Artificial intelligence-based computer-assisted detection/diagnosis (AI-CAD) for screening mammography: Outcomes of AI-CAD in the mammographic interpretation workflow.
Artificial intelligence
Breast cancer screening
Computer-assisted detection
Computer-assisted diagnosis
Mammography
Journal
European journal of radiology open
ISSN: 2352-0477
Titre abrégé: Eur J Radiol Open
Pays: England
ID NLM: 101650225
Informations de publication
Date de publication:
Dec 2023
Dec 2023
Historique:
received:
07
05
2023
revised:
03
07
2023
accepted:
09
07
2023
medline:
24
7
2023
pubmed:
24
7
2023
entrez:
24
7
2023
Statut:
epublish
Résumé
To evaluate the stand-alone diagnostic performances of AI-CAD and outcomes of AI-CAD detected abnormalities when applied to the mammographic interpretation workflow. From January 2016 to December 2017, 6499 screening mammograms of 5228 women were collected from a single screening facility. Historic reads of three radiologists were used as radiologist interpretation. A commercially-available AI-CAD was used for analysis. One radiologist not involved in interpretation had retrospectively reviewed the abnormality features and assessed the significance (negligible vs. need recall) of the AI-CAD marks. Ground truth in terms of cancer, benign or absence of abnormality was confirmed according to histopathologic diagnosis or negative results on the next-round screen. Of the 6499 mammograms, 6282 (96.7%) were in the negative, 189 (2.9%) were in the benign, and 28 (0.4%) were in the cancer group. AI-CAD detected 5 (17.9%, 5 of 28) of the 9 cancers that were intially interpreted as negative. Of the 648 AI-CAD recalls, 89.0% (577 of 648) were marks seen on examinations in the negative group, and 267 (41.2%) of the AI-CAD marks were considered to be negligible. Stand-alone AI-CAD has significantly higher recall rates (10.0% vs. 3.4%, AI-CAD detected 17.9% additional cancers on screening mammography that were initially overlooked by the radiologists. In spite of the additional cancer detection, AI-CAD had significantly higher recall rates in the clinical workflow, in which 89.0% of AI-CAD marks are on negative mammograms.
Identifiants
pubmed: 37484980
doi: 10.1016/j.ejro.2023.100509
pii: S2352-0477(23)00035-7
pmc: PMC10362167
doi:
Types de publication
Journal Article
Langues
eng
Pagination
100509Informations de copyright
© 2023 The Authors.
Déclaration de conflit d'intérêts
Authors have no conflicts of interest regarding this manuscript.
Références
JAMA Netw Open. 2020 Mar 2;3(3):e200265
pubmed: 32119094
Radiology. 2009 Dec;253(3):641-51
pubmed: 19864507
Med Image Anal. 2017 Jan;35:303-312
pubmed: 27497072
Radiology. 2006 Oct;241(1):55-66
pubmed: 16990671
Radiology. 2019 Feb;290(2):305-314
pubmed: 30457482
Eur Radiol. 2022 Nov;32(11):7400-7408
pubmed: 35499564
Radiology. 2006 May;239(2):375-83
pubmed: 16569779
Lancet. 2002 Mar 16;359(9310):909-19
pubmed: 11918907
Radiology. 2017 Apr;283(1):49-58
pubmed: 27918707
Lancet Digit Health. 2020 Mar;2(3):e138-e148
pubmed: 33334578
Eur Radiol. 2021 Aug;31(8):5940-5947
pubmed: 33486604
J Natl Cancer Inst. 2019 Sep 1;111(9):916-922
pubmed: 30834436
J Digit Imaging. 2019 Aug;32(4):618-624
pubmed: 30963339
J Natl Cancer Inst. 2014 Mar;106(3):djt461
pubmed: 24598715
Cancer. 2002 Aug 1;95(3):458-69
pubmed: 12209737
JAMA Oncol. 2020 Oct 1;6(10):1581-1588
pubmed: 32852536
Radiology. 2001 Sep;220(3):781-6
pubmed: 11526282
Eur J Cancer. 2008 Apr;44(6):798-807
pubmed: 18353630
Radiology. 2009 Dec;253(3):632-40
pubmed: 19789234
Nature. 2020 Jan;577(7788):89-94
pubmed: 31894144
J Natl Cancer Inst. 2000 Jul 5;92(13):1081-7
pubmed: 10880551
Radiology. 2012 Oct;265(1):59-69
pubmed: 22723501
AJR Am J Roentgenol. 2022 Sep;219(3):369-380
pubmed: 35018795
Korean J Radiol. 2021 Aug;22(8):1225-1239
pubmed: 33987993
J Natl Cancer Inst. 2008 Jun 18;100(12):876-87
pubmed: 18544742