MoME: Mixture of Multimodal Experts for Cancer Survival Prediction

TL;DR

This paper introduces MoME, a novel multimodal fusion framework for cancer survival prediction that employs progressive encoding and dynamic expert selection to better model complex inter- and intra-modal interactions.

Contribution

It proposes a Biased Progressive Encoding paradigm combined with a Mixture of Multimodal Experts layer for improved multimodal fusion in survival analysis.

Findings

Outperforms existing methods on TCGA datasets

Effectively models complex modality interactions

Demonstrates flexible adaptation to different biomarkers

Abstract

Survival analysis, as a challenging task, requires integrating Whole Slide Images (WSIs) and genomic data for comprehensive decision-making. There are two main challenges in this task: significant heterogeneity and complex inter- and intra-modal interactions between the two modalities. Previous approaches utilize co-attention methods, which fuse features from both modalities only once after separate encoding. However, these approaches are insufficient for modeling the complex task due to the heterogeneous nature between the modalities. To address these issues, we propose a Biased Progressive Encoding (BPE) paradigm, performing encoding and fusion simultaneously. This paradigm uses one modality as a reference when encoding the other. It enables deep fusion of the modalities through multiple alternating iterations, progressively reducing the cross-modal disparities and facilitating complementary interactions. Besides modality heterogeneity, survival analysis involves various biomarkers from WSIs, genomics, and their combinations. The critical biomarkers may exist in different modalities under individual variations, necessitating flexible adaptation of the models to specific scenarios. Therefore, we further propose a Mixture of Multimodal Experts (MoME) layer to dynamically selects tailored experts in each stage of the BPE paradigm. Experts incorporate reference information from another modality to varying degrees, enabling a balanced or biased focus on different modalities during the encoding process. Extensive experimental results demonstrate the superior performance of our method on various datasets, including TCGA-BLCA, TCGA-UCEC and TCGA-LUAD. Codes are available at https://github.com/BearCleverProud/MoME.