Jianwei Zheng1, Guohua Fu2, Islam Abudayyeh3, Magdi Yacoub4, Anthony Chang5, William W Feaster5, Louis Ehwerhemuepha5, Hesham El-Askary1,6, Xianfeng Du2, Bin He2, Mingjun Feng2, Yibo Yu2, Binhao Wang2, Jing Liu2, Hai Yao7, Huimin Chu2, Cyril Rakovski1. 1. Computational and Data Science, Chapman University, Orange, CA, United States. 2. Department of Cardiology, Ningbo First Hospital of Zhejiang University, Hangzhou, China. 3. Department of Cardiology, Loma Linda University, Loma Linda, CA, United States. 4. Harefield Heart Science Center, Imperial College London, London, United Kingdom. 5. CHOC Children's Hospital, Orange, CA, United States. 6. Department of Environmental Sciences, Faculty of Science, Alexandria University, Alexandria, Egypt. 7. Zhejiang Cachet Jetboom Medical Devices Co., Ltd., Hangzhou, China.
Abstract
INTRODUCTION: Multiple algorithms based on 12-lead ECG measurements have been proposed to identify the right ventricular outflow tract (RVOT) and left ventricular outflow tract (LVOT) locations from which ventricular tachycardia (VT) and frequent premature ventricular complex (PVC) originate. However, a clinical-grade machine learning algorithm that automatically analyzes characteristics of 12-lead ECGs and predicts RVOT or LVOT origins of VT and PVC is not currently available. The effective ablation sites of RVOT and LVOT, confirmed by a successful ablation procedure, provide evidence to create RVOT and LVOT labels for the machine learning model. METHODS: We randomly sampled training, validation, and testing data sets from 420 patients who underwent successful catheter ablation (CA) to treat VT or PVC, containing 340 (81%), 38 (9%), and 42 (10%) patients, respectively. We iteratively trained a machine learning algorithm supplied with 1,600,800 features extracted via our proprietary algorithm from 12-lead ECGs of the patients in the training cohort. The area under the curve (AUC) of the receiver operating characteristic curve was calculated from the internal validation data set to choose an optimal discretization cutoff threshold. RESULTS: The proposed approach attained the following performance: accuracy (ACC) of 97.62 (87.44-99.99), weighted F1-score of 98.46 (90-100), AUC of 98.99 (96.89-100), sensitivity (SE) of 96.97 (82.54-99.89), and specificity (SP) of 100 (62.97-100). CONCLUSIONS: The proposed multistage diagnostic scheme attained clinical-grade precision of prediction for LVOT and RVOT locations of VT origin with fewer applicability restrictions than prior studies.
INTRODUCTION: Multiple algorithms based on 12-lead ECG measurements have been proposed to identify the right ventricular outflow tract (RVOT) and left ventricular outflow tract (LVOT) locations from which ventricular tachycardia (VT) and frequent premature ventricular complex (PVC) originate. However, a clinical-grade machine learning algorithm that automatically analyzes characteristics of 12-lead ECGs and predicts RVOT or LVOT origins of VT and PVC is not currently available. The effective ablation sites of RVOT and LVOT, confirmed by a successful ablation procedure, provide evidence to create RVOT and LVOT labels for the machine learning model. METHODS: We randomly sampled training, validation, and testing data sets from 420 patients who underwent successful catheter ablation (CA) to treat VT or PVC, containing 340 (81%), 38 (9%), and 42 (10%) patients, respectively. We iteratively trained a machine learning algorithm supplied with 1,600,800 features extracted via our proprietary algorithm from 12-lead ECGs of the patients in the training cohort. The area under the curve (AUC) of the receiver operating characteristic curve was calculated from the internal validation data set to choose an optimal discretization cutoff threshold. RESULTS: The proposed approach attained the following performance: accuracy (ACC) of 97.62 (87.44-99.99), weighted F1-score of 98.46 (90-100), AUC of 98.99 (96.89-100), sensitivity (SE) of 96.97 (82.54-99.89), and specificity (SP) of 100 (62.97-100). CONCLUSIONS: The proposed multistage diagnostic scheme attained clinical-grade precision of prediction for LVOT and RVOT locations of VT origin with fewer applicability restrictions than prior studies.
Authors: Shuanglun Xie; Maciej Kubala; Jackson J Liang; Tatsuya Hayashi; Jaeseok Park; Irene Lucena Padros; Fermin C Garcia; Pasquale Santangeli; Gregory E Supple; David S Frankel; Erica S Zado; David Lin; Robert D Schaller; Sanjay Dixit; David J Callans; Saman Nazarian; Francis E Marchlinski Journal: J Cardiovasc Electrophysiol Date: 2018-10-08
Authors: Brian P Betensky; Robert E Park; Francis E Marchlinski; Matthew D Hutchinson; Fermin C Garcia; Sanjay Dixit; David J Callans; Joshua M Cooper; Rupa Bala; David Lin; Michael P Riley; Edward P Gerstenfeld Journal: J Am Coll Cardiol Date: 2011-05-31 Impact factor: 24.094
Authors: Jonathan W Dukes; Thomas A Dewland; Eric Vittinghoff; Mala C Mandyam; Susan R Heckbert; David S Siscovick; Phyllis K Stein; Bruce M Psaty; Nona Sotoodehnia; John S Gottdiener; Gregory M Marcus Journal: J Am Coll Cardiol Date: 2015-07-14 Impact factor: 24.094
Authors: Zhuoqiao He; Ming Liu; Min Yu; Nan Lu; Jia Li; Tan Xu; Jinxiu Zhu; Mary Clare O'Gara; Michael O'Meara; Hong Ye; Xuerui Tan Journal: J Cardiovasc Electrophysiol Date: 2018-04-20
Authors: Ruben Doste; Miguel Lozano; Guillermo Jimenez-Perez; Lluis Mont; Antonio Berruezo; Diego Penela; Oscar Camara; Rafael Sebastian Journal: Front Physiol Date: 2022-08-12 Impact factor: 4.755