A hybrid machine learning approach to localizing the origin of ventricular tachycardia using 12-lead electrocardiograms.
A hybrid machine learning approach to localizing the origin of ventricular tachycardia using 12-lead electrocardiograms.
复制标题
DOI：
10.1016/j.compbiomed.2020.104013
复制
发表时间：
2020-11
影响因子：
7.7
通讯作者：
Wang L
中科院分区：
文献类型：
--
作者：
Missel R;Gyawali PK;Murkute JV;Li Z;Zhou S;AbdelWahab A;Davis J;Warren J;Sapp JL;Wang L
机器学习模型可以帮助使用12导联心电图定位室性心动过速（VT）的起源部位。然而，基于人群的模型受到ECG数据内的受试者间解剖变化的影响，而患者特定的模型面临着收集什么起搏数据用于训练的公开挑战。这项研究提出并验证了第一个混合模型，该模型结合了人群和特定于患者的机器学习，以快速“计算机引导的起搏标测”。首先离线训练基于人群的深度学习模型，以解开受试者之间的差异并对VT起源部位进行区域化。给定具有目标室性心动过速的新患者，在线患者特异性模型-在通过基于人群的预测初始化之后-然后通过主动建议下一步起搏的位置并利用每个添加的起搏数据改进预测，逐步引导起搏标测朝向室性心动过速起源部位，从而在真实的时间内构建。在来自38名患者的起搏标测数据上训练群体模型，随后在一名患者上调整患者特异性模型。在一个单独的8名患者队列中对所得混合模型进行了测试，确定了1）193个LV内起搏部位和2）9个具有临床确定出口部位的VT。混合模型在定位LV起搏部位时使用5.4 ± 2.5个起搏部位实现了5.3 ± 2.6 mm的定位误差，与无主动引导的模型相比，使用显著更少量的训练部位实现了显著更高的准确度。所提出的混合模型有可能帮助快速起搏标测室性心动过速的介入目标。
Machine learning models may help localize the site of origin of ventricular tachycardia (VT) using 12-lead electrocardiograms. However, population-based models suffer from inter-subject anatomical variations within ECG data, while patient-specific models face the open challenge of what pacing data to collect for training. This study presents and validates the first hybrid model that combines population and patient-specific machine learning for rapid “computer-guided pace-mapping”. A population-based deep learning model was first trained offline to disentangle inter-subject variations and regionalize the site of VT origin. Given a new patient with a target VT, an on-line patient-specific model -- after being initialized by the population-based prediction -- was then built in real time by actively suggesting where to pace next and improving the prediction with each added pacing data, progressively guiding pace-mapping towards the site of VT origin. The population model was trained on pace-mapping data from 38 patients and the patient-specific model was subsequently tuned on one patient. The resulting hybrid model was tested on a separate cohort of eight patients in localizing 1) 193 LV endocardial pacing sites, and 2) nine VTs with clinically determined exit sites. The hybrid model achieved a localization error of 5.3 ± 2.6 mm using 5.4 ± 2.5 pacing sites in localizing LV pacing sites, achieving a significantly higher accuracy with a significantly smaller amount of training sites in comparison to models without active guidance. The presented hybrid model has the potential to assist rapid pace-mapping of interventional targets in VT.
登录
查看更多内容
影响因子：
8.9
作者：
Koplan, Bruce A.;Stevenson, William G.
通讯作者：
Stevenson, William G.
DOI：
10.1016/j.jacep.2017.02.024
发表时间：
2017-07-01
期刊：
JACC. Clinical electrophysiology
影响因子：
--
作者：
Sapp, John L;Bar-Tal, Meir;Horacek, B Milan
通讯作者：
Horacek, B Milan
DOI：
10.1109/tbme.2017.2756869
发表时间：
2018-07
期刊：
IEEE transactions on bio-medical engineering
影响因子：
--
作者：
Yang T;Yu L;Jin Q;Wu L;He B
通讯作者：
He B
影响因子：
10.6
作者：
Alawad M;Wang L
通讯作者：
Wang L
影响因子：
3.8
作者：
Zhou, Shijie;AbdelWahab, Amir;Horacek, B. Milan
通讯作者：
Horacek, B. Milan