Clustering Analysis of ECG Data Streams

ECG signal is significant for cardiovascular diagnosis. Users may concern about the clustering result of ECG waves in recent time or the whole history. However, most existing stream clustering techniques can't give the two kinds of result at the same time. To tackle this challenge, in this paper, we propose a new stream clustering algorithm, DenstreamD, which can be used to meet the requirement. The core idea of DenstreamD is based on Denstream but to add decay potential core micro-clusters in the online maintenance phase. Comprehensive experiments are conducted using MIT-BIH Long-Term ECG database to demonstrate the effectiveness of proposed algorithms. The experiments show that DenstreamD has better accuracy and efficiency than its original algorithm while obtaining two kinds of clustering results.