Snuffy: Efficient Whole Slide Image Classifier

TL;DR

Snuffy introduces a novel sparse transformer-based MIL pooling method for whole slide image classification that reduces training time, handles domain shifts, and achieves superior accuracy on pathology datasets.

Contribution

The paper presents Snuffy, a new MIL pooling architecture based on sparse transformers, tailored for pathology, with theoretical guarantees and improved performance with limited pre-training.

Findings

Achieves superior accuracy on CAMELYON16 and TCGA datasets.

Theoretically proven to be a universal approximator with minimal layers.

Enables effective continual few-shot pre-training.

Abstract

Whole Slide Image (WSI) classification with multiple instance learning (MIL) in digital pathology faces significant computational challenges. Current methods mostly rely on extensive self-supervised learning (SSL) for satisfactory performance, requiring long training periods and considerable computational resources. At the same time, no pre-training affects performance due to domain shifts from natural images to WSIs. We introduce Snuffy architecture, a novel MIL-pooling method based on sparse transformers that mitigates performance loss with limited pre-training and enables continual few-shot pre-training as a competitive option. Our sparsity pattern is tailored for pathology and is theoretically proven to be a universal approximator with the tightest probabilistic sharp bound on the number of layers for sparse transformers, to date. We demonstrate Snuffy's effectiveness on CAMELYON16 and TCGA Lung cancer datasets, achieving superior WSI and patch-level accuracies. The code is available on https://github.com/jafarinia/snuffy.