Synthetic data for unsupervised polyp segmentation
This work addresses data scarcity and privacy issues in medical imaging for researchers and clinicians, though it is incremental as it builds on existing synthetic data and GAN methods.
The paper tackles the problem of limited annotated medical datasets for polyp segmentation by generating realistic synthetic colon images using 3D technologies and GANs, achieving promising results on five real datasets without any manual annotations.
Deep learning has shown excellent performance in analysing medical images. However, datasets are difficult to obtain due privacy issues, standardization problems, and lack of annotations. We address these problems by producing realistic synthetic images using a combination of 3D technologies and generative adversarial networks. We use zero annotations from medical professionals in our pipeline. Our fully unsupervised method achieves promising results on five real polyp segmentation datasets. As a part of this study we release Synth-Colon, an entirely synthetic dataset that includes 20000 realistic colon images and additional details about depth and 3D geometry: https://enric1994.github.io/synth-colon