JPS6333857B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Background of the Invention The present invention relates to blood pressure monitoring, and more particularly to a systolic blood pressure measurement device that automatically measures systolic blood pressure.

Many types of devices have been used to measure the systolic blood pressure of a subject. An older, simpler device consists of a pressurized cuff, or cuff, combined with a mercury manometer, listening for the Korotkoff sounds with a stethoscope, and reading the pressure sensed by the cuff with a manometer. In a more advanced blood pressure measurement method, the distance between the pressure cuff and the artery wall is accurately measured, and the blood pressure is determined by the frequency shift of the ultrasonic wave reflected by the artery due to the blood flow velocity, that is, the Doppler shift. is measured. Other methods of inserting catheters directly into blood vessels have also been employed.

Oscillographic methods of measuring systolic blood pressure are also known. In this method, the operator observes an indication of the pulsation intensity of the intraarterial pressure. This method is performed indirectly by observing the degree of bouncing, or bouncing, of the top surface of the mercury column of a mercury manometer connected to the cuff, or by measuring the closure that occurs in the intraauricular vessels under pressure. It is carried out according to These oscillographic methods generally determine the systolic blood pressure that should be the maximum applied pressure by observing the occurrence of critical oscillations at the top of the mercury column. When using an ordinary mercury pressure gauge and pressurized cuff, the systolic blood pressure is the maximum pressure when the operator observes the bouncing of the top of the mercury column that occurs when the cuff pressure is slowly and uniformly decreased.
However, the above method is inaccurate for measuring critical vibrations because the mercury column does not respond sensitively to pressure vibrations in a narrow range due to its inertia.

The above method or device for measuring systolic blood pressure comprises:
Narrow pressure pulsations have disadvantages, such as an inaccurate response or the need for complex and expensive measurement equipment.

U.S. Pat. No. 4,009,709 for "Apparatus and Method for Measuring Systolic Blood Pressure" describes a relatively simple and automatic accurate method and apparatus for measuring systolic blood pressure that overcomes the above-mentioned drawbacks. . This device changes the pressure within the cuff attached to the subject adjacent to the blood vessel, measures the amount of variation corresponding to the time-dependent fluctuation component, which is an indication of the pulsating pressure within the blood vessel, with the cuff, and applies The maximum value that the amount of variation reaches in response to a change in pressure is determined, the signal of this maximum value is stored, and the amount of variation is stored for an applied pressure that is greater than the applied pressure when the maximum value occurs. determining when approximately half of the maximum value is reached, and further reading an applied pressure corresponding to approximately half of this maximum value;
A pressure corresponding to the subject's systolic blood pressure is measured. The signal obtained from the cuff is a blood pressure pulse component that is proportional to the amplitude of pulsating pressure within the blood vessel and includes a rapid wavefront signal that occurs during diastole and systole, and a pressurized cuff pressure component that is added from outside the blood vessel by the cuff. Includes the amount of variation proportional to the total amount including .

FIG. 1 shows a block diagram of a portion of the systolic blood pressure measuring device described in US Pat. No. 4,009,709. The block diagram shows a filter network 100 whose output is connected via an amplifier 102 to the input of a peak-to-peak detector 104. Filter network 100 receives input signal 106a on input conductor 106. Filter network 100 receives input signal 106a on input conductor 106. The input signal 106a includes a signal in which a blood pressure pulse component, which is an indication of the pulsating pressure in the blood vessels of the subject, is superimposed on a slowly increasing ramp voltage, which is an indication of the cuff applied pressure, and a fluctuation component of this pulsating pressure. has a sharp rising wavefront signal compared to other components during the rise in blood pressure that changes from diastole to systole. Filter network 100 is configured to remove the effects of the ramp voltage associated with the cuff applied pressure from output waveform 100a. The effect of gradient voltage is
Since the filter network 100 is configured with the amplitude of the input signal 106a assumed to be a component that varies linearly, any irregular disturbance that does not conform to this assumption causes the filter network 100 to
cannot remove such disturbances and may cause errors in the output. The irregular disturbances may include, for example, the activity of the patient's voluntary or involuntary muscles;
This includes artifacts that occur due to various causes, such as pressure-related noise that occurs in the blood pressure measurement environment, oscillations or sudden vibrations of the air tube connecting the cuff, etc. In the device of that patent, the signal obtained from the cuff, as described above, is sent to a filter network to remove the effects of voltage gradients due to cuff pressure. The resulting oscillatory signal is believed to be proportional to the magnitude of the pulsatile pressure within the blood vessel, and a peak-to-peak detector then makes an amplitude measurement that is utilized in signal processing. However, even if a linear characteristic is assumed for the ramp voltage, irregular and uncontrollable deviations may cause errors in the amplitude measurement. For example, if a large disturbance occurs in the gradient voltage due to cuff pressure, the filter 100 will require considerable time to recover, so the reference line 100b (dashed line) measures a fluctuating signal proportional to the amplitude of the pulsating pressure in the blood vessel. There is a possibility that some changes may occur. Therefore, each time peak-to-peak detector 104 is operated in response to sampling signal 108,
The output signal 110a appearing on the conductor 110 may produce an erroneous output from the peak-to-peak detector 104 due to variations in the reference line 100b.

Further, the artifact includes a dichroic wave. Generally, as the heart contracts and expands, the carotid artery,
Pressure pulse waves in the subclavian artery, radial artery, tibial artery, etc. change as shown in the ECG waveform. In the ECG waveform, the initial rapid rise comes from the ventricles contracting and ejecting blood. It then descends to form a trough, which is due in part to blood flowing backwards due to the closure of the aortic valve. The small surge that follows is called a dichrotic wave, and is thought to occur as a reaction to the closure of the aorta. The dichrotic wave has a portion of zero slope and therefore may inadvertently mix into the systolic blood pressure wave. However, with conventional systolic blood pressure measuring devices, it has not been possible to eliminate the influence of such dichrotic waves.

SUMMARY OF THE INVENTION An object of the present invention is to provide a systolic blood pressure measuring device that can prevent the influence of disturbances occurring in the gradient voltage due to cuff pressure.

Another object of the present invention is to provide a systolic blood pressure measuring device that can prevent the influence of disturbance even if an artifact including a zero slope such as a diclothetic wave occurs in a fluctuating signal obtained from a cuff.

Other objects and advantages of the invention will become apparent to those skilled in the art from the following detailed description of the preferred embodiments and accompanying drawings.

The systolic blood pressure measuring device of the present invention includes a transducer that is connected to a cuff that applies variable pressure adjacent to a blood vessel of a subject and that converts blood pressure pulses in the blood vessel into an electrical signal, and an electric signal of the transducer. a differentiator network for converting the signal into a time derivative signal; an accumulator for integrating the time derivative signal to produce an integral signal proportional to the amplitude of each blood pressure pulse; a storage device that stores the maximum value of the signal; a maximum comparator that sequentially compares the above integral signal with the next consecutive integral signal and generates an output that causes the maximum value to be stored in the storage device; and a systole that compares the fractional value with the integral signal at a cuff pressure greater than the cuff pressure when the maximum value is reached, and produces an output when the integral signal is approximately equal to the fractional value. It is composed of a comparator and a readout device that receives the output of the systolic comparator and the electrical signal from the transducer and reads out the applied pressure as the systolic blood pressure of the subject.

In addition, a systolic blood pressure measuring device according to another embodiment of the present invention includes a transducer connected to a cuff that applies variable pressure adjacent to a blood vessel of a subject and converts blood pressure pulses in the blood vessel into electrical signals. a differentiator network for converting the electrical signal of the transducer into a time derivative signal; an accumulator for integrating the time derivative signal to produce an integral signal proportional to the amplitude of each blood pressure pulse; While changing
a storage device that stores the maximum value of the integral signal; a maximum value comparator that sequentially compares the integral signal with the next consecutive integral signal and generates an output that stores the maximum value in the storage device; a dividing device that counts into fractional values, and a cuff pressure greater than the cuff pressure at which the maximum value is reached, compares the fractional value with the integral signal, and produces an output when the integral signal is approximately equal to the fractional value; a systolic comparator; a readout device that receives the output of the systolic comparator and the electric signal from the transducer and reads out the applied pressure as the systolic blood pressure of the subject; and the time derivative signal as a limit level signal. an enabling logic circuit that compares and supplies the integral signal of the accumulator to the storage circuit when the time derivative signal is equal to or higher than the limit level signal.

As mentioned above, the object of the present invention is to provide a novel systolic blood pressure measuring device, which can eliminate the influence of unnecessary components mixed into the signal obtained from the cuff by two methods. . One of the unwanted components is the deviation effect of the linear gradient voltage and the very low frequency disturbance noise that may appear in this gradient voltage. In order to remove such unnecessary components, the systolic blood pressure measurement device of the present invention includes a "differentiation circuit network that converts the electrical signal of the transducer into a time derivative signal" and an "accumulator." That is, after first or second differentiation of the electrical signal representing the blood pressure pulse obtained from the cuff,
By integrating, the deviation of the linear gradient voltage and the low frequency disturbance noise can be completely removed.

The integral signal obtained from the time-derivative signal with unnecessary components removed is the differential signal in a certain range, especially the signal corresponding to the amplitude of the fluctuating signal between the systolic rise and the systolic peak at each blood pressure pulse. It is formed. Furthermore, the maximum value of the integral signal is determined and stored, and then a certain fraction (e.g. 1/2) of that maximum value is determined and stored.
An integral signal corresponding to , along with an applied pressure corresponding to that fraction, is determined. This applied pressure corresponds to the systolic pressure. In the present invention, the maximum amplitude of an electrical signal representing the subject's blood pressure obtained from a transducer is measured at a certain cuff pressure, and the cuff pressure is further increased to approximately half the maximum amplitude of the electrical signal. By measuring the cuff pressure when electrical signals of approximately equal amplitude are generated, it was discovered that this cuff pressure is equal to the subject's systolic blood pressure.Using this principle, contractions can be performed without errors due to artifacts, etc. Periodic blood pressure measurements can be performed.

In addition, as other unnecessary components, there are artifacts including zero-slope parts such as dichrotic waves, but such artifacts can be removed by using effective logic circuits to detect when a time derivative signal exceeding a limit level is generated. This is done by passing the integral signal.

Detailed description of preferred embodiments Embodiments of the present invention will be described below with reference to the drawings.

As shown in the block diagram of FIG. 2 of the systolic blood pressure measurement apparatus of the present invention, an input signal having the same waveform as waveform 106a of FIG. The configuration of the differentiator network 200 is such that the configuration of the differentiator network 200 is such that the signal 2
In addition to performing the first differentiation of 06a, the input signal at a frequency lower than the frequency range is second-order differentiated to remove the deviation effects of the linear gradient voltage and the effects of very low frequency disturbances that may appear on this gradient voltage. .

A filter having the properties required for the differentiating circuit network 200 has a frequency range f ₁ −, as shown in FIG.
For f ₂ , per octave solope -
6db, and has a gain versus frequency characteristic with a slope of -12db for frequencies below _f1 . That is,
In FIG. 5, the frequency range f ₁ -f ₂ corresponds to the bandwidth of the Pac signal that includes fluctuations that are an indication of pulsatile blood pressure. The bandwidth portion of input signal 206a, which is an indication of pulsatile blood pressure, is differentiated and appears at the output of differentiator network 200 as signal 200a (hereinafter referred to as the P signal).

This P signal (200a) is sent to the input end of an integrator 204 via an amplifier 202, and by integrating the P signal over a predetermined interval of each pulse, an output value corresponding to the peak-to-peak pressure of each blood pressure pulse is obtained. . A sample-and-hold circuit 205 connected to integrator 204 detects integrator 20 at the end of each integration interval.
The value appearing at the output of 4 is sampled and held for an intermediate period until the integration of the next blood pressure pulse begins. The integrator 204 and sample-hold circuit 205 are controlled by the reset/integrate/hold signal 2.
08, this signal has the function of controlling the integration period and clearing the integrator before each new integration. The output of sample-and-hold circuit 205 appears as waveform 210a on conductor 210,
The size of this waveform corresponds to the area included in the corresponding portion of the waveform P signal to be integrated.

To understand the basic theory of the present invention, FIGS. 3A to 3D will be explained. As mentioned above, when the amount of variation reaches about half of this maximum value, the cuff applied pressure becomes equal to the systolic blood pressure. The maximum value of this variation is determined by measuring the diastole and systole within a continuous blood pressure pulse. Maximum P-
The pulse exhibiting P amplitude is stored as the maximum value.
Also, the applied cuff pressure is further increased such that the P-P amplitude decreases, and the systolic pressure decreases as the P-P pressure decreases to P
-P Determined by the applied cuff pressure at half the maximum value.

FIG. 3A shows the blood pressure pulse Pac, which is an indication of the pulsatile pressure within the blood vessel. The bottom of the trough of the Pac waveform corresponds to the diastole, more precisely the end diastole of the heartbeat.
The peak SP of the waveform corresponds to the systolic phase of the pulsation. As previously discussed, the signal obtained from the cuff is differentiated to remove ramp voltages and low frequency random disturbances due to the applied pressure, resulting in the P signal, which is the derivative of waveform Pac, as shown in FIG. 3B. Since the waveform Pac exhibits zero slope at both end-diastole (ED) and peak systole (SP), the P signal, which is a derivative waveform, has zero amplitude at these times. Also, Pac, ED and SP
Since the period of contraction rise during this period shows a positive slope, it is above the zero reference line within this period. Since the zero crossing points ED and SP of the P signal waveform correspond to the maximum amplitude points between consecutive Pac pulses, the area above the zero reference of the P signal waveform between the end-diastolic ED and the systolic peak SP corresponds to each blood pressure. The value corresponds to the P-P value of the pulse. This area is determined by integrating the "greater than zero" portion of the P signal waveform.
It should be noted that end-diastole (ED) essentially corresponds to the onset of the rise in peak systole (SP).

Figure 3C clears or resets the integrator before the integration period, then integrates the P signal during this integration period, and finally samples the integral value as an indication of the P-P value for each blood pressure pulse. A hold control signal is shown, which is similar to signal 208 of FIG. When this control sequence of control signals is repeated, the reset operation is represented by R, and the sampling and holding operation is represented by S+H. The sampled integral signal is held for a longer time than the short duration of the S+H signal of FIG. 3C.

The integration result of the P signal waveform between the limits ED and SP is shown in Figure 3D. The amplitude of the integral signal at the systolic peak SP corresponds to the PP value of each blood pressure pulse.

The P signal waveform is integrated during the systolic rise to obtain the P-P value for each blood pressure pulse. In this case, standard circuitry can be used to determine when the P signal waveform crosses the zero reference in the positive direction to begin integration, and when the P signal waveform crosses the zero reference in the negative direction to end integration. Detection and further sample-hold functions are performed. The integrator is reset immediately after sample and hold and typically continues to reset until the next positive going P signal zero crossing. The resulting integral can be considered an indication of the P-P value for each blood pressure pulse. However, due to certain characteristics of the Pac waveform or the presence of artifacts during the diastolic drop, a P signal above the zero reference may appear for short periods of time other than between end-diastole and systolic peak. For example, as shown in FIGS. 3A and 3B, when an artifact (ART) such as a dichratic wave having a zero slope portion occurs just before the end of diastole when the slope of the Pac waveform is relatively flat, the P signal The waveforms represent some of this artifact as “greater than zero” values and are shown in Figures 3C and 3.
Generates the temporary value shown in the box in Figure D.

According to one feature of the invention shown in FIG. 9, a threshold level is provided to distinguish between P values greater than zero associated with systolic rise and other signals, such as artifacts not associated with systolic rise. The amplitude of the P signal associated with the systolic rise is usually much larger than the zero or greater amplitude portion of the signal due to artifacts, etc., and these signals can therefore be distinguished. Integration of the "greater than zero" passes between each ED and SP limit is performed using the identification of the P signal waveform exceeding the limit level during certain "greater than zero" passes.

In FIG. 9, the same reference numerals are used as the corresponding elements in FIG.
0 to produce the P signal, which is then differentiated by amplifier 2.
02 to the respective input terminals of the integrator 204, limit detector 912, and zero passage detector 907. Zero crossing detector 907, although not shown in FIG. 2, may correspond to a device that determines the interval of integration and generates a control signal. Limit detector 912
determines the signal amplitude limit value, and P above this limit value
The signal waveform is assumed to be an indication of the effective systolic rise.
When the P signal waveform exceeds the limit level of limit detector 912, a signal is sent to the input of enable logic circuit 914 indicating that the limit level has been exceeded. Similarly, the valid logic circuit 914 is the zero-crossing detector 90
7, and limits the time when the P signal waveform passes through the zero reference in the positive and negative directions, that is, the time when it crosses the zero reference.

As discussed in more detail with respect to FIG. 4, the enabling logic circuit 914 includes bistable elements that eliminate the effects of artifacts. In other words, the dichrotic wave has a portion of zero slope and therefore may inadvertently mix into the systolic blood pressure wave. Therefore, the limit detector 912 can be provided with a comparator that sends an output to the enable logic circuit 914 to prevent the introduction of dichrotic waves into the systolic blood pressure pulse.

The enable logic output 908' is applied to the reset input of integrator 204 to reset the integrator upon the beginning of the desired integration period beginning with the P signal passing through the zero reference in the positive direction. Effective logic circuit 914
The output 908 of is an optional component sample-
It is applied to the “sample” input of hold circuit 205. Integration between the positive and negative going zero reference crossings of the P signal waveform occurs only if the limit detector 912 produces an output as the P signal waveform within the systolic blood pressure pulse interval is effectively a valid systolic rise. Vessel 2
The integral value accumulated in step 04 is stored. The sample −
The output 910 of the hold circuit 205 is different from the output 210 of FIG.

If the "above zero" portion of the P waveform does not change amplitude within successive pulses, a fixed amplitude limit above the zero reference could be used, but this method is impractical, especially when the pressure signal Pac is at low applied pressures. It cannot be used for blood pressure monitoring using the current oscillographic method, which increases from a small amplitude at , to a large amplitude at high applied pressures, and decreases to a small amplitude at even higher applied pressures. The threshold level, indicated by TRLD in FIG. 3B, is therefore selected to be a function of the amplitude of the systolic rise portion of the P signal that is greater than one or more immediately preceding blood pressure pulses. In the P signal waveform, the amplitude increase (and then decrease) of the continuous systolic rise is gradual, and in the non-systolic rise component of the P signal waveform, the relative amplitude "above zero" is sufficiently small, so the dynamic limit If the level is matched to 50% of the maximum "above zero" amplitude that rises during systole in the P signal waveform of the preceding pulse, it will effectively identify only the "above zero" portion of the P waveform that rises during systole. be able to.

The above dynamic limit level is determined by summing and calculating several preceding systolic rises of the P signal waveform; in this case, the limit level TRLD is 50% of the amplitude of the rise in the immediately preceding systole. The selection level could be greater or less than . An example of a type of analog dynamic limit detector suitable for the above calculations is described in U.S. Pat. No. 3,590,811, entitled "R-Wave Detector for Electrocardiography." A digital device for determining dynamic limit levels is described in detail below.

A systolic blood pressure measuring device according to an embodiment of the present invention will be described in detail with reference to FIGS. 4, 6, and 7. First, the systolic blood pressure measuring device of the present invention will be explained with reference to the functional block diagram shown in FIG. This device digitally processes the analog signal sent from the transducer 23. However, of course, an apparatus that performs analog processing is also possible. In particular, separate electronic devices, separate digital devices, microprocessor technology and equipment, or digital computers may be used. Figures 6 and 7 are
1 is a state diagram and a flowchart, ie, a decision tree, for processing the P signal between successive blood pressure pulses, each corresponding to a continuous blood pressure. Generally, the signal processing process of the blood pressure monitoring device and method of the present invention shown in FIGS. 6 and 7 is as follows:
This corresponds to the operating process of the apparatus of FIG. 4 for integrating signals.

A conventional blood pressure cuff 15 is wrapped around an artery in the subject's arm 11 to apply a selective variable pressure adjacent to the subject's blood vessels. The brachial artery is usually used. A pump 19 and a pressure transducer 23 are attached to the cuff via conduits 17 and 21. The transducer 23 has a transmission function that allows this electrical output to indicate the pressure input up to the limit of the pressure information contained in the pulsating pressure. This transducer is used to measure cuff pressure, which is reflected in waveforms 106a and 206 of FIGS. 1 and 2.
It is the sum of the pressure exerted by the pump and a portion of the pressure generated by blood pressure fluctuations in the artery, as shown in a. That is, a fluctuating signal comprising a fluctuating component corresponding to the amplitude of the blood pressure pulse, which is indicative of the pulsating pressure within the blood vessel, produced by the transducer 23, and a selectively variable pressure component applied externally adjacent to the blood vessel.
Pac is measured. The varying portion of the output of transducer 23 represents the amplitude of the pulsatile blood pressure. The output of transducer 23 is delivered via conductor 24 to one input of multiplexer 25 . The output of transducer 23 passes via conductor 26 and normally closed switch 27 to the input of differentiator network 28 . The output of the differentiating network 28 is applied via a line 29 to an amplifier 30 and then via a line 31 to the other input of the multiplexer 25.

The differentiator network 28 is equivalent to the differentiator network 200 shown in FIGS. 2 and 9, and differentiates the input signal in the f ₁ -f ₂ band of the signal Pac as shown in FIG.
Also, double differentiate frequencies below f ₁ . In this way, only the signal appearing on the conducting wire 31 is represented by the differential (P) of the Pac waveform, and the fluctuation signal Pac is converted into a time derivative signal.

The output of multiplexer 25 is sent via conductor 32 to AD converter 33 . The output of the AD converter 33 is applied to the input terminals of gates 40 and 42 through conductive wires 48, respectively. Clock 34 generates timing pulses which are supplied on lead 35 to a timing controller 36 which switches multiplexer 25, converts analog signals to digital signals, and gates gates 40 and 42. control. One output of the timing controller 36 goes on a lead 44 to the multiplexer 25, the AD converter 33, and the other input of the gate 40, which converts the P signal on lead 31 to a digital signal, which is passed through the lead 48 to the gate 40. given to 40. The other input of timing controller 36 goes on conductor 46 to multiplexer 25 , AD converter 33 , and other inputs of gate 42 and to transducer 23 .
converts the analog signal from the gate into digital form which is applied to gate 42.

The gating signals appearing on conductors 44 and 46 connected to the inputs of gates 40 and 42, respectively, have a sufficiently large duration so that the digitally converted data appearing at the other input of each gate passes through this particular gate. . Of course, the illustrated conductors 44 and 4
The control signals through 6 are mutually exclusive so that data appearing on conductor 31 or 24 is connected to the corresponding gate 40 or 42, respectively.

The period between successive blood pressure pulses is approximately 800-1000 milliseconds, and this systolic rise accounts for approximately 10-20% of this period. The P signal is integrated over the systolic rise by sampling a sufficient number of incremental signals Pi of the P waveform that approximate the area contained within the P waveform. Usually 10–20 of 100–200 msec of systolic rise
A sufficient number of Pi increment signals can be obtained with samples of
The speed of clock 34 and the control signal on lead 44 are also selected depending on the number of samples. At least once during each blood pressure pulse, the control signal on lead 46 is actuated to convert the signal on lead 24 into a digital signal, which is then calculated as an indication of the applied pressure on cuff 15, as described below. - provide as input to the averaging device 83;

Each time a convert-gate signal appears on conductor 44 from timing controller 36, an incremental sample Pi of the P waveform synchronized to the convert-gate signal is applied to gate 4.
Appears at the output end of 0. Incremental sampling of the P waveform is repeated for the duration of the blood pressure pulse and at least during the time that the P waveform is above the zero reference level. The system timing of the timing controller 36 is determined to sample the waveform on lead 24 at least once between each blood pressure pulse at a time that does not conflict with the sampling of the P waveform on lead 31.

Each sampled incremental signal Pi is sent to gate 4
0 to gate 52 with conductor 50 and zero passage detector 9
07 (FIG. 9). Pi to comparator 54
The input is compared to a so-called zero reference value (Pz) supplied on lead 60. The output of the comparator 54 sent out from the conductor 62 is 1 if the Pi value appearing at the input end of this comparator is greater than the zero reference value Pz.
If the display level is equal to or less than Pz, it is set to zero display level. It is assumed that the reference value Pz on the conductor 60 corresponds to the zero reference of FIG. 3B. The actual voltage developed on conductor 60 is adjustable and illustratively expressed as the wiper voltage of potentiometer 20, the terminals of potentiometer 20 being connected to a voltage above and below the voltage corresponding to the appropriate Pz voltage. . To accurately set the correct Pz voltage, switch 27 is momentarily activated so that a "zero" value appears as the Pi output from gate 40 on lead 50, and the Pz value on lead 60 is
It is adjusted to be equal to the value on conductor 50.
Of course, the Pz voltage level appearing on conductor 60 is
It appears in digital form (by an AD converter, not shown) for comparison with the digitized Pi value appearing on conductor 50.

Each time a Pi value exceeds Pz, one output on conductor 62 activates gate 52 to set the particular Pi value on conductor 64.
The signal is sent to the gate output terminal represented by . Therefore, conductor 6
The data appearing in 4 is an indication of Pi values greater than Pz.

Each Pi value appearing on the conductor 64 is shown in FIGS. 2 and 9.
The P waveform is integrated by summing successive increments Pi within a corresponding time period that performs incremental sampling. do. After summing each consecutive Pi increment into a total, Pi
The cumulative total of the increments is sent out on a lead as the output of the accumulator 66. The output of this conductor 68 is applied to the input terminal of the gate 78, and the pulse
It is introduced into an interpulsing circuit or averaging device 39.

The completion of the time interval during which the accumulator 66 accumulates the Pi increments is determined by whether the Pi increments are equal to or less than the zero reference value Pz.
To identify this point of completion, the Pi value on lead 50 and the Pz value on lead 60 are applied as two inputs to comparator 56 such that if Pi is less than or equal to Pz, the output of lead 70 is Set from zero to one.

The conductor 6 is connected to the set input terminal S of the bistable element 72.
The output of the comparator 54 is sent from 2, and Pi is initially Pz
When the output of the comparator 54 changes from zero to one, the Q output of the bistable element 72 is set to one. Similarly, the output of comparator 56 is applied from conductor 70 through delay element 74 to reset input terminal R of bistable element 72, and the output of bistable element 72 is
When Pi becomes less than or equal to zero, it is similarly set from zero to one. In this way, the bistable element 72 has the function of identifying the time interval in which the increment Pi of the P waveform exceeds Pz.

The output of bistable element 72, delivered on conductor 76, is applied to the set input terminal of accumulator 66, and a one output clears, or resets, accumulator 66 to zero. The accumulator 66 is in the reset state not only when the Pi waveform first falls below the Pz reference level, but also throughout the entire period during which it is maintained below the Pz reference level. In this way, the accumulator 6
6, the P waveform is prevented from integrating any part other than the part exceeding the Pz reference level. A delay element 74, which may be a monostable multivibrator or the like, is used for a time smaller than the interval between each successive Pi waveform, until the final integrated value passing through conductor 68 is sent through gate 78, as described below. It has the function of delaying the reset of the bistable element 72 by a time sufficient to delay the reset of the accumulator 66.

A gating pulse on lead 80 is applied to gate 78 at the end of each systolic rise time to reach the limits ED and SP.
The time integral of the P waveform produced during this period appears at the output terminal of gate 78, the output of gate 78 passing through conductor 82 to the beat-to-beat circuit or averaging device 39. At the end of each systolic rise, the Pi value equals the Pz value, at which time the output of comparator 56 resets bistable element 86, producing an output from the terminals. Through the terminal output, the pulse generator 92
generates an inverted gate pulse to gate 78. A gating pulse for gate 78 needs to be generated only at the end of each systolic blood pressure. Therefore, when a V-shaped pulse is generated in the "zero or more" portion of the P wave due to the heart's Dicluate wave, the conductor 50 is passed through so as not to generate the gate pulse.
The Pi increment is compared with the storage limit value sent from conductor 84 in comparator 58, and if this portion is greater than or equal to the storage limit value, bistable element 86 forming active logic circuit 914 (FIG. 9) is set. .

The memory limit value appearing on lead 84 is determined such that the amplitude of the Pi increment is approximately half the maximum amplitude during the systolic blood pressure rise of the previous blood pressure pulse, as described below. The output of comparator 58 included in limit detector 912 shown in FIG.
6 and goes from zero to one when the first Pi increment exceeds the storage limit. Therefore, the Q output value of bistable element 86 is set to one level, and the signal on conductor 70 is sent to the reset input terminal R of bistable element 86.
The Q output value maintains the above 1 level until it is reset. When Pi increment first becomes ≦Pz,
A 0-1 signal transition on conductor 70 occurs. At this moment the output is provided via conductor 90 to pulse generators 92 and 94, respectively.

Pulse generator 92 has a 0-1 input conversion function and generates an output gate pulse on lead 80 which causes the integral value on lead 68 to pass through gate 78. Of course, the delay time provided by delay element 74 is sufficiently large that the integrated value on lead 68 passes through gate 78 before accumulator 66 is reset by the reset signal on lead 76.

To describe the dynamic limit values generated in limit detector 912 and compared to the P waveform to pass or reject "above zero" waveforms as systolic rising waves, a conductor that is an increment of Pi greater than the Pz value The signal on 64 is sent to each input terminal of comparator 95 and gate 96. Comparator 95 controls gate 96 with a signal on conductor 97. Gate 96 has the function of inputting the selected Pi increment to storage 98 via conductor 99. Pi stored in storage device 98
The value is applied at lead 120 to the other input terminal of comparator 95. If the incoming Pi increment oscillates more than the Pi increment stored in 98, the comparator 95
The output always goes from zero level to gate activation one level at conductor 97. The Pi increments thus maintained in storage 98 will represent the maximum amplitude Pi increments for the time of the particular "greater than zero" portion of the P waveform.

At the time the P waveform again passes through the Pz reference level from below, the value stored in storage 98 is an indication of the maximum value of Pi that occurs between the immediately preceding "greater than zero" P waveforms. This retained Pi maximum value is the conductor 12
4 is sent to a "divide by 2" circuit 122 to generate a limit value, which limit value is sent to storage 126.
However, the division of the maximum value of Pi and the storage in the storage device 126 is carried out from the output terminal of the pulse generator 94 to the conductor 128.
It is not executed until the gate signal is received at Conductor 8
Similar to the gating signal of the pulse generator 92 through 0, the limit value determined by the preceding systolic rise is compared with the current "above zero" P waveform and this P
Only when the waveform is recognized as a systolic rise that can be effectively passed through is a gating signal through conductor 128 generated, which occurs once the P waveform falls below the Pz reference level. At the same time, 1/2 of this new P maximum value is sent via conductor 130 to limit value storage 126 and becomes the new limit value.
The next “zero or more” of the P waveform for this new limit value
Passages are compared.

Similarly, a short time after the new limit value has been stored in storage device 126, a "reset" signal through conductor 76 is sent to the "reset" or clear input of storage device 98 to clear storage device 98. do. Of course, the reset of the storage device 98 that stores the Pi maximum value occurs not only when the passage of a P waveform "above zero" is recognized as a systolic rise, but also after the passage of another "above zero" portion of the P waveform. , this latter reset is necessary to avoid inputting the Pi value when artifacts occur when determining the Pi maximum value and generating a new limit value.

To explain the signal value that includes the P waveform integral value between the limit ED and SP and is supplied by the conductor 82,
This signal value is used to determine the systolic rise period for use in the beat-to-beat circuit (averaging device 39) described in the aforementioned US Pat. No. 4,009,709. Briefly, the time integral of the P waveform appearing on lead 82 is provided to an optional averaging device 39 (shown in phantom). The averaging device 39 averages, for example, four consecutive blood pressure pulses Pac, and this average value is applied from a conductor 49 to the respective input terminals of the maximum value comparator 43, the gate 47, and the systolic comparator 63. Comparator 43 controls gate 47 via conductor 45 . Gate 47 causes selected values of the (averaged) P-systolic rise integral representation of the pulse P-P values to be applied from averaging device 39 via lead 49 to memory 53 via lead 51.

The numerical value stored in the storage device 53 is supplied to the comparator 43 via a conductor 55. The signal previously stored as the temporal maximum value of the P-systolic rise integral value is compared in comparator 43 with the current value introduced to comparator 43 on lead 49. If the number supplied to comparator 43 on lead 49 is greater than the number temporarily stored in storage 53 and delivered to comparator 43 on lead 55, the comparator sent out on lead 45 The gate 47 is operated by the signal 43, and the above numerical value is stored in the storage device 53 as a temporary maximum value.

The temporary maximum value of the P-P quantity is introduced via a conductor 57 into a dividing device 59 (divide by 2) where it is halved. This two-divided value is sent to a systolic comparator 63 via a conductor 61. The current value of the P systolic rise period (averaged) is sent from lead 49 through a branch lead to systolic comparator 63 . When the systolic comparator 63 determines that the value supplied to the systolic comparator 63 from conductor 49 through the branch conductor is approximately equal to 1/2 of the temporal maximum value supplied from conductor 61, the systolic comparator 63 sends a command to the switching device 69 to switch the pump 1
9 and evacuates the cuff 15 via the solenoid valve controller 20, the stop and evacuation commands being sent via the conductor 71.

The switching device 69 and the systolic comparator 63 send commands to the readout device 75 via conductors 73 and 93, respectively, to read out the numerical value of the applied pressure applied to the cuff 15 by the pump 19, and calculate the value of the applied pressure (P −
The exact applied pressure corresponding to P) is determined.

The applied pressure value is read out by the readout device 75 using conductors 77 and 79.
sent to. Conductor 77 introduces the applied pressure measurement just before the measured quantity becomes less than half of the maximum value, and conductor 79 indicates the applied pressure at the moment when said measured quantity becomes approximately equal to half of the maximum value. In other words, conductor 7
9 indicates the latest value of applied pressure. The applied pressure value is
Obtained by introducing the digitized waveform from transducer 23 via gate 42 into "final" (current) storage 83 via conductor 24, storage 83 stores a plurality of consecutively occurring current inputs. It has a function of averaging pressure values (for example, 4 values).
As each latest applied pressure value occurs, the average value stored in the storage device 83 is updated with the previous average value, which is sent via a conductor 89 to the storage device 81, which stores the previous average value. remember.
The reading device 75 interpolates the applied pressure values stored in the storage devices 81 and 83, respectively. That is, the reading device 75 calculates continuous applied pressure values from the applied pressure values stored in the storage devices 81 and 83, and determines the intermediate applied pressure that corresponds to the case where it is approximately equal to 1/2 of the maximum value. It has the function of

The analog signal on lead 24 is the systolic peak SP.
It is usually sufficient to utilize the signal from bistable element 86 through lead 90 as a timing input to timing controller 36, for example, for conversion to a digital signal as it occurs. The analog signal on conductor 24, which ultimately appears as a digital signal on conductor 48, essentially consists primarily of a slowly increasing ramp voltage and a very small fluctuation signal indicative of the pulsating pressure superimposed on this ramp voltage. be done. The signal Pac value on lead 24 is therefore approximately proportional to the pressure applied to cuff 15, although lead 24 can be provided with a low pass filter if it is necessary to remove pulsating components.

In the embodiment shown in FIG. 4, for example, the pump 19
Drawings of circuit configurations and control circuits that are obvious to those skilled in the art, such as circuits that supply activation signals that clear or reset all memory elements to appropriate initial states during operation, are omitted. Typically, this initial state represents a "zero" state. However, if desired, limit value storage 126
A "non-zero" minimum limit value may be initially stored in the systolic rise signal, and the process of confirming the validity of the systolic rise signal may be activated at the very first blood pressure pulse.

In addition, P-P is a general indication of each blood pressure pulse.
The systolic rise portion of the P waveform may be integrated within a range somewhat greater or less than the preferred ED-SP limit to obtain the value. In such a case, conductor 6
The Pz reference appearing at zero and utilized by the various comparators can be set to a value slightly above or below zero.

Furthermore, in the above example, the Pi limit value was divided by 2, but if the cuff is configured to apply maximum pressure to one side or the other of the normal position, which is the center of the cuff, another suitable value can be used. We could use a divisor.

Furthermore, in connection with digitizing the P waveform, determining the systolic rise signal limit, and determining the effective systolic rise signal, the embodiment of FIG. This is just an example. For example, determining when the P waveform passes above or below the reference value Pz and whether a valid systolic rise signal has occurred can also be done on the analog P waveform before digitization and is displayed as Pi increments. And the above determination can also be used to control the portion of the P waveform introduced into the accumulator 66. Comparator 56 and bistable elements 72 and 86
can also be replaced with analog circuits with similar functionality.

Those skilled in the art will appreciate that the implementation of the functions shown in FIG. 4 may be accomplished using commercially available components. Other functional blocks, other than the compressed air drive, compressed air drive controller, and initial analog circuitry associated with transducer 23, can be constructed primarily from commercially available microprocessors and other digital signals. In fact, the following discussion of the state block diagram of FIG. 6B and the flowchart (decision tree) of FIG. 7 in relation to the waveforms of FIG. It specifies the operation sequence.

FIG. 6A shows a portion of the P waveform divided into five types of time division states in relation to the signal processing state diagram of FIG. 6B. The detailed decision tree of FIG. 6A and the resulting FIG. 7 will provide those skilled in the art with sufficient information regarding techniques using digital systems, such as microprocessors, to embody the concepts of the present invention. .

To explain the command sequence in FIG. 7, between the systolic peak SP and the end ED of the preceding blood pressure pulse, in other words, before the start of the systolic rise of the current blood pressure pulse, the storage device 98 containing the maximum value of Pi is cleared. Then, a clear signal is given to the storage device (accumulator 66) containing the Pi sum (ie, the P integral value between ED and SP). The short-term increments of the P waveform are then converted to digital values, denoted Pi, which indicate the magnitude of the particular increments. Assume that the system is in state "1" or possibly "2" of waveform P in FIG. 6A, and that Pi is then compared to Pz. If Pi is less than or equal to Pz, the system remains in state "1" and the sequence returns to the "Clear Pi Total" process and repeats this sequence. On the other hand, if Pi is larger than Pz, the state becomes "2", and Pi is larger than the reference Pz.
Since the magnitude of is added to the sum of Pi increments (in accumulator 66), the Pi total value increases.

The program then determines whether a valid systolic rise signal has been recognized based on the memory limit settings. If state "3" in FIGS. 6A and 6B has not been reached, in other words, assuming that the storage limit value has not been set, it is determined in this sequence whether Pi is smaller than the storage limit value. Since the system is still in state "2", the program branches and repeats the sequence starting with the next Pi incremental conversion of the P waveform. The sequence of state “2” continues until Pi≧limit level,
At this point, the memory limit value is set and the state is “2”.
A transition from the state to state "3" is performed.

The above sequence is then memorized by this particular Pi increment.
Determine whether Pi is smaller than the maximum value. Memory Pi
Since the maximum value is determined by the maximum preceding Pi increment, normally Pi will never be less than the stored Pi maximum value during state "3". The above sequence therefore loads the new Pi value into the Pi maximum value storage location and generates a new Pi maximum value.

The above sequence then determines whether Pi is greater than Pz. Pi is in state “3” or state “4”
Assuming that it is, since Pi is larger than Pz,
The above sequence returns to the process where the P waveform is converted to a new Pi increment, and this sequence is repeated. Thus, Pi increments are constantly added to the stored Pi total during state "2" as well as in state "3" and state "4".

When the P waveform reaches its peak, the maximum Pi occurs during the conversion from state “3” to state “4”, and each next Pi
is typically smaller than the maximum Pi value held in storage at the peak. Therefore, the sequence branches to a detour to avoid inputting small Pi values into the Pi maximum value storage location. After switching from state "4" to state "5", the new Pi value is no longer greater than Pi, and the above sequence also confirms the end of the integration, and the Pi value can then be counted as desired. . The counted Pi sum is stored for sliding processing in a location representing the peak-to-peak (P-P) amount of each blood pressure pulse.

Finally, the value stored in the Pi maximum storage location is halved and this value is stored in the location indicating the limit value to become the storage limit value for subsequent pulses. This limit value increases with increasing final Pi max amount within each pulse and similarly decreases with decreasing final Pi max amount.

The setting state of the memory limit value is reset at the end of the current sequence or at the beginning of the next sequence of blood pressure pulses.

For example, due to an artifact, a specific state "2" does not reach the limit level (state "3"), and Pz
It should be noted that the setting state of the storage limit value continues to be in the reset state if it falls below.
Pi increments continuously within the time interval of this state “2”.
Although it is added to the Pi total, the setting state of the memory limit value is not yet set, and the Pi waveform is constantly below Piz.
While clearing the summation storage element, the above sequence reverts to the "Pi sum clear" process, so these increments are not later treated as valid values.

FIG. 8 shows a modified embodiment of the invention, which is suitable for storing a larger portion of the first-order Pi increments of the P waveform. This system reads Pi data with zero abnormality stored in memory,
By outputting to an integrator circuit such as the accumulator 66 of FIG. 4, P waveform integration over appropriate time intervals is performed, and the subsequent processing is similar to that described in FIG. Common elements between the embodiment shown in FIG. 4 and the embodiment shown in FIG. 8 and elements not directly relevant to the understanding of the latter are omitted in FIG. Further, for the same functional elements in the embodiments of FIGS. 4 and 8, the same reference numerals are used in FIG. 8.

Pi increment 50 is sent to the forward input terminal of memory 569, which consists of a last-in-first-out (LIFO) storage element where the most recent input signal is output first. The memory 569 has a word length in “m” units, and this “m”
corresponds to the number of Pi increments that occur during the systolic rise signal time with the maximum expected duration (ie, the slowest expected pulse rate). This number may include approximately 20 to 25 Pi increments. The zero-crossing-limit detection circuit in Figure 4 assumes that the limit level has been exceeded.
It operates to drive the output of the bistable element 86 to 1 the moment the waveform becomes equal to or less than zero. The conversion of said output of bistable element 86 to 1 is sent from lead 90 to the input terminal of a pulse generator 564 which is connected to lead 5.
A pulse output is provided through 65 to the enable input terminal of LIFO drive circuit 566. LIFO drive circuit 566 receives input from timing control circuit 36' via conductor 567 and provides reverse drive pulses to LIFO memory 569 upon enable operation.
The Pi data stored in LIFO memory 569 is
The reverse drive pulse causes the signals to be read out onto conductor 64' in the reverse order of input. Timing control circuit 3
6' provides a drive pulse to the entire contents of memory 569 with sufficient speed to read out the contents of the new Pi increment stored as a continuous input signal in the reverse direction.
However, if the speed of the drive pulses from timing control circuit 36' is slow, it is necessary to inhibit input of new Pi increments 50 to memory 569 during LIFO read operations.

produces a positive transformation of the output of bistable element 86
The Pi increment itself is not stored in memory 569, and the final Pi increment entered is a small positive value.
Thus, when the contents of memory 569 are read in a backward direction, the data appearing on conductor 64' will start at a small positive value and gradually increase in the positive direction, and then the data will increase in the opposite direction to the order in which the data was originally stored. decreases towards the zero standard. Conductor 64' has one input terminal of comparator 560, the other input terminal of comparator 560 being Pz
A conducting wire 60 represents a reference. Comparator 560 is
It goes from zero to one when the data on conductor 64' becomes equal to or less than the Pz reference. The output of comparator 560 is connected to pulse generator 56 via conductor 561.
2, and the comparator 560
When it changes to , an output pulse is generated. The output of pulse generator 562 is sent via conductor 563 to the disable input of LIFO drive circuit 566;
At this point, application of reverse drive pulses to LIFO memory 569 is prohibited. In this way, conductor 6
The Pi data appearing on 4' is time-reversed zero-plus data representing the effective systolic rise signal of the blood pressure pulse.

The data on conductor 64' is transferred to accumulator 66.
and added as described above. The output of accumulator 66 is sent to the input terminal of gate 78 on conductor 68'. A control pulse applied to gate 78 causes the sum accumulated in accumulator 66 to be passed through gate 78 from its output to conductor 82 and then to the processing circuit ( (Drawing omitted).

[Brief explanation of the drawing]

FIG. 1 shows a partial block diagram of a conventional electronic circuit for measuring peak-to-peak pulsating pressure within a blood vessel. FIG. 2 shows a partial block diagram of an electronic circuit for measuring the peak-to-peak magnitude of this pulsating pressure in accordance with the present invention. FIG. 3A shows a time chart of a representative oscillatory pulse of pulsatile pressure within a blood vessel. 3rd B
The figure shows a time chart of the time derivative of the waveform of FIG. 3A. FIG. 3C shows a time chart of control signals by the systolic blood pressure measuring device of the present invention. Third
Figure D shows a time chart of the waveform integral value of Figure 3B. FIG. 4 shows a block diagram of an embodiment of the systolic blood pressure measuring device according to the present invention.
FIG. 5 shows a graph of the gain versus frequency characteristic of a differentiator network used in one preferred embodiment of the present invention. 6th
FIG. 3B is an enlarged view of the time derivative waveform of FIG. 3B showing the effective limit level and the timing of various control states relative to the limit level according to the embodiment of FIG.
FIG. 6B shows a control state diagram according to the embodiments of FIGS. 6A and 4. FIG. 7 shows a flowchart (decision tree) between successive slides of the control sequence used in the embodiment of FIG. 8th
The figure shows a simplified block diagram supplementing FIG. Figure 9 shows a limit detector that measures the peak-to-peak amplitude of the pulsatile blood pressure pulse and the derivative waveform “above zero”.
FIG. 2 is a block diagram illustrating a zero-passing detector for identifying crossings. 19... Pump, 20... Solenoid control device, 23... Pressure transducer, 25... Multiplexer, 27... Normally closed switch, 28...
Differential circuit network, 30...Amplifier, 33...AD converter, 34...Clock, 36...Timing control device, 39...Averaging device, 40, 42, 47, 5
2, 78, 96...gate, 43, 54, 56,
58, 95...Comparator, 53...Storage device, 66
...Acumulator, 63...Contractor comparator,
69...Pump switching device, 72...Bistable element,
74...delay element, 75...reader, 81,
83...Storage device (calculation-averaging device), 86...
Bistable element, 92, 94...Pulse generator, 98
...Storage device, 100 ...Filter circuit network, 10
4...Peak-peak detector, 122...2 divider, 126...Storage device, 200...Differential circuit network, 202...Amplifier, 204...Integrator, 20
5... Sample hold circuit, 560... Comparator, 562, 564... Pulse generator, 566...
...LIFO drive device, 569...LIFO memory, 90
7... Zero passing detector, 912... Limit detector,
914...Peak peak detector.

Claims (1)

[Scope of Claims] 1. A transducer that is connected to a cuff that applies variable pressure adjacent to a blood vessel of a subject and that converts blood pressure pulses in the blood vessel into an electrical signal; a differentiator network for converting the time derivative signal into a time derivative signal; an accumulator for integrating the time derivative signal to produce an integral signal proportional to the amplitude of each blood pressure pulse; a storage signal for storing the maximum value; a maximum value comparator that sequentially compares the integral signal with the next consecutive integral signal and generates an output for storing the maximum value in the storage device; and converting the maximum value to a predetermined fractional value. a dividing device for counting, and a systolic comparison that compares said fractional value with said integral signal at a cuff pressure greater than said cuff pressure when said maximum value is reached and produces an output when said integral signal is substantially equal to said fractional value. A reading device that receives the output of the systolic comparator and the electric signal from the transducer and reads out the applied pressure as the systolic blood pressure of the subject. measuring device. 2 The first claim that the above fractional value is 1/2
The systolic blood pressure measuring device described in Section 1. 3. The systolic blood pressure measuring device according to claim 1, wherein the accumulator is reset to an initial state in response to the level of the time derivative signal. 4. The systolic blood pressure measuring device according to claim 2, wherein the accumulator performs integration when the time derivative signal is above a reference level, and produces an output through a gate when it is below the reference level. 5. The systolic blood pressure measuring device according to claim 3, wherein the accumulator is reset by a delayed signal when the time derivative signal is below a reference level. 6. The systolic blood pressure measuring device according to claim 3, wherein the reference level is 0 and the time derivative signal is greater than 0 during systolic rise. 7. The systolic blood pressure measuring device according to claim 5, wherein the differentiating network converts the electrical signal of the transducer into a first time derivative or a second time derivative. 8. A transducer connected to a cuff that applies variable pressure adjacent to a blood vessel of the subject and converting blood pressure pulses in the blood vessel into an electrical signal; and converting the electrical signal of the transducer into a time derivative signal. an accumulator that integrates the time derivative signal to produce an integral signal proportional to the amplitude of each blood pressure pulse; and a storage device that stores the maximum value of the integral signal while varying the cuff pressure. , a maximum value comparator that sequentially compares the integral signal with the next consecutive integral signal and generates an output that stores the maximum value in the storage device; a dividing device that counts the maximum value into a predetermined fractional value; a systolic comparator that compares the fractional value with the integral signal at a cuff pressure greater than the cuff pressure at which the value is reached, and produces an output when the integral signal is approximately equal to the fractional value; a readout device that receives the output of the transducer and the electrical signal from the transducer and reads out the applied pressure as the systolic blood pressure of the subject; and compares the time derivative signal with a limit level signal, and A systolic blood pressure measuring device comprising: an effective logic circuit that supplies the integrated signal of the accumulator to the storage circuit when the signal is at a limit level or higher. 9. The systolic blood pressure measuring device according to claim 8, wherein the limit level signal is given as a fractional value of the preceding time derivative signal. 10. The systolic blood pressure measuring device according to claim 9, wherein the limit level signal is proportional only to the amount by which the preceding time derivative signal exceeds a reference level. 11. The systolic blood pressure measuring device according to claim 10, wherein the limit level signal is 1/2 of the amount by which the immediately preceding time derivative signal exceeds the reference level.