Synthetic Skin Cancer Image Data Generation Using Generative Adversarial Neural Network

Abstract

In this study, synthetic data generating method using generative adversarial neural network (GAN) for the skin cancer types malignant melanoma and basal-cell carcinoma is presented. GAN is a neural network where two synthetic networks compete. The generator attempts to generate data similar to those measured and the discriminator attempts to classify data as dummy or real. Using medical data in studies is a difficult task due to legal and ethical restrictions. Most of the available data is classified because of patient consent and available data in most cases is not labeled, low quality and/or low quantity. Recent GAN systems can generate labeled high quantity data without any personal discriminative information. In this paper, we used skin cancer images in The International Skin Imaging Collaboration (ISIC) database that have been used for discriminator training. To test our generated images applicability in the medical field studies we have conducted a Turing test with medical experts in various medical fields. Our results indicate that the generated data obtained with our method is a valuable alternative for real medical data.

Keywords

• Deep Learning
• Generative Adversarial Networks
• Image generation
• Medical Image Analysis
• Skin Lesion

References

1. B. Møller, H. Weedon-Fekjær, T. Hakulinen, L. Tryggvadottir, H. Storm, M. Talb¨ack, and T. Haldorsen, “Prediction of cancer incidence in the Nordic countries up to the year 2020,” European journal of cancer prevention: the official journal of the European Cancer Prevention Organization (ECP), vol. 11 Suppl 1, pp. S1–96, 2002.
2. C. F. Heal, B. A. Raasch, P. Buettner, and D. Weedon, “Accuracy of clinical diagnosis of skin lesions,” British Journal of Dermatology, vol. 159, no. 3, pp. 661–668, 2008.
3. I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial networks,” arXiv preprint arXiv: 1406.2661, 2014.
4. J.. Zhu, T. Park, P. Isola, and A. A. Efros. “Unpaired image-to-image translation using cycle-consistent adversarial networks,” in Proceedings of the IEEE international conference on computer vision, pages 2223–2232, 2017.
5. Y. Choi, M. Choi, M. Kim, J. Ha, S. Kim, and J. Choo. “Stargan: Unified generative adversarial networks for multi-domain image-to-image translation,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 8789–8797, 2018.
6. X. Yi, E. Walia, and P. Babyn, “Generative adversarial network in medical imaging: A review,” Medical Image Analysis, vol. 58, p. 101552, 2019.
7. J. M. Wolterink, T. Leiner, M. A. Viergever, and I. Iˇsgum, “Generative adversarial networks for noise reduction in low-dose CT,” IEEE transactions on medical imaging, vol. 36, no. 12, pp. 2536–2545, 2017.
8. S. U. H. Dar, M. Yurt, M. Shahdloo, M. E. Ildız, and T. C¸ Ukur, “Synergistic reconstruction and synthesis via generative adversarial networks for accelerated multi-contrast mri,” arXiv preprint arXiv: 1805.10704, 2018.

9. J. Wang, Y. Zhao, J. H. Noble, and B. M. Dawant, “Conditional Generative Adversarial Networks for Metal Artifact Reduction in CT Images of the Ear,” in International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer vol 11070, pp. 3–11, 2018.
10. C. You, G. Li, Y. Zhang, X. Zhang, H. Shan, M. Li, S. Ju, Z. Zhao, Z. Zhang, W. Cong, “CT super-resolution GAN constrained by the identical, residual, and cycle learning ensemble (GANCIRCLE),” IEEE Transactions on Medical Imaging, vol. 39, no. 1, pp. 188–203, 2019.
11. M. J. Chuquicusma, S. Hussein, J. Burt, and U. Bagci, “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), pp. 240–244, 2018.
12. X. Yi, E. Walia, and P. Babyn, “Unsupervised and semi-supervised learning with Categorical Generative Adversarial Networks assisted by Wasserstein distance for dermoscopy image Classification,” arXiv preprint, arXiv: 1804.03700, 2018.
13. Z. Zhang, L. Yang, and Y. Zheng, “Translating and segmenting multimodal medical volumes with cycle-and shape-consistency generative adversarial network,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 9242-9251, 2018.
14. R. Togo, T. Ogawa and M. Haseyama, "Synthetic Gastritis Image Generation via Loss Function-Based Conditional PGGAN," in IEEE Access, vol. 7, pp. 87448-87457, 2019.
15. C. Han, H. Hayashi, L. Rundo, R. Araki, W. Shimoda, S. Muramatsu, Y. Furukawa, G. Mauri, H. Nakayama, "GAN-based synthetic brain MR image generation," 2018 IEEE 15th International Symposium on Biomedical Imaging, pp. 734-738, 2018.
16. D. Schütte, A., Hetzel, J., Gatidis, S., “Overcoming barriers to data sharing with medical image generation: a comprehensive evaluation.”, NPJ, Digit. Med. 4, 141, 2021.
17. (2021) International Skin Imaging Collaboration Website. [Online] Available: http://www.isdis.net/index.php/isic-project
18. A. Radford, L. Metz, and S. Chintala, “Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks,” arXiv preprint, arXiv: 1511.06434, 2016.
19. A. L. Maas, A. Y. Hannun, and A. Y. Ng, “Rectifier nonlinearities improve neural network acoustic models,” in Proc. ICML, vol. 30, p. 3, 2013.
20. D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint, arXiv: 1412.6980, 2014.
21. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting,” Journal of Machine Learning Research, vol. 15, pp. 1929–1958, 2014.
22. A. M. Turıng, “I.-Computıng Machınery and Intellıgence,” Mind, Volume LIX, Issue 236, pp. 433–460, 1950.