Difficulty Translation in Histopathology Images
arXiv:2004.12535
Abstract
The unique nature of histopathology images opens the door to domain-specific formulations of image translation models. We propose a difficulty translation model that modifies colorectal histopathology images to be more challenging to classify. Our model comprises a scorer, which provides an output confidence to measure the difficulty of images, and an image translator, which learns to translate images from easy-to-classify to hard-to-classify using a training set defined by the scorer. We present three findings. First, generated images were indeed harder to classify for both human pathologists and machine learning classifiers than their corresponding source images. Second, image classifiers trained with generated images as augmented data performed better on both easy and hard images from an independent test set. Finally, human annotator agreement and our model's measure of difficulty correlated strongly, implying that for future work requiring human annotator agreement, the confidence score of a machine learning classifier could be used as a proxy.
arXiv commentAccepted to 2020 Artificial Intelligence in Medicine (AIME) conference. Invited for long oral presentation
References in corpus (6)
- Deep Residual Learning for Image Recognition
- Going Deeper with Convolutions
- Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systems
- Curriculum Learning by Transfer Learning: Theory and Experiments with Deep Networks
- Neural Stain-Style Transfer Learning using GAN for Histopathological Images
- Deep-Learning for Classification of Colorectal Polyps on Whole-Slide Images