Self-Supervision Enhances Instance-based Multiple Instance Learning Methods in Digital Pathology: A Benchmark Study

TL;DR

This study benchmarks various MIL methods in digital pathology, demonstrating that self-supervised learning enhances simple instance-based MIL approaches to outperform complex embedding-based methods in accuracy and interpretability.

Contribution

It provides a comprehensive experimental comparison of MIL strategies, introduces new instance-based MIL methods, and highlights the effectiveness of SSL in improving simple MIL models for pathology.

Findings

Instance-based MIL with SSL matches or exceeds embedding-based MIL performance.

Simple, interpretable MIL models can achieve state-of-the-art results.

SSL significantly boosts the quality of features for MIL in pathology.

Abstract

Multiple Instance Learning (MIL) has emerged as the best solution for Whole Slide Image (WSI) classification. It consists of dividing each slide into patches, which are treated as a bag of instances labeled with a global label. MIL includes two main approaches: instance-based and embedding-based. In the former, each patch is classified independently, and then the patch scores are aggregated to predict the bag label. In the latter, bag classification is performed after aggregating patch embeddings. Even if instance-based methods are naturally more interpretable, embedding-based MILs have usually been preferred in the past due to their robustness to poor feature extractors. However, recently, the quality of feature embeddings has drastically increased using self-supervised learning (SSL). Nevertheless, many authors continue to endorse the superiority of embedding-based MIL. To investigate this further, we conduct 710 experiments across 4 datasets, comparing 10 MIL strategies, 6 self-supervised methods with 4 backbones, 4 foundation models, and various pathology-adapted techniques. Furthermore, we introduce 4 instance-based MIL methods never used before in the pathology domain. Through these extensive experiments, we show that with a good SSL feature extractor, simple instance-based MILs, with very few parameters, obtain similar or better performance than complex, state-of-the-art (SOTA) embedding-based MIL methods, setting new SOTA results on the BRACS and Camelyon16 datasets. Since simple instance-based MIL methods are naturally more interpretable and explainable to clinicians, our results suggest that more effort should be put into well-adapted SSL methods for WSI rather than into complex embedding-based MIL methods.