Explainable localization of premature ventricular contraction using deep learning-based semantic segmentation of 12-lead electrocardiogram

被引:1
作者
Kujime, Kota [1 ]
Seno, Hiroshi [1 ]
Nakajima, Kenzaburo [2 ]
Yamazaki, Masatoshi [1 ,3 ]
Sakuma, Ichiro [1 ]
Yamagata, Kenichiro [4 ]
Kusano, Kengo [2 ]
Tomii, Naoki [1 ]
机构
[1] Univ Tokyo, Grad Sch Engn, Dept Precis Engn, Tokyo, Japan
[2] Natl Cerebral & Cardiovasc Ctr, Dept Cardiovasc Med, Osaka, Japan
[3] Nagano Hosp, Dept Cardiol, Okayama, Japan
[4] Univ Tokyo, Grad Sch Med, Dept Cardiovasc Med, Tokyo, Japan
基金
日本学术振兴会;
关键词
automatic ECG diagnosis; deep neural network; premature ventricular contraction; TRACT TACHYCARDIA ORIGIN; OPTIMAL ABLATION SITE; OUTFLOW TRACT; ECG ALGORITHM; CRITERION; SPECTRUM;
D O I
10.1002/joa3.13096
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
BackgroundPredicting the origin of premature ventricular contraction (PVC) from the preoperative electrocardiogram (ECG) is important for catheter ablation therapies. We propose an explainable method that localizes PVC origin based on the semantic segmentation result of a 12-lead ECG using a deep neural network, considering suitable diagnosis support for clinical application.MethodsThe deep learning-based semantic segmentation model was trained using 265 12-lead ECG recordings from 84 patients with frequent PVCs. The model classified each ECG sampling time into four categories: background (BG), sinus rhythm (SR), PVC originating from the left ventricular outflow tract (PVC-L), and PVC originating from the right ventricular outflow tract (PVC-R). Based on the ECG segmentation results, a rule-based algorithm classified ECG recordings into three categories: PVC-L, PVC-R, as well as Neutral, which is a group for the recordings requiring the physician's careful assessment before separating them into PVC-L and PVC-R. The proposed method was evaluated with a public dataset which was used in previous research.ResultsThe evaluation of the proposed method achieved neutral rate, accuracy, sensitivity, specificity, F1-score, and area under the curve of 0.098, 0.932, 0.963, 0.882, 0.945, and 0.852 on a private dataset, and 0.284, 0.916, 0.912, 0.930, 0.943, and 0.848 on a public dataset, respectively. These quantitative results indicated that the proposed method outperformed almost all previous studies, although a significant number of recordings resulted in requiring the physician's assessment.ConclusionsThe feasibility of explainable localization of premature ventricular contraction was demonstrated using deep learning-based semantic segmentation of 12-lead ECG. Clinical trial registration: M26-148-8.ConclusionsThe feasibility of explainable localization of premature ventricular contraction was demonstrated using deep learning-based semantic segmentation of 12-lead ECG. Clinical trial registration: M26-148-8. Deep learning-based semantic segmentation on multiple leads ECG and a rule-based localization algorithm based on the preceding segmentation results were proposed. Except for neutral cases, which are the recordings requiring the physician's careful assessment, the model outperformed the conventional PVC localization methods.image
引用
收藏
页码:948 / 957
页数:10
相关论文
共 25 条
[21]   Novel transitional zone index allows more accurate differentiation between idiopathic right ventricular outflow tract and aortic sinus cusp ventricular arrhythmias [J].
Yoshida, Naoki ;
Inden, Yasuya ;
Uchikawa, Tomohiro ;
Kamiya, Hiromi ;
Kitamura, Kazuhisa ;
Shimano, Masayuki ;
Tsuji, Yukiomi ;
Hirai, Makoto ;
Murohara, Toyoaki .
HEART RHYTHM, 2011, 8 (03) :349-356
[22]   Electrocardiographic algorithm to identify the optimal target ablation site for idiopathic right ventricular outflow tract ventricular premature contraction [J].
Zhang, Fengxiang ;
Chen, Minglong ;
Yang, Bing ;
Ju, Weizhu ;
Chen, Hongwu ;
Yu, Jian ;
Lau, Chu-Pak ;
Cao, Kejiang ;
Tse, Hung-Fat .
EUROPACE, 2009, 11 (09) :1214-1220
[23]   Machine learning for distinguishing right from left premature ventricular contraction origin using surface electrocardiogram features [J].
Zhao, Wei ;
Zhu, Rui ;
Zhang, Jian ;
Mao, Yangming ;
Chen, Hongwu ;
Ju, Weizhu ;
Li, Mingfang ;
Yang, Gang ;
Gu, Kai ;
Wang, Zidun ;
Liu, Hailei ;
Shi, Jiaojiao ;
Jiang, Xiaohong ;
Kojodjojo, Pipin ;
Chen, Minglong ;
Zhang, Fengxiang .
HEART RHYTHM, 2022, 19 (11) :1781-1789
[24]  
Zheng J., 2019, 12LEAD ECG DATABASE
[25]   A High-Precision Machine Learning Algorithm to Classify Left and Right Outflow Tract Ventricular Tachycardia [J].
Zheng, Jianwei ;
Fu, Guohua ;
Abudayyeh, Islam ;
Yacoub, Magdi ;
Chang, Anthony ;
Feaster, William W. ;
Ehwerhemuepha, Louis ;
El-Askary, Hesham ;
Du, Xianfeng ;
He, Bin ;
Feng, Mingjun ;
Yu, Yibo ;
Wang, Binhao ;
Liu, Jing ;
Yao, Hai ;
Chu, Huimin ;
Rakovski, Cyril .
FRONTIERS IN PHYSIOLOGY, 2021, 12