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

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 mult...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Peng Lu, Chao Liu, Xiaobo Mao, Yvping Zhao, Hanzhang Wang, Hongpo Zhang, Lili Guo
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/4b96fd66e9174b629d7633e646bf4da8
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4b96fd66e9174b629d7633e646bf4da8
record_format	dspace
spelling	oai:doaj.org-article:4b96fd66e9174b629d7633e646bf4da82021-12-02T13:30:34ZFew-shot pulse wave contour classification based on multi-scale feature extraction10.1038/s41598-021-83134-y2045-2322https://doaj.org/article/4b96fd66e9174b629d7633e646bf4da82021-02-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-83134-yhttps://doaj.org/toc/2045-2322Abstract 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.Peng LuChao LiuXiaobo MaoYvping ZhaoHanzhang WangHongpo ZhangLili GuoNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Peng Lu Chao Liu Xiaobo Mao Yvping Zhao Hanzhang Wang Hongpo Zhang Lili Guo Few-shot pulse wave contour classification based on multi-scale feature extraction
description	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.
format	article
author	Peng Lu Chao Liu Xiaobo Mao Yvping Zhao Hanzhang Wang Hongpo Zhang Lili Guo
author_facet	Peng Lu Chao Liu Xiaobo Mao Yvping Zhao Hanzhang Wang Hongpo Zhang Lili Guo
author_sort	Peng Lu
title	Few-shot pulse wave contour classification based on multi-scale feature extraction
title_short	Few-shot pulse wave contour classification based on multi-scale feature extraction
title_full	Few-shot pulse wave contour classification based on multi-scale feature extraction
title_fullStr	Few-shot pulse wave contour classification based on multi-scale feature extraction
title_full_unstemmed	Few-shot pulse wave contour classification based on multi-scale feature extraction
title_sort	few-shot pulse wave contour classification based on multi-scale feature extraction
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/4b96fd66e9174b629d7633e646bf4da8
work_keys_str_mv	AT penglu fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction AT chaoliu fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction AT xiaobomao fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction AT yvpingzhao fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction AT hanzhangwang fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction AT hongpozhang fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction AT liliguo fewshotpulsewavecontourclassificationbasedonmultiscalefeatureextraction
_version_	1718392912443604992

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

Ejemplares similares