基于二维卷积的连续血压预测系统

崔守毅; 杨国伟; 何羽恒; 管静萱; 胡远凝; 廖丹丹; 荆凯

doi:10.20193/j.ices2097-4191.2024.0009

PDF(5622 KB)

集成电路与嵌入式系统 ›› 2024, Vol. 24 ›› Issue (8) : 1-6. DOI: 10.20193/j.ices2097-4191.2024.0009

封面文章

基于二维卷积的连续血压预测系统

作者信息 +

Continuous blood pressure prediction system based on two-dimensional convolution

Author information +

文章历史 +

摘要

针对生命体征信号数字化采集和连续血压预测等需求,设计并实现了一种基于二维卷积的连续血压预测系统。在系统硬件部分使用ESP32模组、AD8232模块和PulseSensor传感器,采集获得的人体心电图(ECG)和光电容积脉搏波(PPG)信号数据并通过MQTT协议传输至服务端处理。本文算法部分使用格拉米角差场(GADF)、二维卷积和模型剪枝技术,设计并训练了使用ECG和PPG信号预测人体连续血压的神经网络模型,并分别在开源数据集和自制数据集中测试了连续血压预测模型的性能。本文系统为重要体征信号采集和连续血压预测提供了一个有效的参考方案。

Abstract

To address the demands of digital acquisition of vital signs signals and continuous blood pressure prediction,this paper designs and constructs a continuous blood pressure prediction system based on two-dimensional (2D) convolution.The system hardware adopts ESP32 module,AD8232 module and PulseSensor sensor to collect the human electrocardiography (ECG) and photoplethysmography (PPG) signal data,which are then transmitted to the server through the MQTT protocol for the consequent processing.Regarding the algorithms of this paper,a neural network model using ECG and PPG signals was designed and trained to predict continuous human blood pressure,employing the Gramian angular difference field (GADF),2D convolution,and model pruning techniques.The performance of the continuous blood pressure prediction model is verified on both classic open-source datasets and self-collected datasets.This system proposed in this paper provides a practical reference scheme for the vital signs signal acquisition and continuous blood pressure prediction.

导出引用

崔守毅, 杨国伟, 何羽恒, 等. 基于二维卷积的连续血压预测系统[J]. 集成电路与嵌入式系统. 2024, 24(8): 1-6 https://doi.org/10.20193/j.ices2097-4191.2024.0009

CUI Shouyi, YANG Guowei, HE Yuheng, et al. Continuous blood pressure prediction system based on two-dimensional convolution[J]. Integrated Circuits and Embedded Systems. 2024, 24(8): 1-6 https://doi.org/10.20193/j.ices2097-4191.2024.0009

中图分类号： TP391

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]

孙芹, 田伟帆, 罗婷婷, 等. 2023年世界卫生组织《全球高血压报告》解读[J]. 中国胸心血管外科临床杂志, 2024, 31(2):203-208.

SUN

, TIAN

W F

, LUO

T T

, et al. Interpretation of the World Health Organization's Global Report on Hypertension in 2023[J]. Chinese Journal of Clinical Thoracic and Cardiovascular Surgery, 2024, 31(2):203-208 (in Chinese).

本文引用 [1]

[2]	NCD Risk Factor Collaboration. Worldwide trends in hypertension prevalence and progress in treatment and control from 1990 to 2019:a pooled analysis of 1201 population-representative studies with 104 million participants[J]. The Lancet, 2021, 398(10304):957-980. 本文引用 [1]

[3]	胡祥芹, 王春梅. 有创血压监测的研究进展[J]. 护理研究, 2008, 38(3):193-195. HU X Q, WANG CH M. Research progress of invasive blood pressure monitoring[J]. Chinese Nursing Research, 2008, 38(3):193-195 (in Chinese). 本文引用 [1]

[4]	张政波, 吴太虎. 无创血压测量技术与进展[J]. 中国医疗器械杂志, 2003, 27(3):196-199. ZHANG ZH B, WU T H. Technology and progress of noninvasive blood pressure measurement[J]. Chinese Journal of Medical Instrumentation, 2003, 27(3):196-199 (in Chinese). 本文引用 [1]

[5]	S SUN, E BRESCH, J MUEHLSTEFF, et al. Systolic blood pressure estimation using ECG and PPG in patients undergoing surgery[J]. Biomedical Signal Processing and Control, 2023, 79(1):104040. 本文引用 [1]

[6]	李家杨, 叶兵, 朱晓冬. STM32的无创连续血压测量系统设计[J]. 单片机与嵌入式系统应用, 2017, 17(4):76-79. LI J Y, YE B, ZHU X D. Design of non-invasive continuous Blood pressure measurement system of STM32[J]. Microcontrollers & Embedded Systems, 2017, 17(4):76-79 (in Chinese). 本文引用 [1]

[7]

J MA

, J

CHOI

, A

HOURLIER-FARGETTE

, et al. Relation between blood pressure and pulse wave velocity for human arteries[J]. Proceedings of the National Academy of Sciences of the United States of America, 2018, 115(44):11144-11149.

https://doi.org/10.1073/pnas.1814392115

https://www.ncbi.nlm.nih.gov/pubmed/30322935

本文引用 [1] 摘要

Continuous monitoring of blood pressure, an essential measure of health status, typically requires complex, costly, and invasive techniques that can expose patients to risks of complications. Continuous, cuffless, and noninvasive blood pressure monitoring methods that correlate measured pulse wave velocity (PWV) to the blood pressure via the Moens-Korteweg (MK) and Hughes Equations, offer promising alternatives. The MK Equation, however, involves two assumptions that do not hold for human arteries, and the Hughes Equation is empirical, without any theoretical basis. The results presented here establish a relation between the blood pressure and PWV that does not rely on the Hughes Equation nor on the assumptions used in the MK Equation. This relation degenerates to the MK Equation under extremely low blood pressures, and it accurately captures the results of in vitro experiments using artificial blood vessels at comparatively high pressures. For human arteries, which are well characterized by the Fung hyperelastic model, a simple formula between and PWV is established within the range of human blood pressures. This formula is validated by literature data as well as by experiments on human subjects, with applicability in the determination of blood pressure from PWV in continuous, cuffless, and noninvasive blood pressure monitoring systems.

[8]	岳铭楷, 黄宝震, 樊明辉, 等. 基于脉搏波速法的无创连续血压系统设计[J]. 单片机与嵌入式系统应用, 2021, 21(9):78-82. YUE M K, HUANG B ZH, FAN M H, et al. Design of non-invasive continuous blood pressure system based on pulse wave velocity method[J]. Microcontrollers & Embedded Systems, 2021, 21(9):78-82 (in Chinese). 本文引用 [1]

[9]	B HUANG, W CHEN, C L LIN, et al. MLP-BP:A novel framework for cuffless blood pressure measurement with PPG and ECG signals based on MLP-Mixer neural networks[J]. Biomedical Signal Processing and Control, 2022(73):103404. 本文引用 [3]

[10]	S BAKER, W XIANG, I ATKINSON. A hybrid neural network for continuous and non-invasive estimation of blood pressure from raw electrocardiogram and photoplethysmogram waveforms[J]. Computer Methods and Programs in Biomedicine, 2021(207):106191. 本文引用 [3]

[11]	Z WANG, T OATES. Encoding time series as images for visual inspection and classification using tiled convolutional neural networks[C]// Workshops at the twenty-ninth AAAI conference on artificial intelligence,Menlo Park,CA,USA, 2015. 本文引用 [1]

[12]	K SIMONYAN, A ZISSERMAN. Very deep convolutional networks for large-scale image recognition[C]// International Conference on Learning Representations (ICLR),San Diego,CA,USA, 2015. 本文引用 [1]

[13]	C SZEGEDY, W LIU, Y JIA, et al. Rabinovich. Going deeper with convolutions[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR),Boston,MA,USA, 2015. 本文引用 [1]

[14]	K HE, X ZHANG, S REN, et al. Deep residual learning for image recognition[C]// IEEE Conference on Computer Vision and Pattern Recognition(CVPR),Las Vegas,NV,USA 2016. 本文引用 [1]

[15]	X DAI, H YIN, N K JHA. NeST:A neural network synthesis tool based on a grow-and-prune paradigm[J]. IEEE Transactions on Computers, 2019, 68(10):1487-1497. 本文引用 [1]

[16]	W WEN, C WU, Y WANG, et al. Learning structured sparsity in deep neural networks[C]// 30th International Conference on Neural Information Processing Systems (NIPS'16),Barcelona,Spain, 2016. 本文引用 [2]

[17]	C CARLSON, V R TURPIN, A SULIMAN, et al. Bed base ballistocardiography:dataset- and ability to track cardiovascular parameters[J]. Sensors, 2021, 21(1):156. 本文引用 [1]

[18]	ATTARPOUR A, MAHNAM A, AMINITABAR A, et al. Cuff-less continuous measurement of blood pressure using wrist and fingertip photo-plethysmograms: Evaluation and feature analysis[J]. Biomedical Signal Processing and Control, 2019(49):212-220. 本文引用 [2]

[19]	THAMBIRAJ G, GANDHI U, MANGALANATHAN U, et al. Investigation on the effect of Womersley number,ECG and PPG features for cuff less blood pressure estimation using machine learning[J]. Biomedical Signal Processing and Control, 2020(60):101942. 本文引用 [2]