NYU Langone AI Tool Closes Radiology Training Gaps With 90% Accuracy
New York – September 30, 2026 -- Researchers at NYU Langone Health have built an artificial intelligence system that tracks the daily case exposure of radiology residents and flags missing pathologies with more than 90 percent accuracy, according to a study published online in Academic Radiology.
AI system pinpoints training gaps that traditional case logs miss
The tool monitors which conditions residents encounter during clinical rotations and automatically identifies underexposed pathologies, then assigns supplemental teaching cases to close the gap. Study author Vinay Prabhu, an associate professor in NYU Grossman School of Medicine's Department of Radiology, said residents seeing 30 cases in a day typically encounter 29 routine exams and only one uncommon condition, leaving gaps in exposure to rarer diseases they will face throughout their careers.