Biological imaging data are often partially confounded or contain unwanted variability. Examples of such phenomena include variable lighting across microscopy image captures, stain intensity variation in histological slides, and batch effects for high throughput drug screening assays. Therefore, to develop "fair" models which generalise well to unseen examples, it is crucial to learn data representations that are insensitive to nuisance factors of variation. In this paper, we present a strategy based on adversarial training, capable of learning unsupervised representations invariant to confounders. As an empirical validation of our method, we use deep convolutional autoencoders to learn unbiased cellular representations from microscopy imaging.
@article{arxiv.1811.06498,
title = {Adjusting for Confounding in Unsupervised Latent Representations of Images},
author = {Craig A. Glastonbury and Michael Ferlaino and Christoffer Nellåker and Cecilia M. Lindgren},
journal= {arXiv preprint arXiv:1811.06498},
year = {2018}
}
Comments
Machine Learning for Health (ML4H) Workshop at NeurIPS 2018 arXiv:1811.07216