A Machine Learning Approach to Objective Cardiac Event Detection

Author

Twomey, N. ; Flach, P.A.

Author_Institution

Dept. of Eng., Univ. of Bristol, Bristol, UK

fYear

2014

fDate

2-4 July 2014

Firstpage

519

Lastpage

524

Abstract

This paper presents an automated framework for the detection of the QRS complex from Electrocardiogram (ECG) signals. We introduce an artefact-tolerant pre-processing algorithm which emphasises a number of characteristics of the ECG that are representative of the QRS complex. With this processed ECG signal we train Logistic Regression and Support Vector Machine classification models. With our approach we obtain over 99.7% detection sensitivity and precision on the MIT-BIH database without using supplementary de-noising or pre-emphasis filters.

Keywords

electrocardiography; learning (artificial intelligence); medical signal detection; regression analysis; support vector machines; ECG signals; MIT-BIH database; QRS complex; artefact-tolerant pre-processing algorithm; detection sensitivity; electrocardiogram signals; logistic regression; machine learning approach; objective cardiac event detection; preemphasis filters; supplementary denoising; support vector machine classification models; Databases; Electrocardiography; Mathematical model; Sensitivity; Support vector machines; Training; Pattern recognition; QRS detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Complex, Intelligent and Software Intensive Systems (CISIS), 2014 Eighth International Conference on

Conference_Location

Birmingham

Print_ISBN

978-1-4799-4326-5

Type

conf

DOI

10.1109/CISIS.2014.75

Filename

6915567