Sunday, November 25 | 10:45 a.m.-10:55 a.m. | SSA01-01 | Arie Crown Theater
Could ultrasound serve as the primary screening test for breast cancer? Perhaps, according to Dr. Wendie Berg, PhD, from the University of Pittsburgh. In this Sunday morning presentation, Berg will discuss results from a study that used American College of Radiology Imaging Network (ACRIN) 6666 data to evaluate outcomes of screening if ultrasound were the primary screening modality.For the study, 2,662 participants from 21 sites in the U.S., Canada, and Argentina completed three annual rounds of screening (a total of 7,473 exams) with mammography and whole-breast ultrasound performed and interpreted by physicians. In the fourth year, each participant had either biopsy or a 12-month clinical follow-up.
The cohort included 110 women who were diagnosed with 111 breast cancer events, 80% of which were invasive. The total number of cancers detected by ultrasound or mammography was comparable, at 58 out of 111 (53%) versus 59 out of 111 (53%). Berg's group also found that the vast majority of the cancers detected by ultrasound were invasive (53 out of 58, or 91%), compared with 41 out of 59 (69%) by mammography.
Cancer detection with ultrasound alone compares favorably to mammography alone, and ultrasound offers greater detection of node-negative invasive cancer than mammography, according to the researchers.










![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)






