RU2782412C2 - Method and device for quantitative assessment of tendency in surrogate blood pressure indicator - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: group of inventions relates to medicine, namely to the quantitative assessment of a tendency in a surrogate blood pressure indicator. A device containing a machine-readable medium is proposed for implementation of a method including obtaining a set of measurement values of a surrogate blood pressure indicator for a subject, characterized in that the method additionally includes obtaining for each measurement value of the surrogate blood pressure indicator an error value indicating a measurement error for the measurement value of the surrogate blood pressure indicator, and analysis of the specified set of measurement values of the surrogate blood pressure indicator and corresponding error values, using Bayesian inference to determine the tendency in the surrogate blood pressure indicator over time. In this case, the stage of obtaining the set of measurement values of the surrogate blood pressure indicator includes for each measurement value of the surrogate blood pressure indicator: obtaining measurement values of the surrogate blood pressure indicator over a time interval; assessment of measurement values of the surrogate blood pressure indicator, obtained within the time interval to determine, whether the surrogate blood pressure indicator is in a stable state; and, if the surrogate blood pressure indicator is not in the stable state, extrapolation of measurement values of the surrogate blood pressure indicator, obtained within the time interval for quantitative assessment of the measurement value of the surrogate blood pressure indicator in the stable state.

EFFECT: group of inventions provides the possibility of more accurate measurement of characteristics of a physiological state of a subject due to improvement of the quantitative assessment of a tendency obtained from measurements of a surrogate blood pressure indicator, in particular, if measurements of the surrogate blood pressure indicator may have measurement errors.

13 cl, 13 dwg

Description

FIELD OF TECHNOLOGY

This description relates to the analysis of a measurement that can be used as a surrogate measure for a subject's blood pressure, and in particular relates to a method and apparatus for quantifying a trend in a surrogate blood pressure measure.

BACKGROUND OF THE INVENTION

The so-called quantitative self-measurement uses technology to measure and collect data and information about a person's daily activities, such as movements, food consumed, amount of exercise, mood, etc. The realization that human skin characteristics can provide monitoring of key physiological parameters (such as heart rate) has led to the introduction of wearable devices such as smartwatches and wristbands to the market.

Along with these wearable devices, so-called smart mirrors are being considered, in which a standard mirror (for example, for use in a bathroom environment) is equipped with one or more sensors to enable the measurement of a number of basic physiological parameters by observing the skin at a certain distance (i.e., from a distance). i.e. non-contact measurement) or through the use of touch (i.e. contact measurement, eg via touch).

Wearable devices and smart mirrors that are designed to provide health-related data typically only measure (or enable measurement) a person's heart rate, heart rate variability, respiratory rate, and oxygen saturation (SpO2) levels.

Blood pressure is another useful human physiological characteristic that can be measured, and the use of wearable devices or smart mirrors to measure blood pressure in an unobtrusive manner is desirable. Typically, obtaining a blood pressure measurement at a distance (ie, non-contact) is difficult, so one or more other physiological characteristics are measured instead, which may be indicative of a subject's blood pressure or indicative of a trend in blood pressure over time. One such “surrogate” physiological characteristic is the pulse wave velocity (PWV). Pulse wave velocity (PWV) is the speed at which a pressure pulse propagates through the human arterial tree and can be measured by observing the time it takes the pulse to travel between two measurement points (known as the pulse transit time (PTT). )) that are at a known distance from each other. The speed of propagation of the pulse wave is obtained by dividing the distance between the measurement points by the propagation time of the pulse wave. By itself, the speed of propagation of the pulse wave is an indicator of arterial pathologies and the general state of the cardiovascular system.

Pulse wave velocity is considered a useful indicator of human blood pressure, and therefore, pulse wave velocity is considered a desirable characteristic for use in an easy-to-use and unobtrusive blood pressure monitoring device. In particular, it is believed that there is a relationship between a person's pulse wave velocity trend and a person's blood pressure trend.

There are several approaches used to measure pulse wave velocity in a non-invasive way. A common approach relies on photoplethysmography (PPG) at two body sites to detect changes in blood volume as pulse signals, with the pulse wave velocity calculated from the distance between the sensors divided by the time difference between the detected pulses. In another embodiment, electrocardiography (ECG) is used in combination with another sensor (for example, PPG, ultrasound, impedance cardiography, etc.), which measures the arrival time of the pulse wave (pulse arrival time, PAT) (i.e., time, for which the pulse reaches a certain point) and calculate the time of arrival of the pulse wave by measuring the time difference between the peak R of the ECG signal and the characteristic points of the signal received by another sensor.

Patent application WO 2017/156084 A2 discloses systems and methods for tracking ballistocardiogram, photoplethysmogram, blood pressure, and irregular heartbeats based on optical imaging of the human body. The ballistocardiogram and photoplethysmogram signals from a similar area of the human body are obtained simultaneously and the delay time between the two signals is used to determine the patient's blood pressure.

Patent application US 2017/0164904 A1 discloses a method for obtaining information about the propagation time of the pulse wave and/or the velocity of the pulse wave. Based on the combination of image frames and the detected movement of body parts, target areas are selected from other fixed body parts and receive information about the time of propagation of the pulse wave and/or the speed of propagation of the pulse wave from the received PPG signals and a certain physical distance between the target areas.

US 9,462,980 B2 discloses a system that predictively predicts a life-threatening condition, with generation occurring based on measurement values of a variety of different vital signs parameters.

DISCLOSURE OF THE INVENTION

As stated above, measurements of pulse wave arrival time, pulse wave propagation time, and/or pulse wave velocity (or measurements of other physiological characteristics) obtained as a function of time can be used as a trend indicator of blood pressure (which may be specifically conditioned) . One way is to determine the trend in blood pressure by applying the classical least-mean-square-error (least-means-square, LMS) method to measurements of pulse wave arrival time, pulse wave propagation time, and/or pulse wave velocity. However, any measurement of a physiological characteristic is accompanied by measurement errors that affect the reliability of the results. This is especially true when using the minimum standard error method to determine the blood pressure trend from the trend in pulse wave arrival time, pulse wave propagation time, and/or pulse wave velocity, since the reliability of the measurements is usually not taken into account.

Measurement errors or measurement uncertainty (i.e. the range in which the actual measurement value can lie) are especially problematic for non-contact measurements and for measurements taken in a short time period (for example, when a person is standing in front of a mirror).

Therefore, there is a need to improve the quantification of the trend derived from surrogate blood pressure measurements (such as pulse wave arrival time, pulse wave propagation time, and/or pulse wave velocity), in particular where surrogate blood pressure measurements may have errors. measurements.

According to a first specific aspect, there is provided a computer-assisted method for quantifying a trend in a surrogate blood pressure score, comprising obtaining a population of surrogate blood pressure measurement values for a subject; for each surrogate blood pressure measurement value, obtaining an error value that indicates a measurement error for the surrogate blood pressure measurement value; and analyzing said set of surrogate blood pressure measurement values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measure over time. This method improves the trend quantification obtained from surrogate blood pressure measurements, where surrogate blood pressure measurements may have measurement errors.

In some embodiments, the step of analyzing includes adjusting each of the surrogate blood pressure measurement values in said population according to the point in the subject's circadian rhythm at which each surrogate blood pressure measurement value was measured; and using Bayesian inference to determine a trend in the surrogate blood pressure measure over time by analyzing the adjusted surrogate blood pressure measurement values and the corresponding error values.

In these embodiments, the adjustment step may include obtaining measurements of a surrogate measure of the subject's circadian rhythm; analysis of measurements of a surrogate indicator of the circadian rhythm to identify the circadian rhythm of the subject; based on the identified circadian rhythm, determining corresponding points in the circadian rhythm at which each of the surrogate blood pressure measurement values in said population was measured; determining an appropriate adjustment for each of the surrogate blood pressure measurement values in said population based on the determined corresponding points in the circadian rhythm; and applying certain appropriate adjustments to the surrogate blood pressure measurement values in said population.

In these embodiments, the step of determining an appropriate adjustment may be based on (i) a linear function of the circadian rhythm surrogate and the blood pressure surrogate; (ii) a mapping function relating the surrogate circadian rate to the surrogate blood pressure; (iii) a predetermined set of adjustments for corresponding points in the circadian rhythm; or (iv) data for a population of other subjects.

In these embodiments, the circadian rhythm surrogate may be heart rate, body temperature, and/or physical activity level.

In some embodiments, the step of obtaining a plurality of surrogate blood pressure measurement values for each surrogate blood pressure measurement value includes: obtaining a plurality of surrogate blood pressure measurement values over a time period; evaluating the measurement values of the surrogate blood pressure indicator obtained during the time period to determine whether the surrogate blood pressure indicator is in a steady state; and including a measurement value of the surrogate blood pressure indicator obtained when the surrogate blood pressure indicator is in a steady state in the set of measurement values of the surrogate blood pressure indicator.

In alternative embodiments, the step of obtaining a plurality of surrogate blood pressure measurement values for each surrogate blood pressure measurement value includes: obtaining a plurality of surrogate blood pressure measurement values over a time period; evaluating the surrogate blood pressure measurement values obtained during the time period to determine if the surrogate blood pressure is in a steady state; and, if the surrogate blood pressure is not in a steady state, extrapolating the measurement values of the surrogate blood pressure obtained during the time period to quantify the measurement value of the surrogate blood pressure at steady state.

In some embodiments, the method further includes the step of providing instructions to the subject to perform a procedure or exercise prior to or during the surrogate blood pressure measurement.

In these embodiments, the method may further include: observing the subject to determine if the subject is performing the procedure or exercise correctly; and providing the subject with feedback or additional instructions to help the subject perform the procedure or exercise correctly.

In some embodiments, the surrogate blood pressure is a pulse wave arrival time (PAT) and the analysis step includes: obtaining measurements of the subject's heart rate; quantifying the pre-ejection period (PEP) for each measured value of the arrival time of the pulse wave from the heart rate at the time of measuring the arrival time of the pulse wave; subtracting the pre-arrival phase score from the corresponding measured pulse wave arrival time value to determine a set of corrected pulse wave arrival time measurement values; and using Bayesian inference to determine a trend in pulse wave arrival time over time by analyzing the corrected pulse wave arrival time measurement values and corresponding error values.

In alternative embodiments, the surrogate blood pressure is any one of pulse wave velocity (PWV), pulse wave arrival time (PAT) and pulse wave propagation time (PTT), heart rate, blood volume distribution, pulse wave morphology change, and general body movements.

In some embodiments, the step of obtaining a plurality of surrogate blood pressure measurement values includes obtaining surrogate blood pressure measurement values using one or more of photoplethysmographic, PPG, sensor, and electrocardiographic, ECG, sensor, accelerometer, image capture device, video capture device, radar transducer, laser Doppler transducer, interference transducer, and moiré transducer.

In some embodiments, the method is implemented by a smart mirror containing one or more sensors to obtain a set of surrogate blood pressure measurement values.

In some embodiments, the method further includes the step of using the determined trend in the surrogate blood pressure over time as an indicator of the trend in blood pressure.

According to a second aspect, a computer program product is provided, comprising a computer-readable medium containing computer-readable code implemented therein, the computer-readable code being configured to, when executed by a suitable computer or processor, cause the computer or processor to execute the method in accordance with the first aspect or any embodiment thereof.

According to a third specific aspect, there is provided a device for quantifying a trend in a surrogate blood pressure measure, the device comprising a processing unit configured to obtain a plurality of measurement values of the surrogate blood pressure measure for a subject; for each surrogate blood pressure measurement value, obtaining an error value indicative of a measurement error for the surrogate blood pressure measurement value; and analyzing the population of surrogate blood pressure measurement values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measure over time. This device provides improved trend quantification derived from surrogate blood pressure measurements, where surrogate blood pressure measurements may have measurement errors.

In some embodiments, the processing unit is configured to analyze a population of surrogate blood pressure measurement values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measurement over time by adjusting each of the surrogate blood pressure measurement values in said population according to with the point in the subject's circadian rhythm at which each surrogate blood pressure measurement value was measured; and using Bayesian inference to determine a trend in the surrogate blood pressure measure over time by analyzing the adjusted surrogate blood pressure measurement values and the corresponding error values.

In these embodiments, the processing unit is configured to correct each of the surrogate blood pressure measurements in said population by obtaining measurements of the subject's surrogate circadian rate; analyzing surrogate circadian rate measurements to identify the subject's circadian rhythm; based on the identified circadian rhythm, determining corresponding points in the circadian rhythm at which each of the surrogate blood pressure measurement values in said population was measured; determining an appropriate adjustment for each of the surrogate blood pressure measurement values in said population based on the determined corresponding points in the circadian rhythm; and applying certain appropriate adjustments to the surrogate blood pressure measurement values in said population.

In these embodiments, the processing unit may be configured to determine an appropriate adjustment based on (i) a linear function of the surrogate circadian rate score and the surrogate blood pressure score; (ii) a mapping function relating the surrogate circadian rate to the surrogate blood pressure; (iii) a predetermined set of adjustments for corresponding points in the circadian rhythm; or (iv) data for a population of other subjects.

In these embodiments, the circadian rhythm surrogate may be heart rate, body temperature, and/or physical activity level.

In some embodiments, the processing unit is configured to obtain a plurality of surrogate blood pressure measurement values by performing the following for each surrogate blood pressure measurement value: obtaining a plurality of surrogate blood pressure measurement values over a time period; evaluating the surrogate blood pressure measurement values obtained during the time period to determine if the surrogate blood pressure is in a steady state; and including a measurement value of the surrogate blood pressure indicator obtained when the surrogate blood pressure indicator is at steady state in the set of measurement values of the surrogate blood pressure indicator.

In alternative embodiments, the processing unit is configured to obtain a plurality of surrogate blood pressure measurement values, which for each surrogate blood pressure measurement value includes: obtaining a plurality of surrogate blood pressure measurement values over a time period; evaluating the surrogate blood pressure measurement values obtained during the time period to determine if the surrogate blood pressure is in a steady state; and if the surrogate blood pressure is not in a steady state, extrapolating the measurement values of the surrogate blood pressure obtained during the time period to quantify the measurement value of the surrogate blood pressure at steady state.

In some embodiments, the processing unit is further configured to provide instructions to the subject, via a user interface, to perform a procedure or exercise prior to or during the surrogate blood pressure measurement.

In these embodiments, the processing unit may be further configured to observe the subject to determine if the subject is performing the procedure or exercise correctly; and providing the subject with feedback or additional instructions through the user interface to help the subject complete the procedure or exercise correctly.

In some embodiments, the surrogate blood pressure measure is a pulse wave arrival time (PAT) and the processing unit is configured to analyze the population of surrogate blood pressure measure values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measure over time. by: obtaining measurements of the subject's heart rate; quantifying the pre-ejection period (PEP) for each measured value of the arrival time of the pulse wave from the heart rate at the time of measuring the arrival time of the pulse wave; subtracting the pre-arrival phase score from the corresponding measured pulse wave arrival time value to determine a set of corrected pulse wave arrival time measurement values; and using Bayesian inference to determine a trend in pulse wave arrival time over time by analyzing the corrected pulse wave arrival time measurement values and corresponding error values.

In alternative embodiments, the surrogate blood pressure is any one of pulse wave velocity (PWV), pulse wave arrival time (PAT) and pulse wave propagation time (PTT), heart rate, blood volume distribution, pulse wave morphology change, and general body movements.

In some embodiments, the processing unit is configured to obtain a plurality of surrogate blood pressure measurement values by obtaining surrogate blood pressure measurement values using one or more of a photoplethysmographic, PPG, sensor, and electrocardiographic, ECG, sensor, accelerometer, image capture device, device video capture, radar sensor, laser doppler sensor, interference sensor, and moiré sensor.

In some embodiments, the device is a smart mirror containing one or more sensors for obtaining a set of surrogate blood pressure measurement values.

In some embodiments, the processing unit is further configured to use a determined trend in the surrogate blood pressure over time as an indicator of the trend in blood pressure.

These and other aspects will be understood from and explained with reference to the implementation(s) described herein below.

BRIEF DESCRIPTION OF THE DRAWINGS

Exemplary embodiments will be described below, by way of example only, with reference to the accompanying drawings, in which:

In FIG. 1 shows two graphs depicting different approaches for trending from a set of measurements;

In FIG. 2 shows a block diagram of a device according to various implementations;

In FIG. 3 shows an image of a device in the form of a smart mirror according to various implementations;

In FIG. 4 is a flow chart showing a method in accordance with various embodiments;

In FIG. 5 is a graph depicting the effect of the circadian rhythm on blood pressure;

In FIG. 6 is a graph depicting changes in pulse wave arrival times over time in response to physical activity;

In FIG. 7 shows two graphs depicting the effect of respiration on the arrival time of the pulse wave;

In FIG. 8 is a graph showing pulse wave arrival time versus heart rate for a set of measurements;

In FIG. 9 is a graph showing the difference between pulse wave arrival time and pre-ejection period (PEP) versus heart rate for a set of measurements;

In FIG. 10 is a graph showing heart rate versus time for a set of measurements;

In FIG. 11 is a graph showing systolic blood pressure versus time for a set of measurements;

In FIG. 12 is a graph showing pulse wave arrival time versus time for a set of measurements; and

In FIG. 13 is a graph showing the difference between the arrival time of the pulse wave and the pre-trigger phase as a function of time for a set of measurements.

IMPLEMENTATION OF THE INVENTION

As used herein, the term "surrogate blood pressure" is any physiological or body characteristic whose measurement can be indicative of a subject's blood pressure or indicative of changes in a subject's blood pressure over time (i.e., a trend). As stated above, one surrogate measure of blood pressure is pulse wave velocity (PWV). Other surrogate blood pressure measures that may be used in accordance with the embodiments described herein include, for example, pulse wave arrival time (PAT), pulse wave propagation time (PTT), heart rate, blood volume distribution (e.g., in the face), changes in pulse wave morphology, general body movements (eg measured using a ballistocardiogram). Thus, measurements of pulse wave arrival time, pulse wave propagation time, and/or pulse wave velocity (or other physiological characteristic) obtained over time can be used as a trend indicator in blood pressure, although the measurement errors associated with measuring these surrogates blood pressure readings can significantly affect the reliability of the resulting trend. Measurement errors or measurement uncertainty (i.e. the range in which the actual measurement value may lie) are especially problematic for non-contact measurements and for measurements taken in a short time period (for example, when a person is standing in front of a mirror), such as measurements taken when the subject is briefly in front of the smart mirror.

One way to determine the trend in a population of measurements is to use the minimum standard error method. However, the MSE method does not take into account the reliability (or error) of the measurements, and this may reduce the accuracy of the trend line. In contrast, Bayesian inference takes measurement error into account and can therefore provide a more reliable trend indication. The Bayesian inference approach allows an objectified way to quantitatively distinguish a linear trend from a constant one. The differences between the minimum mean square error method and the Bayesian method are shown in FIG. one.

и базовые статистические аспекты с

, например, как отношение вероятности

, полученное из:In FIG. 1(a) shows an exemplary set of ten measurements, with each measurement having a corresponding error line representing the measurement error (uncertainty) for each measurement. As shown, the error lines differ between different measurements. The main trend in measurements in FIG. 1(a) is positive (i.e. the measurements generally increase). In FIG. 1(b) shows two different trend lines derived from the set of measurements in FIG. 1(a). The first line, labeled 2, is derived using a standard minimum standard error procedure and indicates a negative trend. The second line, labeled 4, is derived using Bayesian inference and takes into account measurement error (error lines). In particular, the Bayesian inference method uses a much smaller measurement error for the first and ninth measurements (counting from the left side of the graph) to (correctly) identify a positive measurement trend. Additionally, to distinguish the underlying trend in a population of measurements, the Bayesian approach can provide a quantified value as a probability, for example, a linear trend versus constant values ("probability ratio"). The value of this probability ratio can be used as a threshold. It can be expressed in terms of the obtained measurements

and basic statistical aspects with

, for example, as the ratio of the probability

obtained from:

, причём si аппроксимация стандартного отклонения для наблюдения y_i, Δx получено из x_i минус среднее значение всех x_i, а ΔD определено как y_i минус среднее значение всех y_i. Значения

> 5 обычно считают «исчерпывающими» доказательствами наличия линейной тенденции, от 2,5 до 5 - явным признаком наличия линейной тенденции, от 1 до 2,5 - свидетельством наличия линейной тенденции и < 1 - отсутствием указания на линейную тенденцию или ее отсутствием в измерениях. Это измерение также может быть использовано в качестве обратной связи субъекту о том, насколько действительно сильна тенденция (т.е. насколько велика вероятность).where N is the number of observations, ρ is defined through the expression

, where si is an approximation of the standard deviation for observation y _i , Δx is derived from x _i minus the mean of all x _i , and ΔD is defined as y _i minus the mean of all y _i . Values

> 5 is generally considered "comprehensive" evidence of a linear trend, 2.5 to 5 is a clear indication of a linear trend, 1 to 2.5 is evidence of a linear trend, and < 1 is no indication of a linear trend or its absence in the measurements . This measurement can also be used as feedback to the subject on how strong the trend really is (i.e. how likely it is).

Thus, it is proposed to use a Bayesian inference method to determine blood pressure from measurements of a surrogate measure of blood pressure, such as pulse wave velocity (PWV), pulse wave arrival time (PAT), and/or pulse wave propagation time (PTT).

An apparatus that can be used to determine the trend of a surrogate blood pressure score from measurements of one or more surrogate blood pressure scores is shown in FIG. 2. Device 12 includes a processing unit 14 that controls the operation of device 12 and may be configured to perform the methods described herein. Processing unit 14 may be implemented in many different ways, in software and/or hardware, to perform the various functions described herein. Processing unit 14 may include one or more microprocessors or digital signal processors (DSPs) that can be programmed using software or computer code to perform the desired functions and/or to control components of the processing unit 14 to implement the required functions. Processing unit 14 may be implemented as a combination of dedicated hardware for performing certain functions (e.g., amplifiers, preamplifiers, analog-to-digital converters (ADCs) and/or digital-to-analog converters (DACs)) and a processor (e.g., one or more programmable microprocessors, controllers, digital signal processors and related circuits) for other functions. Examples of components that can be used in various embodiments of the present invention include, but are not limited to, conventional microprocessors, digital signal processors, application specific integrated circuits (ASICs), and user programmable logic circuits (FPGAs).

The processing unit 14 is connected to a storage device 16 configured to store data, information and/or signals for use by the processing unit 14 in controlling the operation of the device 12 and/or in performing or implementing the methods described herein. In some embodiments, memory 16 stores computer-readable code that can be executed by processor unit 14 such that processor unit 14 performs one or more functions, including the methods described herein. The storage device 16 may comprise any type of non-transitory computer-readable storage medium such as a cache or system memory, including volatile and non-volatile computer storage devices such as random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), read-only memory device (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), and electrically erasable programmable ROM (EEPROM) implemented in the form of a single-chip storage device, an optical disc (such as a compact disc (CD), digital versatile disc (DVD ), or Blu-ray Disc), hard disk drive, tape storage device, or solid state device, including a flash drive, solid state drive (SSD), memory card, etc.

Device 12 optionally includes interface circuitry 18 to provide data connectivity and/or data exchange with other devices, including any one or more of servers, databases, user devices, and sensors. The connection may be direct or indirect (eg, over the Internet), and thus interface circuitry 18 may provide a connection between device 12 and a network, such as the Internet, via any desired wired or wireless communication protocol. For example, interface circuitry 18 may operate using WiFi, Bluetooth, Zigbee, or any cellular protocol (including, but not limited to, Global System for Mobile Communications (GSM), Universal Mobile Telecommunications System (UMTS), Long Term Evolution ( LTE), Long Term Evolution (LTE) standard with advanced features, etc.). The interface circuit 18 may be connected to the processing unit 14.

In some embodiments, device 12 may also include a user interface (not shown in FIG. 2) comprising one or more components that allow a user of device 12 (e.g., a subject whose blood pressure trend is being measured or quantified) to enter information, data, and/or or commands to device 12, and/or allowing device 12 to output information or data to a user of device 12. The user interface may comprise any suitable input component(s), including, but not limited to, a keyboard, keypad, one or more than buttons, switches, or a dialer, mouse, trackpad, touch screen, stylus, camera, microphone, etc., and the user interface may contain any suitable output component(s), including, but not limited to, a display screen, one or more lamps or lighting elements, one or more loudspeakers, brotherly element, etc.

To determine the trend of a surrogate blood pressure measure, measurements of one or more surrogate blood pressure measures are required. These measurements (or measurement values) are obtained by using one or more surrogate blood pressure sensor(s). The surrogate blood pressure sensor(s) may be part of the device 12 (for example, as shown by the surrogate blood pressure sensor 20 in FIG. 2), or the surrogate blood pressure sensor(s) may be provided separately. from device 12. In the latter case, device 12 may be connected to the surrogate blood pressure sensor(s) to receive surrogate blood pressure measurements directly from the surrogate blood pressure sensor(s), or device 12 may receive surrogate blood pressure measurement values. via an interface circuit 18. The surrogate blood pressure measurement values may be provided to the processing unit 14 for processing, or the measurement values may be stored in the storage device 16 for further processing by the processing unit 14.

The surrogate blood pressure sensor 20 may be any type of sensor capable of measuring a subject's surrogate blood pressure. The surrogate blood pressure sensor 20 may output a measurement signal indicative of the surrogate blood pressure, or the surrogate blood pressure sensor 20 may output a measurement signal that may be analyzed or processed to determine the surrogate blood pressure. In some embodiments, as noted above, a surrogate blood pressure measure may be pulse wave velocity, pulse wave arrival time and/or pulse wave propagation time of the subject, heart rate, blood volume distribution (e.g., in the face), changes in morphology pulse wave, general body movements (for example, measured using a ballistocardiogram), or any other physiological characteristic that can be used as a surrogate measurement for blood pressure (i.e., any other physiological characteristic whose changes correlate or approximately correlate with changes in blood pressure).

To measure, for example, pulse wave velocity or pulse wave propagation time, surrogate blood pressure sensor 20 may include multiple sensors (e.g., two sensors to be placed at different locations on the subject's body to measure the pulse signal at each location), and these sensing elements may be of the same or different types.

In some embodiments, the surrogate blood pressure sensor 20 is or includes one or more photoplethysmogram (PPG) sensors. Each PPG sensor contains a light source for illuminating a part of the subject's body (e.g. finger, wrist, ear, forehead, etc.) and a light sensor for measuring light reflected and/or absorbed by the skin/tissue adjacent to the light source. The light sensor outputs a measurement signal expressing the measured light intensity, and this measurement signal can be processed to identify changes in blood volume as a pulse signal.Usually, the PPG sensor is placed in contact with the skin.The PPG sensor measures the characteristics of blood volume change in superficial or shallow skin tissue (i.e., e. near the surface of the skin), and characteristics expressing the pulse or heartbeat of the subject, such as the time of arrival of the pulse wave, can be obtained from the PPG measurement signal.

In other embodiments, the surrogate blood pressure sensor 20 may be one or more electrocardiographic (ECG) sensors containing one or more electrodes placed on the subject's body, or which may be adjacent to the body, and which measure the electrical activity of the heart. In some embodiments, capacitive ECG electrodes can be used that can measure an ECG signal when a subject contacts the electrodes. For example, the electrodes may be placed on the floor or on a scale array, and the ECG signal may be measured when the subject steps on the electrodes on the floor or scale.

In some embodiments, the surrogate blood pressure sensor 20 may include two PPG sensors located at different locations on the subject's body to measure the time of arrival of the pulse wave at each location (where one PPG sensor is located downstream of the other PPG sensor to allow time of arrival to be measured). one pulse wave in different areas). In other embodiments, the surrogate blood pressure sensor 20 may include a PPG sensor and an ECG sensor. The ECG sensor is used to measure the electrical activity of the heart and, in particular, the R peak, and the PPG sensor is used to measure the time of arrival of the pulse wave at the PPG section of the sensor.

In other embodiments, the surrogate blood pressure sensor 20 may be an accelerometer or other type of motion sensor placed in contact with a part of the subject's body (eg, chest, neck, wrist, finger, etc.) and measuring accelerations (or, in more generally, movements) of body parts. An accelerometer (or other type of motion sensor) may be sensitive enough to skin movements that are caused by the pulsation of blood through blood vessels under the skin to allow the arrival of a blood pulse wave from the measurement signal. In the case of an accelerometer, the accelerometer can measure accelerations along three orthogonal axes (labeled X, Y, and Z, for example) and output three signals, each expressing accelerations along the corresponding one of the axes, or output one signal composed of accelerations measured along three orthogonal axes. The measurement signal from the accelerometer can be processed using signal analysis techniques known in the art to extract the motions due to heartbeat/blood pulsation, and thus it is possible to derive surrogate blood pressure measurement values from acceleration measurements. Other types of motion sensors from which heart rate measurements can be obtained include a gyroscope that measures changes in rotation and orientation. Those skilled in the art will be aware of other types of motion sensors that can be used to measure surrogate blood pressure readings.

It should be understood that an accelerometer (or other motion sensor) can be used in conjunction with a PPG sensor (instead of a second PPG sensor) or an ECG sensor (instead of a PPG sensor) to measure a surrogate blood pressure.

It should be understood that the surrogate blood pressure sensors 20 described above require direct or physical contact with the subject. However, in some embodiments, the surrogate blood pressure sensor 20 measures the surrogate blood pressure without contact, i. without requiring physical contact with the subject. For example, surrogate blood pressure sensor 20 may be an image sensor (eg, camera) for capturing images or a sequence of video frames. The images or sequence of video frames may be processed using image analysis techniques to extract surrogate blood pressure measurements. For example, the images or sequence of video frames may be processed to identify areas of skin within the images or sequence of video frames, and analyze these areas to extract a PPG signal for the skin. This is known as remote PPG (PPG), and techniques for analyzing images or sequences of video frames to extract the PPG signal are known in the art. In addition to UPPG signals, a camera can also (or alternatively) be used to detect local tissue (skin) movements (eg, in the carotid area) due to pulse signals, as well as general body movements caused by the pumping action of the heart. Alternatively, a radar sensor can be used to measure skin movements (ie due to pulses). Other types of sensors that can be used to measure a surrogate blood pressure indicator include laser Doppler sensors, sensors that use coherent radiation to measure interference (speckle patterns), and moiré sensors.

The surrogate blood pressure sensor 20 may be operated at any suitable sampling rate to provide surrogate blood pressure measurements, such as 10 Hertz (Hz) or 50 Hz. This means that, for example, in the case of an accelerometer, the accelerometer is configured to output an acceleration measurement every 1/10th of a second or 1/50th of a second.

Device 12 may be an electronic device or any type of computing device. For example, device 12 may be a server, computer, laptop, tablet, smartphone, and so on. or be a part of it. In some embodiments, device 12 may be, or be part of, a smart mirror or other household or home appliance that the subject uses from time to time. For example, device 12 may be part of a smart mirror located in the bathroom of the subject's home. Alternatively, device 12 may be part of a television. As another alternative, device 12 may be or be part of a wearable device that can be worn by a subject. For example, the wearable device may be a wrist watch, smart watch, bracelet, necklace, other type of jewelry, garment, etc.

It should be understood that a practical implementation of device 12 may include additional components to those shown in FIGS. 1. For example, device 12 may also include a power source, such as a battery, or components to enable device 12 to be connected to a mains power source.

In FIG. 3 shows an exemplary device 12 in the form of a smart mirror. Thus, in FIG. 3, device 12 includes a frame 22, a mirror surface 24, and a surrogate blood pressure sensor 20. The mirror surface 24 may be a conventional reflective surface. Alternatively, the mirror surface 24 may be in the form of a display screen that displays a sequence of video frames received by a video capture device (eg, a camera). The video capture device will be oriented relative to the smart mirror so as to capture a sequence of video frames of everything in front of the smart mirror. The captured sequence of video frames is displayed on the mirror surface 24 (ie display screen) in real time to provide a "reflect" of the subject if the subject is in front of the mirror surface 24.

In FIG. 3 shows two possible surrogate blood pressure sensors 20 (although in some embodiments, device 12 may include only one of these). The first surrogate blood pressure sensor 20a is a camera or other image capturing device, which may be located in the frame 22 (although other locations are possible), acquiring images or a sequence of video frames that can be processed or analyzed using remote PPG techniques to extract surrogate blood pressure measurement values. In embodiments in which the mirror surface 24 is a display screen, the camera 20a may also be used to display a sequence of video frames on the mirror surface 24. Second surrogate blood pressure sensor 20b is a PPG sensor that is also located in frame 22 (although other locations are similarly possible). The light source and light sensor in the PPG sensor 20b may be located in the frame 22 so that a subject can bring a finger or other body part into contact with them to enable the PPG signal to be received.

In the block diagram of FIG. 4 shows a method for quantifying the trend in a subject's surrogate blood pressure score. The method may be performed by the device 12, and in particular the processing unit 14, for example, in response to the execution of computer-readable code stored on storage device 16 or stored on some other suitable computer-readable medium such as an optical disk, mass storage device, magnetic storage device. disks, etc.

In step 101, a set of surrogate blood pressure measurement values for the subject is obtained. In some embodiments, the blood pressure surrogate is any one of pulse wave velocity, pulse wave arrival time, pulse wave propagation time, or any other physiological characteristic that can be used as a surrogate measurement for blood pressure (i.e., any other physiological characteristic, changes in which correlate or approximately correlate with changes in blood pressure).

In some embodiments, implementations may obtain a set of measurement values for two or more surrogate blood pressure measures. In this case, the surrogate blood pressure measures may be two or more of pulse wave velocity, pulse wave arrival time, pulse wave propagation time, and any other physiological characteristic that can be used as a surrogate measurement for blood pressure. The population of surrogate blood pressure measurement values includes measurement values for the surrogate blood pressure measurement obtained at different points in time over a period of time (eg, the population may include measurement values over several hours, days, weeks, etc.). In embodiments where device 12 includes a surrogate blood pressure sensor 20, step 101 may include obtaining a set of surrogate blood pressure measurement values using the surrogate blood pressure sensor 20 and storing them (e.g., on storage device 16) for processing on later stage. In embodiments where device 12 does not include surrogate blood pressure sensor 20, step 101 may include receiving a set of measurement values from another device (e.g., one that includes a surrogate blood pressure sensor) or receiving a set of measurement values from storage device 16.

In step 103, an error value is obtained for each of the surrogate blood pressure measurement values obtained in step 101. Each error value indicates a measurement error for the corresponding surrogate blood pressure measurement value, i. each error value indicates the level of accuracy of the corresponding measurement value. For example, an error value may indicate a measurement error of a measured value as a percentage (for example, with an accuracy within X%), or indicate an error as a range of values (for example, a measured value is accurate to ± Y), or indicate an error as a measurement of a change in a measured value. In some embodiments, step 103 includes quantifying error values via measurement change (determined by conventional statistical methods). In alternative implementations, step 103 may include quantifying error values based on previous measurements (stored in a lookup table). In other alternative implementations, step 103 may include quantifying the error values as a constant value (eg, X% of the corresponding measured value). In another alternative implementation, step 103 may include quantifying error values based on another measurement signal, such as if sensor 20 detects body movement. For example, if there is a functional relationship with the target dimension, such as y = f(x), the change in y can be quantified as dy = (df/dx)*dx, where df/dx is the derivative and dx is the change in x.

Next, in step 105, the set of measurement values of the surrogate blood pressure measure and the set of measurement error values are analyzed using Bayesian inference to determine a trend in the surrogate measure of blood pressure over time. Briefly, at step 105, the slope (gradient) must be determined taking into account the measurement error. This can be done by quantifying the slope (gradient) a as a trend with information obtained from all data sources (i.e., surrogate blood pressure measurement values obtained in step 101), including the calculated error (obtained or determined in step 101). step 103). Mathematically, this can be expressed as follows:

where p expresses the associated probabilities.

In an alternative implementation, the two hypotheses may be tested using a Bayesian approach by comparing the two hypotheses of constant and linear trend. This is expressed as follows:

where C expresses the constant trend hypothesis, M expresses the linear trend hypothesis, and I expresses any other information relevant to the calculation process such as the subject's circadian rhythm, posture, stress, etc. Depending on the ratio of the calculated probabilities O, a preference decision can be made, for example by means of a threshold value, which of the hypotheses is preferred. Thus, for example, in this embodiment, the output of step 105 may indicate that the constant trend hypothesis is preferred, meaning that the surrogate blood pressure measure follows a constant trend (i.e., is relatively static over time), or the output may indicate that the linear trend hypothesis is preferred, which means that the surrogate blood pressure measure increases or decreases linearly.

Since the surrogate blood pressure is a surrogate measurement for blood pressure, the surrogate blood pressure trend determined in step 105 provides a surrogate measurement of the trend in blood pressure.

It should be understood that the accuracy of the trend determined using Bayesian inference depends on the reliability of the surrogate blood pressure measurement values and error values. Therefore, embodiments provide methods for improving the reliability of surrogate blood pressure measurement values by reducing error values for specified measurement values and/or by applying adjustments to surrogate blood pressure measurement values, as well as improving the quantification of error values. These methods are based on the fact that the measurement values of the surrogate measure of blood pressure will be used to derive a trend, and not to determine the absolute value of blood pressure. Measurement errors can include systematic errors and random errors, both of which contribute to the underlying trend. The provided methods improve trend quantification for a surrogate blood pressure measure and standardize measurement error.

These implementation options are described below and may be implemented individually or in any combination. All embodiments can be implemented by device 12 and can be implemented between steps 103 and 105 of the method of FIG. 4. The first implementation relates to adjusting the measurement values of the surrogate blood pressure indicator and the corresponding error values in accordance with the circadian rhythm of the subject. In this embodiment, the subject's circadian rhythm is derived or quantified from measuring a physiological characteristic of the subject, such as heart rate and/or body temperature, and/or measuring the subject's physical activity. The physiological characteristic may be measured by a device separate from device 12.

A second embodiment relates to standardizing error values through the surrogate blood pressure measurement process, for example, by providing prompts to the subject during the measurement process or by performing a standardized procedure or exercise prior to or during the measurement.

A third embodiment relates to adjusting the surrogate blood pressure measurement values and/or error values using measurements of another physiological characteristic, such as heart rate, to enable comparison of different surrogate blood pressure measurement measurements. For example, a third embodiment can be used to compensate for the effects of heart rate associated with pre-ejection phase (PEP) (where pre-event phase is the time interval from the start of ventricular pacing to the opening of the aortic valve).

These three implementations are described below primarily with reference to the smart mirror implementation (eg, as shown in FIG. 3), but it should be understood that the implementations may be implemented in devices 12 having different design options.

As indicated above, the first embodiment relates to adjusting the measurement values of the surrogate blood pressure indicator and the corresponding error values in accordance with the circadian rhythm of the subject. In a specific embodiment of the first embodiment described below, the surrogate blood pressure is the arrival time of the pulse wave, but it should be understood that the first embodiment can be applied to other surrogate blood pressures that may vary according to the subject's circadian rhythm.

Most preferably, when determining a long-term trend in blood pressure (e.g., a trend over weeks or months), daily surrogate blood pressure measurements are required, and the surrogate blood pressure measurement should be obtained at the same time of day, since the subject's circadian rhythm affects surrogate blood pressure measurements and blood pressure. When taking measurements at different times of the day, obtaining a reliable trend measurement is difficult. However, it happens that the subject cannot perform measurements in accordance with this preferred scenario (for example, he may work with a changing schedule). Therefore, adjusting surrogate blood pressure (PAT) measurement values obtained at different times of the day to match them within the circadian rhythm prior to performing trend analysis helps to reliably quantify the underlying trend in blood pressure using surrogate blood pressure, such as the time of arrival of the pulse wave, as a surrogate measure.

Thus, the surrogate blood pressure (PAT) measurement must be "biased" for a bias in the surrogate blood pressure (PAT) measurement due to the influence of the circadian rhythm on blood pressure. Additionally, the error value for the corrected surrogate blood pressure measure must also be corrected.

First, a quantification of the subject's circadian rhythm and, in particular, a quantification of the point in the circadian rhythm/cycle at which each surrogate blood pressure measurement value is obtained is required. A suitable surrogate measurement for quantifying a subject's circadian rhythm is heart rate, body temperature, and/or a measure of the subject's physical activity. For rhythm quantification, a surrogate measure of circadian rhythm should be measured continuously throughout the day, and, for example, heart rate, body temperature, and/or activity measurements of the subject may be obtained continuously using a smartwatch or other wearable device containing a PPG sensor, an accelerometer, or temperature sensor.

The measurement of the surrogate measure of circadian rhythm (ie, heart rate) is then used to correct the measurement values of the measurement of the surrogate measure of blood pressure. For example, during a pulse wave arrival time acquisition, a surrogate circadian rhythm measurement may indicate that the circadian rhythm is at a point where the pulse wave arrival time and blood pressure are above normal for the rest of the rhythm/cycle. Therefore, the measurement value of the surrogate blood pressure indicator can be corrected by decreasing the measurement value by a suitable amount.

In some embodiments, the adjustment applied to the measurement value may be derived from or based on a linear function of the circadian rhythm (ie, heart rate, body temperature, and/or physical activity) measurement. In other words, the adjustment can be derived from (α * S) + β, where S is a measurement of a surrogate measure of circadian rhythm, and where α and β are values to be determined based on other subjects (eg, a population of subjects).

In alternative implementations, the adjustment to be applied to the measurement value may be derived from or based on a mapping function between a point in the circadian rhythm and the desired adjustment. The mapping function can be used if the linear function related to the correction of the circadian rate measurement is unknown or not applicable (ie if the linear function does not correctly describe the effect). The mapping function can be represented as a lookup table that provides the correct adjustment for a particular circadian rate surrogate measurement value.

As another alternative, the method may include performing a calibration phase in which a surrogate measure of circadian rhythm (ie, heart rate, body temperature, and/or physical activity level) is measured along with a surrogate measure of blood pressure. The calibration phase may include measuring a surrogate blood pressure as well as actual blood pressure (eg, using a cuff-based measuring device) and measuring a surrogate circadian rate for a subject over a specific period of time (eg, a week). These measurements can compare and determine the required correction for the surrogate blood pressure measurement values obtained at the specified point in the circadian rhythm. These predetermined adjustments can then be used for subsequent surrogate blood pressure measurements. For example, the graph in Fig. 5 shows a trend line 30 that expresses blood pressure and shows changes during the day due to the circadian rhythm, together with two pulse wave arrival time measurements taken at different times of the day, and therefore at different periods of the circadian rhythm. In this example, the second pulse wave arrival time measurement does not follow the expected circadian rhythm, and this can be taken into account when quantifying the error of the second pulse wave arrival time measurement determined in step 103.

In yet another alternative, the adjustment to be applied to the surrogate blood pressure measurement value based on a particular circadian rhythm point may be determined from data on the relationship between surrogate blood pressure and circadian rhythm for a population (e.g., for the general population, using weighting if necessary according to the specific data for a given subject or for a population of subjects having a similar medical history or medical diagnoses with a given subject).

In addition to correcting the surrogate blood pressure measurement value for the circadian rhythm measurement time, the error value for the measurement value can also be corrected. An error correction may be determined, for example, by a lookup table depicting a surrogate circadian rhythm measurement and the corresponding error correction required. In an embodiment where the measurement of the circadian rhythm surrogate is heart rate, a higher heart rate value results in more error because the subject may be tense and the resulting surrogate is less reliable. The information in the lookup table can be determined from a functional relationship between error values and a circadian rhythm surrogate measure obtained, for example, from population data (eg, matched to the subject) with additional adjustments for the subject's particular disease or circumstances.

With reference to the second embodiment as above, the second embodiment relates to standardizing the measurement process of a surrogate blood pressure indicator, for example, by providing instructions to the subject during the measurement process, or by performing a standardized procedure or exercise before or during the measurement. This is because any movement or action performed by the subject (e.g., initiating a pulse wave arrival time measurement or exercising) results in a short-term arousal of the subject, which in turn results in a short-term change in blood pressure and therefore a short-term a change in a surrogate measure of blood pressure (for example, the time of arrival of a pulse wave). Therefore, pulse wave arrival time measurements for use in determining a long-term trend should be based on pulse wave arrival time measurements obtained while the subject is at rest (and has been at rest for some time) or, more generally, when the subject is in a particular state of health/mobility (for example, while performing a specific exercise or activity). Otherwise, it is difficult to use individual measurements of a surrogate blood pressure indicator to determine a long-term trend.

The graph in Fig. 6 shows how the arrival time of a subject's pulse wave varies with time after the subject begins to perform a physical action, such as exercise. It is shown that the arrival time of a pulse wave increases with time and generally reaches a flat area (steady state). Thus, when measuring the arrival time of the pulse wave, it is most preferable that the measurement is performed after the arrival time of the pulse wave reaches the flat portion.

Therefore, in the second embodiment, correction for such short-term effects of the surrogate blood pressure/blood pressure is required or is required to be minimized by using suitable measurement procedures that smooth and standardize the process of obtaining the surrogate blood pressure.

There are several ways in which the second embodiment can be implemented. In a first embodiment, it is possible to evaluate surrogate blood pressure measurements taken over a short period of time (eg, a few minutes) to determine if the surrogate blood pressure is in a steady state (ie, constant). If so, a surrogate blood pressure measurement value may be obtained, which may be used for trend analysis in step 105. If not, the surrogate blood pressure measurement may continue until a steady state is reached. Thus, real-time evaluation of acquired surrogate blood pressure measurement values (e.g., pulse wave arrival time) can be performed, and the time span in which measurement values are obtained is extended if the surrogate blood pressure measurement is not in a steady state. Steady state can be identified by determining the rate of change of the surrogate blood pressure over time and determining if the surrogate blood pressure is in steady state if the magnitude of the rate of change of the surrogate blood pressure over time is less than a threshold value. In the example shown in FIG. 6, the arrival time of the pulse wave increases with time, and therefore, the measurement value of the arrival time of the pulse wave for use in trend analysis can only be obtained in the time interval indicated by the shaded portion 36. Before this interval, the arrival time of the pulse wave is not in a steady state. condition, and therefore measurements taken at this time will not provide a reliable measurement of the arrival time of the pulse wave. As an alternative to waiting a certain amount of time for the surrogate blood pressure to reach a steady state (which may not be preferred, as this may imply a longer period of time that the subject needs to be monitored), the set of measurements of the surrogate blood pressure may be is obtained over a short period of time (eg 30 seconds) and a steady state or "flat" can be quantified by extrapolation from these measurements. Those skilled in the art will be aware of various methods that can be used to effect such extrapolation.

According to a second variant, the subject may be instructed to perform a predetermined exercise to standardize the surrogate blood pressure measurement process. Thus, the timing of the surrogate blood pressure measurement can be programmed to occur some time after the start of a predetermined exercise to obtain the measurement value when the subject is in a steady state of arousal. In some embodiments, the device 12 may provide the subject with instructions to perform a predetermined exercise. For example, instructions for performing an exercise can be displayed on the mirror surface 24 (display screen) if the device 12 is in the form of a smart mirror. Alternatively, instructions or instructions for performing the exercise may be displayed on a (other) display screen of device 12, or instructions/instructions may be provided using an audio message (eg, a pre-recorded voice message). Instructions for performing the exercise may include instructing the subject to begin the exercise (eg, start breathing calmly) or instructions that guide the subject throughout the exercise (eg, inhale slowly now, exhale slowly now, etc.). The predetermined exercise may include a breathing exercise, a gripping exercise, a posture exercise (eg relating to hand position), etc. In some embodiments in which device 12 includes a camera or other image/video capture device, images from the camera or image/video capture device may be analyzed by processing unit 14 to identify the subject and their actions, movements, and/or posture to check whether the whether the subject is performing a predetermined exercise. In some embodiments, if the processing unit 14 determines that the subject is not performing the predetermined exercise correctly, the processing unit 14 may determine feedback or additional instructions/indications to the subject in order to assist the subject or facilitate the subject's correct performance of the exercise.

As an example of this second embodiment, it has been found that the arrival time of a subject's pulse wave can be modulated by respiration with a variability of up to 15-20 milliseconds (ms). This is shown in FIG. 7, which depicts a signal indicative of the subject's respiration over time with inspiration occurring approximately every 3 seconds (top graph in FIG. 7). The bottom graph in Fig. 7 shows the arrival time of the subject's pulse wave over the same time period as the breath signal in the upper graph of FIG. 7, and it also shows that the arrival time of the pulse wave changes with time in accordance with a pattern similar to the high and low points in the breath signal. Therefore, according to the second embodiment, a better (more reliable) pulse wave arrival time measurement value can be obtained if the subject's breathing is uniform and the pulse wave arrival time is measured only during a well-defined phase of breathing (for example, during expiration or at the end of exhalation). Alternatively, the pulse wave arrival time may be measured over one or more complete breaths and the resulting pulse wave arrival time measurements may be averaged to provide a pulse wave arrival time measurement value for use in trend analysis at step 105. It should be understood that the respiratory status of a subject can be measured using different types of sensors such as a camera, a radar sensor, or an accelerometer.

As mentioned above, in the third embodiment, the surrogate blood pressure measurement values and/or error values are corrected using measurements of another physiological characteristic, such as heart rate, to ensure comparability between different surrogate blood pressure measurements.

The pulse wave arrival time is the sum of the pre-event phase (PEP) and the pulse wave propagation time (PTT). The pulse wave propagation time is useful as a measure of the trend in blood pressure, but the pre-event phase is easily affected by the heart rate. Therefore, heart rate and changes in heart rate must be taken into account when using pulse wave arrival time measurements to determine the trend.

Therefore, the "pre-event phase effect" can be compensated for by properly measuring pulse wave propagation time measurements and correcting for the effect of heart rate. According to the technical paper "Systolic Time Intervals in Heart Failure in Man", Weissler et al., Circulation, Volume XXXVII, Number 2, February 1968, the pre-ejection phase can be quantified from heart rate measurements as follows:

(4)

(four)

(5)

(5)

where HR is the heart rate and M and F refer to the subject's gender.

Therefore, during the arrival of the pulse wave, a correction for the heart rate can be made in accordance with:

(6)

(6)

where PEP ₀ is the male (eg 0.131) or female (eg 0.133) pre-pregnancy phase value and α is the sex-specific factor disclosed in the work by Weissler cited above. This procedure can help reduce the effect of the pre-event phase due to heart rate and positively influence the relationship between pulse wave propagation time, pulse wave arrival time and blood pressure.

An example depicting the requirement to correct the time of arrival of the pulse wave to the heart rate is shown below with reference to a plurality of 13 measurements. Table 1 shows a collection of pulse wave arrival time measurements taken over a period of 14 days, along with quantitative estimates of the pre-fire phase (in milliseconds, ms), heart rate (in beats per minute, bpm), and actual blood pressure measurements (systolic and diastolic blood pressure).

systolic diastolic Heart rate bpm Pre-warning phase ms Time of arrival of the pulse wave
MS Pulse wave arrival time-pre-event phase ms 120 89 87.5 98 259.87 161.87 114.5 89 105.5 90.8 270.50 179.70 115.5 90.5 127.5 82 241.61 159.61 116.5 86.5 106 90.6 272.30 181.70 124.5 87 101.5 92.4 277.83 185.43 116 91 94.5 95.2 290.05 194.85 129.5 89.5 110.5 88.8 240.68 151.88 120.5 84.5 107 90.2 255.75 165.55 122.5 95 95 95 291.75 196.75 124.5 87.5 98.5 93.6 269.43 175.83 118 90.5 92 96.2 265.05 168.85 118 80.5 90.5 96.8 266.00 169.20 125 98.5 103 91.8 271.23 179.43

Table 1

Thus, each row of Table 1 shows a measurement of blood pressure (systolic and diastolic), heart rate, pre-ejection phase score, and measured pulse wave arrival time, each of which was obtained from the subject at approximately the same time on a given day. As can be seen from the heart rate column, the subject's heart rate during measurements varied from one day to the next, and the total heart rate ranged from 87 bpm to 125 bpm. Because of these large differences in heart rate, it is clear that pulse wave arrival time measurements show a dependence on heart rate. This dependency can be reduced by compensating the pre-event phase according to the heart rate, as described above. In other words, the arrival time of the pulse wave and the heart rate are measured, the pre-ignition phase is quantified using the heart rate measurements according to equation (4) or (5) above, and the pulse wave propagation time is determined by subtracting the quantified pre-ignition phase. from the measurement value of the arrival time of the pulse wave.

In FIG. 8-13 show various plots of measurements in FIG. 1. In particular, in FIG. 8 shows a graph of the arrival time of the pulse wave as a function of the heart rate (HR), and this graph shows the dependence of the time of arrival of the pulse wave (due to the pre-event phase) from the heart rate, with the general trend being that the arrival time of the pulse wave decreases with as the heart rate increases. In FIG. 9 shows a graph of the arrival time of the pulse wave-pre-event phase as a function of heart rate, and this graph shows that the ratio of the arrival time of the pulse wave to the heart rate decreases significantly after the removal of the pre-event phase. Thus, adjusting for the calculated effect of pre-event phase with heart rate reduces the effect of heart rate on the measurement of the surrogate blood pressure measure, pulse wave arrival time.

This is more clearly shown in Fig. 10-13, where the third measurement was taken at a time when the heart rate was very high (~128 bpm), but this had no effect on the actual blood pressure, although it is displayed in the pulse wave arrival time measurement. Adjusting the pulse wave arrival time measurement to match the pre-event quantification as above will correct for this change in heart rate for measurements taken over a 14-day period.

Thus, FIG. 10 is a graph of heart rate measurements over time (one measurement per day), and as shown, the third measurement stands out sharply in that it is much higher than the other 12 measurements. In FIG. 11 shows a graph of measurements of arterial systolic pressure (ASP) in mm Hg. in time (similarly, one measurement per day), and as shown, the systolic blood pressure measurement on the third day is similar to the systolic blood pressure measurements obtained on the previous and following days. It is assumed that the ASD measurements in FIG. 11 have an error of 8 mmHg as indicated by the error lines. Fig. 12 is a plot of the arrival time of the pulse wave over time (one measurement per day) and shows that the measurement on the third day is affected by the higher heart rate shown in FIG. 10. It is assumed that in FIG. 12 there is a permanent error of 15 ms. Finally, FIG. 13 is a plot of the pulse wave arrival time - pre-warning phase over time and depicts the effect of the pulse wave arrival time correction on the heart rate.

Therefore, improvements in the quantification of the trend obtained from surrogate blood pressure measurements are provided, in particular where surrogate blood pressure measurements may have measurement errors.

Variations of the disclosed embodiments may be understood by those skilled in the art and implemented by them in their implementation of the principles and methods described herein, from a study of the drawings, description and appended claims. In claims, the word "comprising" does not exclude the presence of other elements or steps, and singular grammatical indicators do not exclude plurality. One processor or other unit may perform the functions of several elements described in the claims. The fact that some measurements are given in mutually different dependent claims does not mean that a combination of these measurements cannot be advantageously used. The computer program may be stored or distributed on a suitable medium, such as an optical storage medium or a solid state storage medium supplied with or as part of other hardware, but it may also be distributed in other forms, such as over the Internet or other wired or wireless telecommunications systems. No reference designations in the claims should be considered as limiting its scope.

Claims (51)

1. A method for quantifying a trend in a blood pressure surrogate, comprising obtaining (101) a set of surrogate blood pressure measurement values for the subject, characterized in that the method further comprises: obtaining (103) for each measurement value of the surrogate blood pressure indicator, an error value indicative of a measurement error for the measurement value of the surrogate blood pressure indicator; and analyzing (105) said set of surrogate blood pressure measurement values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measure over time, wherein the step of obtaining a plurality of surrogate blood pressure measurement values comprises, for each surrogate blood pressure measurement value: obtaining measurement values of a surrogate blood pressure indicator for a period of time; evaluating the surrogate blood pressure measurement values obtained within the time period to determine if the surrogate blood pressure is in a steady state; and, if the surrogate blood pressure is not in a steady state, extrapolating the surrogate blood pressure measurement values obtained within the time span to quantify the surrogate blood pressure measurement value at steady state. 2. The method of claim 1, wherein the analysis step includes: adjusting each of the surrogate blood pressure measurement values in said population according to the point in the subject's circadian rhythm at which each surrogate blood pressure measurement value was measured; and using Bayesian inference to determine the trend in the surrogate blood pressure measure over time by analyzing the adjusted surrogate blood pressure measurement values and the corresponding error values. 3. The method of claim 2, wherein the adjustment step includes: obtaining measurements of a surrogate indicator of the subject's circadian rhythm; analysis of measurements of a surrogate indicator of the circadian rhythm to identify the circadian rhythm of the subject; based on the identified circadian rhythm, determining corresponding points in the circadian rhythm at which each of the surrogate blood pressure measurement values in said population was measured; determining an appropriate adjustment for each of the surrogate blood pressure measurement values in said population based on the determined corresponding points in the circadian rhythm; and applying certain appropriate adjustments to the surrogate blood pressure measurement values in said population. 4. The method according to any one of paragraphs. 1-3, wherein the step of obtaining a plurality of surrogate blood pressure measurement values includes, for each surrogate blood pressure measurement value: obtaining measurement values of a surrogate blood pressure indicator for a period of time; evaluating the surrogate blood pressure measurement values obtained within the time period to determine if the surrogate blood pressure is in a steady state; and including a measurement value of the surrogate blood pressure indicator obtained when the surrogate blood pressure indicator is in a steady state in the set of measurement values of the surrogate blood pressure indicator. 5. The method according to any one of paragraphs. 1-4, further comprising the step of the subject is provided with instructions to perform a procedure or exercise prior to or during the surrogate blood pressure measurement. 6. The method according to any one of paragraphs. 1-5, in which the method is performed by means of a smart mirror (12) containing one or more sensors (20) to obtain a set of surrogate blood pressure measurement values. 7. A computer-readable medium for quantifying a trend in a surrogate blood pressure indicator, comprising a computer-readable code, the computer-readable code being configured to, when executed by a suitable computer or processor (14), cause the computer or processor to execute the method of any one of claims. 1-6. 8. A device (12) for quantifying a trend in a surrogate blood pressure indicator, comprising a processing unit (14) configured to obtaining a set of surrogate blood pressure measurement values for a subject, characterized in that the device is additionally configured to: obtaining, for each measurement value of the surrogate blood pressure indicator, an error value indicative of a measurement error for the measurement value of the surrogate blood pressure indicator; and analyzing said set of surrogate blood pressure measurement values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measure over time, wherein the processing unit (14) is configured to obtain a plurality of surrogate blood pressure measurement values by performing the following for each surrogate blood pressure measurement value: obtaining measurement values of a surrogate blood pressure indicator for a period of time; evaluating the surrogate blood pressure measurement values obtained within the time period to determine if the surrogate blood pressure is in a steady state; and including a measurement value of the surrogate blood pressure indicator obtained when the surrogate blood pressure indicator is in a steady state, in the specified set of measurement values of the surrogate blood pressure indicator. 9. The apparatus (12) of claim 8, wherein the processing unit (14) is configured to analyze a set of surrogate blood pressure measurement values and corresponding error values using Bayesian inference to determine a trend in the surrogate blood pressure measurement over time by: adjusting each of the surrogate blood pressure measurement values in said population according to the point in the subject's circadian rhythm at which each surrogate blood pressure measurement value was measured; and using Bayesian inference to determine the trend in the surrogate blood pressure measure over time by analyzing the adjusted surrogate blood pressure measurement values and the corresponding error values. 10. The apparatus (12) of claim 9, wherein the processing unit (14) is configured to correct each of the surrogate blood pressure measurement values in said population by: obtaining measurements of a surrogate indicator of the subject's circadian rhythm; analyzing measurements of a surrogate circadian rate indicator to identify the subject's circadian rhythm; based on the identified circadian rhythm, determining corresponding points in the circadian rhythm at which each of the surrogate blood pressure measurement values in said population was measured; determining an appropriate adjustment for each of the surrogate blood pressure measurement values in said population based on the determined corresponding points in the circadian rhythm, and applying certain appropriate adjustments to the surrogate blood pressure measurement values in said population. 11. Device (12) according to any one of paragraphs. 8-10, wherein the processing unit (14) is configured to obtain a plurality of surrogate blood pressure measurement values by performing the following for each surrogate blood pressure measurement value: obtaining measurement values of a surrogate blood pressure indicator for a period of time; evaluating the surrogate blood pressure measurement values obtained within the time period to determine if the surrogate blood pressure is in a steady state; and, if the surrogate blood pressure is not in a steady state, extrapolating the measurement values of the surrogate blood pressure obtained within the time span to calculate the measurement value of the surrogate blood pressure at steady state. 12. The device (12) according to any one of paragraphs. 8-11, in which the processing unit (14) is further configured to providing, via a user interface, instructions to a subject to perform a procedure or exercise before or during the surrogate blood pressure measurement. 13. Device (12) according to any one of paragraphs. 8-12, wherein the device is a smart mirror containing one or more sensors (20) for obtaining a set of surrogate blood pressure measurement values.

Images