Towards Explainable Graph Representations in Digital Pathology

TL;DR

This paper introduces a post-hoc explainer for graph-based representations in digital pathology, enhancing interpretability by highlighting key biological entities, with demonstrated effectiveness in breast cancer subtyping.

Contribution

It presents a generic, post-hoc explanation method for graph representations in digital pathology, focusing on biologically relevant entities like cells and interactions.

Findings

Effective in generating comprehensive explanations

Applicable to various topological representations

Demonstrated in breast cancer subtyping

Abstract

Explainability of machine learning (ML) techniques in digital pathology (DP) is of great significance to facilitate their wide adoption in clinics. Recently, graph techniques encoding relevant biological entities have been employed to represent and assess DP images. Such paradigm shift from pixel-wise to entity-wise analysis provides more control over concept representation. In this paper, we introduce a post-hoc explainer to derive compact per-instance explanations emphasizing diagnostically important entities in the graph. Although we focus our analyses to cells and cellular interactions in breast cancer subtyping, the proposed explainer is generic enough to be extended to other topological representations in DP. Qualitative and quantitative analyses demonstrate the efficacy of the explainer in generating comprehensive and compact explanations.