Weakly supervised pan-cancer segmentation tool

TL;DR

This paper introduces a weakly supervised multi-instance learning method for cancer segmentation that uses slide-level annotations, offering a faster and more robust alternative to pixel-level annotations, with strong performance across diverse datasets.

Contribution

The paper presents a novel weakly supervised multi-instance learning approach that effectively utilizes slide-level annotations for tumor segmentation in various cancer subtypes.

Findings

Achieves superior performance in out-of-distribution testing

Demonstrates robustness across different cancer subtypes

Uses annotations that are quick to obtain in clinical routine

Abstract

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.