Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Dec 21;5(2):e220072.
doi: 10.1148/ryai.220072. eCollection 2023 Mar.

ADMANI: Annotated Digital Mammograms and Associated Non-Image Datasets

Helen M L Frazer  1 , Jennifer S N Tang  1 , Michael S Elliott  1 , Katrina M Kunicki  1 , Brendan Hill  1 , Ravishankar Karthik  1 , Chun Fung Kwok  1 , Carlos A Peña-Solorzano  1 , Yuanhong Chen  1 , Chong Wang  1 , Osamah Al-Qershi  1 , Samantha K Fox  1 , Shuai Li  1 , Enes Makalic  1 , Tuong L Nguyen  1 , Daniel F Schmidt  1 , Prabhathi Basnayake Ralalage  1 , Jocelyn F Lippey  1 , Peter Brotchie  1 , John L Hopper  1 , Gustavo Carneiro  1 , Davis J McCarthy  1
Affiliations

ADMANI: Annotated Digital Mammograms and Associated Non-Image Datasets

Helen M L Frazer et al. Radiol Artif Intell. .

Abstract

Supplemental material is available for this article. Keywords: Mammography, Screening, Convolutional Neural Network (CNN) Published under a CC BY 4.0 license. See also the commentary by Cadrin-Chênevert in this issue.

Keywords: Convolutional Neural Network (CNN); Mammography; Screening.

PubMed Disclaimer

Conflict of interest statement

Disclosures of conflicts of interest: H.M.L.F. Financial grant awarded by Australian Government Medical Research Future Fund (MRFF) funded development, in-kind support provided by the five partners (BreastScreen Victoria [BSV], St Vincent's Hospital Melbourne [SVHM], St Vincent's Institute of Medical Research [SVI], University of Melbourne [UOM], University of Adelaide [UOA]); honorary member of AI Advisory Committee for Royal Australian and New Zealand College of Radiologists. J.S.N.T. Employee of annalise.ai. M.S.E. No relevant relationships. K.M.K. No relevant relationships. B.H. Employed at SVI during 2021 and performed data engineering work for the results in this manuscript. R.K. No relevant relationships. C.F.K. No relevant relationships. C.A.P.S. No relevant relationships. Y.C. No relevant relationships. C.W. No relevant relationships. O.A.Q. No relevant relationships. S.K.F. No relevant relationships. S.L. No relevant relationships. E.M. No relevant relationships. T.L.N. No relevant relationships. D.F.S. National Breast Cancer Foundation Grant IIRS-18-093. P.B.R. No relevant relationships. J.F.L. No relevant relationships. P.B. Employee of annalise.ai. J.L.H. No relevant relationships. G.C. Employment at the UOA; project funded by MRFF grant; support for attending meetings and/or travel from MRFF grant; two patent applications. D.J.M. This manuscript arose from work supported by a grant awarded by the MRFF of Australia (no. MRFAI000090: BRAIx), grant payments made to author's institution and SVI and partially distributed to other partner institutions (SVHM, UOM, UOA, BSV); this work was also supported by a Ramaciotti Health Investment Grant from the Ramaciotti Foundation in Australia (Faster, cheaper, more accurate breast cancer screening: training computers to detect disease).

Figures

Overview of a BreastScreen Victoria screening episode with key clinical outcomes. CC = craniocaudal, L = left, MLO = mediolateral oblique, R = right, 2D = two-dimensional.
Figure 1:
Overview of a BreastScreen Victoria screening episode with key clinical outcomes. CC = craniocaudal, L = left, MLO = mediolateral oblique, R = right, 2D = two-dimensional.
A typical four-view mammogram from a screening episode. The final screening episode outcome for this woman was true positive, with a screen-detected cancer on the left breast as annotated. The left breast images were classified with this final screening episode outcome along with the associated reading and assessment outcomes. The lesion on the right breast was also indicated for assessment and determined to be benign. As such, the right breast images, although associated with a true-positive screening episode, were classified with the less prognostically significant reading (recall for assessment) and assessment (assessed normal with benign lesion) outcome. CC = craniocaudal, L = left, MLO = mediolateral, R = right.
Figure 2:
A typical four-view mammogram from a screening episode. The final screening episode outcome for this woman was true positive, with a screen-detected cancer on the left breast as annotated. The left breast images were classified with this final screening episode outcome along with the associated reading and assessment outcomes. The lesion on the right breast was also indicated for assessment and determined to be benign. As such, the right breast images, although associated with a true-positive screening episode, were classified with the less prognostically significant reading (recall for assessment) and assessment (assessed normal with benign lesion) outcome. CC = craniocaudal, L = left, MLO = mediolateral, R = right.
Manufacturer distribution across the datasets. ADMANI = Annotated Digital Mammograms and Associated Non-Image data.
Figure 3:
Manufacturer distribution across the datasets. ADMANI = Annotated Digital Mammograms and Associated Non-Image data.
See this image and copyright information in PMC

References

    1. American Institute for Cancer Research . Breast cancer . World Cancer Research Fund . https://www.wcrf.org/dietandcancer/breast-cancer/. Accessed August 11, 2021 .
    1. Australian Institute of Health and Welfare . BreastScreen Australia monitoring report 2020 . Australian Institute of Health and Welfare; ; 2020. .
    1. Morrell S , Taylor R , Roder D , Dobson A . Mammography screening and breast cancer mortality in Australia: an aggregate cohort study . J Med Screen 2012. ; 19 ( 1 ): 26 – 34 . - PubMed
    1. Rodriguez-Ruiz A , Lång K , Gubern-Merida A , et al. . Stand-alone artificial intelligence for breast cancer detection in mammography: comparison with 101 radiologists . J Natl Cancer Inst 2019. ; 111 ( 9 ): 916 – 922 . - PMC - PubMed
    1. Rodríguez-Ruiz A , Krupinski E , Mordang J-J , et al. . Detection of breast cancer with mammography: effect of an artificial intelligence support system . Radiology 2019. ; 290 ( 2 ): 305 – 314 . - PubMed

LinkOut - more resources