Leveraging ECG Images for Predicting Ejection Fraction using Machine Learning Algorithms.

IF 1.4 Q3 CARDIAC & CARDIOVASCULAR SYSTEMS
Abhyuday Kumara Swamy, Vivek Rajagopal, Deepak Krishnan, Paramita Auddya Ghorai, Anagha Choukhande, Santhosh Rathnam Palani, Deepak Padmanabhan, Emmanuel Rupert, Devi Prasad Shetty, Pradeep Narayan
{"title":"Leveraging ECG Images for Predicting Ejection Fraction using Machine Learning Algorithms.","authors":"Abhyuday Kumara Swamy, Vivek Rajagopal, Deepak Krishnan, Paramita Auddya Ghorai, Anagha Choukhande, Santhosh Rathnam Palani, Deepak Padmanabhan, Emmanuel Rupert, Devi Prasad Shetty, Pradeep Narayan","doi":"10.1016/j.ihj.2025.03.009","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>The capability to accurately predict the ejection fraction (EF) from an electrocardiogram (ECG) holds significant and valuable clinical implications. Various algorithms based on ECG images are currently being evaluated, with most methods requiring raw signal data from ECG devices. In this study, our objective was to train and validate a neural network on a readily available ECG trace image graph to determine the presence or absence of left ventricular dysfunction (LVD).</p><p><strong>Methods: </strong>12-lead ECG trace images paired with their echocardiogram reports performed on the same day were selected. A DenseNet121 model, using ECG images as input, was trained to identify EF <50%. and then externally validated.</p><p><strong>Results: </strong>1,19,281 ECG-echocardiogram pairs were used for model development. The model demonstrated comparable performance in both the internal test data and external validation data. The area under receiver operating characteristic and precision-recall curves were 0.92 and 0.78, respectively, for the internal test data and 0.88 and 0.74, respectively, for the external validation data. The model accurately identified more than 85% of cases with EF <50% in both datasets.</p><p><strong>Conclusions: </strong>Actual images of ECGs with simple pre-processing and model architecture can be used as a reliable tool to screen for LVD. The use of images expands the reach of these algorithms to geographies with resource and technological limitations.</p>","PeriodicalId":13384,"journal":{"name":"Indian heart journal","volume":" ","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Indian heart journal","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1016/j.ihj.2025.03.009","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"CARDIAC & CARDIOVASCULAR SYSTEMS","Score":null,"Total":0}
引用次数: 0

Abstract

Introduction: The capability to accurately predict the ejection fraction (EF) from an electrocardiogram (ECG) holds significant and valuable clinical implications. Various algorithms based on ECG images are currently being evaluated, with most methods requiring raw signal data from ECG devices. In this study, our objective was to train and validate a neural network on a readily available ECG trace image graph to determine the presence or absence of left ventricular dysfunction (LVD).

Methods: 12-lead ECG trace images paired with their echocardiogram reports performed on the same day were selected. A DenseNet121 model, using ECG images as input, was trained to identify EF <50%. and then externally validated.

Results: 1,19,281 ECG-echocardiogram pairs were used for model development. The model demonstrated comparable performance in both the internal test data and external validation data. The area under receiver operating characteristic and precision-recall curves were 0.92 and 0.78, respectively, for the internal test data and 0.88 and 0.74, respectively, for the external validation data. The model accurately identified more than 85% of cases with EF <50% in both datasets.

Conclusions: Actual images of ECGs with simple pre-processing and model architecture can be used as a reliable tool to screen for LVD. The use of images expands the reach of these algorithms to geographies with resource and technological limitations.

求助全文
约1分钟内获得全文 求助全文
来源期刊
Indian heart journal
Indian heart journal CARDIAC & CARDIOVASCULAR SYSTEMS-
CiteScore
2.60
自引率
6.70%
发文量
82
审稿时长
52 days
期刊介绍: Indian Heart Journal (IHJ) is the official peer-reviewed open access journal of Cardiological Society of India and accepts articles for publication from across the globe. The journal aims to promote high quality research and serve as a platform for dissemination of scientific information in cardiology with particular focus on South Asia. The journal aims to publish cutting edge research in the field of clinical as well as non-clinical cardiology - including cardiovascular medicine and surgery. Some of the topics covered are Heart Failure, Coronary Artery Disease, Hypertension, Interventional Cardiology, Cardiac Surgery, Valvular Heart Disease, Pulmonary Hypertension and Infective Endocarditis. IHJ open access invites original research articles, research briefs, perspective, case reports, case vignette, cardiovascular images, cardiovascular graphics, research letters, correspondence, reader forum, and interesting photographs, for publication. IHJ open access also publishes theme-based special issues and abstracts of papers presented at the annual conference of the Cardiological Society of India.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信