Fig. 1From: Deep learning for predicting future lesion emergence in high-risk breast MRI screening: a feasibility studyStudy overview for identifying anomalies in screening CE-MRI of the breast associated with future lesions. A deep learning model detects anomalies in the appearance change of breast CE-MRIs between screening visits. An elevated anomaly score in negative scans (BI-RADS 1) predicts the emergence of lesions in future screening scans. Anomalies may serve as a basis for dynamic risk profiling in breast cancer screening to optimize individual screening intervals. BI-RADS Breast Imaging and Reporting Data System, CE-MRI Contrast-enhanced magnetic resonance imagingBack to article page