WCSE

Analysis and Identification of Pulse Signals for Patients with Duodenal Bulbar Ulcer Based on Wavelet Packet Transform and Sample Entropy

Abstract— Objective. The aim of this study is to classify and identify the pulse signals of healthy individuals and patients with duodenal bulb ulcer (DBU) using machine learning algorithm to provide objective parameters for the clinical application of the Traditional Chinese Medicine (TCM) pulse diagnosis. Methods. In order to effectively identify the chronic disease DBU by pulse signals, the fusion of energy features with time-frequency characteristics and sample entropy feature are used to address the shortcomings of traditional pulse feature extraction methods, and the fused feature vectors are used as the input data of the K-nearest neighbor (KNN) classifier for model training and classification identification. Results. There are significant differences in the energy feature and sample entropy feature between pulse signals of normal subjects and patients, respectively. After the two features are fused and fed into the classifier, the results show that the differences are more obvious, which verifies the effectiveness of the proposed feature fusion method. Compared with the traditional methods, the feature fusion method proposed in this paper can effectively represent more accurate pulse information with the accuracy of 89.3\%. Conclusion. Therefore, it can be determined that the linear and nonlinear fused features used in this study can be used as a favorable disease predictor based on pulse signals, and the method also provides a new perspective for pulse diagnosis in TCM.

Index Terms— pulse signal, energy feature, sample entropy

Lin Fan
School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi 'an Key Laboratory of Big Data and Intelligent Computing, Xi'an, China
Xiaokang Zhang
School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, China
Zhongmin Wang
School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi 'an Key Laboratory of Big Data and Intelligent Computing, Xi'an, China
Rong Zhang
School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi 'an Key Laboratory of Big Data and Intelligent Computing, Xi'an, China
Yan He
School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi 'an Key Laboratory of Big Data and Intelligent Computing, Xi'an, China

[Download]

Cite: Lin Fan, Xiaokang Zhang, Zhongmin Wang, Jincheng Zhang, Rong Zhang, Yan He, " Analysis and Identification of Pulse Signals for Patients with Duodenal Bulbar Ulcer Based on Wavelet Packet Transform and Sample Entropy, " WCSE 2022 Spring Event: 2022 9th International Conference on Industrial Engineering and Applications, pp. 1439-1446, Sanya, China, April 15-18, 2022.