Pulse Wave Recognition of Pregnancy at Three Stages Based on 1D CNN and GRU

Abstract

AbstractThe aim of the present study is to achieve the discrimination of pulse at each stage of pregnancy by 1D convolutional neural network (1D CNN) and gated recurrent unit(GRU) classifier. Firstly, the pulse signals of Chi acquisition position were collected from 160 healthy pregnancy women. Secondly, a new deep learning classifier was proposed by combining 1D CNN and GRU technologies for pulse classification that learns the representation directly from the wave signal. Finally, the classifier proposed is used to classify the pregnancy pulse at three stages of pregnancy. The classifier proposed combines the advantages of CNN and GRU, which greatly improve the accuracy of pregnancy pulse identification. The classification accuracy of three stages of pregnancy pulse achieved satisfactory accuracy of 85%, 88% and 86%, respectively. Furthermore, the average sensitivity, precision and F1-score can reach 88.18%, 86.25% and 87.42%, respectively. The experiment results demonstrated that the method has a good recognition effect and promoted the objective development of TCM.KeywordsTCMPregnancy pulse1D CNNGRU