A new time-decay radiomics integrated network (TRINet) for breast cancer risk prediction

Hong Hui Yeoh, Fredrik Strand, Raphael C.-W. Phan, Kartini Rahmat, Maxine Tan

Abstract: Highlights•New deep learning model integrating radiomic features for risk prediction.•Introduce time-decay attention to focus on recent mammograms for better prediction.•Novel time embedding additive hazard layer for multi-year risk forecasting.•Continual learning with bilateral asymmetry detection for new dataset adaption.•Achieved AUC performance comparable to state-of-the-art methods.

External IDs:dblp:journals/mia/YeohSPRT26