Retrospective evaluation of a CE-marked AI system, including 1,017,208 mammography screening examinations
Question Results from large mammography screening samples not used in training AI algorithms are important to consider when planning prospective studies and implementation. Findings More than 80% of the screening-detected cancers were classified as positive by AI when considering 3% of the examinations with the highest AI risk score as positive. Clinical relevance A lack of radiologists is a challenge in mammographic screening. Our findings support other studies that suggest the use of AI to reduce screen-reading workload.
Authors
- Solveig Hofvind (ORCID: https://orcid.org/0000-0003-0178-8939)
- Tone Hovda (ORCID: https://orcid.org/0000-0001-9786-9455)
- Lars A. Akslen (ORCID: https://orcid.org/0000-0003-2710-9543)
- Marie Burns Bergan
- Jonas Gjesvik (ORCID: https://orcid.org/0009-0005-6756-3566)
- Marthe Larsen (ORCID: https://orcid.org/0000-0001-5437-0363)
Institutions
- Norwegian Institute of Public Health (NO)
- Vestre Viken Hospital Trust (NO)
- Haukeland University Hospital (NO)
- Cancer Registry of Norway (NO)
- University of Bergen (NO)
- UiT The Arctic University of Norway (NO)
Publication Details
- Journal
- European Radiology
- Published
- 2025-03-26
- DOI
- https://doi.org/10.1007/s00330-025-11521-4
- Citations
- 4
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 9.92
Funders
- Kreftforeningen
- Norwegian Institute of Public Health