Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Aug 27;84(9):815-828.
doi: 10.1016/j.jacc.2024.05.062.

Deep Learning-Based Electrocardiogram Analysis Predicts Biventricular Dysfunction and Dilation in Congenital Heart Disease

Joshua Mayourian  1 , Addison Gearhart  1 , William G La Cava  2 , Akhil Vaid  3 , Girish N Nadkarni  3 , John K Triedman  1 , Andrew J Powell  1 , Rachel M Wald  4 , Anne Marie Valente  1 , Tal Geva  1 , Son Q Duong  5 , Sunil J Ghelani  6
Affiliations

Deep Learning-Based Electrocardiogram Analysis Predicts Biventricular Dysfunction and Dilation in Congenital Heart Disease

Joshua Mayourian et al. J Am Coll Cardiol. .

Abstract

Background: Artificial intelligence-enhanced electrocardiogram (AI-ECG) analysis shows promise to detect biventricular pathophysiology. However, AI-ECG analysis remains underexplored in congenital heart disease (CHD).

Objectives: The purpose of this study was to develop and externally validate an AI-ECG model to predict cardiovascular magnetic resonance (CMR)-defined biventricular dysfunction/dilation in patients with CHD.

Methods: We trained (80%) and tested (20%) a convolutional neural network on paired ECG-CMRs (≤30 days apart) from patients with and without CHD to detect left ventricular (LV) dysfunction (ejection fraction ≤40%), RV dysfunction (ejection fraction ≤35%), and LV and RV dilation (end-diastolic volume z-score ≥4). Performance was assessed during internal testing and external validation on an outside health care system using area under receiver-operating curve (AUROC) and area under precision recall curve.

Results: The internal and external cohorts comprised 8,584 ECG-CMR pairs (n = 4,941; median CMR age 20.7 years) and 909 ECG-CMR pairs (n = 746; median CMR age 25.4 years), respectively. Model performance was similar for internal testing (AUROC: LV dysfunction 0.87; LV dilation 0.86; RV dysfunction 0.88; RV dilation 0.81) and external validation (AUROC: LV dysfunction 0.89; LV dilation 0.83; RV dysfunction 0.82; RV dilation 0.80). Model performance was lowest in functionally single ventricle patients. Tetralogy of Fallot patients predicted to be at high risk of ventricular dysfunction had lower survival (P < 0.001). Model explainability via saliency mapping revealed that lateral precordial leads influence all outcome predictions, with high-risk features including QRS widening and T-wave inversions for RV dysfunction/dilation.

Conclusions: AI-ECG shows promise to predict biventricular dysfunction/dilation, which may help inform CMR timing in CHD.

Keywords: artificial intelligence; cardiovascular magnetic resonance; congenital heart disease; tetralogy of Fallot; ventricular function.

PubMed Disclaimer

Conflict of interest statement

Funding Support and Author Disclosures This work was supported in part by the Thrasher Research Fund Early Career Award (Dr Mayourian), National Institutes of Health T32 Research Methods in Pediatric Cardiovascular Disease Award Number: 5T32HL007572-38 (Dr Gearhart), and National Institutes of Health grant R00-LM012926 (Prof La Cava). Dr Nadkarni has consultancy agreements with AstraZeneca, BioVie, GLG Consulting, Pensieve Health, Reata, Renalytix, Siemens Healthineers, and Variant Bio; has received research funding from Goldfinch Bio and Renalytix; has received honoraria from AstraZeneca, BioVie, Lexicon, Daiichi-Sankyo, Menarini Health, and Reata; has patents or royalties with Renalytix; owns equity and stock options in Pensieve Health and Renalytix as a scientific cofounder; owns equity in Verici Dx; has received financial compensation as a scientific board member and advisor to Renalytix; serves on the advisory board of Neurona Health; and serves in an advisory or leadership role for Pensieve Health and Renalytix; none of these relationships played a role in the design or conduct of this study. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.

Figures

Figure 1:
Figure 1:. STROBE diagram showing initial patient selection and the final cohort.
Filtering at each data processing stage is shown. Patient partitioning for training (80%) and testing (20%) is shown, with external validation at Mount Sinai. Abbreviations: cardiovascular magnetic resonance (CMR); left ventricle (LV); right ventricle (RV); ejection fraction (EF); end-diastolic volume (EDV).
Figure 2:
Figure 2:. External Validation of Electrocardiogram-Based Deep Learning Model Performance.
Model performance evaluated with receiver operating and precision-recall curves on internal (left) and external (right) cohorts for: (A) left ventricular ejection fraction (LVEF) ≤40%; (B) Left ventricular end-diastolic volume (LVEDV) z-score ≥4; (C) right ventricular ejection fraction (RVEF) ≤35%; and (D) Right ventricular end-diastolic volume (RVEDV) z-score ≥4. AUROC and AUPRC metric values for each model and outcome are inset. Dotted line represents chance. 95% confidence intervals are shown using bootstrapping. Abbreviations: positive predictive value (PPV).
Figure 3:
Figure 3:. Subgroup Model Performance.
Forest plot showing area under the receiver operating curve (AUROC; red) and precision recall curve (AUPRC; black) performance when stratifying by age and CMR subgroups for the following outcomes: left ventricular ejection fraction (LVEF) ≤40%, left ventricular end-diastolic volume (LVEDV) z-score ≥4, right ventricular ejection fraction (RVEF) ≤35%, and right ventricular end-diastolic volume (RVEDV) z-score ≥4. Dotted lines represent overall cohort performance, with metrics inset above each forest plot. 95% confidence intervals are shown using bootstrapping. Abbreviations: left ventricle (LV); right ventricle (RV); single ventricle (SV).
Figure 4:
Figure 4:. Survival Analysis in Tetralogy of Fallot Based on AI-ECG Predictions.
Kaplan-Meier survival curves on the Tetralogy of Fallot (ToF) subgroup for patients deemed high-risk (orange) versus low-risk (blue) based on AI-ECG predictions of left ventricular ejection fraction (LVEF) and right ventricular ejection fraction (RVEF) outcomes, using the Youden Index as a cutoff. Statistics based on log-rank testing. Number at risk inset below plots. 95% confidence intervals are shown in shaded regions.
Figure 5:
Figure 5:. Explainability of AI-ECG Outcome Predictions.
Averaged median electrocardiogram waveform from the 25 highest (red) and 25 lowest (green) predictions for left ventricular ejection fraction (LVEF), left ventricular end-diastolic volume (LVEDV), right ventricular ejection fraction (RVEF), and right ventricular end-diastolic volume (RVEDV). Saliency mapping shows more (dark blue) and less (light blue) contributory regions of the ECG in the background of each lead waveform.
Central Illustration:
Central Illustration:. AI-ECG to Predict Biventricular Size and Function.
An artificial intelligence-enhanced ECG (AI-ECG) algorithm trained on ECG- cardiovascular magnetic resonance (CMR) pairs at Boston Children’s Hospital was predictive of right (RV) and left (LV) ventricular dysfunction and dilation in a congenital heart disease cohort, with external validation and model explainability. Abbreviations: ejection fraction (EF); end-diastolic volume (EDV).
See this image and copyright information in PMC

References

    1. Candelino M, Tagi VM, Chiarelli F. Cardiovascular risk in children: a burden for future generations. Ital J Pediatr 2022;48:57. - PMC - PubMed
    1. Meyer SL, St Clair N, Powell AJ, Geva T, Rathod RH. Integrated Clinical and Magnetic Resonance Imaging Assessments Late After Fontan Operation. J Am Coll Cardiol 2021;77:2480–2489. - PubMed
    1. Knauth AL, Gauvreau K, Powell AJ et al. Ventricular size and function assessed by cardiac MRI predict major adverse clinical outcomes late after tetralogy of Fallot repair. Heart 2008;94:211–6. - PubMed
    1. Ghai A, Silversides C, Harris L, Webb GD, Siu SC, Therrien J. Left ventricular dysfunction is a risk factor for sudden cardiac death in adults late after repair of tetralogy of Fallot. J Am Coll Cardiol 2002;40:1675–80. - PubMed
    1. Valente AM, Gauvreau K, Assenza GE et al. Contemporary predictors of death and sustained ventricular tachycardia in patients with repaired tetralogy of Fallot enrolled in the INDICATOR cohort. Heart 2014;100:247–53. - PMC - PubMed

MeSH terms

LinkOut - more resources