35P Deep learning for overall survival risk prediction in early breast cancer using H&E-stained images and clinicopathological variables
Nikos Tsiknakis, Dimitrios Salgkamis, Evangelos Tzoras, Georgios C. Manikis, X. Liu, Kostas Marias, Balázs Ács, Johan Hartman, Mats Hellström, H. Johansson, A. Andersson, S. Loibl, Michael Untch, C. Denkert, Paul Jank, Ioannis Zerdes, Alexios Matikas, Jonas C. S. Bergh, Theodoros Foukakis · ESMO Open · 2025
Overall survival (OS) is a critical endpoint as it represents an unambiguous and clinically relevant endpoint in oncology trials. The aim of this study is to develop and validate an artificial intelligence/deep learning model to estimate a risk score for each patient based on OS prediction.