Open AccessTo explain or not to explain?—Artificial intelligence explainability in clinical decision support systems
Author(s) -
Julia Amann,
Dennis Vetter,
Stig Nikolaj Blomberg,
Helle Collatz Christensen,
Megan Coffee,
Sara Gerke,
Thomas Gilbert,
Thilo Hagendorff,
Sune Holm,
Michelle Livne,
Andy Spezzatti,
Inga Strümke,
Roberto Zicari,
Vince I. Madai,
AUTHOR_ID
Publication year - 2022
Publication title -
plos digital health
Language(s) - English
Resource type - Journals
ISSN - 2767-3170
DOI - 10.1371/journal.pdig.0000016
Subject(s) - context (archaeology) , key (lock) , clinical decision support system , computer science , normative , decision support system , process (computing) , artificial intelligence , intelligent decision support system , knowledge management , risk analysis (engineering) , medicine , computer security , paleontology , philosophy , epistemology , biology , operating system
Explainability for artificial intelligence (AI) in medicine is a hotly debated topic. Our paper presents a review of the key arguments in favor and against explainability for AI-powered Clinical Decision Support System (CDSS) applied to a concrete use case, namely an AI-powered CDSS currently used in the emergency call setting to identify patients with life-threatening cardiac arrest. More specifically, we performed a normative analysis using socio-technical scenarios to provide a nuanced account of the role of explainability for CDSSs for the concrete use case, allowing for abstractions to a more general level. Our analysis focused on three layers: technical considerations, human factors, and the designated system role in decision-making. Our findings suggest that whether explainability can provide added value to CDSS depends on several key questions: technical feasibility, the level of validation in case of explainable algorithms, the characteristics of the context in which the system is implemented, the designated role in the decision-making process, and the key user group(s). Thus, each CDSS will require an individualized assessment of explainability needs and we provide an example of how such an assessment could look like in practice.