# Fusion is not one-size-fits-all: Cross-Modal Representation Alignment for Time-to-Event Modeling

arXiv: https://arxiv.org/abs/2606.15038 · HTML: https://mlsanity.com/paper/2606.15038/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 23 references that are in the map.
- First posted on arXiv 13 Jun 2026.

## Builds on (most cited first, 12 of 12)

- [Foundations and Trends in Multimodal Machine Learning: Principles, Challenges, and Open Questions](https://mlsanity.com/paper/2209.03430/) (2022)
- [MOTOR: A Time-To-Event Foundation Model For Structured Medical Records](https://mlsanity.com/paper/2301.03150/) (2023)
- [Deep Learning for Survival Analysis: A Review](https://mlsanity.com/paper/2305.14961/) (2023)
- [MedImageInsight: An Open-Source Embedding Model for General Domain Medical Imaging](https://mlsanity.com/paper/2410.06542/) (2024)
- [Enhancing Representation in Radiography-Reports Foundation Model: A Granular Alignment Algorithm Using Masked Contrastive Learning](https://mlsanity.com/paper/2309.05904/) (2023)
- [Max-Margin Contrastive Learning](https://mlsanity.com/paper/2112.11450/) (2021)
- [INSPECT: A Multimodal Dataset for Pulmonary Embolism Diagnosis and Prognosis](https://mlsanity.com/paper/2311.10798/) (2023)
- [A scoping review on multimodal deep learning in biomedical images and texts](https://mlsanity.com/paper/2307.07362/) (2023)
- [Learning the Unlearned: Mitigating Feature Suppression in Contrastive Learning](https://mlsanity.com/paper/2402.11816/) (2024)
- [MedInsight: A Multi-Source Context Augmentation Framework for Generating Patient-Centric Medical Responses using Large Language Models](https://mlsanity.com/paper/2403.08607/) (2024)
- [Time-to-Event Pretraining for 3D Medical Imaging](https://mlsanity.com/paper/2411.09361/) (2024)
- [TorchSurv: A Lightweight Package for Deep Survival Analysis](https://mlsanity.com/paper/2404.10761/) (2024)

## Abstract (arXiv)

Accurate time-to-event (TTE) prediction from multimodal clinical data remains challenging due to modality imbalance and distribution shift. We introduce a foundation model-driven framework for cross-modal representation alignment between CT imaging and longitudinal EHR data, designed to generalize across tasks and institutions. CT and EHR modalities are encoded independently using domain-specific foundation models and aligned in a shared latent space through four principled fusion strategies: late fusion, contrastive alignment, cross-attention, and co-attention. We evaluate two clinically distinct TTE tasks: pulmonary embolism (PE) mortality and cardiovascular disease (CVD) outcomes, on large-scale multi-institutional cohorts (PE: N=3,099 train; 1,098 internal; 435 external; CVD: N=2,951 train; 837 internal; 682 external). Fusion consistently improves concordance index by 1.5-5.4% over unimodal baselines when modalities contribute comparably. Overall, contrastive multimodal fusion, particularly with CLMBR representations, provided the most consistent and statistically robust improvements, especially for PE mortality prediction. For MACE, cross-attention (one-hot) achieved the highest internal performance and image-guided co-attention achieved the best external performance. We therefore introduce a generalizable foundation model-based cross-modal alignment framework and provide the first systematic analysis of fusion behavior under modality imbalance in TTE prediction. Our results establish task-aware multimodal alignment as a necessary design principle for robust generalization and scalable clinical deployment.