Pseudo-Bag Mixup Augmentation for Multiple Instance Learning-Based Whole Slide Image Classification

TL;DR

This paper introduces Pseudo-bag Mixup (PseMix), a novel data augmentation method for MIL-based WSI classification that enhances model performance, robustness, and generalization without extra computational costs.

Contribution

It proposes PseMix, a generalization of Mixup for WSIs using pseudo-bags, improving training efficiency and model robustness in MIL-based WSI classification.

Findings

PseMix improves classification accuracy of state-of-the-art MIL models.

PseMix enhances model robustness to occlusion and noise.

PseMix boosts generalization in challenging test scenarios.

Abstract

Given the special situation of modeling gigapixel images, multiple instance learning (MIL) has become one of the most important frameworks for Whole Slide Image (WSI) classification. In current practice, most MIL networks often face two unavoidable problems in training: i) insufficient WSI data and ii) the sample memorization inclination inherent in neural networks. These problems may hinder MIL models from adequate and efficient training, suppressing the continuous performance promotion of classification models on WSIs. Inspired by the basic idea of Mixup, this paper proposes a new Pseudo-bag Mixup (PseMix) data augmentation scheme to improve the training of MIL models. This scheme generalizes the Mixup strategy for general images to special WSIs via pseudo-bags so as to be applied in MIL-based WSI classification. Cooperated by pseudo-bags, our PseMix fulfills the critical size alignment and semantic alignment in Mixup strategy. Moreover, it is designed as an efficient and decoupled method, neither involving time-consuming operations nor relying on MIL model predictions. Comparative experiments and ablation studies are specially designed to evaluate the effectiveness and advantages of our PseMix. Experimental results show that PseMix could often assist state-of-the-art MIL networks to refresh their classification performance on WSIs. Besides, it could also boost the generalization performance of MIL models in special test scenarios, and promote their robustness to patch occlusion and label noise. Our source code is available at https://github.com/liupei101/PseMix.