HK1027496B

HK1027496B - Pulse wave diagnostic apparatus

Info

Publication number: HK1027496B
Application number: HK00106478.8A
Authority: HK
Inventors: 天野和彦; 上马场和夫; 石山仁; 笠原宏
Original assignee: 精工爱普生株式会社
Priority date: 1997-11-20
Filing date: 1998-11-20
Publication date: 2008-06-06

Description

Pulse wave diagnostic device

[ technical field ] A method for producing a semiconductor device

The present invention relates to a pulse wave diagnostic device suitable for specifying the type of pulse, a blood pressure monitoring device using average blood pressure and pulse pressure as parameters, a pulse wave waveform monitoring device using parameters relating to the repeating unit of an arterial blood pressure waveform, and a pharmacological action monitoring device.

[ background of the invention ]

Generally, a pulse wave is a wave of blood that is transmitted within a blood vessel and is emitted from the heart. Therefore, it is known that various medical information can be obtained by detecting and analyzing the pulse wave. Further, as the study of the pulse wave is advanced, it has been found that various information that cannot be obtained from the blood pressure and the heart rate alone can be obtained by analyzing the pulse wave collected from the human body by various methods, and diagnosis can be performed based on the information.

The present inventors have focused on the relationship between the shape of the pulse wave and its distortion in PCT/JP96/01254 (the name of the invention: a body condition diagnosis apparatus and a body condition control apparatus), detected and processed the pulse wave of a subject, thereby calculating the distortion of the pulse wave, and specified the shape of the pulse wave based on the distortion, thereby diagnosing the body condition of the subject.

Here, the relationship between the shape of the pulse waveform and the distortion rate thereof described in the above application will be briefly described.

First, pulse waveforms are classified into various types and shapes, and here, a typical pulse waveform obtained by a classification method of chinese medicine, which is one of oriental traditional medicine, is described. Fig. 45A to 45C are diagrams showing the shapes of typical pulse waveforms obtained by this classification method.

Fig. 45A shows a pulse waveform called 'flat pulse', which is a pulse condition of a normal healthy person. As shown in the figure, the 'PingMai' is characterized by smooth and moderate pulse, and stable and disordered rhythm.

Next, the pulse wave shown in fig. 45B is called "a" smooth pulse ", and is a pulse condition of a person with abnormal blood flow conditions, and is characterized by a sharp rise and a sharp fall, and a deep aortic artery, and the peak thereafter is much higher than the normal value. The 'smooth pulse' is thought to be caused by edema, liver and kidney diseases, respiratory diseases, gastrointestinal diseases, inflammation and other diseases, and makes the pulse flow smoothly and smoothly.

The pulse wave shown in fig. 45B is called "chordal pulse" and is a pulse condition of a person whose vascular wall tone increases, and is characterized by a high-voltage state that continues for a certain period of time without immediately falling after a rapid rise. The 'string pulse' is considered to reflect diseases such as liver and gall diseases, skin diseases, hypertension, and painful diseases, and is caused by the fact that the vascular wall is tensed and the elasticity is reduced due to the tension of the autonomic nervous system, and the influence of rhythmic exercise of the pressed blood is hard to be expressed.

In fig. 45A to 45C, the vertical axis and the horizontal axis represent blood pressure (mmHg) and time (sec), respectively.

The pulse condition of the pulse waveform and the distortion rate d have the relationship shown in fig. 46. Here, the distortion rate d of the pulse waveform is determined by the following equation (1).

img id="idf0001" file="C9880351600041.GIF" wi="489" he="72" img-content="drawing" img-format="GIF"/

Furthermore, in the formula (1), A₁Is the amplitude of the fundamental component of the pulse wave, A₂、A₃、……A_nThe amplitudes of the 2 nd, 3 rd and nth harmonic components of the pulse wave, respectively.

Therefore, when the pulse waveform of the subject is detected, the amplitude a is obtained by performing, for example, FFT (fast fourier transform) processing₁～A_nThen, the distortion d is calculated, and the pulse condition of the pulse waveform can be quantitatively specified by using the correlation shown in FIG. 46.

As shown in FIG. 46, the distortion rate d is in the range of 0.98 to 1.22 when it is determined to be a smooth pulse, in the range of 0.92 to 1.10 when it is determined to be a flat pulse, and in the range of 0.73 to 0.94 when it is determined to be a chordal pulse.

In this case, when the distortion rate d of the pulse waveform is in the range of 0.98 to 1.10, it can be determined as either smooth or flat. When the distortion d of the pulse waveform is within the range of 0.92 to 0.94, it can be determined as either a flat pulse or a chordal pulse. Therefore, it is difficult for the conventional pulse wave diagnostic apparatus to accurately determine the pulse condition.

In non-invasive blood pressure monitoring, a sphygmomanometer that measures and displays a systolic blood pressure and a diastolic blood pressure is used.

However, even if the highest blood pressure and the lowest blood pressure are the same, there are various blood pressure waveforms. Therefore, there is a problem that the blood pressure characteristics of each person cannot be sufficiently expressed by only the maximum blood pressure and the minimum blood pressure.

In addition, although the average blood pressure is an important parameter for understanding the blood pressure state of each person, the average blood pressure cannot be known if only the maximum blood pressure and the minimum blood pressure are measured.

On the other hand, the pulse taking of the chinese medicine and the indian traditional medicine is performed based on the shape of the pulse wave perceived by the examiner by pressing the radial artery at the distal part of the wrist of the patient with his/her finger properly, that is, the change of the pressing force corresponding to the change of the blood pressure of the radial artery perceived by the examiner's finger.

For example, in chinese medicine, pulse wave shapes sensed by applying a pressing force to the radial artery are roughly classified into 3 types, which are called a flat pulse, a slippery pulse, and a chordal pulse, respectively, as described above. The feeling of the normal pulse is smooth and the rhythm is stable and not disordered, which is the pulse condition of healthy people. The smooth veins feel smooth and flow back and forth, indicating an abnormal blood flow. The feeling of the string pulse is that the stretched pulse is caused by the tense and aging of the vessel wall.

However, such a diagnostic method based on the pulse wave shape has a problem of objectivity and reproducibility because classification of the pulse wave shape depends on the feeling of the examiner.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a pulse wave diagnostic device capable of objectively and accurately determining a pulse condition.

Another object of the present invention is to provide a blood pressure monitoring device capable of displaying the state of blood pressure in more detail based on information on the highest blood pressure and the lowest blood pressure, and capable of monitoring parameters related to blood pressure noninvasively.

It is still another object of the present invention to provide a pulse waveform monitoring apparatus which can make an objective examination based on the shape of a pulse wave and has reproducibility.

[ disclosure of the invention ]

(1) A pulse wave diagnosis device of the present invention includes:

a pulse wave detection device for detecting a pulse wave waveform of a body;

a tidal wave feature extraction device for extracting a characteristic of a tidal wave from the pulse waveform and generating tidal wave feature information;

a beat wave feature extraction device for extracting the features of the beat wave from the pulse wave waveform and generating beat wave feature information;

and a pulse condition determining device for determining the pulse condition of the body based on the tidal wave characteristic information and the gravitational wave characteristic information.

(2) In the pulse wave diagnostic device according to item (1), the tidal wave characteristic extraction unit preferably generates tidal wave characteristic information based on an amplitude change of the tidal wave in a time domain, and the dicrotic wave characteristic extraction unit preferably generates dicrotic wave characteristic information based on an amplitude change of the dicrotic wave in a time domain.

(3) In the pulse wave diagnostic apparatus according to item (2), it is preferable that the amplitude changes of the tidal wave and the gravitational wave in the time domain are calculated by differentiating the pulse wave waveform once or 2 times with respect to time.

(4) A pulse wave diagnosis device of the present invention includes:

a pulse wave detection device for detecting a pulse wave waveform of a body;

a spectrum analyzer for performing spectrum analysis on the pulse waveform;

a tidal wave feature extraction device for extracting a feature of a tidal wave from the pulse waveform based on an analysis result of the spectrum analysis device and generating tidal wave feature information;

a beat wave characteristic extraction unit for extracting a beat wave characteristic from the pulse wave waveform based on an analysis result of the spectrum analysis unit and generating beat wave characteristic information;

(5) The pulse wave diagnostic apparatus according to item (4),

the tidal wave feature extracting means preferably specifies a duration of a tidal wave in the pulse waveform, extracts a feature of the tidal wave from the pulse waveform based on an analysis result of the spectrum analyzing means during the specified duration, and generates tidal wave feature information,

preferably, the characteristics extracting means specifies a duration of the beat wave in the pulse waveform, extracts characteristics of the beat wave from the pulse waveform based on an analysis result of the spectrum analyzing means during the specified duration, and generates the characteristics information of the beat wave.

(6) The pulse wave diagnostic device according to item (4) or (5),

the spectrum analyzing device preferably performs FFT processing on the pulse waveform.

(7) The pulse wave diagnostic device according to item (4) or (5),

the spectrum analyzing device preferably performs wavelet transform processing on the pulse waveform.

(8) The pulse wave diagnostic device according to any one of items (1) to (7),

preferably, the pulse condition judging means has informing means for informing the pulse condition judged by the pulse condition judging means.

(9) A pulse wave diagnosis device of the present invention includes:

a pulse wave detection device for detecting a pulse wave waveform from a detection site of a body;

and an autocorrelation calculating device for calculating autocorrelation data representing autocorrelation of the pulse wave waveform detected by the pulse wave detecting device.

And a pulse condition data generating device for generating pulse condition data indicating the type of the pulse wave based on the autocorrelation data.

(10) The pulse wave diagnostic apparatus according to item (9),

preferably, the pulse data generating device generates the pulse data by comparing the autocorrelation data with a predetermined threshold value.

(11) The pulse wave diagnostic apparatus according to item (10),

preferably, the pulse data generating device includes a minimum value detecting unit for detecting a minimum value of the autocorrelation data during one heartbeat period,

the comparison unit compares the minimum value detected by the minimum value detection unit with the threshold value to generate the pulse condition data.

(12) The pulse wave diagnostic apparatus according to item (10),

preferably, the pulse data generating device includes a minimum value detecting unit for detecting an average minimum value of the autocorrelation data by averaging the minimum values detected in several cardiac cycles,

the comparison unit compares the average minimum value detected by the minimum value detection unit with the threshold value to generate the pulse condition data.

(13) The pulse wave diagnostic apparatus according to item (9),

preferably, the pulse data generating device includes a time measuring unit, a calculating unit, and a comparing unit, the time measuring unit compares the autocorrelation data with a predetermined threshold value, measures a time interval in which the autocorrelation data is higher than the threshold value and a time interval in which the autocorrelation data is lower than the threshold value,

the calculation unit calculates the ratio of the time interval measured by the time measurement unit to one heart cycle,

the comparison unit compares the result calculated by the calculation unit with a predetermined threshold value to generate the pulse condition data.

(14) The pulse wave diagnostic apparatus according to item (13),

preferably, the calculation unit calculates a ratio of the time interval measured by the time measurement unit to a single heart cycle, and calculates an average value of the calculation results.

(15) The pulse wave diagnostic apparatus according to item (9),

preferably, the pulse data generating device includes a change rate calculating section and a change rate comparing section, the change rate calculating section detects a change rate of the autocorrelation data based on the autocorrelation data,

the change rate comparing unit compares the change rate detected by the change rate calculating unit with a predetermined threshold value to generate the pulse condition data.

(16) The pulse wave diagnostic apparatus according to item (15),

preferably, the change rate comparing unit detects a maximum value of the change rate, and generates the pulse condition data by comparing the maximum value of the change rate with the threshold value.

(17) The pulse wave diagnostic apparatus according to item (9),

the pulse data generating device preferably includes a minimum value detecting unit for detecting a minimum value of the autocorrelation data for 1 heartbeat period, a 1 st comparing unit, a time measuring unit, a calculating unit, and a 2 nd comparing unit,

a 1 st comparing section for comparing the minimum value detected by the minimum value detecting section with a predetermined 1 st threshold value and generating pulse condition data indicating a chordal pulse when the minimum value is lower than the 1 st threshold value,

the time measuring unit compares the autocorrelation data with a predetermined 2 nd threshold value, measures a time interval during which the autocorrelation data is higher than the 2 nd threshold value or a time interval during which the autocorrelation data is lower than the 2 nd threshold value,

the 2 nd comparing unit generates the pulse condition data indicating a flat pulse or a smooth pulse by comparing the result calculated by the calculating unit with a predetermined 3 rd threshold value.

(18) The pulse wave diagnostic apparatus according to item (9),

the pulse data generating device preferably includes a minimum value detecting unit for detecting a minimum value of the autocorrelation data for 1 heartbeat period, a 1 st comparing unit, a change rate calculating unit, and a 2 nd comparing unit,

a 1 st comparing section for comparing the minimum value detected by the minimum value detecting section with the 1 st threshold value and generating pulse condition data indicating a chordal pulse when the minimum value is lower than the 1 st threshold value,

the change rate calculation unit detects a change rate of the autocorrelation data based on the autocorrelation data,

the 2 nd comparing unit generates the pulse condition data indicating a flat pulse or a smooth pulse by comparing the change rate detected by the change rate calculating unit with a predetermined threshold value.

(19) The pulse wave diagnostic device according to the item (17) or (18),

the autocorrelation data represents an autocorrelation coefficient, and the 1 st threshold value used in the comparison operation of the 1 st comparing section is preferably about 0.25.

(20) The pulse wave diagnostic apparatus according to item (17),

the autocorrelation data represents an autocorrelation coefficient, and the 2 nd threshold value used for the comparison operation of the time measuring section is preferably set in the range of 0.4 to 0.8.

(21) The pulse wave diagnostic device according to any one of (9) to (20),

preferably, the body motion detecting device includes a body motion detecting device for detecting a body motion waveform indicating the body motion, and a body motion component removing device for generating a body motion component in the pulse wave waveform from the body motion waveform, removing the body motion component from the pulse wave waveform, and generating a body motion-removed pulse wave waveform,

the autocorrelation calculating means calculates autocorrelation data representing autocorrelation based on the body motion-removed pulse waveform, not on the pulse waveform.

(22) The pulse wave diagnostic apparatus according to item (21),

preferably, the body motion detecting device includes a judging device for judging whether the body is moving or not based on the body motion waveform detected by the body motion detecting device,

the body motion component removing device stops the body motion removing operation and outputs the pulse waveform without outputting the body motion removing pulse waveform when the body motion is not present as a result of the determination by the determining device.

(23) The pulse wave diagnostic device according to any one of (9) to (20),

it is preferable to have 1 st wavelet transform means, body motion detecting means, 2 nd wavelet transform means, body motion component removing means and inverse wavelet transform means,

a 1 st wavelet transformation means for performing wavelet transformation on the pulse wave waveform detected by the pulse wave detection means and generating pulse wave analysis data for each frequency band,

the body motion detecting device detects the body motion and outputs a body motion waveform,

a 2 nd wavelet transformation means for performing wavelet transformation on the body motion waveform detected by the body motion detection means and generating body motion analysis data for each frequency region,

the body motion component removing device subtracts the body motion analysis data from the pulse wave analysis data to generate body motion-removed pulse wave analysis data from which body motion has been removed,

the inverse wavelet transform device performs inverse wavelet transform on the body motion removed pulse wave analysis data to generate a body motion removed pulse wave waveform.

(24) The pulse wave diagnostic device according to any one of (9) to (20),

it is preferable to have wavelet transform means, body motion component removing means and inverse wavelet transform means,

a wavelet transformation device for performing wavelet transformation on the pulse wave detected by the pulse wave detection device and generating pulse wave analysis data for each frequency region,

a body motion component removing device for removing a frequency component corresponding to a predetermined body motion from the pulse wave analysis data to generate body motion-removed pulse wave analysis data,

(25) A pulse wave diagnosis device of the present invention includes:

an autocorrelation calculating device for calculating autocorrelation data representing autocorrelation with respect to the pulse wave analysis data in a certain frequency region,

(26) A pulse wave diagnosis device of the present invention includes:

a 1 st wavelet transformation device for performing wavelet transformation on the pulse waveform detected by the pulse wave detection device and generating pulse wave analysis data for each frequency region,

a body motion detecting device for detecting the body motion and outputting a body motion waveform,

a body motion component removing device for subtracting the body motion analysis data from the pulse wave analysis data to generate body motion-removed pulse wave analysis data from which body motion has been removed,

an autocorrelation calculating device for calculating autocorrelation data representing autocorrelation with respect to the body motion-removed pulse wave analysis data in a certain frequency region,

(27) A pulse wave diagnosis device of the present invention includes:

(28) The pulse wave diagnostic device according to any one of (9) to (27),

preferably, the pulse condition data generating device includes a notifying device for notifying the pulse condition data generated by the pulse condition data generating device.

(29) A blood pressure monitoring device of the present invention includes:

an arterial pressure waveform detection unit that continuously measures arterial blood pressure and detects an arterial pressure waveform;

and a mean blood pressure calculation unit for calculating a mean blood pressure from the arterial pressure waveform.

A blood pressure monitoring device of the present invention includes an average blood pressure calculation unit that calculates an average blood pressure from an arterial pressure waveform detected by an arterial pressure waveform detection unit. Therefore, the average blood pressure can be monitored using the arterial pressure waveform detected by the arterial pressure waveform detecting unit.

(30) The blood pressure monitoring device according to the item (29),

it is preferable that the blood pressure calculating unit further includes a pulse pressure calculating unit for calculating a pulse pressure, which is a pressure difference between the highest blood pressure and the lowest blood pressure, from the arterial pressure waveform.

According to the present invention, the pulse pressure can be monitored using the arterial pressure waveform detected by the arterial pressure waveform detecting unit.

(31) The blood pressure monitoring device according to the (30) above,

it is preferable that the blood pressure measuring device further includes a blood pressure converting unit that converts the arterial pressure waveform detected by the arterial pressure waveform detecting unit into a cardiac arterial pressure waveform, which is an arterial pressure waveform at a position corresponding to a height of a heart, wherein the average blood pressure calculating unit calculates an average blood pressure from the cardiac arterial pressure waveform, and the pulse pressure calculating unit calculates a pulse pressure from the cardiac arterial pressure waveform.

According to the present invention, the blood pressure conversion unit converts the arterial pressure waveform detected by the arterial pressure waveform detection unit into an arterial pressure waveform at a position corresponding to the height of the heart, that is, a cardiac arterial pressure waveform. Then, based on the heart position arterial pressure waveform, the average blood pressure calculating unit calculates the average blood pressure, and the pulse pressure calculating unit calculates the pulse pressure. Therefore, at least one parameter of the average blood pressure and the pulse pressure of the artery corresponding to the height of the heart can be monitored using the arterial pressure waveform detected by the arterial pressure waveform detecting unit.

(32) The blood pressure monitoring device according to item (30) or (31),

preferably, the blood pressure determination device further includes a blood pressure determination information storage unit that stores blood pressure determination information in advance, and a blood pressure determination unit that performs blood pressure determination based on the blood pressure determination information and at least one parameter of the average blood pressure and the pulse pressure.

According to the present invention, the blood pressure monitoring apparatus can perform, for example, the determination of hypertension, hypotension, normality, and the like based on at least one parameter of average blood pressure and pulse pressure and the blood pressure determination information stored in advance.

(33) The blood pressure monitoring device according to any one of (30) to (32),

preferably, the blood pressure monitor further includes an output unit that outputs at least one of information corresponding to the average blood pressure, information corresponding to the pulse pressure, and information corresponding to the blood pressure determination.

According to the present invention, the blood pressure monitoring device can output at least one of information corresponding to the average blood pressure, information corresponding to the pulse pressure, and information corresponding to the blood pressure determination, for example, in the form of a numerical value, a graph, a voltage, or the like, by the output unit.

(34) A blood pressure monitoring device of the present invention includes:

the pulse pressure calculation unit calculates a pulse pressure, which is a pressure difference between the highest blood pressure and the lowest blood pressure, from the arterial pressure waveform.

(35) The blood pressure monitoring device according to (34),

it is preferable that the blood pressure monitor further includes a hypertension calculation unit for calculating a hypertension from the pulse pressure and a hypotension calculation unit for calculating a hypotension from the pulse pressure and the hypertension.

According to the present invention, the hypertension calculating unit determines the hypertension based on the pulse pressure by using the characteristic that the hypertension can be expressed as a linear function of the pulse pressure. Since the pulse pressure is a pressure difference between the highest blood pressure and the lowest blood pressure, the lowest blood pressure calculation unit can calculate the lowest blood pressure if the highest blood pressure and the pulse pressure are known.

(36) A pulse wave shape monitoring device of the present invention includes:

the dicrotic wave height calculating unit calculates a blood pressure difference between the dicrotic wave and the dicrotic wave peak obtained from the arterial pressure waveform, that is, a dicrotic wave height.

According to the present invention, the pulse wave shape monitoring device can calculate the pulse wave height from the arterial pressure waveform obtained by continuously measuring the arterial blood pressure by the arterial pressure waveform detecting unit.

(37) The pulse wave shape monitoring device according to the item (36),

it is preferable that the blood pressure waveform calculation unit further includes a pulse pressure difference ratio calculation unit for calculating a pulse pressure difference ratio, which is a ratio of a pulse pressure difference obtained from the arterial pressure waveform between the blood pressure of the pulse and the hypotensive pressure, and a pulse pressure difference obtained from the hypotensive pressure waveform between the maximum blood pressure and the hypotensive pressure.

According to the present invention, the pulse wave shape monitoring device can calculate the pulse wave height and the pulse pressure difference ratio based on the arterial pressure waveform obtained by continuously measuring the arterial blood pressure by the arterial pressure waveform detecting unit.

(38) The pulse wave shape monitoring device according to item (37),

preferably, the blood pressure/pulse pressure ratio calculating unit further includes a blood pressure/pulse pressure ratio calculating unit for calculating a ratio of the average blood pressure obtained from the arterial pressure waveform to a pulse pressure, which is a pressure difference between the highest blood pressure and the lowest blood pressure.

According to the present invention, the pulse wave shape monitoring device can calculate the pulse wave height, the pulse pressure difference ratio, and the average blood pressure pulse ratio from the arterial pressure waveform obtained by continuously measuring the arterial blood pressure by the arterial pressure waveform detecting unit.

(39) The pulse wave shape monitoring device according to the item (38),

preferably, the blood pressure conversion unit further includes a blood pressure conversion unit for converting the arterial pressure waveform detected by the arterial pressure waveform detection unit into an arterial pressure waveform at a position corresponding to the height of the heart, i.e., a cardiac arterial pressure waveform,

the heart pulse height calculating unit calculates a blood pressure difference between the heart pulse and the heart pulse peak, which is obtained from the heart position arterial pressure waveform, that is, a heart pulse height.

The heart pressure difference ratio calculating unit calculates a heart pressure difference ratio, which is a ratio of a heart pressure difference between the heart pressure and the lowest blood pressure, which is obtained from the heart artery pressure waveform, to a heart pressure difference between the heart pressure and the lowest blood pressure, which is a pulse pressure.

The average blood pressure-to-pulse pressure ratio calculation unit calculates an average blood pressure-to-pulse pressure ratio, which is a ratio of an average blood pressure obtained from the heart position arterial pressure waveform to a pulse pressure, which is a pressure difference between a highest blood pressure and a lowest blood pressure.

According to the present invention, the blood pressure conversion unit converts the arterial pressure waveform detected by the arterial pressure waveform detection unit into an arterial pressure waveform at a position corresponding to the height of the heart, that is, a cardiac arterial pressure waveform. Then, based on the heart position arterial pressure waveform, a beat height calculating unit calculates a beat height, a beat pressure difference ratio calculating unit calculates a beat pressure difference ratio, and an average blood pressure/pulse ratio calculating unit calculates an average blood pressure/pulse ratio. Therefore, at least one parameter of the pulse height of the artery, the pulse pressure difference ratio, and the average blood pressure pulse ratio corresponding to the height of the heart can be monitored using the arterial pressure waveform detected by the arterial pressure waveform detecting unit.

(40) The pulse wave shape monitoring device according to the item (36),

preferably, the pulse wave shape determination device further includes a pulse wave shape determination information storage unit that stores pulse wave shape determination information in advance, and a pulse wave shape determination unit that determines the shape of the pulse wave based on the pulse wave height and the pulse wave shape determination information.

According to the present invention, the pulse wave shape determination unit can determine the pulse wave shape based on the pulse wave height and the pulse wave shape determination information.

(41) The pulse wave shape monitoring device according to item (37),

preferably, the pulse wave shape judging device further comprises a pulse wave shape judging information storage section and a pulse wave shape judging section for storing the pulse wave shape judging information in advance,

the pulse wave shape determination unit determines the pulse wave shape based on the pulse wave height, the pulse pressure difference ratio, and the pulse wave shape determination information.

According to the present invention, the pulse wave shape determination unit can determine the pulse wave shape based on the pulse wave height, the pulse pressure difference ratio, and the pulse wave shape determination information.

(42) The pulse wave shape monitoring device according to the item (38),

the pulse wave shape determination unit determines the pulse wave shape based on the pulse wave height, the pulse pressure difference ratio, the average blood pressure-to-pulse pressure ratio, and the pulse wave shape determination information.

According to the present invention, the pulse wave shape determination unit can determine the pulse wave shape based on the pulse wave height, the pulse pressure difference ratio, the average blood pressure-to-pulse pressure ratio, and the pulse wave shape determination information.

(43) The pulse wave shape monitoring device according to the item (36),

preferably, the pulse wave generator further includes an output unit for outputting at least one of information corresponding to the pulse wave height and information corresponding to the pulse wave shape.

According to the present invention, the pulse wave shape monitoring device can output at least one of information corresponding to the level of the pulse wave and information corresponding to the shape of the pulse wave from the output unit, for example, in the form of a numerical value, a graph, a voltage, or the like.

(44) The pulse wave shape monitoring device according to item (37),

preferably, the pulse wave generator further includes an output unit that outputs at least one of information corresponding to the pulse wave height, information corresponding to the pulse pressure difference ratio, and information corresponding to the pulse wave shape.

According to the present invention, the pulse wave shape monitoring device can output at least one of information corresponding to the level of the pulse wave, information corresponding to the pressure difference ratio of the pulse wave, and information corresponding to the shape of the pulse wave, for example, in the form of a numerical value, a graph, a voltage, or the like, from the output unit.

(45) The pulse wave shape monitoring device according to the item (38),

preferably, the blood pressure monitor further includes an output unit that outputs at least one of information corresponding to the pulse height, information corresponding to the pulse pressure difference ratio, information corresponding to the average blood pressure/pulse pressure ratio, and information corresponding to the pulse wave shape.

According to the present invention, the pulse wave shape monitoring device can output at least one of information corresponding to the pulse height, information corresponding to the pulse pressure difference ratio, information corresponding to the average blood pressure/pulse pressure ratio, and information corresponding to the pulse wave shape, for example, in the form of a numerical value, a graph, a voltage, or the like, from the output unit.

(46) A pulse wave shape monitoring device of the present invention includes:

a pulse wave detection unit for detecting a pulse wave waveform of a body;

the pulsatory pressure difference ratio calculating unit calculates a pulsatory pressure difference ratio, which is a pressure difference between the pulsatory pressure and the lowest pressure obtained from the pulse wave waveform, that is, a ratio of the pulsatory pressure difference to a pressure difference between the highest pressure and the lowest pressure, that is, a pulse pressure.

Unlike the invention described in (36), the systolic pressure ratio calculation unit only needs to obtain a ratio and thus does not need an absolute blood pressure value. Therefore, instead of the arterial pressure waveform detecting unit in the invention described in (36), a pulse wave detecting unit may be used, and only the waveform (pulse wave) corresponding thereto may be detected.

(47) A pulse wave shape monitoring device of the present invention includes:

a pulse wave detection unit for detecting a pulse wave waveform of a body;

an average pressure/pulse pressure ratio calculating unit calculates an average pressure/pulse pressure ratio, which is a ratio of the average pressure obtained from the pulse waveform to a pulse pressure, which is a pressure difference between the highest pressure and the lowest pressure.

Unlike the invention described in (16), the average blood pressure-to-pulse pressure ratio calculating unit only needs to obtain a ratio and thus does not need an absolute blood pressure value. Therefore, instead of the arterial pressure waveform detecting unit in the invention described in (16), a pulse wave detecting unit may be used, and only the waveform (pulse wave) corresponding thereto may be detected.

(48) A blood pressure monitoring device of the present invention includes:

a mean blood pressure calculation unit for calculating a mean blood pressure from the arterial pressure waveform;

(49) A blood pressure monitoring device of the present invention includes:

the pulsation pressure difference ratio calculation unit calculates a pulsation pressure difference ratio, which is a ratio of a pulsation pressure difference, which is a pressure difference between the blood pressure of the pulsation and the lowest blood pressure, which is obtained from the arterial pressure waveform, and a pulse pressure, which is a pressure difference between the highest blood pressure and the lowest blood pressure.

(50) A blood pressure monitoring device of the present invention includes:

the average blood pressure/pulse pressure ratio calculation unit calculates an average blood pressure/pulse pressure ratio, which is a ratio of an average blood pressure obtained from the arterial pressure waveform to a pulse pressure, which is a pressure difference between the highest blood pressure and the lowest blood pressure.

(51) A blood pressure monitoring device of the present invention includes:

a pulse pressure calculation unit that calculates a pulse pressure, which is a pressure difference between the highest blood pressure and the lowest blood pressure, from the arterial pressure waveform;

(52) A blood pressure monitoring device of the present invention includes:

(53) A blood pressure monitoring device of the present invention includes:

(54) A pulse wave shape monitoring device of the present invention includes:

and a purge expansion pressure calculation unit for calculating a blood pressure difference between the systolic blood pressure and the systolic blood pressure obtained from the arterial pressure waveform, that is, a purge expansion pressure.

(55) The pulse wave shape monitoring device according to the item (54),

preferably, the pulse wave shape judging device further includes a pulse wave shape judging information storage unit for storing pulse wave shape judging information in advance, and a pulse wave shape judging unit for judging the shape of the pulse wave based on the above-mentioned discharge stretching pressure and the above-mentioned pulse wave shape judging information.

(56) A pulse wave shape monitoring device of the present invention includes:

a pulse wave detection unit for detecting a pulse wave waveform of a body;

and a pulsar diastolic pressure ratio calculating unit for calculating a pulse pressure ratio, which is a pressure difference between the systolic blood pressure and the systolic blood pressure, which is obtained from the pulse waveform, and a pressure difference between the systolic blood pressure and the systolic blood pressure, which is the pulsar diastolic blood pressure.

(57) The pulse wave shape monitoring device according to the item (56),

preferably, the pulse wave shape judging device further includes a pulse wave shape judging information storage unit for storing pulse wave shape judging information in advance, and a pulse wave shape judging unit for judging the shape of the pulse wave based on the above-mentioned pulse-off pressure ratio.

(58) The pharmacological action monitoring device of the present invention comprises:

(59) The pharmacological action monitoring device according to item (58),

preferably, the blood pressure sensor further includes a pulse height calculating unit for calculating a blood pressure difference between the pulse and the pulse peak obtained from the arterial pressure waveform, that is, a pulse height.

(60) The pharmacological action monitoring device of the present invention comprises:

a pulse wave detection unit that detects a pulse wave waveform from a body;

and a expelled-diastolic-pressure-ratio calculating unit for calculating a expelled-diastolic pressure ratio, which is a ratio of a pressure difference between the systolic pressure and the diastolic pressure, that is, a pressure difference between the expelled-diastolic pressure and the systolic pressure, that is, a pulse pressure, obtained from the pulse waveform.

(61) The pharmacological action monitoring device according to item (60),

preferably, the pulse wave height ratio calculating unit further includes a pulse wave height ratio calculating unit for calculating a pulse pressure ratio, which is a ratio of a pulse pressure, which is a pressure difference between the pulse wave and the pulse wave peak, that is, a pressure difference between the systolic pressure and the diastolic pressure, obtained from the pulse wave waveform.

[ brief description of the drawings ]

Fig. 1 is a block diagram showing a configuration of a pulse wave diagnostic apparatus according to embodiment 1.

Fig. 2 is a flowchart for explaining the operation of the pulse wave diagnostic device according to embodiment 1.

Fig. 3 is a block diagram showing the configuration of a pulse wave diagnostic apparatus according to embodiment 2.

Fig. 4 is a block diagram showing an example of the configuration of the wavelet transform unit according to embodiment 2.

Fig. 5 is a block diagram showing the configuration of the waveform shaping unit according to embodiment 2.

Fig. 6 is a flowchart showing the operation of the waveform shaping unit according to embodiment 2.

Fig. 7 is a diagram for explaining the operation of the tidal wave and dicrotic wave detection unit according to embodiment 2.

Fig. 8 is a diagram showing correspondence with an electrocardiographic waveform, an aortic blood pressure waveform, and a distal blood pressure waveform.

Fig. 9 is a diagram for explaining the correspondence between the pulse waveform and the waveform parameter.

FIG. 10 is a graph showing the difference (y) between blood pressure values₅-y₄) And distortion rate d.

Fig. 11 is a diagram showing the result of spectral analysis of the veins.

Fig. 12 is a graph showing the result of spectral analysis of the flat pulse.

Fig. 13 is a diagram showing the result of spectrum analysis of a chordal vein.

Fig. 14 is a graph showing the amplitudes of the tidal wave and the heartbeat wave of each pulse condition.

Fig. 15 is a graph showing harmonic components of each pulse condition.

Fig. 16 is a diagram showing autocorrelation coefficients of chordal arteries.

Fig. 17 is a graph showing the autocorrelation coefficient of the flat pulse.

Fig. 18 is a graph showing autocorrelation coefficients of a smooth vein.

Fig. 19 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 3.

Fig. 20 is a flowchart showing the operation of the pulse wave diagnostic apparatus according to embodiment 3.

Fig. 21 is a block diagram showing a pulse data generating unit according to embodiment 4.

Fig. 22 is a diagram showing a change rate of autocorrelation data of a typical pulse waveform.

Fig. 23 is a flowchart showing the operation of the pulse wave diagnostic apparatus according to embodiment 4.

Fig. 24 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 5.

Fig. 25 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 6.

Fig. 26 is a diagram showing pulse wave analysis data of a pulse wave waveform over time.

Fig. 27 is a timing chart for explaining the operation of the body motion eliminating unit according to embodiment 6.

Fig. 28 is a diagram showing pulse wave correction data MKDa in the period Tc in embodiment 6.

Fig. 29 is a diagram showing the body motion correction data TKDa in the period Tc in the embodiment 6.

Fig. 30 is a diagram showing body motion removed pulse wave data MKDaj after body motion component removal in example 6.

Fig. 31 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 7.

Fig. 32 is a block diagram of a body motion eliminating unit according to embodiment 7.

Fig. 33 is a diagram showing an example of body motion-removed pulse wave data according to embodiment 7.

Fig. 34 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 8.

Fig. 35 is a block diagram of the 1 st wavelet transform unit of embodiment 8.

Fig. 36 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 9.

Fig. 37A is a diagram showing a state in which the wristwatch-type pulse wave diagnostic device is mounted.

Fig. 37B is a diagram showing a pulse wave detection unit of the wristwatch-type pulse wave diagnostic device.

Fig. 37C is a diagram showing a connection portion provided in the main body of the wristwatch-type pulse wave diagnostic device.

Fig. 38 is a diagram showing a configuration example of the pulse wave detection unit.

Fig. 39A is a diagram showing an external appearance of another wristwatch-type pulse wave diagnostic device.

Fig. 39B is a diagram showing a mounted state of the pulse wave diagnostic device shown in fig. 39A.

Fig. 40 is a diagram showing an external configuration of the necklace-type pulse wave diagnostic apparatus.

Fig. 41 is a diagram showing a state in which the pulse wave detecting unit of the pulse wave diagnostic apparatus shown in fig. 40 is placed in the carotid artery.

Fig. 42 is a diagram showing an external configuration of the eyeglass-type pulse wave diagnostic apparatus.

Fig. 43 is a diagram showing an external configuration of the card pulse wave diagnostic apparatus.

Fig. 44A is a diagram showing an external configuration of the pedometer type pulse wave diagnostic device.

Fig. 44B is a diagram showing a mounted state of the pulse wave diagnostic device shown in fig. 44A.

Fig. 45A is a diagram showing a typical flat pulse waveform.

Fig. 45B is a diagram showing a typical pulse waveform of a pulse.

Fig. 45C is a diagram showing a typical chordal pulse waveform.

Fig. 46 is a diagram showing a correlation between the distortion rate and the shape of the pulse wave waveform.

Fig. 47 is an explanatory diagram showing a structure for recording an arterial pressure waveform of a radial artery according to the theory of embodiment 10.

Fig. 48 is a diagram showing a typical arterial pressure waveform.

Fig. 49 is a graph showing the results of an experiment using the structure shown in fig. 47, and shows the relationship between the average blood pressure and the diastolic blood pressure.

Fig. 50 is a graph showing the results of an experiment using the structure shown in fig. 47, and shows the relationship between the average blood pressure and the systolic blood pressure.

FIG. 51 is a graph showing the results of an experiment using the structure shown in FIG. 47, and shows the relationship between pulse pressure and systolic blood pressure.

Fig. 52 is a block diagram showing a blood pressure monitoring apparatus according to embodiment 10.

FIG. 53 shows the pulse height Δ BP in each pulse shape according to the theory of the 11 th embodiment_DA graph of the distribution of (c).

FIG. 54 shows the pulse pressure difference ratio BP as a theoretical basis for example 11_DdΔ BP and Bobo height Δ BP_DA graph of the relationship between.

FIG. 55 shows the kick-out expansion pressure Δ BPp and the dicrotic wave height Δ BP according to the theory of the embodiment 11_DA graph of the relationship between.

Fig. 56 is a block diagram showing the configuration of a pulse wave shape monitoring device according to embodiment 11.

Fig. 57 is a block diagram showing the configuration of the pharmacological action monitoring device according to embodiment 12.

Fig. 58 is a block diagram showing a modification of the wavelet transform unit configured by a filter bank.

Fig. 59 is a block diagram showing a modification example in which the inverse wavelet transform unit is configured by a filter bank.

Fig. 60 is a face view showing a modification of the notification device.

Fig. 61 is a diagram showing an example of a transmission-type photoelectric pulse wave sensor according to a modification.

Fig. 62 is a diagram showing a modification example in the case where the photoelectric pulse wave sensor is used in the spectacle-type pulse wave monitoring device.

Fig. 63 is a graph showing changes in pressure of each part in an arterial pressure waveform after administration of a hypotensive agent.

Fig. 64 is a graph showing arterial pressure waveforms before and after administration of a hypotensive agent.

[ best mode for carrying out the invention ]

EXAMPLE 1

First, a pulse wave diagnosis device according to embodiment 1 of the present invention will be described.

1.1 theoretical basis for example 1

Needless to say, the heart is driven out of blood by repeated contraction and expansion. Here, the time at which blood flows out of the heart due to 1 cycle of contraction and expansion is referred to as the expulsion time. When the number of contractions of the heart per unit time, that is, the number of beats, increases due to exercise, catecholamines such as epinephrine are released, and as a result, the expulsion time tends to become short. This means that the contractile force of the cardiac muscle increases.

As the contraction time increases, the amount of blood flowing out tends to increase in 1 cycle of systolic expansion of the heart.

When a person performs an exercise or other activity, a large amount of oxygen must be supplied to the cardiac muscle and skeletal muscle, and thus the product of the stroke number and the stroke volume, that is, the flow rate of blood delivered by the heart per unit time increases. Here, the ejection time is shortened as a result of the increase in the stroke number, and the stroke volume is conversely reduced. However, since the increase rate of the stroke rate is higher than the decrease rate of the stroke volume, the product of the stroke rate and the stroke volume increases as a whole.

Next, the relationship between the motion of the heart and the blood waveform will be described. In the electrocardiographic waveform shown in fig. 8, the period from the R point to the end point U of the T wave is generally referred to as a ventricular systole, which corresponds to the above-described expulsion time. The period from the point U to the next point R is referred to as a ventricular diastole. Here, in the ventricular systole, the contraction of the ventricles does not occur uniformly, but gradually contracts from the outside to the inside. Therefore, the blood pressure waveform at the aortic origin on the posterior side of the heart has a convex shape in the ventricular systole from the aortic valve opening to closing, as shown in fig. 8.

Corresponding to the blood pressure waveform at the aorta origin, the blood pressure waveform at the distal portion (radial artery), that is, the pulse wave waveform at the distal portion is also as shown in fig. 8. The shape is changed to such a shape that the heart first produces the 1 st wave (ejection wave) called a kick wave due to blood pulsation, then the 2 nd wave (tidal wave) called a tidal wave due to reflection at a bifurcated portion of a blood vessel near the heart, and then the 3 rd wave (dicrotic wave) called a dicrotic wave occurs due to the occurrence of a dicrotic wave as a result of the closing of the aortic valve.

Therefore, in the pulse waveform, the period from the lowest point of the blood pressure value to the rebroadcasting corresponds to the ventricular systole, and the period from the rebroadcasting to the lowest point of the blood pressure value in the next cycle corresponds to the ventricular diastole.

Here, the point corresponding to the opening of the aortic valve in the pulse waveform is the point of the minimum value of the blood pressure value. The point corresponding to the aortic valve closing in the pulse waveform, that is, the rebleed, is the 3 rd minimum point counted from the minimum point in the time axis, and is the 2 nd minimum point counted from the minimum point in the time axis.

Note that the distal blood pressure waveform shown in fig. 8, that is, the pulse wave waveform is actually delayed in time from the aortic blood pressure waveform, and for convenience of explanation, the time delay is ignored and the phases are made to coincide.

Next, the distal blood pressure waveform, i.e., the pulse wave waveform, will be discussed. Since the pulse wave waveform detected from the tip of the subject is a pressure wave in which blood passes through a closed system including a heart as a pulsating pump and a vascular system as a catheter, the pulse wave waveform 1 is affected by the diameter of a blood vessel, the contraction and expansion of the blood vessel, and the viscous impedance of blood, in addition to the pump function of the heart, i.e., the cardiac function state. Therefore, it is considered that when the pulse waveform is detected and analyzed, the state of the cardiac function can be evaluated in addition to the state of the arterial system of the subject, and a medical professional in oriental medicine can diagnose the state of the body from the form of pulsation.

Here, it is discussed which part of the pulse waveform should be analyzed.

First, the present inventors determined waveform parameters for determining the shape characteristics of the pulse waveform as shown in fig. 9. That is, the following waveform parameters were determined.

(1) A time t from a point P0 (minimum value point) at which the blood pressure value of the pulse waveform is minimum and is the rising bottom point of one heartbeat to a rising start point P6 of the next beat₆，

(2) Blood pressure value (difference) y of peak-valley points (maximum value point and minimum value point) P1-P5 appearing in sequence in pulse wave waveform₁～y₅，

(3) The elapsed time t from the bottom point P0 (minimum point) at the pulse start time to the appearance of the peak-bottom points P1 to P5₁～t₅。

Furthermore, at this time, y₁～y₅Each represents a relative blood pressure value based on the blood pressure value at the valley point P0.

The present inventors actually detected the pulse wave of 74 healthy adults aged 22 to 46 and found the waveform parameters, respectively, and, on the other hand, similarly to the above-mentioned PCT/JP96/01254, FFT-processed the pulse wave waveform and found the distortion rate d of the pulse wave waveform using the above-mentioned expression (1).

The present inventors also studied the correlation between the distortion rate d and the respective waveform parameters and the difference between these parameters, and found that the distortion rate d and the amplitude of the dicnodic wave from the onset of the dicrotic beat, that is, the blood pressure difference (y)₅-y₄) Has strong correlation relationship, and the correlation coefficient (R) thereof²) Is 0.77. FIG. 10 showsThe correlation relationship.

Thus, the present inventors conducted the following analysis on the hypothesis that a medical professional in oriental medicine diagnoses a pulse condition by feeling the characteristics of a dicrotic wave (dicrotic wave) and a tidal wave (tidal wave) with a finger.

In the analysis, FFT processing is performed on each pulse waveform used by a medical professional in oriental medicine to determine a pulse condition, and each higher harmonic component with respect to the fundamental component is calculated. Fig. 11 shows the analysis results of the smooth arteries, fig. 12 shows the analysis results of the flat arteries, and fig. 13 shows the analysis results of the chordal arteries.

In fig. 11 to 13, f1 is a fundamental wave, f2 is a 2 th harmonic, f3 is a 3 rd harmonic, … f10 is a 10 th harmonic, and amplitudes and phases thereof are shown, respectively, and further, a waveform Wf1 is a waveform in which a fundamental wave f1 and a 2 nd harmonic f2 are added, a waveform Wf2 is a waveform in which a fundamental wave f1 is added to a 3 rd harmonic f3, and a waveform … Wf9 is a waveform in which a fundamental wave f1 is added to a 10 th harmonic f 10.

Here, when comparing the original waveform determined as a smooth pulse by the medical professional shown in fig. 11 with the original waveform determined as a flat pulse shown in fig. 12, it is understood that the two waveforms are similar to each other, but the height of the heavy pulse (dicrotic notch) of the smooth pulse is low and the amplitude of the heavy pulse (dicrotic wave) of the flat pulse is large. Further, when the synthesized waveforms are compared, it is found that the pulmonic and pulmonic heavy waves (dicnodic waves) can substantially reproduce the original waveform by the waveform Wf3 added from the fundamental wave f1 to the 4 th harmonic f 4.

On the other hand, in the chordal vein shown in fig. 13, the tidal wave (tidal wave) thereof can substantially reproduce the original waveform with the waveform Wf6 added from the fundamental wave f1 to the 7 th harmonic f 7.

Fig. 14 is a diagram showing the amplitude of a dicrotic wave and the amplitude of a tidal wave in the pulse wave waveforms of the pulse conditions described in fig. 11 to 13. The amplitude of the dicnodic wave (dicnodic wave) is 7.3mmHg for the smooth pulse, 10.6mmHg for the flat pulse, and 2.9mmHg for the small chordal pulse. In addition, the amplitude of the slippery and flat tidal waves (tidal wave) is 0, while the amplitude of the chordal tidal waves (tidal wave) is 3.8 mmHg.

Thus, it can be said that the pulsus and flat beat waves (dicnodic wave) have some characteristics, and the characteristics of the beat waves (dicnodic wave) can be reflected from the fundamental waves f1 to the 4 th harmonic f 4. While tidal waves (tidalwave) of chordal arteries are somewhat characteristic, the characteristics of tidal waves (tidalwave) can be reflected from high frequency components, such as the 5 th harmonic f 5-7 th harmonic f 7.

Fig. 15 is a graph showing the ratio of the amplitude of each harmonic to the amplitude of the fundamental wave in percentage by pulse condition. Here, when the ratio (f2+ f3+ f4)/f1 of the sum of the amplitudes of the 2 nd harmonic f2 to the 4 th harmonic f4 to the amplitude of the fundamental wave is considered, the smooth pulse is 1.74 and the flat pulse is 1.5. Therefore, the slippery pulse and the flat pulse can be discriminated from each other based on the value. When the ratio (f5+ f6+ f7)/f1 of the sum of the amplitudes of the 5 th harmonic f5 to the 7 th harmonic f7 to the amplitude of the fundamental wave is considered, the smooth pulse is 0.36, the flat pulse is 0.26, and the chordal pulse is 0.42, and therefore, it can be determined whether or not the harmonic is a chordal pulse from this value.

Constitution of pulse wave diagnostic apparatus (1.2)

The pulse wave diagnostic apparatus of the present embodiment is configured based on the above-described theory, and performs spectrum analysis on a pulse wave waveform detected from the body of a subject, extracts a tidal wave (tidewave) component and a dicrodic wave (dicrodic wave), and determines a pulse condition based on the extracted result. The external appearance of the pulse wave diagnostic apparatus will be described in the following section "external appearance of each of the above embodiments 10".

Fig. 1 is a block diagram showing a functional configuration of a pulse wave diagnostic device according to the present embodiment. In the figure, the pulse wave detecting unit 10 detects, for example, a pulse wave waveform of a distal portion (for example, a radial artery) of the subject, and outputs the detection signal to the motion removing unit 30 as MH.

On the other hand, the body motion detector 20 is configured by, for example, an acceleration sensor, detects the body motion of the subject, and outputs the detection signal to the waveform processor 21 as TH. The waveform processing unit 21 is configured by a low-pass filter or the like, and performs waveform shaping processing on the signal TH output from the body motion detecting unit 20 to output it as a signal MHt indicating a body motion component. The body motion removing unit 30 subtracts a signal MHt indicating a body motion component from the signal MH output from the pulse wave detecting unit 10, and outputs the subtracted signal as a signal MHj indicating a pulse wave component.

The pulse wave diagnostic apparatus of the present embodiment is an apparatus that processes a pulse wave waveform detected from the body of a subject, but when the subject performs any operation, a signal MHt indicating a body motion component of the subject is superimposed on a signal MH detected from the pulse wave detecting unit 10 in addition to a signal MHj indicating a pulse wave component. Therefore, the signal MH output from the pulse wave detection unit 10 cannot accurately reflect the pulse wave shape of the subject, when MH is MHt + MHj.

On the other hand, since the blood flow is influenced by blood vessels and tissues, the body motion component MHt included in the signal MH is not the signal TH itself indicating the body motion of the subject but a signal which is inactivated.

Therefore, the body motion component removing unit 30 performs waveform shaping on the signal TH directly representing the body motion of the subject output from the body motion detecting unit 20 by the waveform processing unit 21, uses the shaped signal as the signal MHt of the body motion component, subtracts the shaped signal from the signal MH output from the pulse wave detecting unit 10, removes the influence of the body motion, and outputs the shaped signal as the signal MHj representing the pulse wave component. The form, number of steps, constant, and the like of the low-pass filter in the waveform processing unit 21 are determined by actual measurement data.

However, if the body motion component removing unit 30 is operated to remove the body motion even without the body motion, the noise of the body motion detecting unit 20 decreases the signal-to-noise ratio of the output signal of the body motion component removing unit 30, and power is consumed by the body motion removing operation. Therefore, the present embodiment is provided with the determination section 22. The determination unit 22 generates a control signal C for determining the presence or absence of body motion from the body motion waveform TH. Specifically, the determination is made by comparing the body motion waveform TH with a threshold value. The threshold is set in advance so as to be able to determine the presence or absence of body motion, taking into account the noise level of the body motion detection unit 20. When the control signal C indicates no body motion, the operations of the waveform processing unit 21 and the body motion component removal unit 30 are stopped. At this time, the body motion component removing unit 30 outputs the pulse waveform MH as it is. Therefore, the signal-to-noise ratio of the output signal of the body motion component removing unit 30 can be improved, and the power consumption of the apparatus can be reduced.

Next, the FFT processing unit 40 performs FFT processing on the signal MHj representing the pulse component, and performs spectrum analysis of the pulse component. As a result of the analysis by the FFT processing, a plurality of spectral lines can be obtained, and the frequency and energy level of each spectral line can be calculated. The FFT processing unit 40 compares these spectrum data, and specifies the spectrum line with the highest energy level as the fundamental wave f1 of the pulse wave component. Each harmonic is specified by taking an integral multiple of the frequency of the fundamental wave f 1. Pulse wave analysis data MKD indicating the energy levels of the fundamental wave f1 and the 2 nd harmonic f2 to 10 th harmonic f10 is generated and output.

Next, the tidal wave characteristic extraction unit 50 generates tidal wave characteristic data TWD indicating the characteristics of tidal waves (tidal waves) from the pulse wave analysis data MKD. As described above, since the tidal wave (tidawave) characteristics can be expressed by the ratio of the 5 th harmonic f5 to 7 th harmonic f7 of the pulse waveform to the fundamental wave f1, the tidal wave characteristic extraction unit 50 generates the tidal wave characteristic data TWD according to the following equation.

TWD＝(f5+f6+f7)/f1

Next, the repeating wave characteristic extraction unit 60 generates repeating wave characteristic data DWD indicating a characteristic of a repeating wave (dicrotic wave) from the pulse wave analysis data MKD. As described above, since the dicrotic wave (dicrotic wave) characteristic can be expressed by the ratio of the sum of the harmonics 2 nd 2 to 4 th f4 of the pulse waveform to the fundamental wave f1, the dicrotic wave characteristic extraction unit 60 generates the tidal wave characteristic data DWD according to the following equation.

TWD＝(f2+f3+f4)/f1

Next, the pulse wave determination unit 70 determines a pulse condition based on the tidal wave feature data TWD and the hyperbola feature data DWD, and generates pulse condition data ZD indicating the type of the pulse condition of the subject. Specifically, first, the tidal wave feature data TWD is compared with the 1 st threshold value, and when the tidal wave feature data TWD is higher than the 1 st threshold value, the pulse condition data ZD1 indicating that it is a chordal pulse is generated. The 1 st threshold is set in advance so that it can be determined whether or not a chordal pulse is present, and is set to 0.41 in this example.

On the other hand, when the tidal wave feature data TWD is lower than the 1 st threshold, the rebubular feature data DWD is compared with the 2 nd threshold, and when the rebubular feature data DWD is lower than the 2 nd threshold, the pulse condition data ZD2 indicating that it is a flat pulse is generated. On the other hand, when the bob characteristic data DWD is higher than the 2 nd threshold, the pulse condition data ZD3 indicating that it is a smooth pulse is generated. Here, the 2 nd threshold is set in advance so that whether it is a flat pulse or a smooth pulse can be determined, and in this example, it is set to 1.62.

The notification unit 80 is a device for displaying the pulse condition data ZD or outputting the pulse condition data ZD to the outside by sound or the like, and displays characters corresponding to each pulse condition, such as 'smooth pulse', 'flat pulse', 'string pulse', or displays symbols such as icons. Therefore, the 3 rd person such as the subject and the doctor can know the pulse condition.

Operation of the pulse wave diagnostic apparatus is [ (1.3 ]

Next, the operation of the pulse wave diagnostic device according to embodiment 1 will be described with reference to fig. 2.

First, a body motion component accompanying the body motion of the subject is superimposed on the signal MH output from the pulse wave detection unit 10, and the body motion component is removed by the body motion component removal unit 30 to become a signal MHj indicating only the pulse wave component, which is supplied to the FFT processing unit 40 (steps S1, S2).

Next, the FFT processing unit 40 performs FFT processing on the signal MHj to generate the fundamental wave f1 of the pulse wave component and the harmonics f2 to f10 as the pulse wave analysis data MKD, and supplies the pulse wave analysis data MKD to the tidal wave characteristics extraction unit 50 and the rebleed characteristics extraction unit 60 (S3).

The tidal wave characteristic extraction unit 50 then calculates the ratio of the 5 th harmonic f5 to 7 th harmonic f7 of the pulse waveform representing the characteristics of the tidal wave to the fundamental wave f1, and generates tidal wave characteristic data TWD. The rebubular feature extraction unit 60 calculates the ratio of the sum of the 2 nd harmonic f2 to the 4 th harmonic f4 of the pulse waveform representing the feature of the feature wave to the fundamental wave f1, and generates the rebubular feature data DWD (S4).

In this way, when extracting the characteristics of the tidal wave (tidal wave) and the dicrotic wave (dicrotic wave) of the pulse waveform, the pulse condition determination unit 70 first compares the tidal wave characteristic data TWD with the 1 st threshold (S5). If the tidal wave feature data is higher than the 1 st threshold (0.41), the process proceeds to S6, and pulse profile data ZD1 representing the chordal pulse is generated.

On the other hand, if the tidal wave feature data is lower than the 1 st threshold (0.41) and the determination result at S5 is no, the process proceeds to S7, and the pulse condition is determined based on the bobble-wave feature data DWD. At this time, the pulse condition determination unit 70 determines whether or not the repetitive pulse feature data DWD is lower than the 2 nd threshold (1.62), and if so, proceeds to S8 to generate pulse condition data ZD2 indicating a flat pulse. On the other hand, if the rebubble feature data DWD is higher than the 2 nd threshold (1.62), the determination result of S7 goes to S9, and pulse data ZD3 indicating a smooth pulse is generated.

As described above, in embodiment 1, from the viewpoint of determining a pulse condition from the tidal wave and the hyperbaric wave of the pulse waveform, a medical professional skilled in pulse diagnosis can objectively and accurately determine the pulse condition by performing spectrum analysis on the pulse waveform and extracting the characteristics of the tidal wave and the hyperbaric wave, based on the fact that the characteristics of the tidal wave and the hyperbaric wave can be reflected on predetermined harmonics.

EXAMPLE 2

Next, a pulse wave diagnostic device according to embodiment 2 of the present invention will be described.

In the above-described embodiment 1, the pulse waveform is subjected to FFT processing based on the fact that the characteristic portions of the slippery pulse and the flat pulse are in the heavy wave (dicrotic wave) and the characteristic portions of the chordal pulse are in the tidal wave (tidal wave), frequency components corresponding to the heavy wave and the tidal wave are extracted from the analysis result, respectively, and the pulse condition is determined based on the extracted frequency components.

In the spectrum analysis by the FFT processing, when there is no data for a relatively long time, an analysis result cannot be obtained. Therefore, in general FFT processing, a pulse waveform of several cycles is often processed.

However, the dicrotic and tidal waves occur during a period of the pulse waveform. Therefore, by performing the spectrum analysis only for the period of the heavy pulse wave and the tidal wave in the pulse waveform, the accuracy of determining the pulse condition can be improved.

Therefore, the 2 nd embodiment specifies a pulse condition by using a wavelet transform in which a spectrum analysis and a time domain analysis are simultaneously performed on a pulse waveform.

2.1 general constitution of pulse wave diagnostic apparatus

Fig. 3 shows a configuration of a pulse wave diagnostic apparatus according to embodiment 2. The pulse wave diagnostic apparatus shown in fig. 3 is the same as the pulse wave diagnostic apparatus of embodiment 1 shown in fig. 1, except that a wavelet transform unit 41 is used instead of the FFT processing unit 40, and the internal configurations of the tidal wave specifying the time positions of the tidal wave and the rebleed wave, the rebleed wave detection unit 42, the tidal wave characteristic extraction unit 50, the rebleed characteristic extraction unit 60, and the pulse condition determination unit 70 are different. The following describes different points.

2.2 section of wavelet transform

First, the configuration of the wavelet transform unit 41 will be described in detail together with the drawings.

Generally, in time-frequency analysis for simultaneously capturing a signal in both time and frequency, a wavelet is a unit for dividing a signal into respective parts. The wavelet transform indicates the size of each part of a signal divided in such units. To define the wavelet transform, a function ψ (x) localized in both time and frequency is introduced as a mother wavelet as a basic function. Here, the wavelet transform of mother wavelet ψ (x) of function f (x) is defined as follows.

img id="idf0002" file="C9880351600271.GIF" wi="516" he="55" img-content="drawing" img-format="GIF"/

In equation (2), b is a parameter used when the mother wavelet ψ (x) is shifted (parallel shift), and a is a parameter when it is stretched. Therefore, in equation (2), the wavelet ψ ((x-b)/a) is a result of shifting the mother wavelet ψ (x) in parallel by b and stretching a. In this case, the amplitude of the mother wavelet ψ (x) is elongated in accordance with the scale parameter a, and therefore 1/a is an amount corresponding to the frequency. The wavelet transform unit 41 is configured to be able to perform the calculation of the formula (2), and its detailed configuration is as shown in fig. 4.

In fig. 4, the signal MHj output from the body motion component removing unit 30 is supplied to the waveform shaping unit 400 and the a/D conversion unit 410. The waveform shaping unit 400 generates the control signal CS and the clock signal CK in synchronization with the pulse waveform MHj.

Here, fig. 5 shows a block diagram of the waveform shaping unit 400. In fig. 5, ringing filter 401 is a high-Q filter having a center frequency of 2.2Hz and a passband of 0.8Hz to 3.5 Hz. Since the fundamental wave component of the pulse waveform is usually in the range of 0.8Hz to 3.5Hz, the fundamental wave component is extracted after the pulse waveform MHj passes through the ringing filter 401. For example, as shown in fig. 6, when the pulse waveform MHj is input to the ringing filter 401, the output thereof becomes a waveform shown as 401out in fig. 6.

Next, the zero-cross detection circuit 402 is configured by a comparator or the like, and generates a rectangular wave by comparing the signal output from the ringing filter 401 with the ground level. The rectangular wave is a signal synchronized with the pulse. For example, if the output signal of the ringing filter 401 is a signal indicated by 401out in fig. 6, the output signal of the zero-cross detection circuit 402 becomes a signal indicated by 402out in the figure.

Next, the comparison unit 403, the loop filter 404, the voltage-controlled oscillator circuit 405, and the frequency divider circuit 406 constitute a phase-locked loop. When one input terminal of the comparison section 403 is supplied with the output signal of the zero-cross detection circuit 402 and the other terminal is supplied with the output signal of the frequency dividing circuit 406, the comparison section 403 outputs an error signal corresponding to the phase difference between the two. When the error signal is supplied to the vco 405 through the loop filter 404, the vco 405 outputs the clock signal CK. The clock signal CK is divided by 1/N by the frequency dividing circuit 406 and fed back to the other input terminal of the comparison section. For example, if the frequency division ratio is 1/8, the frequency of the clock signal CK is 8 times the frequency of the output signal of the zero-cross detection circuit 402, as shown by 405out in fig. 6. Then, the clock signal CK is frequency-divided 1/2 by the frequency dividing circuit 407 and is output as the control signal CS shown in fig. 6.

The pulse waveform MHj is converted into a digital signal by the a/D converter 410 shown in fig. 4 and then stored in the 1 st memory 420 and the 2 nd memory 430. Here, the write enable terminal of the 1 st memory 420 directly supplies the control signal CS, and the write enable terminal of the 2 nd memory 430 supplies the control signal CS inverted by the inverter 440. Therefore, the 1 st and 2 nd memories 420 and 430 alternately store the pulse waveform in units of clock cycles.

Further, reference numeral 450 denotes a multiplexer which selectively outputs the pulse wave data MD alternately read from the 1 st and 2 nd memories 420 and 430 to the basic function development unit W. In this way, the pulse data MD is read from the 2 nd memory 430 during the write period of the 1 st memory 420, and the pulse data MD is written into the 2 nd memory 430 during the read period of the 1 st memory 420.

Next, the basic function expansion unit W is configured to perform the calculation of the above equation (2), and performs the calculation in a clock cycle when the clock signal CK is supplied. The basic function expansion unit W is composed of a basic function storage unit W1 that stores the mother wavelet ψ (x), a scale conversion unit W2 that converts the scale parameter a, a buffer memory W3, a parallel shift unit W4 that performs shifting, and a multiplication unit W5. Note that, as the mother wavelet ψ (x) stored in the basic function storage unit W1, not only Gabal wavelets but also mexican hat wavelets, Haar wavelets, Meyer wavelets, Shannon wavelets, and the like can be used, and in this example, a mother wavelet called symlets5 is used, and the characteristics of the pulse waveform can be extracted favorably.

First, when the mother wavelet ψ (x) is read from the basic function storage unit W1, the scale transformation unit W2 transforms the scale parameter a. Here, the scale parameter a is a parameter corresponding to the period, and therefore, when a increases, the mother wavelet ψ (x) extends on the time axis. In this case, since the data amount of the mother wavelet ψ (x) stored in the basic function storage unit W1 is constant, the data amount per unit time decreases as a increases. To compensate for this, the scaling unit W2 performs interpolation processing and thinning processing to generate the function ψ (x/a) when a becomes small. The data is temporarily stored in the buffer memory W3.

Next, the parallel shift unit W4 reads out the function ψ (x/a) from the buffer W3 at a timing corresponding to the shift parameter B, and generates the function ψ ((x-B)/a) after the function ψ (x/a) is shifted in parallel.

Next, the multiplication unit W4 multiplies the variable 1/a by the value^1/2The function ψ ((x-b)/a) and the pulse wave data MD are multiplied, and then wavelet transform is performed in units of heart beats to generate pulse wave analysis data MKD. In this example, the pulse wave analysis data MKD is outputted after being divided into frequency regions of, for example, 0Hz to 0.5Hz, 0.5Hz to 1.0Hz, 1.0Hz to 1.5Hz, 1.5Hz to 2.0Hz, 2.0Hz to 2.5Hz, 2.5Hz to 3.0Hz, 3.0Hz to 3.5Hz, and 3.5Hz to 4.0 Hz.

2.3 department of examination of tidal wave and Bobo wave

The tide/dicrotic wave detector 42 compares a certain frequency range in which the pulse wave analysis data MKD exists with a threshold value, and generates control signals ct and cd for specifying the time positions of the tide wave (tidal wave) and the dicrotic wave (dicrotic wave), respectively.

For example, in the figures7 in the pulse waveform MHj, t_WIs the tidal wave, d_WIs a dicrotic wave. When wavelet transform is performed on the pulse waveform MHj, a wavelet analysis result shown in fig. 7 can be obtained. The analysis result shows the values of the pulse wave analysis data MKD in a shaded form, and the darker the value of the pulse wave analysis data MKD, the lighter the value of the pulse wave analysis data MKD, the smaller the value of the pulse wave analysis data MKD.

As is clear from this figure, in the time interval corresponding to the tidal wave (tidal wave), there are an area a and an area B surrounded by a portion (white) having a small value of the pulse wave analysis data MKD. The small value of the pulse analysis data MKD in the Y-axis direction indicating the frequency indicates that the energy of the pulse waveform MHj is low during this time, and the partial pulse waveform MHj becomes flat. For example, the boundary between area A and area B and the tidal wave t_WThe peak points of (a) and (b) are consistent.

Therefore, when a certain frequency region is focused, the tidal wave t can be detected based on the change of the frequency region_WThe time position of (c). In this example, the specific tidal wave t shown in fig. 7 is generated by comparing the pulse wave analysis data MKD in the frequency range X with a threshold value_WAnd (c) control signals ct at time positions t 1-t 2. On the other hand, the specific heartbeat wave d as shown in fig. 7 is generated by comparing the pulse wave analysis data MKD in the frequency range X with the threshold value_WThe control signal cd at the time positions t3 to t 4.

2.4 tidal wave characteristics extraction section and Chogbaud characteristics extraction section

First, the tidal wave feature extraction unit 50 specifies the time position of the tidal wave based on the control signal ct, and adds data in a specified frequency range to the pulse wave analysis data MKD during the specified time period.

For example, as described in other embodiment (6) of 14 described later shown in fig. 58, the wavelet transform unit 41 is configured by a filter bank, and the characteristics of the high-pass filter 1A and the low-pass filter 1B are changed by a clock signal CK synchronized with the pulse waveform MHj. At this time, when the pulse wave analysis data corresponding to the fundamental wave f1 of the pulse waveform MHj is M × 1, the characteristic part of the tidal wave is represented by the 5 th harmonic f5 to 7 th harmonic f7, and therefore, the tidal wave characteristic extraction data TWD is calculated by adding M × 5, M × 6, and M × 7.

Next, the beat wave feature extraction unit 60 specifies the time position of the beat wave based on the control signal cd, and adds data in a specified frequency region from the pulse wave analysis data MKD during the specified time period.

For example, as described in the tidal wave feature extraction unit 50, when the pulse wave analysis data corresponding to the fundamental wave f1 of the pulse wave waveform MHj is M × 1, the characteristic portion of the dicrotic wave is expressed by the 2 nd harmonic f2 to 4 th harmonic f 4. Therefore, the repetitive beat feature extraction data DWD is calculated by adding M × 2, M × 3, and M × 4.

As described above, the tidal wave feature extraction unit 50 and the rebubular feature extraction unit 60 according to embodiment 2 extract features of a tidal wave (tidal wave) and a rebubular wave (dicrotic wave) not only from the frequency domain but also from the time domain because of the property of performing waveform analysis by skillfully utilizing the divided frequency domain and time domain of wavelet transform, and therefore, can extract features with high accuracy.

2.5 pulse condition determining section and informing section

The pulse condition determination unit 70 compares the tidal wave feature extraction data TWD and the bode feature extraction data DWD with a threshold value to determine the pulse conditions of the smooth pulse, the flat pulse, and the chordal pulse, and generates the pulse condition data ZD. The notification unit 80 also notifies the 3 rd subject, doctor, etc. by displaying the pulse data ZD, sound, etc. as in embodiment 1.

As described above, according to the pulse wave diagnostic apparatus of embodiment 2, the time of pulse wave analysis can be limited to the period during which tidal waves (tidal waves) and dicrotic waves (dicrotic waves) exist, and spectral analysis can be performed by performing wavelet transform on the pulse wave waveform detected from the subject. As a result, the characteristics of tidal wave (tidal wave) and dicrotic wave (dicrotic wave) can be extracted with high accuracy, and accurate pulse condition determination can be performed.

3 EXAMPLE 3

3.1 principle of embodiment 3

The pulse waveform indicates the pulsation when the blood flow sent to the aorta by the contraction of the heart is propagated through the artery, and therefore has a constant cycle synchronized with the pulsation of the heart. As an analysis method for analyzing a periodic waveform and extracting its characteristic, there is a method such as a spectrum analysis typified by FFT (fast fourier transform), but the present inventors have also noticed an autocorrelation function that can be processed by a simple operation.

Here, the irregular variation is represented by x (T), and if there is a periodic variation of the period T in x (T), x (T) can be given by the following equation.

x (t) ═ x (t ± nT) where n is 0, 1, 2, …

That is, if the period is shifted by an integral multiple of the period, the waveform overlaps the original waveform. If the periodicity of the irregular fluctuation x (t) is strong, the waveform is similar to the original waveform as long as the period is shifted by an integral multiple of the period on the time axis. Therefore, it is only necessary to examine how much the waveform shifted by a certain time τ is different from the original waveform, and to discriminate the cycle component in the shift, the correlation between x (t) and x (t + τ) is required.

When x (t) is a time-dependent irregularity, the autocorrelation function can be defined as the average of the product of two variables separated by time τ, as given by the following equation.

C(τ)＝E[x(t)x(t+τ)]

Here, while E is the ensemble average, it may be replaced with a time average in the normal establishment process. Therefore, the autocorrelation function C (τ) can be expressed by the following equation.

img id="idf0003" file="C9880351600311.GIF" wi="604" he="52" img-content="drawing" img-format="GIF"/

Further, an autocorrelation function normalized by τ to 0 with respect to the autocorrelation function C (τ) is referred to as an autocorrelation coefficient R (τ), and the autocorrelation coefficient can be given by the following equation.

R(τ)＝C(τ)/C(0)

＝E[x(t)x(t+τ)]/E[x²(t)]

Here, fig. 16 is the autocorrelation coefficients of the chordal arteries shown in fig. 45C, fig. 17 is the autocorrelation coefficients of the flat arteries shown in fig. 45A, and fig. 18 is the autocorrelation coefficients of the smooth arteries shown in fig. 45B. When these figures are compared, it can be seen that the autocorrelation coefficient R (τ) fluctuates with the change of the chordal pulse → the flat pulse → the slippery pulse. Specifically, the chordal pulse is a smooth waveform, the equatorial pulse has three peaks, the size of the peak is small, and the synovial pulse is composed of two peaks. That is, the autocorrelation coefficient R (τ) reflects the characteristics of the waveform of each pulse wave. Further, since the waveform shape of the autocorrelation coefficient R (τ) is obtained from the pulse waveform for a certain long time, the characteristics of the pulse waveform can be extracted with high accuracy.

Further, since the instantaneous value of the autocorrelation coefficient R (τ) is different from the instantaneous value of the pulse waveform and is obtained by time averaging, even if the signal-to-noise ratio of the pulse waveform detected from the body is poor, the signal-to-noise ratio can be improved by averaging the noise. Further, since the autocorrelation coefficient R (τ) is obtained by normalizing the autocorrelation function, it is effective to compare pulse waveforms having different analysis amplitude values.

For the above reasons, the present inventors have invented a pulse wave diagnostic apparatus for determining a physical condition by skillfully utilizing the properties of the autocorrelation coefficient R (τ) to specify a pulse wave.

Circuit configuration of 3.2 pulse wave diagnostic apparatus

Next, a circuit configuration of the pulse wave diagnostic apparatus will be described with reference to fig. 19. Fig. 19 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus.

The pulse wave diagnostic device 1 is roughly composed of a pulse wave detection unit 10 that detects a pulse wave MH, an autocorrelation calculation unit 210 that calculates autocorrelation data RD based on the pulse wave MH, a pulse data generation unit 220 that generates pulse data ZD based on the autocorrelation data RD, and a display unit 230.

First, the autocorrelation function calculation unit 210 is composed of a waveform processing unit 211, a memory 212, a multiplication unit 213, an average value calculation unit 214, and a normalization unit 215. The memory 212 stores at least a pulse waveform MH of 1 heartbeat period. However, if the autocorrelation function C (τ) is calculated from the above equation (3), x (T + τ) must be integrated from-T/2 to T/2 (T is infinite), and since x (T) is a pulse waveform MH synchronized with the heart cycle in this example, it is sufficient to perform calculation based on at least 1 cycle of the pulse waveform MH. Therefore, in this example, the pulse waveform MH is shaped by the waveform processing section 211 to become a square wave synchronized with the pulse period, and the write control signal WE for controlling the write operation of the memory 212 is generated from this square wave. For example, the write control signal WE writes the pulse waveform MH of 4 cycles into the memory 212. Next, the multiplication unit 213 reads out the pulse waveform MH1 corresponding to x (t) and the pulse waveform MH2 corresponding to x (t + τ), multiplies them together, and outputs the multiplication result. Then, the average value calculation unit 214 calculates an overall average MH1 · MH2 corresponding to the overall average of x (t) x (t + τ), and outputs the calculation result.

Next, the normalization operation unit 215 normalizes the calculation result of the average value calculation unit by the autocorrelation function C (0) to generate autocorrelation data RD. Utensil for cleaning buttockIn general, by x²(t), namely MH1²The ensemble average of (a) is normalized. Therefore, the autocorrelation data RD indicates the autocorrelation coefficient R (τ) of the pulse wave shape MH.

Here, the autocorrelation coefficient R (τ) of the pulse waveform MH of each pulse condition shown in fig. 45A to 45 is examined. First, the minimum value of the autocorrelation coefficient R (τ) of the chordal pulse shown in fig. 16 is 0.2, the minimum value of the autocorrelation coefficient R (τ) of the flat pulse shown in fig. 17 is 0.32, and the minimum value of the autocorrelation coefficient R (τ) of the smooth pulse shown in fig. 1 is 0.3. Therefore, when the minimum value of the autocorrelation coefficient is lower than 0.25, it can be determined that it is a chordal pulse.

Next, the proportion of the autocorrelation coefficient R (τ) to the 1-heart period over a period of 0.5 or more is 60% for the flat pulse (see fig. 17) and 34% for the smooth pulse (see fig. 18). Therefore, it can be determined that more than 47% of the cases are smooth pulses and less than 47% are smooth pulses.

The pulse condition data generating unit 220 determines a pulse condition based on the determination criterion, and includes a minimum value detecting unit 221 and a 1 st comparing unit 222 for determining whether the pulse is a chordal pulse, a time measuring unit 223 for determining whether the pulse is a flat pulse, an arithmetic unit 224 and a 2 nd comparing unit 225, and a data generating unit 226 for generating pulse condition data ZD.

First, the minimum value detection unit 221 detects the minimum value of the autocorrelation data RD during a period corresponding to 1 heart cycle. The 1 st comparing section 222 determines whether or not the minimum value is less than 0.25, and outputs the result to the data generating section 226.

Next, the time measuring section 223 compares the autocorrelation data RD with a predetermined threshold value (0.5), and measures the time interval in which the autocorrelation data RD is higher than the threshold value (0.5). The calculation unit 224 calculates the ratio of the measured time interval to the 1-heart cycle. The 1-heart cycle is supplied from the waveform processing unit 211. The 2 nd comparing unit 225 determines whether or not the calculation result exceeds a predetermined threshold value (47%).

Next, the data generation unit 226 generates the pulse condition data ZD based on the determination results of the 1 st comparison unit 222 and the 2 nd comparison unit 225. First, when the 1 st comparing unit 222 judges that the value is less than 0.25, the data ZD1 indicating a string pulse is generated. When the 2 nd comparing unit 255 determines that the proportion of the time during which the autocorrelation data RD exceeds 0.5 exceeds 47%, pulse data ZD2 indicating a flat pulse is generated. When the proportion of the time is judged to be less than 47%, pulse condition data ZD3 representing the smooth pulse is generated.

Next, the display unit 230 is composed of a ROM, a control circuit, and a liquid crystal display device. When the pulse data ZD is supplied to the display section 230, the control circuit detects this, reads out the characters stored in the ROM, and displays them on the liquid crystal display. Characters used may be characters such as 'Pingmai', 'chordai', and 'Royal', or may be specific symbols or icons. This enables the user or doctor to be informed of the health condition.

In this way, the function of the pulse wave diagnostic device is configured, but in an actual device, the autocorrelation calculating unit 210 and the pulse wave data generating unit 220 are configured by a CPU, a memory, and the like. At this time, the CPU executes various arithmetic processing and comparison processing according to a control program stored in a part of the ROM and generates the pulse condition data ZD.

3.3 actions of pulse wave diagnostic apparatus

Next, the operation of the pulse wave diagnostic device according to embodiment 3 will be described with reference to the drawings. Fig. 20 is a flowchart showing the operation of the pulse wave diagnostic apparatus.

First, the pulse wave detection unit 10 detects a pulse wave MH (S1).

When the pulse waveform MH is supplied to the autocorrelation calculating section 210, the pulse waveform MH is written into the memory 212, delayed by a predetermined time, and then read. The autocorrelation function is calculated based on the read pulse waveform MH, and autocorrelation data RD indicating an autocorrelation coefficient R (τ) is generated by normalizing the calculation result (S2).

Then, the pulse data generating unit 220 generates the pulse data ZD from the autocorrelation data RD as described below. First, when the minimum value detection unit 221 detects the minimum value of the autocorrelation data RD corresponding to the 1-heart beat period (S3), the 1 st comparison unit 222 determines whether or not the minimum value is less than 0.25 (S4). If the minimum value is less than 0.25, the process proceeds to S5, where the result of determination is yes, and the data generator 226 determines a chordal pulse and generates pulse data ZD 1.

If the minimum value is higher than 0.25, it is determined whether the result is a smooth pulse or a flat pulse. At this time, the time measuring unit 223 measures the time interval in which the autocorrelation data RD exceeds 0.5 during the period corresponding to 1 heartbeat period (S6), and the arithmetic unit 24 calculates the ratio of the measured time interval to the 1 heartbeat period.

Then, the 2 nd comparing unit 225 determines whether or not the calculation result of the calculating unit 224 exceeds 47% (S8), and if it exceeds 47%, the routine proceeds to S9, where it is determined that the pulse is flat, and the data generating unit 226 generates pulse data ZD 2. If not more than 47%, the flow proceeds to S10, where it is determined as a slippery pulse, and the data generator 226 generates pulse data ZD 3.

Thus, according to the pulse wave diagnostic device 1 of embodiment 3, since the pulse condition is determined based on the autocorrelation data RD of the pulse wave form MH, the subject and the doctor can understand the physical condition of the pulse wave from the form of the pulse wave without knowledge of the pulse diagnosis.

3.4 modification of embodiment 3

(1) In the above-described embodiment 3, the minimum value detecting unit 221 calculates the autocorrelation data RD during 1 heartbeat period, but may detect an average minimum value by averaging the respective minimum values of the autocorrelation data RD detected during a plurality of heartbeat periods. In this case, since the minimum value is an averaged value, it is possible to suppress variation in the minimum value due to noise, and it is possible to improve the accuracy of determining whether the pulse is a chordal pulse or not.

(2) In the above-described embodiment 3, the arithmetic unit 224 calculates the ratio of the measured time interval in the 1-heart cycle, but may average the ratios detected in a plurality of heart cycles and output the average. In this case, since the ratio is an averaged value, it is possible to suppress the fluctuation due to noise, and therefore, it is possible to improve the accuracy of determining whether the pulse is flat or smooth.

(3) When the autocorrelation coefficients R (τ) of typical flat and smooth pulses are compared (see fig. 17 and 18), it is found that the values of the autocorrelation coefficients R (τ) are different in the width on the time axis from the autocorrelation coefficient R (τ) in the range of 0.4 to 0.8. Therefore, the time measuring section 223 may measure a time interval higher than any threshold value specified in the range of 0.4 to 0.8, and the 2 nd comparing section may determine whether it is a flat pulse or a smooth pulse using a value corresponding to the threshold value.

4 EXAMPLE 4

Next, a pulse wave diagnosis device according to embodiment 4 will be described.

Circuit configuration of 4.1 pulse wave diagnostic apparatus

The circuit configuration of the pulse wave diagnostic apparatus according to embodiment 4 is the same as that of embodiment 3 except for the configuration of the pulse wave data generating section 220.

Next, the circuit configuration of the pulse data generating unit 220 according to embodiment 4 will be described. Fig. 21 is a block diagram of a pulse data generating unit 220 according to embodiment 4.

The pulse condition data generating unit 220 of the present example includes a minimum value detecting unit 221 and a 1 st comparing unit 222 for determining whether the pulse is a chordal pulse, a change rate calculating unit 227 for determining whether the pulse is a flat pulse or a smooth pulse, a maximum value detecting unit 228 and a 2 nd comparing unit 225, and a data generating unit 226 for generating pulse condition data ZD based on the determination result.

First, the 1 st comparing unit 222 determines whether the pulse condition is a chordal pulse by determining whether or not the minimum value of the autocorrelation data RD detected by the minimum value detecting unit 221 exceeds 0.25, as in embodiment 3.

Next, the change rate calculation unit 227 calculates the change rate RDd of the autocorrelation data RD. For example, if the autocorrelation data RD is as shown in fig. 16 to 18, the change rates RDd of the chordal, the platic and the slippery pulses are as shown in fig. 22. From this figure, the maximum value of the rate of change of the smooth pulse RDd is approximately 0.1, and the maximum value of the flat pulse is 0.072. This corresponds to the pulse waveform MH of the flat pulse and the pulse waveform MH of the smooth pulse having a larger fluctuation, and whether the pulse is the flat pulse or the smooth pulse can be determined from the maximum value of the change rate RDd. Therefore, when the maximum value of the change rate RDd exceeds 0.085, it can be determined as a smooth pulse, and when not, it can be determined as a flat pulse.

Next, the maximum value detection unit 228 detects the maximum value of the change rate RDd for a predetermined time period that is longer than at least 1 heartbeat period. The 2 nd comparing unit 225 compares the maximum value of the change rate with a predetermined threshold value (0.85), and determines whether or not the maximum value of the change rate RDd exceeds the threshold value.

Next, the data generation unit 226 generates the pulse condition data ZD based on the determination results of the 1 st comparison unit 222 and the 2 nd comparison unit 225. First, when the 1 st comparing unit 222 determines that the minimum value is less than 0.25, pulse condition data ZD1 indicating a chordal pulse is generated. The 2 nd comparing unit 225 generates pulse data ZD2 indicating a flat pulse when the maximum value of the change rate RDd is judged to be lower than 0.085, and generates pulse data ZD3 indicating a smooth pulse when the maximum value is judged to be higher than 0.085.

4.2 actions of pulse wave diagnostic apparatus

Next, the operation of the pulse wave diagnostic device according to embodiment 4 will be described with reference to the drawings. Fig. 23 is a flowchart showing the operation of the pulse wave diagnostic apparatus. The operations from S1 to S5 are the same as those of the pulse wave diagnostic device according to embodiment 3 described with reference to fig. 20, and therefore the description thereof is omitted here.

First, in S11, when the change rate calculation unit 227 detects the change rate RDd of the autocorrelation data RD, the maximum value detection unit 228 detects the maximum value of the intra-period change rate RDd corresponding to the 1-heart period (S12). For example, if the calculated rate of change RDd is a smooth pulse as shown in fig. 22, the maximum value detected is approximately 0.1.

Next, the 2 nd comparing unit 225 determines whether or not the detected maximum value exceeds 0.85, and if it is less than 0.85, the routine proceeds to S14, where it is determined that the pulse is flat, and the data generating unit 226 generates pulse data ZD 2. If the value is higher than 0.85, the routine proceeds to S15, where it is determined as a pulse, and the data generator 226 generates pulse data ZD 3.

In this way, in embodiment 4, since the maximum value of the change rate RDd of the autocorrelation data RD of the level pulse and the smooth pulse is different and the maximum value of the change rate RDd is compared with the threshold value to determine whether the level pulse or the smooth pulse is present, even a person who has no knowledge of pulse diagnosis can know the correct pulse condition.

5 EXAMPLE 5

In the pulse wave diagnostic apparatus 1 according to the above-described embodiments 3 and 4, the autocorrelation data RD is generated based on the pulse wave form MH output from the pulse wave detection unit 10, and the pulse condition is determined. However, when the subject walks or performs daily activities, the blood flow fluctuates due to body movements. Therefore, the body motion component is superimposed on the pulse wave MH output from the pulse wave detection unit 10. Since the magnitude of the body motion component depends on the degree of the movement of the subject, the body motion component superimposed on the pulse wave MH increases when the amount of movement of the subject is large, and it is difficult to accurately determine the pulse condition. Therefore, in embodiment 5, the body motion component is removed from the pulse waveform MH, and the pulse condition is determined based on this.

Circuit configuration of pulse wave diagnosis apparatus (5.1)

Fig. 24 is a block diagram showing a circuit configuration of the pulse wave diagnostic apparatus according to embodiment 5. The autocorrelation unit 210, the pulse data generation unit 220, and the display unit 230 in this example have the same configurations as those described in embodiment 3 or 4, and therefore, the description thereof is omitted here. Note that the body motion component removing unit 30 and its preceding components, that is, the pulse wave detecting unit 10, the body motion detecting unit 20, the determining unit 22, and the waveform processing unit 21, are the same as those described in embodiment 1, and therefore, the description thereof will be omitted.

Since the body motion-removed pulse waveform MHj is generated by the above configuration, the autocorrelation calculating unit 10 can generate the autocorrelation data RD from the body motion-removed pulse waveform MHj. Therefore, according to the pulse wave diagnostic apparatus 1, the pulse condition can be specified without being affected by body motion.

6 EXAMPLE 6

Next, a pulse wave diagnosis device according to embodiment 6 will be described.

Fig. 25 is a block diagram of the pulse wave diagnostic apparatus 1 according to embodiment 6. The body motion component TH is detected by the body motion detecting unit 20 and the waveform processing unit 21 in embodiment 6 as in embodiment 5, but is different from the body motion removal by the wavelet transform described in embodiment 5.

6.1 1 st and 2 nd wavelet transform units and 1 st and 2 nd frequency correction units

In fig. 25, the 1 st wavelet transform unit 243 performs wavelet transform on the pulse wave waveform MH output from the pulse wave detection unit 10, and generates pulse wave analysis data MKD. The 2 nd wavelet transforming section 245 performs wavelet transformation on the pulse wave waveform MH output from the pulse wave detecting section 10, and generates pulse wave analysis data TKD. Note that the 1 st and 2 nd wavelet transform units 243 and 245 are the same as those described in embodiment 2.

Fig. 26 is a graph showing pulse wave analysis data MKD of the pulse waveform MH in part of time. In the figure, the period T is near the peak P4, and the pulse wave analysis data MKD is obtained at time intervals obtained by dividing the period T into 8 parts. However, in the wavelet transform, the frequency resolution and the time resolution have a trade-off relationship, and therefore, pulse wave analysis data can be obtained at a relatively short time interval at the expense of the frequency resolution.

Next, the 1 st frequency correction unit 244 performs frequency correction on the pulse wave analysis data MKD. In the above formula (2), 1/a is present^1/2When comparing data of different frequency intervals, it is necessary to correct the influence of the term. The 1 st frequency correction section 244 is provided for this purpose, and multiplies the wavelet data WD by the coefficient a^1/2Pulse correction data MKDa is generated. This makes it possible to perform correction in accordance with each frequency, thereby making the power density per unit frequency constant. The 2 nd frequency corrector 246 performs frequency correction to generate body motion correction data from the body motion analysis data TKD, as in the 1 st frequency corrector 244TKDa。

6.2 department of removal of body movement component

The body motion component removing unit 240 subtracts the body motion correction data TKDa from the pulse wave correction data MKDa to generate body motion removed pulse wave data MKDaj. This point will be specifically explained below. In the following description, it is assumed that the subject holds the cup by hand and returns the cup to its original position. At this time, the pulse wave MH shown in fig. 27 is detected by the pulse wave detecting unit 10, and the body motion waveform TH is detected by the body motion detecting unit 20.

Here, the body motion waveform TH increases from time T1, reaches a peak at time T2, then gradually decreases, passes through the 0 level at time T3, reaches a negative peak at time T4, and returns to the 0 level again at time T5. However, since the body motion waveform TH is detected by the body motion detector 20 using an acceleration sensor, the time T3 corresponds to the time when the subject lifts the cup to the maximum, the time T1 corresponds to the time when the lift starts, and the time T4 corresponds to the time when the cup is put down. Therefore, the period from the time T1 to the time T4 is a period in which a body motion exists. The pulse waveform MHj of fig. 27 is assumed to be a pulse waveform when there is no body motion. In this example, the fundamental wave frequency of the pulse waveform is 1.3 Hz.

Here, the pulse wave correction data MKDa of the period Tc (see fig. 27) is shown in fig. 28, and the body motion correction data TKDa of the period Tc is shown in fig. 29. As seen from this figure, in the body motion waveform TH, frequency components having a large level exist in a frequency band of 0.0Hz to 1.0 Hz. When the pulse wave correction data MKDa and the body motion correction data TKDa are supplied to the body motion component removing unit 240, the body motion component removing unit 240 subtracts the body motion correction data TKDa from the pulse wave correction data MKDa, and generates body motion-component-removed pulse wave data MKDaj shown in fig. 30. This eliminates the influence of body motion even when the body motion exists.

Section for determination 6.3

The determination unit 22 compares the body motion waveform TH with a predetermined threshold value, generates a control signal C indicating the presence or absence of body motion, and supplies the control signal C to the waveform processing unit 21, the 2 nd wavelet transform unit 245, and the 2 nd frequency correction unit 246. Therefore, when there is no body motion, the waveform processing unit 21, the 2 nd wavelet transform unit 245, and the 2 nd frequency correction unit 246 stop operating, and it is possible to achieve reduction in the arithmetic processing time, reduction in the power consumption, and improvement in the signal-to-noise ratio.

6.4 department of inverse wavelet transform

The inverse wavelet transform unit 247 performs inverse wavelet transform on the body motion removed pulse wave data MKDaj to generate a body motion removed pulse wave waveform MHj. In this case, the inverse wavelet transform unit 247 performs inverse wavelet transform shown in equation (4) and then re-synthesizes waveforms.

As in embodiment 5, the autocorrelation calculating unit 210 generates autocorrelation data RD based on the body motion removal pulse waveform MHj.

Thereafter, when the pulse condition data generation unit 220 generates the pulse condition data ZD from the autocorrelation data RD, the display unit 230 displays characters (chordal, flat, and smooth) representing the pulse condition data ZD or symbols corresponding to each pulse condition. Therefore, the 3 rd person such as the subject and the doctor can recognize the pulse condition.

As described above, according to embodiment 6, since the autocorrelation data RD is generated based on the body motion removal pulse waveform MHj from which the body motion has been removed, the subject can continuously detect the pulse condition even in daily life.

7 EXAMPLE 7

In embodiments 5 and 6, the body motion waveform TH is detected by the body motion detecting unit 20, the pulse waveform TH is compared with the body motion waveform TH, the body motion component included in the frequency component of the pulse waveform MH is eliminated, and the autocorrelation function RD is calculated, thereby specifying the pulse condition. However, the body motion detection unit 20 and the waveform processing unit 21 are required, and therefore the configuration is complicated. Embodiment 7 is made in view of this point, and provides a pulse wave diagnostic device capable of accurately diagnosing a pulse condition with a simple configuration even in the case of a body motion.

Fig. 31 is a block diagram of the pulse wave diagnostic device of embodiment 7, and is the same as the pulse wave diagnostic device 1 of embodiment 7 shown in fig. 25 except that the internal configuration of the body motion detecting unit 20, the waveform processing unit 21, the 2 nd wavelet transform unit 245 and the 2 nd frequency correcting unit 246 are omitted, and the internal configuration of the body motion component removing unit 240 is omitted. Hereinafter, the difference will be described.

The body motion component removing unit 240 separates and removes body motion components from the pulse correction data MKDa to generate body motion removed pulse wave data MKDaj. Here, the body motion removing unit 240 utilizes the following properties of body motion.

Body movement is caused by up-and-down movement of the wrist and swinging of the wrist during walking, and in daily life, a human body hardly moves instantaneously. Therefore, in daily life, the frequency of the body motion waveform TH is not so high, and is usually in the range of 0Hz to 1 Hz. In this case, the fundamental frequency of the pulse waveform MH is mostly in the range of 1Hz to 2 Hz. Therefore, in daily life, the frequency of the body motion waveform TH is in a frequency region lower than the fundamental frequency of the pulse waveform MH.

On the other hand, in a light exercise, the frequency of the body motion waveform TH is increased mainly by the influence of the wrist swing, but the fundamental frequency of the pulse wave MH is also increased because the heart rate increases with the amount of exercise. Therefore, even during the movement, the frequency of the body motion waveform TH is usually in a frequency range lower than the fundamental frequency of the pulse waveform MH.

The body motion component removing section 240 separates the body motion components based on this point, and filters out a frequency range lower than the fundamental wave component of the pulse waveform MH. At this time, if a body motion component exists in a frequency region higher than the fundamental wave component of the pulse waveform MH, the detection accuracy of the cardiac function is lowered. However, since the probability that the body motion component is in a frequency region lower than the fundamental wave component of the pulse wave MH is high, the state of the pulse condition can be diagnosed with high accuracy.

Fig. 32 is a block diagram of the body motion component removing unit 240. The waveform shaping unit 301 performs waveform shaping on the pulse waveform MH to generate a reset pulse synchronized with the pulse waveform MH. The counter 302 counts clock pulses, not shown, and resets the count value by the reset pulse. The average value calculation circuit 303 calculates an average value of the count values of the counter 302. In this case, the average value calculated by the average value calculation circuit 303 corresponds to the average period of the pulse waveform MH. Therefore, the fundamental frequency of the pulse waveform MH can be known by referring to the average value.

Next, the substitution circuit 304 specifies a frequency range including the fundamental wave frequency of the pulse waveform MH based on the average value. For example, when the average value is 0.71 seconds, the fundamental frequency is 1.4Hz, and therefore, the specific frequency region is 1Hz to 1.5 Hz. Then, the substitution circuit 304 substitutes the pulse correction data MKDa to '0' in a frequency region smaller than the specific frequency region and generates the body motion removed pulse wave data MKDaj. This makes it possible to filter out a component in a frequency range lower than the fundamental wave frequency of the pulse waveform MH. At this time, the pulse wave component is also replaced with '0' together with the body motion component, but since the characteristic portion of the pulse wave MH is in a frequency region higher than the fundamental wave frequency, even if it is replaced with '0', it has little influence on the pulse wave.

For example, when the pulse waveform MH (fundamental wave frequency is 1.3Hz) shown in fig. 27 is detected by the pulse wave detection unit 10, the pulse wave correction data MKDa during Tc is as shown in fig. 28.

In this case, since the frequency range specified by the substitution circuit 304 is 1Hz to 1.5Hz, the frequency ranges to be substituted are 0.5Hz to 1Hz and 0Hz to 0.5 Hz. Therefore, 0 to 1Hz in the pulse correction data MKDa is replaced with '0', and the body motion removed pulse data MKDaj shown in fig. 33 is generated.

When the pulse correction data MKDa thus obtained is converted into the body motion removed pulse waveform MHj by the inverse wavelet transform 247, the autocorrelation calculating unit 210 can generate autocorrelation data RD from the body motion removed pulse waveform MHj as in the 5 th embodiment. When the pulse condition generating unit 220 generates the pulse condition data ZD based on the autocorrelation data RD, the display unit 230 displays characters (chordal, flat, and smooth) indicating the pulse condition data ZD or symbols corresponding to each pulse condition. Therefore, the 3 rd person such as the subject and the doctor can recognize the pulse condition.

As described above, according to embodiment 7, since the autocorrelation data RD is generated based on the body motion removal pulse waveform MHj from which the body motion is removed, the pulse condition can be continuously detected even in the daily life of the subject.

In embodiment 7, the body motion component is removed by skillfully utilizing the property of the body motion in which the probability of the body motion component being in a frequency region lower than the fundamental wave frequency component of the pulse wave MH is high. Therefore, the body motion detection unit 20 and the waveform processing unit 21, which are necessary in embodiments 3 and 4, can be omitted, and the pulse condition can be accurately diagnosed even when there is body motion.

8 EXAMPLE 8

In the above-described embodiments 3 and 4, the autocorrelation data RD is generated based on the pulse waveform MH, and in the above-described embodiments 5 to 7, the autocorrelation data RD is generated based on the body motion removal pulse waveform MHj. In the wavelet transform, the analysis result can be obtained for each divided time frequency. Here, the characteristic portions of the pulse waveform MH and the body motion removal pulse waveform MHj are positive or negative peak values. Further, when these peaks occur, the value of the wavelet transform analysis data becomes large in a relatively high frequency domain. Therefore, if a certain frequency region is focused and autocorrelation data RD is generated for analysis data of the focused frequency region, a pulse condition can be specified accordingly. The 8 th embodiment is proposed in accordance with this point.

Fig. 34 is a block diagram showing a pulse wave diagnosis device according to embodiment 8. In the figure, when the pulse wave detection unit 10 detects the pulse wave MH, the 1 st wavelet transform unit 243 performs wavelet transform on the pulse wave MH to generate pulse wave analysis data MKDf corresponding to the frequency region of the features. For example, as in the wavelet transform shown in fig. 26, the frequency bin is divided into 8, and if the frequency bin to be noted is 3.0Hz to 2.5Hz, M16 to M86 are output as pulse wave analysis data MKDf.

In this case, the structure of the 1 st wavelet transform task 243 is shown in fig. 35. Comparing fig. 35 with fig. 4, it can be seen that the scale converting unit W2 is omitted. This is because the scale converter W2 just needs to store the mother wavelet corresponding to the frequency region of interest in the basis function storage W1 by converting the scale parameter a corresponding to the period.

The pulse wave analysis data MKDf thus generated represents the temporal change of the characteristic portion of the pulse wave waveform MH, and therefore, the pulse wave waveform MH can be efficiently analyzed by calculating the pulse wave analysis data MKDf. Therefore, the autocorrelation calculating unit 210 of the present example generates autocorrelation data RD indicating the autocorrelation coefficient of the pulse wave analysis data MKDf.

Next, the pulse data generating unit 220 generates the pulse data ZD from the autocorrelation data RD. At this time, the pulse data generating unit 220 performs an arithmetic operation on the autocorrelation data RD, compares the arithmetic operation result with a set threshold value that can specify a pulse, and generates the pulse data ZD. When the pulse condition data ZD thus obtained is supplied to the display unit 230, characters called a chordal pulse, a level pulse, and a smooth pulse are displayed, and thus the 3 rd person such as the subject and the doctor can recognize the pulse condition.

9 EXAMPLE 9

Embodiment 9 is a case where body motion using a wavelet is eliminated and used in the pulse wave diagnostic apparatus 1 of embodiment 8.

Fig. 36 is a block diagram of a pulse wave diagnostic device 1 according to embodiment 9. In this example, the 1 st wavelet transform unit 243 and the 2 nd wavelet transform unit 245 are configured as shown in fig. 35, as in the case of the 8 th embodiment described above. Therefore, the 1 st and 2 nd wavelet transform units 243 and 245 generate the pulse wave analysis data MKDf and the body motion analysis data TKDf corresponding to only the noted frequency regions. Further, the body motion component removing section 240 subtracts the body motion analysis data TKDf from the pulse wave analysis data MKDf and generates body motion removal analysis data MKDaj.

In this way, when generating the body motion removal analysis data MKDfaj, the autocorrelation calculating unit 210 generates autocorrelation data RD representing an autocorrelation coefficient from the body motion removal analysis data MKDfaj, as in embodiment 8. Next, the pulse data generating unit 220 generates the pulse data ZD from the autocorrelation data RD. At this time, the pulse data generating unit 220 performs an arithmetic operation on the autocorrelation data RD, compares the arithmetic operation result with a set threshold value that can specify a pulse, and generates the pulse data ZD. When the pulse condition data ZD thus obtained is supplied to the display unit 230, characters called a chordal pulse, a level pulse, and a smooth pulse are displayed, and thus the 3 rd person such as the subject and the doctor can recognize the pulse condition.

Thus, according to embodiment 9, since the wavelet transform is performed to remove the body motion with attention paid to a certain frequency region, it is not necessary to compare the results of the wavelet transform in different frequency regions. Therefore, the 1 st and 2 nd frequency correction units 244 and 246 can be omitted. Further, since the autocorrelation data RD is directly generated from the body motion removal analysis data MKDfaj, the inverse wavelet transform unit 247 can be omitted.

(10) appearance Structure of each of the above-mentioned embodiments

Next, a few examples of the external configuration of the pulse wave diagnostic apparatus according to embodiments 1 to 9 will be described.

10.1 wrist-watch type A

First, a configuration example of the wrist-watch type pulse wave diagnostic apparatus 1 according to each of the above embodiments will be described with reference to fig. 37A to 37C.

As shown in fig. 37A and 37B, the pulse wave diagnostic device 1 is mainly composed of a device main body 100 having a wristwatch structure, a cable 101 connected to the device main body 100, and a pulse wave detecting unit 10 provided at the tip of the cable 101.

In which the device body 100 is equipped with a band 102. Specifically, the band 102 is wound around the left wrist of the subject in a direction facing the device body 100 in the direction of 12 o 'clock of the wristwatch, and the other end is fixed in the direction of 6 o' clock.

A connection portion 103 is also provided in the 6 o' clock direction of the apparatus main body 100. The connection portion 103 is fitted with a freely detachable connector 104, which is in turn an end of the cable 101.

When the connector 104 is removed, as shown in fig. 37C, the connection unit 103 is provided with an LED113 and a photoelectric tube 114 for data transmission, in addition to the contacts 111 and 112 connected to the cable 101.

On the other hand, as shown in fig. 37B, the pulse wave detecting unit 10 is fitted around the base of the index finger of the subject and shielded from light by the sensor fixing band 11. Thus, when the pulse wave detecting part 10 is fitted around the root of the finger, the cable can be short, and thus the movement is not hindered. When the body temperature distribution from the palm to the finger is measured, the temperature of the base of the finger is not low because the blood flow is stable, as compared with the case where the temperature of the tip of the finger is significantly low in the cold. Therefore, when the pulse wave detection unit 10 is attached to the base of the finger, the pulse wave waveform can be accurately detected even when the user goes out on a cold day.

A display unit 110 formed of a liquid crystal panel is provided on the surface of the apparatus main body 100. The display unit 110 has a segment display area, an array display area, and the like, and displays the current time and the diagnosis content. That is, the display unit 110 corresponds to the notification unit 80 or the display unit 230 in each embodiment.

On the other hand, an acceleration sensor, not shown, is mounted inside the apparatus main body 100 to detect body motion caused by the swinging of the wrist and the vertical movement of the body of the subject. That is, the acceleration sensor corresponds to the body motion detecting unit 20 of each embodiment.

A CPU (not shown) for controlling various calculations and conversions is provided inside the apparatus main body 100, and button switches SW1 and SW2 for instructing various operations are provided on the outer edge of the apparatus main body 100.

10.1.1 detailed Structure of pulse wave detection section

Next, the configuration of the pulse wave detection unit 10 will be described with reference to fig. 38.

As shown in the figure, the pulse wave detection unit 10 is composed of an LED12, a photoelectric cell 13, and the like. When the switch SW is turned on and the power supply voltage is applied, illumination light is emitted from the LED 12. The irradiation light is reflected by the blood vessel and tissue of the subject, and the reflected light is received by the photoelectric cell 13. Accordingly, the photocurrent of the photoelectric tube 13 is converted into a voltage and then output as a signal of the pulse wave detection unit 10.

Here, the emission wavelength of the LED is selected in the vicinity of the peak of the absorption wavelength of hemoglobin in blood. Therefore, the light receiving level changes according to the blood flow rate. Therefore, by detecting the light reception level, the pulse waveform can be detected.

Further, as the LED12, a blue LED of InGaN group (indium-gallium-nitrogen group) is suitable. The emission spectrum of the blue LED has an emission peak at 450nm, for example, and the emission wavelength thereof ranges from 350nm to 600 nm. As the photoelectric cell 13 corresponding to the LED having such light emission characteristics, the present embodiment uses GaAsP family (gallium-arsenic-phosphorus family). The light receiving wavelength region of the photoelectric cell 13 is, for example, a region having a main sensitivity in the range of 300nm to 600nm, and 300nm or less also belongs to the sensitivity region.

When such a blue LED and a photoelectric tube are combined and a pulse wave is detected in an overlapping region, that is, in a wavelength range from 300nm to 600nm, there are the following advantages.

First, since it is difficult for light in the wavelength region of 700nm or less to transmit through the tissue of the finger in the light including the external light, even if the external light irradiates the finger portion not covered by the sensor fixing band, the light cannot reach the photoelectric cell 13 through the tissue of the finger, and only light in the wavelength region having no influence on the detection can reach the photoelectric cell 13. On the other hand, since light in a wavelength region longer than 300nm is almost completely absorbed by the skin surface, the substantial light receiving wavelength region is from 300nm to 700nm even if the light receiving wavelength region is set to 700nm or less. Therefore, even if the finger is not covered by the large part, the influence of the external light can be suppressed. In addition, hemoglobin in blood has a large absorption coefficient for light having a wavelength of 300nm to 700nm, which is several times to several hundred times larger than that for light having a wavelength of 880 nm. Therefore, if light in a wavelength region (from 300nm to 700nm) having strong absorption characteristics corresponding to the absorption characteristics of hemoglobin is used as detection light as in this example, the detection peak value thereof often changes according to the change in blood flow rate, and the sensitivity thereof often changes, so that the signal-to-noise ratio of the pulse wave MH that changes according to the change in blood flow rate can be improved.

10.2 wrist-watch type B

Next, another configuration example of the wristwatch-type pulse wave diagnostic device 1 will be described with reference to fig. 39A and 39B. In this configuration, the pulse waveform of the subject is detected not photoelectrically by an LED, a photoelectric tube, or the like, but by a pressure sensor.

As shown in fig. 39A, the pulse wave diagnostic apparatus 1 is provided with a pair of bands 102 and 102, and an elastic rubber 131 of a pressure sensor 130 is provided on the fastening side of a fastener 120 on one side thereof so as to protrude outward. The band 102 having the fastener 120 is configured by wrapping an fpc (flexible Printed circuit) bottom plate, which supplies a detection signal of the pressure sensor, with a soft plastic (not shown in detail).

In use, as shown in fig. 39B, the elastic rubber 131 provided on the fastener 120 is to be placed in the vicinity of the radial artery 140, and the pulse wave diagnostic apparatus 1 of a wristwatch structure is wound around the left wrist 150 of the subject. Therefore, the pulse wave can be detected constantly. Furthermore, the winding manner is not similar to the use state of a common watch.

When such elastic rubber 131 is pressed against the vicinity of the radial artery 140 of the subject, the blood flow fluctuation (i.e., pulse wave) of the artery is transmitted to the pressure sensor 130 through the elastic rubber 131, and the pressure sensor 130 detects the fluctuation as blood pressure.

As another example of the pulse wave detecting unit 10, a cuff band may be put on a finger and the cuff band may be continuously pressurized to detect a harmonic component of a pulse wave waveform, as described in japanese patent application laid-open No. 5-192620, thereby extracting characteristics of a tidal wave and a dicrotic wave.

10.3 neck ring type

It is also conceivable that the pulse wave diagnostic apparatus 1 according to each embodiment is configured as a collar as shown in fig. 40.

In this figure, the pressure sensor 130 is provided at the tip of the cable 101, and is attached to the carotid artery of the subject using an adhesive tape 170, for example, as shown in fig. 41. In fig. 40, the main components of the device are incorporated in a device body 100 having a hollow portion and shaped like a brooch, and a display portion 110, switches SW1 and SW2 are provided on the front surface thereof. A part of the cable 101 is embedded in the chain 160, and a signal MH output from the pressure sensor 130 is transmitted to the apparatus main body 100.

10.4 spectacle type

As an example of the form of the pulse wave diagnostic apparatus 1 according to each of the above embodiments, a glasses form as shown in fig. 42 is further considered.

As shown in the drawing, the device body is divided into a case 100a and a case 100b, which are mounted on the temples 181, respectively, and are electrically connected to each other via leads buried in the temples 181. The side of the case 100a on the side of the mirror 182 is provided with a liquid crystal panel 183, and at one end of the side, a mirror 184 is fixed at a given angle. The box 100a also houses a drive circuit for the liquid crystal panel 183 including a light source (not shown) and a circuit for creating display data, which constitute the display unit 110. Light emitted from the light source is reflected by the mirror 184 via the liquid crystal panel 183 and is projected onto the mirror 182. In addition, the main part of the apparatus is housed in the box 100b, and on the upper side thereof, switches SW1, SW2 are provided.

On the other hand, the pressure sensor 130 is electrically connected to the case 100b via a cable 101, and is attached to the carotid artery as in the case of the neck-coil structure. Further, lead wires connecting the case 100a and the case 100b may be routed along the temple 181. In this example, the apparatus main body is divided into the cartridge 100a and the cartridge 100b, but they may be integrated into one cartridge. Further, the mirror 183 may be movable at an adjustable angle.

10.5 card type

As another example of the form, a card form as shown in fig. 43 is also conceivable. The card-like device body 100 can be placed in, for example, a left chest pocket of a subject. The pressure sensor 130 is electrically connected to the apparatus body 100 via the cable 101, and is attached to the carotid artery of the subject, as in the case of the neck-ring type and the spectacle type.

10.6 pedometer type

As another example of the embodiment, a pedometer type as shown in fig. 44A is further considered. The pedometer main body 100 is attached to a waist belt 191 of a subject as shown in fig. 44B. The pressure sensor 130 is electrically connected to the apparatus body 100 via a cable 101, fixed to the femoral artery of the femoral joint of the subject by an adhesive tape, and further protected by a protective band 192. In this case, it is preferable to take measures such as sewing the cable 101 in clothes so as not to hinder the daily life of the subject.

11 EXAMPLE 10

11.1 theoretical basis for example 10

Fig. 47 shows a structure used by the present inventors to record an arterial pressure waveform of a radial artery. The continuous blood pressure monitor 580 (CBM-2000 manufactured by コ - リ ) shown in the figure includes an arm cuff blood pressure measuring unit 582, a radial artery sensor unit 584, a controller 586 for these units, and a personal computer 588 connected to the controller 586.

Fig. 48 is a diagram showing a typical arterial pressure waveform measured by such a device, that is, a blood pressure waveform of a radial artery. The arterial blood pressure waveform shown in this figure, as described above, generally includes a kick wave (ejection wave) having a highest peak, a tidal wave (tidalwave) having a second highest peak, and a dicnodic wave (dicnodic wave) having a 3 rd peak. Furthermore, the peak value of the kick-off wave and the systolic blood pressure BP_sysAnd (7) corresponding. Dilated blood pressure BP_dirCorresponding to the lowest blood pressure in the blood pressure waveform. Furthermore, the systolic blood pressure BP_sysAnd dilating blood pressure BP_dirThe pressure difference of (b) is called pulse pressure Δ BP. Further, the mean blood pressure BP_meanThe blood pressure waveform is averaged over time.

In the continuous blood pressure monitoring device 580 shown in the figure, the cuff blood pressure measurement unit 582 measures the systolic blood pressure BP_sysAnd dilating blood pressure BP_dir. The radial artery sensor 584 detects a pulse wave corresponding to the blood pressure waveform of the radial artery, and the systolic blood pressure BP measured by the cuff blood pressure measurement 582_sysAnd dilating blood pressure BP_dirAnd correcting to obtain an arterial pressure waveform.

The present inventors measured the arterial pressure waveform of the radial artery of 74 healthy adults aged 22 to 46 years by using the continuous blood pressure monitor 580, and the measurement was performed on the seat after 15 minutes of rest in the fasting state.

Fig. 49 to 51 are diagrams showing results obtained by plotting data obtained by such measurement. That is, FIG. 49 shows the mean blood pressure BP_meanAnd dilating blood pressure BP_dirHas a strong linear relationship, and the correlation coefficient r between the two relations is 0.95, so that the two relations have a strong correlation relationship. FIG. 50 shows the mean blood pressure BP_meanAnd systolic blood pressure BP_sysThere is also a strong linear relationship (correlation coefficient r is 0.87). FIG. 51 shows the systolic blood pressure BP_dysAnd dilating blood pressure BP_dirDifferential pressure of, i.e. pulse pressure Δ BP and systolic blood pressure BP_dysThere is a strong linear relationship (correlation coefficient r is 0.86). Thus, it can be seen that the mean blood pressure BP is obtained_meanAnd pulse pressure Δ BP may reflect the state of blood pressure.

11.2 construction of blood pressure monitor

Fig. 52 is a block diagram showing the configuration of a blood pressure monitoring device 500 according to the present embodiment. As shown in the drawing, the blood pressure monitoring device 500 includes an arterial pressure waveform detection unit 504, a blood pressure conversion unit 516, an average blood pressure calculation unit 508, a pulse pressure calculation unit 512, a blood pressure determination information storage unit 520, a blood pressure determination unit 524, and an output unit 528.

The arterial pressure detection unit 504 continuously measures the arterial blood pressure and detects an arterial pressure waveform. The arterial pressure waveform detection unit 504 may be formed of, for example, the radial artery sensor unit 584 shown in fig. 47, the pulse wave detection unit 10 shown in fig. 37b, and the pressure sensor 130 shown in fig. 39A, 40, 41, 42, 43, and 44A.

The blood pressure converter 516 converts the arterial pressure waveform detected by the arterial pressure waveform detector 504 into an arterial pressure waveform at a position corresponding to the height of the heart, that is, a cardiac arterial pressure waveform. For example, the blood pressure converter 516 may be configured to use the arterial pressure waveform detected by the arterial pressure waveform detector 504 as an arterial pressure waveform corrected by the systolic blood pressure and the diastolic blood pressure measured by the cuff sphygmomanometer.

The average blood pressure calculation unit 508 calculates the average blood pressure BP from the arterial pressure waveform or the cardiac arterial pressure waveform_mean。

The pulse pressure calculating unit 512 calculates the maximum blood pressure (systolic blood pressure BP) from the arterial pressure waveform_sys) And the lowest blood pressure (diastolic blood pressure BP)_dir) I.e. the pulse pressure Δ BP.

The blood pressure determination information storage unit 520 stores blood pressure determination information in advance. The blood pressure determination information is information of, for example, a threshold blood pressure value that is a critical point of high blood pressure and normal blood pressure and a threshold blood pressure value that is a critical point of low blood pressure and normal blood pressure for the average blood pressure and pulse pressure. Furthermore, mean blood pressure and pulse pressure and systolic blood pressure BP_sys(hypertension) and diastolic blood pressure BP_dirThe (lowest blood pressure) has a high correlation and can be used as an index indicating the state of blood pressure as shown in fig. 49 to 51 described above.

The blood pressure determination unit 524 determines the average blood pressure BP from the average blood pressure BP_meanAnd at least one of the pulse pressure Δ BP and the blood pressure determination information stored in the blood pressure determination information storage unit 520. For example, using the obtained mean blood pressure BP_meanAnd pulse pressure Δ BP and pre-stored blood pressure determination information, for example, to determine hypertension, hypotension, normality, or the like.

The output unit 528 outputs the average blood pressure BP_meanAt least one of the corresponding information, the information corresponding to the pulse pressure Δ BP, and the information corresponding to the blood pressure determination. The output unit 528 may output at least one of the information corresponding to the average blood pressure, the information corresponding to the pulse pressure, and the information corresponding to the blood pressure determination, for example, as a numerical value, a graphic, or the like to a liquid crystal display device, a CRT, a printer, or the like, or may output the information as voltage, digital information, or the like corresponding to the information.

11.3 operation of blood pressure monitor

The operation of the blood pressure monitoring device 500 according to the present embodiment will be described with reference to fig. 52.

First, the arterial pressure waveform detection unit 504 continuously measures the arterial blood pressure using the pressure sensor 130 and the like described above to detect an arterial pressure waveform.

Next, the blood pressure converter 516 converts the arterial pressure waveform detected by the arterial pressure waveform detector 504 into an arterial pressure waveform at a position corresponding to the height of the heart, that is, a cardiac arterial pressure waveform.

Next, the average blood pressure calculation unit 508 calculates the average blood pressure BP from the arterial pressure waveform or the cardiac arterial pressure waveform outputted from the arterial pressure waveform detection unit 504 or the blood pressure conversion unit 516_mean。

In parallel with this, the pulse pressure calculating unit 512 calculates the systolic Blood Pressure (BP) from the arterial pressure waveform or the cardiac arterial pressure waveform outputted from the arterial pressure waveform detecting unit 504 or the blood pressure converting unit 516_sys) And the lowest blood pressure (diastolic blood pressure BP)_dir) I.e. the pulse pressure Δ BP.

Next, the mean blood pressure BP is calculated_meanAnd at least one of the pulse pressures Δ BP is input to the blood pressure determination unit 524, and the blood pressure determination unit 524 determines whether the blood pressure is high blood pressure, low blood pressure, normal blood pressure, or the like, for example, based on these data and the blood pressure determination information stored in the blood pressure determination information storage unit 520.

Then, the output unit 528 outputs the average blood pressure BP_meanAt least one of the corresponding information, the information corresponding to the pulse pressure Δ BP, and the information corresponding to the blood pressure determination. The output unit includes, for example, a liquid crystal display device, a CRT, a printer, and the like, and may display the information as numerical values, graphics, and the like, or may output the information as voltage, digital information, and the like corresponding to the information.

As described above, according to the blood pressure monitoring device 500 of the present invention, the average blood pressure or the pulse pressure can be calculated from the arterial pressure waveform detected by the arterial pressure waveform detecting unit 504, and can be monitored. Furthermore, the blood pressure monitoring apparatus 500 uses the obtained average blood pressure BP_meanAnd pulse pressure Δ BP and pre-stored blood pressure determination information, for example, to determine hypertension, hypotension, normality, or the like.

12 EXAMPLE 11

12.1 theoretical basis for example 11

The present inventors further performed the following experiments using data obtained from the experiments described in the section "theoretical basis of example 10".

That is, the obtained radial artery pressure waveform is classified into typical arterial pressure waveforms shown in fig. 45A, 45B, and 45C, i.e., the flat pulse, the slippery pulse, and the chordal pulse, by referring to a shape model of the pulse wave shape in chinese medicine (department of traditional chinese medicine, the tongue and pulse diagnoses in chinese medical science, written in fistular, chinese pulse diagnosis research, published by shanghai medical college (1991)).

As a result, it was found that the blood pressure difference between the Bob and the peak of the Bob, i.e., the Bob height Δ BP, was used_D(see fig. 48) can be classified into a smooth pulse, a flat pulse and a chordal pulse artificially as shown in fig. 53. I.e. Δ BP of the smooth vein_DIs 11 + -4 mmHg, and has a flat pulse delta BP_DIs 7 + -2 mmHg, Δ BP of chordal vein_DIs 3+ -1 mmHg, and 1% of human discrimination errors among 3 clusters are confirmed.

Further, fig. 54 shows the bob pressure difference ratio BP_DdΔ BP and Bobo height Δ BP_DHas a linear relationship between the blood pressure difference and the lowest blood pressure, i.e. the blood pressure difference BP_Dd(see fig. 48) and the pulse pressure Δ BP, which is the pressure difference between the highest blood pressure and the lowest blood pressure. As can be seen from the graph, the correlation coefficient between them is-0.86, and there is a strong correlation relationship. Thus, the pressure difference ratio BP of the Bob is known_DdThe,/Δ BP can also be used as an index for artificially classifying the typical pulse wave shapes in the chinese medical pulse diagnosis, i.e., the smooth, flat and chordal pulse.

Further, fig. 55 shows the difference between the systolic blood pressure and the systolic blood pressure, i.e., the kick-out extension pressure Δ BPp and the systolic blood pressure Δ BP_DHave a linear relationship therebetween. As can be seen from this figure, the correlation coefficient between them is 0.77, and there is a strong correlation. Therefore, it is understood that the expulsion distension pressure Δ BPp may be used as an index for artificially classifying the typical pulse wave shapes in the chinese medical pulse diagnosis, i.e., the smooth, flat, and chordal pulses.

Further, fig. 45A, 45B and 45C show typical arterial pressure waveforms relatively classified into pulse wave shapes of the smooth pulse, the rough pulse and the chordal pulse in the chinese medical pulse diagnosis, and it can be seen from these figures that the average blood pressure BP_meanRatio of pulse pressure Δ BP, i.e. mean blood pressure-pulse pressure ratio BP_meanThe/Δ BP increases in the order of smooth, flat, chordal. Therefore, the mean blood pressure-to-pulse pressure ratio BP is used_meanThe/Δ BP can classify pulse wave shapes, i.e., smooth, flat, and chordal.

Composition of 12.2 pulse wave shape monitoring device

Fig. 56 is a block diagram showing the configuration of a pulse wave shape monitoring device 540 according to the present embodiment. As shown in the figure, the pulse wave shape monitoring device 540 includes an arterial pressure waveform detecting unit 504, a pulse wave height calculating unit 544, a pulse pressure difference ratio calculating unit 548, a mean blood pressure pulse pressure ratio calculating unit 552, a pulse-width-to-pulse-width ratio calculating unit 554, a pulse wave shape determination information storing unit 556, a pulse wave shape determining unit 560, and an output unit 564. Note that, although not shown in fig. 56, a blood pressure conversion unit may be provided after the arterial pressure waveform detection unit 504, at the stage of the pulse height calculation unit 544, the pulse pressure difference ratio calculation unit 548, and before the mean blood pressure pulse pressure ratio calculation unit 552.

The arterial pressure waveform detection unit 504 continuously measures arterial blood pressure and detects an arterial pressure waveform. The arterial pressure waveform detection unit 504 may be formed of, for example, the radial artery sensor unit 584 shown in fig. 47, the pulse wave detection unit 10 shown in fig. 37B, and the pressure sensor 130 shown in fig. 39A, 40, 41, 42, 43, and 44A.

The blood pressure conversion unit converts the arterial pressure waveform detected by the arterial pressure waveform detection unit into an arterial pressure waveform at a position corresponding to the height of the heart, that is, a cardiac arterial pressure waveform.

The beat height calculating unit 544 calculates a beat height Δ BP, which is a blood pressure difference between a beat (dicrotic notch) and a beat (dicrotic wave) peak obtained from the arterial pressure waveform or the cardiac arterial pressure waveform_D(refer to FIG. 48).

The systolic pressure ratio calculating unit 548 calculates systolic blood pressure and diastolic blood pressure (extended blood pressure BP) obtained from the arterial pressure waveform or the cardiac arterial pressure waveform_dir) Pressure difference of (1), i.e. the Bob pressure difference BP_DdThe pulse pressure difference ratio BP, which is the ratio of the pulse pressure Δ BP to the pressure difference between the highest blood pressure and the lowest blood pressure_Dd/ΔBP。

An average blood pressure and pulse pressure calculating unit 552 calculates an average blood pressure BP obtained from an arterial pressure waveform or a cardiac arterial pressure waveform_meanThe ratio of the pulse pressure Δ BP to the pressure difference between the highest blood pressure and the lowest blood pressure, i.e., the mean blood pressure-pulse pressure ratio BP_mean/ΔBP。

An expelled-diastolic-pressure- Δ BPp calculating unit 554 calculates a systolic blood pressure BP and a dicrotic notch obtained from an arterial pressure waveform or a cardiac arterial pressure waveform_sysThe difference in blood pressure therebetween, i.e., the kick-out expansion pressure Δ BPp (see fig. 48).

The pulse wave shape determination information storage 556 stores pulse waves in advanceShape determination information. The pulse shape determination information is, for example, a pair pulse height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanThe/Δ BP or the kick-out pressure Δ BPp is threshold information as a threshold value of each pulse wave shape, for example, a smooth pulse, a flat pulse, or a string pulse in a chinese medical pulse. Furthermore, these pulse shapes can be obtained by using the bob wave height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanThe/Δ BP, or the kick-out expansion pressure Δ BPp is classified as shown in fig. 53, 54, 45A, 45B, 45C, and 55.

The pulse shape determination unit 560 determines the pulse shape according to the pulse height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanAt least one of/Δ BP and kick-out expansion pressure Δ BPp and pulse wave shape determination information are used to perform pulse wave shape determination. Thus, the use of the Bobo height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanAt least one of the/Δ BP and the kick-out expansion pressure Δ BPp and the prestored pulse wave shape determination information may be used to determine, for example, a smooth pulse, a flat pulse, a string, and the like in a chinese medical pulse.

The output unit 564 outputs the output signal and the bob wave height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanAt least one of the information corresponding to the kick-out expansion pressure Δ BPp and the information for determining the pulse wave shape. The output unit 528 may output numerical values, graphics, and the like to a liquid crystal display device, a CRT, a printer, and the like, or may output voltage, digital information, and the like corresponding to the information.

12.3 operation of pulse wave shape monitoring apparatus

The operation of the pulse wave shape monitoring device 540 according to the present embodiment will be described with reference to fig. 56.

Next, the blood pressure conversion unit converts the arterial pressure waveform detected by the arterial pressure waveform detection unit 504 into an arterial pressure waveform at a position corresponding to the height of the heart, that is, a cardiac arterial pressure waveform. Further, in the following step, when the arterial pressure waveform is used, the step may be omitted.

Then, the beat height calculation unit 544 calculates a beat height Δ BP, which is a blood pressure difference between the peak values of the beat (dicrotic notch) and the beat (dicrotic wave) obtained from the arterial pressure waveform or the cardiac arterial pressure waveform_D(refer to FIG. 48).

The beat pressure difference ratio calculating unit 548 and the beat height Δ BP_DCalculating the systolic blood pressure and the diastolic Blood Pressure (BP) from the arterial pressure waveform or the cardiac arterial pressure waveform_dir) Pressure difference of (1), i.e. the Bob pressure difference BP_DdThe pulse pressure difference ratio BP, which is the ratio of the pulse pressure Δ BP to the pressure difference between the highest blood pressure and the lowest blood pressure_Dd/ΔBP。

The average blood pressure and pulse pressure calculating unit 552 calculates the average blood pressure BP obtained from the pulse pressure waveform or the cardiac pulse pressure waveform in parallel with the above operation_meanThe ratio of the pulse pressure Δ BP to the pressure difference between the highest blood pressure and the lowest blood pressure, i.e., the mean blood pressure-pulse pressure ratio BP_mean/ΔBP。

Similarly, the expelled-diastolic-pressure calculating unit 554 calculates a systolic (systolic) blood pressure BP and a heartbeat (systolic) obtained from the arterial pressure waveform or the cardiac arterial pressure waveform in parallel with the above-described operation_sysThe difference in blood pressure therebetween, i.e., the kick-out expansion pressure Δ BPp (see fig. 48).

Next, the pulse wave shape determination unit 560 determines the pulse wave height Δ BP according to the repeating pulse wave height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanAt least one of the/Δ BP and the kick-out expansion pressure Δ BPp and the pulse wave shape determination information determine the shape of the pulse wave, for example, the smooth pulse, the flat pulse, the chordal pulse, and the like in the pulse diagnosis of chinese medicine.

The output unit 564 outputs the output signal and the bob wave height Δ BP_DCorresponding information, and the pressure difference ratio BP of the Bob_DdInformation corresponding to/Δ BP, and blood pressure/pulse pressure ratio BP_meanAt least one of information corresponding to/Δ BP, information corresponding to the kick-out expansion pressure Δ BPp, and pulse wave shape determination information. The output unit may output, for example, numerical values, graphics, and the like to a liquid crystal display device, a CRT, a printer, and the like, or may output voltage, digital information, and the like corresponding to these pieces of information.

Thus, according to the pulse wave shape monitoring device 540 of the present invention, the pulse wave height Δ BP can be calculated from the arterial pressure waveform detected by the arterial pressure waveform detecting unit 504_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanAt least one of/Δ BP, or a kick-out extension pressure Δ BPp. The pulse wave shape monitoring device 540 uses the pulse wave height Δ BP_DPressure difference ratio BP of Bobo_DdΔ BP, mean blood pressure to pulse pressure ratio BP_meanThe pulse wave shape determination information and at least one of/Δ BP or the kick-out expansion pressure Δ BPp may determine the pulse wave shape. Furthermore, the pulse wave shape monitoring device 540 can compare the pulse wave height Δ BP with the pulse wave height Δ BP_DCorresponding information, and the pressure difference ratio BP of the Bob_DdInformation corresponding to/Δ BP, and blood pressure/pulse pressure ratio BP_meanAt least one of the information corresponding to/Δ BP, the information corresponding to the kick-out expansion pressure Δ BPp, and the information corresponding to the pulse wave shape is output as a numerical value, a pattern, or a voltage, for example.

13 EXAMPLE 12

Principle of 13.1

The present inventor monitored the aforementioned kick-out diastolic pressure Δ BPp (differential pressure between the systolic blood pressure and the pulsus obtained from the pulse pressure waveform) and the pulsus wave height Δ BP using a configuration substantially identical to that of the pulse wave shape monitoring apparatus used in embodiment 11_D(pressure difference between the dicrotic beat and the dicrotic beat peak). As a result, it was confirmed that the discharge extension pressure can be changed as described below by administering a certain drug, for example, a pressure-lowering agent.

FIG. 63 is a graph showing the measurement of mean blood pressure BP in the case where a hypotensive agent reserpine is administered at 0 minutes, and the administration of the first 60 minutes and the second 60 minutes_meanDriving out expanding pressure delta BPp and repeating pressure difference BP_Dd(pressure difference between Bob blood pressure and the lowest blood pressure) and Bob-wave height Δ BP_D(pressure difference between the dicrotic pulse and the dicrotic pulse peak). In fig. 64, the waveform with control is the arterial pressure waveform of 1 pulsation averaged 60 minutes before administration of the pressure-lowering agent, and the waveform with NF (15min) is the arterial pressure waveform of 1 pulsation averaged every 10 seconds 15 minutes after administration of the pressure-lowering agent. From these figures, it can be seen that the blood pressure BP was averaged by administering a hypotensive agent_meanDecreased, expelling distending pressure Δ BPp increased, and dicrotic wave height Δ BP_DIs increased. The present embodiment relates to a pharmacological effect monitoring device based on such observation of pharmacological effects.

13.2 composition and Effect of pharmacological Effect monitoring device

As shown in fig. 57, the pharmacological effect monitoring device 570 does not include the heartbeat pressure difference ratio calculating unit 548 and the mean blood pressure/pulse pressure ratio calculating unit 552 of the 11 th embodiment. In the pharmacological effect monitoring device 570, the pulse wave shape determination unit is replaced with a pharmacological effect determination unit 574, and the pulse wave shape determination information storage unit is replaced with a pharmacological effect determination information storage unit 572. Except for this, the pharmacological action monitoring device 570 has the same configuration as the pulse wave shape monitoring device 540 of embodiment 11.

With such a configuration, when the arterial pressure waveform is detected by the arterial pressure waveform detecting unit 504, the pharmacological effect monitoring device 570 inputs the waveform information to the pulsar expansion pressure calculating unit 554 and the dicrotic wave height calculating unit 544. Then, the kick-out/extension pressure calculating unit 554 calculates kick-out/extension pressure, which is a difference between systolic blood pressure (systolic blood pressure) and systolic blood pressure, and outputs the kick-out/extension pressure data to the pharmacological effect determining unit 574 and the output unit 564. The heartbeat-height calculating unit 544 calculates the blood pressure difference between the heartbeat and the heartbeat peak, that is, the heartbeat height, and outputs the data to the pharmacological effect determining unit 574 and the output unit 564.

The pharmacological effect determination unit 574 determines the pharmacological effect from the input data of the expelled diastolic pressure and the data of the bobble height based on the pharmacological effect determination information stored in the pharmacological effect determination information storage unit 572, and outputs the result to the output unit 564. The output unit 564 informs the user of the values of the kick-out extension pressure and the bob height and the determination result of the pharmacological effect determination unit in the form of image information, audio information from a speaker, voltage output, and the like, for example, via an LCD.

In the pharmacological effect monitoring device 570, the pulse wave detector may be used to detect the wave of blood flowing through the blood vessel from the heart, that is, the pulse wave, instead of the arterial pressure waveform detector 504. In this case, since the diastolic blood pressure calculating unit 554 and the dicrotic-wave-height calculating unit 544 cannot calculate the absolute blood pressure, the data can be accurately compared by the systolic blood pressure and the diastolic blood pressure, which are normalized by the differential pressure between the systolic blood pressure and the diastolic blood pressure, and the systolic blood pressure.

Further, the pharmacological effect monitoring device 570 having the kick-out diastolic pressure calculation unit 554 and the sinogram height ratio calculation unit 544 has been described above, but the pharmacological effect monitoring device may have only one of the kick-out diastolic pressure calculation unit 554 and the sinogram height ratio calculation unit 544.

14 other modifications

The embodiments of the present invention have been described above, but the present invention is not limited to the above-described embodiments, and various modifications may be made within the spirit of the present invention or within the scope equivalent to the scope of the claims.

(1) In the above-described embodiment 1, the tidal wave feature extraction unit 50 generates the tidal wave feature data TWD using (f2+ f3+ f4)/f1, and the rebubular feature extraction unit 60 generates the rebubular feature data DWD using (f5+ f6+ f7)/f1, but the present invention is not limited thereto, and any harmonic component can be used to extract the features as long as the features of the tidal wave and the rebubular wave can be extracted.

(2) In the above-described embodiment 2, the time positions of the tidal wave and the dicrotic wave are specified based on the pulse wave analysis data, but the present invention is not limited thereto, and any method may be used as long as the time positions can be specified. For example, the time positions of the tidal wave and the beat wave may be specified by differentiating the signal MHj to calculate the zero crossing point of the signal MHj and showing the peak point of the signal MHj.

(3) In the above-described embodiment 2, the signal MHj is subjected to spectrum analysis using wavelet transform, but it is also possible to extract a tidal wave and a reb wave from the signal MHj using a window function and subject it to FFT processing.

(4) In embodiments 1 and 2, for the purpose of extracting the characteristics of the tidal wave and the dicrotic wave, spectral analysis is performed by FFT processing (embodiment 1), and temporal spectral analysis is performed by wavelet transform (embodiment 2), but the present invention is not limited to this, and any method may be used as long as the characteristics can be extracted from the amplitudes of the tidal wave and the dicrotic wave. For example, the amplitude of the tidal wave and the amplitude of the dicrotic wave may be determined by differentiating the signal MHj 2 times to emphasize the peaks of the signal MHj, and extracting the respective amplitudes as features.

(5) In the above-described 6 th and 7 th embodiments, the 1 st frequency correction unit or the 2 nd frequency correction unit is used, but this part may be omitted.

(6) The wavelet transform and the inverse wavelet transform performed by the above-described embodiments may also use a filter bank. The configuration of the filter bank used in the wavelet transform is shown in fig. 58, for example. In the figure, the filter bank is composed of 3 stages, and its basic units are a high-pass filter 1A and a decimation filter 1C, a low-pass filter 1B and a decimation filter 1C. The high-pass filter 1A and the low-pass filter 1B divide the signal into predetermined frequency bands and output a high-frequency component and a low-frequency component, respectively. In this example, assuming that the frequency band of the pulse wave data MD is 0Hz to 4Hz, the passband of the first-stage high-pass filter is set to 2Hz to 4Hz, while the passband of the first-stage low-pass filter is set to 0Hz to 2 Hz. In addition, the decimation filter 1C taps out 1 data every 1 sample.

When the data thus generated is supplied to the secondary stage, band division and thinning are repeated, and finally data M1 to M8 are obtained in which the band of 0Hz to 4Hz is divided into 8 parts.

Further, the high-pass filter 1A and the low-pass filter 1B may be constituted by transversal filters including delay elements (D flip-flops) inside thereof. However, the heart rate of a person is in the range of 40-200, and the fundamental frequency of the pulse wave MH varies from moment to moment according to different body states. In this case, by changing the divided frequency band in synchronization with the fundamental wave frequency, information for dynamically tracking the body state can be obtained. Therefore, the divided frequency band can be adaptively changed by using the pulse waveform MH as a clock to be supplied to the transversal filter.

In the pulse wave analysis data MKD, typical frequency components reflecting the characteristics of the pulse wave waveform MH are the fundamental wave, the 2 nd harmonic, and the 3 rd harmonic. Therefore, the pulse condition may be determined using a part of the data M × 1 to M × 8 output from the filter bank. In this case, if the filter bank is synchronized with the pulse waveform MH as described above, the high-pass filter 1A, the low-pass filter 1B, and the decimation filter 1C can be omitted, and the configuration can be simplified.

Next, fig. 59 shows an example of the configuration of an inverse filter bank used for the inverse wavelet transform. In the figure, the filter bank is composed of 3 stages, and its basic units are a high-pass filter 2A and an interpolation filter 2C, a low-pass filter 2B and an interpolation filter 2C, and an adder 2D. The high-pass filter 2A and the low-pass filter 2B divide the signal into predetermined frequency bands and output a high-frequency component and a low-frequency component, respectively. In addition, the interpolation filter 2C interpolates 1 sample every 2 samples.

Here, in order to reproduce the waveform, it is necessary to use a fully reconstructed filter bank for the filter bank shown in fig. 58 and the filter bank shown in fig. 59. In this case, it is necessary that the characteristics of the high-pass filters 1A and 2A and the low-pass filters 1B and 2B have the following relationship.

H0(-Z)F0(Z)+H1(-Z)F1(Z)＝0

H0(Z)F0(Z)+H1(-Z)F1(Z)＝2Z^-L

Further, the high-pass filter 2A and the low-pass filter 2B may be constituted by transversal filters including delay elements (D flip-flops) inside thereof. In order to synchronize the filter bank used in the wavelet transform unit 10 with the fundamental wave frequency of the pulse waveform MH and to vary the divided frequency band, the clock signal supplied thereto may be supplied to the high-pass filter 2A and the low-pass filter 2B when synchronized with the pulse waveform MH.

(7) In the above embodiment, the display unit 30 is described as an example of the notification device, but the following means may be mentioned as means for notifying a person by the device. It is appropriate to classify these means into the five sense organs. It is needless to say that these means may be used not only alone but also in combination of a plurality of means. As described below, for example, the visually impaired person can understand the content of the notification by using a non-visual means, and the hearing impaired person can understand the content of the notification by using a non-auditory means, so that a device which is convenient for the disabled user can be configured.

First, as an auditory informing means, there are apparatuses and the like aiming at informing the analysis and diagnosis result of cardiac function and the like or for warning. For example, there are a piezoelectric element and a speaker in addition to a buzzer. In addition, as a specific example, it is conceivable that a person to be notified carries a portable radio paging receiver, and the device side calls the portable radio paging receiver when notification is performed. In addition, when these devices are used for notification, it is often the case that some information is to be conveyed together not only for notification. At this time, the level of the sound volume or the like shown below may be changed in correspondence with the content of the information to be conveyed. Such as tone, volume, timbre, and type of music (tunes, etc.).

Next, the visual notification means is used for the purpose of notifying various information and measurement results by the apparatus or for the purpose of warning. The following instruments can be considered as means for achieving these objects. For example, a display device, a CRT (cathode ray tube display device), an LCD (liquid crystal display device), a printer, an X-Y plotter, a lamp, and the like. Further, a glasses type slide projector is used as a special display device. In addition, the following forms of notification are considered. For example, in numerical value notification, in addition to digital display and analog display, there are graphic display, shading of display colors, direct numerical value display, bar graph display when numerical values are displayed with gradation, a circle graph, a face expression graph, and the like. The facial expression chart is, for example, as shown in fig. 60.

Second, the tactile notification means may be used for warning purposes. Such means are as follows. First, there is a means for electrical stimulation, in which a shape memory alloy protruding from the back surface is provided in a portable device such as a wristwatch, and the stimulation is performed by energizing the shape memory alloy. Further, there are mechanical stimulation means, which are classified into a configuration in which stimulation is performed by a protrusion (for example, a less sharp needle) that can be inserted from the back side and retracted from a portable device such as a wristwatch, a configuration in which stimulation is performed by vibration of the device main body 100 of a wristwatch, and the like.

Next, the olfactory notification means may be configured to include a discharge mechanism such as a perfume in the device, to correspond the fragrance to the content of the notification, and to discharge the perfume corresponding to the content of the notification. Incidentally, as the discharge mechanism of the perfume and the like, a micro pump and the like are preferably used.

It is needless to say that these means may be used not only alone but also in combination of a plurality of means.

(8) In the above embodiments, the pulse wave detection unit 10 was described as an example of the pulse wave detection device, but the present invention is not limited to this, and any device may be used as long as it can detect the pulsation.

For example, the pulse wave detection unit 10 uses reflected light, but may use transmitted light. However, light having a wavelength region of 700nm or less is difficult to transmit through the tissues of the finger. Therefore, when the transmitted light is used, the light of 600nm to 1000nm wavelength is irradiated from the light emitting section, the irradiated light is transmitted in the order of tissue → blood vessel → tissue, and the change in the amount of the transmitted light is detected. Since the transmitted light is absorbed by hemoglobin in blood, the pulse waveform can be detected by detecting the change in the amount of transmitted light.

In this case, the light emitting section preferably uses laser light emitting diodes of InGaAs group (indium-gallium-arsenic) and GaAs group (gallium-arsenic). However, since the external light having a wavelength of 600nm to 1000nm easily penetrates the tissue, the signal-to-noise ratio of the pulse signal is lowered when the external light enters the light receiving section. Therefore, the polarized laser light can be irradiated from the light emitting portion, and the light receiving portion can receive the transmitted light through the polarization filter. Thus, the pulse wave signal can be detected with a high signal-to-noise ratio without being affected by external light.

In this case, as shown in fig. 61, the light emitting section 146 is provided on the fixed side of the fixing member 145, and the light receiving section 147 is provided on the wristwatch body side. At this time, the light emitted from the light emitting section 146 passes through the blood vessel 143, and then passes through the space between the radius 148 and the ulna 142 to reach the light receiving section 147. Further, when using transmitted light, it is necessary to transmit the irradiated light through the tissue, and the wavelength is preferably 600nm to 1000nm in consideration of the absorption of the tissue.

Fig. 62 shows an example in which the detection site is an ear. The clamp 190 and the clamp 191 are energized by a spring 192 and can rotate about a shaft 193. Further, the holder 190 and the holder 191 are provided with a light emitting portion 194 and a light receiving portion 195. When the pulse wave detecting section is used, the pulse wave is detected after the ear is held by the holder 190 and the holder 191.

(9) In embodiment 9, the body motion detecting unit 20 detects the body motion waveform TH, compares the pulse wave analysis data MKDf with the body motion analysis data TKDf, eliminates the body motion component, calculates the autocorrelation data RD, and specifies the pulse condition based on the autocorrelation data RD. However, since the body motion components mostly occur in a low frequency region lower than the fundamental wave frequency of the pulse wave form MH, if the frequency region to be monitored is selected in a high frequency region higher than the fundamental wave frequency of the pulse wave form MH, the body motion detection unit 20, the waveform processing unit 21, the determination unit 22, the 2 nd wavelet transform unit 45, and the body motion component removal unit 240 can be omitted. That is, in the pulse wave diagnosis apparatus 1 shown in fig. 34, if the frequency region to be monitored is selected to be in a high frequency region higher than the fundamental wave frequency of the pulse wave MH, the pulse condition can be accurately specified even if there is a body motion.

(10) In the 10 th and 11 th embodiments, examples are shown in which the blood pressure monitor and the pulse wave shape monitor are formed as separate devices. However, the pulse wave shape monitoring apparatus according to embodiment 11 may be included in the blood pressure monitoring apparatus according to embodiment 10.

(11) In addition, in embodiment 11, an example of a pulse wave shape monitoring device provided with a pulse wave height calculation unit 544, a pulse pressure difference ratio calculation unit 548, a mean blood pressure pulse pressure ratio calculation unit, and a diastolic pressure calculation unit is shown, but at least one of these calculation units may be provided.

Claims

1. A pulse wave diagnostic device, comprising:

a pulse wave detection device for detecting a pulse wave waveform of a body;

a spectrum analyzer for performing spectrum analysis on the pulse waveform;

a tidal wave feature extraction device for extracting a feature of a tidal wave from the pulse waveform and generating tidal wave feature information based on a ratio [ (f5+ f6+ f7)/f1] of the sum of the amplitudes f5 to f7 of the 5 th to 7 th harmonics and the fundamental wave amplitude f1, which are analysis results of the spectrum analysis device;

a rebleed wave feature extraction device for extracting the feature of the rebleed wave from the pulse wave waveform and generating the rebleed wave feature information according to the analysis result of the frequency spectrum analysis device, namely the ratio [ (f2+ f3+ f4)/f1] of the sum of the amplitudes f 2-f 4 of the 2 nd-4 th higher harmonics and the amplitude f1 of the fundamental wave;

2. The pulse wave diagnostic apparatus of claim 1,

the tidal wave feature extraction means measures a duration of a tidal wave in the pulse waveform, extracts a feature of the tidal wave from the pulse waveform based on an analysis result of the spectrum analysis means during the duration, and generates tidal wave feature information,

the characteristics extracting device measures the duration of the pulse wave in the pulse wave waveform, extracts the characteristics of the pulse wave from the pulse wave waveform according to the analysis result of the spectrum analyzing device in the duration, and generates the characteristics information of the pulse wave.

3. The pulse wave diagnostic device according to claim 1 or 2,

the spectrum analysis device performs FFT processing on the pulse waveform.

4. The pulse wave diagnostic device according to claim 1 or 2,

the spectrum analysis device performs wavelet transform processing on the pulse waveform.

5. The pulse wave diagnostic device according to claim 1 or 2,

the pulse condition judging device is provided with a informing device for informing the pulse condition judged by the pulse condition judging device.

6. The pulse wave diagnostic apparatus of claim 3,

7. The pulse wave diagnostic apparatus of claim 4,