Share：
Share this content in WeChat
X
[Chinese][PDF]583
• Special Focus •
Application value of diffusion-weighted imaging reconstructed based on deep learning in benign and malignant differentiation of pulmonary lesions
LI Jie  XIA Yi  XU Meiling  LIN Xiaoqing  JIANG Song  DAI Jiankun  JIANG Xin'ang  SUN Guangyuan  LIU Shiyuan  FAN Li 

Cite this article as: LI J, XIA Y, XU M L, et al. Application value of diffusion-weighted imaging reconstructed based on deep learning in benign and malignant differentiation of pulmonary lesions[J]. Chin J Magn Reson Imaging, 2024, 15(10): 15-21. DOI:10.12015/issn.1674-8034.2024.10.004.


[Abstract] Objective To evaluate the impact of deep learning reconstruction (DLR) on the image quality of pulmonary diffusion-weighted imaging (DWI) and to explore the value of DLR in the identification of benign and malignant pulmonary lesions.Materials and Methods In this prospective study, 61 patients with pulmonary nodules or masses (including 49 malignant and 12 benign cases) were recruited. Each patient underwent T2WI and DWI imaging using a 3.0 T MRI scanner, with DWI images reconstructed using both conventional reconstruction (ConR) and deep learning reconstruction (DLR). Two radiologists with 4 and 10 years of experience independently evaluated the overall image quality, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and apparent diffusion coefficient (ADC) of the lesion. Interobserver agreement on subjective scores was assessed using Kappa values, while intra-class correlation coefficient (ICC) were used to evaluate interobserver agreement on SNR, CNR, and ADC values. Wilcoxon rank sum tests were used to compare the differences between DLR DWI and ConR DWI in terms of subjective scores, SNR, CNR, and ADC. Mann-Whitney tests were performed to compare differences between benign and malignant lesions. The diagnostic performance of ADC for identifying benign and malignant lesions was evaluated using receiver operating characteristics (ROC) curves, with the area under the curve (AUC) compared between ConR and DLR using the DeLong test.Results Both DLR and ConR images showed good interobserver agreement in subjective scoring (Kappa>0.60). In objective assessments, SNR and ADC demonstrated excellent interobserver consistency (ICC>0.75), while CNR showed only fair interobserver agreement (ICC>0.40). Compared to ConR DWI, DLR DWI had higher overall image quality scores (P=0.003), lesion SNR (P<0.001), and higher ADC values (P=0.017). Additionally, the CNR of DLR DWI was higher than that of ConR DWI, but the difference was not significant (P=0.258). For both ConR and DLR DWI, the ADC of malignant lesions was significantly lower than that of benign lesions (P<0.05). ROC curve analysis indicated that DLR DWI (AUC=0.891) had higher diagnostic performance in distinguishing between benign and malignant lesions compared to ConR DWI (AUC=0.808), with a significant difference observed by DeLong test (P=0.044).Conclusions DLR DWI significantly improves overall image quality and enhances the SNR of images, offering superior diagnostic performance for distinguishing benign from malignant lesions compared to ConR DWI.
[Keywords] lung cancer；pulmonary lesions；diffusion-weighted imaging；deep learning reconstruction；magnetic resonance imaging

LI Jie1, 2   XIA Yi2   XU Meiling2   LIN Xiaoqing1, 2   JIANG Song2   DAI Jiankun3   JIANG Xin'ang2   SUN Guangyuan4   LIU Shiyuan2   FAN Li2*  

1 College of Health Sciences and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

2 Department of Radiology, Second Affiliated Hospital of Naval Medical University, Shanghai 200003, China

3 GE Healthcare, Shanghai 200120, China

4 Department of Thoracic Surgery, Second Affiliated Hospital of Naval Medical University, Shanghai 200003, China

Corresponding author: FAN L, E-mail: fanli0930@163.com

Conflicts of interest   None.

Received  2024-04-15
Accepted  2024-07-05
DOI: 10.12015/issn.1674-8034.2024.10.004
Cite this article as: LI J, XIA Y, XU M L, et al. Application value of diffusion-weighted imaging reconstructed based on deep learning in benign and malignant differentiation of pulmonary lesions[J]. Chin J Magn Reson Imaging, 2024, 15(10): 15-21. DOI:10.12015/issn.1674-8034.2024.10.004.

[1]
SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249. DOI: 10.3322/caac.21660.
[2]
FILLI L, GHAFOOR S, KENKEL D, et al. Simultaneous multi-slice readout-segmented echo planar imaging for accelerated diffusion-weighted imaging of the breast[J]. Eur J Radiol, 2016, 85(1): 274-278. DOI: 10.1016/j.ejrad.2015.10.009.
[3]
CHEN N K, GUIDON A, CHANG H C, et al. A robust multi-shot scan strategy for high-resolution diffusion weighted MRI enabled by multiplexed sensitivity-encoding (MUSE)[J]. Neuroimage, 2013, 72: 41-47. DOI: 10.1016/j.neuroimage.2013.01.038.
[4]
DRATSCH T, SIEDEK F, ZÄSKE C, et al. Reconstruction of shoulder MRI using deep learning and compressed sensing: a validation study on healthy volunteers[J/OL]. Eur Radiol Exp, 2023, 7(1): 66 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/37880546/. DOI: 10.1186/s41747-023-00377-2.
[5]
KIM M, LEE S M, PARK C, et al. Deep learning-enhanced parallel imaging and simultaneous multislice acceleration reconstruction in knee MRI[J]. Invest Radiol, 2022, 57(12): 826-833. DOI: 10.1097/RLI.0000000000000900.
[6]
CHAUDHARI A S, SANDINO C M, COLE E K, et al. Prospective deployment of deep learning in MRI: a framework for important considerations, challenges, and recommendations for best practices[J]. J Magn Reson Imaging, 2021, 54(2): 357-371. DOI: 10.1002/jmri.27331.
[7]
PARK J C, PARK K J, PARK M Y, et al. Fast T2-weighted imaging with deep learning-based reconstruction: evaluation of image quality and diagnostic performance in patients undergoing radical prostatectomy[J]. J Magn Reson Imaging, 2022, 55(6): 1735-1744. DOI: 10.1002/jmri.27992.
[8]
ALLEN T J, HENZE BANCROFT L C, UNAL O, et al. Evaluation of a deep learning reconstruction for high-quality T2-weighted breast magnetic resonance imaging[J]. Tomography, 2023, 9(5): 1949-1964. DOI: 10.3390/tomography9050152.
[9]
KIM B, LEE C M, JANG J K, et al. Deep learning-based imaging reconstruction for MRI after neoadjuvant chemoradiotherapy for rectal cancer: effects on image quality and assessment of treatment response[J]. Abdom Radiol, 2023, 48(1): 201-210. DOI: 10.1007/s00261-022-03701-3.
[10]
CHAZEN J L, TAN E T, FIORE J, et al. Rapid lumbar MRI protocol using 3D imaging and deep learning reconstruction[J]. Skeletal Radiol, 2023, 52(7): 1331-1338. DOI: 10.1007/s00256-022-04268-2.
[11]
KISO K, TSUBOYAMA T, ONISHI H, et al. Effect of deep learning reconstruction on respiratory-triggered T2-weighted MR imaging of the liver: a comparison between the single-shot fast spin-echo and fast spin-echo sequences[J]. Magn Reson Med Sci, 2024, 23(2): 214-224. DOI: 10.2463/mrms.mp.2022-0111.
[12]
LEE K L, KESSLER D A, DEZONIE S, et al. Assessment of deep learning-based reconstruction on T2-weighted and diffusion-weighted prostate MRI image quality[J/OL]. Eur J Radiol, 2023, 166: 111017 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/37541181/. DOI: 10.1016/j.ejrad.2023.111017.
[13]
DUAN C H, DENG H, XIAO S, et al. Fast and accurate reconstruction of human lung gas MRI with deep learning[J]. Magn Reson Med, 2019, 82(6): 2273-2285. DOI: 10.1002/mrm.27889.
[14]
LEBEL R M. Performance characterization of a novel deep learning-based MR image reconstruction pipeline[EB/OL]. 2020: arXiv: 2008.06559. [2024-04-14] (2020-08-14). http://arxiv.org/abs/2008.06559
[15]
WAN Q, LEI Q, WANG P, et al. Intravoxel incoherent motion diffusion-weighted imaging of lung cancer: comparison between turbo spin-echo and echo-planar imaging[J]. J Comput Assist Tomogr, 2020, 44(3): 334-340. DOI: 10.1097/RCT.0000000000001004.
[16]
ZHENG Y, LI J, CHEN K, et al. Comparison of conventional DWI, intravoxel incoherent motion imaging, and diffusion kurtosis imaging in differentiating lung lesions[J/OL]. Front Oncol, 2021, 11: 815967 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/35127530/. DOI: 10.3389/fonc.2021.815967.
[17]
BRONCANO J, STEINBRECHER K, MARQUIS K M, et al. Diffusion-weighted imaging of the chest: a primer for radiologists[J/OL]. Radiographics, 2023, 43(7): e220138 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/37347699/. DOI: 10.1148/rg.220138.
[18]
IIMA M, PARTRIDGE S C, LE BIHAN D. Six DWI questions you always wanted to know but were afraid to ask: clinical relevance for breast diffusion MRI[J]. Eur Radiol, 2020, 30(5): 2561-2570. DOI: 10.1007/s00330-019-06648-0.
[19]
CHEN Q, FANG S, YUCHEN Y, et al. Clinical feasibility of deep learning reconstruction in liver diffusion-weighted imaging: improvement of image quality and impact on apparent diffusion coefficient value[J/OL]. Eur J Radiol, 2023, 168: 111149 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/37862927/. DOI: 10.1016/j.ejrad.2023.111149.
[20]
AFAT S, HERRMANN J, ALMANSOUR H, et al. Acquisition time reduction of diffusion-weighted liver imaging using deep learning image reconstruction[J]. Diagn Interv Imaging, 2023, 104(4): 178-184. DOI: 10.1016/j.diii.2022.11.002.
[21]
SAUER S T, CHRISTNER S A, LOIS A M, et al. Deep learning k-space-to-image reconstruction facilitates high spatial resolution and scan time reduction in diffusion-weighted imaging breast MRI[J/OL]. J Magn Reson Imaging, 2023 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/37974498/. DOI: 10.1002/jmri.29139.
[22]
BAE S H, HWANG J, HONG S S, et al. Clinical feasibility of accelerated diffusion weighted imaging of the abdomen with deep learning reconstruction: comparison with conventional diffusion weighted imaging[J/OL]. Eur J Radiol, 2022, 154: 110428 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/35797791/. DOI: 10.1016/j.ejrad.2022.110428.
[23]
YANG A, FINKELSTEIN M, KOO C, et al. Impact of deep learning image reconstruction methods on MRI throughput[J/OL]. Radiol Artif Intell, 2024, 6(3): e230181 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/38506618/. DOI: 10.1148/ryai.230181.
[24]
KIRYU S, AKAI H, YASAKA K, et al. Clinical impact of deep learning reconstruction in MRI[J/OL]. Radiographics, 2023, 43(6): e220133 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/37200221/. DOI: 10.1148/rg.220133.
[25]
YAQUB M, JINCHAO J C, ARSHID K, et al. Deep learning-based image reconstruction for different medical imaging modalities[J/OL]. Comput Math Methods Med, 2022, 2022: 8750648 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/35756423/. DOI: 10.1155/2022/8750648.
[26]
LI J X, WU B L, HUANG Z, et al. Whole-lesion histogram analysis of multiple diffusion metrics for differentiating lung cancer from inflammatory lesions[J/OL]. Front Oncol, 2022, 12: 1082454 [2024-04-14]. https://pubmed.ncbi.nlm.nih.gov/36741699/. DOI: 10.3389/fonc.2022.1082454.
[27]
MAHDAVI RASHED M, NEKOOEI S, NOURI M, et al. Evaluation of DWI and ADC sequences' diagnostic values in benign and malignant pulmonary lesions[J]. Turk Thorac J, 2020, 21(6): 390-396. DOI: 10.5152/TurkThoracJ.2020.19007.
[28]
ÇAKMAK V, UFUK F, KARABULUT N. Diffusion-weighted MRI of pulmonary lesions: comparison of apparent diffusion coefficient and lesion-to-spinal cord signal intensity ratio in lesion characterization[J]. J Magn Reson Imaging, 2017, 45(3): 845-854. DOI: 10.1002/jmri.25426.
[29]
MATOBA M, TONAMI H, KONDOU T, et al. Lung carcinoma: diffusion-weighted mr imaging: preliminary evaluation with apparent diffusion coefficient[J]. Radiology, 2007, 243(2): 570-577. DOI: 10.1148/radiol.2432060131.
[30]
UTO T, TAKEHARA Y, NAKAMURA Y, et al. Higher sensitivity and specificity for diffusion-weighted imaging of malignant lung lesions without apparent diffusion coefficient quantification[J]. Radiology, 2009, 252(1): 247-254. DOI: 10.1148/radiol.2521081195.
[31]
GÜMÜŞTAŞ S, INAN N, AKANSEL G, et al. Differentiation of malignant and benign lung lesions with diffusion-weighted MR imaging[J]. Radiol Oncol, 2012, 46(2): 106-113. DOI: 10.2478/v10019-012-0021-3.
[32]
ZHU Q, REN C, XU J J, et al. Whole-lesion histogram analysis of mono-exponential and bi-exponential diffusion-weighted imaging in differentiating lung cancer from benign pulmonary lesions using 3 T MRI[J]. Clin Radiol, 2021, 76(11): 846-853. DOI: 10.1016/j.crad.2021.07.003.

PREV Feasibility study of deep learning-based MRI image reconstruction algorithms for myocardial delayed enhancement in unrecognized myocardial infarction
NEXT Value analysis of deep learning model based on DCE-MRI images in the differential diagnosis of benign and malignant breast tumors
  


LINK


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