| Cardiovascular disease is one of the dangerous diseases that seriously hazard thehealth of human life, while Electrocardiogram (ECG) is an effective method used inclinic to detect cardiovascular diseases. Thus, automatically and correctly analyzingECG with computer program has been a hot issue among national and internationalscholars for many years.Premature Ventricular Contraction (PVC) is the most familiar arrhythmia.According to the characters of the electrocardiogram of PVC, paper designed andrealized the detection of PVC waves by two methods. Firstly, founded on the relatedcharacters of R and S waves of the electrocardiogram of PVC, paper proposed the R-Sdistance and amplitude detection method. On the bases of standards that used in mostpresent cardiogram monitors, paper mainly discussed another PVC detectionmethod-Linear Nerve Network (LNN) method. This method employed four charactersof the QRS waves to realize the detection, including R-R distance, polarity and meanamplitude of R waves, distance of QRS waves. Paper discussed the designing and thetraining of the Nerve Network in detail. The validities of the two methods werecertified by MIT/BIH standard arrhythmia database.The final purpose of the automatic detection of the QRS waves is to substitutedoctors by computer in the clinic classification and diagnosis of QRS waves. Paperintroduced fuzzy set and membership function of fuzzing mathematics to fuzzilycharacterize the slope of the cardiograms, so that the membership grade set and thecharacters of the cardiograms can be acquired. On the bases of membership grade set,paper proposed a method to measure the similarity grade of cardiogram waves andclassify them accordingly. Here paper also used MIT/BIH standard arrhythmiadatabase to test the method. The results showed that the concept of the cardiogramsimilarity grade proposed in the paper is meaningful and that the correspondingclassification method is feasible. |
PDF Full Text Request