RMDL: Recalibrated multi-instance deep learning for whole slide gastric image classification

TL;DR

This paper introduces RMDL, a recalibrated multi-instance deep learning approach that improves gastric cancer diagnosis from large whole slide images by selecting discriminative regions and modeling their dependencies, achieving higher accuracy.

Contribution

The paper proposes a novel RMDL framework that captures instance dependencies and recalibrates features, along with a large annotated gastric histopathology dataset for evaluation.

Findings

Significant accuracy improvement over existing methods

Effective selection of discriminative instances

Framework generalizes to other cancer diagnosis tasks

Abstract

The whole slide histopathology images (WSIs) play a critical role in gastric cancer diagnosis. However, due to the large scale of WSIs and various sizes of the abnormal area, how to select informative regions and analyze them are quite challenging during the automatic diagnosis process. The multi-instance learning based on the most discriminative instances can be of great benefit for whole slide gastric image diagnosis. In this paper, we design a recalibrated multi-instance deep learning method (RMDL) to address this challenging problem. We first select the discriminative instances, and then utilize these instances to diagnose diseases based on the proposed RMDL approach. The designed RMDL network is capable of capturing instance-wise dependencies and recalibrating instance features according to the importance coefficient learned from the fused features. Furthermore, we build a large whole-slide gastric histopathology image dataset with detailed pixel-level annotations. Experimental results on the constructed gastric dataset demonstrate the significant improvement on the accuracy of our proposed framework compared with other state-of-the-art multi-instance learning methods. Moreover, our method is general and can be extended to other diagnosis tasks of different cancer types based on WSIs.