CAMIL: Context-Aware Multiple Instance Learning for Cancer Detection and Subtyping in Whole Slide Images

TL;DR

CAMIL introduces a context-aware MIL model that leverages neighboring tile dependencies and contextual priors to improve accuracy and interpretability in cancer detection and subtyping from whole slide images.

Contribution

The paper presents CAMIL, a novel architecture that incorporates neighbor-constrained attention and contextual priors into MIL for better cancer diagnosis from WSIs.

Findings

Achieved high AUCs of 97.5%, 95.9%, and 88.1% on different cancer detection tasks.

Outperformed existing state-of-the-art methods in accuracy.

Enhanced interpretability by identifying diagnostically relevant regions.

Abstract

The visual examination of tissue biopsy sections is fundamental for cancer diagnosis, with pathologists analyzing sections at multiple magnifications to discern tumor cells and their subtypes. However, existing attention-based multiple instance learning (MIL) models used for analyzing Whole Slide Images (WSIs) in cancer diagnostics often overlook the contextual information of tumor and neighboring tiles, leading to misclassifications. To address this, we propose the Context-Aware Multiple Instance Learning (CAMIL) architecture. CAMIL incorporates neighbor-constrained attention to consider dependencies among tiles within a WSI and integrates contextual constraints as prior knowledge into the MIL model. We evaluated CAMIL on subtyping non-small cell lung cancer (TCGA-NSCLC) and detecting lymph node (CAMELYON16 and CAMELYON17) metastasis, achieving test AUCs of 97.5\%, 95.9\%, and 88.1\%, respectively, outperforming other state-of-the-art methods. Additionally, CAMIL enhances model interpretability by identifying regions of high diagnostic value.