#machine learning
Mar 2026
Longitudinal Risk Prediction in Mammography with Privileged History Distillation
SEM-HD consistently improves long-horizon AUC and pAUC over longitudinal models evaluated without history, particularly in the clinically relevant low false-positive-rate region and recovers much of the performance gap with respect to full-history inference across datasets and backbones.
Banafsheh Karimian, Alexis Guichemerre, Soufiane Belharbi et al.
· arXiv.org · 0 citations