Apparatus for continuously measuring biological information

Abstract

This apparatus for continuously measuring biological information continuously obtains an optimized systolic blood pressure estimation value, and can monitor a changing state of blood pressure in hemodialysis treatment or the like. The apparatus comprises: an optical blood flow measuring means for detecting a blood flow pulse wave from a living body; an electrocardiographic signal detecting means for detecting an electrocardiographic signal from the living body; an actual biosignal measuring means for detecting an actually measured biosignal value by actually measuring a target biosignal; a biosignal calculating means for calculating the target biosignal on the basis of a characteristic signal obtained from the optical blood flow measuring means and the electrocardiographic signal detecting means; and a biosignal processing means for outputting the actually measured signal obtained by the actual biosignal measuring means, the calculated signal obtained by the biosignal calculating means, and a biosignal optimized through mathematical optimization using the actually measured biosignal.

Description

Continuous biological information measurement device

[0001] The present invention relates to a biological information continuous measurement device that can continuously obtain an optimized estimate of a subject's systolic blood pressure using an electrocardiogram and a blood flow pulse wave.

[0002] Current blood pressure monitors use the oscillometric method, in which a cuff is wrapped around the upper arm, etc., and then pressurized, and then the pressure is reduced by checking the fluctuations in cuff pressure (pressure pulse wave), which reflects the vibrations of the blood vessel walls that are synchronized with the heartbeat, to determine blood pressure values.However, this type of blood pressure monitor applies pressure with the cuff that is strong enough to block the blood vessels, causing a feeling of pressure on the subject and making it difficult to measure blood pressure continuously.

[0003] Therefore, a method has been proposed in which the pulse wave transit time (PTT) related to blood pressure values ​​is calculated based on an electrocardiogram obtained from an electrocardiograph, and the pulse wave obtained from a pulse wave meter worn on a fingertip and the distance from the heart to the part where the pulse wave meter sensor is worn are measured, and the systolic blood pressure and diastolic blood pressure are estimated from a formula for calculating blood pressure-related values. This calculation method requires the distance value from the heart of the person being measured to the pulse wave measuring unit, and may require pre-setting for measurement. For example, Ryuta Mizuguchi, "Wireless monitoring of continuous systolic blood pressure in daily life using a neckband device," Life Support, Vol. 34, No. 1, 2022, states that if certain vascular conditions are met, the PTT can be estimated. －２ However, it is described that the pulse wave is proportional to the systolic blood pressure and that the systolic blood pressure can be obtained as an estimated value, and there is no need to measure the distance from the subject's heart to the pulse wave measuring unit.

[0004] In Japanese Patent Application Laid-Open No. 2020-142070, the systolic blood pressure change rate SBPφi, which is blood pressure information, is calculated from an electrocardiogram signal and a blood flow pulse wave signal as follows: SBPφi = PWi α (α:0.25~0.5) or SBPφi=(Tu / Cbb) α It is described that the pulse wave index (PWi) (Tu: upstroke time, Cbb: pulse wave propagation time) is expressed as follows.

[0005] Japanese Patent Publication No. 2020-142070 describes that the blood pressure state (rate of change) of the estimated systolic blood pressure (SBP) is measured using blood flow components and an electrocardiogram, and that information on the estimated blood pressure value can be obtained.

[0006] Japanese Patent Application Laid-Open No. 2020-142070

[0007] Ryuta Mizuguchi, Continuous Wireless Monitoring of Systolic Blood Pressure in Daily Life Using a Neckband Device, Life Support, Vol. 34, No. 1, 2022

[0008] In this way, blood pressure information can be obtained simply by attaching and wearing a probe sensor or electrodes, without applying pressure to the living body, which broadens the scope of long-term monitoring of blood pressure-related information. However, blood pressure information obtained numerically often contains errors, and although the results are obtained as estimated values, it is preferable to be able to obtain more accurate values ​​of continuous changes in blood pressure.

[0009] When blood pressure information is calculated, it is only a rough estimate, and generally, when blood flow is closed within the body, pinpoint blood pressure measurement is required. On the other hand, in fields such as hemodialysis and apheresis, where blood is extracted externally and operations such as the removal of waste products and fluid removal are performed, the blood flow is equivalent to an open circuit, resulting in a rhythm different from the biological rhythm. Therefore, blood pressure fluctuations, such as sudden hypotension, can occur that affect life prognosis. Therefore, a system that can monitor changes in blood pressure information and blood flow information over time and detect sudden drops in blood pressure is desirable.

[0010] Furthermore, there is a need for a system that allows patients to receive dialysis treatment with less physical and mental stress by monitoring their physical and mental conditions, blood pressure, etc., and by carefully adjusting factors that place a burden on the body during dialysis, such as the amount of water removed and the temperature of the dialysis fluid. Meanwhile, the above-mentioned literature reports that the estimation of systolic blood pressure based on the above-mentioned formula can provide accurate estimations when certain conditions are met, such as when the vascular condition is constant, but errors occur if these conditions are not met numerically. In particular, the margin of error tends to be large when the blood volume is reduced, such as during hemodialysis.

[0011] The present invention has been achieved in view of the above-mentioned problems, and provides a continuous biological information measurement device comprising: an optical blood flow measurement means for detecting a blood flow pulse wave from a living body; an electrocardiogram signal detection means for detecting an electrocardiogram signal from the living body; a biological signal measurement means for actually measuring a target biological signal and detecting a measured value biological signal; a biological signal calculation means for calculating the target biological signal based on a characteristic signal obtained from the optical blood flow measurement means and the electrocardiogram signal detection means; and a biological signal processing means for mathematically optimizing the measured signal obtained by the biological signal measurement means, the calculated signal obtained by the biological signal calculation means, and the measured biological signal, and outputting an optimized biological signal.

[0012] According to the continuous biological information measuring device of the present invention, for the calculation formula when estimating systolic blood pressure, systolic blood pressure change, etc., variable multipliers are added to each parameter such as RR, C, Tu shown in the following formula (1), and various values ​​are further substituted for these variables, and the value obtained from the calculation formula is compared with the corresponding actually measured biological information, and for example, the value that minimizes the difference is found, and the multiplier value when this is minimized is used to set an algorithm (arithmetic formula), thereby reducing errors and obtaining optimized biological information.

[0013] The continuous biological information measuring device of the present invention can measure continuous estimated systolic blood pressure values ​​non-invasively by measuring an electrocardiogram and blood flow pulse waves, and can therefore collect information on blood pressure such as systolic blood pressure values, changes in systolic blood pressure, etc. Furthermore, the continuous biological information measuring device of the present invention can continuously display, as biological information, arteriosclerosis information and estimated blood pressure information obtained from blood flow information.

[0014] Furthermore, the continuous biological information measurement device of the present invention detects fluctuations in blood pressure and blood flow-related information during hemodialysis treatment and other blood purification therapies, and responds predictively to these fluctuations, enabling stabilization processing and achieving gentle dialysis treatment.

[0015] Furthermore, the continuous biological information measuring device of the present invention displays information related to blood pressure fluctuations in real time in treatments that affect blood pressure fluctuations, such as blood purification therapy, which controls the water content of the blood or removes uremic toxins, enabling monitoring that allows predictive knowledge of blood pressure fluctuations in the patient.

[0016] FIG. 1 is a diagram showing one embodiment of the biological information continuous measurement device of the present invention. FIG. 2 is a block diagram for explaining one embodiment of the biological information continuous measurement device of the present invention. FIG. 3 is a diagram for explaining one embodiment of the present invention. FIG. 4 is a diagram showing the results of an experiment conducted to explain one embodiment of the present invention. FIG. 5 is a diagram showing the results of an experiment conducted to explain one embodiment of the present invention. FIGS. 6(a) and (b) are diagrams for explaining a method for measuring blood pressure changes due to exercise load. FIG. 7 is a graph showing the measurement results of blood pressure changes. FIG. 8 is a graph showing continuous blood pressure measurement values ​​according to the present invention.

[0017] Examples of optical blood flow measurement means in the present invention include reflective or transmissive sensors using a laser light emitting element or LED as a light source, such as a combination of an LED and a light receiving sensor, or a combination of a laser light emitting element and a light receiving sensor. Examples of light receiving sensors include a light receiving sensor that receives reflected or transmitted light and performs photoelectric conversion. LEDs are used to measure blood flow in peripheral areas such as earlobes and fingertips, while laser light emitting elements enable measurement in thicker body parts such as hands and feet and provide more accurate values.

[0018] The electrocardiogram signal detecting means in the present invention may be a general electrocardiograph, for example, an electrocardiograph that has two electrodes attached across the heart and one electrode as a ground, and obtains so-called QRS waves (waveforms representing ventricular excitation) from the potentials detected by these electrodes. The electrodes may be attached by sticking them on the body or may be configured to be used by touching the electrodes to the skin with the hand.

[0019] In the present disclosure, examples of feature values ​​determined from an electrocardiogram signal and a blood flow pulse wave signal include time information from the R-wave peak phase to the peak value of the second derivative of the pulse wave (pulse wave propagation time), pulse wave rise time (also called upstroke time) from the peak value of the second derivative of the pulse wave to the pulse wave peak, which is indicated as the force with which blood flow sent out from the heart pushes against blood vessels, and the time between one R wave and the next R wave (RR interval).

[0020] In the present disclosure, examples of measured biosignals include systolic blood pressure, diastolic blood pressure, and the like measured by a biosignal measurement means, for example, an existing sphygmomanometer, such as a pressure-type blood pressure measuring device, a pulse wave transit time measuring device, or a tonometry-type blood pressure measuring device. Furthermore, in the present disclosure, examples of feature signals include one or more signals selected from a pulse wave period signal, a pulse wave amplitude value, a pulse wave transit time, a pulse wave rise time (Tu), an R-R wave time signal, an R wave amplitude, and a QRS width. In one embodiment, the feature signal is composed of the RR interval, the pulse wave transit time, and the pulse wave rise time (Tu) in an electrocardiogram waveform. The variable multiplier in the present invention is a numerical value such as an integer or a decimal, and examples thereof include a plurality of numerical values ​​that increase or decrease at regular intervals.

[0021] The biological information algorithm in the present invention is, for example, the equation for predictively calculating the rate of change of the systolic blood pressure (SBP) shown in Japanese Patent Application Laid-Open No. 2020-142070: SBPφi=(Tu / Cbb) α This formula was then converted to obtain the estimated systolic blood pressure for each patient:

[0022] Here, X is a unique value for each patient, C is the pulse wave transit time (msec), Tu is the pulse wave rise time (msec), and RR is the RR interval (msec). The unique value X for each patient is calculated by multiplying the systolic blood pressure value measured with an existing blood pressure monitor at the time of initial setting before calculating the estimated systolic blood pressure value for each patient by the Tu value, C value, and RR value measured at the same timing by [(Tu / C)] ０．５ / (0.82RR-0.15) ０．１In some cases, the characteristic value X for each patient may be obtained by dividing the characteristic value X by the average value X of the characteristic values ​​obtained by performing multiple measurements at the time of initial setting. ａｖｅ The systolic blood pressure value based on the systolic blood pressure value measured at the time of initial setting can be always obtained from the characteristic value X and the C value, Tu value, and RR value that change over time, according to the above formula (1).

[0023] For example, when mathematically optimizing equation (1), an optimization solver among multiplier optimization methods is preferable. For example, as shown in equation (2), a multiplier is added to equation (1) and the multiplier value is determined by finding the minimum difference value between the equation (1) and the actual measured value.

[0024] where K is an individual eigenvalue, and Tu ａｖｅ is the mean pulse wave rise time (msec), and C ａｖｅ is the mean pulse wave transit time (msec), and RR ａｖｅ is the average RR interval (msec). ａｖｅ , C ａｖｅ , and RR ａｖｅ indicates the average values ​​of Tu, C, and RR extracted from a predetermined number of blood flow pulse waves. Note that the eigenvalue K is calculated in the same way as the eigenvalue X.

[0025] These multipliers are used as variables and are represented by x(i), y(i), z(i), a(i), and examples thereof include a sequence type in which different numerical values ​​are listed in order, a random number type, etc. The variable multiplier in the present invention may be any multiplier that is used in mathematical optimization, and an example thereof is a multiplier in which an appropriate numerical value is variable and assigned to each element of equation (1) for optimization.

[0026] The above equations (2) and (3) provide estimated blood pressure values. Optimization in the present invention refers to the setting of a calculation algorithm with multipliers, coefficients, etc. that approximate calculated values ​​to actual measured values. An example of optimization using multipliers in mathematical optimization is shown, for example, in Excel (registered trademark) Solver. By finding a multiplier that minimizes the difference between the value calculated by the calculation formula to obtain the desired information and the actual measured value, the algorithm can be converted to one that can output values ​​closer to the actual measured values.

[0027] Another example of an algorithm for optimizing the above formula (1) is shown in formula (3).

[0028] Equation (3) is an equation in which the symbol i in equation (2) is replaced with the symbol (n, m) for ease of explanation. The symbol (n, m) indicates, for example, a state in which m variables are stored in each of n groups.

[0029] The present invention only requires that the determined parameters are substituted into equation (1), equation (2) or (3) is formed based on equation (1) to apply a mathematical optimization algorithm or an optimization solver algorithm, and that the multiplier is determined by multiplier optimization using, for example, Excel (registered trademark) solver (software included with the spreadsheet software Excel (registered trademark) (trademark)). For example, it is sufficient that optimization can be performed based on biological information for which actual measurements can be obtained as reference values, such as an arteriosclerosis calculation algorithm or a blood flow calculation algorithm.

[0030] The algorithms described in the present invention include, for example, equations (1) to (3), but also include intermediate equations and other equations that combine multiple equations to calculate biological information. Note that equations (1) to (3) indicate equations for obtaining an estimated systolic blood pressure using blood flow information and electrocardiogram information, but the optimization method described in the present invention can also be applied to other blood pressure calculation equations, blood flow calculation equations, and other equations that use algorithms to obtain biological information.

[0031] Therefore, in one embodiment, the algorithm is a calculation formula for estimating and calculating either the systolic blood pressure value or the amount of change therein.

[0032] In one embodiment, the biological signal processing means executes a program having the steps of: extracting feature values ​​from an electrocardiogram signal and a blood flow pulse wave signal to form an algorithm; adding variable multipliers to the algorithm, and calculating the algorithm for each variable multiplier to obtain a difference between the value obtained and an actual measurement value; and determining a multiplier that minimizes the sum of squares of the difference values.

[0033] The biosignal processing means executes a program including, for example, the steps of: extracting feature values ​​from an electrocardiogram signal and a blood flow pulse wave signal, substituting the extracted feature values ​​into an algorithm, adding a predetermined variable multiplier to the feature value in the algorithm, substituting a predetermined numerical value for the multiplier of the algorithm, obtaining a difference between an output value of the algorithm into which the sequence value has been substituted and an actual measurement value, obtaining the sum of squares of the difference values, determining a multiplier when the difference value is minimized, and outputting an optimized biosignal value based on the algorithm formed with the determined multiplier. The program is stored in a storage medium connected to a web server, a cloud server, a personal computer, a single-board computer, etc., and is executed on the computer after being called from the storage medium.

[0034] Therefore, in one embodiment, the present invention comprises: an optical blood flow measuring means for detecting a blood flow pulse wave from a living body; an electrocardiogram signal detecting means for detecting an electrocardiogram signal from the living body; an actual blood pressure measuring means for actually measuring a target biological signal (for example, a systolic (maximum) blood pressure value) to obtain an actual measured blood pressure value; a characteristic value calculating means for calculating a characteristic value (such as the appearance time of an R wave in an electrocardiogram, the rise time of a pulse wave, and the peak time of a pulse wave) from the optical blood flow measuring means and the electrocardiogram signal detecting means; and the following steps (i), (ii) and (iii): (i) a step of forming a sum of squares of values ​​calculated by varying a multiplier of the algorithm (the following formula (2) or (3)) formed by adding a variable multiplier to a mathematical formula formed from the characteristic value; (ii) a step of determining, as an optimal value, a multiplier that minimizes the difference between the sum of squares and the actual measured blood pressure value; and (iii) a step of inputting the characteristic value into the algorithm in which the determined optimal value is substituted as the multiplier, and calculating a blood pressure value. The present invention relates to a continuous biological information measuring device, and a signal processing means for performing the above.

[0035] In the present invention, when values ​​related to biological information are calculated from an algorithm, a more accurate blood pressure estimation value can be obtained by forming an optimization algorithm that optimizes the algorithm. For example, mathematical optimization can be applied to obtain blood pressure-related values ​​(such as systolic blood pressure) formed from pulse wave rise time (Tu), pulse wave propagation time, etc., which are obtained in relation to blood pressure estimation data. Therefore, examples of embodiments include a form in which the present invention is applied to a wearable device such as a wristwatch, and a display device that adjusts the time axis of items that can detect sudden changes in blood pressure by monitoring blood pressure and blood flow during blood purification such as hemodialysis without burdening the patient.

[0036] The actual measurement data that serves as the basis for optimization may be obtained, for example, only before measurement or at regular intervals. For example, when estimating and measuring blood pressure, an existing blood pressure monitor such as a cuff-type blood pressure monitor may be used. The actual measurement data may be used as a reference value, once before and after the start of dialysis, or at regular intervals when the patient is restrained for nearly four hours, such as during dialysis treatment. In some cases, the actual measurement data may be obtained only once at the beginning of dialysis treatment.

[0037] Next, an embodiment of the present invention will be described in detail with reference to Figure 1. Reference numeral 101 denotes an electrocardiogram input unit. The electrocardiogram input unit is, for example, composed of a plurality of electrodes and electrical lead wires connected to each electrode, and is used by attaching the electrodes in a state in which they sandwich the heart. Examples of electrodes include a combination of conductive adhesive gel and electrical lead wires, as well as electrodes made of a conductive material that come into contact with fingers or hands.

[0038] Reference numeral 102 denotes a photoelectric detection unit, which is composed of an LED and a light-receiving element, or a laser light-emitting element and a light-receiving element, or both. Reference numeral 103 denotes an electrocardiogram detection unit, which includes an analog noise filter and an analog amplifier, and generates an analog signal of at least a QRS wave. The electrocardiogram detection unit 103 may generate one lead from multiple electrode inputs, or may generate multiple different types of leads and generate an analog waveform for determining the average value of the fluctuating R-wave phase.

[0039] Reference numeral 104 denotes a pulse wave input unit, which receives an electrical signal obtained by transmitting or reflecting laser light or LED light irradiated on a living body by the photoelectric detection unit 102, amplifies the input electrical signal, and outputs the signal. Reference numeral 105 denotes an electrocardiogram AD converter, which converts an analog electrocardiogram signal into a digital electrocardiogram signal and outputs it. Reference numeral 106 denotes a blood flow pulse wave AD converter, which converts the input analog blood flow pulse wave into a digital signal. Reference numeral 107 denotes an electrocardiogram filter, which is formed by a digital filter and is sufficient as long as it passes at least the high frequency band required for detecting QRS waves. The digital filter may be an existing algorithm, or an analog filter may be used in some cases, such as when the computing power is insufficient.

[0040] Reference numeral 108 denotes a blood flow pulse wave filter, which is formed by a digital filter and can accurately obtain the amplitude and time of characteristic portions of the blood flow pulse wave after passing through the filter. The electrocardiogram filter 107 and blood flow pulse wave filter 108 may be incorporated into an arithmetic unit 109. Reference numeral 109 denotes a computer processor including a digital signal processor (DSP), a central processing unit (CPU), and memories such as random access memory (RAM) and read-only memory (ROM). The arithmetic unit executes programs stored in the memory, calculates time intervals such as the RR interval, pulse wave rise time (Tu), and pulse wave transit time, calculates amplitude values, stores and executes mathematical optimization algorithms, and outputs display signals. Reference numeral 110 denotes a storage unit, which temporarily or continuously stores the obtained optimized blood pressure-related information, pulse wave transit time values, Tu values, etc. The storage unit 110 may be, for example, a solid-state drive (SSD), hard disk (HD), RAM, or non-volatile memory with the capacity to record blood pressure-related value information for at least the duration of the dialysis treatment (typically approximately four hours), as shown in the case of dialysis treatment, or may be a device or system capable of storing the information externally, such as in the cloud or on a server.

[0041] Reference numeral 111 denotes a blood pressure signal input unit. An example of a blood pressure signal input unit is a cuff that can be wrapped around the upper arm, wrist, or the like and inflated. Reference numeral 112 denotes a blood pressure measuring device. Examples of this blood pressure measuring device include existing pressure-type blood pressure measuring devices, as long as they are capable of measuring at least systolic blood pressure and outputting it as a digital value. If the blood pressure measuring device displays blood pressure values ​​numerically, the blood pressure values ​​can be manually input into the calculation unit 109. Alternatively, if the blood pressure measuring device has the function of outputting blood pressure value signals as digital data to an external device, the digital data can be automatically input into the calculation unit 109, which has the function of receiving this digital data. Reference numeral 113 denotes a display unit, which includes a liquid crystal display (LCD) and an LED array and can continuously display blood flow analog waveforms, optimized blood pressure (systolic blood pressure) information, pulse wave transit time (PWT), and Tu in real time. The display unit 113 may be integrated with the calculation unit 109, or may be connected wirelessly or wired to a tablet terminal, smartphone terminal, or other personal computer (PC) terminal and display information on the LCD terminal. Furthermore, the display unit 113 may be configured to issue an alert (light, sound, etc.) when an unusual signal form occurs.

[0042] Next, the operation of one embodiment of the biological information continuous measurement device shown in Figure 1 will be described. First, before starting measurement, the electrodes of the electrocardiogram input unit 101 are attached to the electrocardiogram measurement site of the living body, and the blood flow pulse wave sensor of the photoelectric detection unit 102 is attached to the fingertip, earlobe, or other site where blood flow pulse waves can be detected. Furthermore, the blood pressure measurement cuff forming the blood pressure signal input unit 111 is wrapped around the upper arm. During initial setup before starting measurement, the blood pressure measuring device 112 is activated to obtain values ​​for maximum blood pressure (systolic blood pressure) and minimum blood pressure (diastolic blood pressure), which are temporarily stored in the memory unit 110, and measurement of the electrocardiogram and blood flow pulse wave begins.

[0043] An electrocardiogram signal input from an electrocardiogram input unit 101 is detected by an electrocardiogram detection unit 103 for detecting QRS waves, and is converted into a digital signal by an electrocardiogram AD converter 105 and output to an electrocardiogram filter 107. A blood flow pulse wave AD converter 106 converts the analog blood flow pulse wave signal into a digital blood flow pulse wave signal, which is output to a blood flow pulse wave filter 108. The blood flow pulse wave filter 108 performs filtering to extract waveforms required for detecting Tu (pulse wave rise time) and the end point of the pulse wave propagation time (PTT or C) on the blood flow pulse wave. When the outputs of the electrocardiogram filter 107 and the blood flow pulse wave filter 108 are input to the arithmetic unit 109, an algorithm for calculating the pulse wave transit time, RR (the time width between peaks of the R wave in the electrocardiogram), etc. is calculated, an algorithm for obtaining predetermined target information is called, numerical values ​​such as the pulse wave transit time are input, and target values ​​(e.g., estimated systolic blood pressure and estimated diastolic blood pressure) are calculated.

[0044] The calculated target value is temporarily stored in the storage unit 110, and the algorithm is optimized to bring this value closer to the actually measured value. The multipliers constituting the algorithm are adjusted, an optimization algorithm is formed, and the optimization algorithm is stored in the storage unit 110. When the actual measured value is measured for the first time, the measurement is carried out as is, and a value that is closer to the target biological information-related value based on the optimization algorithm is obtained. The obtained value is displayed on the liquid crystal display of the display unit 113, and in some cases, is displayed in parallel with other values, for example, with the time axis compressed, so that doctors, technicians, and nurses can refer to it and use it to predict changes in blood pressure and detect other changes in physical condition.

[0045] Furthermore, for example, after a predetermined time has elapsed during hemodialysis, the arithmetic unit 109 activates the blood pressure measuring device 112 again by operating a timer function or the like, to obtain an actual measurement value and obtain a more accurate and optimized target blood pressure-related value. The number of actual measurements to obtain the optimized target biological value may be, for example, only during the dialysis treatment, in the initial stage, or at intervals of 1 hour, 30 minutes, 15 minutes, or the like, which may be adjusted as appropriate.

[0046] Next, a flowchart illustrating one embodiment of the present invention will be described with reference to FIGS. 2 and 3. Reference numeral 201 denotes a start step, which indicates a state in which preparations are complete to begin hemodialysis treatment and measurement. In this case, the electrocardiogram input unit 101 shown in FIG. 1, which is equipped with electrocardiogram measurement electrodes, is attached to the patient's body, and the photoelectric detection unit 102, such as a blood flow pulse wave sensor, is attached, for example, to a fingertip. Furthermore, the blood pressure measurement cuff constituting the blood pressure signal input unit 111 shown in FIG. 1 is attached to the upper arm. Reference numeral 202 denotes a measurement start step for determining whether hemodialysis treatment has started. If measurement has started (yes), the process proceeds to a step for measuring actual blood pressure values.

[0047] If measurement is already ongoing (No), the process proceeds to step 207, in which it is determined whether a comparative measured systolic blood pressure value RSBP has already been set. Reference numeral 203 is a step of measuring measured blood pressure values, in which a cuff previously wrapped around the upper arm is inflated and blood pressure (e.g., systolic blood pressure) is measured. In this embodiment, since the purpose is to obtain a blood pressure estimate, the reference blood pressure is measured using an existing method as a measured value. Reference numeral 204 is a step of temporarily or continuously storing measured blood pressure information, for example, storing individual measured blood pressure values ​​for purposes such as performing multiple blood pressure measurements to obtain a mean blood pressure. Reference numeral 205 is a step of setting a comparative measured systolic blood pressure value, in which a systolic blood pressure value RSBPn is set among the measured blood pressure values. The systolic blood pressure value RSBPn may represent, for example, multiple systolic blood pressure values ​​obtained by multiple (n) measurements or a systolic blood pressure value obtained by a single measurement.

[0048] Reference numeral 206 denotes a step for setting a comparative measured systolic blood pressure value RSBP. This step involves averaging or selecting one of the systolic blood pressure values ​​RSBPn obtained through multiple measurements, or determining a systolic blood pressure value RSBPn (n = 1) obtained through a single measurement as the comparative measured systolic blood pressure value RSBP. In this embodiment, it is preferable to determine a systolic blood pressure value RSBPn (n = 1) obtained through a single measurement, which is a blood pressure measurement that does not impose a burden on the patient, as the comparative measured systolic blood pressure value RSBP. Reference numeral 207 denotes a step for determining whether a comparative measured systolic blood pressure value RSBP for optimization has already been set. If it has already been set (yes), the process proceeds to step 208. If it has not been set or if it will be set by periodic measurement (no), the process proceeds to step 203, where a measured blood pressure value is measured.

[0049] Reference numeral 208 denotes a step of replacing the measured blood pressure value with a previously set comparative measured systolic blood pressure value RSBP. Reference numeral 209 denotes a step of inputting a blood flow pulse wave and an electrocardiogram signal, and converts the analog electrical signals of the blood flow pulse wave and the electrocardiogram signal obtained from the electrocardiogram input unit 101 and the photoelectric detection unit 102 shown in FIG. 1 into digital signals. Reference numeral 210 denotes a filtering step, which is implemented by a program routine for a known digital filter and which removes other noise information to obtain the desired waveform information.

[0050] Reference numeral 211 denotes a step of converting the data into blood flow pulse wave data and electrocardiogram data, for example, into a data string with a common time axis. Reference numeral 212 denotes a step of obtaining the RR interval, pulse wave rise time (Tu), and pulse wave propagation time (C) from the blood flow pulse wave data and electrocardiogram data and temporarily storing them to be substituted into equation (3).

[0051] Reference numeral 213 is a step of determining whether an average value can be calculated, and is a step of repeatedly measuring to obtain the average value and temporarily storing the measured value. Reference numeral 214 is a step of calculating an average value RR by averaging a predetermined number of RR intervals, etc. ａｖｅ , Tu ａｖｅ , Cａｖｅ This is the step to obtain

[0052] The terminal indicated by reference number 00 in Fig. 2 is connected to the terminal indicated by reference number 00 in Fig. 3. In Fig. 3, reference number 215 is a step of setting variables (n, m) to be substituted into the multipliers a, x, y, and z. The variables (n, m) represent a configuration in which there are n groups (n1, n2, n3...nn), each group including m mutually different numerical values ​​(m1, m2, m3...mm).

[0053] Reference numeral 216 indicates a step of specifying the first group of the first group n. Reference numeral 217 indicates a state of specifying the first m value (m1) of the specified group. The variable m is just an example, and may be appropriately selected depending on the variable interval and the properties of the variable. Reference numeral 218 indicates an algorithm that combines a variable multiplier with equation (3), and is a step of calculating the CSBPn,m value using the m-th variable of the nth group as a multiplier.

[0054] Reference numeral 219 is a step for calculating and temporarily storing the difference between CSBPn,m and the comparative measured systolic blood pressure value RSBP obtained by actual measurement. The measured value is the most recent measured value. The most recent may refer to a value measured at a predetermined time interval or, if only the value measured at the start of measurement is used, the measured value at the start of measurement. Reference numeral 220 is a step for determining whether the mth value in the nth group is the last value (mm). Reference numeral 221 is a step for specifying the next mth value if the mth value is not the last value. Reference numeral 222 is a step for calculating the sum of squares of m calculated difference values ​​when the mth value in the nth group is the last value (mm). In this step, m difference value data for the nth group are temporarily stored.

[0055] Reference numeral 223 is a step of determining whether the nth group is the last group, reference numeral 224 is a step of specifying the (n+1)th group that follows the nth group, and reference numeral 225 is a step of determining the minimum value of the sum of squares.

[0056] In this embodiment, the minimum value is determined from the n sums of squares calculated using the m numerical values ​​of each group as multipliers. However, if a target minimum value is set in advance, a numerical value close to that target value may be minimized, or the values ​​of all n groups may not be calculated, and the numerical value obtained when the CSBPn,m value hardly changes along the way may be used.

[0057] Reference numeral 226 is a step for determining the values ​​of x, y, z, and a, and the corresponding x, y, z, and a values ​​are found from the minimum CSBPn,m values. Reference numeral 225 is an end step, which indicates the state when the routine shown in Figures 2 and 3 is completed, and constitutes and outputs an algorithm that is ready to calculate an optimized systolic blood pressure (SBP) value. This optimized systolic blood pressure (SBP) calculation algorithm operates in the routine work corresponding to the main routine until the next actual blood pressure is measured.

[0058] 2 and 3. When starting hemodialysis treatment, the device is operated while preparations are being made by attaching electrocardiogram electrodes, a sensor for measuring blood flow pulse waves, and a blood pressure measurement cuff to the upper arm (step 201). Since this embodiment is a routine for determining an algorithm for calculating an estimated systolic blood pressure, the systolic blood pressure estimation algorithm may already have been determined or actual blood pressure may have already been measured at step 201. Therefore, if actual blood pressure measurements have already been made and a comparative actual systolic blood pressure value RSBP has been determined (step 207), the previously set comparative actual systolic blood pressure value RSBP is used (step 208).

[0059] If the comparative measured systolic blood pressure value RSBP has not been set (step 207 (yes)), the process proceeds to step 203, where the measured blood pressure value is measured. The blood pressure measuring device can be a conventional sphygmomanometer. The sphygmomanometer may be configured to automatically operate upon receiving a command signal from the arithmetic unit 109 and simultaneously start measurement. The blood pressure values ​​measured by the sphygmomanometer are temporarily stored in, for example, the storage unit 110 shown in FIG. 1 via the arithmetic unit 109 (step 204), and the systolic blood pressure value RSBPn is set from the temporarily stored blood pressure values ​​(step 205). When an average value is to be calculated from multiple systolic blood pressure values ​​RSBPn, a predetermined number of systolic blood pressure values ​​RSBPn are read in step 206, and the average value is used as the comparative measured systolic blood pressure value RSBP. When an average value is not required and the comparative measured systolic blood pressure value RSBP is set from a single measurement, one systolic blood pressure value RSBPn is used as the comparative measured systolic blood pressure value RSBP.

[0060] 1, blood flow pulse wave and ECG signals are input via ECG input unit 101 and photoelectric detector 102, and converted to digital signals by ECG AD converter 105 and blood flow pulse wave AD converter 106 (step 209). A digital filter extracts signals with a required frequency band and noise removed (step 210). The blood flow pulse wave data and ECG data are then formed into data on the same time axis (step 211). Parameters such as RR, Tu, and C are calculated and temporarily stored (step 212). These parameters are obtained for each heartbeat, and are integrated over the number of heartbeats within a predetermined time interval (step 213) to calculate average values ​​(step 214). Next, RR, Tu, and C are substituted into equation (3) to set multipliers for constructing an optimization algorithm (step 215).

[0061] The multipliers assigned until a single value is determined are, for example, n groups, each containing m different values. This grouping may also be used by dividing a string of numbers for a single multiplier by a fixed number. Here, the nth group is specified (step 216). The first value m1 in this n1th group is then specified (step 217), and this is substituted into the formula for CSBPn,m to obtain a value (step 218).

[0062] The difference between this value and the actual measured systolic blood pressure value RSBP for comparison is obtained (step 219). It is determined whether this m value corresponds to the last value (step 220). If it is not the last (no), the process proceeds to step 221, where the next m value is specified, and the CSBPn,m value is obtained in step 218. If it is the last (yes), the average of the difference values ​​for n1 groups is taken, and then the sum of squares DFS is obtained (step 222). In step 223, it is determined whether the group for which the sum of squares DFS was obtained is the last group. If it is not the last group (no), the next n+1th group is specified (step 224), and the process proceeds to step 217, where the first m value of the new n+1th group is substituted.

[0063] If it is the last group (yes), the minimum DFS value is determined in step 225, and the multipliers (a, x, y, z) from equation (3) corresponding to the minimum value are determined (step 226). Upon this determination, the routine ends (step 227), and these multipliers are used to form an optimized equation (3) until the next actual blood pressure measurement.

[0064] From the above operational explanation, an optimized systolic blood pressure estimation equation can be formed, and continuous estimated blood pressure values ​​can be obtained.

[0065] Next, an example of an actual clinical experiment based on the optimized systolic blood pressure estimation formula shown in the above embodiment will be described.

[0066] Example 11: Clinical experiment in hemodialysis treatment A cuff was attached to the upper arm of a hemodialysis patient to measure actual blood pressure, and electrocardiogram electrodes were attached to the right side of the chest, the left side of the chest, and the clavicle to measure electrocardiograms. Furthermore, to detect blood flow pulse wave signals, a sensor for a pulse wave meter or a laser blood flow meter was attached to the fingertip on the side where blood access was to be punctured.

[0067] Before the start of dialysis, electrocardiogram and blood flow pulse wave measurements were started. After dialysis began, blood pressure was measured using an upper arm cuff immediately before dialysis and every 10 to 30 minutes after the start of dialysis to obtain actual blood pressure values. Hemodialysis was performed for a typical period (approximately 3 to 5 hours). The actual blood pressure values ​​measured at regular intervals before and during dialysis, as well as waveform records using an electrocardiograph, pulse wave, or blood flow meter, were converted into digital signals, which were then displayed numerically and recorded in a data format (CSV, TXT, etc.).

[0068] After dialysis was completed, the recorded data was filtered using a digital filter, and the heart rate (RR interval), pulse wave transit time (C), and pulse wave rise time (Tu) were extracted. Average values ​​of these parameters were calculated over a predetermined time interval (30 seconds before and after the actual blood pressure measurement). The average value of the above analysis results within the time period during which the actual blood pressure was recorded using the cuff was calculated and substituted into equation (1).

[0069] Equation (2) is generated by adding multipliers (a, x, y, z) to equation (1) based on the most recent measured blood pressure value, which is measured periodically. Equation (2) is then subjected to mathematical optimization (using, for example, Excel® Solver) to minimize the difference between the calculated and actual blood pressure values ​​while varying the multipliers (a, x, y, z), thereby determining the multipliers (a, x, y, z) and obtaining an optimized SBP estimate. K is an individual value used to further approximate the SBP value to the actual blood pressure value. Between the previous and next measured blood pressure values, the heart rate (RR time), pulse wave propagation time (C), and pulse wave rise time (Tu) were substituted into equation (1) based on the previous measured blood pressure value, and the calculated blood pressure correlation value = f(a, x^b, y^c, z^d) (a, b, c, d: constants, x: heart rate, y: pulse wave velocity, z: pulse wave rise time) shown in equation (2) was formed to determine the multipliers (a, x, y, z) and obtain an estimated SBP (systolic blood pressure).

[0070] The results are shown in Figure 4. The measured systolic blood pressure value and the blood pressure values ​​measured at regular intervals roughly match, making continuous blood pressure measurement possible. In this case, a more accurate formula can be derived by optimizing data from multiple people and multiple times.

[0071] In Example 1 above, data is processed after dialysis treatment to obtain continuous systolic blood pressure data, but it is also possible to measure blood pressure fluctuations in real time using the optimization algorithms shown in Figures 2 and 3.

[0072] Example 2: Measurement of Blood Pressure Changes Due to Exercise Load In this example, a healthy subject was systematically loaded with an electrocardiogram and blood flow pulse wave measurements for blood pressure estimation, and an estimated systolic blood pressure was calculated. More specifically, an electrocardiograph (on the left upper chest) and a blood flow meter or pulse wave meter sensor (on the earlobe) were attached to the exercising subject (healthy subject). Measurement of blood flow pulse wave and electrocardiogram was initiated, and blood pressure was measured using an upper arm cuff at the following times to obtain actual blood pressure values. Measurements were taken in the following order: normal → low exercise load → normal (2 times) → medium exercise load → normal (2 times) → high exercise load → normal (2 times), with the number of times shown in parentheses.

[0073] The measured blood pressure values ​​and waveform records obtained using an electrocardiograph, pulse wave, or blood flow meter were converted into digital signals, converted into data in a data format (CSV, TXT, etc.), and temporarily stored. After the exercise load was completed, the heart rate (RR time), pulse wave transit time (C), and pulse wave rise time (Tu) were extracted from the recorded data. Average values ​​of these parameters were calculated over a predetermined time interval (30 seconds before and after the actual blood pressure measurement). The average value of the above analysis results within the time period during which the actual blood pressure was recorded using the cuff was derived and substituted into equation (1).

[0074] Based on the most recent actual blood pressure value measured periodically, multipliers (a, x, y, z) were added to equation (1) to form equation (2) or (3). While varying the multipliers (a, x, y, z) for equation (2) or (3), mathematical optimization (e.g., using Excel (registered trademark) Solver) was performed to minimize the difference from the actual blood pressure value, thereby determining the multipliers (a, x, y, z) and obtaining an optimized SBP estimate. K is an individual value that further approximates the SBP estimate to the actual blood pressure value. K is a preset value that is specific to each subject.

[0075] The results are shown in Figure 5. Using equation (2) with the optimized multiplier, the estimated systolic blood pressure showed roughly the same changes as the actually measured blood pressure. In Examples 1 and 2 above, all measurements were first taken, converted into digital signals in real time, and then stored in a format such as TXT or CSV. Finally, the R-R interval, Tu, and C values ​​were calculated. However, these examples merely demonstrate the effects of one embodiment of the present invention. Alternatively, the analog values ​​of the electrocardiogram and blood flow pulse wave signals may be converted into digital values, and then the R-R value, C value, and Tu value may be calculated in real time. These values ​​may then be substituted into equation (2) or (3). An algorithm may then be determined that determines the optimized multiplier from the calculation of the variable multiplier, thereby determining the estimated blood pressure.

[0076] Example 3 (Measurement of Blood Pressure Changes Due to Exercise Load (Repeated Squatting and Standing) in a Healthy Subject (Male in His 30s)) Measurement of blood pressure changes due to exercise load on a healthy subject will be described with reference to Figure 6. In Figure 6(a), 601 is the biological information continuous measurement device of the above embodiment, 602 is a pulse wave input unit, 603 is an electrocardiogram input unit, and 602 and 603 correspond to 101 and 102 shown in Figure 1, respectively. 604 is a pulse wave sensor, which is formed with a light source and a light-receiving transistor in a reflective or transmissive configuration and is configured to be worn by clamping it around the earlobe. 605 is an electrocardiogram electrode, consisting of one lead electrode. An example of the electrocardiogram electrode 605 is a self-adhesive type, but other fixing members such as a fixing belt may also be used. Reference numeral 606 denotes a comparative continuous sphygmomanometer. Here, a Finapres sphygmomanometer (Finometer model 1 manufactured by Finapres Medical Systems BV) was used as the comparative measurement device. Reference numeral 607 denotes a fingertip sensor for the comparative continuous sphygmomanometer, which is composed of a cuff that is wrapped around a fingertip and inflates and deflates with air pressure, and a sensor that detects the pressure pulse wave at the fingertip. The comparative continuous sphygmomanometer fingertip sensor 607 and the reference cuff-type sphygmomanometer 608 are conventional pressure-cuff-type sphygmomanometers used for calibrating the biological information continuous measurement device 601 and the comparative continuous sphygmomanometer 606 of the above embodiment, and for measuring reference blood pressure. Reference numeral 609 denotes an upper arm cuff, which is exemplified by an existing combination of a cuff that inflates and deflates with air pressure and a sensor for blood pressure measurement, connected to the pressure-cuff-type sphygmomanometer 608. The upper arm cuff may also be used to calibrate the blood pressure values ​​calculated by the comparative continuous sphygmomanometer 606. Reference numeral 610 denotes a chair on which the subject 6M sits when calibrating the biological information continuous measurement device 601 of the above embodiment and the comparative continuous sphygmomanometer, and on which the subject 6M sits during a low-temperature stress experiment. In Fig. 6(b), reference numeral 611 denotes a cold water container, which is large enough to fit a hand and contains cold water at a temperature suitable for the stress.

[0077] Attachment of Measuring Equipment 1. For subject 6M (the above-mentioned healthy individual), an electrocardiogram electrode 605 connected to a one-lead electrocardiogram input unit 603 was attached to the upper left chest, and a counter electrode (not shown) was attached to the abdomen. A pulse wave sensor 604 connected to a pulse wave input unit 602 was clamped and attached to the left earlobe. 2. An upper arm cuff 609 of a reference cuff-type sphygmomanometer 608 was attached to the right upper arm. For comparison, a cuff of a fingertip sensor 607 for a comparison continuous sphygmomanometer of an existing comparison continuous sphygmomanometer 606 was attached to a fingertip.

[0078] Calibration of the comparative continuous sphygmomanometer 1. The reference upper arm cuff 609 was attached to the upper arm. The comparative continuous sphygmomanometer 606 was turned on, and the comparative continuous sphygmomanometer 606 was calibrated while the subject was seated comfortably in the chair 610. 2. After calibration was complete, the upper arm cuff 609 of the comparative continuous sphygmomanometer 606 was removed, and the upper arm cuff 609 of the cuff-type sphygmomanometer 608 was attached to the upper arm. Note that since both are upper arm cuffs and the same cuff may be used, the same number is used to indicate them.

[0079] Measurement 1. Blood pressure measurement was started using the biological information continuous measurement device 601 of the above embodiment, the comparative continuous sphygmomanometer 606, and the cuff-type sphygmomanometer 608. Blood pressure measurements using the cuff-type sphygmomanometer 608 were performed in the following order: normal (2 times) → squatting (high load) (1 time) → normal (2 times) → squatting (high load) (1 time) → normal (3 times), for a number of times equal to or greater than the number shown in parentheses. 2. Blood pressure information was measured using the biological information continuous measurement device 601, and the blood pressure ratio was calculated. Continuous blood pressure values ​​using the comparative continuous sphygmomanometer 606 and actual blood pressure values ​​using the cuff-type sphygmomanometer 608 were obtained and compared.

[0080] Results The results are shown in Figure 7. As shown in Figure 7, when an exercise load is applied, the blood pressure (SBP) estimated value obtained by the biological information continuous measurement device 601 of the above embodiment was found to track the intermittent blood pressure value obtained by the cuff-type blood pressure monitor 608 and the continuous blood pressure value obtained by the comparative continuous blood pressure monitor 606, with respect to the changing blood pressure value. Furthermore, as a comparative example, it was found that the blood pressure estimated value also roughly matched the blood pressure estimated value obtained by the comparative continuous blood pressure monitor 606.

[0081] Example 4 (Measurement of blood pressure changes due to low temperature load in a healthy subject (male in his 30s)) A blood pressure monitor was attached according to the above-mentioned steps 1 and 2.

[0082] Measurement 1. Blood pressure measurement was started using the biological information continuous measurement device 601 of the above embodiment, the comparative continuous sphygmomanometer 606, and the cuff-type sphygmomanometer 608. Measurements using the cuff-type sphygmomanometer 608 were performed in the following order: normal (2 times) → immersing the hand opposite the upper arm cuff 609 in ice water in the cold water container 611 (for approximately 60 to 90 seconds) (1 time) as shown in Figure 6(b) → normal (3 times) → immersing the hand opposite the upper arm cuff 609 in ice water in the cold water container 611 (for approximately 60 to 90 seconds) (1 time) → normal (3 times), for a number of times equal to or greater than the number shown in parentheses. 2. Using the above formula (2), the calculated blood pressure ratio was compared with the actual blood pressure measured using the comparative continuous sphygmomanometer 606 and the cuff-type sphygmomanometer 608.

[0083] Results The results are shown in Figure 8. As shown in Figure 8, according to the embodiment of the present invention, blood pressure values ​​that followed blood pressure fluctuations due to low temperature stress could be obtained for the continuous blood pressure measurements taken by the reference cuff-type sphygmomanometer 608 and the comparative continuous sphygmomanometer 606.

[0084] The present invention can measure continuous systolic blood pressure values ​​without invasive procedures such as pressurization, and can be applied to more wearable biometric detection devices in treatments such as hemodialysis, in which blood is extracted from the body for blood purification and then returned to the body after purification.

[0085] 101 Electrocardiogram input unit 102 Photoelectric detection unit 103 Electrocardiogram detection unit 104 Pulse wave input unit 105 Electrocardiogram AD converter 106 Blood flow pulse wave AD converter 107 Electrocardiogram filter 108 Blood flow pulse wave filter 109 Arithmetic unit 110 Memory unit 111 Blood pressure signal input unit 112 Blood pressure measuring device 601 Biological information continuous measurement device 602 Pulse wave input unit 603 Electrocardiogram input unit 604 Pulse wave sensor 605 Electrocardiogram electrode 606 Comparison continuous sphygmomanometer 607 Fingertip sensor for comparison continuous sphygmomanometer 608 Reference cuff-type sphygmomanometer 609 Upper arm cuff

Claims

1. A continuous biological information measuring device comprising: an optical blood flow measuring means for detecting blood flow pulse waves from a living body; an electrocardiogram signal detecting means for detecting an electrocardiogram signal from the living body; a biological signal measuring means for actually measuring a target biological signal and detecting a measured value biological signal; a biological signal calculating means for calculating the target biological signal based on a characteristic signal obtained from said optical blood flow measuring means and said electrocardiogram signal detecting means; and a biological signal processing means for mathematically optimizing the measured signal obtained by said biological signal measuring means, the calculated signal obtained by said biological signal calculating means, and the measured biological signal, and outputting an optimized biological signal.

2. The continuous biological information measuring device according to claim 1, wherein the biological signal processing means executes a program having the steps of: extracting feature values ​​from an electrocardiogram signal and a blood flow pulse wave signal to form an algorithm; adding variable multipliers to the algorithm and calculating the algorithm for each variable multiplier to obtain the difference between the value obtained and the actual measured value; and determining the multiplier that minimizes the sum of squares of the difference values.

3. A device for continuously measuring biological information according to claim 1, wherein the blood flow pulse wave is detected by a reflective or transmissive sensor using a laser light emitting element or an LED as a light source.

4. The device for continuously measuring biological information according to claim 1, wherein the actual measured biological signal is measured by an existing pressure-type blood pressure measuring device, a pulse wave propagation time measuring device, or a tonometry-type blood pressure measuring device.

5. The device for continuously measuring biological information according to claim 1, wherein the characteristic signal comprises one or more selected from a pulse wave period signal, a pulse wave amplitude value, a pulse wave propagation time, a pulse wave rise time, an R wave-R wave time signal, an R wave amplitude, and a QRS width.

6. The device for continuously measuring vital signs according to claim 2, wherein the algorithm is a formula for estimating and calculating either the systolic blood pressure value or the amount of change therein, or both.

7. The device for continuously measuring biological information according to claim 1, wherein the characteristic signals are the time between an R wave and the next R wave in an electrocardiogram waveform (RR interval), pulse wave propagation time (C), and pulse wave rise time (Tu).

8. The algorithm (where X is a unique value for each individual, C is the pulse wave propagation time (msec), Tu is the pulse wave rise time (msec), and RR is the RR interval (msec).) The device for continuously measuring biological information according to claim 2.

9. The algorithm with the variable multipliers (a(i), x(i), y(i), z(i)) is (where K is the unique value for each individual, and Tu ａｖｅ is the mean pulse wave rise time (msec), and C ａｖｅ is the mean pulse wave transit time (msec), and RR ａｖｅ 3. The biological information continuous measuring device according to claim 2, wherein: ∑ i = 1 i ⁢ ...

10. A continuous biological information measuring device comprising: optical blood flow measuring means for detecting blood flow pulse waves from a living body; electrocardiogram signal detecting means for detecting electrocardiogram signals from a living body; actual blood pressure measuring means for actually measuring target biological signals to obtain actual measured blood pressure values; characteristic value calculating means for calculating characteristic values ​​from said optical blood flow measuring means and said electrocardiogram signal detecting means; and signal processing means for executing the following steps (i), (ii) and (iii): (i) a step of forming a sum of squares of values ​​calculated by varying a multiplier of the algorithm described in claim 8 or 9; (ii) a step of determining, as an optimal value, a multiplier that minimizes the difference between said sum of squares and the actual measured blood pressure value; and (iii) a step of inputting said characteristic value into said algorithm in which the determined optimal value is substituted as the multiplier, and calculating a blood pressure value.

Images