Automatic Classification of Myocardial Infarction Using Spline Representation of Single-Lead Derived Vectorcardiography

Abstract

Myocardial infarction (MI) is one of the most prevalent cardiovascular diseases worldwide and most patients suffer from MI without awareness. Therefore, early diagnosis and timely treatment are crucial to guarantee the life safety of MI patients. Most wearable monitoring devices only provide single-lead electrocardiography (ECG), which represents a major limitation for their applicability in diagnosis of MI. Incorporating the derived vectorcardiography (VCG) techniques can help monitor the three-dimensional electrical activities of human hearts. This study presents a patient-specific reconstruction method based on long short-term memory (LSTM) network to exploit both intra- and inter-lead correlations of ECG signals. MI-induced changes in the morphological and temporal wave features are extracted from the derived VCG using spline approximation. After the feature extraction, a classifier based on multilayer perceptron network is used for MI classification. Experiments on PTB diagnostic database demonstrate that the proposed system achieved satisfactory performance to differentiating MI patients from healthy subjects and to localizing the infarcted area.

On the history of vectorcardiography: past, present, future

“…В контексте векторкардиографии зародился ряд новых направлений исследований: периодическая динамика реполяризации [40,41]; синтез ВКГ из одного отведения ЭКГ [42]; привязка ВКГ к структурам сердца на основе пациент-специфических моделей [43].…”

Reliable Detection of Myocardial Ischemia Using Machine Learning Based on Temporal-Spatial Characteristics of Electrocardiogram and Vectorcardiogram

“…VCG features extracted automatically are employed to detect ischemic beats ( Häggmark et al, 2008 ; Aranda Hernandez et al, 2018 ; Braun et al, 2020 ; Chuang et al, 2020 ) or frames ( Rahul et al, 2021 ), discriminate myocardial ischemia ( Deng et al, 2017 ), localize the culprit infarct-related arteries in acute myocardial infarction ( Le et al, 2020 ), and screen for the presence of scar tissue in the myocardium ( Uyen Chau et al, 2018 ).…”

“…The number of features is highly diverse in existing studies. 12 nonlinear ECG features ( Liu J. et al, 2020 ), 72 multiscale energy and eigenspace ECG features ( Sharma et al, 2015 ), 288 ECG features in time, frequency, nonlinear, and entropy domains ( Liu et al, 2019 ), 7 ( Correa et al, 2013 ) and 52 features of 3-lead ECG ( Chuang et al, 2020 ), 290 multidimensional parameters of 5-lead VCG ( Braun et al, 2020 ), 22 features of 6-lead VCG ( Dehnavi et al, 2011 ), and 322 pseudo-VCG features ( Aranda Hernandez et al, 2018 ). These features are fed into machine learning (ML) models ( Aranda Hernandez et al, 2018 ; Chuang et al, 2020 ) to develop computer-aided diagnostic models of myocardial ischemia, achieving higher efficiency and accuracy than manual inspection.…”

Continuous-Time Attention for Sequential Learning

“…All rights reserved. a patient by using several biomedical signals acquired by different sensors or diagnosis facilities where the sampling resolutions in different signals are varied (Lipton, Kale, and Wetzel 2016;Che et al 2018;El Naqa et al 2018;Chuang et al 2020). Some other practical applications in presence of irregular-sampled data (Elman 1990) include financial marketing (Bauer, Schölkopf, and Peters 2016), weather forecasting (Shi et al 2015) and traffic engineering (Wang et al 2017), to name a few.…”