Computer Aided detection for fibrillations and flutters using deep convolutional neural network

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62690094%3A18450%2F19%3A50015691" target="_blank" >RIV/62690094:18450/19:50015691 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.sciencedirect.com/science/article/pii/S0020025519301884" target="_blank" >https://www.sciencedirect.com/science/article/pii/S0020025519301884</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.ins.2019.02.065" target="_blank" >10.1016/j.ins.2019.02.065</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Computer Aided detection for fibrillations and flutters using deep convolutional neural network

Popis výsledku v původním jazyce

Fibrillations and flutters are serious diseases influence the normal functioning of the heart. Among the most frequently occurring heart disorders belong atrial fibrillation (Afib), atrial flutter (Afl), and ventricular fibrillation (Vfib). Nowadays, heart failures are mostly detected by electrocardiogram (ECG) device by examining the signal transferred from electrodes placed on the human body to the output display. The signal is examined by professional health personnel, who are looking for an obvious pattern representing the normal or abnormal rhythm of the heart. Nevertheless, information from ECG can be distorted by noise on data transmission. Moreover,problematic pattern does not have to be so much different from normal and it can be difficult to recognize them just by human eye even by an expert in the field. An automated computer-aided diagnosis (CAD) is an approach to make decision support for elimination of these lacks. For early diagnosis, CAD tool should work in like real-time system without big time consuming and dependency on data and measuring differences of each device. This paper proposes a novel approach of a CAD system to the detection of fibrillations and flutters by our 8-layer deep convolutional neural network. Proposed model requires only basic data normalization without pre-processing and feature extraction from raw ECG samples. We have achieved the accuracy, specificity, and sensitivity of 98.45%, 99.27%, and 99.87% respectively. Designed system can be directly implemented like decision support system in clinical environment.

Název v anglickém jazyce

Computer Aided detection for fibrillations and flutters using deep convolutional neural network

Popis výsledku anglicky

Fibrillations and flutters are serious diseases influence the normal functioning of the heart. Among the most frequently occurring heart disorders belong atrial fibrillation (Afib), atrial flutter (Afl), and ventricular fibrillation (Vfib). Nowadays, heart failures are mostly detected by electrocardiogram (ECG) device by examining the signal transferred from electrodes placed on the human body to the output display. The signal is examined by professional health personnel, who are looking for an obvious pattern representing the normal or abnormal rhythm of the heart. Nevertheless, information from ECG can be distorted by noise on data transmission. Moreover,problematic pattern does not have to be so much different from normal and it can be difficult to recognize them just by human eye even by an expert in the field. An automated computer-aided diagnosis (CAD) is an approach to make decision support for elimination of these lacks. For early diagnosis, CAD tool should work in like real-time system without big time consuming and dependency on data and measuring differences of each device. This paper proposes a novel approach of a CAD system to the detection of fibrillations and flutters by our 8-layer deep convolutional neural network. Proposed model requires only basic data normalization without pre-processing and feature extraction from raw ECG samples. We have achieved the accuracy, specificity, and sensitivity of 98.45%, 99.27%, and 99.87% respectively. Designed system can be directly implemented like decision support system in clinical environment.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

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

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2019

Kód důvěrnosti údajů

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

Název periodika

Information sciences

ISSN

0020-0255

e-ISSN

—

Svazek periodika

486

Číslo periodika v rámci svazku

červen

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

9

Strana od-do

231-239

Kód UT WoS článku

000464301300016

EID výsledku v databázi Scopus

2-s2.0-85062047707