Detection of abnormality in the electrocardiogram without prior knowledge by using the quantisation error of a self-organising map, tested on the European ischaemia database

Most systems for the automatic detection of abnormalities in the ECG requir e prior knowledge of normal and abnormal ECG morphology from pre-existing d atabases. An automated system for abnormality detection has been developed based on learning normal ECG morphology directly from the patient The quant isation error from a self-organising map 'learns' the form of the patient's ECG and detects any change in its morphology. The system does not require prior knowledge of normal and abnormal morphologies. It was tested on 76 re cords from the European Society of Cardiology database and detected 40.5% o f those first abnormalities declared by the database to be ischaemic. The s ystem also responded to abnormalities arising from ECG axis changes and slo w baseline drifts and revealed that ischaemic episodes are often followed b y long-term changes in ECG morphology.