Abstract
This paper illustrates the use of a combined neural network model based on Stacked Generalization method for classification of electrocardiogram (ECG) beats. In conventional Stacked Generalization method, the combiner learns to map the base classifiers' outputs to the target data. We claim adding the input pattern to the base classifiers' outputs helps the combiner to obtain knowledge about the input space and as the result, performs better on the same task. Experimental results support our claim that the additional knowledge according to the input space, improves the performance of the proposed method which is called Modified Stacked Generalization. In particular, for classification of 14966 ECG beats that were not previously seen during training phase, the Modified Stacked Generalization method reduced the error rate for 12.41% in comparison with the best of ten popular classifier fusion methods including Max, Min, Average, Product, Majority Voting, Borda Count, Decision Templates, Weighted Averaging based on Particle Swarm Optimization and Stacked Generalization.
Highlights
Accurate and computationally efficient means of classifying electrocardiography (ECG) arrhythmias has been the subject of considerable research effort in recent years
Among the various abnormalities related to functioning of the human heart, premature ventricular contraction (PVC) is one the most important arrhythmias
Unlike the Stacked Generalization method, in this paper we propose to feed the combiner with both the input sample and outputs of the base classifiers simultaneously
Summary
Accurate and computationally efficient means of classifying electrocardiography (ECG) arrhythmias has been the subject of considerable research effort in recent years. The results of the studies have demonstrated that the Wavelet Transformation is the most promising method to extract features from the ECG signals [2] [5,6] [10]. Wavelet coefficients obtained from the decomposition process are considered as the filtered signal in the sub bands Features extracted from these coefficients can efficiently represent the characteristics of the original signal in different details [20,21]. Combining classifiers based on the fusion of outputs of a set of different classifiers has been developed as a method for improving the recognition rate of classification problems [43,44,45]. Experimental results indicate that our proposed combining method performs better than other combining methods
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call