Autoregressive EHR Foundation Models with Multimodal Inputs
Abstract
Autoregressive foundation models trained on tokenized electronic health records (EHRs) can support zero-shot clinical prediction, yet most operate on structured event codes alone, and do not incorporate multiple modalities in a principled way. We present a framework for conditioning such models on auxiliary clinical modalities, including ECG waveforms, chest X-ray images, and clinical notes, using modality-specific latent compression and gated cross-attention with temporal alignment. We investigate two key design choices: (1) how to compress long per-modality sequences (e.g., ECG time series) before they enter the multi-modal cross-attention. This feature may be essential to reduce compute overheads and may be beneficial for generalization; (2) how the choice of pretrained encoder for each modality impacts downstream performance. Through controlled ablations on MIMIC-IV, we show that the best latent-compression configurations outperforms both uncompressed cross-attention and mean pooling. Encoder choice has a clear within-modality effect, with stronger pretrained encoders consistently outperforming weaker alternatives. We further show that merely adding auxiliary modalities does not guarantee improvement on ICU mortality prediction over an EHR-only baseline. This implies that careful design of the fusion architecture and an appropriate evaluation in the clinical context are required.
Cite
@article{arxiv.2607.22264,
title = {Autoregressive EHR Foundation Models with Multimodal Inputs},
author = {Yuxuan Liu and Joshua Placidi and Jinpei Han and Alfred John Balston and Marek Rei and A. Aldo Faisal},
journal= {arXiv preprint arXiv:2607.22264},
year = {2026}
}
Comments
5 pages excluding references and supplements, 2 figures and 2 tables, Proceedings of the Workshop on Structured Data for Health at the 43rd International Conference on Machine Learning, Seoul, South Korea