SparseConvMIL: Sparse Convolutional Context-Aware Multiple Instance Learning for Whole Slide Image Classification

TL;DR

This paper introduces SparseConvMIL, a novel method that leverages spatial relationships between tiles in whole slide images using sparse maps and CNNs, achieving state-of-the-art classification results.

Contribution

It presents a new MIL approach that exploits spatial dependencies of tiles via sparse maps and CNNs, improving classification accuracy for whole slide images.

Findings

Achieved state-of-the-art performance on cancer subtype classification.

Improved representation learning of instances.

Enhanced classification by exploiting spatial context.

Abstract

Multiple instance learning (MIL) is the preferred approach for whole slide image classification. However, most MIL approaches do not exploit the interdependencies of tiles extracted from a whole slide image, which could provide valuable cues for classification. This paper presents a novel MIL approach that exploits the spatial relationship of tiles for classifying whole slide images. To do so, a sparse map is built from tiles embeddings, and is then classified by a sparse-input CNN. It obtained state-of-the-art performance over popular MIL approaches on the classification of cancer subtype involving 10000 whole slide images. Our results suggest that the proposed approach might (i) improve the representation learning of instances and (ii) exploit the context of instance embeddings to enhance the classification performance. The code of this work is open-source at {github censored for review}.