Multi-Channel Auto-Encoders and a Novel Dataset for Learning Domain Invariant Representations of Histopathology Images

TL;DR

This paper introduces the Multi-Channel Auto-Encoder (MCAE) for learning domain-invariant features in histopathology images, demonstrating improved generalization across different data domains and a new synthetic dataset for evaluation.

Contribution

The paper extends the Dual-Channel Auto-Encoder to handle multiple domains and introduces a synthetic dataset generated with CycleGANs for robust evaluation.

Findings

MCAE produces less domain-sensitive features than existing methods.

MCAE outperforms StaNoSA by 5 percentage points in tissue classification.

The synthetic dataset enables testing of model generalization to appearance variations.

Abstract

Domain shift is a problem commonly encountered when developing automated histopathology pipelines. The performance of machine learning models such as convolutional neural networks within automated histopathology pipelines is often diminished when applying them to novel data domains due to factors arising from differing staining and scanning protocols. The Dual-Channel Auto-Encoder (DCAE) model was previously shown to produce feature representations that are less sensitive to appearance variation introduced by different digital slide scanners. In this work, the Multi-Channel Auto-Encoder (MCAE) model is presented as an extension to DCAE which learns from more than two domains of data. Additionally, a synthetic dataset is generated using CycleGANs that contains aligned tissue images that have had their appearance synthetically modified. Experimental results show that the MCAE model produces feature representations that are less sensitive to inter-domain variations than the comparative StaNoSA method when tested on the novel synthetic data. Additionally, the MCAE and StaNoSA models are tested on a novel tissue classification task. The results of this experiment show the MCAE model out performs the StaNoSA model by 5 percentage-points in the f1-score. These results show that the MCAE model is able to generalise better to novel data and tasks than existing approaches by actively learning normalised feature representations.