Cargando…

Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?

We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning (ML) models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely...

Descripción completa

Detalles Bibliográficos
Autores principales: Yoon, Chang Ho, Torrance, Robert, Scheinerman, Naomi
Formato: Online Artículo Texto
Lenguaje:English
Publicado: BMJ Publishing Group 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9411871/
https://www.ncbi.nlm.nih.gov/pubmed/34006600
http://dx.doi.org/10.1136/medethics-2020-107102
_version_ 1784775362456059904
author Yoon, Chang Ho
Torrance, Robert
Scheinerman, Naomi
author_facet Yoon, Chang Ho
Torrance, Robert
Scheinerman, Naomi
author_sort Yoon, Chang Ho
collection PubMed
description We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning (ML) models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely addressed by current methods of appraising medical interventions like pharmacological therapies; second, a number of judicial precedents underpinning medical liability and negligence are compromised when ‘autonomous’ ML recommendations are considered to be en par with human instruction in specific contexts; third, explainable algorithms may be more amenable to the ascertainment and minimisation of biases, with repercussions for racial equity as well as scientific reproducibility and generalisability. We conclude with some reasons for the ineludible importance of interpretability, such as the establishment of trust, in overcoming perhaps the most difficult challenge ML will face in a high-stakes environment like healthcare: professional and public acceptance.
format Online
Article
Text
id pubmed-9411871
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher BMJ Publishing Group
record_format MEDLINE/PubMed
spelling pubmed-94118712022-09-12 Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned? Yoon, Chang Ho Torrance, Robert Scheinerman, Naomi J Med Ethics Clinical Ethics We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning (ML) models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely addressed by current methods of appraising medical interventions like pharmacological therapies; second, a number of judicial precedents underpinning medical liability and negligence are compromised when ‘autonomous’ ML recommendations are considered to be en par with human instruction in specific contexts; third, explainable algorithms may be more amenable to the ascertainment and minimisation of biases, with repercussions for racial equity as well as scientific reproducibility and generalisability. We conclude with some reasons for the ineludible importance of interpretability, such as the establishment of trust, in overcoming perhaps the most difficult challenge ML will face in a high-stakes environment like healthcare: professional and public acceptance. BMJ Publishing Group 2022-09 2021-05-18 /pmc/articles/PMC9411871/ /pubmed/34006600 http://dx.doi.org/10.1136/medethics-2020-107102 Text en © Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY. Published by BMJ. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.
spellingShingle Clinical Ethics
Yoon, Chang Ho
Torrance, Robert
Scheinerman, Naomi
Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
title Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
title_full Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
title_fullStr Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
title_full_unstemmed Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
title_short Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
title_sort machine learning in medicine: should the pursuit of enhanced interpretability be abandoned?
topic Clinical Ethics
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9411871/
https://www.ncbi.nlm.nih.gov/pubmed/34006600
http://dx.doi.org/10.1136/medethics-2020-107102
work_keys_str_mv AT yoonchangho machinelearninginmedicineshouldthepursuitofenhancedinterpretabilitybeabandoned
AT torrancerobert machinelearninginmedicineshouldthepursuitofenhancedinterpretabilitybeabandoned
AT scheinermannaomi machinelearninginmedicineshouldthepursuitofenhancedinterpretabilitybeabandoned