Impact of artificial intelligence in breast cancer screening with mammography.
Artificial intelligence
BI-RADS classification
Breast cancer
Mammography
Journal
Breast cancer (Tokyo, Japan)
ISSN: 1880-4233
Titre abrégé: Breast Cancer
Pays: Japan
ID NLM: 100888201
Informations de publication
Date de publication:
Nov 2022
Nov 2022
Historique:
received:
25
10
2021
accepted:
29
05
2022
pubmed:
29
6
2022
medline:
26
10
2022
entrez:
28
6
2022
Statut:
ppublish
Résumé
To demonstrate that radiologists, with the help of artificial intelligence (AI), are able to better classify screening mammograms into the correct breast imaging reporting and data system (BI-RADS) category, and as a secondary objective, to explore the impact of AI on cancer detection and mammogram interpretation time. A multi-reader, multi-case study with cross-over design, was performed, including 314 mammograms. Twelve radiologists interpreted the examinations in two sessions delayed by a 4 weeks wash-out period with and without AI support. For each breast of each mammogram, they had to mark the most suspicious lesion (if any) and assign it with a forced BI-RADS category and a level of suspicion or "continuous BI-RADS 100". Cohen's kappa correlation coefficient evaluating the inter-observer agreement for BI-RADS category per breast, and the area under the receiver operating characteristic curve (AUC), were used as metrics and analyzed. On average, the quadratic kappa coefficient increased significantly when using AI for all readers [κ = 0.549, 95% CI (0.528-0.571) without AI and κ = 0.626, 95% CI (0.607-0.6455) with AI]. AUC was significantly improved when using AI (0.74 vs 0.77, p = 0.004). Reading time was not significantly affected for all readers (106 s without AI and vs 102 s with AI; p = 0.754). When using AI, radiologists were able to better assign mammograms with the correct BI-RADS category without slowing down the interpretation time.
Identifiants
pubmed: 35763243
doi: 10.1007/s12282-022-01375-9
pii: 10.1007/s12282-022-01375-9
pmc: PMC9587927
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
967-977Informations de copyright
© 2022. The Author(s).
Références
Technol Cancer Res Treat. 2022 Jan-Dec;21:15330338221075172
pubmed: 35060413
J Digit Imaging. 2019 Aug;32(4):625-637
pubmed: 31011956
Acad Radiol. 2011 Feb;18(2):129-42
pubmed: 21232681
Med Image Anal. 2017 Jan;35:303-312
pubmed: 27497072
Lancet Digit Health. 2020 Sep;2(9):e468-e474
pubmed: 33328114
JAMA Intern Med. 2015 Nov;175(11):1828-37
pubmed: 26414882
Radiol Artif Intell. 2020 Nov 04;2(6):e190208
pubmed: 33937844
Acad Radiol. 2019 Jul;26(7):915-922
pubmed: 30268720
Clin Radiol. 2017 Aug;72(8):694.e1-694.e6
pubmed: 28381334
Nature. 2020 Jan;577(7788):89-94
pubmed: 31894144
Cochrane Database Syst Rev. 2006 Oct 18;(4):CD001877
pubmed: 17054145
Bull Cancer. 2019 Jul - Aug;106(7-8):684-692
pubmed: 31047637
Psychol Bull. 1968 Oct;70(4):213-20
pubmed: 19673146
Radiol Artif Intell. 2019 Jul 31;1(4):e180096
pubmed: 32076660
N Engl J Med. 2007 Apr 5;356(14):1399-409
pubmed: 17409321
Semin Oncol Nurs. 2017 May;33(2):141-155
pubmed: 28365057
JAMA Netw Open. 2020 Mar 2;3(3):e200265
pubmed: 32119094
J Am Coll Radiol. 2021 Jan;18(1 Pt A):79-86
pubmed: 33058789
BMJ. 2021 Sep 1;374:n1872
pubmed: 34470740
Eur J Radiol. 2013 Mar;82(3):388-97
pubmed: 22483607
Radiology. 2019 Feb;290(2):305-314
pubmed: 30457482
AJR Am J Roentgenol. 2019 Feb 19;:1-2
pubmed: 30779657
Eur Radiol. 2021 Nov;31(11):8682-8691
pubmed: 33948701
Diagn Interv Imaging. 2020 Dec;101(12):811-819
pubmed: 32819886
Nat Med. 2021 Feb;27(2):244-249
pubmed: 33432172
Eur Radiol. 2021 Sep;31(9):7058-7066
pubmed: 33744991
Eur Radiol. 2021 Mar;31(3):1687-1692
pubmed: 32876835
Radiology. 2008 Oct;249(1):47-53
pubmed: 18682584
Radiology. 2021 Jul;300(1):57-65
pubmed: 33944627
Acad Radiol. 2017 Jan;24(1):60-66
pubmed: 27793579
Presse Med. 2019 Oct;48(10):1101-1111
pubmed: 31676215
Expert Rev Med Devices. 2019 May;16(5):351-362
pubmed: 30999781