Disease Detection by Feature Extraction of ECG Signal based on ANFIS
Publication Date : 13/10/2017
Electrocardiogram signals are used to determine the operation of heart. Due to their variability, it is quite difficult to examine them directly. Considering this fact, several techniques have been developed by researchers to understand the signals by extracting features from them. The defined wavelet techniques can extract the features but the accuracy in detecting disease has decreased. In order to enhance the classification process of extracted features, ANFIS structure has been introduced in this paper. The proposed ANFIS structure has the capability of training and then evaluating the ECG signals to determine the person with their health status. The proposed algorithm evaluates two different types of diseases i.e. bradycardia and tachycardia. For the experimental analysis total 27 signals have been considered which have been taken from physiobank databases website. From the simulation analysis, it has been concluded that the proposed technique outperforms in comparison with traditional wavelet technique. The performance parameters which have been used to ensure the performance of the individual techniques are Accuracy, Mean Square Error and Root Mean Square Error. Consequently, the proposed technique stands out with 85% accuracy in comparison with traditional technique which relies at 70% only.
No. of Downloads :