AI may be able to detect breast cancer more than 4 years before diagnosis

Publicly released:
International
CC-0. https://pixabay.com/photos/pink-ribbon-3715346/
CC-0. https://pixabay.com/photos/pink-ribbon-3715346/

Commercial artificial intelligence (AI) tools for detecting breast cancer based on screening mammograms can estimate the development of the disease years before diagnosis, according to Norwegian and US scientists. The team tested the AI on historical mamograms of 116,495 women aged 50 to 69 with no history of breast cancer produced in 2004. The women were followed up until 2018 to see if they developed breast cancer. The team found AI scores were higher for breasts developing than for breasts not developing cancer four to six years before the eventual detection of cancer.  AI could help identify women at high risk of breast cancer, allowing for an earlier cancer diagnosis and the chance of better treatment outcomes, the authors conclude.

Media release

From: JAMA

AI Algorithm for Subclinical Breast Cancer Detection

About The Study: In this retrospective cohort study of women undergoing screening mammography, mean absolute artificial intelligence (AI) scores were higher for breasts developing vs not developing cancer 4 to 6 years before their eventual detection. These findings suggest that commercial AI algorithms developed for breast cancer detection may identify women at high risk of a future breast cancer, offering a pathway for personalized screening approaches that can lead to earlier cancer diagnosis.

Attachments

Note: Not all attachments are visible to the general public. Research URLs will go live after the embargo ends.

Research JAMA, Web page The URL will go live after the embargo ends
Journal/
conference:
JAMA Network Open
Research:Paper
Organisation/s: Norwegian Institute of Public Health, Norway
Funder: The study was supported by the Cancer Society of Norway, and the Pink Ribbon Campaign sponsored this study (grant 214931). Drs Lee and Miglioretti are funded in part by the National Cancer Institute (grants R01CA262023 and P01CA154292).We thank Lunit Inc for the research use of their commercial AI algorithm.
Media Contact/s
Contact details are only visible to registered journalists.