基于光纤传感的呼吸与心跳信号采集方法 下载: 1158次
As important physiological indicators of human body, respiration and heart rate can reflect the presence of certain diseases including heart diseases. Photoelectric sensors are more resistant to electromagnetic interference and have a longer service life, which now have a wide range of applications and are used by many researchers to detect physiological parameters. The corresponding measurement methods are mainly divided into two types: non-invasive and invasive. An invasive measurement requires the devices such as electrode pads to contact with body, which can be extremely constricting for the subject and not be easily used in daily life. In contrast, the non-invasive measurement avoids this problem. It is a convenient and comfortable way to measure at home. Since electronic sensors are susceptible to electromagnetic interference, they are not suitable for long-term measurements. Therefore, we need to select a proper sensor to detect physiological information. When a non-invasive physiological information measurement is implemented, the signal is easily disturbed by high-frequency noises and motion artifacts, which reduces the detection accuracy. In order to improve the detection accuracy, one needs to process the acquired signals for de-noising.
We design a signal acquisition platform based on fiber Bragg grating(FBG) sensors, which contains three detection channels, each containing three sensors linked in series. Firstly, the detection device is placed on the bed to collect the signals and demodulate them. This design is to allow subjects to be free from the limitation of lying flat positions. Second, we select the signals acquired by two sensors with the highest energy from the nine detected signals to find the average value of these two sensor output signals and de-trend them. Third, the noise in the signals is removed by using variational mode decomposition (VMD) combined with the improved wavelet threshold function algorithm. The signals are decomposed into a series of intrinsic mode function (IMF) components by using the VMD algorithm. We calculate the correlation coefficient between each component and the original signal, and use the coefficient to determine whether each component is valid or not. The effective IMF components are de-noised again using an improved wavelet threshold function. Finally, we determine whether the motion artifacts are present in the signal or removed, separate the respiratory signal from the heartbeat signal using a band-pass filter, and calculate the respective frequencies using Fourier transform.
We use a denoising method based on VMD combined with an improved wavelet threshold function. Compared with other threshold functions, the estimated wavelet coefficient amplitudes obtained from the processing of our designed threshold function have less deviation from the true amplitudes (Fig. 3). The speed of approximating the true amplitude is faster. It proves to be superior. To verify the performance of the proposed method, we select three comparative algorithms to conduct simulation experiments. We use signal to noise ratio (SNR), root mean squared error (RMSE) , and percent root mean square difference (PRD) to evaluate the denoising performance. The 5 dB25 dB Gaussian white noise is added to the simulated signal. The denoising performance is also verified in the actual acquired signals. From the simulation results (Tables 2, 3, and 4), the SNR after denoising is 30.287 by adding the 25 dB noise. At the same time, the RMSE and PRD are 0.2597 and 3.0595, respectively. The proposed method is superior in these three indicators compared with other methods. The calculated SNR value after de-noising can reach 15.8232 dB with additional 5 dB noise. Even if the signal has a low signal-to-noise ratio, the proposed algorithm still has a good de-noising performance. Results of the actual experiment can be seen in Figs. 11 and 12. The signal obtained after denoising by the proposed algorithm is smooth and the burrs in the signal have been removed. This is due to the fact that VMD overcomes the mode aliasing and endpoint effects of empirical mode decomposition (EMD) in the decomposition process. It has a good decomposition effect on low frequency signals. And we use the correlation coefficient to select the valid and invalid signals, and successfully remove most of the invalid signals (Table 5). The improved wavelet threshold function in this paper can well remove the residual noise in the signal. In general, the proposed algorithm can remove the noise in the signal better than other algorithms.
We propose a method to acquire respiration and heartbeat signals based on FBG sensors. A combined variational mode decomposition with improved wavelet threshold function (VMD-IWT) noise reduction algorithm is used to remove noise interference existing in signals. The simulation results show that our proposed algorithm realizes the best SNR, RMSE, and other indicators, and makes actual signals smoother after noise reduction. We use a band-pass filter to separate signals and calculate their frequencies. The maximum error rate of heart rate is 8.75% with respect to the reference value, and the maximum deviation of respiration rate from the reference value is 1 bpm, which proves the better accuracy of the proposed method. This provides a more convenient and economical way to monitor health at home.
1 引言
传感与数据处理技术的发展使居家健康检测成为可能[1-2],心率与呼吸率作为重要的生理指标,可用于检测某些心血管和呼吸暂停等疾病[3]。心率与呼吸率的采集方式主要分为侵入式和非侵入式两种,侵入式检测需要将电极片等与人体接触以获取信息,因此束缚性较强、舒适性欠佳,非侵入式检测则避免了该问题,已逐渐成为居家健康检测的研究重点。
实施非侵入式检测时,信号易受工频噪声干扰以及肢体活动造成的运动伪影和基线漂移的影响[4-5],从而检测的准确性降低。降噪处理技术在各个领域都得到了广泛的应用[6-7]。为了减少生理信号中存在的噪声干扰,文献[8]利用经验模态分解联合独立分量分析(EMD-ICA)方法去除信号中存在的高频噪声,该方法取得了较好的降噪效果,但EMD分解得到的固有模态函数(IMF) 分量存在模态混叠和端点效应等问题;文献[9]利用集合经验模态分解(EEMD)去除信号中的高频噪声,但没有相关指标判断噪声信号与有用信号的分界点;文献[10]提出自适应噪声的完全集合经验模态分解(CEEMDAN)算法,有效消除了高频噪声和基线漂移的干扰,但该算法计算复杂度高,难以应用于实际;文献[11]提出利用自回归模型和维纳滤波器来重建包含运动伪影的信号片段,但重建信号的幅值与真实信号幅值之间存在一定偏差;文献[12]利用平滑度先验滤波算法和小波滤波来消除高频噪声的干扰,该方法取得了较好的结果,但需严格限制受试者的肢体活动;文献[13]采用小波变换和均方根滤波方法去除信号中的噪声和运动伪影,但易受分解层数的影响。目前的信号降噪处理方法去除了部分高频噪声或运动伪影噪声等,但信号内的残余噪声未得到完全去除。
针对以上问题,本文提出了一种变分模态分解(VMD)联合改进小波阈值函数(IWT)的降噪方法。首先利用所设计的光纤布拉格光栅(FBG)传感器阵列采集人体呼吸与心跳信号,选取有效的信号后,通过VMD算法将信号分解为一系列的IMF分量,并计算各分量与原始信号的相关系数,将信号分为有用信号和高频噪声信号,直接剔除高频噪声信号。其次,针对有用信号中的残余噪声,利用改进的小波阈值函数进行二次去噪。最后,判断信号中是否存在运动伪影并将其剔除,再次提取呼吸与心跳信号。
2 基本原理
2.3 采集平台设计
FBG传感器具有体积小、灵敏度高和易于复用等优点[14],因此将其作为传感检测单元。其工作原理是将外界物理量转化为自身中心波长的变化,中心波长表达式为
心脏跳动时会使人体产生微弱振动,呼吸时人体胸腔也会发生位移变化,当两者作用于FBG传感器时会使其产生形变,从而导致其中心波长发生漂移。通过检测FBG传感器中心波长漂移量Δλ,便可以检测心跳与呼吸的变化情况。单个传感器检测的范围较小,因此不适用于在较大面积的床上进行生理信号的检测。为了解决这一问题,本文将三个传感器(其中心波长分别为1547,1549,1550 nm)进行串联以扩大测量范围,同时,还设计了三个传感通道(CH1, CH2, CH3),每个通道之间的距离约为10 cm。为了保护传感器,将连接好的传感器粘贴于聚碳酸酯板上,如
位于人体背部不同位置的传感器所采集到的信号振动幅度不同[16],振幅越大表示信号的能量越大。为了从9个FBG传感器中选取有用的信号,本文参考文献[16]的方法,从这些信号中各选取一段稳定的信号片段并计算其能量大小,计算公式为
2.4 降噪算法设计
由于信号在采集过程中易受噪声的干扰,而现有的降噪算法无法较好地去除信号中的噪声,因此本文提出了一种变分模态分解联合改进小波阈值函数(VMD-IWT)的信号降噪算法。
VMD算法可将信号分解为一系列具有稀疏特性且从低频到高频排列的IMF分量,每个IMF分量具有相应的中心频率和带宽[17]。VMD的实质是在满足各分量之和等于原始输入信号的约束条件下寻找k1个模态,同时使得模态带宽和最小。
VMD获得的IMF分量包含有用信号与噪声信号,因此需要将有用信号选取出来。相关系数可表征信号之间的相关性,IMF分量与原信号的相关系数越大,则IMF分量包含的有用信息越多。若计算获得的相关系数P大于所设定的阈值,则保留该IMF分量并将其视为有用信号,否则视为无效信号,直接舍弃。VMD算法可去除信号中的大部分高频噪声,但仍存在未完全去除的残余噪声。因此,本文对小波阈值函数进行了改进,使用改进后的算法对信号进行二次降噪。
小波阈值函数降噪的原理就是在选定的小波基函数下对含噪信号进行N层分解[18],每层信号可分解为低频系数和高频系数两个部分。对于分解出的每一层,对其低频系数继续进行分解,对高频系数使用阈值函数进行降噪,后续各层继续执行该操作。当分解层数达到设定值后,进行信号重构。
阈值函数分为硬阈值与软阈值两种[19],硬阈值函数在设定的阈值λ处不连续,信号重构时会产生伪吉布斯现象;而软阈值函数能较好地避免该问题,但使用软阈值函数会使重构信号存在恒定的幅值偏差。为了解决该问题,本文提出了一种改进的阈值函数,以减少估计系数与真实数之间的幅值偏差,即
该阈值函数的连续性证明为
该阈值函数的偏差性证明为
本文将改进的阈值函数与软阈值函数、硬阈值函数、文献[20]和文献[21]所提方法进行对比,仿真结果如
基于上述研究,设计的VMD-IWT算法流程如
2.5 呼吸与心跳信号提取
采集过程中受试者可能会出现肢体运动,这会使信号发生畸变即运动伪影,如
由于心跳与呼吸信号的频率范围不同,因此可以利用带通滤波器将信号进行分离。其中,呼吸信号的滤波范围为0.1~0.8 Hz,心跳信号的滤波范围为1.0~3.5 Hz。因此本文通过FFT将信号由时域转换到频域进行计算。心跳与呼吸率分别为
3 仿真分析
3.1 仿真环境设置
为了验证本文提出的降噪算法的有效性,参照文献[22]构建了心跳与呼吸的合成信号,如
图 6. 仿真信号。(a)纯净信号;(b)添加了5 dB高斯白噪声后的信号;(c)添加了15 dB高斯白噪声后的信号;(d)添加了25 dB高斯白噪声后的信号
Fig. 6. Simulated signals. (a) Pure signal; (b) signal after adding 5 dB Gaussian white noise; (c) signal after adding 15 dB Gaussian white noise; (d) signal after adding 25 dB Gaussian white noise
3.2 评价指标与对比方法
本文选取的信噪比(SNR)、均方根差(RMSE)和均方根差百分比(PRD)是评价信号质量的重要指标。SNR为信号与噪声功率之比,SNR值越大表示降噪效果越好,其计算公式为
RMSE主要用来衡量期望输出值与实际值之间的偏差,RMSE值越小表示降噪效果越好,其计算公式为
PRD通过计算原信号与降噪后信号之间的误差来表征降噪性能,PRD越小代表降噪效果越好,其计算公式为
本文选取EMD-ICA[8]、EEMD[9]、CEEMDAN[10]三种降噪算法进行对比。由于直接观察降噪后的信号图形无法准确判断降噪效果的优劣,因此需要利用相关指标对算法的降噪效果进行评价。其中,本文所提算法中的VMD的层数是根据EMD算法自适应分解得到的,为了获得分解后有用的IMF分量,在添加25 dB高斯白噪声的仿真实验环境下,VMD的层数设置为7,并计算各IMF分量与原信号的相关系数,结果如
表 1. 仿真信号相关系数的计算结果
Table 1. Calculation results of correlation coefficients of simulated signal
|
3.3 仿真结果与分析
表 2. SNR的计算结果
Table 2. Calculation results of SNR
|
表 3. RMSE的计算结果
Table 3. Calculation results of RMSE
|
表 4. PRD的计算结果
Table 4. Calculation results of PRD
|
为了便于观察,将上述计算指标的结果转换成直方图的形式,如
图 7. 各评价指标对比图。(a) SNR指标;(b) RMSE指标;(c) PRD指标
Fig. 7. Comparison chart of each evaluation metric. (a) SNR metric; (b) RMSE metric; (c) PRD metric
4 实验验证
4.1 实验设计
本实验在恒温的室内环境下进行呼吸与心跳信号的采集并结合本文的降噪算法进行处理,降噪后的信号用于呼吸率与心率的计算。实验采集系统如
4.2 信号降噪处理
首先利用VMD算法将信号分解成11个IMF分量,如
表 5. 采集信号相关系数的计算结果
Table 5. Calculation results of correlation coefficients of acquisition signal
|
图 11. 各算法降噪后的效果图。(a) EEMD算法;(b) CEEMDAN算法;(c) EMD-ICA算法;(d)本文算法
Fig. 11. Effect plots of each algorithm after denoising. (a) EEMD algorithm; (b) CEEMDAN algorithm; (c) EMD-ICA algorithm; (d) proposed method
图 12. 信号细节图。(a) EEMD算法;(b) CEEMDAN算法;(c) EMD-ICA算法;(d) 本文算法
Fig. 12. Detailed plots of signals. (a) EEMD algorithm; (b) CEEMDAN algorithm; (c) EMD-ICA algorithm; (d) proposed method
4.3 信号的提取实验
为了将呼吸与心跳信号从采集到的混合信号中提取出来,本文利用带通滤波器将降噪后的信号进行分离,如
图 13. 信号分离及频率计算。(a)原始信号;(b)降噪后的信号;(c)呼吸信号;(d)呼吸频谱图;(e)心跳信号;(f)心跳频谱图
Fig. 13. Signal separation and frequency calculation. (a) Original signal; (b) denoised signal; (c) respiratory signal; (d) respiratory spectrogram; (e) heartbeat signal; (f) heartbeat spectrogram
表 6. 呼吸与心率的计算结果
Table 6. Calculation results of respiration and heart rate
|
结果表明,本文方法获得的呼吸率与参考值基本一致,其中呼吸次数最大偏差为1 beat/min,心跳次数与参考值的最大误差率为8.75%。根据心率测量准确度要求[23],误差不超过输入心率的±10%或5 beat/min,本文方法的误差率满足要求。表明所设计方法可较准确地检测呼吸率与心率,具有一定的实用性。
5 结论
提出了一种基于FBG传感器的呼吸与心跳信号采集方法。针对信号中存在的噪声干扰问题,提出了一种VMD-IWT联合降噪算法。仿真实验结果表明,所提算法的SNR、RMSE等指标较好,降噪后的信号更平滑。提取降噪后的呼吸与心跳信号,计算结果表明,与参考值相比,采集的心率最大误差率为8.75%,呼吸与参考值的最大偏差为1 beat/min,证明所提方法具有较好的准确性,为居家健康检测提供了一种更为便捷经济的方式。未来工作将继续研究受试者在说话或打鼾情况下的心率与呼吸率的准确测量。
[1] Xie Q S, Wang M, Zhao Y, et al. A personalized beat-to-beat heart rate detection system from ballistocardiogram for smart home applications[J]. IEEE Transactions on Biomedical Circuits and Systems, 2019, 13(6): 1593-1602.
[2] Lu C Y, Wu S, Jiang C X, et al. Weak harmonic signal detection method in chaotic interference based on extended Kalman filter[J]. Digital Communications and Networks, 2019, 5(1): 51-55.
[3] LokaveeS, TantrakulV, PengjiamJ, et al. A sleep monitoring system using force sensor and an accelerometer sensor for screening sleep apnea[C]//2021 13th International Conference on Knowledge and Smart Technology (KST), January 21-24, 2021, Bangsaen, Chonburi, Thailand. New York: IEEE Press, 2021: 208-213.
[4] Bicen A, Whittingslow D C, Inan O T. Template-based statistical modeling and synthesis for noise analysis of ballistocardiogram signals: a cycle-averaged approach[J]. IEEE Journal of Biomedical and Health Informatics, 2019, 23(4): 1516-1525.
[5] Zhang H B, Huang Q, Li F W, et al. A network security situation prediction model based on wavelet neural network with optimized parameters[J]. Digital Communications and Networks, 2016, 2(3): 139-144.
[6] 杨菊花, 张琳婧, 陈光武, 等. 基于SVD/小波的MEMS陀螺误差分析及降噪处理[J]. 重庆邮电大学学报(自然科学版), 2020, 32(2): 322-328.
Yang J H, Zhang L J, Chen G W, et al. Error analysis and noise reduction of MEMS gyro based on SVD/wavelet[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2020, 32(2): 322-328.
[7] 吕俊瑞, 罗学刚, 岐世峰, 等. 基于模糊度量视觉特征的非局部均值去噪[J]. 重庆邮电大学学报(自然科学版), 2018, 30(3): 408-415.
Lü J R, Luo X G, Qi S F, et al. Improved non-local means image denoising algorithm using visual features based on fuzzy metric[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2018, 30(3): 408-415.
[8] 姜星, 耿读艳, 张园园, 等. 基于EMD-ICA的心冲击信号降噪研究[J]. 中国生物医学工程学报, 2019, 38(2): 138-145.
Jiang X, Geng D Y, Zhang Y Y, et al. BCG signal de-noising method research based on EMD-ICA[J]. Chinese Journal of Biomedical Engineering, 2019, 38(2): 138-145.
[9] Chang K M. Ensemble empirical mode decomposition for high frequency ECG noise reduction[J]. Biomedizinische Technik/Biomedical Engineering, 2010, 55(4): 193-201.
[10] 耿读艳, 王晨旭, 赵杰, 等. 基于CEEMDAN-PE的心冲击信号降噪方法研究[J]. 仪器仪表学报, 2019, 40(6): 155-161.
Geng D Y, Wang C X, Zhao J, et al. Research on BCG signal de-noising method based on CEEMDAN and PE[J]. Chinese Journal of Scientific Instrument, 2019, 40(6): 155-161.
[11] Alivar A, Carlson C, Suliman A, et al. Motion artifact detection and reduction in bed-based ballistocardiogram[J]. IEEE Access, 2019, 7: 13693-13703.
[12] 孔令琴, 吴育恒, 庞宗光, 等. 基于心冲击描记术的心率检测方法[J]. 中国激光, 2020, 47(2): 0207042.
[13] Wen X, Huang Y Q, Wu X M, et al. A feasible feature extraction method for atrial fibrillation detection from BCG[J]. IEEE Journal of Biomedical and Health Informatics, 2020, 24(4): 1093-1103.
[14] 陈勇, 陈亚武, 刘志强, 等. 基于光纤布拉格光栅传感的齿轮故障检测方法[J]. 中国激光, 2020, 47(3): 0304007.
[15] 方震, 白忠瑞, 陈贤祥, 等. 基于压电陶瓷传感器的非接触式精准逐拍心率提取方法研究[J]. 电子与信息学报, 2021, 43(5): 1472-1479.
Fang Z, Bai Z R, Chen X X, et al. Unconstrained accurate beat-to-beat heart rate extraction based on piezoelectric ceramics sensor[J]. Journal of Electronics & Information Technology, 2021, 43(5): 1472-1479.
[16] 赵林, 彭敏, 杨翔宇, 等. 基于压电陶瓷的睡眠信息检测方法[J]. 仪器仪表学报, 2018, 39(7): 245-252.
Zhao L, Peng M, Yang X Y, et al. Detection method of sleep information based on piezoceramic[J]. Chinese Journal of Scientific Instrument, 2018, 39(7): 245-252.
[17] Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.
[18] 杨勇, 方针, 方海斌, 等. 基于小波去噪的MEMS陀螺仪随机误差校准算法[J]. 重庆邮电大学学报(自然科学版), 2020, 32(1): 99-104.
Yang Y, Fang Z, Fang H B, et al. An algorithm for random error calibration of MEMS gyroscope based on wavelet denoising[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2020, 32(1): 99-104.
[19] 李想, 汪立新, 段志强. 改进自适应小波降噪在激光陀螺仪信号处理中的应用[J]. 激光与光电子学进展, 2020, 57(21): 210401.
[20] Chen Y, Cheng Y N, Liu H L. Application of improved wavelet adaptive threshold de-noising algorithm in FBG demodulation[J]. Optik, 2017, 132: 243-248.
[21] 尹丽, 陈富民, 张琦, 等. 采用集合经验模态分解和改进阈值函数的心电自适应去噪方法[J]. 西安交通大学学报, 2020, 54(1): 101-107.
Yin L, Chen F M, Zhang Q, et al. ECG adaptive denoising method based on EEMD and improved threshold function[J]. Journal of Xi'an Jiaotong University, 2020, 54(1): 101-107.
[22] Leal A G, Díaz C R, Leitão C, et al. Polymer optical fiber-based sensor for simultaneous measurement of breath and heart rate under dynamic movements[J]. Optics & Laser Technology, 2019, 109: 429-436.
[23] 郭健, 陈雨行, 王丽荣, 等. 非接触式呼吸与心率信号采集系统[J]. 光学 精密工程, 2019, 27(6): 1354-1361.
Article Outline
李玉环, 陈勇, 刘焕淋, 江涛. 基于光纤传感的呼吸与心跳信号采集方法[J]. 中国激光, 2022, 49(4): 0406004. Yuhuan Li, Yong Chen, Huanlin Liu, Tao Jiang. Respiration and Heartbeat Signal Acquisition Method Based on Fiber Optic Sensing[J]. Chinese Journal of Lasers, 2022, 49(4): 0406004.