Narrow Results

This paper presents a new algorithm for reliable pattern classification in the Electrocardiogram (ECG) based on Support Vector Machine (SVM). Among all ECG components, QRS complex is the most significant feature. Once the positions of the QRS complexes are found, a more detailed examination of the ECG signal can be carried out, in order to study the complete cardiac period. This paper presents SVM as a QRS detector in ECG signal. Two different preprocessing methods are applied for the generation of features. First involves digital filtering to remove base line wander and power line interference while the second involves entropy criterion for feature generation. The processed signal is further used for QRS detection using SVM. This algorithm implements an idea of supervised learning i.e. learning through examples. Algorithm performance was evaluated against the CSE ECG Database. The numerical results indicated that the algorithm finally achieved about 98.7% of the detection rate and also it could function reliably even under the condition of poor signal quality in the ECG data. The successful detection depends strongly on the quality of the learning set (selection of cases), data representation and the mathematical basis of the classifier.