JPH06261871A - Processor for data on circulation action measurement - Google Patents

Abstract

PURPOSE:To provide the processor cable of exactly making circulation action measurement. CONSTITUTION:A blood pressure signal 28 is filtered by a low-pass filter having about several Hz cut-off frequency. A waveform 30 and a waveform 32 are thereby obtd. The max. value of the blood pressure signal 28 in the rising period gamma1 of this waveform 30 is defined as SBP. The min. value of the blood pressure signal 28 in the falling period delta1 of the waveform 32 is defined as DBP. The number of pulsations is computed in accordance with the interval between the DBPs.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a circulatory dynamics measurement data processing device, and more particularly to improvement of analysis accuracy.

[0002]

2. Description of the Related Art In order to know the influence of drug administration, blood pressure (systolic blood pressure, average blood pressure, diastolic blood pressure), pulse rate, etc. due to drug administration are measured. For example, a change in blood pressure in a blood vessel is measured by a pressure sensor, and based on this, systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse rate, etc. are calculated.

FIG. 20 shows a waveform obtained by measuring the blood pressure in the blood vessel of a rat. In this example, one second is 3 seconds. The maximum blood pressure value is SBP in these 3 seconds.
And the minimum blood pressure is DBP. Further, the pulse rate is calculated by measuring the number of points where the waveform crosses the threshold value Th and halving the number. The SBP, DBP and pulse rate thus obtained are stored in the memory. Similarly, SBP, D for the next one measurement time
Store BP and pulse rate in memory. By repeating this, changes in SBP, DBP and pulse rate after drug administration can be known.

[0004]

However, the above-described conventional measurement has the following problems. Figure 2
Fig. 1 shows a diagram in which changes in blood pressure of rats were measured. In the figure, the reason that the waveform is disturbed at the α portion is that the rat moved. However, the conventional measurement method has a problem that the blood pressure in this portion is recognized as SBP or DBP. In addition, there is a problem that the pulse rate is measured more than the original pulse rate because of the α portion. On the other hand, as shown in the β part, when the blood pressure rises as a whole, the waveform does not cross the threshold value Th and there is a problem that the pulse rate is measured low. That is, there is a possibility that accurate measurement cannot be performed.

An object of the present invention is to solve the above problems and to provide a periodic data processing device capable of performing accurate measurement.

[0006]

According to another aspect of the present invention, there is provided a circulatory dynamics measurement data processing apparatus which measures a circulatory dynamics measurement parameter having periodicity and converts the circulatory dynamics measurement parameter into an electric signal, and receives an electric signal from the measurement sensor. A recognition region that determines at least one of a falling period and an ascending period of the output based on the outputs of the periodicity detection filter means and the periodicity detection filter means that pass only a component of a frequency near the frequency corresponding to the periodicity. It is characterized in that the determining means and the representative value extracting means for extracting the representative value of each cycle based on the electric signal from the measurement sensor in each of the falling period and the rising period determined by the recognition area determining means.

The circulatory dynamics measurement data processing apparatus according to claim 2 removes the unnecessary representative value by comparing the representative value extracted by the representative value extracting means with the representative value of another adjacent cycle. It has a feature.

The circulatory dynamics measurement data processing device according to claim 3 is characterized in that the pulse rate or the heart rate is calculated based on the time interval from the previous representative value to the present representative value. In the hemodynamic measurement data processing device according to claim 4, the representative values for calculating the pulse rate are diastolic blood pressure and systolic blood pressure, and the representative values for calculating the heart rate are left ventricular end diastolic pressure, It is characterized by left ventricular systolic pressure.

According to another aspect of the circulatory dynamics measurement data processing device of the present invention, the measurement sensor is a blood pressure measurement sensor for measuring blood pressure, and the representative value extraction means is a blood pressure measurement sensor in the falling period determined by the recognition area determination means. Based on the blood pressure signal from, based on the blood pressure signal from the blood pressure measurement sensor in the ascending period determined by the diastolic blood pressure calculating means for calculating the diastolic blood pressure, or the recognition area determining means,
At least one of the systolic blood pressure calculating means for calculating the systolic blood pressure is provided.

According to a sixth aspect of the present invention, in the circulatory dynamics data processing device, the periodicity detection filter means is provided with a falling period filter having a transmission frequency characteristic close to the frequency near the minimum point of the blood pressure signal and a frequency near the maximum point of the blood pressure signal. A rising period filter having a transmission frequency characteristic close to that of the recognition region determining means, and the recognition area determining means is divided into falling period determining means and rising period determining means, and the falling period determining means determines the falling period based on the output of the falling period filter. And the rising period determining means determines the rising period based on the output of the rising period filter.

According to a seventh aspect of the present invention, in the circulatory dynamics measurement data processing device, the measurement sensor is a pressure measurement sensor for measuring the left ventricular pressure, and the representative value extracting means differentiates the left ventricular pressure signal to differentiate the left ventricular pressure. A left ventricular end-diastolic pressure is calculated based on the left ventricular pressure signal at the time when the left ventricular pressure differential signal is output in the falling period determined by the left ventricular pressure differential means that outputs a signal and the recognition area determination means. At least one of left ventricular systolic pressure calculation means for calculating left ventricular systolic pressure based on the left ventricular pressure signal during the ascending period determined by the ventricular end-diastolic pressure calculation means or the recognition area determination means. It is characterized by that.

According to another aspect of the circulatory dynamics measurement data processing device of the present invention, the periodicity detection filter means includes a falling period filter having a transmission frequency characteristic close to a frequency near a local minimum point of the left ventricular pressure signal and the left ventricular pressure signal. It is configured by a rising period filter having a transmission frequency characteristic close to the frequency near the maximum point, and the recognition region determining means is divided into falling period determining means and rising period determining means, and the falling period determining means is
The falling period is determined based on the output of the falling period filter, and the rising period determination means is configured to determine the rising period based on the output of the rising period filter.

In the circulatory dynamics measurement data processing device of claim 9, the measurement sensor is a blood flow measuring sensor for measuring the blood flow, and the representative value extracting means differentiates the blood flow signal to obtain the blood flow differential signal. Output blood flow differentiating means, and
A blood flow volume calculation means immediately before rising, which calculates a blood flow volume immediately before rising based on the blood flow volume signal at the time when the blood flow volume differential signal is output in the falling period determined by the recognition area determination means,
Alternatively, at least one of the maximum blood flow rate calculating means for calculating the maximum blood flow rate based on the blood flow rate signal in the rising period determined by the recognition area determining means is provided.

According to a tenth aspect of the present invention, there is provided a blood flow velocity measurement sensor for measuring blood flow velocity by the measuring sensor, wherein the representative value extracting means differentiates the blood flow velocity signal to obtain a blood flow velocity signal. The blood flow velocity differentiating means for outputting the velocity differential signal and the blood flow velocity signal immediately before rising are calculated based on the blood flow velocity signal at the time when the blood flow velocity differential signal is output within the falling period determined by the recognition region determining means. At least one of maximum blood flow velocity calculation means for calculating the maximum blood flow velocity based on the blood flow velocity signal in the rise period determined by the recognition area determination means. It is characterized by being.

According to another aspect of the circulatory dynamics measurement data processing apparatus, the periodicity detecting filter means is a falling period filter having a transmission frequency characteristic close to a frequency near a local minimum point of a blood flow rate signal or a blood flow velocity signal, and blood. It is configured by a rising period filter having a transmission frequency characteristic close to the frequency near the maximum point of the flow rate signal or the blood flow velocity signal, and the recognition area determining means is divided into the falling period determining means and the rising period determining means, and the falling period determining means. Determines the falling period based on the output of the falling period filter, and the rising period determining means,
The feature is that the rising period is determined based on the output of the rising period filter.

According to another aspect of the circulatory dynamics measurement data processing apparatus of the present invention, the circulatory dynamics measurement data arranged in time series is inputted, and the processed data is outputted at a desired sampling interval via a low frequency pass filter means. A dynamic measurement data processing device, wherein the pass frequency of the low frequency pass filter means is changed according to the sampling frequency corresponding to the sampling interval, and the pass frequency is substantially the same as or slightly higher than the sampling frequency. It is characterized by

[0017]

In the circulatory dynamics measurement data processing apparatus according to the first aspect, the periodicity detecting filter means extracts the periodic component, determines the falling period or the rising period based on this, and extracts the representative value in this period. I am trying to do it. Therefore, the representative value in each period can be accurately extracted by excluding data having an abnormal cycle.

In the circulatory dynamics measurement data processing device of the second aspect, unnecessary representative values are removed by comparing the extracted representative values with the representative values of other adjacent cycles. Therefore, abnormal data can be excluded.

In the circulatory dynamics measurement data processing device of the third and fourth aspects, the pulse rate or heart rate is calculated based on the time interval from the previous representative value to the present representative value.
Therefore, an accurate pulse rate or heart rate can be obtained.

In the circulatory dynamics measurement data processing device of the fifth aspect, the periodicity detecting filter means extracts the periodic component, determines the falling period or the rising period based on this, and based on the blood pressure signal in the falling period. Then, the diastolic blood pressure is calculated, or the systolic blood pressure is calculated based on the blood pressure signal in the rising period. Therefore, it is possible to accurately calculate the diastolic blood pressure or the systolic blood pressure while excluding the blood pressure signal having an abnormal cycle.

In the circulatory dynamics measurement data processing device of claims 6, 8 and 10, a periodicity detection filter is constituted by the falling period filter and the rising period filter, and the falling period is determined by the output of the falling period filter. The rising period is determined by the output of the rising period filter. Therefore, the falling period and the rising period can be determined more accurately.

According to another aspect of the circulatory dynamics measurement data processing apparatus of the present invention, the periodicity detection filter means extracts the periodical component, determines the falling period or the rising period based on this, and differentiates the left ventricular pressure signal. To obtain the left ventricular end-diastolic pressure based on the left ventricular intra-diastolic pressure signal at the time of the differential signal output during the falling period, or to calculate the left ventricular systolic internal pressure based on the left ventricular intra-ventricular pressure signal during the rising period. I am trying. Therefore, it is possible to accurately calculate the left ventricular end diastolic pressure or the left ventricular systolic pressure while excluding the left ventricular pressure signal having an abnormal cycle.

In the circulatory dynamics measurement data processing apparatus of claim 9, the periodicity detection filter means extracts the periodical component, determines the falling period or the rising period based on this, and determines the differential signal of the blood flow signal. Then, the blood flow volume immediately before rising is calculated based on the blood flow signal at the time of outputting the differential signal in the falling period, or the maximum blood flow is calculated based on the blood flow signal during the rising period. Therefore, the blood flow volume immediately before rising or the maximum blood flow volume can be accurately calculated while excluding the blood flow volume signal having an abnormal cycle.

In the circulatory dynamics measurement data processing device of the tenth aspect, the periodicity detection filter means extracts the periodic component, determines the falling period or the rising period based on this, and differentiates the blood flow velocity signal. To obtain the blood flow velocity immediately before rising based on the blood flow velocity signal at the time of output of the differential signal within the falling period, or to calculate the maximum blood flow velocity based on the blood flow velocity signal during the rising period. I have to. Therefore, the blood flow velocity immediately before rising or the maximum blood flow velocity can be accurately calculated while eliminating the blood flow velocity signal having an abnormal cycle.

In the circulatory dynamics measurement data processing apparatus of claim 12, the pass frequency of the low frequency pass filter means is changed according to the sampling frequency corresponding to the sampling interval, and the pass frequency is slightly higher than the sampling frequency. I am trying. Therefore, appropriate data can be obtained according to the sampling interval.

[0026]

【Example】

—Basic Configuration— FIG. 2 is a conceptual diagram of a circulation dynamics measurement data processing device according to an embodiment of the present invention. Blood pressure measuring sensors (not shown) are attached to the animals A _{1 to} A ₈ as the specimens. The output of each sensor is input to the 8-channel A / D converter 2 through the amplification amplifier 3 and converted into a digital signal. The converted data is input to the computer 4 as a blood pressure signal. The computer 4 calculates and stores representative values such as systolic blood pressure and diastolic blood pressure based on this blood pressure signal.

FIG. 3 shows the blood pressure measuring sensor used in this embodiment. The conduit 26 provided at the tip is inserted into the blood vessel of the sample. A thin tube 28 filled with saline is connected to the conduit 26, and the blood pressure measurement sensor 6 is further connected to the conduit 26. Therefore, the change in blood pressure in the blood vessel is guided to the blood pressure measurement sensor 6 through the narrow tube 28. As a result, the blood pressure measurement sensor 6 outputs an analog signal according to the blood pressure.

FIG. 4 is a block diagram of the blood pressure data processing device shown in FIG. The outputs from the blood pressure measurement sensors 6 _{1 to} 6 ₈ attached to each sample are amplified by the amplification amplifier 3 and are
It is given to the D converter 2. In response to a command from the CPU 12, one digitally converted blood pressure signal is selected and given to the input port 10. The input port 10 is connected to the bus line 24. In the bus line 24,
The CPU 12, ROM 14, RAM 16, CRT 18, recorder 20, and hard disk 22 are connected. C
The PU 12 takes in a blood pressure signal from the input port 10 and performs processing according to a program stored in the ROM 14.

-Processing of Blood Pressure Signal- FIG. 5 shows a flowchart of a program stored in the ROM 14. First, the CPU 12 sequentially takes in the outputs of the blood pressure measurement sensors 6 _{1 to} 6 ₈ as digital signals in step S ₀ . Here, for convenience of description, only the output of the blood pressure measurement sensor 6 ₁ will be described, but the same processing is performed for the other blood pressure measurement sensors 6 _{2 to} 6 ₈ .

The captured digital blood pressure signal is R
It will be stored in AM16. When this blood pressure signal is shown as a waveform, it becomes like 28 in FIG. The CPU 12 digitally filters this blood pressure signal (step S ₁ ). A flowchart of the filtering process is shown in FIG. First, the CPU 12 filters the blood pressure signal for noise removal (step S).
₆ ). For example, 100 Hz when the sample is a rat
The above frequency components are cut.

Next, the low-pass filtering process for the rising period is performed (step S ₇ ). The transmission frequency of this low-pass filter is near the maximum point of the blood pressure signal (γ in FIG. 1A).
Select to pass the frequency component of the part. For example, when the sample is a rat, frequency components of 8 Hz or less are transmitted. As a result, the low frequency component 30 as shown in FIG. 1A is obtained. The obtained low-frequency component 30 has a delay time due to the filtering process.

Next, the low-pass filtering process for the falling period is performed (step S ₈ ). The transmission frequency of this low-pass filter is near the minimum point of the blood pressure signal (δ in FIG. 1B).
The frequency component of the part is transmitted. Especially, since the time interval from the minimum point to the maximum point is small, the delay time of the filtering process cannot be made too large,
The cutoff frequency cannot be too low. For example, when the sample is a rat, the frequency component of 15 Hz or less is transmitted.

The blood pressure data filtered for noise removal, the filtered data for the rising period, and the filtered data for the falling period are stored in the RAM 16, respectively.

As described above, in this embodiment, each filtering process is performed by the digital filter (in this embodiment, the second-order IIR Butterworth type filter process is performed). However, an analog filter may be used. .

Returning to FIG. 5, in step S ₂ , it is detected whether or not the minimum point X of the data subjected to the filtering process for the falling period appears. If the minimum point X does not appear, the process returns to step S ₀ to collect the next blood pressure data.

When the minimum point X appears, the CPU 12 makes RA
M16 on the basis of the stored filtering data to extract the representative value (Step S ₃ in FIG. 5). FIG. 7 shows a flowchart of the representative value extraction processing. First, CPU1
2 calculates the data rising period γ ₁ of the low frequency component (see 30 in FIG. 1A) that has been subjected to the rising period filter processing. Next, the maximum value of the blood pressure data within this period γ ₁ is obtained, and this is set as the systolic blood pressure (SBP) (step S ₉ ). Similarly, the data falling period δ of the low frequency component (see 32 in FIG. 1B) that has been subjected to the falling period filtering process.
₁ is calculated, and obtains the minimum value of blood pressure data in the falling period δ _1, which diastolic and blood pressure (DBP) (Step S _10).

Next, the blood pressure data from the previous DBP to the current DBP period are integrated and averaged to obtain a mean blood pressure (MBP).
Is calculated (step S ₁₁ ). Furthermore, from DBP to SB
It calculates a time until the current DBP from the time and the last DBP to P, determining the pulse rate by multiplying the 60 to its inverse (step S _12). Then, it calculates the difference between the previous DBP and current DBP (step S _13). Additionally, to compute the absolute value of the difference of the waveform to the current DBP from the previous DBP, this is the blood pressure waveform displacement value (step S _14). Next, the blood pressure gradient (blood pressure gradient) θ of the previous DBP and the current DBP is calculated (step S ₁₅ ). The representative value obtained as described above is stored in the hard disk 22, as shown in FIG. Further, as shown in FIG. 10, the CPU 12 can display the extraction result of the representative value on the CRT 18.

When the extraction and storage of the representative value are completed, the rejection test is started (step S ₄ ). FIG. 8 shows a flowchart of the rejection test process. In this embodiment, it is performed rejection test for the representative value of all calculated in step S3 (step S ₁₅ ~S _21). In step S _15, it is doing a rejection test of SBP. It is determined whether or not the SBP subject to the rejection test has an abnormal value when viewed from the five SBPs before and after. In this embodiment, the average value and the variance value of the five SBPs before and after are calculated, and when the variance value is 10% or more of the average value, the Smirno of 1% is calculated.
v It is judged by performing a rejection test. In step S ₁₆ following, the rejection test in the same manner for the other representative value.

The rejection test is performed as described above, and if any one of the representative values is determined to be an abnormal value, all the one set of representative values is rejected. That is, it is erased from the hard disk 22 and is not used in the subsequent processing. In this way, abnormal data can be eliminated. For example, when abnormal data as shown by α in FIG. 21 is obtained as a blood pressure signal, the representative value of this portion is far from the representative value of other normal portions. Therefore, the above rejection test eliminates the data in the portion of α, which enables accurate data processing.

Next, in step S ₅ , an average value or the like for each representative time is calculated for each representative value based on only the normal data from which the abnormal data is excluded. After that, the process returns to step S ₀ again to continue the processing.

As described above, in this embodiment, the falling period and the rising period are determined by the digital filter,
The representative value is calculated based on the blood pressure data within this period. Therefore, the representative value can be accurately extracted. Also, the blood pressure data near the frequency of the original blood pressure change is transmitted by the digital filter. Further, since the pulse rate is calculated based on the interval between DBPs, the pulse rate can be calculated accurately. Further, since the rejection test is performed, abnormal data can be easily excluded, and accurate measurement can be performed.

In the above description, one blood pressure measuring sensor 6 is used.
_Although the processing of the blood pressure signal from _{1 has} been described, the CPU 1
2 also takes in blood pressure signals from the other blood pressure measurement sensors 6 _{2 to} 6 ₈ by time division and performs the same processing. Thus the representative value calculated by is stored in the hard disk 22 for each sensor ₆₁ through _8. Also, as shown in FIG.
Representative values corresponding to the _eight sensors 6 _{1 to} 6 ₈ are displayed on the CRT 18
Can be displayed at the same time. In the figure, the screen is 8
Is split, part of Ch1 is pulse rate is representative value based on pressure signals from the blood pressure measuring sensor 6 ₁ (HR), SBP,
The graph of the time change of MBP and DBP is shown.
Ch2~Ch8 are respectively displays a value representative of the blood pressure signal from the blood pressure measuring sensor 6 _{2-6 8.} In another embodiment, the number of channels of the A / D converter 2 is increased / decreased,
The number of specimens that can be measured simultaneously can be selected.

-Processing of Left Ventricular Pressure Signal- In the above, the processing of the blood pressure signal in which the blood pressure in the blood vessel is measured by the sensor has been described. Similar processing can be performed for the left ventricular pressure. As the sensor for measuring the pressure in the left ventricle, the same sensor as that shown in FIG. 3 can be used. Therefore, the hardware configuration is as shown in FIG.

A program for processing the left ventricular pressure signal stored in the ROM 14 is shown in the flow chart of FIG. A / D conversion in step S ₃₀ is the same as the processing in the case of the blood pressure signal of FIG. FIG. 16A graphically illustrates the left ventricular pressure signal 36. The digitally converted left ventricular pressure signal is sequentially stored in the RAM 16. Then, CPU 12 performs a filtering process (step S _31). Figure 1 shows the flowchart of the filtering process.
3 shows. First, in step _S36a , filtering processing for noise removal is performed. When the sample is a rat, the frequency component of 96 Hz or higher is cut, and when the sample is a dog, the frequency component of 40 Hz or higher is cut. Noise-cut left ventricular pressure data is RA
It is stored in M16.

Then, filtering for periodicity detection is performed (step _S37a ). The blood pressure signal was asymmetric for rising and falling (that is, different in frequency), but the left ventricular pressure signal is almost symmetrical for rising and falling, as shown in FIG. 16A. Therefore, the filtering for the rising period detection and the filtering for the falling period detection can be the same. In this example, when the sample is a rat, the component of 6 Hz or more is cut, and when the sample is a dog, the component of 4 Hz or more is cut. The data subjected to the filtering process of periodicity detection is stored in the RAM 16.

The filtered data for periodicity detection is shown as a waveform at 38 in FIG. 16A. This waveform is delayed by a predetermined time with respect to the left ventricular pressure signal waveform 36 by the filtering process.

Returning to FIG. 12, in step S ₃₂ ,
The maximum point Y of the data filtered by the periodicity detection
Is detected. If the local maximum point Y has not our table to collect the next left ventricular pressure data returns to step S _30.

When the maximum point Y appears, the CPU 12 determines RA
M16 on the basis of the stored filtering data to extract the representative value (step S ₃₃ in FIG. 12). 14
The flowchart of the representative value extraction processing is shown in FIG. First, C
The PU 12 calculates the data falling period δ ₂ and the data rising period γ ₂ of the low-frequency component (see 38 in FIG. 16A) that has been subjected to the representative value extracting filter process. Next, find the maximum value of the left ventricular pressure data within the data rise period γ ₂ ,
This is the left ventricular systolic pressure (LVSP) (step S ₃₆ ).

Next, the CPU 12 first-order differentiates the left ventricular pressure data stored in the RAM 16 (step S ₃₇ ). The waveform of the differentiated data is shown in Fig. 1.
6B. Next, the minimum value of this differential data within the total period of the period δ ₂ and the period γ ₂ is obtained, and this is set as the left ventricular pressure maximum expansion velocity (Min dp / dt) (step S ₃₈ ). Further, the left ventricular pressure data at the rising point Z (see FIG. 16B) of the differential data is obtained and set as the left ventricular end diastolic pressure (LVEDP) (step S ₃₉ ). Next, the maximum value of this differential data within the total period of the period δ ₂ and the period γ ₂ is obtained, and this is set as the left ventricular pressure maximum contraction rate (Max dp / dt) (step S ₄₀ ). In step S ₄₁ , LVEDP to LV
The time to SP is calculated, and the time between the previous LVEDP and the current LVEDP (LVEDP-LVED
P interval) is calculated. Furthermore, LVEDP-LVEDP
The heart rate is calculated based on the interval (step S ₄₁ ). The representative value obtained as described above is the hard disk 22.
Memorized in.

[0050] Next, the rejection test for the obtained representative value (step S ₃₄ in FIG. 12). A flowchart of the rejection test is shown in FIG. In this embodiment, as in the case of blood pressure data (see FIG. 8), the Smirnov rejection test is performed based on the five data before and after. If any one of the representative values is determined to be an abnormal value,
Discard all the representative values of the set. That is, it is erased from the hard disk 22 and is not used in the subsequent processing.

Next, in step S _35, based on only the normal data abnormal data is eliminated, the average value or the like for each predetermined time is calculated for each representative value. Then, to continue the process returns to step S ₃₀ again.

-Processing of Blood Flow Volume Signal and Blood Flow Velocity Signal- Similar to the blood pressure signal and left ventricular pressure signal, the blood flow volume signal and blood flow velocity signal can be processed.
However, a blood flow velocity sensor is required to obtain the blood flow rate signal and the blood flow velocity signal. For example, pulse Doppler blood flow meter (such as PD-20 manufactured by Prime Tech Co., Ltd.)
Should be used. Therefore, the blood pressure measurement sensor 6 of FIG.
It may be connected to the blood flow velocity sensors instead of the _{1-6 8.}

FIG. 17A shows a blood flow rate signal (or blood flow velocity signal) 42 from the blood flow velocity measuring sensor. An output waveform of the periodicity detection filtering process is shown by 44.
Further, FIG. 17B shows a blood flow rate signal (or blood flow velocity signal).
42 is a differential waveform of 42. The calculation of the rising period γ ₃ , the falling period δ ₃ , the blood flow rate immediately before rising or the blood flow velocity immediately before rising (DBF), the maximum blood flow rate or the maximum blood flow velocity (SBF), and the like are the same as in the case of left ventricular pressure data. . Note that the average blood flow volume or the average blood flow velocity (MBF) is the previous DBF, respectively.
To the current DBF are calculated by averaging blood flow rates or blood flow velocities.

-Output of Data- The representative value is stored in the hard disk 22 by the above processing. This representative value can be confirmed by outputting it to the CRT 18 or the recorder 20. FIG. 18A shows an example in which a temporal change in pulse rate (HR) systolic blood pressure (SBP) mean blood pressure (MBP) diastolic blood pressure (DBP) obtained based on the blood pressure signal is output to the recorder 20. The horizontal axis represents time, and one scale is 1
It's time. The drug was administered at the time indicated by the broken line 50.

Since the output graph of FIG. 18A shows even minute changes, it is not sufficient for observing the entire course of a long time. On the other hand, when it is desired to observe changes immediately after drug administration, it is preferable that fine changes can be perfectly expressed. Therefore, in this embodiment, an appropriate output graph can be obtained by performing the following filtering process.

FIG. 19 shows a flowchart of the output filtering process. First, in step S ₅₀ , the representative value desired to be input and output of the collection time interval t _p is selected. Here, the collection time interval t _p refers to a desired minimum time interval when output is displayed. If the collection time interval t _p is large, minute changes are removed, and the overall tendency is easily captured, and if the collection time interval t _p is small, even minute changes can be accurately represented.
For example, in the example of FIG. 18A, if the collection time interval t _{p is set} to about 10 minutes, minute changes can be removed and the overall tendency can be clearly captured.

Next, the CPU 12 calculates the cutoff frequency fc for the filtering process according to the following equation based on the input collection time interval t _p (step S ₅₁ ).

[0058]

[Number 1] fc = 1 / t _p ········· (1) For example, if the collection time interval _{t p} is 10 minutes, the cutoff frequency fc becomes 0.0016Hz.

Next, the CPU 12 reads the representative value of interest from the hard disk 22 and calculates the cutoff frequency f.
Filtering processing is performed based on c (step S ₅₂ ). The data thus obtained is stored in the hard disk 22.

When the recorder 20 displays the data obtained by filtering the data shown in FIG. 18A (with a cutoff frequency of 0.0016 Hz), it becomes as shown in FIG. 18B. As is clear from comparison with FIG. 18A, the overall long-term trend is clearly displayed.

The cut-off frequency fc may be set to be slightly lower than that of the expression (1).

-Other Examples- In the above-mentioned examples, blood pressure data, left ventricular pressure data, and blood flow data were targeted as circulatory dynamics measurement data, but other circulatory dynamics measurement data may be targeted.

Further, in the above-mentioned embodiment, each means is constituted by using the CPU, but all or part of them may be constituted by hardware logic.

[0064]

According to the circulatory dynamics measurement data processing apparatus of the first aspect, the periodicity detecting filter means extracts the periodicity component, and based on this, the falling period or the rising period is determined, and the representative value in this period. I am trying to extract. Therefore, the representative value in each period can be accurately extracted by excluding data having an abnormal cycle. That is, the reliability of measurement data can be improved. In the circulatory dynamics measurement data processing device of the second aspect, unnecessary representative values are removed by comparing the extracted representative values with the representative values of other adjacent cycles. Therefore, abnormal data can be excluded.

In the circulatory dynamics measurement data processing device of the third and fourth aspects, the pulse rate or the heart rate is calculated based on the time interval from the previous representative value to the present representative value.
Therefore, an accurate pulse rate or heart rate can be obtained. That is, there is no possibility of detecting an incorrect pulse rate or heart rate due to abnormal data.

In the circulatory dynamics measurement data processing device of the fifth aspect, the periodicity detecting filter means extracts the periodic component, determines the falling period or the rising period based on this, and based on the blood pressure signal in the falling period. Then, the diastolic blood pressure is calculated, or the systolic blood pressure is calculated based on the blood pressure signal in the rising period. Therefore, it is possible to accurately calculate the diastolic blood pressure or the systolic blood pressure while excluding the blood pressure signal having an abnormal cycle.

In the circulatory dynamics measurement data processing device according to claims 6, 8 and 10, the periodicity detection filter is constituted by the falling period filter and the rising period filter, and the falling period is determined by the output of the falling period filter. The rising period is determined by the output of the rising period filter. Therefore, the falling period and the rising period can be determined more accurately.

In the circulatory dynamics measurement data processing device according to the seventh aspect, the periodicity detecting filter means extracts the periodic component, determines the falling period or the rising period based on this, and differentiates the left ventricular pressure signal. To obtain the left ventricular end-diastolic pressure based on the left ventricular intra-diastolic pressure signal at the time of the differential signal output during the falling period, or to calculate the left ventricular systolic internal pressure based on the left ventricular intra-ventricular pressure signal during the rising period. I am trying. Therefore, it is possible to accurately calculate the left ventricular end diastolic pressure or the left ventricular systolic pressure while excluding the left ventricular pressure signal having an abnormal cycle.

In the circulatory dynamics measurement data processing device of the ninth aspect, the periodicity detection filter means extracts the periodic component, determines the falling period or the rising period based on this, and determines the differential signal of the blood flow signal. Then, the blood flow volume immediately before rising is calculated based on the blood flow signal at the time of outputting the differential signal in the falling period, or the maximum blood flow is calculated based on the blood flow signal during the rising period. Therefore, the blood flow volume immediately before rising or the maximum blood flow volume can be accurately calculated while excluding the blood flow volume signal having an abnormal cycle.

In the circulatory dynamics measurement data processing device of the tenth aspect, the periodicity detection filter means extracts the periodic component, determines the falling period or the rising period based on this, and differentiates the blood flow velocity signal. To obtain the blood flow velocity immediately before rising based on the blood flow velocity signal at the time of output of the differential signal within the falling period, or to calculate the maximum blood flow velocity based on the blood flow velocity signal during the rising period. I have to. Therefore, the blood flow velocity immediately before rising or the maximum blood flow velocity can be accurately calculated while eliminating the blood flow velocity signal having an abnormal cycle.

In the circulatory dynamics measurement data processing apparatus of claim 12, the pass frequency of the low frequency pass filter means is changed according to the sampling frequency corresponding to the sampling interval, and the pass frequency is slightly higher than the sampling frequency. I am trying. Therefore, appropriate data can be obtained according to the sampling interval.

[Brief description of drawings]

FIG. 1 is a diagram showing blood pressure data processing based on a circulatory dynamics measurement data processing apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing an external appearance of a circulation dynamics measurement data processing device according to an embodiment of the present invention.

3 is a diagram showing a blood pressure sensor used in the device of FIG.

FIG. 4 is a block diagram of the apparatus of FIG.

5 is a flowchart of a data processing program stored in a ROM 14. FIG.

FIG. 6 is a flowchart of filtering processing.

FIG. 7 is a flowchart of a representative value extraction process.

FIG. 8 is a flowchart of a rejection test process.

FIG. 9 is a diagram showing an example of representative values stored in a hard disk 22.

10 is a screen for representative value extraction displayed on the CRT 18. FIG.

FIG. 11 is a screen showing a transition of representative values displayed simultaneously on eight channels on the CRT.

12 is a flowchart of a data processing program stored in a ROM 14. FIG.

FIG. 13 is a flowchart of filtering processing.

FIG. 14 is a flowchart of a representative value extraction process.

FIG. 15 is a flowchart of a rejection test process.

FIG. 16 is a diagram for showing left ventricular pressure data processing based on the hemodynamic measurement data processing apparatus according to the embodiment of the present invention.

FIG. 17 is a diagram showing blood flow rate data processing or blood flow velocity data processing based on the circulatory dynamics measurement data processing apparatus according to the embodiment of the present invention.

FIG. 18 is an example in which the recorder 20 outputs the change of the representative value with time and the change of the representative value with time after the filtering process.

FIG. 19 is a flowchart showing a filtering process for output.

FIG. 20 is a graph showing measurement of conventional blood pressure data.

FIG. 21 is a graph showing measurement of conventional blood pressure data.

[Explanation of symbols]

2 ... A / D converter 6 _{1 to} 6 ₈ ... Blood pressure measurement sensor 12 ... CPU 14 ... ROM 16 ... RAM 18 ... CRT 20 ... Recorder 22 ... Hard disk

Claims (12)

[Claims] 1. A measurement sensor that measures a circulatory dynamics measurement parameter having periodicity and converts it into an electric signal, receives an electric signal from the measurement sensor, and outputs only a component of a frequency near a frequency corresponding to the periodicity. The periodicity detection filter means to be passed, the recognition region determining means for determining at least one of the falling period and the rising period of the output based on the output of the periodicity detection filter means, the falling period and the rising period determined by the recognition region determining means A circulatory dynamics measurement data processing device, comprising: a representative value extracting means for extracting a representative value of each cycle based on an electrical signal from the measurement sensor in. 2. The circulatory dynamics measurement data processing device according to claim 1, wherein unnecessary representative values are removed by comparing the representative values extracted by the representative value extracting means with representative values of other adjacent cycles. Characterized by 3. The circulatory dynamics measurement data processing device according to claim 1, wherein the pulse rate or heart rate is calculated based on the time interval from the previous representative value to the present representative value. 4. The hemodynamic measurement data processing device according to claim 3, wherein the representative values for calculating the pulse rate are diastolic blood pressure and systolic blood pressure, and the representative values for calculating the heart rate are left. It is characterized by end-diastolic pressure and left-ventricular systolic pressure. 5. The circulatory dynamics measurement data processing device according to claim 1, wherein the measurement sensor is a blood pressure measurement sensor for measuring blood pressure, and the representative value extraction means is the blood pressure in the falling period determined by the recognition area determination means. A systolic blood pressure is calculated based on the blood pressure signal from the blood pressure measurement sensor for calculating the diastolic blood pressure based on the blood pressure signal from the measurement sensor, or the blood pressure measurement sensor during the ascending period determined by the recognition region determining means. At least one of the systolic blood pressure calculating means for 6. The circulatory dynamics data processing apparatus according to claim 5, wherein the periodicity detection filter means includes a falling period filter having a transmission frequency characteristic close to a frequency near a local minimum point of the blood pressure signal and a local maximum point of the blood pressure signal. The rising region filter having a transmission frequency characteristic close to the frequency of ## EQU1 ## separates the recognition region determining means into the falling period determining means and the rising period determining means, and the falling period determining means is based on the output of the falling period filter. The circulatory dynamics data processing device, wherein the falling period is determined, and the rising period determining means determines the rising period based on the output of the rising period filter. 7. The circulatory dynamics measurement data processing device according to claim 1, wherein the measurement sensor is a pressure measurement sensor for measuring the pressure in the left ventricle, and the representative value extracting means differentiates the pressure signal in the left ventricle to obtain the left ventricle. Left ventricular end-diastolic pressure is calculated based on the left ventricular pressure differential means that outputs an internal pressure differential signal and the left ventricular pressure signal at the time when the left ventricular pressure differential signal is output within the falling period determined by the recognition area determination means. Or at least one of left ventricular systolic pressure calculating means for calculating the left ventricular systolic pressure based on the left ventricular pressure signal in the ascending period determined by the left ventricular end-diastolic pressure calculating means. What is characterized by. 8. The circulatory dynamics measurement data processing apparatus according to claim 7, wherein the periodicity detection filter means includes a falling period filter having a transmission frequency characteristic close to a frequency near a local minimum point of the left ventricular pressure signal and the left ventricular pressure. The recognition region determining means is divided into a falling period determining means and an ascending period determining means, and the falling period determining means is a falling period filter. The circulatory dynamics measurement data processing device, wherein the falling period is determined based on the output of the rising period determining means, and the rising period determining means determines the rising period based on the output of the rising period filter. 9. The circulatory dynamics measurement data processing device according to claim 1, wherein the measurement sensor is a blood flow measurement sensor for measuring blood flow, and the representative value extracting means differentiates the blood flow signal to differentiate the blood flow. Blood flow rate differentiating means for outputting a signal, and blood flow volume before rising is calculated based on the blood flow volume signal at the time when the blood flow volume differential signal is output within the falling period determined by the recognition area determining means. means,
Alternatively, at least one of maximum blood flow rate calculation means for calculating the maximum blood flow rate based on the blood flow rate signal in the rising period determined by the recognition area determination means is provided. 10. The circulatory dynamics measurement data processing device according to claim 1, wherein the measurement sensor is a blood flow velocity measurement sensor for measuring a blood flow velocity, and the representative value extracting means differentiates a blood flow velocity signal, Blood flow velocity differentiating means for outputting a blood flow velocity differential signal and a blood flow velocity immediately before rising based on the blood flow velocity signal at the time when the blood flow velocity differential signal is output within the falling period determined by the recognition area determining means. Or at least one of maximum blood flow velocity calculation means for calculating the maximum blood flow velocity based on the blood flow velocity signal in the rise period determined by the recognition area determination means. What is characterized by being 11. The circulatory dynamics measurement data processing device according to claim 9, wherein the periodicity detection filter means has a falling period filter having a transmission frequency characteristic close to a frequency near a minimum point of a blood flow signal or a blood flow velocity signal. And a rising period filter having a transmission frequency characteristic close to the frequency near the maximum point of the blood flow signal or the blood flow velocity signal, and the recognition region determining means is divided into the falling period determining means and the rising period determining means. The circulatory dynamics measurement is characterized in that the determining means determines the falling period based on the output of the falling period filter, and the rising period determining means determines the rising period based on the output of the rising period filter. Data processing device. 12. A circulatory dynamics measurement data processing apparatus for inputting measurement data of circulatory dynamics arranged in time series and outputting processed data at a desired sampling interval via a low-frequency pass filter means, Circulation dynamics measurement, characterized in that the pass frequency of the low-frequency pass filter means is changed according to the sampling frequency corresponding to the sampling interval, and the pass frequency is set to a frequency substantially the same as or slightly higher than the sampling frequency. Data processing processor.