Chinese Journal of Tissue Engineering Research ›› 2010, Vol. 14 ›› Issue (43): 8069-8072.doi: 10.3969/j.issn.1673-8225.2010.43.023

Previous Articles     Next Articles

Human identification using electrocardiograms based on generic algorithm–back propagation neural network

Shi Li, Zhu Min-jie   

  1. Department of Electrical Engineering, Zhengzhou University, Zhengzhou  450001, Henan Province, China
  • Online:2010-10-22 Published:2010-10-22
  • Contact: Zhu Min-jie, Master, Department of Electrical Engineering, Zhengzhou University, Zhengzhou 450001, Henan Province, China zhumj@zzu.edu.cn
  • About author:Shi Li☆, Doctor, Professor, Doctoral supervisor, Department of Electrical Engineering, Zhengzhou University, Zhengzhou 450001, Henan Province, China shili@zzu.edu.cn
  • Supported by:

    the National Natural Science Foundation of China, No. 60971110*

PDF

897

Abstract:

BACKGROUND: Electrocardiogram (ECG) is stable and simple to collect, and prevents reproduction or replication. Therefore, ECG identification technology has important realistic meaning and broad application prospect.
OBJECTIVE: To observe the application of ECG human identification based on generic algorithm (GA) BP neural network.
METHODS: A novel approach using ECG based on GA for human identification was investigated. First, the noise was reduced in preprocessing and features of P, QRS complex and T waves from lead II were extracted using wavelet transform. Second, the features of the amplitude and the interval related to human identification were selected. Then, the BP neural network classifier with optimization of GA for human identification was designed. Finally, the validity of the proposed method was verified using the clinical data.
RESULTS AND CONCLUSION: A total of 30 people served as samples and the average identification accuracy was 96.3%. Results showed that ECG human identification based on generic algorithm (GA) BP neural network has high accuracy and can be effectively used in ECG human identification.

CLC Number: