A Radiology-focused Review of Predictive Uncertainty for AI Interpretability in Computer-assisted Segmentation.
Bayesian Network (BN)
Ethics
Quantification
Segmentation
Journal
Radiology. Artificial intelligence
ISSN: 2638-6100
Titre abrégé: Radiol Artif Intell
Pays: United States
ID NLM: 101746556
Informations de publication
Date de publication:
Nov 2021
Nov 2021
Historique:
received:
22
01
2021
revised:
09
08
2021
accepted:
25
08
2021
entrez:
6
12
2021
pubmed:
7
12
2021
medline:
7
12
2021
Statut:
epublish
Résumé
The recent advances and availability of computer hardware, software tools, and massive digital data archives have enabled the rapid development of artificial intelligence (AI) applications. Concerns over whether AI tools can "communicate" decisions to radiologists and primary care physicians is of particular importance because automated clinical decisions can substantially impact patient outcome. A challenge facing the clinical implementation of AI stems from the potential lack of trust clinicians have in these predictive models. This review will expand on the existing literature on interpretability methods for deep learning and review the state-of-the-art methods for predictive uncertainty estimation for computer-assisted segmentation tasks. Last, we discuss how uncertainty can improve predictive performance and model interpretability and can act as a tool to help foster trust.
Identifiants
pubmed: 34870219
doi: 10.1148/ryai.2021210031
pmc: PMC8637228
doi:
Types de publication
Journal Article
Review
Langues
eng
Pagination
e210031Informations de copyright
2021 by the Radiological Society of North America, Inc.
Déclaration de conflit d'intérêts
Disclosures of Conflicts of Interest: B.M. Institution received grant from Defence Research and Development Canada (grant number CFPMN2-017-McMaster) for the Innovation for Defence Excellence and Security program presented by the DRDC. K.Z. Consultant for the Centre for Probe Development and Commercialization. T.E.D. No relevant relationships. M.D.N. Data analytics consultant for MCI OneHealth; CEO and cofounder of TBI Finder.
Références
Psychol Sci. 2018 Apr;29(4):504-520
pubmed: 29466077
J Med Ethics. 2020 Mar;46(3):205-211
pubmed: 31748206
Lancet Digit Health. 2019 Oct;1(6):e271-e297
pubmed: 33323251
Sci Data. 2017 Sep 05;4:170117
pubmed: 28872634
IEEE Trans Med Imaging. 2015 Oct;34(10):1993-2024
pubmed: 25494501
Radiol Artif Intell. 2019 Mar 27;1(2):190021
pubmed: 33937789
Radiol Artif Intell. 2020 May 27;2(3):e190043
pubmed: 32510054
Neurocomputing. 2019 Sep 3;335:34-45
pubmed: 31595105
Nature. 2015 May 28;521(7553):436-44
pubmed: 26017442
Nat Commun. 2020 Jul 22;11(1):3673
pubmed: 32699250
NPJ Digit Med. 2021 Feb 19;4(1):31
pubmed: 33608629
IEEE Trans Med Imaging. 2020 Dec;39(12):3868-3878
pubmed: 32746129
Nat Med. 2018 Sep;24(9):1342-1350
pubmed: 30104768