Share：
Share this content in WeChat
X
[Chinese][PDF]691
• Technical Article •
Value of T2 Flair sequence based on deep learning in improving image quality of white matter hyperintensities
ZHAO Rusheng  XU Lulu  LI Qing  XU Yicheng  ZHANG Jiulou  RONG Fanling 

Cite this article as: ZHAO R S, XU L L, LI Q, et al. Value of T2 Flair sequence based on deep learning in improving image quality of white matter hyperintensities[J]. Chin J Magn Reson Imaging, 2024, 15(11): 117-122. DOI:10.12015/issn.1674-8034.2024.11.018.


[Abstract] Objective To explore the application value of T2 fluid-attenuated inversion recovery (Flair) sequence based on deep learning reconstruction (DLR) algorithm in improving the image quality of white matter hyperintensities (WMH).Materials and Methods Fifty patients with suspected cerebral ischemic disease were prospectively recruited. Both the conventional T2 FLAIR sequence and the high-resolution T2 Flair sequence, utilizing the DLR algorithm, were conducted on the patients. The DLR Flair sequence selected for this study retained the pre-processed images that have undergone conventional reconstruction algorithms without DLR processing (referred to as Pre-DLR). Subjective evaluations were performed on three groups of images using a 4-point scale to assess image sharpness, gray-white matter contrast, cerebrospinal fluid-choroid plexus contrast, WMH display, and overall image quality. Comparisons were made between the number of WMH detections, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of WMH in three sets of images.Results In the subjective evaluation, the DLR group of images scored higher than the conventional group and Pre-DLR group in terms of image sharpness, gray-white matter contrast, cerebrospinal fluid-choroid plexus contrast, WMH display, and overall image quality (all P<0.05). In terms of WMH counting, the DLR group identified a higher number of WMHs than the conventional group (P<0.05), while there was no statistical difference with the Pre-DLR group. In the objective evaluation, the DLR group showed higher SNR and CNR of WMH compared to the conventional group and Pre-DLR group (all P<0.05).Conclusions Compared to conventional sequences, the high-resolution T2 Flair sequence combined with the DLR algorithm can improve WMH image quality and detect more subtle WMH lesions without increasing scan time.
[Keywords] white matter hyperintensities；magnetic resonance imaging；deep learning；reconstruction algorithm；high resolution；image quality

ZHAO Rusheng1   XU Lulu1   LI Qing1   XU Yicheng2   ZHANG Jiulou1   RONG Fanling1*  

1 Department of Radiology, the First Affiliated Hospital of Nanjing Medical University, Nanjing210029, China

2 Siemens Healthineers Ltd., Shanghai200126, China

Corresponding author: RONG F L, E-mail: flrong1204@163.com

Conflicts of interest   None.

Received  2024-07-08
Accepted  2024-11-05
DOI: 10.12015/issn.1674-8034.2024.11.018
Cite this article as: ZHAO R S, XU L L, LI Q, et al. Value of T2 Flair sequence based on deep learning in improving image quality of white matter hyperintensities[J]. Chin J Magn Reson Imaging, 2024, 15(11): 117-122. DOI:10.12015/issn.1674-8034.2024.11.018.

[1]
ALBER J, ALLADI S, BAE H J, et al. White matter hyperintensities in vascular contributions to cognitive impairment and dementia (VCID): knowledge gaps and opportunities[J/OL]. Alzheimers Dement, 2019, 5: 107-117 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/31011621/. DOI: 10.1016/j.trci.2019.02.001.
[2]
PARENT O, BUSSY A, DEVENYI G A, et al. Assessment of white matter hyperintensity severity using multimodal magnetic resonance imaging[J/OL]. Brain Commun, 2023, 5(6): fcad279 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/37953840/. DOI: 10.1093/braincomms/fcad279.
[3]
RYU W S, WOO S H, SCHELLINGERHOUT D, et al. Stroke outcomes are worse with larger leukoaraiosis volumes[J]. Brain, 2017, 140(1): 158-170. DOI: 10.1093/brain/aww259.
[4]
ZHANG S, HU Y Y, YANG H L, et al. Value of white matter hyperintensity volume and total white matter volume for evaluating cognitive impairment in patients with cerebral small-vessel disease[J/OL]. Front Aging Neurosci, 2023, 15: 1096808 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/37065461/. DOI: 10.3389/fnagi.2023.1096808.
[5]
CHANG X X. Current advances in deep learning-based studies on early diagnosis of cerebral small vessel disease[J]. J Pract Radiol, 2023, 39(1): 148-150, 162. DOI: 10.3969/j.issn.1002-1671.2023.01.037.
[6]
DU F, XU H, WEN X X, et al. Application value of AI-assisted compressed sensing T2 fluid attenuated inversion recovery sequence in acute cerebral infarction[J]. J Pract Radiol, 2022, 38(12): 1917-1920. DOI: 10.3969/j.issn.1002-1671.2022.12.002.
[7]
CAO J J, LIU N, ZHONG M M, et al. Application of artificial intelligence-assisted compressed sensing technology in brain 3D T2-FLAIR sequence acquisition and evaluation of white matter hyperintensity[J]. Chin J Magn Reson Imag, 2024, 15(2): 135-139, 146. DOI: 10.12015/issn.1674-8034.2024.02.020.
[8]
DONG X Y, NAO J F. Fluid-attenuated inversion recovery vascular hyperintensities in anterior circulation acute ischemic stroke: associations with cortical brain infarct volume and 90-day prognosis[J]. Neurol Sci, 2019, 40(8): 1675-1682. DOI: 10.1007/s10072-019-03909-0.
[9]
GWO C Y, ZHU D C, ZHANG R. Brain white matter hyperintensity lesion characterization in 3D T2 fluid-attenuated inversion recovery magnetic resonance images: shape, texture, and their correlations with potential growth[J/OL]. Front Neurosci, 2022, 16: 1028929 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/36507337/. DOI: 10.3389/fnins.2022.1028929.
[10]
HAHN S, YI J, LEE H J, et al. Image quality and diagnostic performance of accelerated shoulder MRI with deep learning-based reconstruction[J]. AJR Am J Roentgenol, 2022, 218(3): 506-516. DOI: 10.2214/AJR.21.26577.
[11]
HUANG C, HU J W, LI L, et al. Artificial intelligence-assisted compressed sensing technology for lumbar MR scanning[J]. Chin J Med Imag Technol, 2023, 39(7): 1080-1083. DOI: 10.13929/j.issn.1003-3289.2023.07.029.
[12]
YANG B W, CHENG H, LIU M, et al. Evaluation of image quality of deep learning-based reconstruction of prostate T2WI and diagnostic performance for prostate cancer in transition zone[J]. Chin J Radiol, 2023, 57(11): 1208-1214. DOI: 10.3760/cma.j.cn112149-20230703-00452.
[13]
HERRMANN J, KOERZDOERFER G, NICKEL D, et al. Feasibility and implementation of a deep learning MR reconstruction for TSE sequences in musculoskeletal imaging[J/OL]. Diagnostics, 2021, 11(8): 1484 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/34441418/. DOI: 10.3390/diagnostics11081484.
[14]
YANG J, LI Q G, WU T, et al. A comparative study on the enhancement of high resolution coronal image quality by deep learning reconstruction[J]. Chin J Magn Reson Imag, 2023, 14(5): 21-24, 30. DOI: 10.12015/issn.1674-8034.2023.05.005.
[15]
DO W J, SEO S, HAN Y, et al. Reconstruction of multicontrast MR images through deep learning[J]. Med Phys, 2020, 47(3): 983-997. DOI: 10.1002/mp.14006.
[16]
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.
[17]
SCHLEMPER J, CABALLERO J, HAJNAL J V, et al. A deep cascade of convolutional neural networks for dynamic MR image reconstruction[J]. IEEE Trans Med Imaging, 2018, 37(2): 491-503. DOI: 10.1109/TMI.2017.2760978.
[18]
LIANG D, CHENG J, KE Z W, et al. Deep magnetic resonance image reconstruction: inverse problems meet neural networks[J]. IEEE Signal Process Mag, 2020, 37(1): 141-151. DOI: 10.1109/MSP.2019.2950557.
[19]
ARSHAD M, QURESHI M, INAM O, et al. Transfer learning in deep neural network-based receiver coil sensitivity map estimation[J]. MAGMA, 2021, 34(5): 717-728. DOI: 10.1007/s10334-021-00919-y.
[20]
NOORDMAN C R, YAKAR D, BOSMA J, et al. Complexities of deep learning-based undersampled MR image reconstruction[J/OL]. Eur Radiol Exp, 2023, 7(1): 58 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/37789241/. DOI: 10.1186/s41747-023-00372-7.
[21]
WU Z L, LIAO W B, YAN C, et al. Deep learning based MRI reconstruction with transformer[J/OL]. Comput Methods Programs Biomed, 2023, 233: 107452 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/36924533/. DOI: 10.1016/j.cmpb.2023.107452.
[22]
YOON J H, NICKEL M D, PEETERS J M, et al. Rapid imaging: recent advances in abdominal MRI for reducing acquisition time and its clinical applications[J]. Korean J Radiol, 2019, 20(12): 1597-1615. DOI: 10.3348/kjr.2018.0931.
[23]
KANIEWSKA M, DEININGER-CZERMAK E, GETZMANN J M, et al. Application of deep learning-based image reconstruction in MR imaging of the shoulder joint to improve image quality and reduce scan time[J]. Eur Radiol, 2023, 33(3): 1513-1525. DOI: 10.1007/s00330-022-09151-1.
[24]
GRISWOLD M A, JAKOB P M, HEIDEMANN R M, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA)[J]. Magn Reson Med, 2002, 47(6): 1202-1210. DOI: 10.1002/mrm.10171.
[25]
GONZÁLEZ BALLESTER M A, ZISSERMAN A P, BRADY M. Estimation of the partial volume effect in MRI[J]. Med Image Anal, 2002, 6(4): 389-405. DOI: 10.1016/s1361-8415(02)00061-0.
[26]
LI C Q, MAGLAND J F, RAJAPAKSE C S, et al. Implications of resolution and noise for in vivo micro-MRI of trabecular bone[J]. Med Phys, 2008, 35(12): 5584-5594. DOI: 10.1118/1.3005598.
[27]
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.
[28]
TAJIMA T, AKAI H, YASAKA K, et al. Usefulness of deep learning-based noise reduction for 1.5 T MRI brain images[J/OL]. Clin Radiol, 2023, 78(1): e13-e21 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/36116967/. DOI: 10.1016/j.crad.2022.08.127.
[29]
SUN S, TAN E T, MINTZ D N, et al. Evaluation of deep learning reconstructed high-resolution 3D lumbar spine MRI[J]. Eur Radiol, 2022, 32(9): 6167-6177. DOI: 10.1007/s00330-022-08708-4.
[30]
ALMANSOUR H, HERRMANN J, GASSENMAIER S, et al. Deep learning reconstruction for accelerated spine MRI: prospective analysis of interchangeability[J/OL]. Radiology, 2023, 306(3): e212922 [2024-07-07]. https://pubmed.ncbi.nlm.nih.gov/36318032/. DOI: 10.1148/radiol.212922.

PREV Application research of imaging genomics in preoperative prediction of microsatellite stability of endometrial cancer using mp-MRI
NEXT Feasibility study of 3D high-resolution compressed sensing contrast-enhanced whole-body MRA imaging technology
  


LINK


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