RSNA 2021 Women's Imaging Preview

Deep-learning tool triages women with decreased breast density

By Amerigo Allegretto, staff writer

Wednesday, December 1 | 3:00 p.m.-4:00 p.m. | SSBR09-4 | Room TBA
In this Wednesday talk, researchers will present findings from their study of nearly 2,700 women that used a convolutional neural network in predicting mammographic density percentage from MRI scans.

Bas van der Velden, PhD, from University Medical Center Utrecht will talk about how the deep-learning tool (VolparaDensity version 1.5, Volpara Health) showed high specificity in correctly triaging more than half the number of women with decreased breast density, switching them from receiving MRI-only screening to receive mammographic screening.

The team included 4,300 MRI exams and trained their network through two screening rounds.

"AI-based triaging of women with extremely dense breasts from MRI-only screening back to mammography screening is feasible, enabling radiologists to reduce workload and reallocate resources," the study authors wrote.

Find out more about the network's performance in this session.