The Explainability Paradox: Challenges for xAI in Digital Pathology

Abstract

The increasing prevalence of digitized workflows in diagnostic pathology opens the door to life-saving applications of artificial intelligence (AI). Explainability is identified as a critical component for the safety, approval and acceptance of AI systems for clinical use. Despite the cross-disciplinary challenge of building explainable AI (xAI), very few application- and user-centric studies in this domain have been carried out. We conducted the first mixed-methods study of user interaction with samples of state-of-the-art AI explainability techniques for digital pathology. This study reveals challenging dilemmas faced by developers of xAI solutions for medicine and proposes empirically-backed principles for their safer and more effective design.

@article{EVANS2022,
title = {The explainability paradox: Challenges for xAI in digital pathology},
journal = {Future Generation Computer Systems},
year = {2022},
issn = {0167-739X},
doi = {https://doi.org/10.1016/j.future.2022.03.009},
url = {https://www.sciencedirect.com/science/article/pii/S0167739X22000838},
author = {Theodore Evans and Carl Orge Retzlaff and Christian Geißler and Michaela Kargl and Markus Plass and Heimo Müller and Tim-Rasmus Kiehl and Norman Zerbe and Andreas Holzinger},
keywords = {Explainable AI, Digital pathology, Usability, Trust, Artificial intelligence},
abstract = {The increasing prevalence of digitized workflows in diagnostic pathology opens the door to life-saving applications of artificial intelligence (AI). Explainability is identified as a critical component for the safety, approval and acceptance of AI systems for clinical use. Despite the cross-disciplinary challenge of building explainable AI (xAI), very few application- and user-centric studies in this domain have been carried out. We conducted the first mixed-methods study of user interaction with samples of state-of-the-art AI explainability techniques for digital pathology. This study reveals challenging dilemmas faced by developers of xAI solutions for medicine and proposes empirically-backed principles for their safer and more effective design.}
}
Authors:
Theodore Evans, Christian Geißler, Carl Orge Retzlaff, Michaela Kargl, Markus Plass, Heimo Müller, Tim-Rasmus Kiehl, Norman Zerbe, Andreas Holzinger
Category:
Journal
Year:
2022
Location:
Future Generation Computer Systems
Link: