M4: Multi-Proxy Multi-Gate Mixture of Experts Network for Multiple Instance Learning in Histopathology Image Analysis

TL;DR

The paper introduces M4, a multi-task learning model using a multi-gate mixture of experts for simultaneous genetic mutation prediction from histopathology images, improving efficiency and inter-task modeling.

Contribution

It presents a novel multi-proxy multi-gate mixture of experts architecture tailored for multi-task learning in histopathology image analysis, addressing limitations of single-task methods.

Findings

Significant performance improvements over state-of-the-art methods.

Effective multi-task learning for multiple genetic mutations.

Enhanced modeling of complex pathological image features.

Abstract

Multiple instance learning (MIL) has been successfully applied for whole slide images (WSIs) analysis in computational pathology, enabling a wide range of prediction tasks from tumor subtyping to inferring genetic mutations and multi-omics biomarkers. However, existing MIL methods predominantly focus on single-task learning, resulting in not only overall low efficiency but also the overlook of inter-task relatedness. To address these issues, we proposed an adapted architecture of Multi-gate Mixture-of-experts with Multi-proxy for Multiple instance learning (M4), and applied this framework for simultaneous prediction of multiple genetic mutations from WSIs. The proposed M4 model has two main innovations: (1) utilizing a mixture of experts with multiple gating strategies for multi-genetic mutation prediction on a single pathological slide; (2) constructing multi-proxy expert network and gate network for comprehensive and effective modeling of pathological image information. Our model achieved significant improvements across five tested TCGA datasets in comparison to current state-of-the-art single-task methods. The code is available at:https://github.com/Bigyehahaha/M4.