DGMIL: Distribution Guided Multiple Instance Learning for Whole Slide Image Classification

TL;DR

This paper introduces DGMIL, a novel MIL framework for whole slide image classification that leverages the inherent feature distribution of histopathological data to improve accuracy and patch localization.

Contribution

DGMIL models feature distributions explicitly and uses a pseudo label refinement strategy, offering a new approach to MIL without complex discriminative architectures.

Findings

Achieves state-of-the-art results on CAMELYON16 and TCGA Lung Cancer datasets.

Effectively separates positive and negative instances in feature space.

Improves both classification accuracy and patch localization performance.

Abstract

Multiple Instance Learning (MIL) is widely used in analyzing histopathological Whole Slide Images (WSIs). However, existing MIL methods do not explicitly model the data distribution, and instead they only learn a bag-level or instance-level decision boundary discriminatively by training a classifier. In this paper, we propose DGMIL: a feature distribution guided deep MIL framework for WSI classification and positive patch localization. Instead of designing complex discriminative network architectures, we reveal that the inherent feature distribution of histopathological image data can serve as a very effective guide for instance classification. We propose a cluster-conditioned feature distribution modeling method and a pseudo label-based iterative feature space refinement strategy so that in the final feature space the positive and negative instances can be easily separated. Experiments on the CAMELYON16 dataset and the TCGA Lung Cancer dataset show that our method achieves new SOTA for both global classification and positive patch localization tasks.