Weakly supervised pan-cancer segmentation tool
This work addresses the need for efficient and robust pan-cancer segmentation tools in clinical settings, reducing reliance on tedious pixel-level annotations.
The paper tackles the problem of tumor segmentation in solid cancer subtypes by proposing a weakly supervised multi-instance learning approach that uses slide-level annotations, achieving superior performance in out-of-distribution, out-of-location, and out-of-domain testing sets.
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.