Yi-Ting Li, Kuang-Chien Chiang, Alexander Te-Wei Shieh, Tetsuji Kitano, Yosuke Nabeshima, Chung-Yen Lee, Kang Liu, Kuan-Yu Lai, Meng-Han Tsai, Li-Ting Ho, Wen-Jone Chen, Masaaki Takeuchi, Tzung-Dau Wang, Li-Tan Yang
{"title":"利用深度学习将血流动力学显著性主动脉瓣反流的心电图与超声心动图参数关联起来。","authors":"Yi-Ting Li, Kuang-Chien Chiang, Alexander Te-Wei Shieh, Tetsuji Kitano, Yosuke Nabeshima, Chung-Yen Lee, Kang Liu, Kuan-Yu Lai, Meng-Han Tsai, Li-Ting Ho, Wen-Jone Chen, Masaaki Takeuchi, Tzung-Dau Wang, Li-Tan Yang","doi":"10.6515/ACS.202411_40(6).20240918B","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Of all electrocardiogram (ECG) deep-learning (DL) models used to detect left-sided valvular heart diseases, aortic regurgitation (AR) has been the hardest to detect. Moreover, to what extent ECGs could detect AR-related left ventricular (LV) remodeling and dysfunction is unknown.</p><p><strong>Objectives: </strong>We aimed to evaluate the ability of DL-based ECG models to predict LV remodeling parameters associated with hemodynamically significant AR.</p><p><strong>Methods: </strong>From 573 consecutive patients, 1457 12-lead ECGs close to baseline transthoracic echocardiograms confirming ≥ moderate-severe AR and before aortic valve surgery were retrospectively collected. A ResNet-based model was used to predict LV ejection fraction (LVEF), LV end-diastolic dimension (LVEDD), LV end-systolic dimension index (LVESDi), LV mass index (LVMi), LV end-diastolic volume index (LVEDVi), LV end-systolic volume index (LVESVi), and bicuspid aortic valve (BAV) from the ECGs. Five-fold cross-validation was used for model development (80%) with the held-out testing set (20%) to evaluate its performance.</p><p><strong>Results: </strong>Our DL model achieved area under receiver operating characteristic curves (AUROCs) of 0.77, 0.80, and 0.87 for discriminating LVEF < 55%, < 50%, and < 40%. For LVEDD > 65 mm, LVESDi > 30 mm/m<sup>2</sup>, LVESVi > 45 ml/m<sup>2</sup>, LVEDVi > 99 ml/m<sup>2</sup>, LVMi > 158 mm/m<sup>2</sup>, and BAV, our model also achieved significant results, with AUROCs of 0.83, 0.85, 0.84, 0.81, 0.78, and 0.74, respectively. The SHapley Additive exPlanation values showed that our model focused on the QRS complex while making decisions.</p><p><strong>Conclusions: </strong>Our DL model found correlations between ECGs and parameters indicating LV remodeling and dysfunction in patients with significant AR. Analyzing ECGs with DL models may assist in the timely detection of LV dysfunction and screening for the necessity of additional echocardiography exams, especially when echocardiography might not be readily available.</p>","PeriodicalId":6957,"journal":{"name":"Acta Cardiologica Sinica","volume":"40 6","pages":"762-780"},"PeriodicalIF":1.8000,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11579687/pdf/","citationCount":"0","resultStr":"{\"title\":\"Correlating Electrocardiograms with Echocardiographic Parameters in Hemodynamically-Significant Aortic Regurgitation Using Deep Learning.\",\"authors\":\"Yi-Ting Li, Kuang-Chien Chiang, Alexander Te-Wei Shieh, Tetsuji Kitano, Yosuke Nabeshima, Chung-Yen Lee, Kang Liu, Kuan-Yu Lai, Meng-Han Tsai, Li-Ting Ho, Wen-Jone Chen, Masaaki Takeuchi, Tzung-Dau Wang, Li-Tan Yang\",\"doi\":\"10.6515/ACS.202411_40(6).20240918B\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Of all electrocardiogram (ECG) deep-learning (DL) models used to detect left-sided valvular heart diseases, aortic regurgitation (AR) has been the hardest to detect. Moreover, to what extent ECGs could detect AR-related left ventricular (LV) remodeling and dysfunction is unknown.</p><p><strong>Objectives: </strong>We aimed to evaluate the ability of DL-based ECG models to predict LV remodeling parameters associated with hemodynamically significant AR.</p><p><strong>Methods: </strong>From 573 consecutive patients, 1457 12-lead ECGs close to baseline transthoracic echocardiograms confirming ≥ moderate-severe AR and before aortic valve surgery were retrospectively collected. A ResNet-based model was used to predict LV ejection fraction (LVEF), LV end-diastolic dimension (LVEDD), LV end-systolic dimension index (LVESDi), LV mass index (LVMi), LV end-diastolic volume index (LVEDVi), LV end-systolic volume index (LVESVi), and bicuspid aortic valve (BAV) from the ECGs. Five-fold cross-validation was used for model development (80%) with the held-out testing set (20%) to evaluate its performance.</p><p><strong>Results: </strong>Our DL model achieved area under receiver operating characteristic curves (AUROCs) of 0.77, 0.80, and 0.87 for discriminating LVEF < 55%, < 50%, and < 40%. For LVEDD > 65 mm, LVESDi > 30 mm/m<sup>2</sup>, LVESVi > 45 ml/m<sup>2</sup>, LVEDVi > 99 ml/m<sup>2</sup>, LVMi > 158 mm/m<sup>2</sup>, and BAV, our model also achieved significant results, with AUROCs of 0.83, 0.85, 0.84, 0.81, 0.78, and 0.74, respectively. The SHapley Additive exPlanation values showed that our model focused on the QRS complex while making decisions.</p><p><strong>Conclusions: </strong>Our DL model found correlations between ECGs and parameters indicating LV remodeling and dysfunction in patients with significant AR. Analyzing ECGs with DL models may assist in the timely detection of LV dysfunction and screening for the necessity of additional echocardiography exams, especially when echocardiography might not be readily available.</p>\",\"PeriodicalId\":6957,\"journal\":{\"name\":\"Acta Cardiologica Sinica\",\"volume\":\"40 6\",\"pages\":\"762-780\"},\"PeriodicalIF\":1.8000,\"publicationDate\":\"2024-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11579687/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Acta Cardiologica Sinica\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.6515/ACS.202411_40(6).20240918B\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"CARDIAC & CARDIOVASCULAR SYSTEMS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Acta Cardiologica Sinica","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.6515/ACS.202411_40(6).20240918B","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"CARDIAC & CARDIOVASCULAR SYSTEMS","Score":null,"Total":0}
Correlating Electrocardiograms with Echocardiographic Parameters in Hemodynamically-Significant Aortic Regurgitation Using Deep Learning.
Background: Of all electrocardiogram (ECG) deep-learning (DL) models used to detect left-sided valvular heart diseases, aortic regurgitation (AR) has been the hardest to detect. Moreover, to what extent ECGs could detect AR-related left ventricular (LV) remodeling and dysfunction is unknown.
Objectives: We aimed to evaluate the ability of DL-based ECG models to predict LV remodeling parameters associated with hemodynamically significant AR.
Methods: From 573 consecutive patients, 1457 12-lead ECGs close to baseline transthoracic echocardiograms confirming ≥ moderate-severe AR and before aortic valve surgery were retrospectively collected. A ResNet-based model was used to predict LV ejection fraction (LVEF), LV end-diastolic dimension (LVEDD), LV end-systolic dimension index (LVESDi), LV mass index (LVMi), LV end-diastolic volume index (LVEDVi), LV end-systolic volume index (LVESVi), and bicuspid aortic valve (BAV) from the ECGs. Five-fold cross-validation was used for model development (80%) with the held-out testing set (20%) to evaluate its performance.
Results: Our DL model achieved area under receiver operating characteristic curves (AUROCs) of 0.77, 0.80, and 0.87 for discriminating LVEF < 55%, < 50%, and < 40%. For LVEDD > 65 mm, LVESDi > 30 mm/m2, LVESVi > 45 ml/m2, LVEDVi > 99 ml/m2, LVMi > 158 mm/m2, and BAV, our model also achieved significant results, with AUROCs of 0.83, 0.85, 0.84, 0.81, 0.78, and 0.74, respectively. The SHapley Additive exPlanation values showed that our model focused on the QRS complex while making decisions.
Conclusions: Our DL model found correlations between ECGs and parameters indicating LV remodeling and dysfunction in patients with significant AR. Analyzing ECGs with DL models may assist in the timely detection of LV dysfunction and screening for the necessity of additional echocardiography exams, especially when echocardiography might not be readily available.
期刊介绍：
Acta Cardiologica Sinica welcomes all the papers in the fields related to cardiovascular medicine including basic research, vascular biology, clinical pharmacology, clinical trial, critical care medicine, coronary artery disease, interventional cardiology, arrythmia and electrophysiology, atherosclerosis, hypertension, cardiomyopathy and heart failure, valvular and structure cardiac disease, pediatric cardiology, cardiovascular surgery, and so on. We received papers from more than 20 countries and areas of the world. Currently, 40% of the papers were submitted to Acta Cardiologica Sinica from Taiwan, 20% from China, and 20% from the other countries and areas in the world. The acceptance rate for publication was around 50% in general.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
