UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Tang, Jialu, Xia, Tong, Lu, Yuan, Saeed, Aaqib
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914279855226880
author Tang, Jialu
Xia, Tong
Lu, Yuan
Saeed, Aaqib
author_facet Tang, Jialu
Xia, Tong
Lu, Yuan
Saeed, Aaqib
contents Accurate clinical prognosis requires synthesizing structured Electronic Health Records (EHRs) with real-time physiological signals like the Electrocardiogram (ECG). Large Language Models (LLMs) offer a powerful reasoning engine for this task but struggle to natively process these heterogeneous, non-textual data types. To address this, we propose UniPACT (Unified Prognostic Question Answering for Clinical Time-series), a unified framework for prognostic question answering that bridges this modality gap. UniPACT's core contribution is a structured prompting mechanism that converts numerical EHR data into semantically rich text. This textualized patient context is then fused with representations learned directly from raw ECG waveforms, enabling an LLM to reason over both modalities holistically. We evaluate UniPACT on the comprehensive MDS-ED benchmark, it achieves a state-of-the-art mean AUROC of 89.37% across a diverse set of prognostic tasks including diagnosis, deterioration, ICU admission, and mortality, outperforming specialized baselines. Further analysis demonstrates that our multimodal, multi-task approach is critical for performance and provides robustness in missing data scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2601_17916
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR
Tang, Jialu
Xia, Tong
Lu, Yuan
Saeed, Aaqib
Machine Learning
Accurate clinical prognosis requires synthesizing structured Electronic Health Records (EHRs) with real-time physiological signals like the Electrocardiogram (ECG). Large Language Models (LLMs) offer a powerful reasoning engine for this task but struggle to natively process these heterogeneous, non-textual data types. To address this, we propose UniPACT (Unified Prognostic Question Answering for Clinical Time-series), a unified framework for prognostic question answering that bridges this modality gap. UniPACT's core contribution is a structured prompting mechanism that converts numerical EHR data into semantically rich text. This textualized patient context is then fused with representations learned directly from raw ECG waveforms, enabling an LLM to reason over both modalities holistically. We evaluate UniPACT on the comprehensive MDS-ED benchmark, it achieves a state-of-the-art mean AUROC of 89.37% across a diverse set of prognostic tasks including diagnosis, deterioration, ICU admission, and mortality, outperforming specialized baselines. Further analysis demonstrates that our multimodal, multi-task approach is critical for performance and provides robustness in missing data scenarios.
title UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR
topic Machine Learning
url https://arxiv.org/abs/2601.17916