Inter-patient ECG classification using deep convolutional neural networks

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

2 Citations (Scopus)

Abstract

In this paper we present fully automatic interpatient electrocardiogram (ECG) signal classification method using deep convolutional neural networks (CNN). ECG is simple and non-invasive way to measure the electric activity of the heart and it provides valuable information about the condition of the heart. It is widely utilized for detecting different abnormalities in heartbeat. Identifying and classification abnormalities is timeconsuming, because it often requires analyzing each heartbeat of the ECG recording. Therefore, automatic classification of the arrhythmias using machine-learning technologies can bring various benefits. In this paper, we are focusing on inter-patient arrhythmia classification, where separate patient data is used in training and test phase. This method is more realistic in clinical environment, where trained model needs to operate with patients, whose ECG data was not available during the training phase. Our proposed method gives 92% sensitivity, 97% positive predictivity and 23% false positive rate for normal heartbeats. For supraventricular ectopic beat, our approach gives 62% sensitivity, 56% positive predictivity and 2% false positive rate. For ventricular ectopic beat, our method gives 89% sensitivity, 51% positive predictivity and 6% false positive rate. These results from our fully automatic feature learning approach are on par with solutions that require manual feature engineering.

Original languageEnglish
Title of host publication2018 21st Euromicro Conference on Digital System Design
EditorsNikos Konofaos, Martin Novotny, Amund Skavhaug
PublisherInstitute of Electrical and Electronic Engineers IEEE
Pages421-425
Number of pages5
ISBN (Electronic)978-1-5386-7377-5
ISBN (Print)978-1-5386-7378-2
DOIs
Publication statusPublished - 15 Oct 2018
MoE publication typeNot Eligible
EventEuromicro Conference on Digital System Design, DSD 2018 - Prague, Czech Republic
Duration: 29 Aug 201831 Aug 2018
Conference number: 21

Conference

ConferenceEuromicro Conference on Digital System Design, DSD 2018
Abbreviated titleDSD 2018
CountryCzech Republic
CityPrague
Period29/08/1831/08/18

Fingerprint

Electrocardiography
Neural networks
Learning systems

Keywords

  • Biomedical signals
  • Convolutional neural networks (CNN)
  • Electrocardiogram (ECG)
  • Inter-patient paradigm

Cite this

Takalo-Mattila, J., Kiljander, J., & Soininen, J. P. (2018). Inter-patient ECG classification using deep convolutional neural networks. In N. Konofaos, M. Novotny, & A. Skavhaug (Eds.), 2018 21st Euromicro Conference on Digital System Design (pp. 421-425). [8491848] Institute of Electrical and Electronic Engineers IEEE. https://doi.org/10.1109/DSD.2018.00077
Takalo-Mattila, Janne ; Kiljander, Jussi ; Soininen, Juha Pekka. / Inter-patient ECG classification using deep convolutional neural networks. 2018 21st Euromicro Conference on Digital System Design. editor / Nikos Konofaos ; Martin Novotny ; Amund Skavhaug. Institute of Electrical and Electronic Engineers IEEE, 2018. pp. 421-425
@inproceedings{d7d3e48cfab74a1eaba94028ed1fa633,
title = "Inter-patient ECG classification using deep convolutional neural networks",
abstract = "In this paper we present fully automatic interpatient electrocardiogram (ECG) signal classification method using deep convolutional neural networks (CNN). ECG is simple and non-invasive way to measure the electric activity of the heart and it provides valuable information about the condition of the heart. It is widely utilized for detecting different abnormalities in heartbeat. Identifying and classification abnormalities is timeconsuming, because it often requires analyzing each heartbeat of the ECG recording. Therefore, automatic classification of the arrhythmias using machine-learning technologies can bring various benefits. In this paper, we are focusing on inter-patient arrhythmia classification, where separate patient data is used in training and test phase. This method is more realistic in clinical environment, where trained model needs to operate with patients, whose ECG data was not available during the training phase. Our proposed method gives 92{\%} sensitivity, 97{\%} positive predictivity and 23{\%} false positive rate for normal heartbeats. For supraventricular ectopic beat, our approach gives 62{\%} sensitivity, 56{\%} positive predictivity and 2{\%} false positive rate. For ventricular ectopic beat, our method gives 89{\%} sensitivity, 51{\%} positive predictivity and 6{\%} false positive rate. These results from our fully automatic feature learning approach are on par with solutions that require manual feature engineering.",
keywords = "Biomedical signals, Convolutional neural networks (CNN), Electrocardiogram (ECG), Inter-patient paradigm",
author = "Janne Takalo-Mattila and Jussi Kiljander and Soininen, {Juha Pekka}",
year = "2018",
month = "10",
day = "15",
doi = "10.1109/DSD.2018.00077",
language = "English",
isbn = "978-1-5386-7378-2",
pages = "421--425",
editor = "Nikos Konofaos and Martin Novotny and Amund Skavhaug",
booktitle = "2018 21st Euromicro Conference on Digital System Design",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
address = "United States",

}

Takalo-Mattila, J, Kiljander, J & Soininen, JP 2018, Inter-patient ECG classification using deep convolutional neural networks. in N Konofaos, M Novotny & A Skavhaug (eds), 2018 21st Euromicro Conference on Digital System Design., 8491848, Institute of Electrical and Electronic Engineers IEEE, pp. 421-425, Euromicro Conference on Digital System Design, DSD 2018, Prague, Czech Republic, 29/08/18. https://doi.org/10.1109/DSD.2018.00077

Inter-patient ECG classification using deep convolutional neural networks. / Takalo-Mattila, Janne; Kiljander, Jussi; Soininen, Juha Pekka.

2018 21st Euromicro Conference on Digital System Design. ed. / Nikos Konofaos; Martin Novotny; Amund Skavhaug. Institute of Electrical and Electronic Engineers IEEE, 2018. p. 421-425 8491848.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Inter-patient ECG classification using deep convolutional neural networks

AU - Takalo-Mattila, Janne

AU - Kiljander, Jussi

AU - Soininen, Juha Pekka

PY - 2018/10/15

Y1 - 2018/10/15

N2 - In this paper we present fully automatic interpatient electrocardiogram (ECG) signal classification method using deep convolutional neural networks (CNN). ECG is simple and non-invasive way to measure the electric activity of the heart and it provides valuable information about the condition of the heart. It is widely utilized for detecting different abnormalities in heartbeat. Identifying and classification abnormalities is timeconsuming, because it often requires analyzing each heartbeat of the ECG recording. Therefore, automatic classification of the arrhythmias using machine-learning technologies can bring various benefits. In this paper, we are focusing on inter-patient arrhythmia classification, where separate patient data is used in training and test phase. This method is more realistic in clinical environment, where trained model needs to operate with patients, whose ECG data was not available during the training phase. Our proposed method gives 92% sensitivity, 97% positive predictivity and 23% false positive rate for normal heartbeats. For supraventricular ectopic beat, our approach gives 62% sensitivity, 56% positive predictivity and 2% false positive rate. For ventricular ectopic beat, our method gives 89% sensitivity, 51% positive predictivity and 6% false positive rate. These results from our fully automatic feature learning approach are on par with solutions that require manual feature engineering.

AB - In this paper we present fully automatic interpatient electrocardiogram (ECG) signal classification method using deep convolutional neural networks (CNN). ECG is simple and non-invasive way to measure the electric activity of the heart and it provides valuable information about the condition of the heart. It is widely utilized for detecting different abnormalities in heartbeat. Identifying and classification abnormalities is timeconsuming, because it often requires analyzing each heartbeat of the ECG recording. Therefore, automatic classification of the arrhythmias using machine-learning technologies can bring various benefits. In this paper, we are focusing on inter-patient arrhythmia classification, where separate patient data is used in training and test phase. This method is more realistic in clinical environment, where trained model needs to operate with patients, whose ECG data was not available during the training phase. Our proposed method gives 92% sensitivity, 97% positive predictivity and 23% false positive rate for normal heartbeats. For supraventricular ectopic beat, our approach gives 62% sensitivity, 56% positive predictivity and 2% false positive rate. For ventricular ectopic beat, our method gives 89% sensitivity, 51% positive predictivity and 6% false positive rate. These results from our fully automatic feature learning approach are on par with solutions that require manual feature engineering.

KW - Biomedical signals

KW - Convolutional neural networks (CNN)

KW - Electrocardiogram (ECG)

KW - Inter-patient paradigm

UR - http://www.scopus.com/inward/record.url?scp=85056487499&partnerID=8YFLogxK

U2 - 10.1109/DSD.2018.00077

DO - 10.1109/DSD.2018.00077

M3 - Conference article in proceedings

SN - 978-1-5386-7378-2

SP - 421

EP - 425

BT - 2018 21st Euromicro Conference on Digital System Design

A2 - Konofaos, Nikos

A2 - Novotny, Martin

A2 - Skavhaug, Amund

PB - Institute of Electrical and Electronic Engineers IEEE

ER -

Takalo-Mattila J, Kiljander J, Soininen JP. Inter-patient ECG classification using deep convolutional neural networks. In Konofaos N, Novotny M, Skavhaug A, editors, 2018 21st Euromicro Conference on Digital System Design. Institute of Electrical and Electronic Engineers IEEE. 2018. p. 421-425. 8491848 https://doi.org/10.1109/DSD.2018.00077