DeepSurv model using mammography features and age has similar performance to modern models
By Elana Gotkine HealthDay Reporter
FRIDAY, Sept. 19, 2025 (HealthDay News) — A deep learning model based on mammographic features and age predicts cardiovascular risk with performance comparable to traditional risk equations, according to a study published online Sept. 16 in Heart.
Jennifer Yvonne Barraclough, Ph.D., from The George Institute for Global Health in Sydney, Australia, and colleagues developed and tested a deep learning algorithm for cardiovascular risk prediction based on routine mammography images (DeepSurv). The concordance index was used to compare model performance with standard risk prediction models.
Overall, 3,392 of the 49,196 women included in the study, with median follow-up of 8.8 years, experienced a first major cardiovascular event. The researchers found that the concordance index was 0.72 for the DeepSurv model using mammography features and participant age, and the model performed similarly to modern models containing age and clinical variables, such as the New Zealand PREDICT tool and the American Heart Association PREVENT equations.
“By integrating CV risk screening with breast screening through the use of mammograms — something many women already engage with at a stage in life when their cardiovascular risk increases — we can identify and potentially prevent two major causes of illness and death at the same time,” study coauthor Clare Arnott, Ph.D., also from the George Institute for Global Health, said in a statement.
Several authors disclosed ties to the biopharmaceutical and medical technology industries.
Editorial (subscription or payment may be required)
Copyright © 2025 HealthDay. All rights reserved.
