Pairwise Feature Interactions to Predict Arrhythmic Risk of Brugada Syndrome

Sharen Lee; Jiandong Zhou; Konstantinos P Letsas; Ka Hou Christien Li; Tong Liu; Sven Zumhagen; Eric Schulze-Bahr; Gary Tse; Qingpeng Zhang

doi:10.23919/CinC53138.2021.9662913

Pairwise Feature Interactions to Predict Arrhythmic Risk of Brugada Syndrome

Sharen Lee
, Jiandong Zhou
, Konstantinos P Letsas
, Ka Hou Christien Li
, Tong Liu
, Sven Zumhagen
, Eric Schulze-Bahr
, Gary Tse^*
, Qingpeng Zhang^*

^*Corresponding author for this work

Department of Data Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

Electrocardiographic (ECG) indices were used for risk stratification in Brugada syndrome (BrS). However, nonlinear interactions between risk factors were ignored. Therefore, we adapted a generalized additive model with pair-wise interactions (GA2M) to predict BrS with spontaneous ventricular tachycardia/fibrillation (VT/VF) as outcomes based on specific ECG markers. A total of 191 adult patients with BrS from three centres (Germany, Greece and Hong Kong) were included for analysis. Depolarization and repolarization ECG markers were measured from the right precordial leads (V1 to V3). The proposed GA2M-based risk prediction model successfully identified a set of risk factors and their pairwise interactions in addition to the dispersion of repolarization/total repolarization (Tpeak- Tend x mean QT)). The model outperformed the baseline logistic model based on the same set of ECG measurements. In conclusion, the inclusion of pairwise interactions improved predictive performance and enabled more effective risk stratification in BrS.

Original language	English
Title of host publication	2021 Computing in Cardiology (CinC)
Publisher	IEEE
Number of pages	4
ISBN (Electronic)	9781665479165
ISBN (Print)	9781665467216
DOIs	https://doi.org/10.23919/CinC53138.2021.9662913
Publication status	Published - Sept 2021
Event	48th Computing in Cardiology Conference (CinC 2021) - Hotel Passage, Brno, Czech Republic Duration: 13 Sept 2021 → 15 Sept 2021 http://www.cinc2021.org/

Publication series

Name	Computing in Cardiology
ISSN (Print)	2325-8861
ISSN (Electronic)	2325-887X

Conference

Conference	48th Computing in Cardiology Conference (CinC 2021)
Place	Czech Republic
City	Brno
Period	13/09/21 → 15/09/21
Internet address	http://www.cinc2021.org/

Access Output

10.23919/CinC53138.2021.9662913

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{c94e7d5cbd704561a3f32c2b3eed4582,

title = "Pairwise Feature Interactions to Predict Arrhythmic Risk of Brugada Syndrome",

abstract = "Electrocardiographic (ECG) indices were used for risk stratification in Brugada syndrome (BrS). However, nonlinear interactions between risk factors were ignored. Therefore, we adapted a generalized additive model with pair-wise interactions (GA2M) to predict BrS with spontaneous ventricular tachycardia/fibrillation (VT/VF) as outcomes based on specific ECG markers. A total of 191 adult patients with BrS from three centres (Germany, Greece and Hong Kong) were included for analysis. Depolarization and repolarization ECG markers were measured from the right precordial leads (V1 to V3). The proposed GA2M-based risk prediction model successfully identified a set of risk factors and their pairwise interactions in addition to the dispersion of repolarization/total repolarization (Tpeak- Tend x mean QT)). The model outperformed the baseline logistic model based on the same set of ECG measurements. In conclusion, the inclusion of pairwise interactions improved predictive performance and enabled more effective risk stratification in BrS.",

author = "Sharen Lee and Jiandong Zhou and Letsas, \{Konstantinos P\} and \{Christien Li\}, \{Ka Hou\} and Tong Liu and Sven Zumhagen and Eric Schulze-Bahr and Gary Tse and Qingpeng Zhang",

year = "2021",

month = sep,

doi = "10.23919/CinC53138.2021.9662913",

language = "English",

isbn = "9781665467216",

series = "Computing in Cardiology",

publisher = "IEEE",

booktitle = "2021 Computing in Cardiology (CinC)",

address = "United States",

note = "48th Computing in Cardiology Conference (CinC 2021) ; Conference date: 13-09-2021 Through 15-09-2021",

url = "http://www.cinc2021.org/",

}

Lee, S, Zhou, J, Letsas, KP, Christien Li, KH, Liu, T, Zumhagen, S, Schulze-Bahr, E, Tse, G & Zhang, Q 2021, Pairwise Feature Interactions to Predict Arrhythmic Risk of Brugada Syndrome. in 2021 Computing in Cardiology (CinC). Computing in Cardiology, IEEE, 48th Computing in Cardiology Conference (CinC 2021), Brno, Czech Republic, 13/09/21. https://doi.org/10.23919/CinC53138.2021.9662913

TY - GEN

T1 - Pairwise Feature Interactions to Predict Arrhythmic Risk of Brugada Syndrome

AU - Lee, Sharen

AU - Zhou, Jiandong

AU - Letsas, Konstantinos P

AU - Christien Li, Ka Hou

AU - Liu, Tong

AU - Zumhagen, Sven

AU - Schulze-Bahr, Eric

AU - Tse, Gary

AU - Zhang, Qingpeng

PY - 2021/9

Y1 - 2021/9

N2 - Electrocardiographic (ECG) indices were used for risk stratification in Brugada syndrome (BrS). However, nonlinear interactions between risk factors were ignored. Therefore, we adapted a generalized additive model with pair-wise interactions (GA2M) to predict BrS with spontaneous ventricular tachycardia/fibrillation (VT/VF) as outcomes based on specific ECG markers. A total of 191 adult patients with BrS from three centres (Germany, Greece and Hong Kong) were included for analysis. Depolarization and repolarization ECG markers were measured from the right precordial leads (V1 to V3). The proposed GA2M-based risk prediction model successfully identified a set of risk factors and their pairwise interactions in addition to the dispersion of repolarization/total repolarization (Tpeak- Tend x mean QT)). The model outperformed the baseline logistic model based on the same set of ECG measurements. In conclusion, the inclusion of pairwise interactions improved predictive performance and enabled more effective risk stratification in BrS.

AB - Electrocardiographic (ECG) indices were used for risk stratification in Brugada syndrome (BrS). However, nonlinear interactions between risk factors were ignored. Therefore, we adapted a generalized additive model with pair-wise interactions (GA2M) to predict BrS with spontaneous ventricular tachycardia/fibrillation (VT/VF) as outcomes based on specific ECG markers. A total of 191 adult patients with BrS from three centres (Germany, Greece and Hong Kong) were included for analysis. Depolarization and repolarization ECG markers were measured from the right precordial leads (V1 to V3). The proposed GA2M-based risk prediction model successfully identified a set of risk factors and their pairwise interactions in addition to the dispersion of repolarization/total repolarization (Tpeak- Tend x mean QT)). The model outperformed the baseline logistic model based on the same set of ECG measurements. In conclusion, the inclusion of pairwise interactions improved predictive performance and enabled more effective risk stratification in BrS.

UR - https://www.scopus.com/pages/publications/85124718332

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85124718332&origin=recordpage

U2 - 10.23919/CinC53138.2021.9662913

DO - 10.23919/CinC53138.2021.9662913

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781665467216

T3 - Computing in Cardiology

BT - 2021 Computing in Cardiology (CinC)

PB - IEEE

T2 - 48th Computing in Cardiology Conference (CinC 2021)

Y2 - 13 September 2021 through 15 September 2021

ER -

Pairwise Feature Interactions to Predict Arrhythmic Risk of Brugada Syndrome

Abstract

Publication series

Conference

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this