Standard ECG Lead I Prospective Estimation Study from Far-field Bipolar Leads on the Left Upper Arm: A Neural Network Approach

Pedro Vizcaya, Gilberto Perpiñan, David McEneaney, OJ Escalona

Research output: Contribution to journalArticle

Abstract

In this study, the feasibility of interpreting heart rhythms from far-field bipolar ECG arm-band lead recordings on the left-upper-arm (LUA), is evaluated in a clinical multichannel arm-ECG mapping database (N = 153 subjects) for the prospective development of long-term heart rhythm monitoring from comfortable arm wearable devices. A preliminary multivariable linear regression analysis on ECG chest Lead I from 10 selected far-field bipolar leads along the left arm, indicated that 3 of them in the LUA were relevant and worth evaluating in more detail from a heart rhythm information perspective.
To derive a good and effective estimation process, a time series non-linear regression point estimator, using an artificial neural network with 2 lags was investigated, showing a correlation coefficient of up to 0.969 for a single subject. Then, a vector approach was adopted for the whole LUA database, aiming to develop a subject independent estimation process of the P-QRS-T waveform interval and its heart rhythm attributes in the standard chest Lead I. In the same study, the first 96 coefficients, of the Discrete Cosine Transform on the P-QRS-T interval were used as a means for reducing the dimensionality of the input space, with a loss of just 0.1% in power, and reducing the dimensionality to just 5% of the original size. The trained ANN for ECG Lead I estimation from one upper arm Lead-1 showed a correlation coefficient above 80% on a beat-to-beat basis, an improvement on all but 1.34% of the beats estimated for a typical train/test partition of the LUA database. The non-triviality of the results was tested with random and intentional true negatives. Information theory analytics revealed that there is an estimated information of 1.6 bits/beat between LUA armband bipolar leads and the standard Lead I.
LanguageEnglish
Article numberBSPC 1471
Pages171-180
Number of pages10
JournalBiomedical Signal Processing and Control
Volume51
DOIs
Publication statusPublished - 4 Mar 2019

Fingerprint

Electrocardiography
Arm
Lead
Prospective Studies
Neural networks
Discrete cosine transforms
Information theory
Databases
Linear regression
Regression analysis
Time series
Thorax
Information Theory
Monitoring
Feasibility Studies
Linear Models
Regression Analysis
Equipment and Supplies

Keywords

  • Artificial neural network
  • Biological information theory
  • Biomedical monitoring
  • Electrocardiography
  • Mutual information
  • Wearable sensors

Cite this

@article{ba5f2d53d3594251b7dcaa99ef92526f,
title = "Standard ECG Lead I Prospective Estimation Study from Far-field Bipolar Leads on the Left Upper Arm: A Neural Network Approach",
abstract = "In this study, the feasibility of interpreting heart rhythms from far-field bipolar ECG arm-band lead recordings on the left-upper-arm (LUA), is evaluated in a clinical multichannel arm-ECG mapping database (N = 153 subjects) for the prospective development of long-term heart rhythm monitoring from comfortable arm wearable devices. A preliminary multivariable linear regression analysis on ECG chest Lead I from 10 selected far-field bipolar leads along the left arm, indicated that 3 of them in the LUA were relevant and worth evaluating in more detail from a heart rhythm information perspective. To derive a good and effective estimation process, a time series non-linear regression point estimator, using an artificial neural network with 2 lags was investigated, showing a correlation coefficient of up to 0.969 for a single subject. Then, a vector approach was adopted for the whole LUA database, aiming to develop a subject independent estimation process of the P-QRS-T waveform interval and its heart rhythm attributes in the standard chest Lead I. In the same study, the first 96 coefficients, of the Discrete Cosine Transform on the P-QRS-T interval were used as a means for reducing the dimensionality of the input space, with a loss of just 0.1{\%} in power, and reducing the dimensionality to just 5{\%} of the original size. The trained ANN for ECG Lead I estimation from one upper arm Lead-1 showed a correlation coefficient above 80{\%} on a beat-to-beat basis, an improvement on all but 1.34{\%} of the beats estimated for a typical train/test partition of the LUA database. The non-triviality of the results was tested with random and intentional true negatives. Information theory analytics revealed that there is an estimated information of 1.6 bits/beat between LUA armband bipolar leads and the standard Lead I.",
keywords = "Artificial neural network, Biological information theory, Biomedical monitoring, Electrocardiography, Mutual information, Wearable sensors",
author = "Pedro Vizcaya and Gilberto Perpi{\~n}an and David McEneaney and OJ Escalona",
year = "2019",
month = "3",
day = "4",
doi = "10.1016/j.bspc.2019.01.020",
language = "English",
volume = "51",
pages = "171--180",

}

Standard ECG Lead I Prospective Estimation Study from Far-field Bipolar Leads on the Left Upper Arm: A Neural Network Approach. / Vizcaya, Pedro; Perpiñan, Gilberto; McEneaney, David; Escalona, OJ.

Vol. 51, BSPC 1471, 04.03.2019, p. 171-180.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Standard ECG Lead I Prospective Estimation Study from Far-field Bipolar Leads on the Left Upper Arm: A Neural Network Approach

AU - Vizcaya, Pedro

AU - Perpiñan, Gilberto

AU - McEneaney, David

AU - Escalona, OJ

PY - 2019/3/4

Y1 - 2019/3/4

N2 - In this study, the feasibility of interpreting heart rhythms from far-field bipolar ECG arm-band lead recordings on the left-upper-arm (LUA), is evaluated in a clinical multichannel arm-ECG mapping database (N = 153 subjects) for the prospective development of long-term heart rhythm monitoring from comfortable arm wearable devices. A preliminary multivariable linear regression analysis on ECG chest Lead I from 10 selected far-field bipolar leads along the left arm, indicated that 3 of them in the LUA were relevant and worth evaluating in more detail from a heart rhythm information perspective. To derive a good and effective estimation process, a time series non-linear regression point estimator, using an artificial neural network with 2 lags was investigated, showing a correlation coefficient of up to 0.969 for a single subject. Then, a vector approach was adopted for the whole LUA database, aiming to develop a subject independent estimation process of the P-QRS-T waveform interval and its heart rhythm attributes in the standard chest Lead I. In the same study, the first 96 coefficients, of the Discrete Cosine Transform on the P-QRS-T interval were used as a means for reducing the dimensionality of the input space, with a loss of just 0.1% in power, and reducing the dimensionality to just 5% of the original size. The trained ANN for ECG Lead I estimation from one upper arm Lead-1 showed a correlation coefficient above 80% on a beat-to-beat basis, an improvement on all but 1.34% of the beats estimated for a typical train/test partition of the LUA database. The non-triviality of the results was tested with random and intentional true negatives. Information theory analytics revealed that there is an estimated information of 1.6 bits/beat between LUA armband bipolar leads and the standard Lead I.

AB - In this study, the feasibility of interpreting heart rhythms from far-field bipolar ECG arm-band lead recordings on the left-upper-arm (LUA), is evaluated in a clinical multichannel arm-ECG mapping database (N = 153 subjects) for the prospective development of long-term heart rhythm monitoring from comfortable arm wearable devices. A preliminary multivariable linear regression analysis on ECG chest Lead I from 10 selected far-field bipolar leads along the left arm, indicated that 3 of them in the LUA were relevant and worth evaluating in more detail from a heart rhythm information perspective. To derive a good and effective estimation process, a time series non-linear regression point estimator, using an artificial neural network with 2 lags was investigated, showing a correlation coefficient of up to 0.969 for a single subject. Then, a vector approach was adopted for the whole LUA database, aiming to develop a subject independent estimation process of the P-QRS-T waveform interval and its heart rhythm attributes in the standard chest Lead I. In the same study, the first 96 coefficients, of the Discrete Cosine Transform on the P-QRS-T interval were used as a means for reducing the dimensionality of the input space, with a loss of just 0.1% in power, and reducing the dimensionality to just 5% of the original size. The trained ANN for ECG Lead I estimation from one upper arm Lead-1 showed a correlation coefficient above 80% on a beat-to-beat basis, an improvement on all but 1.34% of the beats estimated for a typical train/test partition of the LUA database. The non-triviality of the results was tested with random and intentional true negatives. Information theory analytics revealed that there is an estimated information of 1.6 bits/beat between LUA armband bipolar leads and the standard Lead I.

KW - Artificial neural network

KW - Biological information theory

KW - Biomedical monitoring

KW - Electrocardiography

KW - Mutual information

KW - Wearable sensors

UR - http://www.scopus.com/inward/record.url?scp=85062338670&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2019.01.020

DO - 10.1016/j.bspc.2019.01.020

M3 - Article

VL - 51

SP - 171

EP - 180

M1 - BSPC 1471

ER -