Cross-Task Pretraining for Cross-Organ Cross-Scanner Adenocarcinoma Segmentation

TL;DR

This paper presents a cross-task pretraining approach to improve domain generalization in histopathological image segmentation across different organs and scanners, addressing domain shift challenges in medical imaging.

Contribution

The study introduces cross-task pretraining as a novel strategy to mitigate domain shift in adenocarcinoma segmentation across organs and scanners.

Findings

Cross-task pretraining outperforms standard training methods.

Pretraining on one dataset and fine-tuning improves segmentation on unseen organs.

Combining datasets does not always yield better results than pretraining.

Abstract

This short abstract describes a solution to the COSAS 2024 competition on Cross-Organ and Cross-Scanner Adenocarcinoma Segmentation from histopathological image patches. The main challenge in the task of segmenting this type of cancer is a noticeable domain shift encountered when changing acquisition devices (microscopes) and also when tissue comes from different organs. The two tasks proposed in COSAS were to train on a dataset of images from three different organs, and then predict segmentations on data from unseen organs (dataset T1), and to train on a dataset of images acquired on three different scanners and then segment images acquired with another unseen microscope. We attempted to bridge the domain shift gap by experimenting with three different strategies: standard training for each dataset, pretraining on dataset T1 and then fine-tuning on dataset T2 (and vice-versa, a strategy we call \textit{Cross-Task Pretraining}), and training on the combination of dataset A and B. Our experiments showed that Cross-Task Pre-training is a more promising approach to domain generalization.