Clustering and visualization of ECG signals

Abstract

Holter electrocardiographic (ECG) recordings are ambulatory long-term registers that are used to detect heart diseases. These recordings normally include more than one channel and their durations are up to 24 hours. The principal problem of the cardiologists is the manual inspection of the whole Holter ECG in order to find all those beats which morphologically differ from the normal beats. In this paper we present our method. Firstly, we apply a grid clustering technique. Secondly, we use a special density-based clustering algorithm, named Optics. Then we visualize every heart beat in the record, heartbeats in a cluster, furthermore we represent every cluster with median of heartbeats. We can perform manual. With this method the ECG is easily analyzed and the time of processing is optimized.