Overview

Schema Version

https://atlas.rsna.org/schemas/2025-11/dataset.json

Name

Region of Southern Denmark Population-wide Mammography Screening Cohort (2014–2018)

Link

https://pubs.rsna.org/doi/10.1148/ryai.230529

Indexing

Keywords: Mammography, Screening, Double reading, Arbitration, Artificial intelligence, Triaging, Population-based, Denmark, Lunit INSIGHT MMG

Content: BR, RS

RadLex: RID28817, RID45946, RID10357, RID38811

SNOMED: 254837009

Author(s)

Mohammad T. Elhakim

Sarah W. Stougaard

Ole Graumann

Mads Nielsen

Oke Gerke

Lisbet B. Larsen

Benjamin S. B. Rasmussen

Organization(s)

Odense University Hospital

University of Southern Denmark

Aarhus University Hospital

Aarhus University

University of Copenhagen

Danish Clinical Quality Program–National Clinical Registries (RKKP)

Danish Breast Cancer Cooperative Group (DBCG)

Danish Quality Database on Mammography Screening (DKMS)

License

Text: Article is CC BY 4.0; dataset provided by third parties; no dataset-specific license stated.

URL: https://creativecommons.org/licenses/by/4.0/

Funding

Supported by the Innovation Fund by the Region of Southern Denmark (grant 10240300).

Ethical review

Approved by the national ethics committee (identifier D1763009); need for individual informed consent was waived.

Comments

Retrospective, multicenter cohort of all screening mammograms performed in the Region of Southern Denmark between August 4, 2014, and August 15, 2018, used to simulate AI-integrated screening scenarios and compare with standard double reading with arbitration.

Date

Published: 2024-09-04

Created: 2014-08-04

References

[1] Elhakim MT, Stougaard SW, Graumann O, et al.. "AI-integrated Screening to Replace Double Reading of Mammograms: A Population-wide Accuracy and Feasibility Study". Radiology: Artificial Intelligence. 2024-09-04. doi:10.1148/ryai.230529. PMID: 39230423. PMCID: PMC11605135.

[2] Kühl J, Elhakim MT, Stougaard SW, et al.. "Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms". European Radiology. 2024-01-01. PMID: 37938386. PMCID: PMC11166831. Available from: https://pubmed.ncbi.nlm.nih.gov/37938386/

Dataset

Motivation

To evaluate accuracy and feasibility of AI-integrated screening scenarios replacing one or both readers compared with standard double reading with arbitration.

Sampling

Consecutive cohort including all screening mammograms in the Region of Southern Denmark between Aug 4, 2014, and Aug 15, 2018; exclusions for insufficient image quality, missing images, lack of follow-up, unsupported image types, and technical issues are enumerated.

Missing information

Image file formats, resolution, and detailed acquisition parameters are not specified in the article.

Relationships between instances

Screening mammograms are linked to individual participants; AI produced per-view lesion scores aggregated to per-breast and examination-level scores; outcomes linked to registry-verified cancer status within 24 months or until next screening.

External data

Follow-up and cancer outcomes obtained by matching with Danish national breast cancer and screening databases (RKKP, DBCG, DKMS).

Confidentiality

Clinical screening data linked with national registries; handled under ethics approval; individual consent waived.

Sensitive data

Yes, contains health data including screening mammograms and cancer outcomes.