly coordinated patients.Materials and Methods A total of 32 patients with poor T2-FLAIR imaging by conventional parallel imaging (PI) technique were selected, and the CNN-SS technique was used for T2-FLAIR complementary scanning. The image quality, lesion display of the two methods were compared and analyzed.Results The total brain scanning time of T2-FLAIR sequence was 33 seconds in the CNN-SS group and 126 seconds in the PI control group. In terms of image quality, the scores of image quality and lesion sharpness in CNN-SS group were (3.75±0.51) points and (3.68±0.55) points, respectively, significantly improved compared with PI group (2.41±0.71) points and (2.52±0.96) points (P＜0.001).Conclusions The parallel acquisition CNN-SS technique can be effectively applied to cranial magnetic resonance imaging in patients who are unable to cooperate effectively, which can not only significantly shorten the examination time, reduce the image artifacts, and improve the image quality. It is helpful for timely diagnosis of potential lesions, so as to avoid missed diagnosis and delayed treatment.
[Keywords] fluid attenuated inversion recovery；single-shot；convolutional neural network；brain；magnetic resonance imaging；parallel imaging

LIU Kai1   SUN Haitao1*   CHEN Caizhong1   WANG Jian1   WEN Xixi2   ZENG Mengsu1  

1 Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai Institute of Medical Imaging, Shanghai 200032, China

2 Shanghai United Imaging Healthcare Co., Ltd., Shanghai 201807, China

Corresponding author: Sun HT, E-mail: sht1720@163.com

Conflicts of interest   None.

Received  2022-07-29
Accepted  2022-12-20
DOI: 10.12015/issn.1674-8034.2023.01.020
Cite this article as: LIU K, SUN H T, CHEN C Z, et al. Application value of convolutional neural network single-shot technique in brain magnetic resonance imaging in poorly coordinated patients[J]. Chin J Magn Reson Imaging, 2023, 14(1): 111-115. DOI:10.12015/issn.1674-8034.2023.01.020.

[1]
RYU K H, BAEK H J, MOON J I, et al. Initial clinical experience of synthetic MRI as a routine neuroimaging protocol in daily practice: a single-center study[J]. J De Neuroradiol, 2020, 47(2): 151-160. DOI: 10.1016/j.neurad.2019.03.002.
[2]
SZELES D M, MILANO N J, MOSS H J, et al. Brain reserve in a case of cognitive resilience to severe leukoaraiosis[J]. J Int Neuropsychol Soc, 2021, 27(1): 99-108. DOI: 10.1017/S1355617720000569.
[3]
HU F Y, SUN Z H, ZHAO J L, et al. Comparison of MRI characteristics of short-term prognosis in patients with acute cerebral infarction with different degree of leukoaraiosis[J]. J Pract Radiol, 2019, 35(8): 1217-1219. DOI: 10.3969/j.issn.1002-1671.2019.08.004.
[4]
JIANG L, ZHOU L L, AI Z P, et al. Prediction of the onset time of acute stroke by deep learning based on DWI and FLAIR[J]. Chin J Radiol, 2021, 55(8): 811-816. DOI: 10.3760/cma.j.cn112149-20201027-01187.
[5]
ZHAI D Y, ZHU S G, ZHANG W, et al. Infarct morphology assessment in patients with carotid artery/middle cerebral artery occlusion using fast fluid-attenuated inversion recovery (FLAIR) vascular hyperintensity (FVH)[J]. PLoS One, 2017, 12(11): e0188078. DOI: 10.1371/journal.pone.0188078.
[6]
ORLOWSKI H, SHARMA A, ALVI F, et al. Evaluation of 2D FLAIR hyperintensity of the optic nerve and optic nerve head and visual parameters in idiopathic intracranial hypertension[J]. J De Neuroradiol, 2022, 49(2): 193-197. DOI: 10.1016/j.neurad.2021.10.002.
[7]
CHAN H P, SAMALA R K, HADJIISKI L M, et al. Deep learning in medical image analysis[J]. Adv Exp Med Biol, 2020, 1213: 3-21. DOI: 10.1007/978-3-030-33128-3_1.
[8]
CHANDRA S S, BRAN LORENZANA M, LIU X W, et al. Deep learning in magnetic resonance image reconstruction[J]. J Med Imaging Radiat Oncol, 2021, 65(5): 564-577. DOI: 10.1111/1754-9485.13276.
[9]
ZEGERS C M L, POSCH J, TRAVERSO A, et al. Current applications of deep-learning in neuro-oncological MRI[J]. Phys Med, 2021, 83: 161-173. DOI: 10.1016/j.ejmp.2021.03.003.
[10]
WANG Y D, SONG Y, XIE H B, et al. Reconstruction of under-sampled magnetic resonance image based on convolutional neural network[J]. Chin J Magn Reson Imaging, 2018, 9(6): 453-459. DOI: 10.12015/issn.1674-8034.2018.06.010.
[11]
PARK S, GACH H M, KIM S, et al. Autoencoder-inspired convolutional network-based super-resolution method in MRI[J/OL]. IEEE J Transl Eng Health Med, 2021, 9: 1800113 [2022-07-28]. https://ieeexplore.ieee.org/document/9417192 DOI: 10.1109/JTEHM.2021.3076152.
[12]
HAMMERNIK K, KLATZER T, KOBLER E, et al. Learning a variational network for reconstruction of accelerated MRI data[J]. Magn Reson Med, 2018, 79(6): 3055-3071. DOI: 10.1002/mrm.26977.
[13]
SUN L Y, FAN Z W, DING X H, et al. Region-of-interest undersampled MRI reconstruction: a deep convolutional neural network approach[J]. Magn Reson Imaging, 2019, 63: 185-192. DOI: 10.1016/j.mri.2019.07.010.
[14]
OH C, KIM D, CHUNG J Y, et al. A k-space-to-image reconstruction network for MRI using recurrent neural network[J]. Med Phys, 2021, 48(1): 193-203. DOI: 10.1002/mp.14566.
[15]
CIANFONI A, MARTIN M G, DU J, et al. Artifact simulating subarachnoid and intraventricular hemorrhage on single-shot, fast spin-echo fluid-attenuated inversion recovery images caused by head movement: a trap for the unwary[J]. AJNR Am J Neuroradiol, 2006, 27(4): 843-849.
[16]
SHENG R F, ZHENG L Y, JIN K P, et al. Single-breath-hold T2WI liver MRI with deep learning-based reconstruction: a clinical feasibility study in comparison to conventional multi-breath-hold T2WI liver MRI[J]. Magn Reson Imaging, 2021, 81: 75-81. DOI: 10.1016/j.mri.2021.06.014.
[17]
SHIGEMOTO Y, SATO N, KIMURA Y, et al. Postoperative changes in the brain: assessment with serial T2WI/FLAIR MR images in non-neoplastic patients[J]. Epilepsy Res, 2019, 154: 149-151. DOI: 10.1016/j.eplepsyres.2019.05.012.
[18]
SHANG W J, CHEN H B, SHU L M, et al. The association between FLAIR vascular hyperintensity and stroke outcome varies with time from onset[J]. AJNR Am J Neuroradiol, 2019, 40(8): 1317-1322. DOI: 10.3174/ajnr.A6142.
[19]
LOHÖFER F K, KAISSIS G A, RASPER M, et al. Magnetic resonance cholangiopancreatography at 3 Tesla: image quality comparison between 3D compressed sensing and 2D single-shot acquisitions[J]. Eur J Radiol, 2019, 115: 53-58. DOI: 10.1016/j.ejrad.2019.04.002.
[20]
SHANBHOGUE K, TONG A, SMEREKA P, et al. 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[J]. Eur Radiol, 2021, 31(11): 8447-8457. DOI: 10.1007/s00330-021-08008-3.
[21]
WANG Z, WANG J, DETRE J A. Improved data reconstruction method for GRAPPA[J]. Magn Reson Med, 2005, 54(3): 738-742. DOI: 10.1002/mrm.20601.
[22]
MURPHY M, ALLEY M, DEMMEL J, et al. Fast-SPIRiT compressed sensing parallel imaging MRI: scalable parallel implementation and clinically feasible runtime[J]. IEEE Trans Med Imaging, 2012, 31(6): 1250-1262. DOI: 10.1109/TMI.2012.2188039.
[23]
GU J X, WANG Z H, KUEN J, et al. Recent advances in convolutional neural networks[J]. Pattern Recognit, 2018, 77(C): 354-377. DOI: 10.1016/j.patcog.2017.10.013.
[24]
LIU K, CHEN C Z, WEN X X, et al. Comparison of compressed sensing and parallel imaging applied to contrastGenhanced MRI of liver[J]. J Pract Radiol, 2019, 35(10): 1665-1667,1701. DOI: 10.3969/j.issn.1002-1671.2019.10.028.
[25]
IKEDA H, OHNO Y, MURAYAMA K, et al. Compressed sensing and parallel imaging accelerated T2 FSE sequence for head and neck MR imaging: comparison of its utility in routine clinical practice[J/OL]. Eur J Radiol, 2021, 135: 109501 [2022-07-28]. https://www.ejradiology.com/article/S0720-048X(20)30691-4/fulltext. DOI: 10.1016/j.ejrad.2020.109501.
[26]
DELATTRE B M A, BOUDABBOUS S, HANSEN C, et al. Compressed sensing MRI of different organs: ready for clinical daily practice?[J]. Eur Radiol, 2020, 30(1): 308-319. DOI: 10.1007/s00330-019-06319-0.
[27]
DONG C, LOY C C, HE K M, et al. Image super-resolution using deep convolutional networks[J]. IEEE Trans Pattern Anal Mach Intell, 2015, 38(2): 295-307. DOI: 10.1109/TPAMI.2015.2439281.
[28]
KIDOH M, SHINODA K, KITAJIMA M, et al. Deep learning based noise reduction for brain MR imaging: tests on phantoms and healthy volunteers[J]. Magn Reson Med Sci, 2020, 19(3): 195-206. DOI: 10.2463/mrms.mp.2019-0018.
[29]
TIAN C W, FEI L K, ZHENG W X, et al. Deep learning on image denoising: an overview[J]. Neural Netw, 2020, 131: 251-275. DOI: 10.1016/j.neunet.2020.07.025.
[30]
NYBERG E, SANDHU G S, JESBERGER J, et al. Comparison of brain MR images at 1.5T using BLADE and rectilinear techniques for patients who move during data acquisition[J]. AJNR Am J Neuroradiol, 2012, 33(1): 77-82. DOI: 10.3174/ajnr.A2737.

PREV Multimodal MRI manifestations and pathological basis of peritumoral infiltration of glioma in rat
NEXT Feasibility study of fast arterial spin labeling imaging in brain on 3.0 T MRI scanner
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn