Combination model achieved higher AUC for interval and long-term cancers grouped together
By Elana Gotkine HealthDay Reporter
TUESDAY, Aug. 29, 2023 (HealthDay News) — Combining artificial intelligence (AI) with mammography texture models improves breast cancer risk assessment, according to a study published online Aug. 29 in Radiology.
Andreas D. Lauritzen, Ph.D., from the University of Copenhagen in Denmark, and colleagues conducted a retrospective study of Danish women consecutively screened for breast cancer to examine whether risk assessment improves when combining a diagnostic AI system for lesions and a mammographic texture model. The individual and combined performance of the AI and texture models were used for the prediction of future cancers in women with a negative screening mammogram, including those with interval cancers and long-term cancers.
A total of 119,650 women were included in the screening cohort, of whom 320 developed interval cancers and 1,401 developed long-term cancers. The researchers found that for interval and long-term cancer grouped together, the combination model achieved a higher area under the curve (AUC) than the diagnostic AI or the texture risk models (AUC, 0.73 versus 0.70 and 0.66, respectively). Overall, 44.1 and 33.7 percent of interval cancers and long-term cancers, respectively, were accounted for by the 10 percent of women with the highest combined risk identified by the combination model.
“Using our model, risk can be assessed with the same performance as the clinical risk models but within seconds from screening and without introducing overhead in the clinic,” Lauritzen said in a statement.
Several authors disclosed ties to the medical device industry.
Editorial (subscription or payment may be required)
Copyright © 2023 HealthDay. All rights reserved.
