摘要
为了在睡眠时以非侵入方式监测心冲击信号(BCG)和呼吸信号,使用电阻式薄膜压力传感器嵌入床垫中,将变分模态分解(VMD)算法引入到二维生理信号提取过程.信号经床垫中的柔性压力传感器,通过硬件低通滤波、数字去趋势(DFA)后,利用VMD算法分解出生理信号中心冲击信号与呼吸信号的潜在分量,通过自适应选取有效分量重构BCG信号与呼吸信号.基于Hilbert变换,对比VMD、经验模态分解(EMD)、互补集合经验模态分解(CEEMD)分量的瞬时频率.VMD在0~3.0 Hz内的混叠情况相对于EMD与CEEMD得到改善.采用BlandAltman法,对标准结果和实验重构结果进行一致性评价.结果表明,利用VMD法所得BCG与呼吸信号分别有93.75%和92.5%的点在95%一致性标准界限内,有较高的一致性.
A resistive film pressure sensor was embedded in the mattress,and the variational modal decomposition(VMD)algorithm was introduced into the two-dimensional physiological signal extraction process in order to monitor the ballistocardiogram(BCG)and respiratory signal in a non-invasive manner during sleep.The VMD algorithm was used to decompose the potential components of the BCG signal and respiratory signal in the physiological signal after the signal passes through the flexible pressure sensor in the mattress,hardware low-pass filtering,and digital detrending(DFA).The effective components were adaptively selected to reconstruct the BCG signal and the respiratory signal.The instantaneous frequencies of VMD,empirical mode decomposition(EMD),and complementary set empirical mode decomposition(CEEMD)components were compared based on Hilbert transform.The aliasing situation of VMD in 0~3.0 Hz was improved compared with EMD and CEEMD.The BlandAltman method was used to evaluate the consistency of the standard results and experimental reconstruction results.Results show that 93.75%and 92.5%of the BCG and respiratory signals obtained by the VMD method are within the standard limit of 95%consistency,and there is a high consistency.
作者
童基均
柏雁捷
潘剑威
杨佳锋
蒋路茸
TONG Ji-jun;BAI Yan-jie;PAN Jian-wei;YANG Jia-feng;JIANG Lu-rong(School of Information Sciences and Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China;Department of Neurosurgery,the First Affiliated Hospital,School of Medicine,Zhejiang University,Hangzhou 310003,China)
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2020年第10期2058-2066,共9页
Journal of Zhejiang University：Engineering Science
基金
浙江省重点研发计划资助项目(2015C03023)
浙江理工大学基本科研业务费资助项目(2019Q042)
浙江理工大学“521人才培养计划”资助项目。
