Benchmarking beat classification algorithms

Author

Nabney, IT ; Evans, DJ ; Tenner, J. ; Gamlyn, L.

Author_Institution

Cardionetics Inst. of Bioinformatics, Aston Univ., Birmingham, UK

fYear

2001

fDate

2001

Firstpage

529

Lastpage

532

Abstract

This study compares the accuracy of a range of advanced and classical pattern recognition algorithms for beat and arrhythmia classification from ECG using a principled statistical framework. These are to be used in an application where no patient-specific adaptation of the features or of the model is possible, which means that models must be able to generalise across subjects. Our results demonstrate that non-linear classification models offer significant advantages in ECG beat classification and that, with a principled approach to feature selection, pre-processing and model development, it is possible to get robust inter-subject generalisation even on ambulatory data

Keywords

electrocardiography; feature extraction; generalisation (artificial intelligence); medical signal processing; signal classification; statistics; ECG; ambulatory data; arrhythmia classification; beat classification algorithm benchmarking; feature selection; generalisable models; model development; nonlinear classification models; patient-specific adaptation; pattern recognition algorithm accuracy; pre-processing; principled statistical framework; robust inter-subject generalisation; Cardiology; Classification algorithms; Data visualization; Electrocardiography; Feature extraction; Heart rate variability; Histograms; Pattern recognition; Principal component analysis; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Computers in Cardiology 2001

Conference_Location

Rotterdam

ISSN

0276-6547

Print_ISBN

0-7803-7266-2

Type

conf

DOI

10.1109/CIC.2001.977709

Filename

977709