Synthetic Training with Generative Adversarial Networks for Segmentation of Microscopies
Medical imaging is often burdened with small available annotated data. In case of supervised deep learning algorithms a large amount of data is needed. One common strategy is to augment the given dataset for increasing the amount of training data. Recent researches show that the generation of synthetic images is a possible strategy to expand datasets. Especially, generative adversarial networks (GAN)s are promising candidates for generating new annotated training images. This work combines recent architectures of Generative Adversarial Networks in one pipeline to generate medical original and segmented image pairs for semantic segmentation. Results of training a U-Net with incorporated synthetic images as addition to common data augmentation are showing a performance boost compared to training without synthetic images from 77.99% to 80.23% average Jaccard Index.
Unable to display preview. Download preview PDF.
- 1.Ronneberger O, Fischer P, Brox T; Springer. U-net: convolutional networks for biomedical image segmentation. Proc MICCAI. 2015; p. 234-241.Google Scholar
- 2.Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets. Adv Neural Inf Process Syst. 2014; p. 2672-2680.Google Scholar
- 3.Frid-Adar M, Diamant I, Klang E, et al. GAN-based synthetic medical image augmentation for increased CNN Performance in Liver Lesion Classification. CoRR. 2018;abs/1803.01229.Google Scholar
- 4.Arjovsky M, Chintala S, Bottou L. Wasserstein generative adversarial networks. Proc 34th Int Conf Mach Learn. 2017;70:214-223.Google Scholar
- 5.Zhang H, Xu T, Li H, et al. StackGAN: text to photo-realistic image synthesis with stacked generative adversarial networks. CoRR. 2016;abs/1612.03242.Google Scholar
- 6.Wang T, Liu M, Zhu J, et al. High-resolution image synthesis and semantic manipulation with conditional GANs. CoRR. 2017;abs/1711.11585.Google Scholar
- 7.Gulrajani I, Ahmed F, Arjovsky M, et al. Improved training of Wasserstein gans. Adv Neural Inf Process Syst. 2017; p. 5769-5779.Google Scholar