The ACROBAT 2022 Challenge: Automatic Registration Of Breast Cancer Tissue

TL;DR

This paper presents the ACROBAT 2022 challenge, benchmarking eight algorithms on the largest dataset to date for aligning breast cancer tissue images, revealing insights into method performance and influencing factors.

Contribution

It introduces a large-scale WSI registration benchmark, compares multiple algorithms, and analyzes factors affecting registration accuracy in breast cancer tissue images.

Findings

Different registration methods can achieve high accuracy.

Clinical covariates influence registration performance.

The dataset enables comprehensive benchmarking.

Abstract

The alignment of tissue between histopathological whole-slide-images (WSI) is crucial for research and clinical applications. Advances in computing, deep learning, and availability of large WSI datasets have revolutionised WSI analysis. Therefore, the current state-of-the-art in WSI registration is unclear. To address this, we conducted the ACROBAT challenge, based on the largest WSI registration dataset to date, including 4,212 WSIs from 1,152 breast cancer patients. The challenge objective was to align WSIs of tissue that was stained with routine diagnostic immunohistochemistry to its H&E-stained counterpart. We compare the performance of eight WSI registration algorithms, including an investigation of the impact of different WSI properties and clinical covariates. We find that conceptually distinct WSI registration methods can lead to highly accurate registration performances and identify covariates that impact performances across methods. These results establish the current state-of-the-art in WSI registration and guide researchers in selecting and developing methods.