Abstract
Acquiring large quantities of data and annotations is effective for developing high-performing deep learning models, but is difficult and expensive to do in the healthcare context. Adding synthetic training data using generative models offers a low-cost method to deal effectively with the data scarcity challenge, and can also address data imbalance and patient privacy issues. In this study, we propose a comprehensive framework that fits seamlessly into model development workflows for medical image analysis. We demonstrate, with datasets of varying size, (i) the benefits of generative models as a data augmentation method; (ii) how adversarial methods can protect patient privacy via data substitution; (iii) novel performance metrics for these use cases by testing models on real holdout data. We show that training with both synthetic and real data outperforms training with real data alone, and that models trained solely with synthetic data approach their real-only counterparts. Code is available at https://github.com/Global-Health-Labs/US-DCGAN.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bargsten, L., Schlaefer, A.: SpeckleGAN: a generative adversarial network with an adaptive speckle layer to augment limited training data for ultrasound image processing. IJCARS 15, 1427–1436 (2020)
Bhoil, R., Ahluwalia, A., Chopra, R., Surya, M., Bhoil, S.: Signs and lines in lung ultrasound. J. Ultrason. 21(86), 225–233 (2021)
Chuquicusma, M.J., et al.: How to fool radiologists with GANs? a visual turing test for lung cancer diagnosis. In: ISBI, pp. 240–244. IEEE (2018)
Frid-Adar, M., et al.: Synthetic data augmentation using GAN for improved liver lesion classification. In: ISBI, pp. 289–293. IEEE (2018)
Goodfellow, I., et al.: Generative adversarial networks. In: NeurIPS (2014)
Haq, D.Z., Fatichah, C.: Ultrasound image synthetic generating using deep convolution generative adversarial network for breast cancer identification. IPTEK J. Technol. Sci. 34(1), 12 (2023)
Hu, Y., et al.: Freehand ultrasound image simulation with spatially-conditioned generative adversarial networks. In: Cardoso, M.J., et al. (eds.) CMMI/SWITCH/RAMBO -2017. LNCS, vol. 10555, pp. 105–115. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-67564-0_11
Koh, H., et al.: Acoustic simulation for transcranial focused ultrasound using GAN-based synthetic CT. IEEE J. Biomed. Health Inform. 26(1), 161–171 (2021)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. Commun. ACM 60(6), 84–90 (2017)
Kulhare, S., et al.: Ultrasound-based detection of lung abnormalities using single shot detection convolutional neural networks. In: Stoyanov, D., et al. (eds.) POCUS/BIVPCS/CuRIOUS/CPM -2018. LNCS, vol. 11042, pp. 65–73. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01045-4_8
Lahiri, A., Ayush, K., Kumar Biswas, P., Mitra, P.: Generative adversarial learning for reducing manual annotation in semantic segmentation on large scale miscroscopy images: automated vessel segmentation in retinal fundus image as test case. In: Proceedings of CVPR Workshops, pp. 42–48 (2017)
Liang, J., Chen, J.: Data augmentation of thyroid ultrasound images using generative adversarial network. In: IEEE IUS, pp. 1–4. IEEE (2021)
Nie, D., et al.: Medical image synthesis with deep convolutional adversarial networks. IEEE TBME 65(12), 2720–2730 (2018)
Ouyang, J., et al.: Weakly semi-supervised detection in lung ultrasound videos. In: IPMI 2023, pp. 195–207 (2023)
Pang, T., Wong, J.H.D., Ng, W.L., Chan, C.S.: Semi-supervised GAN-based radiomics model for data augmentation in breast ultrasound mass classification. Comput. Meth. Prog. Biomed. 203, 106018 (2021)
Radford, A., et al.: Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434 (2015)
Shea, D., et al.: Deep learning video classification of lung ultrasound features associated with pneumonia. In: CVPR 2023, pp. 3102–3111. IEEE (2023)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2015)
Volpicelli, G., et al.: International evidence-based recommendations for point-of-care lung ultrasound. Int. J. Med. Inform. 129, 413–422 (2019)
World Health Organization (WHO): Pneumonia (2016). https://www.who.int/mediacentre/factsheets/fs331/en
Xu, Q., et al.: An empirical study on evaluation metrics of generative adversarial networks. arXiv preprint arXiv:1806.07755 (2018)
Acknowledgements
We thank the Bill and Melinda Gates Foundation Trust for their generous support, and Xinliang Zheng, Ben Wilson, Meihua Zhu, Cynthia Gregory, and Kenton Gregory for their roles in managing the lung ultrasound project.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Yu, M. et al. (2023). How Good Are Synthetic Medical Images? An Empirical Study with Lung Ultrasound. In: Wolterink, J.M., Svoboda, D., Zhao, C., Fernandez, V. (eds) Simulation and Synthesis in Medical Imaging. SASHIMI 2023. Lecture Notes in Computer Science, vol 14288. Springer, Cham. https://doi.org/10.1007/978-3-031-44689-4_8
Download citation
DOI: https://doi.org/10.1007/978-3-031-44689-4_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44688-7
Online ISBN: 978-3-031-44689-4
eBook Packages: Computer ScienceComputer Science (R0)
