Quality Assurance in Cardiac Disease Diagnostic using Computerised Feature Extraction of ECG Signal

摘要

The bio-potentials generated by the muscles of the heart result in an electrical signal called electrocardiogram (ECG). It is one of the most important physiological parameter, which is being extensively used for knowing the state of cardiac patients. Electrocardiograph enables a rapid observation of the behavior of the heart activity in terms of heart rhythm, conduction intervals, morphological aspects of the waves (P,Q,R,S,T) associated to depolarization and repolarization of the auricles and ventricles. Feature extraction of ECG is most essential task in the manual and automated ECG analysis for use in instruments like ECG monitors, Holter tape recorders and scanners, ambulatory ECG recorders and analysers. Reliable detection of the QRS complex in either a normal or an abnormal ECG and its analysis is the first and foremost task in almost every ECG signal analysis system aimed at the diagnostic interpretation of ECG. A good amount of research work has been carried out during the last five decades for the accurate and reliable detection of QRS segment in the ECG signal. In the present paper, a bibliographic survey of different feature extraction methods has been presented. A comparative study of the work carried out in the department of Electrical Engineering of Indian Institute of Technology Roorkee has also been discussed. The element of subjectivity is totally eliminated in cardiac disease diagnostic by using computerised feature extraction of the ECG signal. All modern state-of-art hospials use computer based ECG machines for feature extraction and disease diagnostic with a view to assure quality of their health care services.