Mind the Gap: Scanner-induced domain shifts pose challenges for representation learning in histopathology

TL;DR

This paper investigates the use of self-supervised pre-training to address scanner-induced domain shifts in histopathology, revealing limited benefits for tumor segmentation and providing insights into scanner effects.

Contribution

The study introduces Barlow Triplets for learning scanner-invariant representations and analyzes their impact on histopathology domain shifts.

Findings

Self-supervised pre-training aligns scanner representations.

Limited improvement in tumor segmentation performance.

Insights into scanner characteristics affecting downstream tasks.

Abstract

Computer-aided systems in histopathology are often challenged by various sources of domain shift that impact the performance of these algorithms considerably. We investigated the potential of using self-supervised pre-training to overcome scanner-induced domain shifts for the downstream task of tumor segmentation. For this, we present the Barlow Triplets to learn scanner-invariant representations from a multi-scanner dataset with local image correspondences. We show that self-supervised pre-training successfully aligned different scanner representations, which, interestingly only results in a limited benefit for our downstream task. We thereby provide insights into the influence of scanner characteristics for downstream applications and contribute to a better understanding of why established self-supervised methods have not yet shown the same success on histopathology data as they have for natural images.