The Explanation Necessity for Healthcare AI

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mamalakis, Michail, de Vareilles, Héloïse, Murray, Graham, Lio, Pietro, Suckling, John
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912374835904512
author Mamalakis, Michail
de Vareilles, Héloïse
Murray, Graham
Lio, Pietro
Suckling, John
author_facet Mamalakis, Michail
de Vareilles, Héloïse
Murray, Graham
Lio, Pietro
Suckling, John
contents Explainability is a critical factor in enhancing the trustworthiness and acceptance of artificial intelligence (AI) in healthcare, where decisions directly impact patient outcomes. Despite advancements in AI interpretability, clear guidelines on when and to what extent explanations are required in medical applications remain lacking. We propose a novel categorization system comprising four classes of explanation necessity (self-explainable, semi-explainable, non-explainable, and new-patterns discovery), guiding the required level of explanation; whether local (patient or sample level), global (cohort or dataset level), or both. To support this system, we introduce a mathematical formulation that incorporates three key factors: (i) robustness of the evaluation protocol, (ii) variability of expert observations, and (iii) representation dimensionality of the application. This framework provides a practical tool for researchers to determine the appropriate depth of explainability needed, addressing the critical question: When does an AI medical application need to be explained, and at what level of detail?
format Preprint
id arxiv_https___arxiv_org_abs_2406_00216
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The Explanation Necessity for Healthcare AI
Mamalakis, Michail
de Vareilles, Héloïse
Murray, Graham
Lio, Pietro
Suckling, John
Artificial Intelligence
Explainability is a critical factor in enhancing the trustworthiness and acceptance of artificial intelligence (AI) in healthcare, where decisions directly impact patient outcomes. Despite advancements in AI interpretability, clear guidelines on when and to what extent explanations are required in medical applications remain lacking. We propose a novel categorization system comprising four classes of explanation necessity (self-explainable, semi-explainable, non-explainable, and new-patterns discovery), guiding the required level of explanation; whether local (patient or sample level), global (cohort or dataset level), or both. To support this system, we introduce a mathematical formulation that incorporates three key factors: (i) robustness of the evaluation protocol, (ii) variability of expert observations, and (iii) representation dimensionality of the application. This framework provides a practical tool for researchers to determine the appropriate depth of explainability needed, addressing the critical question: When does an AI medical application need to be explained, and at what level of detail?
title The Explanation Necessity for Healthcare AI
topic Artificial Intelligence
url https://arxiv.org/abs/2406.00216