Algorithm for Clustering Analysis of ECG Data

Zetao Lin; Yaozheng Ge; Guoliang Tao

doi:10.1109/IEMBS.2005.1615302

Algorithm for Clustering Analysis of ECG Data

Conf Proc IEEE Eng Med Biol Soc. 2005:2005:3857-60. doi: 10.1109/IEMBS.2005.1615302.

Authors

Zetao Lin¹, Yaozheng Ge, Guoliang Tao

Affiliation

¹ The State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University, Hangzhou 310027 China.

PMID: 17281072
DOI: 10.1109/IEMBS.2005.1615302

Abstract

To satisfy the difficult requirements of ECG analysis such as large data volume, high accuracy and real-time, a classification algorithm for arrhythmia based on clustering analysis is developed. According to things-of-one-kind-come-together principle, this algorithm uses the similarity of heart cases of the same category and, at the same time, incorporates the factor of individual differences. It analyzes arrhythmia by clustering QRS complex waveforms and applies rhythm analysis as the subordinate method. Verified by eight records of MIT-BIH arrhythmia standard heart electricity database, the clustering correct rate reaches above 90%, which shows that this algorithm can analyze arrhythmia effectively.