@inproceedings{cai-etal-2025-supreme,
title = "{S}u{P}re{ME}: A Supervised Pre-training Framework for Multimodal {ECG} Representation Learning",
author = "Cai, Mingsheng and
Jiang, Jiuming and
Huang, Wenhao and
Liu, Che and
Arcucci, Rossella",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.633/",
doi = "10.18653/v1/2025.findings-emnlp.633",
pages = "11817--11844",
ISBN = "979-8-89176-335-7",
abstract = "Cardiovascular diseases are a leading cause of death and disability worldwide. Electrocardiogram (ECG) is critical for diagnosing and monitoring cardiac health, but obtaining large-scale annotated ECG datasets is labor-intensive and time-consuming. Recent ECG Self-Supervised Learning (eSSL) methods mitigate this by learning features without extensive labels but fail to capture fine-grained clinical semantics and require extensive task-specific fine-tuning. To address these challenges, we propose $\textbf{SuPreME}$, a $\textbf{Su}$pervised $\textbf{Pre}$-training framework for $\textbf{M}$ultimodal $\textbf{E}$CG representation learning. SuPreME is pre-trained using structured diagnostic labels derived from ECG report entities through a one-time offline extraction with Large Language Models (LLMs), which help denoise, standardize cardiac concepts, and improve clinical representation learning. By fusing ECG signals with textual cardiac queries instead of fixed labels, SuPreME enables zero-shot classification of unseen conditions without further fine-tuning. We evaluate SuPreME on six downstream datasets covering 106 cardiac conditions, achieving superior zero-shot AUC performance of 77.20{\%}, surpassing state-of-the-art eSSLs by 4.98{\%}. Results demonstrate SuPreME{'}s effectiveness in leveraging structured, clinically relevant knowledge for high-quality ECG representations."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cai-etal-2025-supreme">
<titleInfo>
<title>SuPreME: A Supervised Pre-training Framework for Multimodal ECG Representation Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mingsheng</namePart>
<namePart type="family">Cai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiuming</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenhao</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Che</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rossella</namePart>
<namePart type="family">Arcucci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>Cardiovascular diseases are a leading cause of death and disability worldwide. Electrocardiogram (ECG) is critical for diagnosing and monitoring cardiac health, but obtaining large-scale annotated ECG datasets is labor-intensive and time-consuming. Recent ECG Self-Supervised Learning (eSSL) methods mitigate this by learning features without extensive labels but fail to capture fine-grained clinical semantics and require extensive task-specific fine-tuning. To address these challenges, we propose SuPreME, a Supervised Pre-training framework for Multimodal ECG representation learning. SuPreME is pre-trained using structured diagnostic labels derived from ECG report entities through a one-time offline extraction with Large Language Models (LLMs), which help denoise, standardize cardiac concepts, and improve clinical representation learning. By fusing ECG signals with textual cardiac queries instead of fixed labels, SuPreME enables zero-shot classification of unseen conditions without further fine-tuning. We evaluate SuPreME on six downstream datasets covering 106 cardiac conditions, achieving superior zero-shot AUC performance of 77.20%, surpassing state-of-the-art eSSLs by 4.98%. Results demonstrate SuPreME’s effectiveness in leveraging structured, clinically relevant knowledge for high-quality ECG representations.</abstract>
<identifier type="citekey">cai-etal-2025-supreme</identifier>
<identifier type="doi">10.18653/v1/2025.findings-emnlp.633</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.633/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>11817</start>
<end>11844</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SuPreME: A Supervised Pre-training Framework for Multimodal ECG Representation Learning
%A Cai, Mingsheng
%A Jiang, Jiuming
%A Huang, Wenhao
%A Liu, Che
%A Arcucci, Rossella
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F cai-etal-2025-supreme
%X Cardiovascular diseases are a leading cause of death and disability worldwide. Electrocardiogram (ECG) is critical for diagnosing and monitoring cardiac health, but obtaining large-scale annotated ECG datasets is labor-intensive and time-consuming. Recent ECG Self-Supervised Learning (eSSL) methods mitigate this by learning features without extensive labels but fail to capture fine-grained clinical semantics and require extensive task-specific fine-tuning. To address these challenges, we propose SuPreME, a Supervised Pre-training framework for Multimodal ECG representation learning. SuPreME is pre-trained using structured diagnostic labels derived from ECG report entities through a one-time offline extraction with Large Language Models (LLMs), which help denoise, standardize cardiac concepts, and improve clinical representation learning. By fusing ECG signals with textual cardiac queries instead of fixed labels, SuPreME enables zero-shot classification of unseen conditions without further fine-tuning. We evaluate SuPreME on six downstream datasets covering 106 cardiac conditions, achieving superior zero-shot AUC performance of 77.20%, surpassing state-of-the-art eSSLs by 4.98%. Results demonstrate SuPreME’s effectiveness in leveraging structured, clinically relevant knowledge for high-quality ECG representations.
%R 10.18653/v1/2025.findings-emnlp.633
%U https://aclanthology.org/2025.findings-emnlp.633/
%U https://doi.org/10.18653/v1/2025.findings-emnlp.633
%P 11817-11844
Markdown (Informal)
[SuPreME: A Supervised Pre-training Framework for Multimodal ECG Representation Learning](https://aclanthology.org/2025.findings-emnlp.633/) (Cai et al., Findings 2025)
ACL