Magnification Generalization for Histopathology Image Embedding

TL;DR

This paper introduces a novel approach for histopathology image embedding that generalizes across different magnification levels using a meta-learning technique, addressing a gap in existing domain adaptation and generalization methods.

Contribution

First to address magnification generalization in histopathology embedding using a meta-learning approach based on MASF and MAML concepts.

Findings

Effective magnification generalization on breast cancer dataset

Outperforms existing methods in cross-magnification embedding tasks

Demonstrates robustness across four different magnification levels

Abstract

Histopathology image embedding is an active research area in computer vision. Most of the embedding models exclusively concentrate on a specific magnification level. However, a useful task in histopathology embedding is to train an embedding space regardless of the magnification level. Two main approaches for tackling this goal are domain adaptation and domain generalization, where the target magnification levels may or may not be introduced to the model in training, respectively. Although magnification adaptation is a well-studied topic in the literature, this paper, to the best of our knowledge, is the first work on magnification generalization for histopathology image embedding. We use an episodic trainable domain generalization technique for magnification generalization, namely Model Agnostic Learning of Semantic Features (MASF), which works based on the Model Agnostic Meta-Learning (MAML) concept. Our experimental results on a breast cancer histopathology dataset with four different magnification levels show the proposed method's effectiveness for magnification generalization.