JPH05115448A - Waveform processor - Google Patents

Abstract

PURPOSE:To realize the measurement with high accuracy by eliminating a waveform component exerting an adverse influence on an object measurement from among observed waveforms. CONSTITUTION:This processor is constituted of a pulse wave measuring part 1, a parameter part 2 for calculating a pattern of a parameter of amplitude, etc., from an observed pulse wave, a holding part 3 for storing the pattern of this parameter, a pattern holding part 4 for holding a standard pattern of the parameter, a comparing part 5 for comparing the pattern of an observed parameter with that of a standard parameter, a correcting part 6 for correcting an observed pulse wave envelope based on the result of this comparison, and a blood pressure estimating part 7.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a waveform processing device used in a vibration method type electronic blood pressure monitor or the like.

[0002]

2. Description of the Related Art Generally, a vibration method electronic blood pressure monitor extracts a pulse wave component superimposed on the cuff pressure in the process of stopping the blood flow in an artery with a cuff and then slowly discharging the cuff pressure. For example, the amplitude of each pulse wave is calculated as the parameter, and a predetermined algorithm is applied to the envelope of what is arranged in correspondence with the change in cuff pressure to determine the systolic blood pressure and the optimal blood pressure.

[0003]

In the above-mentioned conventional electronic blood pressure monitor employing the vibration method, the blood pressure is determined by using the envelope curve having the pulse wave amplitude as a parameter, so that the number of pulse waves is sufficient. When it is not observed or when the pulse wave includes an abnormal waveform due to body movement or arrhythmia, this has a bad influence, and there is a problem that accurate pulse wave estimation and eventually blood pressure determination cannot be performed.

The present invention has been made by paying attention to the above-mentioned problems, and a waveform capable of realizing accurate measurement by removing a waveform component which adversely affects a target measurement from the observed waveform. It is intended to provide a processing device.

[0005]

The waveform processing apparatus of the present invention comprises a waveform observing means, a parameter calculating means for calculating a pattern of a predetermined parameter from the observed waveform, and a standard pattern of parameters. Storage means for holding, comparison means for comparing the pattern of the calculated parameter with the held standard pattern, and correcting the parameter of the observed waveform based on the comparison result of the comparison means. Compensation means.

In this waveform processing apparatus, the pattern parameter pattern of the waveform observed by the observing means and the standard parameter pattern stored and held are compared. If the two patterns do not match, the comparison result is The parameters of the observed waveform are corrected. When this waveform processing device is used, for example, as a device for obtaining the envelope of the pulse wave amplitude of an electronic sphygmomanometer, it can be corrected by comparison with the fact that a standard pulse wave envelope cannot be obtained due to body movement or arrhythmia. So
It is possible to measure blood pressure with high accuracy without being adversely affected by body movement and arrhythmia.

[0007]

The present invention will be described in more detail with reference to the following examples. FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment of the present invention. In this embodiment, the waveform processing device of the present invention is applied to a pulse wave processing device of an electronic sphygmomanometer adopting a vibration method.

The electronic blood pressure monitor of this embodiment includes a pulse wave measuring unit (observing unit) 1, a parameter calculating unit 2, a holding unit 3, a pattern holding unit 4, a comparing unit 5, a correcting unit 6, and a blood pressure. The estimation unit 7 is included. The pattern holding unit 4, the comparison unit 5, and the correction unit 6 form a parameter correction unit 8. The pulse wave measurement unit 1 includes a cuff, a pressure sensor, a pressure pump,
A well-known part such as an exhaust valve is included, and a known pulse waveform is measured in the process of pressurizing or depressurizing the cuff. The parameter calculation unit 2 calculates parameters for displaying the wave shape such as amplitude and average value of the pulse waveform measured by the pulse wave measurement unit 1. The holding unit 3 stores and holds the parameters calculated by the parameter calculation unit 2 in time series. Pattern holding unit 4
Holds the number of extreme values of the standard pulse wave envelope and the slope (maximum amplitude-average amplitude) pattern of the envelope in the form of a membership function rule, which is referred to by the matching unit of the comparing unit 5.

As shown in FIG. 2, the comparison unit 5 is composed of a data reading unit 11, a calculation unit 12 and a collation unit 13, and reads the amplitude data among the data held in the storage unit 3 to obtain a maximum value, The number of maximum / minimum values and the change before and after the maximum value (inclination of the envelope) are calculated and collated with the standard pattern of the pattern holding unit 4 to detect the presence / absence of an abnormality and the abnormal pulse wave. For the matching, a fuzzy inference engine built in the matching unit 13 is used. The result of the comparison is the time series number of abnormal data,
A flag indicating normal or abnormal is output.

The correction unit 6 receives the signal from the comparison unit 5,
If normal, nothing is done and the parameters of the holding unit 3 are sent to the blood pressure estimating unit 7 as they are. In the case of abnormality, interpolation is performed using the data before and after the abnormality data number n (see FIG. 3). The interpolation method shown in FIG. 3 is a case of linear interpolation, and the interpolation value Y _n determined by this interpolation is the interpolation value Y _n = (Y _{n + 1} −Y _n-1 ) / 2 + Y _n−1 (1) Becomes After interpolating all the parameters used for blood pressure estimation, the interpolated data is sent to the blood pressure estimating unit 7.

The blood pressure estimating unit 7 estimates blood pressure from parameters such as pulse wave amplitude. A well-known algorithm may be used as the estimation algorithm. For example, in the cuff pressure-pulse wave amplitude characteristic, the peak value of the pulse wave amplitude is 0.
The cuff pressure on the high cuff pressure side for the amplitude of 5 is estimated as the systolic blood pressure, and the cuff pressure on the low cuff pressure side for the amplitude of 0.7 for the peak value of the pulse wave amplitude is estimated as the diastolic blood pressure.

Now, the processing of the comparison section 5 will be described more specifically. Here, the extreme value is detected, and fuzzy inference is performed using the number of polar areas and each maximum value and the average of the slopes before and after it, and the normal maximum value and the abnormal maximum value are classified. If it is an envelope as shown in FIG. 4, there are three extreme values, n, t, and w, and the average slope before and after this maximum value is obtained as A−S * = (absolute value a + absolute value b) / 2. ..

However, * is the data number of the maximum value, a is the slope before the maximum value, and b is the slope after the maximum value. The pattern holding unit 4 has a minimum value P-S shown in FIG. 5, a maximum value P-B shown in FIG. 6, an average value A-S of slopes before and after the maximum value shown in FIG. 7, and FIG. Conclusion The membership function of DEC is stored, and the following rules for abnormality detection are also stored.

* If (if) P-B is large then (if it is) abnormal * if P-Sis large then abnormal * if P-B is small and A-S is small then normal * if P-S is Small and A-Sis Large then Abnormality Next, the normal / abnormality determination processing of the pulse wave envelope will be described with reference to the flows shown in FIGS. 9 and 10.

At the start of operation, the variable n (data number of amplitude arranged in time series) is first set to 0 [step ST
(Hereinafter abbreviated as ST) 1]. Next, the n + 1th amplitude A
_{The nth} amplitude A _n is subtracted from _{n + 1} to obtain S _n, and similarly, the n + 2nd amplitude A _{n + 2} to the n + 1th amplitude A
_{N + 1} is subtracted to obtain S _n (ST2), and it is determined whether S _n and S _{n + 1} have the same sign (ST3). If the signs are the same, the amplitude is at least rising or falling, so the variable n is incremented by 1 (ST4), and the processes of steps ST2 to ST4 are repeated.

Eventually, when the maximum point or the minimum point is reached, S
_{Since n} and S _{n + 1} have different signs, the determination in step ST3 is NO, and the process proceeds to step ST5.
In step ST5, it is determined whether the change in code from the change value S _n to S _{n + 1} is a change from ten to one. If so, it means that include a maximum value therebetween, and +1 counts the maximum value of the number P-B (ST6), the average of the absolute value and the absolute value of S _{n + 1} of S _n, that is, the maximum The average value of the slopes before and after the value is calculated and stored (ST7). If the sign change in ST3 is from 1 to 10, step ST5
Is NO, and in this case, the minimum value is included between them, so the minimum value number P−S is incremented by 1 (S
T8). After step ST7 or step ST8, the process proceeds to step ST9. Then, it is determined whether or not the data has ended. If it is not the final data, the variable n is incremented by 1 (ST10), the process returns to step ST2, and the processes of ST2 to ST10 are repeated until the end of the data. As a result, the number of maximum values, the average of the slopes before and after the maximum value, and the number of minimum values of the pulse wave envelope are obtained.

For all amplitude data, step S
When the processes of T2 to ST10 are completed, the determination in step ST9 becomes YES, and the process then proceeds to the abnormality determination process. First,
The variable S is set to 0 (ST11), and the held data P-
Fuzzy inference is executed by inputting S, P-B and A-S _(S) (ST12). If the determination result is abnormal (ST13), the data number is registered (ST1
4) If the determination result is normal, leave it as it is and set the variable S
Until the maximum value P-B is reached (ST15), the variable S is advanced by 1, the process returns to step ST12, and step ST1.
The processes of 2 to ST16 are repeated. The variable S is the maximum number P-
When it reaches B, the determination of normality / abnormality of the data corresponding to the maximum value number is completed, so this process is completed.

In the above embodiment, the case where the present invention is applied to the electronic sphygmomanometer has been described. However, the present invention can be applied to other various waveform processing devices.

[0019]

According to the present invention, if the pattern of the observed waveform is different from the pattern of the standard parameter, it is judged as abnormal by comparison and appropriate correction is applied, so that the accuracy of the waveform processing is improved. improves. Especially when used as an electronic sphygmomanometer, the estimation accuracy of pulse pressure is improved, and even when there are few observable pulse waves during pressurization, accurate estimation of pulse pressure is possible, etc. There are advantages.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing an internal configuration of a comparison unit that constitutes the embodiment.

FIG. 3 is a diagram for explaining an abnormal pulse wave envelope and an interpolation method thereof in the above embodiment.

FIG. 4 is a diagram for explaining the number of maximum values in a pulse wave envelope and the slope average before and after the maximum value.

FIG. 5 is a diagram showing a membership function of a minimum value.

FIG. 6 is a diagram showing a membership function of a maximum value.

FIG. 7 is a diagram showing a membership function of an average value of slopes before and after the maximum value.

FIG. 8 is a diagram showing a membership function as an output of fuzzy inference.

FIG. 9 is a flowchart for explaining an abnormality determination process of the embodiment apparatus.

FIG. 10 is a flowchart illustrating an abnormality determination process of the apparatus according to the embodiment.

[Explanation of symbols]

 1 pulse wave measurement unit 2 parameter calculation unit 3 storage unit 4 pattern holding unit 5 comparison unit 6 correction unit

Claims (1)

[Claims] 1. A waveform observation means, a parameter calculation means for calculating a pattern of a predetermined parameter from the observed waveform, a storage means for holding a standard pattern of parameters, and the calculated parameter pattern. A waveform processing device comprising: a comparison unit that compares the held standard pattern; and a correction unit that corrects the observed waveform parameter based on the comparison result of the comparison unit.