AI-Assisted Mammography Interpretation Across Training Levels: A Prospective Pilot Study

Prospective, single-center study of 240 mammograms found AI-assisted interpretation impacted reader performance metrics across training levels (general radiologist to technologist).

• Design: Prospective, single-center diagnostic accuracy study; 240 mammographic studies (358 breasts) interpreted by 4 readers (general radiologist, senior resident, junior resident, technologist) with and without AI assistance (tools: MammoScreen, INSIGHT MMG).
• Limitation: Single center, small reader pool (4 readers); pilot nature limits generalizability and subgroup comparisons.

RadPigeon summaries are original and for information only. They are not clinical advice.