MetaHistoSeg: A Python Framework for Meta Learning in Histopathology Image Segmentation

TL;DR

MetaHistoSeg is a Python framework that facilitates meta-learning and transfer learning for histopathology image segmentation, providing tools and datasets to improve cross-domain generalization and rapid experimentation.

Contribution

It introduces a flexible framework and a benchmark dataset for meta-learning in histopathology, addressing cross-domain generalization challenges.

Findings

Meta-learning and transfer learning perform comparably on average.

Task-specific benefits vary between meta-learning and transfer learning.

The framework enables rapid model development and benchmarking.

Abstract

Few-shot learning is a standard practice in most deep learning based histopathology image segmentation, given the relatively low number of digitized slides that are generally available. While many models have been developed for domain specific histopathology image segmentation, cross-domain generalization remains a key challenge for properly validating models. Here, tooling and datasets to benchmark model performance across histopathological domains are lacking. To address this limitation, we introduce MetaHistoSeg - a Python framework that implements unique scenarios in both meta learning and instance based transfer learning. Designed for easy extension to customized datasets and task sampling schemes, the framework empowers researchers with the ability of rapid model design and experimentation. We also curate a histopathology meta dataset - a benchmark dataset for training and validating models on out-of-distribution performance across a range of cancer types. In experiments we showcase the usage of MetaHistoSeg with the meta dataset and find that both meta-learning and instance based transfer learning deliver comparable results on average, but in some cases tasks can greatly benefit from one over the other.