Generative Image Translation for Data Augmentation in Colorectal Histopathology Images

TL;DR

This paper introduces a CycleGAN-based method to generate synthetic colorectal polyp images for data augmentation, improving CNN detection performance and maintaining image realism as validated by pathologists.

Contribution

The study presents a novel application of CycleGANs with a filtration module to generate realistic histopathology images, enhancing data diversity for better classification.

Findings

Generated images are indistinguishable from real images in Turing tests.

Data augmentation with generated images improves CNN AUC by over 10%.

Path-Rank-Filter enhances the quality of synthetic images.

Abstract

We present an image translation approach to generate augmented data for mitigating data imbalances in a dataset of histopathology images of colorectal polyps, adenomatous tumors that can lead to colorectal cancer if left untreated. By applying cycle-consistent generative adversarial networks (CycleGANs) to a source domain of normal colonic mucosa images, we generate synthetic colorectal polyp images that belong to diagnostically less common polyp classes. Generated images maintain the general structure of their source image but exhibit adenomatous features that can be enhanced with our proposed filtration module, called Path-Rank-Filter. We evaluate the quality of generated images through Turing tests with four gastrointestinal pathologists, finding that at least two of the four pathologists could not identify generated images at a statistically significant level. Finally, we demonstrate that using CycleGAN-generated images to augment training data improves the AUC of a convolutional neural network for detecting sessile serrated adenomas by over 10%, suggesting that our approach might warrant further research for other histopathology image classification tasks.