Journal of Clinical Engineering · 1984 · 60 citations · 0 references
EngineeringBiomedical EngineeringCardiologyBiomedical Signal ProcessingElectrophysiologyTime Series AnalysisCardiac Signal ProcessingCardiovascular MonitoringElectrocardiogram AnalysisElectrocardiographyPhysiological Signal AcquisitionTemplate MatchingArrhythmia DetectionArrhythmia RecognitionCardiac Monitoring SystemsEcg Anomaly DetectionEcg Fiducial Point DetectionMulti-lead Ecg AnalysisQrs Complex DetectionTemplate-based Matching
A microcomputer system which detects QRS complexes in real-time is presented. The system is unique in that it implements a template matching algorithm using moving-window correlation on a microcomputer. Incoming ECG data are continuously correlated with multiple previously stored templates. Normal or arrhythmic QRS complexes are detected whenever correlation with one of the templates exceeds a preset threshold. Use of moving-window correlation eliminates the problems often encountered when performing correlation based upon fiducial point alignment. Viability of the technique has been tested by processing recorded ECG data taken from six human patients.