A Contrastive Variational AutoEncoder for NSCLC Survival Prediction with Missing Modalities

2026-02-19

AI Summary

This study addresses the challenge of robust survival prediction in non-small cell lung cancer (NSCLC), where severe missingness in clinical multimodal data often undermines existing methods. To this end, the authors propose the Multimodal Contrastive Variational Autoencoder (MCVAE), which employs modality-specific variational encoders to model uncertainty, integrates a gated fusion mechanism to dynamically balance contributions from available modalities, and incorporates cross-modal contrastive learning with multitask loss to enhance representation alignment and robustness. During training, random modality masking is applied to improve generalization under arbitrary missing patterns. Evaluated on TCGA-LUAD and TCGA-LUSC datasets, MCVAE significantly outperforms current approaches, maintaining strong disease-specific survival prediction performance even under high missing rates, while also revealing that multimodal fusion does not universally confer performance gains.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Predicting survival outcomes for non-small cell lung cancer (NSCLC) patients is challenging due to the different individual prognostic features. This task can benefit from the integration of whole-slide images, bulk transcriptomics, and DNA methylation, which offer complementary views of the patient's condition at diagnosis. However, real-world clinical datasets are often incomplete, with entire modalities missing for a significant fraction of patients. State-of-the-art models rely on available data to create patient-level representations or use generative models to infer missing modalities, but they lack robustness in cases of severe missingness. We propose a Multimodal Contrastive Variational AutoEncoder (MCVAE) to address this issue: modality-specific variational encoders capture the uncertainty in each data source, and a fusion bottleneck with learned gating mechanisms is introduced to normalize the contributions from present modalities. We propose a multi-task objective that combines survival loss and reconstruction loss to regularize patient representations, along with a cross-modal contrastive loss that enforces cross-modal alignment in the latent space. During training, we apply stochastic modality masking to improve the robustness to arbitrary missingness patterns. Extensive evaluations on the TCGA-LUAD (n=475) and TCGA-LUSC (n=446) datasets demonstrate the efficacy of our approach in predicting disease-specific survival (DSS) and its robustness to severe missingness scenarios compared to two state-of-the-art models. Finally, we bring some clarifications on multimodal integration by testing our model on all subsets of modalities, finding that integration is not always beneficial to the task.

Problem

• NSCLC survival prediction
• missing modalities
• multimodal integration
• clinical data incompleteness
• robustness

Innovation

• Multimodal Contrastive Variational AutoEncoder
• missing modalities
• cross-modal contrastive learning
• stochastic modality masking
• survival prediction