WO2024031845A2 - 一种基于多点校准的无袖带动态血压测量方法 - Google Patents

一种基于多点校准的无袖带动态血压测量方法 Download PDF

Info

Publication number
WO2024031845A2
WO2024031845A2 PCT/CN2022/128483 CN2022128483W WO2024031845A2 WO 2024031845 A2 WO2024031845 A2 WO 2024031845A2 CN 2022128483 W CN2022128483 W CN 2022128483W WO 2024031845 A2 WO2024031845 A2 WO 2024031845A2
Authority
WO
WIPO (PCT)
Prior art keywords
blood pressure
ppg
measured
waveform
blood
Prior art date
Application number
PCT/CN2022/128483
Other languages
English (en)
French (fr)
Inventor
邢晓曼
宋明轩
董文飞
Original Assignee
苏州国科医工科技发展(集团)有限公司
济南国科医工科技发展有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州国科医工科技发展(集团)有限公司, 济南国科医工科技发展有限公司 filed Critical 苏州国科医工科技发展(集团)有限公司
Publication of WO2024031845A2 publication Critical patent/WO2024031845A2/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/021—Measuring pressure in heart or blood vessels
    • A61B5/02108—Measuring pressure in heart or blood vessels from analysis of pulse wave characteristics
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/021—Measuring pressure in heart or blood vessels
    • A61B5/02108—Measuring pressure in heart or blood vessels from analysis of pulse wave characteristics
    • A61B5/02116—Measuring pressure in heart or blood vessels from analysis of pulse wave characteristics of pulse wave amplitude
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Definitions

  • the present invention relates to blood pressure measurement, and in particular to a cuffless dynamic blood pressure measurement method based on multi-point calibration.
  • Hypertension can cause cardiovascular disease, aortic aneurysm, stroke, peripheral vascular disease, chronic kidney disease, etc., and has a high risk of death. Hypertensive patients are also a high-risk group in epidemics. For example, infection with the new coronavirus may cause more severe lung failure, and the mortality rate is much higher than that of other patients. Therefore, early prevention and control of hypertension is of extremely great significance.
  • Diagnosing and managing hypertension requires continuous monitoring of blood pressure, but mood swings, physical activity, and sleep can all affect blood pressure.
  • the currently commonly used ambulatory sphygmomanometers in clinical practice are cuff-type sphygmomanometers.
  • the inflation and deflation operations can easily interfere with sleep and aggravate subjects' anxiety.
  • the measurement frequency is low, with only about fifty data points per day, and cannot reflect instantaneous increases in blood pressure.
  • due to the high price and cumbersome preparation work it is difficult to achieve monitoring in daily home scenarios.
  • a wrist watch is used to measure blood pressure in real time, but non-interfering wrist blood pressure measurement generally extracts PPG features and performs blood pressure fitting through multiple linear regression or machine learning.
  • the PPG waveform often causes trend reversal due to interference from other factors; and the wrist blood volume waveform contains limited information, especially when the pressure is unstable or the temperature is low, and when you are emotionally stressed, it causes vasoconstriction and serious baseline drift, leading to inaccurate measurements. .
  • one of the purposes of the present invention is to provide a cuffless dynamic blood pressure measurement method based on multi-point calibration, which can avoid trend reversal caused by interference from other factors and baseline drift caused by vasoconstriction.
  • a cuffless dynamic blood pressure measurement method based on multi-point calibration including the following steps:
  • Data collection Use multiple sets of sensors to measure the contact pressure on the wrist, and use the contact pressure measured by multiple sets of sensors to subtract to obtain the contact pressure difference to offset the error caused by baseline drift;
  • V is the blood volume corresponding to the specified transmural pressure P transm
  • V 0 is the V value corresponding to zero P transm
  • V max is the maximum blood volume
  • C max is the maximum compliance at zero P transm
  • the contact pressure is equal to blood pressure (BP) minus the transmural pressure P transm
  • the number of equations exceeds the number of unknowns and can be further reduced. small error;
  • Measure cardiac output measure cardiac output CO through a three-axis acceleration sensor
  • the force output by the heart beat is proportional to m*a.
  • the resolution of the three-axis acceleration sensor is set to less than or equal to 0.5 mg.
  • each group of sensors includes a light source and a detector.
  • a four-element elastic cavity model is used to generate the PPG waveform specifically as follows:
  • a four-element elastic cavity model is used to simulate the response of peripheral blood vessels
  • a four-element elastic cavity model was used to simulate the response of peripheral blood vessels.
  • the specific details are as follows: the heart is represented as a current source, the left ventricular ejection is represented as q in , p 1 represents the central blood pressure, and the arterial tree system passes the lumped resistance as R , the lumped compliance of the central aorta is C 1 , the total compliance of the distal artery is C 2 and the blood flow inertia is L for modeling, after modeling
  • q 0 is the maximum input amplitude related to cardiac output; q 0 is calculated by SV ⁇ /(120 ⁇ Ts), where SV is the stroke volume, T s is the duration of left ventricular ejection, and T is The duration of the cardiac cycle, ⁇ T s is the peak time during cardiac contraction;
  • the resistance to blood flow is the sum of all individual resistances in the microcirculation, resulting in the following equation:
  • the PPG waveform generated by peripheral blood pressure p2 is obtained by solving differential equation (3).
  • the generated and measured PPG waveforms are compared in pairs to confirm the applicable range of the linear model of the measured PPG waveform.
  • the measured PPG generated based on the actual measurement is Data, select a parameter combination that can actually exist to form a credible interval for the simulated database.
  • selecting a parameter combination that can actually exist to form a credible interval of the simulation database is specifically: pairwise comparison of the generated and measured PPG waveforms, if a set of simulation parameters generated If the waveform matches at least one reference measurement waveform, it is considered to be real; adjust the simulated blood pressure value p2 to match the reference mean arterial pressure, calculate the corresponding PPG waveform through the pressure-volume conversion formula, and normalize the PPG waveform amplitude After normalization, the average deviation is calculated and the points on the measured PPG waveform with small average deviation are retained.
  • the average deviation PPGsim(k) is the calculated waveform point
  • PPGref(k) is the measured waveform point
  • n is the number of waveform points
  • the range of k is 1 to n.
  • the average deviation e is less than 0.005.
  • the cuffless dynamic blood pressure measurement method of the present invention based on multi-point calibration realizes dynamic monitoring and compensation of pressure through multiple sensor groups; it protects the PPG waveform from distortion, making it possible to measure blood pressure based on the four-element elastic cavity model.
  • the evaluation of hemodynamic parameters is more accurate; the three-axis acceleration sensor realizes the direct measurement of cardiac pulsatility and is used to calibrate the amplitude of the pulse wave, making the blood pressure evaluation trend correlation positive and reducing the opposite trend evaluation. possibility.
  • Figure 1 is a flow chart of the cuffless ambulatory blood pressure measurement method based on multi-point calibration according to the present invention
  • Figure 2 is a schematic diagram of multiple groups of sensors
  • Figure 3 is a schematic diagram of the changes in PPG amplitude under different peripheral resistance
  • Figure 4 is a schematic diagram of the changes in PPG amplitude under different blood vessel elasticities
  • Figure 5 is the equivalent circuit diagram of the four-component elastic cavity model.
  • a component when referred to as being “fixed to” another component, it can be directly on the other component or another intermediate component may be present through which it is fixed.
  • a component When a component is said to be “connected” to another component, it can be directly connected to the other component or there may be another intermediate component present at the same time.
  • a component When a component is said to be “disposed on” another component, it can be directly located on the other component or another intervening component may be present.
  • the terms “vertical,” “horizontal,” “left,” “right” and similar expressions are used herein for illustrative purposes only.
  • Figure 1 is a flow chart of the cuffless dynamic blood pressure measurement method based on multi-point calibration of the present invention.
  • the cuffless dynamic blood pressure measurement method based on multi-point calibration of the present invention includes the following steps:
  • a cuffless dynamic blood pressure measurement method based on multi-point calibration including the following steps:
  • Data collection Use multiple sets of sensors to measure the contact pressure on the wrist, and use the contact pressure measured by multiple sets of sensors to subtract to obtain the contact pressure difference to offset the error caused by baseline drift;
  • V is the blood volume corresponding to the specified transmural pressure P transm
  • V 0 is the V value corresponding to zero P transm
  • V max is the maximum blood volume
  • C max is the maximum compliance at zero P transm
  • the contact pressure is equal to blood pressure (BP) minus the transmural pressure P transm
  • the contact pressure is equal to blood pressure (BP) minus the transmural pressure P transm
  • the contact pressure is equal to blood pressure (BP) minus the transmural pressure P transm
  • BP blood pressure
  • Measure cardiac output measure cardiac output CO through a three-axis acceleration sensor
  • the data collection steps are as follows: Set up multiple sets of sensors on the measurement end of the watch.
  • the multiple sets of sensors measure the contact pressure on the wrist.
  • the contact pressures measured by the multiple sets of sensors are subtracted to obtain the contact pressure difference to offset the baseline drift. error.
  • the design of two sets of light sources and two sets of sensors is taken as an example (as shown in Figure 2).
  • the specific steps to obtain the measured blood volume wave PPG are:
  • V is the blood volume corresponding to the specified transmural pressure P transm
  • V 0 is the V value corresponding to zero P transm
  • V max is the maximum blood volume
  • C max is the maximum compliance at zero P transm ; since the contact pressure difference is equal to the transmural Pressure difference, blood volume V is obtained through the pressure-volume conversion formula, thereby obtaining the measured blood volume wave PPG,
  • a calib can be used to evaluate peripheral circulation resistance and then average blood pressure.
  • FIG. 3 is a schematic diagram of the changes in PPG amplitude under different peripheral resistance.
  • C 1 1.3ml/mmHg
  • C 2 0.25ml/mmHg
  • L 0.01mmHg.s 2 /ml
  • T s 0.35s
  • ⁇ 1/3
  • Figure 4 is a schematic diagram of changes in PPG amplitude under different blood vessel elasticities.
  • C 2 0.25ml/mmHg
  • L 0.01mmHg.s 2 /ml
  • R 0.8mmHg.s/ml
  • T s 0.35s
  • ⁇ 1/3.
  • the optimal blood pressure solution obtained above is applicable to a certain range, and will produce large errors outside the specific range. Therefore, the real measured PPG needs to be combined with the simulated PPG to confirm the applicable range of the linear model of the measured PPG waveform.
  • the four-element elastic cavity model is used to generate the PPG waveform as follows: the four-element elastic cavity model is used to simulate the response of peripheral blood vessels; each variable included in the four-element elastic cavity model is discretely traversed, covering all possible parameter combinations, to generate Waveform-blood pressure database.
  • the heart is represented as a current source.
  • q in is related to left ventricular ejection
  • p 1 represents the central blood pressure
  • the arterial tree system through the lumped resistance (R), the lumped compliance of the central great arteries (C 1 ), the total compliance of the distal arteries (C 2 ) and
  • the blood flow inertia (L) is modeled.
  • the resistance to blood flow is the sum of all individual resistances in the microcirculation. Since our focus is on PPG and the compliance of the wrist vessels is negligible compared to C1 and C2 , two series resistors ( R1 and R2 ) are used at the peripheral site.
  • the blood pressure at the wrist is assumed to be p 3 , which is proportional to the peripheral blood pressure (p 2 ).
  • vascular compliance depends greatly on age. Blood flow inertia mainly depends on the cross-sectional area and length of blood vessels. The combination of peripheral vascular compliance and resistance determines the rate of PPG decay on the falling edge. In order to simplify the model, we used fixed parameters and assumed that the subject was in steady state, obtaining the formula
  • Ventricular ejection is approximated by equation (2), where q 0 is the maximum input amplitude related to cardiac output. q 0 is calculated by SV ⁇ /(120 ⁇ T s ).
  • SV refers to stroke volume.
  • T s is the duration of left ventricular ejection. T is the duration of the cardiac cycle, which is fixed at 0.8 s in the current study. We eliminate the influence of heart rate by fixing T because it is not directly related to PPG morphology.
  • ⁇ T s is the peak time during systole, and diastolic blood flow is set to 0.
  • the resistance to blood flow is the sum of all individual resistances in the microcirculation, resulting in the following equation:
  • the PPG waveform generated by peripheral blood pressure p2 is obtained by solving differential equation (3).
  • a waveform generated by a set of simulation parameters is considered realistic if it matches at least one reference measured waveform. Assuming that there are N reference measurements and a total of M simulations are performed, the following process is performed M ⁇ N times. If a match is found, the corresponding set of simulation parameters (including q 0 ) is recorded as a trusted combination. If an analog parameter group matches multiple reference waveforms, only one analog parameter group is recorded. It is worth noting that we only use the correlation of the PPG profile to avoid amplitude uncertainties caused by contact pressure and ambient temperature.
  • the matching process between simulated and measured waveforms is as follows: First, adjust the simulated blood pressure value to match the reference mean arterial pressure (MAP), which can be achieved through changes in q 0 . Since q 0 is determined by SV and T s , and SV has a certain range in the human body, we choose SV ⁇ [50ml-150ml] to limit the value range of q 0 .
  • the corresponding PPG waveform is calculated according to formula (1), interpolated to 81 points, and the amplitude is normalized to [0 1]. Each point of the waveform is represented as PPG sim (k), where k ranges from 1 to 81.
  • the measured PPG is standardized in the same way and is recorded as PPG ref (k).
  • PPGsim(k) is the calculated waveform point
  • PPGref(k) is the measured waveform point
  • n is the number of waveform points
  • the range of k is 1 to n.
  • n is 81. Calculate the average deviation. Only Simulated PPG waveforms less than 0.005 are considered realistic.
  • This application aims at the shortcomings of traditional sphygmomanometers such as low sampling rate and discomfort, as well as the shortcomings of wearable wristwatches where the optical volume wave amplitude is susceptible to interference.
  • This application achieves dynamic monitoring and compensation of pressure through multiple sensor groups; protects the PPG waveform and avoids interference. Due to distortion, the evaluation of hemodynamic parameters based on the four-element elastic cavity model is more accurate; the direct measurement of cardiac pulsation force is achieved through the three-axis acceleration sensor, and is used to calibrate the amplitude of the pulse wave, making the blood pressure assessment trend relevant Being positive reduces the possibility of opposite trend assessments.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Cardiology (AREA)
  • Engineering & Computer Science (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physiology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Vascular Medicine (AREA)
  • Biomedical Technology (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)

Description

一种基于多点校准的无袖带动态血压测量方法 技术领域
本发明涉及血压测量，尤其是涉及基于多点校准的无袖带动态血压测量方法。
背景技术
高血压能够引发心血管疾病、主动脉瘤、中风、周边血管疾病、慢性肾病等，具有高死亡风险。高血压患者在流行病中也属高危人群，譬如新型冠状病毒的感染可能会引发更严重的肺衰竭，病死率远高于其他患者。因此，高血压疾病的早期防控具有极为重大的意义。
确诊及管理高血压病需要持续监测血压，但是情绪波动、体力活动和睡眠均能够对血压产生影响。目前临床普遍使用的动态血压计为袖带式血压计，充放气操作容易干扰睡眠，加重受试者焦虑；测量频率低，每天仅有五十个左右的数据点，不能反映血压的瞬时涨落；最后，由于价格昂贵，准备工作繁琐，很难实现日常居家场景下的监测。
现有技术中有采用腕表实时测量血压，但无干扰性的腕部血压测量一般通过提取PPG特征，并通过多元线性回归或机器学习的方式进行血压拟合。PPG波形经常会由于其他因素干扰产生趋势反转；并且腕部血液容积波波形含信息量有限，尤其是压力不稳定或者温度低，情绪紧张的时候引起血管收缩，基线漂移严重，导致测量不准确。
发明内容
为了克服现有技术的不足，本发明的目的之一在于提供一种基于多点校准的无袖带动态血压测量方法，能够避免其他因素干扰产生趋势反转以及血管收缩导致的基线漂移。
本发明的目的之一采用如下技术方案实现：
一种基于多点校准的无袖带动态血压测量方法，包括以下步骤：
数据采集：用多组传感器测量腕部的接触压力，利用多组传感器测量的接触压力相减得到接触压力差的方式抵消基线漂移带来的误差；
得到测量血液容积波:由于接触压力差等于透壁压差，通过压力-体积转换公式
Figure PCTCN2022128483-appb-000001
得到测量血容积波PPG，V是指定透壁压P transm对应的血容量，V 0是零P transm对应的V值，V max是最大血容量；C max是零P transm时的最大顺应性；由于接触压力等于血压(BP)减去透壁压P transm，在未知参量有4个的时候，应用四个方程来求解，在多光源-传感器的设计中，方程数量超过未知量，可以进 一步减小误差；
判断是否需要调节腕表：当接触压力增大，脉动变小，则说明佩戴过紧，需要调整表带；当腕表不需要调节时，继续通过测量血容积波PPG拟合血压；
测量心输出量：通过三轴加速度传感器测量心输出量CO；
校准测量PPG幅值：在任一时刻，校准幅值A calib(t)＝A(t)/CO(t)*CO(t 0)，A(t)为PPG幅值，CO(t)为任一时刻心输出量，CO(t 0)为初始时刻心输出量；在公式(1)存在个体差异的情况下，三轴加速度测量可从独立模态增加血容积测量的准确性。
根据外周循环阻力和血管弹性获得测量血压最优解：不同外周循环阻力和不同血管弹性下测量PPG幅值会产生变化，根据外周循环阻力和血管弹性获得测量血压最优解；
确认测量血压最优解的适用范围：采用四元件弹性腔模型生成PPG波形，对生成和测量的PPG波形进行成对比较，用以确认测量的PPG波形的线性模型或机器学习模型的适用范围。
进一步的，在所述测量心输出量步骤中，设体重为m，心跳引起的腕部最大加速度为a，则心脏搏动输出的力正比于m*a，在PPG评估模型中，心输出量参数CO正比于m*a，设CO＝k*m*a，k为比例系数。
进一步的，在所述测量心输出量步骤中，三轴加速度传感器的分辨率设置为小于等于0.5mg。
进一步的，在数据采集步骤中，每一组所述传感器包括光源和探测器。
进一步的，所述确认测量血压最优解的适用范围步骤中采用四元件弹性腔模型生成PPG波形具体为：
采用四元件弹性腔模型来模拟周边血管的反应；
对四元件弹性腔模型包含的每一个变量进行离散遍历，涵盖所有可能的参数组合，生成波形-血压数据库。
进一步的，采用了四元件弹性腔模型来模拟周边血管的反应具体为：将心脏表示为电流源，左心室射血表示为q in，p 1代表中央血压，动脉树系统通过集总阻力为R，中央大动脉的集总顺应性为C 1，远端动脉的总顺应性为C 2和血流惯性为L进行建模，建模后
Figure PCTCN2022128483-appb-000002
q 0是与心输出量有关的最大输入振幅；q 0通过SV·π/(120·Ts)来计算，SV是指每搏心输出量，T s是左心室射血的持续时间，T是心动周期的持续时间，αT s是心脏收缩期间的峰值时间；
根据四元件弹性腔模型的等效电路中对血液流动的阻力是微循环中所有单个阻力的总和，得到以下方程：
Figure PCTCN2022128483-appb-000003
通过求解微分方程(3)获得外周血压p 2生成的PPG波形。
进一步的，所述确认测量血压最优解的适用范围步骤中对生成和测量的PPG波形进行成对比较，用以确认测量的PPG波形的线性模型的适用范围具体为：根据实测产生的测量PPG数据，选择能够真实存在的参数组合，形成模拟数据库的可信区间。
进一步的，所述根据实测产生的测量PPG数据，选择能够真实存在的参数组合，形成模拟数据库的可信区间具体为：对生成和测量的PPG波形进行成对比较，如果一组仿真参数生成的波形至少与一个参考测量波形相匹配，则被认为是可真实存在的；调整模拟血压值p 2以匹配参考平均动脉压,通过压力-体积转换公式计算相应的PPG波形，并将PPG波形幅度归一化后计算平均偏差，保留平均偏差小的测量PPG波形上的点。
进一步的，平均偏差
Figure PCTCN2022128483-appb-000004
PPGsim(k)为计算波形点，PPGref(k)为实测波形点，n为波形点数量，k的范围是1到n。
进一步的，平均偏差e小于0.005。
相比现有技术，本发明基于多点校准的无袖带动态血压测量方法通过多个传感器组实现了对压力的动态监测及补偿；保护PPG波形，免于扭曲，使得基于四元件弹性腔模型的血流动力学参数评估更为准确；通过三轴加速度传感器实现了对心脏搏动力的直接测量，并用以校准脉搏波的幅值，使得血压评估趋势相关性为正，降低了趋势评估相反的可能性。
附图说明
图1为本发明基于多点校准的无袖带动态血压测量方法的流程图；
图2为多组传感器示意图；
图3为不同外周阻力下PPG幅值的变化示意图；
图4为不同血管弹性下PPG幅值的变化示意图；
图5为四元件弹性腔模型的等效电路图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
需要说明的是，当组件被称为“固定于”另一个组件，它可以直接在另一个组件上或者也可以存在另一中间组件，通过中间组件固定。当一个组件被认为是“连接”另一个组件，它可以是直接连接到另一个组件或者可能同时存在另一中间组件。当一个组件被认为是“设置于”另一个组件，它可以是直接设置在另一个组件上或者可能同时存在另一中间组件。本文所使用的术语“垂直的”、“水平的”、“左”、“右”以及类似的表述只是为了说明的目的。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。
图1为本发明基于多点校准的无袖带动态血压测量方法的流程图，本发明基于多点校准的无袖带动态血压测量方法包括以下步骤：
一种基于多点校准的无袖带动态血压测量方法，包括以下步骤：
数据采集：用多组传感器测量腕部的接触压力，利用多组传感器测量的接触压力相减得到接触压力差的方式抵消基线漂移带来的误差；
得到测量血液容积波:通过压力-体积转换公式
Figure PCTCN2022128483-appb-000005
得到测量血容积波PPG，V是指定透壁压P transm对应的血容量，V 0是零P transm对应的V值，V max是最大血容量；C max是零P transm时的最大顺应性；由于接触压力等于血压(BP)减去透壁压P transm，在未知参量有4个的时候，应用四个方程来求解，在多光源-传感器的设计中，方程数量超过未知量，可以进一步减小误差；具体的，在两光源-两探测器的设计中，可实现4组方程，完成求解。在多光源-传感器的设计中，方程数量超过未知量，可以进一步减小误差。
判断是否需要调节腕表：当接触压力增大，脉动变小，则说明佩戴过紧，需要调整表带；当腕表不需要调节时，继续通过测量血容积波PPG拟合血压；
测量心输出量：通过三轴加速度传感器测量心输出量CO；
校准测量PPG幅值：在任一时刻，校准幅值A calib(t)＝A(t)/CO(t)*CO(t0)，A(t)为PPG幅值，CO(t)为任一时刻心输出量，CO(t 0)为初始时刻心输出量；在公式(1)存在个体差异的情况下，三轴加速度测量可从独立模态增加血容积测量的准确性。
根据外周循环阻力和血管弹性获得测量血压最优解：不同外周循环阻力和不同血管弹性下测量PPG幅值会产生变化，根据外周循环阻力和血管弹性获得测量血压最优解；
确认测量血压最优解的适用范围：采用四元件弹性腔模型生成PPG波形，对生成和测量的PPG波形进行成对比较，用以确认测量的PPG波形的线性模型或机器学习模型的适用范围。
数据采集步骤具体为：在腕表的测量端，设立多组传感器，多组传感器测量腕部的接触压力，利用多组传感器测量的接触压力相减得到接触压力差的方式抵消基线漂移带来的误差。在本实施例中，以2组光源，2组传感器的设计为例(如图2所示)。
得到测量血液容积波PPG步骤具体为：
在腕部，最适合的血容积-压力关系为
Figure PCTCN2022128483-appb-000006
V是指定透壁压P transm对应的血容量，V 0是零P transm对应的V值，V max是最大血容量；C max是零P transm时的最大顺应性；由于接触压力差等于透壁压差，通过压力-体积转换公式得到血容量V，从而得到测量血容积波PPG，
判断是否需要调节腕表：在压力适中的情况下，压力偏大，脉动变大，一般波动可通过公式1进行调整补偿；压力过大的时候，会出现血管坍缩，压力越大，脉动反而变小。我们根据S1D1、S1D2、S2D1、S2D2等四对信号进行分析，如果发现接触压力增大，脉动变小，则说明佩戴过紧，影响血液循环，必须要调整压力，使得血流通畅。当腕表不需要调节时，继续通过测量血容积波PPG拟合血压。
测量心输出量步骤具体为：
由于PPG的幅值极易受到压力、温度等因素影响，进而影响血压评估的准确度，我们采用更为直接的力学测量方式，通过三轴加速度传感器评估心脏搏动输出的力，并用以校准测量PPG的幅值。校准流程如下：
(1)设体重为m，心跳引起的腕部最大加速度为a，则心脏搏动输出的力正比于m*a，在PPG评估模型中，心输出量参数(cardiac output,CO)正比于m*a，设CO＝k*m*a，k为比例系数。
(2)设在初始时刻t 0,血压为P 0，PPG的幅值为A 0。
(3)在任一时刻，校准幅值A calib(t)＝A(t)/CO(t)*CO(t 0)；A calib可用来评估外周循环阻力，进而评估平均血压。
根据外周循环阻力和血管弹性获得测量血压最优解具体为：
平均血压MAP＝CO/R＝k*m*a/R，其中MAP为平均血压，R为外周循环阻力。血压脉压差PP与主动脉血管弹性C 1相关，与A calib密切相关。因此经心脏动力校准的PPG，其测量更为准确，对趋势的预测更偏正相关。图3为不同外周阻力下PPG幅值的变化示意图，图中C 1＝1.3ml/mmHg，C 2＝0.25ml/mmHg，L＝0.01mmHg.s 2/ml，T s＝0.35s，α＝1/3。图4为不同血管弹性下PPG幅值的变化示意图，图中C 2＝0.25ml/mmHg，L＝0.01mmHg.s 2/ml，R＝0.8mmHg.s/ml，T s＝0.35s，α＝1/3。
确认测量血压最优解的适用范围具体为：
上述获得的血压最优解适用于一定范围，在特定范围外会产生较大误差，因此真实测量的PPG需要和仿真模拟的PPG相结合，确认测量的PPG波形的线性模型的适用范围。
具体的，采用四元件弹性腔模型生成PPG波形为：采用四元件弹性腔模型来模拟周边血管的反应；对四元件弹性腔模型包含的每一个变量进行离散遍历，涵盖所有可能的参数组合，生成波形-血压数据库。
如图5所示，心脏被表示为电流源。q in与左心室射血相关，p 1代表中央血压，动脉树系统通过集总阻力(R)，中央大动脉的集总顺应性(C 1)，远端动脉的总顺应性(C 2)和血流惯性(L)进行建模。根据Poiseuille定律，对血液流动的阻力是微循环中所有单个阻力的总和。由于我们的重点是PPG，与C 1和C 2相比，腕部血管的顺应性可以忽略，因此在外围部位使用了两个串联电阻(R 1和R 2)。腕部的血压假定为p 3，与外围血压(p 2)成正比。血管顺应性很大程度上取决于年龄。血流惯性主要取决于血管的横截面积和长度。外周血管顺应性和阻力的结合决定了PPG的下降沿衰减速率。为了简化模型，我们使用了固定参数，假设受试者处于稳态，得到公式
Figure PCTCN2022128483-appb-000007
心室射血由公式(2)近似，其中q 0是与心输出量有关的最大输入振幅。q 0通过SV·π/(120·T s)来计算。此处SV是指每搏心输出量。T s是左心室射血的持续时间。T是心动周期的持续时间，在当前研究中固定为0.8s，我们通过固定T来排除心率的影响，因为它与PPG形态没有直接关系。αT s是心脏收缩期间的峰值时间，舒张期血流设为0。
根据四元件弹性腔模型的等效电路中对血液流动的阻力是微循环中所有单个阻力的总和，得到以下方程：
Figure PCTCN2022128483-appb-000008
通过求解微分方程(3)获得外周血压p 2生成的PPG波形。
对生成和测量的PPG波形进行成对比较，用以确认测量的PPG波形的线性模型的适用范围：根据实测产生的测量PPG数据，选择能够真实存在的参数组合，形成模拟数据库的可信区间。
对生成和测量的波形进行成对比较。如果一组仿真参数生成的波形至少与一个参考测量波形相匹配，则被认为是可真实存在的。假设有N次参考测量，且总共进行了M次模拟，以下过程进行了M×N次。如果找到匹配项，则将相应的一组模拟参数(包括q 0)记录为可信组合。如果一个模拟参数组匹配多个参考波形，仅记录一组模拟参数。值得注意的是，我们仅使用PPG轮廓的相关性，以避免由接触压力和环境温度引起的幅值不确定性。
模拟和实测波形的匹配流程如下：首先，调整模拟血压值以匹配参考平均动脉压(MAP)，可通过q 0的变化来达成。由于q 0由SV和T s决定，并且SV在人体中具有一定的范围，因此我们选择SV∈[50ml-150ml]来限制q 0的取值范围。根据公式(1)计算相应的PPG波形，将其内插值到81个点，并将幅度归一化为[0 1]。波形的每个点表示为PPG sim(k)，k的范围是1到81。实测PPG的标准化处理方式相同，记为PPG ref(k)。我们按照公式
Figure PCTCN2022128483-appb-000009
Figure PCTCN2022128483-appb-000010
PPGsim(k)为计算波形点，PPGref(k)为实测波形点，n为波形点数量，k的范围是1到n。在本实施例中，n为81。计算平均偏差。只有
Figure PCTCN2022128483-appb-000011
小于0.005的模拟PPG波形才被认为是可真实存在的。
本申请针对传统血压计采样率低，不舒适的缺点以及可穿戴腕表光学容积波幅值易受干扰的缺点，通过多个传感器组实现了对压力的动态监测及补偿；保护PPG波形，免于扭曲，使得基于四元件弹性腔模型的血流动力学参数评估更为准确；通过三轴加速度传感器实现了对心脏搏动力的直接测量，并用以校准脉搏波的幅值，使得血压评估趋势相关性为正，降低了趋势评估相反的可能性。
以上实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进演变，都是依据本发明实质技术对以上实施例做的等同修饰与演变，这些都属于本发明的保护范围。

Claims (10)

  1. 一种基于多点校准的无袖带动态血压测量方法，其特征在于，包括以下步骤：
    数据采集：用多组传感器测量腕部的接触压力，利用多组传感器测量的接触压力相减得到接触压力差的方式抵消基线漂移带来的误差；
    得到测量血液容积波:由于接触压力差等于透壁压差，通过压力-体积转换公式
    Figure PCTCN2022128483-appb-100001
    得到测量血容积波PPG，V是指定透壁压P transm对应的血容量，V 0是零P transm对应的V值，V max是最大血容量；C max是零P transm时的最大顺应性；由于接触压力等于血压(BP)减去透壁压P transm，在未知参量有4个的时候，应用四个方程来求解，在多光源-传感器的设计中，方程数量超过未知量，可以进一步减小误差；
    判断是否需要调节腕表：当接触压力增大，脉动变小，则说明佩戴过紧，需要调整表带；当腕表不需要调节时，继续通过测量血容积波PPG拟合血压；
    测量心输出量：通过三轴加速度传感器测量心输出量CO；
    校准测量PPG幅值：在任一时刻，校准幅值A calib(t)＝A(t)/CO(t)*CO(t 0)，A(t)为PPG幅值，CO(t)为任一时刻心输出量，CO(t 0)为初始时刻心输出量；在公式(1)存在个体差异的情况下，三轴加速度测量可从独立模态增加血容积测量的准确性；
    根据外周循环阻力和血管弹性获得测量血压最优解：不同外周循环阻力和不同血管弹性下测量PPG幅值会产生变化，根据外周循环阻力和血管弹性获得测量血压最优解；
    确认测量血压最优解的适用范围：采用四元件弹性腔模型生成PPG波形，对生成和测量的PPG波形进行成对比较，用以确认测量的PPG波形的线性模型或机器学习模型的适用范围。
  2. 根据权利要求1所述的基于多点校准的无袖带动态血压测量方法，其特征在于：在所述测量心输出量步骤中，设体重为m，心跳引起的腕部最大加速度为a，则心脏搏动输出的力正比于m*a，在PPG评估模型中，心输出量参数CO正比于m*a，设CO＝k*m*a，k为比例系数。
  3. 根据权利要求1所述的基于多点校准的无袖带动态血压测量方法，其特征在于：在所述测量心输出量步骤中，三轴加速度传感器的分辨率设置为小于等于0.5mg。
  4. 根据权利要求1所述的基于多点校准的无袖带动态血压测量方法，其特征在于：在数据采集步骤中，每一组所述传感器包括光源和探测器。
  5. 根据权利要求1所述的基于多点校准的无袖带动态血压测量方法，其特征在于：所述确认测量血压最优解的适用范围步骤中采用四元件弹性腔模 型生成PPG波形具体为：
    采用四元件弹性腔模型来模拟周边血管的反应；
    对四元件弹性腔模型包含的每一个变量进行离散遍历，涵盖所有可能的参数组合，生成波形-血压数据库。
  6. 根据权利要求5所述的基于多点校准的无袖带动态血压测量方法，其特征在于：采用了四元件弹性腔模型来模拟周边血管的反应具体为：将心脏表示为电流源，左心室射血表示为q in，p 1代表中央血压，动脉树系统通过集总阻力为R，中央大动脉的集总顺应性为C 1，远端动脉的总顺应性为C 2和血流惯性为L进行建模，建模后
    Figure PCTCN2022128483-appb-100002
    q 0是与心输出量有关的最大输入振幅；q 0通过SV·π/(120·T s)来计算，SV是指每搏心输出量，T s是左心室射血的持续时间，T是心动周期的持续时间，αT s是心脏收缩期间的峰值时间；
    根据四元件弹性腔模型的等效电路中对血液流动的阻力是微循环中所有单个阻力的总和，得到以下方程：
    Figure PCTCN2022128483-appb-100003
    通过求解微分方程(3)获得外周血压P 2生成的PPG波形。
  7. 根据权利要求5所述的基于多点校准的无袖带动态血压测量方法，其特征在于：所述确认测量血压最优解的适用范围步骤中对生成和测量的PPG波形进行成对比较，用以确认测量的PPG波形的线性模型或机器学习模型的适用范围具体为：根据实测产生的测量PPG数据，选择能够真实存在的参数组合，形成模拟数据库的可信区间。
  8. 根据权利要求7所述的基于多点校准的无袖带动态血压测量方法，其特征在于：所述根据实测产生的测量PPG数据，选择能够真实存在的参数组合，形成模拟数据库的可信区间具体为：对生成和测量的PPG波形进行成对比较，如果一组仿真参数生成的波形至少与一个参考测量波形相匹配，则被认为是可真实存在的；调整模拟血压值p 2以匹配参考平均动脉压,通过压力-体积转换公式计算相应的PPG波形，并将PPG波形幅度归一化后计算平均偏差，保留平均偏差小的测量PPG波形上的点。
  9. 根据权利要求8所述的基于多点校准的无袖带动态血压测量方法，其特征在于：平均偏差
    Figure PCTCN2022128483-appb-100004
    PPGsim(k)为计算波形点，PPGref(k)为实测波形点，n为波形点数量，k的范围是1到n。
  10. 根据权利要求9所述的基于多点校准的无袖带动态血压测量方法，其特征在于：平均偏差e小于0.005。
PCT/CN2022/128483 2022-08-09 2022-10-31 一种基于多点校准的无袖带动态血压测量方法 WO2024031845A2 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210952060.8 2022-08-09
CN202210952060.8A CN115336990B (zh) 2022-08-09 2022-08-09 一种基于多点校准的无袖带动态血压测量方法

Publications (1)

Publication Number Publication Date
WO2024031845A2 true WO2024031845A2 (zh) 2024-02-15

Family

ID=83952198

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/128483 WO2024031845A2 (zh) 2022-08-09 2022-10-31 一种基于多点校准的无袖带动态血压测量方法

Country Status (2)

Country Link
CN (1) CN115336990B (zh)
WO (1) WO2024031845A2 (zh)

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1698536A (zh) * 2004-05-20 2005-11-23 香港中文大学 采用自动补偿的无袖带式连续血压测量方法
CN108523867B (zh) * 2018-03-28 2020-12-29 武汉麦咚健康科技有限公司 一种自校准ppg无创血压测量方法及系统
KR20200054719A (ko) * 2018-11-12 2020-05-20 삼성전자주식회사 혈압 캘리브레이션 시점 검출 방법 및 장치
CN111870237B (zh) * 2020-09-04 2022-08-16 平安科技（深圳）有限公司 血压检测装置、设备及介质
KR20220107909A (ko) * 2021-01-25 2022-08-02 삼성전자주식회사 혈압 추정 장치 및 방법

Also Published As

Publication number Publication date
CN115336990A (zh) 2022-11-15
CN115336990B (zh) 2023-08-25

Similar Documents

Publication Publication Date Title
US20210212575A1 (en) Method of measuring central pressure waveform with features preserved
JP6904796B2 (ja) 生体情報推定装置及び方法並びに血圧情報推定装置
US11298029B2 (en) Blood pressure measuring apparatus, blood pressure measuring method, electronic device, and computer readable storage medium
CN105943005B (zh) 基于光电绿光脉搏与心电图混合的无创血压检测装置
Fortino et al. PPG-based methods for non invasive and continuous blood pressure measurement: An overview and development issues in body sensor networks
US6758822B2 (en) Method and apparatus for measuring cardiac output
CN106889979B (zh) 一种基于心电信号和血氧容积波的连续性血压测量系统
JP5985355B2 (ja) 血液量測定方法および測定装置
US20060074322A1 (en) Measuring systolic blood pressure by photoplethysmography
Marzorati et al. Chest wearable apparatus for cuffless continuous blood pressure measurements based on PPG and PCG signals
CN106618504A (zh) 一种具备自适应校准的高准确度连续无创血压测量装置
TW201400086A (zh) 一種藉壓脈帶壓力脈波震盪訊號以預估中央動脈脈搏壓的方法及其裝置
JP6670376B2 (ja) 心弾動図（ｂｃｇ波）の複数の基準点の間で測定された時間差から大動脈の脈拍の伝搬時間（大動脈ｐｔｔ）を見積もる方法と装置。
Shao et al. An optimization study of estimating blood pressure models based on pulse arrival time for continuous monitoring
WO2024031845A2 (zh) 一种基于多点校准的无袖带动态血压测量方法
TWI615127B (zh) 計算血壓的方法及其裝置
Bose et al. Improving the performance of continuous non-invasive estimation of blood pressure using ECG and PPG
US20190110702A1 (en) A Method Of Deriving Systolic Blood Pressure And/Or Diastolic Blood Pressure Of A Subject
Sharman et al. Measurements of arterial pressure and flow in vivo
JPS61247431A (ja) 非観血連続血圧計
EP4173555A1 (en) Device, system and method for calibrating a blood pressure surrogate for use in monitoring a subject's blood pressure
EP4173556A1 (en) Device, system and method for calibrating a blood pressure surrogate for use in monitoring a subject's blood pressure
CN108309272A (zh) 一种基于服务器的云计算血压计
Ozsahin et al. The design of a noninvasive blood pressure measurement device
Jobbágy et al. The Effect of Occlusion with the Cuff

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22954785

Country of ref document: EP

Kind code of ref document: A2