The most widely used signal in clinical practice is the electrocardiogram (ECG). ECG conveys information regarding the electrical function of the heart, by altering the shape of its constituent waves, namely the P, QRS, and T waves. Thus, the required tasks of ECG processing are the reliable recognition of these waves, and the accurate measurement of clinically important parameters measured from the temporal distribution of the ECG constituent waves. The purpose of this paper is the classification of ventricular ectopic beats (VEB's). This research includes noise handling, feature extraction, and neural classification, all integrated in a three-stage procedure. Thirty features extracted from the morphology of the QRS segment, are reduced to seven coefficients using principal component analysis (PCA) and two coefficients using linear predictive coding (LPC) technique in addition to two other temporal parameters were used separately as the input of two neural network classifiers. The neural classifiers were tested on the MIT-BIH database and high scores were obtained for both sensitivity and specificity (84.88% and 91.92% respectively using ACP technique, and 76.17% and 88.95% using LPC method). This study confirms the power of artificial neural networks in the classification of normal and abnormal VEB beats. Clinical use of this method, however, still requires further investigation.
Navigation Security Module with Real-Time Voice Command Recognition System