Monday, December 2 | 3:50 p.m.-4:00 p.m. | SSE02-06 | Room E451B
In this session, University of Pittsburgh researchers will discuss how whole-breast ultrasound performed by a technologist can find more cancers in women with dense breast tissue even after screening digital breast tomosynthesis (DBT) -- although it does increase the recall rate.Screening DBT's efficacy for women with dense breasts is uncertain, according to presenter Dr. Wendie Berg, PhD, and colleagues. The researchers sought to investigate the supplemental cancer detection rate from double reading DBT and compare it to adding technologist-performed whole-breast handheld screening ultrasound.
The study included 6,258 women between the ages of 40 and 75 with heterogeneously dense or extremely dense breasts who underwent yearly technologist-performed ultrasound after DBT for three years. Of 13,874 analyzable screens, 91 cancers were diagnosed, for a rate of 6.6 per 1,000 women screened.
Supplemental ultrasound found 0.9 cancers per 1,000 screens, compared with second-read DBT, which found 0.65 cancers per 1,000 screens. But ultrasound's recall rate was higher, at 50.4 per 1,000 women, compared with double-read DBT's rate of 36.9 per 1,000 women.
The findings indicate that more research is needed to determine ultrasound's supplemental screening role, Berg and colleagues concluded.
"Noninvasive methods to improve cancer detection, including double reading and screening ultrasound, merit consideration," the group noted.












![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)






