Pulse Waveforms Classification Based on Wavelet Network

Abstract

This paper proposes a new algorithm to classify pulse waveforms based on discrete wavelet network. This paper selects 4-order discrete Daubechies wavelet as the wavelet node of this wavelet network to classify six pulse patterns distinctive in shape. 600 pulse records are used to train this wavelet network and 300 pulse records are used to test the classifier's performance. The test results show that this approach has 83% agreement rate with the experienced experts. Compared with traditional classification methods, it needs not the experience in feature extraction.

Classification of Pulse Waveforms Using Edit Distance with Real Penalty

“…To provide a comprehensive performance evaluation of the proposed methods, we compare the classification rates of EDKC and GEKC with several achieved accuracies in the recent literature [19,[21][22][23]. Table 4 lists the sizes of the data set, the number of pulse waveform classes, and the achieved classification rates of several recent pulse waveform EURASIP Journal on Advances in Signal Processing…”

“…classifiers, including improved dynamic time warping (IDTW) [19], decision tree (DT-M4) [22], artificial neural network [21], and wavelet network [23]. From Table 4, one can see that GEKC achieves higher accuracy than wavelet network [23] and artificial neural network [21]. Moreover, although IDTW and DT-M4 reported somewhat higher classification rates than our methods, the size of the data set used in our experiments is much larger than those used in these two methods, and DT-M4 is only tested on a 3-class problem.…”

“…However, because of the complicated intra-class variation in pulse patterns and the inevitable influence of local time shifting in pulse waveforms, it has remained a difficult problem for automatic pulse waveform classification. Although researchers have developed several pulse waveform classification methods such as artificial neural network [18,20,21], decision tree [22], and wavelet network [23], most of them are only tested on small data sets and usually cannot achieve satisfactory classification accuracy.…”

Modern Pulsometer for Traditional Indian Medicine

“…The least works has been carried out on different patients, of different age groups. In different resolutions and by using different techniques it has been mentioned in subfigures from (a) through(o) are from [4], [12], [13], [5], [14], [15], [16], [17], [18], [19], [20], [21], [22], [9], [23] respectively.…”

A Framework for Automatic Time-Domain Characteristic Parameters Extraction of Human Pulse Signals

“…Therefore, quantitative methods are needed. Much effort is being spent on pulse analysis, such as the classification of pulse waveforms [1][2][3][4][5][6][7] and cardiovascular assistant diagnosis [8][9][10][11]. In pulse diagnosis, time-domain parameters can reflect the specificity of pulse signals.…”