Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification

Quazi Abidur Rahman, Larisa Tereshchenko, Matthew Kongkatong, Theodore Abraham, M. Roselle Abraham, Hagit Shatkay

    Research output: Contribution to journalArticle

    27 Citations (Scopus)

    Abstract

    Hypertrophic cardiomyopathy (HCM) is a cardiovascular disease where the heart muscle is partially thickened and blood flow is (potentially fatally) obstructed. A test based on electrocardiograms (ECG) that record the heart electrical activity can help in early detection of HCM patients. This paper presents a cardiovascular-patient classifier we developed to identify HCM patients using standard 10-second, 12-lead ECG signals. Patients are classified as having HCM if the majority of their recorded heartbeats are recognized as characteristic of HCM. Thus, the classifier's underlying task is to recognize individual heartbeats segmented from 12-lead ECG signals as HCM beats, where heartbeats from non-HCM cardiovascular patients are used as controls. We extracted 504 morphological and temporal features-both commonly used and newly-developed ones-from ECG signals for heartbeat classification. To assess classification performance, we trained and tested a random forest classifier and a support vector machine classifier using 5-fold cross validation. We also compared the performance of these two classifiers to that obtained by a logistic regression classifier, and the first two methods performed better than logistic regression. The patient-classification precision of random forests and of support vector machine classifiers is close to 0.85. Recall (sensitivity) and specificity are approximately 0.90. We also conducted feature selection experiments by gradually removing the least informative features; the results show that a relatively small subset of 264 highly informative features can achieve performance measures comparable to those achieved by using the complete set of features.

    Original languageEnglish (US)
    Article number7094280
    Pages (from-to)505-512
    Number of pages8
    JournalIEEE Transactions on Nanobioscience
    Volume14
    Issue number5
    DOIs
    StatePublished - Jul 1 2015

    Fingerprint

    Hypertrophic Cardiomyopathy
    Electrocardiography
    Classifiers
    Support vector machines
    Logistic Models
    Logistics
    Lead
    Cardiomyopathies
    Myocardium
    Cardiovascular Diseases
    Muscle
    Feature extraction
    Blood
    Sensitivity and Specificity

    Keywords

    • Electrocardiogram
    • feature selection
    • hypertrophic cardiomyopathy
    • machine learning
    • patient classification

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Biomedical Engineering
    • Biotechnology
    • Computer Science Applications
    • Bioengineering
    • Medicine (miscellaneous)
    • Pharmaceutical Science

    Cite this

    Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification. / Rahman, Quazi Abidur; Tereshchenko, Larisa; Kongkatong, Matthew; Abraham, Theodore; Roselle Abraham, M.; Shatkay, Hagit.

    In: IEEE Transactions on Nanobioscience, Vol. 14, No. 5, 7094280, 01.07.2015, p. 505-512.

    Research output: Contribution to journalArticle

    Rahman, Quazi Abidur ; Tereshchenko, Larisa ; Kongkatong, Matthew ; Abraham, Theodore ; Roselle Abraham, M. ; Shatkay, Hagit. / Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification. In: IEEE Transactions on Nanobioscience. 2015 ; Vol. 14, No. 5. pp. 505-512.
    @article{094d816f47764e7f9cfe4393f137c548,
    title = "Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification",
    abstract = "Hypertrophic cardiomyopathy (HCM) is a cardiovascular disease where the heart muscle is partially thickened and blood flow is (potentially fatally) obstructed. A test based on electrocardiograms (ECG) that record the heart electrical activity can help in early detection of HCM patients. This paper presents a cardiovascular-patient classifier we developed to identify HCM patients using standard 10-second, 12-lead ECG signals. Patients are classified as having HCM if the majority of their recorded heartbeats are recognized as characteristic of HCM. Thus, the classifier's underlying task is to recognize individual heartbeats segmented from 12-lead ECG signals as HCM beats, where heartbeats from non-HCM cardiovascular patients are used as controls. We extracted 504 morphological and temporal features-both commonly used and newly-developed ones-from ECG signals for heartbeat classification. To assess classification performance, we trained and tested a random forest classifier and a support vector machine classifier using 5-fold cross validation. We also compared the performance of these two classifiers to that obtained by a logistic regression classifier, and the first two methods performed better than logistic regression. The patient-classification precision of random forests and of support vector machine classifiers is close to 0.85. Recall (sensitivity) and specificity are approximately 0.90. We also conducted feature selection experiments by gradually removing the least informative features; the results show that a relatively small subset of 264 highly informative features can achieve performance measures comparable to those achieved by using the complete set of features.",
    keywords = "Electrocardiogram, feature selection, hypertrophic cardiomyopathy, machine learning, patient classification",
    author = "Rahman, {Quazi Abidur} and Larisa Tereshchenko and Matthew Kongkatong and Theodore Abraham and {Roselle Abraham}, M. and Hagit Shatkay",
    year = "2015",
    month = "7",
    day = "1",
    doi = "10.1109/TNB.2015.2426213",
    language = "English (US)",
    volume = "14",
    pages = "505--512",
    journal = "IEEE Transactions on Nanobioscience",
    issn = "1536-1241",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "5",

    }

    TY - JOUR

    T1 - Utilizing ECG-Based Heartbeat Classification for Hypertrophic Cardiomyopathy Identification

    AU - Rahman, Quazi Abidur

    AU - Tereshchenko, Larisa

    AU - Kongkatong, Matthew

    AU - Abraham, Theodore

    AU - Roselle Abraham, M.

    AU - Shatkay, Hagit

    PY - 2015/7/1

    Y1 - 2015/7/1

    N2 - Hypertrophic cardiomyopathy (HCM) is a cardiovascular disease where the heart muscle is partially thickened and blood flow is (potentially fatally) obstructed. A test based on electrocardiograms (ECG) that record the heart electrical activity can help in early detection of HCM patients. This paper presents a cardiovascular-patient classifier we developed to identify HCM patients using standard 10-second, 12-lead ECG signals. Patients are classified as having HCM if the majority of their recorded heartbeats are recognized as characteristic of HCM. Thus, the classifier's underlying task is to recognize individual heartbeats segmented from 12-lead ECG signals as HCM beats, where heartbeats from non-HCM cardiovascular patients are used as controls. We extracted 504 morphological and temporal features-both commonly used and newly-developed ones-from ECG signals for heartbeat classification. To assess classification performance, we trained and tested a random forest classifier and a support vector machine classifier using 5-fold cross validation. We also compared the performance of these two classifiers to that obtained by a logistic regression classifier, and the first two methods performed better than logistic regression. The patient-classification precision of random forests and of support vector machine classifiers is close to 0.85. Recall (sensitivity) and specificity are approximately 0.90. We also conducted feature selection experiments by gradually removing the least informative features; the results show that a relatively small subset of 264 highly informative features can achieve performance measures comparable to those achieved by using the complete set of features.

    AB - Hypertrophic cardiomyopathy (HCM) is a cardiovascular disease where the heart muscle is partially thickened and blood flow is (potentially fatally) obstructed. A test based on electrocardiograms (ECG) that record the heart electrical activity can help in early detection of HCM patients. This paper presents a cardiovascular-patient classifier we developed to identify HCM patients using standard 10-second, 12-lead ECG signals. Patients are classified as having HCM if the majority of their recorded heartbeats are recognized as characteristic of HCM. Thus, the classifier's underlying task is to recognize individual heartbeats segmented from 12-lead ECG signals as HCM beats, where heartbeats from non-HCM cardiovascular patients are used as controls. We extracted 504 morphological and temporal features-both commonly used and newly-developed ones-from ECG signals for heartbeat classification. To assess classification performance, we trained and tested a random forest classifier and a support vector machine classifier using 5-fold cross validation. We also compared the performance of these two classifiers to that obtained by a logistic regression classifier, and the first two methods performed better than logistic regression. The patient-classification precision of random forests and of support vector machine classifiers is close to 0.85. Recall (sensitivity) and specificity are approximately 0.90. We also conducted feature selection experiments by gradually removing the least informative features; the results show that a relatively small subset of 264 highly informative features can achieve performance measures comparable to those achieved by using the complete set of features.

    KW - Electrocardiogram

    KW - feature selection

    KW - hypertrophic cardiomyopathy

    KW - machine learning

    KW - patient classification

    UR - http://www.scopus.com/inward/record.url?scp=84939539063&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84939539063&partnerID=8YFLogxK

    U2 - 10.1109/TNB.2015.2426213

    DO - 10.1109/TNB.2015.2426213

    M3 - Article

    C2 - 25915962

    AN - SCOPUS:84939539063

    VL - 14

    SP - 505

    EP - 512

    JO - IEEE Transactions on Nanobioscience

    JF - IEEE Transactions on Nanobioscience

    SN - 1536-1241

    IS - 5

    M1 - 7094280

    ER -