Saved in:
Bibliographic Details
Main Authors: Ning, Catherine, Ma, Yu, Wang, Cindy Beini, McMahon, Sean, Radojevic, Joseph, Zweibel, Steven, Bertsimas, Dimitris
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.25942
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908999390068736
author Ning, Catherine
Ma, Yu
Wang, Cindy Beini
McMahon, Sean
Radojevic, Joseph
Zweibel, Steven
Bertsimas, Dimitris
author_facet Ning, Catherine
Ma, Yu
Wang, Cindy Beini
McMahon, Sean
Radojevic, Joseph
Zweibel, Steven
Bertsimas, Dimitris
contents Left ventricular ejection fraction (LVEF) assessment depends on echocardiography, limiting access in primary care and resource-constrained settings. We developed a multimodal machine-learning framework that combines engineered 12-lead ECG timeseries features with structured EHR variables to classify LVEF into four clinically used strata: normal (>50%), mildly reduced (40-50%), moderately reduced (30-40%), and severely reduced (<30%). To support model explainability, we identified the most influential ECG and EHR features via SHAP attributions. Using retrospective data from Hartford HealthCare, we trained XGBoost models on 36,784 ECG-echocardiogram pairs from 30,952 outpatients and evaluated temporal generalizability on 19,966 ECGs from a subsequent period. The multimodal model achieved one-vs-rest AUROCs of 0.95 (severe), 0.92 (moderate), 0.82 (mild), and 0.91 (normal), outperforming ECG-only and EHR-only baselines, and maintained performance under temporal validation. This work supports ECG-based, multimodal LVEF stratification as a practical screening and triage aid to prioritize confirmatory imaging where resources are limited.
format Preprint
id arxiv_https___arxiv_org_abs_2604_25942
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Multimodal and Explainable Machine Learning Approach to Diagnosing Multi-Class Ejection Fraction from Electrocardiograms
Ning, Catherine
Ma, Yu
Wang, Cindy Beini
McMahon, Sean
Radojevic, Joseph
Zweibel, Steven
Bertsimas, Dimitris
Machine Learning
Left ventricular ejection fraction (LVEF) assessment depends on echocardiography, limiting access in primary care and resource-constrained settings. We developed a multimodal machine-learning framework that combines engineered 12-lead ECG timeseries features with structured EHR variables to classify LVEF into four clinically used strata: normal (>50%), mildly reduced (40-50%), moderately reduced (30-40%), and severely reduced (<30%). To support model explainability, we identified the most influential ECG and EHR features via SHAP attributions. Using retrospective data from Hartford HealthCare, we trained XGBoost models on 36,784 ECG-echocardiogram pairs from 30,952 outpatients and evaluated temporal generalizability on 19,966 ECGs from a subsequent period. The multimodal model achieved one-vs-rest AUROCs of 0.95 (severe), 0.92 (moderate), 0.82 (mild), and 0.91 (normal), outperforming ECG-only and EHR-only baselines, and maintained performance under temporal validation. This work supports ECG-based, multimodal LVEF stratification as a practical screening and triage aid to prioritize confirmatory imaging where resources are limited.
title A Multimodal and Explainable Machine Learning Approach to Diagnosing Multi-Class Ejection Fraction from Electrocardiograms
topic Machine Learning
url https://arxiv.org/abs/2604.25942