Volume 52, Issue 6 pp. 1852-1858

Original Research

High-Resolution Breast MRI Reconstruction Using a Deep Convolutional Generative Adversarial Network

Kun Sun MD, PhD,

Kun Sun MD, PhD

Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Liangqiong Qu PhD,

Liangqiong Qu PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Chunfeng Lian PhD,

Chunfeng Lian PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Yongsheng Pan PhD,

Yongsheng Pan PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Dan Hu PhD,

Dan Hu PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Bingqing Xia MD,

Bingqing Xia MD

Department of Radiology, International Peace Maternity and Child Health Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Search for more papers by this author

Xinyue Li MD,

Xinyue Li MD

Department of Radiology, Ruijin Hospital Luwan Branch, Shanghai Jiaotong University School of Medicine, Shanghai, China

Search for more papers by this author

Weimin Chai MD, PhD,

Weimin Chai MD, PhD

Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Search for more papers by this author

Fuhua Yan MD,PhD,

Corresponding Author

Fuhua Yan MD,PhD

[email protected]

Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Address reprint requests to: Dinggang Shen, Department of Research and Development,Shanghai United Imaging Intelligence Co.,Ltd.,Shanghai,China. E-mail: [email protected]

Fuhua Yan, Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China. E-mail: [email protected]

Search for more papers by this author

Dinggang Shen PhD,

Corresponding Author

Dinggang Shen PhD

[email protected]

Department of Research and Development, Shanghai United Imaging Intelligence Co.,Ltd., Shanghai, China

Department of Brain and Cognitive Engineering, Korea University, Seoul 02841, Republic of Korea

Address reprint requests to: Dinggang Shen, Department of Research and Development,Shanghai United Imaging Intelligence Co.,Ltd.,Shanghai,China. E-mail: [email protected]

Fuhua Yan, Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China. E-mail: [email protected]

Search for more papers by this author

Kun Sun MD, PhD,

Kun Sun MD, PhD

Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Liangqiong Qu PhD,

Liangqiong Qu PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Chunfeng Lian PhD,

Chunfeng Lian PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Yongsheng Pan PhD,

Yongsheng Pan PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Dan Hu PhD,

Dan Hu PhD

Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Search for more papers by this author

Bingqing Xia MD,

Bingqing Xia MD

Department of Radiology, International Peace Maternity and Child Health Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Search for more papers by this author

Xinyue Li MD,

Xinyue Li MD

Department of Radiology, Ruijin Hospital Luwan Branch, Shanghai Jiaotong University School of Medicine, Shanghai, China

Search for more papers by this author

Weimin Chai MD, PhD,

Weimin Chai MD, PhD

Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Search for more papers by this author

Fuhua Yan MD,PhD,

Corresponding Author

Fuhua Yan MD,PhD

[email protected]

Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

Address reprint requests to: Dinggang Shen, Department of Research and Development,Shanghai United Imaging Intelligence Co.,Ltd.,Shanghai,China. E-mail: [email protected]

Fuhua Yan, Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China. E-mail: [email protected]

Search for more papers by this author

Dinggang Shen PhD,

Corresponding Author

Dinggang Shen PhD

[email protected]

Department of Research and Development, Shanghai United Imaging Intelligence Co.,Ltd., Shanghai, China

Department of Brain and Cognitive Engineering, Korea University, Seoul 02841, Republic of Korea

Address reprint requests to: Dinggang Shen, Department of Research and Development,Shanghai United Imaging Intelligence Co.,Ltd.,Shanghai,China. E-mail: [email protected]

Fuhua Yan, Department of Radiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China. E-mail: [email protected]

Search for more papers by this author

First published: 12 July 2020

https://doi.org/10.1002/jmri.27256

Citations: 4

# Co-first authors The first two authors contributed equally to this work.

Share a link

Email
Wechat
Bluesky

Abstract

Background

A generative adversarial network could be used for high-resolution (HR) medical image synthesis with reduced scan time.

Purpose

To evaluate the potential of using a deep convolutional generative adversarial network (DCGAN) for generating HR_pre and HR_post images based on their corresponding low-resolution (LR) images (LR_pre and LR_post).

Study Type

This was a retrospective analysis of a prospectively acquired cohort.

Population

In all, 224 subjects were randomly divided into 200 training subjects and an independent 24 subjects testing set.

Field Strength/Sequence

Dynamic contrast-enhanced (DCE) MRI with a 1.5T scanner.

Assessment

Three breast radiologists independently ranked the image datasets, using the DCE images as the ground truth, and reviewed the image quality of both the original LR images and the generated HR images. The BI-RADS category and conspicuity of lesions were also ranked. The inter/intracorrelation coefficients (ICCs) of mean image quality scores, lesion conspicuity scores, and Breast Imaging Reporting and Data System (BI-RADS) categories were calculated between the three readers.

Statistical Test

Wilcoxon signed-rank tests evaluated differences among the multireader ranking scores.

Results

The mean overall image quality scores of the generated HR_pre and HR_post were significantly higher than those of the original LR_pre and LR_post (4.77 ± 0.41 vs. 3.27 ± 0.43 and 4.72 ± 0.44 vs. 3.23 ± 0.43, P < 0.0001, respectively, in the multireader study). The mean lesion conspicuity scores of the generated HR_pre and HR_post were significantly higher than those of the original LR_pre and LR_post (4.18 ± 0.70 vs. 3.49 ± 0.58 and 4.35 ± 0.59 vs. 3.48 ± 0.61, P < 0.001, respectively, in the multireader study). The ICCs of the image quality scores, lesion conspicuity scores, and BI-RADS categories had good agreements among the three readers (all ICCs >0.75).

Data Conclusion

DCGAN was capable of generating HR of the breast from fast pre- and postcontrast LR and achieved superior quantitative and qualitative performance in a multireader study.

Level of Evidence

Technical Efficacy Stage

2 J. MAGN. RESON. IMAGING 2020;52:1852–1858.

REFERENCES

1Mann RM, Cho N, Moy LJR. Breast MRI: State of the art. Radiology 2019; 292: 520-536.
10.1148/radiol.2019182947
PubMed Web of Science® Google Scholar
2Lehman CD, Gatsonis C, Kuhl CK, et al. MRI evaluation of the contralateral breast in women with recently diagnosed breast cancer. N Engl J Med 2007; 356: 1295-1303.
10.1056/NEJMoa065447
CAS PubMed Web of Science® Google Scholar
3Heller SL, Moy L. MRI breast screening revisited. J Magn Reson Imaging 2019; 49: 1212-1221.
10.1002/jmri.26547
PubMed Web of Science® Google Scholar
4Deike-Hofmann K, Koenig F, Paech D, et al. Abbreviated MRI protocols in breast cancer diagnostics. J Magn Reson Imaging 2019; 49: 647-658.
10.1002/jmri.26525
PubMed Web of Science® Google Scholar
5Konyer NB, Ramsay EA, Bronskill MJ, Plewes DB. Comparison of MR imaging breast coils. Radiology 2002; 222: 830-834.
10.1148/radiol.2223001310
PubMed Web of Science® Google Scholar
6Taşkın F, Polat Y, Erdoğdu İH, et al. Problem-solving breast MRI: Useful or a source of new problems? Diagn Interv Radiol 2018; 24: 255-261.
10.5152/dir.2018.17504
PubMed Web of Science® Google Scholar
7Brandzel S, Rosenberg DE, Johnson D, et al. Women's experiences and preferences regarding breast imaging after completing breast cancer treatment. Patient Prefer Adherence 2017; 11: 199-204.
10.2147/PPA.S122244
PubMed Web of Science® Google Scholar
8LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015; 521: 436-444.
10.1038/nature14539
CAS PubMed Web of Science® Google Scholar
9Nie D, Trullo R, Lian J, et al. Medical image synthesis with deep convolutional adversarial networks. IEEE Trans Biomed Eng 2018; 65: 2720-2730.
10.1109/TBME.2018.2814538
PubMed Web of Science® Google Scholar
10Kamnitsas K, Ledig C, Newcombe VF, et al. Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med Image Anal 2017; 36: 61-78.
10.1016/j.media.2016.10.004
PubMed Web of Science® Google Scholar
11Rouhi R, Jafari M, Kasaei S, Keshavarzian P. Benign and malignant breast tumors classification based on region growing and CNN segmentation. Exp Syst Appl 2015; 42: 990-1002.
10.1016/j.eswa.2014.09.020
Web of Science® Google Scholar
12Truhn D, Schrading S, Haarburger C, et al. Radiomic versus convolutional neural networks analysis for classification of contrast-enhancing lesions at multiparametric breast MRI. Radiology 2018; 290: 290-297.
10.1148/radiol.2018181352
PubMed Web of Science® Google Scholar
13Akselrod-Ballin A, Chorev M, Shoshan Y, et al. Predicting breast cancer by applying deep learning to linked health records and mammograms. Radiology 2019; 292: 331-342.
10.1148/radiol.2019182622
PubMed Web of Science® Google Scholar
14Zhao Z, Jiao L, Zhao J, Gu J, Zhao J. Discriminant deep belief network for high-resolution SAR image classification. Pattern Recogn 2017; 61: 686-701.
10.1016/j.patcog.2016.05.028
Web of Science® Google Scholar
15Kingma DP, Welling M. Auto-encoding variational Bayes. arXiv preprint arXiv:1312.6114. 2013.
Google Scholar
16Goodfellow I. 2016 tutorial: Generative adversarial networks. arXiv Preprint arXiv:1701.00160.2016.
Google Scholar
17Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets. Proceedings of the advances in neural information processing systems. Montreal: NIPS; 2014. p 2672–2680.
Google Scholar
18Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434. 2015.
Google Scholar
19Chuquicusma MJ, Hussein S, Burt J, et al. How to fool radiologists with generative adversarial networks? A visual Turing test for lung cancer diagnosis. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018): Washington, IEEE; 2018. p 240–244.
Google Scholar
20Salehinejad H, Valaee S, Dowdell T, et al. Generalization of deep neural networks for chest pathology classification in x-rays using generative adversarial networks. In: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): Washington; IEEE; 2018, p 990–994.
Google Scholar
21Fang W, Zhang F, Sheng VS, et al. A method for improving CNN-based image recognition using DCGAN. Comput Mater Continua 2018; 57: 167-178.
10.32604/cmc.2018.02356
Web of Science® Google Scholar
22Chaudhari AS, Fang Z, Kogan F, et al. Super-resolution musculoskeletal MRI using deep learning. Magn Reson Med 2018; 80: 2139-2154.
10.1002/mrm.27178
PubMed Web of Science® Google Scholar
23Chen Y, Shi F, Christodoulou AG, et al. Efficient and accurate MRI super- resolution using a generative adversarial network and 3D multi-level densely connected network. International Conference on Medical Image Computing and Computer-Assisted Intervention. Berlin: Springer; 2018. p 91-99.
10.1007/978-3-030-00928-1_11
Google Scholar
24Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 2002; 17: 825-841.
10.1006/nimg.2002.1132
PubMed Web of Science® Google Scholar
25Rosten E, Drummond T. Machine learning for high-speed corner detection. European Conference on Computer Vision. Berlin: Springer; 2006. p 430-443.
10.1007/11744023_34
Google Scholar
26Zhu J-Y, Park T, Isola P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks. In: Proceedings of the IEEE International Conference on Computer Vision. San Francisco: IEEE; 2017. p 2223–2232.
Google Scholar
27Isola P, Zhu J-Y, Zhou T, et al. Image-to-image translation with conditional adversarial networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. San Francisco: IEEE; 2017. p 1125–1134.
Google Scholar
28Li C, Wand M. Precomputed real-time texture synthesis with markovian generative adversarial networks. In: European Conference on Computer Vision. Amsterdam: Springer; 2016. p 702–716.
Google Scholar
29Ledig C, Theis L, Huszár F, et al. Photo-realistic single image super-resolution using a generative adversarial network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. San Francisco: IEEE; 2017. p 4681–4690.
Google Scholar
30Kingma DP, Ba J. Adam. A method for stochastic optimization. arXiv preprint arXiv:14126980.2014.
Google Scholar
31Desai AD, Gold GE, Hargreaves BA, et al. Technical considerations for semantic segmentation in MRI using convolutional neural networks. arXiv preprint arXiv:190201977.2019.
Google Scholar
32Qu L, Zhang Y, Wang S, Yap PT, Shen D. Synthesized 7T MRI from 3T MRI via deep learning in spatial and wavelet domains. Med Image Anal 2020; 62:101663.
10.1016/j.media.2020.101663
PubMed Web of Science® Google Scholar
33Amin J, Sharif M, Raza M, Saba T, Anjum MA. Brain tumor detection using statistical and machine learning method. Comput Methods Programs Biomed 2019; 177: 69-79.
10.1016/j.cmpb.2019.05.015
PubMed Web of Science® Google Scholar
34Hore A, Ziou D. Image quality metrics: PSNR vs. SSIM. In: 2010 20th International Conference on Pattern Recognition. Istanbul: IEEE 2010. p 2366–2369.
Google Scholar
35Chen Y, Xiao T, Li C, et al. Model-based convolutional De-aliasing network learning for parallel MR imaging. International Conference on Medical Image Computing and Computer-Assisted Intervention: Berlin: Springer; 2019. p 30-38.
Google Scholar
36Sun K, Zhu H, Chai W, et al. Whole-lesion histogram and texture analyses of breast lesions on inline quantitative DCE mapping with CAIPIRINHA-Dixon-TWIST-VIBE. Eur Radiol 2020; 30: 57-65.
10.1007/s00330-019-06365-8
PubMed Web of Science® Google Scholar
37Li Z, Ai T, Hu Y, et al. Application of whole-lesion histogram analysis of pharmacokinetic parameters in dynamic contrast-enhanced MRI of breast lesions with the CAIPIRINHA-Dixon-TWIST-VIBE technique. J Magn Reson Imaging 2018; 47: 91-96.
10.1002/jmri.25762
PubMed Web of Science® Google Scholar
38Nie D, Trullo R, Lian J, et al. Medical image synthesis with context-aware generative adversarial networks. Med Image Comput 2017; 10435: 417-425.
Google Scholar
39Jin CB, Kim H, Liu M, et al. Deep CT to MR synthesis using paired and unpaired data. Sensors (Basel) 2019; 19: 2361.
10.3390/s19102361
Google Scholar
40Emami H, Dong M, Nejad-Davarani SP, Glide-Hurst CK. Generating synthetic CTs from magnetic resonance images using generative adversarial networks. Med Phys 2018; 45: 3627-3636.
10.1002/mp.13047
Web of Science® Google Scholar
41Arjovsky M, Chintala S, Bottou L. Wasserstein GAN. arXiv preprint arXiv:1701.07875.2017.
Google Scholar

Citing Literature

Volume52, Issue6

December 2020

Pages 1852-1858

High-Resolution Breast MRI Reconstruction Using a Deep Convolutional Generative Adversarial Network

Abstract

Background

Purpose

Study Type

Population

Field Strength/Sequence

Assessment

Statistical Test

Results

Data Conclusion

Level of Evidence

Technical Efficacy Stage

REFERENCES

Citing Literature

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley

High-Resolution Breast MRI Reconstruction Using a Deep Convolutional Generative Adversarial Network

Abstract

Background

Purpose

Study Type

Population

Field Strength/Sequence

Assessment

Statistical Test

Results

Data Conclusion

Level of Evidence

Technical Efficacy Stage

REFERENCES

Citing Literature

References

Related

Information