Interpretable Assessment of ST-Segment Deviation in ECG Time Series

Israel Campero Jurado; Andrejs Fedjajevs; Joaquin Vanschoren; Aarnout Brombacher

doi:10.3390/s22134919

Interpretable Assessment of ST-Segment Deviation in ECG Time Series

Israel Campero Jurado (Corresponding author)
, Andrejs Fedjajevs
, Joaquin Vanschoren
, Aarnout Brombacher

Research output: Contribution to journal › Article › Academic › peer-review

10 Citations (Scopus)

163 Downloads (Pure)

Abstract

Nowadays, even with all the tremendous advances in medicine and health protocols, cardiovascular diseases (CVD) continue to be one of the major causes of death. In the present work, we focus on a specific abnormality: ST-segment deviation, which occurs regularly in highperformance athletes and elderly people, serving as a myocardial infarction (MI) indicator. It is usually diagnosed manually by experts, through visual interpretation of the printed electrocardiography (ECG) signal. We propose a methodology to detect ST-segment deviation and quantify its scale up to 1 mV by extracting statistical, point-to-point beat characteristics and signal quality indexes (SQIs) from single-lead ECG.We do so by applying automated machine learning methods to find the best hyperparameter configuration for classification and regression models. For validation of our method, we use the ST-T database from Physionet; the results show that our method obtains 98.30% accuracy in the case of a multiclass problem and 99.87% accuracy in the case of binarization.

Original language	English
Article number	4919
Number of pages	24
Journal	Sensors
Volume	22
Issue number	13
DOIs	https://doi.org/10.3390/s22134919
Publication status	Published - 1 Jul 2022

Bibliographical note

Funding Information:
Funding: We are grateful for the support and funding of this work to ITEA INNO4HEALTH 19008 Project. This work made use of the Dutch national e-infrastructure with the support of the SURF Cooperative using grant no. EINF-2491. This research was partially supported by TAILOR, a project funded by EU Horizon 2020 research and innovation programme under GA No 952215.

Funding

Funding: We are grateful for the support and funding of this work to ITEA INNO4HEALTH 19008 Project. This work made use of the Dutch national e-infrastructure with the support of the SURF Cooperative using grant no. EINF-2491. This research was partially supported by TAILOR, a project funded by EU Horizon 2020 research and innovation programme under GA No 952215.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

automated machine learning
electrocardiogram
hyperparameter optimization
myocardial infarction
ST-segment
time series
Myocardial Infarction/diagnosis
Time Factors
Humans
Electrocardiography
Aged

Access to Document

10.3390/s22134919

pdfFinal published version, 50.5 MBLicence: CC BY

Cite this

@article{eddbc328b5f3474b8b2e0d35ab384872,

title = "Interpretable Assessment of ST-Segment Deviation in ECG Time Series",

abstract = "Nowadays, even with all the tremendous advances in medicine and health protocols, cardiovascular diseases (CVD) continue to be one of the major causes of death. In the present work, we focus on a specific abnormality: ST-segment deviation, which occurs regularly in highperformance athletes and elderly people, serving as a myocardial infarction (MI) indicator. It is usually diagnosed manually by experts, through visual interpretation of the printed electrocardiography (ECG) signal. We propose a methodology to detect ST-segment deviation and quantify its scale up to 1 mV by extracting statistical, point-to-point beat characteristics and signal quality indexes (SQIs) from single-lead ECG.We do so by applying automated machine learning methods to find the best hyperparameter configuration for classification and regression models. For validation of our method, we use the ST-T database from Physionet; the results show that our method obtains 98.30\% accuracy in the case of a multiclass problem and 99.87\% accuracy in the case of binarization.",

keywords = "automated machine learning, electrocardiogram, hyperparameter optimization, myocardial infarction, ST-segment, time series, Myocardial Infarction/diagnosis, Time Factors, Humans, Electrocardiography, Aged",

author = "Jurado, \{Israel Campero\} and Andrejs Fedjajevs and Joaquin Vanschoren and Aarnout Brombacher",

note = "Funding Information: Funding: We are grateful for the support and funding of this work to ITEA INNO4HEALTH 19008 Project. This work made use of the Dutch national e-infrastructure with the support of the SURF Cooperative using grant no. EINF-2491. This research was partially supported by TAILOR, a project funded by EU Horizon 2020 research and innovation programme under GA No 952215. ",

year = "2022",

month = jul,

day = "1",

doi = "10.3390/s22134919",

language = "English",

volume = "22",

journal = "Sensors",

issn = "1424-8220",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "13",

}

TY - JOUR

T1 - Interpretable Assessment of ST-Segment Deviation in ECG Time Series

AU - Jurado, Israel Campero

AU - Fedjajevs, Andrejs

AU - Vanschoren, Joaquin

AU - Brombacher, Aarnout

N1 - Funding Information: Funding: We are grateful for the support and funding of this work to ITEA INNO4HEALTH 19008 Project. This work made use of the Dutch national e-infrastructure with the support of the SURF Cooperative using grant no. EINF-2491. This research was partially supported by TAILOR, a project funded by EU Horizon 2020 research and innovation programme under GA No 952215.

PY - 2022/7/1

Y1 - 2022/7/1

N2 - Nowadays, even with all the tremendous advances in medicine and health protocols, cardiovascular diseases (CVD) continue to be one of the major causes of death. In the present work, we focus on a specific abnormality: ST-segment deviation, which occurs regularly in highperformance athletes and elderly people, serving as a myocardial infarction (MI) indicator. It is usually diagnosed manually by experts, through visual interpretation of the printed electrocardiography (ECG) signal. We propose a methodology to detect ST-segment deviation and quantify its scale up to 1 mV by extracting statistical, point-to-point beat characteristics and signal quality indexes (SQIs) from single-lead ECG.We do so by applying automated machine learning methods to find the best hyperparameter configuration for classification and regression models. For validation of our method, we use the ST-T database from Physionet; the results show that our method obtains 98.30% accuracy in the case of a multiclass problem and 99.87% accuracy in the case of binarization.

AB - Nowadays, even with all the tremendous advances in medicine and health protocols, cardiovascular diseases (CVD) continue to be one of the major causes of death. In the present work, we focus on a specific abnormality: ST-segment deviation, which occurs regularly in highperformance athletes and elderly people, serving as a myocardial infarction (MI) indicator. It is usually diagnosed manually by experts, through visual interpretation of the printed electrocardiography (ECG) signal. We propose a methodology to detect ST-segment deviation and quantify its scale up to 1 mV by extracting statistical, point-to-point beat characteristics and signal quality indexes (SQIs) from single-lead ECG.We do so by applying automated machine learning methods to find the best hyperparameter configuration for classification and regression models. For validation of our method, we use the ST-T database from Physionet; the results show that our method obtains 98.30% accuracy in the case of a multiclass problem and 99.87% accuracy in the case of binarization.

KW - automated machine learning

KW - electrocardiogram

KW - hyperparameter optimization

KW - myocardial infarction

KW - ST-segment

KW - time series

KW - Myocardial Infarction/diagnosis

KW - Time Factors

KW - Humans

KW - Electrocardiography

KW - Aged

UR - https://www.scopus.com/pages/publications/85133678314

U2 - 10.3390/s22134919

DO - 10.3390/s22134919

M3 - Article

C2 - 35808421

AN - SCOPUS:85133678314

SN - 1424-8220

VL - 22

JO - Sensors

JF - Sensors

IS - 13

M1 - 4919

ER -

Interpretable Assessment of ST-Segment Deviation in ECG Time Series

Abstract

Bibliographical note

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this