Wednesday, December 3 | 12:15 p.m.-12:45 p.m. | BRS274 | Lakeside Learning Center
Screening recall rates for 2D digital mammography went down after digital breast tomosynthesis (DBT) was added into clinical practice, suggesting that DBT offers yet another benefit, according to research to be presented at this Wednesday poster session.Dr. Nicole Lee, of Dartmouth-Hitchcock Medical Center, and colleagues compared aggregate digital mammography recall rates from February 2010 through January 2012, before DBT began to be used, with rates from February 2012 through January 2014, after the technology was implemented.
They found that 2D recall rates were, indeed, lower after DBT began to be used, going from 8.81% to 7.89%. DBT also appeared to lower 2D digital mammography recall rates for women with heterogeneously dense tissue (10.68% before and 9.29% after), as well as for women with scattered dense tissue (8.58% versus 7.85%). The differences were statistically significant.
"The etiology of this effect is unclear but may be related to a learning curve from experience with [DBT]," Lee and colleagues wrote.












![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)






