Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12-lead electrocardiogram

Khaled Rjoob, RR Bond, D Finlay, V. E. McGilligan, Stephen James Leslie, Aleeha Iftikhar, D Guldenring, Ali Rababah, Charles Knoery, Anne McShane, Aaron Peace

Research output: Contribution to journalArticle

Abstract

Background: Electrocardiogram (ECG) lead misplacement can adversely affect ECG diagnosis and subsequent clinical decisions. V1 and V2 are commonly placed superior of their correct position. The aim of the current study was to use machine learning approaches to detect V1 and V2 lead misplacement to enhance ECG data quality. Method: ECGs for 453 patients, (normal n=151, Left Ventricular Hypertrophy (LVH) n=151, Myocardial Infarction n=151) were extracted from body surface potential maps. These were used to extract both the correct and incorrectly placed V1 and V2 leads. The prevalence for correct and incorrect leads were 50%. Sixteen features were extracted in three different domains: time-based, statistical and time-frequency features using a wavelet transform. A hybrid feature selection approach was applied to select an optimal set of features. To ensure optimal model selection, five classifiers were used and compared. The aforementioned feature selection approach and classifiers were applied for V1 and V2 misplacement in three different positions: first, second and third intercostal spaces (ICS). Results: The accuracy for V1 misplacement detection was 93.9%, 89.3%, 72.8% in the first, second and third ICS respectively. In V2, the accuracy was 93.6%, 86.6% and 68.1% in the first, second and third ICS respectively. There is a noticeable decline in accuracy when detecting misplacement in the third ICS which is expected.
LanguageEnglish
Number of pages10
JournalJournal of Electrocardiology
Early online date24 Aug 2019
DOIs
Publication statusE-pub ahead of print - 24 Aug 2019

Fingerprint

Electrocardiography
Electrodes
Thorax
Wavelet Analysis
Left Ventricular Hypertrophy
Myocardial Infarction
Lead
Machine Learning

Keywords

  • Machine learning
  • Feature extraction
  • Body surface potential maps
  • Lead misplacement
  • Electrode misplacement
  • Chest leads

Cite this

Rjoob, Khaled ; Bond, RR ; Finlay, D ; McGilligan, V. E. ; Leslie, Stephen James ; Iftikhar, Aleeha ; Guldenring, D ; Rababah, Ali ; Knoery, Charles ; McShane, Anne ; Peace, Aaron. / Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12-lead electrocardiogram. In: Journal of Electrocardiology. 2019.
@article{0520a491660943d59a768f5adba3a6cf,
title = "Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12-lead electrocardiogram",
abstract = "Background: Electrocardiogram (ECG) lead misplacement can adversely affect ECG diagnosis and subsequent clinical decisions. V1 and V2 are commonly placed superior of their correct position. The aim of the current study was to use machine learning approaches to detect V1 and V2 lead misplacement to enhance ECG data quality. Method: ECGs for 453 patients, (normal n=151, Left Ventricular Hypertrophy (LVH) n=151, Myocardial Infarction n=151) were extracted from body surface potential maps. These were used to extract both the correct and incorrectly placed V1 and V2 leads. The prevalence for correct and incorrect leads were 50{\%}. Sixteen features were extracted in three different domains: time-based, statistical and time-frequency features using a wavelet transform. A hybrid feature selection approach was applied to select an optimal set of features. To ensure optimal model selection, five classifiers were used and compared. The aforementioned feature selection approach and classifiers were applied for V1 and V2 misplacement in three different positions: first, second and third intercostal spaces (ICS). Results: The accuracy for V1 misplacement detection was 93.9{\%}, 89.3{\%}, 72.8{\%} in the first, second and third ICS respectively. In V2, the accuracy was 93.6{\%}, 86.6{\%} and 68.1{\%} in the first, second and third ICS respectively. There is a noticeable decline in accuracy when detecting misplacement in the third ICS which is expected.",
keywords = "Machine learning, Feature extraction, Body surface potential maps, Lead misplacement, Electrode misplacement, Chest leads",
author = "Khaled Rjoob and RR Bond and D Finlay and McGilligan, {V. E.} and Leslie, {Stephen James} and Aleeha Iftikhar and D Guldenring and Ali Rababah and Charles Knoery and Anne McShane and Aaron Peace",
year = "2019",
month = "8",
day = "24",
doi = "10.1016/j.jelectrocard.2019.08.017",
language = "English",
journal = "Journal of Electrocardiology",
issn = "0022-0736",
publisher = "Elsevier",

}

Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12-lead electrocardiogram. / Rjoob, Khaled; Bond, RR; Finlay, D; McGilligan, V. E.; Leslie, Stephen James; Iftikhar, Aleeha; Guldenring, D; Rababah, Ali; Knoery, Charles; McShane, Anne; Peace, Aaron.

In: Journal of Electrocardiology, 24.08.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12-lead electrocardiogram

AU - Rjoob, Khaled

AU - Bond, RR

AU - Finlay, D

AU - McGilligan, V. E.

AU - Leslie, Stephen James

AU - Iftikhar, Aleeha

AU - Guldenring, D

AU - Rababah, Ali

AU - Knoery, Charles

AU - McShane, Anne

AU - Peace, Aaron

PY - 2019/8/24

Y1 - 2019/8/24

N2 - Background: Electrocardiogram (ECG) lead misplacement can adversely affect ECG diagnosis and subsequent clinical decisions. V1 and V2 are commonly placed superior of their correct position. The aim of the current study was to use machine learning approaches to detect V1 and V2 lead misplacement to enhance ECG data quality. Method: ECGs for 453 patients, (normal n=151, Left Ventricular Hypertrophy (LVH) n=151, Myocardial Infarction n=151) were extracted from body surface potential maps. These were used to extract both the correct and incorrectly placed V1 and V2 leads. The prevalence for correct and incorrect leads were 50%. Sixteen features were extracted in three different domains: time-based, statistical and time-frequency features using a wavelet transform. A hybrid feature selection approach was applied to select an optimal set of features. To ensure optimal model selection, five classifiers were used and compared. The aforementioned feature selection approach and classifiers were applied for V1 and V2 misplacement in three different positions: first, second and third intercostal spaces (ICS). Results: The accuracy for V1 misplacement detection was 93.9%, 89.3%, 72.8% in the first, second and third ICS respectively. In V2, the accuracy was 93.6%, 86.6% and 68.1% in the first, second and third ICS respectively. There is a noticeable decline in accuracy when detecting misplacement in the third ICS which is expected.

AB - Background: Electrocardiogram (ECG) lead misplacement can adversely affect ECG diagnosis and subsequent clinical decisions. V1 and V2 are commonly placed superior of their correct position. The aim of the current study was to use machine learning approaches to detect V1 and V2 lead misplacement to enhance ECG data quality. Method: ECGs for 453 patients, (normal n=151, Left Ventricular Hypertrophy (LVH) n=151, Myocardial Infarction n=151) were extracted from body surface potential maps. These were used to extract both the correct and incorrectly placed V1 and V2 leads. The prevalence for correct and incorrect leads were 50%. Sixteen features were extracted in three different domains: time-based, statistical and time-frequency features using a wavelet transform. A hybrid feature selection approach was applied to select an optimal set of features. To ensure optimal model selection, five classifiers were used and compared. The aforementioned feature selection approach and classifiers were applied for V1 and V2 misplacement in three different positions: first, second and third intercostal spaces (ICS). Results: The accuracy for V1 misplacement detection was 93.9%, 89.3%, 72.8% in the first, second and third ICS respectively. In V2, the accuracy was 93.6%, 86.6% and 68.1% in the first, second and third ICS respectively. There is a noticeable decline in accuracy when detecting misplacement in the third ICS which is expected.

KW - Machine learning

KW - Feature extraction

KW - Body surface potential maps

KW - Lead misplacement

KW - Electrode misplacement

KW - Chest leads

UR - https://www.sciencedirect.com/science/article/pii/S0022073619303851

U2 - 10.1016/j.jelectrocard.2019.08.017

DO - 10.1016/j.jelectrocard.2019.08.017

M3 - Article

JO - Journal of Electrocardiology

T2 - Journal of Electrocardiology

JF - Journal of Electrocardiology

SN - 0022-0736

ER -