Shape Analysis of Consecutive Beats May Help in the Automated Detection of Atrial Fibrillation

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68081731%3A_____%2F18%3A00509016" target="_blank" >RIV/68081731:_____/18:00509016 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.22489/CinC.2018.036" target="_blank" >http://dx.doi.org/10.22489/CinC.2018.036</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.22489/CinC.2018.036" target="_blank" >10.22489/CinC.2018.036</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Shape Analysis of Consecutive Beats May Help in the Automated Detection of Atrial Fibrillation

Popis výsledku v původním jazyce

Background: Atrial fibrillation (AF) is associated with a higher risk of heart failure or death. AF may be episodic and patients with suspected AF are equipped with Holter ECG devices for several days. However, automated detection of AF in an ECG signal remains problematic, as was shown by the results of the PhysioNet Challenge 2017. Here, we introduce a simple yet robust logistic regression model for AF detection. nMethod: The detrended signal is filtered (1-35 Hz) and normalized. QRS detection based on envelograms (10-35 Hz) reveals QRS complexes. Five features are exfracted from the ECG signal describing RR stability as well as the shape stability of areas preceding QRS complexes. Features were exfracted for 1,517 recordings from the PhysioNet Challenge 2017 public dataset (758 AF recordings and 759 recordings with normal rhythm, other arrhythmia or noisy signal). The recordings were split in a 70/30 % ratio for the purposes of training and testing. nResults: The results showed a sensitivity and specificity of 93 % and 90 %, respectively (AUC 0.96). The presented model was also tested on the MIT-AFDB public database, showing sensitivity and specificity of 89 % and 88 %, respectively. However, tests on an independent private dataset revealed lower specificity when pathologies which are not widely present in the training dataset are common in the tested ECG signal.

Název v anglickém jazyce

Shape Analysis of Consecutive Beats May Help in the Automated Detection of Atrial Fibrillation

Popis výsledku anglicky

Background: Atrial fibrillation (AF) is associated with a higher risk of heart failure or death. AF may be episodic and patients with suspected AF are equipped with Holter ECG devices for several days. However, automated detection of AF in an ECG signal remains problematic, as was shown by the results of the PhysioNet Challenge 2017. Here, we introduce a simple yet robust logistic regression model for AF detection. nMethod: The detrended signal is filtered (1-35 Hz) and normalized. QRS detection based on envelograms (10-35 Hz) reveals QRS complexes. Five features are exfracted from the ECG signal describing RR stability as well as the shape stability of areas preceding QRS complexes. Features were exfracted for 1,517 recordings from the PhysioNet Challenge 2017 public dataset (758 AF recordings and 759 recordings with normal rhythm, other arrhythmia or noisy signal). The recordings were split in a 70/30 % ratio for the purposes of training and testing. nResults: The results showed a sensitivity and specificity of 93 % and 90 %, respectively (AUC 0.96). The presented model was also tested on the MIT-AFDB public database, showing sensitivity and specificity of 89 % and 88 %, respectively. However, tests on an independent private dataset revealed lower specificity when pathologies which are not widely present in the training dataset are common in the tested ECG signal.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20601 - Medical engineering

Projekt

<a href="/cs/project/LO1212" target="_blank" >LO1212: ALISI - Centrum pokročilých diagnostických metod a technologií</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2018

Kód důvěrnosti údajů

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

Název statě ve sborníku

Computing in Cardiology 2018

ISBN

—

ISSN

2325-887X

e-ISSN

—

Počet stran výsledku

4

Strana od-do

8743764

Název nakladatele

IEEE

Místo vydání

New York

Místo konání akce

Maastricht

Datum konání akce

23. 9. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000482598700075