A new phase space analysis algorithm for cardiac arrhythmia detection

Computer-assisted cardiac arrhythmia recognition is critical for prompt and efficient management of cardiac disorders. In this paper, a new phase space analysis-based algorithm has been proposed for electrocardiogram (ECG) signals to facilitate the detection of cardiac arrhythmia. A phase space density plot was obtained by mapping the distribution of points in the phase space of ECG signals and the phase space density values within a predefined window were used for arrhythmia detection. Classification was performed using a multilayer back propagation neural network and an overall accuracy of 98.55% was obtained for classifying cardiac arrhythmias under consideration. The proposed technique will be a useful tool in the assessment of the cardiovascular status of patients in critical care units.