Quantifying Explainers of Graph Neural Networks in Computational Pathology

11/25/2020
by   Guillaume Jaume, et al.
13

Explainability of deep learning methods is imperative to facilitate their clinical adoption in digital pathology. However, popular deep learning methods and explainability techniques (explainers) based on pixel-wise processing disregard biological entities' notion, thus complicating comprehension by pathologists. In this work, we address this by adopting biological entity-based graph processing and graph explainers enabling explanations accessible to pathologists. In this context, a major challenge becomes to discern meaningful explainers, particularly in a standardized and quantifiable fashion. To this end, we propose herein a set of novel quantitative metrics based on statistics of class separability using pathologically measurable concepts to characterize graph explainers. We employ the proposed metrics to evaluate three types of graph explainers, namely the layer-wise relevance propagation, gradient-based saliency, and graph pruning approaches, to explain Cell-Graph representations for Breast Cancer Subtyping. The proposed metrics are also applicable in other domains by using domain-specific intuitive concepts. We validate the qualitative and quantitative findings on the BRACS dataset, a large cohort of breast cancer RoIs, by expert pathologists.

READ FULL TEXT

page 1

page 7

page 18

page 19

research
07/01/2020

Towards Explainable Graph Representations in Digital Pathology

Explainability of machine learning (ML) techniques in digital pathology ...
research
11/03/2022

Exploring Explainability Methods for Graph Neural Networks

With the growing use of deep learning methods, particularly graph neural...
research
07/03/2019

A Semi-Supervised Framework for Automatic Pixel-Wise Breast Cancer Grading of Histological Images

Throughout the world, breast cancer is one of the leading causes of fema...
research
02/22/2021

Hierarchical Cell-to-Tissue Graph Representations for Breast Cancer Subtyping in Digital Pathology

Cancer diagnosis and prognosis for a tissue specimen are heavily influen...
research
06/27/2021

An XAI Approach to Deep Learning Models in the Detection of Ductal Carcinoma in Situ

During the last decade or so, there has been an insurgence in the deep l...
research
03/21/2023

Unlocking Layer-wise Relevance Propagation for Autoencoders

Autoencoders are a powerful and versatile tool often used for various pr...
research
06/14/2023

Explainable and Position-Aware Learning in Digital Pathology

Encoding whole slide images (WSI) as graphs is well motivated since it m...

Please sign up or login with your details

Forgot password? Click here to reset