Few-shot pulse wave contour classification based on multi-scale feature extraction

Peng Lu,Xiaobo Mao,Hanzhang Wang,Lili Guo,Yvping Zhao,Hongpo Zhang,Chao Liu

doi:10.1038/s41598-021-83134-y

Abstract

The annotation procedure of pulse wave contour (PWC) is expensive and time-consuming, thereby hindering the formation of large-scale datasets to match the requirements of deep learning. To obtain better results under the condition of few-shot PWC, a small-parameter unit structure and a multi-scale feature-extraction model are proposed. In the small-parameter unit structure, information of adjacent cells is transmitted through state variables. Simultaneously, a forgetting gate is used to update the information and retain long-term dependence of PWC in the form of unit series. The multi-scale feature-extraction model is an integrated model containing three parts. Convolution neural networks are used to extract spatial features of single-period PWC and rhythm features of multi-period PWC. Recursive neural networks are used to retain the long-term dependence features of PWC. Finally, an inference layer is used for classification through extracted features. Classification experiments of cardiovascular diseases are performed on photoplethysmography dataset and continuous non-invasive blood pressure dataset. Results show that the classification accuracy of the multi-scale feature-extraction model on the two datasets respectively can reach 80% and 96%, respectively.

Highlights

The annotation procedure of pulse wave contour (PWC) is expensive and time-consuming, thereby hindering the formation of large-scale datasets to match the requirements of deep learning
Machine-learning methods represented by convolution neural n etworks[8,9] (CNN), recursive neural n etworks[10,11] (RNN), and support vector m achines[12,13] are suitable for complex non-linear sequences
In literature[24], the representation of speech signals from an original network is automatically learned by CNN, and the temporal representation of features is learned by Long short-term memory (LSTM); In literature[25], the features of wearable sensor data is learned by CNN, and the time dependence between actions are modeled by LSTM

Summary

Introduction

The annotation procedure of pulse wave contour (PWC) is expensive and time-consuming, thereby hindering the formation of large-scale datasets to match the requirements of deep learning. To obtain better results under the condition of few-shot PWC, a small-parameter unit structure and a multi-scale feature-extraction model are proposed. Results show that the classification accuracy of the multi-scale feature-extraction model on the two datasets respectively can reach 80% and 96%, respectively. Statistical methods represented by artificial time-frequency domain feature extraction are suitable for stationary s equences[7]. Long short-term memory (LSTM), a variant of RNN, has the ability of mining long-distance time-series data information[15]. It is extensively used in machine translation[16,17], fault diagnosis[18,19], speech recognition[20,21], and electrocardiogram classification[22,23]. PWC is a kind of few-shot data that difficultly meets the training needs of complex deep n etworks[26]

Methods

Results

Conclusion