The vast majority of semantic segmentation approaches rely on pixel-level annotations that are tedious and time consuming to obtain and suffer from significant inter and intra-expert variability. To address these issues, recent approaches have leveraged categorical annotations at the slide-level, that in general suffer from robustness and generalization. In this paper, we propose a novel weakly supervised multi-instance learning approach that deciphers quantitative slide-level annotations which are fast to obtain and regularly present in clinical routine. The extreme potentials of the proposed approach are demonstrated for tumor segmentation of solid cancer subtypes. The proposed approach achieves superior performance in out-of-distribution, out-of-location, and out-of-domain testing sets.
@article{arxiv.2105.04269,
title = {Weakly supervised pan-cancer segmentation tool},
author = {Marvin Lerousseau and Marion Classe and Enzo Battistella and Théo Estienne and Théophraste Henry and Amaury Leroy and Roger Sun and Maria Vakalopoulou and Jean-Yves Scoazec and Eric Deutsch and Nikos Paragios},
journal= {arXiv preprint arXiv:2105.04269},
year = {2021}
}