ECG Abnormalities Recognition Using Convolutional Network With Global Skip Connections and Custom Loss Function

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F20%3APU137393" target="_blank" >RIV/00216305:26220/20:PU137393 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9344393" target="_blank" >https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9344393</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.22489/CinC.2020.189" target="_blank" >10.22489/CinC.2020.189</a>

Jazyk výsledku

angličtina

Název v původním jazyce

ECG Abnormalities Recognition Using Convolutional Network With Global Skip Connections and Custom Loss Function

Popis výsledku v původním jazyce

The latest trends in clinical care and telemedicine suggest a demand for a reliable automated electrocardiogram (ECG) signal classification methods. In this paper, we present customized deep learning model for ECG classification as a part of the Physionet/CinC Challenge 2020. The method is based on modified ResNet type convolutional neural network and is capable to automatically recognize 24 cardiac abnormalities using 12-lead ECG. We have adopted several preprocessing and learning techniques including custom tailored loss function, dedicated classification layer and Bayesian threshold optimization which have major positive impact on the model performance. At the official phase of the Challenge, our team BUT-Team - reached a challenge validation score of 0.696, and the full test score of 0.202, placing us 21 out of 40 in the official ranking. This implies that our method performed well on data from the same source (reached first place with validation score), however, it has very poor generalization to data from different sources.

Název v anglickém jazyce

ECG Abnormalities Recognition Using Convolutional Network With Global Skip Connections and Custom Loss Function

Popis výsledku anglicky

The latest trends in clinical care and telemedicine suggest a demand for a reliable automated electrocardiogram (ECG) signal classification methods. In this paper, we present customized deep learning model for ECG classification as a part of the Physionet/CinC Challenge 2020. The method is based on modified ResNet type convolutional neural network and is capable to automatically recognize 24 cardiac abnormalities using 12-lead ECG. We have adopted several preprocessing and learning techniques including custom tailored loss function, dedicated classification layer and Bayesian threshold optimization which have major positive impact on the model performance. At the official phase of the Challenge, our team BUT-Team - reached a challenge validation score of 0.696, and the full test score of 0.202, placing us 21 out of 40 in the official ranking. This implies that our method performed well on data from the same source (reached first place with validation score), however, it has very poor generalization to data from different sources.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

30201 - Cardiac and Cardiovascular systems

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2020

Kód důvěrnosti údajů

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

Název statě ve sborníku

Computing in Cardiology 2020

ISBN

978-1-7281-7382-5

ISSN

2325-8861

e-ISSN

—

Počet stran výsledku

4

Strana od-do

1-4

Název nakladatele

IEEE

Místo vydání

NEW YORK

Místo konání akce

Rimini

Datum konání akce

13. 9. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000657257000244