Electrocardiographic Discrimination of Long QT Syndrome Genotypes: A Comparative Analysis and Machine Learning Approach

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21460%2F25%3A00383083" target="_blank" >RIV/68407700:21460/25:00383083 - isvavai.cz</a>

Alternative codes found

RIV/68407700:21730/25:00383083

Result on the web

<a href="https://doi.org/10.3390/s25072253" target="_blank" >https://doi.org/10.3390/s25072253</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.3390/s25072253" target="_blank" >10.3390/s25072253</a>

Result language

angličtina

Original language name

Electrocardiographic Discrimination of Long QT Syndrome Genotypes: A Comparative Analysis and Machine Learning Approach

Original language description

Long QT syndrome (LQTS) presents a group of inheritable channelopathies with prolonged ventricular repolarization, leading to syncope, ventricular tachycardia, and sudden death. Differentiating LQTS genotypes is crucial for targeted management and treatment, yet conventional genetic testing remains costly and time-consuming. This study aims to improve the distinction between LQTS genotypes, particularly LQT3, through a novel electrocardiogram (ECG)-based approach. Patients with LQT3 are at elevated risk due to arrhythmia triggers associated with rest and sleep. Employing a database of genotyped long QT syndrome E-HOL-03-0480-013 ECG signals, we introduced two innovative parameterization techniques—area under the ECG curve and wave transformation into the unit circle—to classify LQT3 against LQT1 and LQT2 genotypes. Our methodology utilized single-lead ECG data with a 200 Hz sampling frequency. The support vector machine (SVM) model demonstrated the ability to discriminate LQT3 with a recall of 90% and a precision of 81%, achieving an F1-score of 0.85. This parameterization offers a potential substitute for genetic testing and is practical for low frequencies. These single-lead ECG data could enhance smartwatches’ functionality and similar cardiovascular monitoring applications. The results underscore the viability of ECG morphology-based genotype classification, promising a significant step towards streamlined diagnosis and improved patient care in LQTS.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

R - Projekt Ramcoveho programu EK

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Name of the periodical

Sensors

ISSN

1424-8220

e-ISSN

1424-8220

Volume of the periodical

25

Issue of the periodical within the volume

7

Country of publishing house

CH - SWITZERLAND

Number of pages

15

Pages from-to

—

UT code for WoS article

001465613900001

EID of the result in the Scopus database

2-s2.0-105002452853