Skip to main content
Video s3
    Details
    Poster
    Presenter(s)
    Julien Duforest Headshot
    Display Name
    Julien Duforest
    Affiliation
    Affiliation
    Université Lille1, CNRS, Centrale Lille, Junia, Université Polytechnique Hauts-de-France
    Country
    Country
    France
    Author(s)
    Display Name
    Julien Duforest
    Affiliation
    Affiliation
    Université Lille1, CNRS, Centrale Lille, Junia, Université Polytechnique Hauts-de-France
    Display Name
    Benoit Larras
    Affiliation
    Affiliation
    Junia
    Display Name
    Deepu John
    Affiliation
    Affiliation
    University College Dublin
    Display Name
    Olev Märtens
    Affiliation
    Affiliation
    Tallinn University of Technology
    Display Name
    Antoine Frappé
    Affiliation
    Affiliation
    Univ. Lille
    Abstract

    To prevent cardiovascular diseases (CVD), the electrocardiogram (ECG) of the patient must be recorded for a long period of time. For an effective diagnostic, the recording system must easily adapt to new patients and function for long periods of time. This paper presents the use of slopes in event-driven signal as a feature to detect beats and arrhythmias in electrocardiograms (ECG) with 98.3% accuracy and adaptability to new patients specification.