Hardware Optimization & Cad System of an Electrocardiograph Circuit for Lead II

Abstract

This thesis provides electronic implementation of electrocardiogram (ECG) circuit by using instrumentation amplifier as bio-potential amplifier in such a manner that it reduces noise, common voltage, DC offset value and RF interference from the existing circuit. Electrocardiogram (ECG) signal's noise reduction is critical for ECG automatic diagnosis and analysis, and the de-noising result directly affects the accuracy of ECG parameter extraction even the patients' illness diagnosis and analysis. Noise and common voltage can be removed from ECG using driven right leg circuit or by using isolator circuit. DC offset can be removed by using integrator as feedback. By using filters, we can reduce RF interference. In our project, we have used instrumentation amplifier as a bio-potential amplifier. We have also made a CAD system which helps the doctor in analyzing whether patient waveform is normal or abnormal. We have used the knowledge of automatic classification of ECG signals with feature selection and extraction techniques. 83 samples of 12 lead systems were collected from various doctors out of whom we have used the lead 2 samples of the ECG waveforms. . In the feature extraction module, for extracting the features morphological feature extraction method was employed. In the classification module we used different classifiers like k-NN, SVM & SSVM. In this thesis, we have accurately classified and differentiated normal and abnormal waveforms. Experiments reveal that the OCA is 71.43%