Beat Analysis of Dimensionality Reduced Perspective Streams from Electrocardiogram Signal Data

Abstract Purpose Although electrocardiogram (ECG) has been proven as a biometric for human identification, applying biometric technology remains challenging with diverse heart rate circumstances in which high intensity heart rate caused waveform deformation may not be known in advance when ECG templates are registered. Methods A calibration method that calculates the ratio of the length of an unidentified electrocardiogram signal to the length of an electrocardiogram template is proposed in this paper. Next, the R peak is used as an axis anchor point of a trigonometric projection (TP) to attain the displacement value. Finally, the unidentified ECG signal is calibrated according to the generated trigonometric value, which corresponds to the trigonometric projection degree of the ratio and the attained displacement measurement. Results The results reveal that the proposed method provides superior overall performance compared with that of the conventional downsampling method, based on the percentage root mean square difference (PRD), correlation coefficients, and mean square error (MSE). Conclusion The curve fitting equation directly maps from the heart rate levels to the TP degree without prior registration information. The proposed ECG calibration method offers a more robust system against heart rate interference when conducting ECG identification.

Download Full-text

Atrial Fibrillation Detection by the Combination of Recurrence Complex Network and Convolution Neural Network

Journal of Probability and Statistics ◽

10.1155/2019/8057820 ◽

2019 ◽

Vol 2019 ◽

pp. 1-9 ◽

Cited By ~ 5

Author(s):

Xiaoling Wei ◽

Jimin Li ◽

Chenghao Zhang ◽

Ming Liu ◽

Peng Xiong ◽

...

Keyword(s):

Neural Network ◽

Atrial Fibrillation ◽

Complex Network ◽

Individual Variation ◽

Deep Neural Network ◽

Detection Algorithm ◽

Convolution Neural Network ◽

Electrocardiogram Signal ◽

Sensitivity Specificity ◽

Atrial Fibrillation Detection

In this paper, R wave peak interval independent atrial fibrillation detection algorithm is proposed based on the analysis of the synchronization feature of the electrocardiogram signal by a deep neural network. Firstly, the synchronization feature of each heartbeat of the electrocardiogram signal is constructed by a Recurrence Complex Network. Then, a convolution neural network is used to detect atrial fibrillation by analyzing the eigenvalues of the Recurrence Complex Network. Finally, a voting algorithm is developed to improve the performance of the beat-wise atrial fibrillation detection. The MIT-BIH atrial fibrillation database is used to evaluate the performance of the proposed method. Experimental results show that the sensitivity, specificity, and accuracy of the algorithm can achieve 94.28%, 94.91%, and 94.59%, respectively. Remarkably, the proposed method was more effective than the traditional algorithms to the problem of individual variation in the atrial fibrillation detection.

Download Full-text