Uncertainty Estimation in Medical Image Classification: Systematic Review.
deep learning
medical image classification
medical imaging
network calibration
out-of-distribution detection
uncertainty estimation
Journal
JMIR medical informatics
ISSN: 2291-9694
Titre abrégé: JMIR Med Inform
Pays: Canada
ID NLM: 101645109
Informations de publication
Date de publication:
02 Aug 2022
02 Aug 2022
Historique:
received:
14
01
2022
accepted:
04
06
2022
revised:
11
04
2022
entrez:
2
8
2022
pubmed:
3
8
2022
medline:
3
8
2022
Statut:
epublish
Résumé
Deep neural networks are showing impressive results in different medical image classification tasks. However, for real-world applications, there is a need to estimate the network's uncertainty together with its prediction. In this review, we investigate in what form uncertainty estimation has been applied to the task of medical image classification. We also investigate which metrics are used to describe the effectiveness of the applied uncertainty estimation. Google Scholar, PubMed, IEEE Xplore, and ScienceDirect were screened for peer-reviewed studies, published between 2016 and 2021, that deal with uncertainty estimation in medical image classification. The search terms "uncertainty," "uncertainty estimation," "network calibration," and "out-of-distribution detection" were used in combination with the terms "medical images," "medical image analysis," and "medical image classification." A total of 22 papers were chosen for detailed analysis through the systematic review process. This paper provides a table for a systematic comparison of the included works with respect to the applied method for estimating the uncertainty. The applied methods for estimating uncertainties are diverse, but the sampling-based methods Monte-Carlo Dropout and Deep Ensembles are used most frequently. We concluded that future works can investigate the benefits of uncertainty estimation in collaborative settings of artificial intelligence systems and human experts. RR2-10.2196/11936.
Sections du résumé
BACKGROUND
BACKGROUND
Deep neural networks are showing impressive results in different medical image classification tasks. However, for real-world applications, there is a need to estimate the network's uncertainty together with its prediction.
OBJECTIVE
OBJECTIVE
In this review, we investigate in what form uncertainty estimation has been applied to the task of medical image classification. We also investigate which metrics are used to describe the effectiveness of the applied uncertainty estimation.
METHODS
METHODS
Google Scholar, PubMed, IEEE Xplore, and ScienceDirect were screened for peer-reviewed studies, published between 2016 and 2021, that deal with uncertainty estimation in medical image classification. The search terms "uncertainty," "uncertainty estimation," "network calibration," and "out-of-distribution detection" were used in combination with the terms "medical images," "medical image analysis," and "medical image classification."
RESULTS
RESULTS
A total of 22 papers were chosen for detailed analysis through the systematic review process. This paper provides a table for a systematic comparison of the included works with respect to the applied method for estimating the uncertainty.
CONCLUSIONS
CONCLUSIONS
The applied methods for estimating uncertainties are diverse, but the sampling-based methods Monte-Carlo Dropout and Deep Ensembles are used most frequently. We concluded that future works can investigate the benefits of uncertainty estimation in collaborative settings of artificial intelligence systems and human experts.
INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)
UNASSIGNED
RR2-10.2196/11936.
Identifiants
pubmed: 35916701
pii: v10i8e36427
doi: 10.2196/36427
pmc: PMC9382553
doi:
Types de publication
Journal Article
Review
Langues
eng
Pagination
e36427Informations de copyright
©Alexander Kurz, Katja Hauser, Hendrik Alexander Mehrtens, Eva Krieghoff-Henning, Achim Hekler, Jakob Nikolas Kather, Stefan Fröhling, Christof von Kalle, Titus Josef Brinker. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 02.08.2022.
Références
Med Image Anal. 2020 May;62:101653
pubmed: 32172037
Med Image Anal. 2020 Oct;65:101790
pubmed: 32801096
Med Image Anal. 2020 Aug;64:101724
pubmed: 32497870
J Clin Med. 2019 Aug 17;8(8):
pubmed: 31426482
Sci Rep. 2017 Dec 19;7(1):17816
pubmed: 29259224
IEEE Access. 2021 Jun 02;9:85442-85454
pubmed: 34812397
Comput Biol Med. 2021 Aug;135:104418
pubmed: 34052016
PLoS Med. 2021 Mar 29;18(3):e1003583
pubmed: 33780438
Med Image Anal. 2021 Feb;68:101855
pubmed: 33260116