Cargando…

Multi-category classification of left ventricle ejection fraction using a convolutional neural network

BACKGROUND: Screening for left ventricular (LV) systolic dysfunction (defined as ejection fraction ≤35%) based on data from a standard 12-lead electrocardiogram (ECG) has become well established when standard digital ECGs are available–8 independent leads sampled at least 250 hertz for 10 seconds. A...

Descripción completa

Detalles Bibliográficos
Autores principales: Carter, R, Hardway, H, Johnson, P, Douglass, E, Adedinsewo, D
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9779755/
http://dx.doi.org/10.1093/ehjdh/ztac076.2778
_version_ 1784856687399665664
author Carter, R
Hardway, H
Johnson, P
Douglass, E
Adedinsewo, D
author_facet Carter, R
Hardway, H
Johnson, P
Douglass, E
Adedinsewo, D
author_sort Carter, R
collection PubMed
description BACKGROUND: Screening for left ventricular (LV) systolic dysfunction (defined as ejection fraction ≤35%) based on data from a standard 12-lead electrocardiogram (ECG) has become well established when standard digital ECGs are available–8 independent leads sampled at least 250 hertz for 10 seconds. As the algorithm has been incorporated into various clinical scenarios and ancillary research projects, a limitation of the binary classification at 35% has become apparent. PURPOSE: The objective of this study was to develop and validate a deep learning-based algorithm that would classify LVEF into three categories based on only the digital ECG input. METHODS: After IRB approval, native digital resting ECGs acquired between 1/1/2010 and 12/31/2021 on patients seen in Mayo Clinic in Jacksonville were extracted from the institutional electronic ECG database management system (MUSE, GE Healthcare). These ECGs were matched with transthoracic echocardiograms obtained up to four days prior or 30 days after the ECGs acquisition. A convolutional neural network consisting of 8 layers of convolutions, batch normalization and pooling was trained using Keras and Tensorflow with hyper-parameter optimization for L1 and L2 regularization, learning rate adjustments, and class weights to predict three classes of LVEF: ≤35%, 36–51%, and ≥52% based on clinical relevance. The primary measure of overall performance was the detection of LVEF ≤35%; however, the triad of model predictions was also considered in translating the model output to human interpretable findings. RESULTS: A total of 30,153 patients (60,169 ECG pairings; mean age 63 years; 48% male) were randomly split at the patient level into training (24,172 patients), validation (2,973 patients) and testing (3,008 patients). The trained model provided robust discrimination in the withheld testing data – AUROC of 0.941 (95% CI: 0.931 to 0.950). Using the optimal model threshold based on Youden's index from the validation data (0.186), sensitivity and specificity were estimated to be 87.9% (95% CI: 83.8% to 91.2%) and 86.3% (95% CI: 85.4% to 87.2%) in the testing data. In instances where discordant predictions were observed, the posterior distribution of model probabilities provide additional insights into the possible underlying value of LVEF (Figure 1). CONCLUSIONS: The utilization of a multi-category deep learning classification model for the detection of reduced ejection fraction adds new dimensions to the use of AI technologies on digital ECGs. This work shows high discrimination can still be obtained when using three classes of LVEF. FUNDING ACKNOWLEDGEMENT: Type of funding sources: None.
format Online
Article
Text
id pubmed-9779755
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Oxford University Press
record_format MEDLINE/PubMed
spelling pubmed-97797552023-01-27 Multi-category classification of left ventricle ejection fraction using a convolutional neural network Carter, R Hardway, H Johnson, P Douglass, E Adedinsewo, D Eur Heart J Digit Health Abstracts BACKGROUND: Screening for left ventricular (LV) systolic dysfunction (defined as ejection fraction ≤35%) based on data from a standard 12-lead electrocardiogram (ECG) has become well established when standard digital ECGs are available–8 independent leads sampled at least 250 hertz for 10 seconds. As the algorithm has been incorporated into various clinical scenarios and ancillary research projects, a limitation of the binary classification at 35% has become apparent. PURPOSE: The objective of this study was to develop and validate a deep learning-based algorithm that would classify LVEF into three categories based on only the digital ECG input. METHODS: After IRB approval, native digital resting ECGs acquired between 1/1/2010 and 12/31/2021 on patients seen in Mayo Clinic in Jacksonville were extracted from the institutional electronic ECG database management system (MUSE, GE Healthcare). These ECGs were matched with transthoracic echocardiograms obtained up to four days prior or 30 days after the ECGs acquisition. A convolutional neural network consisting of 8 layers of convolutions, batch normalization and pooling was trained using Keras and Tensorflow with hyper-parameter optimization for L1 and L2 regularization, learning rate adjustments, and class weights to predict three classes of LVEF: ≤35%, 36–51%, and ≥52% based on clinical relevance. The primary measure of overall performance was the detection of LVEF ≤35%; however, the triad of model predictions was also considered in translating the model output to human interpretable findings. RESULTS: A total of 30,153 patients (60,169 ECG pairings; mean age 63 years; 48% male) were randomly split at the patient level into training (24,172 patients), validation (2,973 patients) and testing (3,008 patients). The trained model provided robust discrimination in the withheld testing data – AUROC of 0.941 (95% CI: 0.931 to 0.950). Using the optimal model threshold based on Youden's index from the validation data (0.186), sensitivity and specificity were estimated to be 87.9% (95% CI: 83.8% to 91.2%) and 86.3% (95% CI: 85.4% to 87.2%) in the testing data. In instances where discordant predictions were observed, the posterior distribution of model probabilities provide additional insights into the possible underlying value of LVEF (Figure 1). CONCLUSIONS: The utilization of a multi-category deep learning classification model for the detection of reduced ejection fraction adds new dimensions to the use of AI technologies on digital ECGs. This work shows high discrimination can still be obtained when using three classes of LVEF. FUNDING ACKNOWLEDGEMENT: Type of funding sources: None. Oxford University Press 2022-12-22 /pmc/articles/PMC9779755/ http://dx.doi.org/10.1093/ehjdh/ztac076.2778 Text en Reproduced from: European Heart Journal, Volume 43, Issue Supplement_2, October 2022, ehac544.2778, https://doi.org/10.1093/eurheartj/ehac544.2778 by permission of Oxford University Press on behalf of the European Society of Cardiology. The opinions expressed in the Journal item reproduced as this reprint are those of the authors and contributors, and do not necessarily reflect those of the European Society of Cardiology, the editors, the editorial board, Oxford University Press or the organization to which the authors are affiliated. The mention of trade names, commercial products or organizations, and the inclusion of advertisements in this reprint do not imply endorsement by the Journal, the editors, the editorial board, Oxford University Press or the organization to which the authors are affiliated. The editors and publishers have taken all reasonable precautions to verify drug names and doses, the results of experimental work and clinical findings published in the Journal. The ultimate responsibility for the use and dosage of drugs mentioned in this reprint and in interpretation of published material lies with the medical practitioner, and the editors and publisher cannot accept liability for damages arising from any error or omissions in the Journal or in this reprint. Please inform the editors of any errors. © The Author(s) 2022. https://creativecommons.org/licenses/by-nc/4.0/This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
spellingShingle Abstracts
Carter, R
Hardway, H
Johnson, P
Douglass, E
Adedinsewo, D
Multi-category classification of left ventricle ejection fraction using a convolutional neural network
title Multi-category classification of left ventricle ejection fraction using a convolutional neural network
title_full Multi-category classification of left ventricle ejection fraction using a convolutional neural network
title_fullStr Multi-category classification of left ventricle ejection fraction using a convolutional neural network
title_full_unstemmed Multi-category classification of left ventricle ejection fraction using a convolutional neural network
title_short Multi-category classification of left ventricle ejection fraction using a convolutional neural network
title_sort multi-category classification of left ventricle ejection fraction using a convolutional neural network
topic Abstracts
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9779755/
http://dx.doi.org/10.1093/ehjdh/ztac076.2778
work_keys_str_mv AT carterr multicategoryclassificationofleftventricleejectionfractionusingaconvolutionalneuralnetwork
AT hardwayh multicategoryclassificationofleftventricleejectionfractionusingaconvolutionalneuralnetwork
AT johnsonp multicategoryclassificationofleftventricleejectionfractionusingaconvolutionalneuralnetwork
AT douglasse multicategoryclassificationofleftventricleejectionfractionusingaconvolutionalneuralnetwork
AT adedinsewod multicategoryclassificationofleftventricleejectionfractionusingaconvolutionalneuralnetwork