JP2021040882A - Cardiopulmonary function state change estimation system, cardiopulmonary function state change estimation device, cardiopulmonary function state change estimation method, and cardiopulmonary function state change estimation program - Google Patents

Abstract

To provide highly accurate cardiopulmonary function state change estimation system, cardiopulmonary function state change estimation device, cardiopulmonary function state change estimation method, and cardiopulmonary function state change estimation program.SOLUTION: A cardiopulmonary function state change estimation system includes: a loading device for giving a load to a person to be measured who is doing exercise; and a cardiopulmonary function state change estimation device having a data acquisition unit for acquiring a plurality of pieces of biological marker data related to a cardiopulmonary function of the person to be measured from the person to be measured who is doing exercise at each of a plurality of time points, and a cardiopulmonary function state change point estimation unit for constructing a multi-dimensional movement vector between two points of the plurality of time points respectively on the basis of a multi-dimensional vector at the plurality of time points composed of the biological marker data, and estimating a local maximum point of a curvature degree of two continuous multi-dimensional movement vectors as a change point of a cardiopulmonary function state of the person to be measured.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a cardiopulmonary function state change estimation system, a cardiopulmonary function state change estimation device, a cardiopulmonary function state change estimation method, and a cardiopulmonary function state change estimation program.

In recent years, the effects of exercise on maintaining and improving health have been clarified in many clinical studies, and exercise therapy has come to be implemented as a method for treating lifestyle-related diseases such as diabetes and heart disease. The center of exercise therapy is aerobic exercise such as walking and cycling. In order to perform aerobic exercise more safely and effectively, it is important for patients and the like to exercise at the optimum exercise intensity for each individual (hereinafter referred to as "optimal exercise intensity").
The optimum exercise intensity is generally estimated by a CPX test (cardiopulmonary exercise load test) in which exhaled gas is collected and analyzed while gradually applying an exercise load to the person to be measured. In the CPX test, exercise intensity (anaerobic metabolism) near the boundary between aerobic exercise and anaerobic exercise, which is the change point of cardiopulmonary function, is obtained from the oxygen uptake and carbon dioxide emission of the measurement subject obtained by breath gas analysis. The threshold, hereinafter referred to as AT (Anaerobic Threshold)), is determined. In the CPX test, the anaerobic metabolism threshold AT is regarded as the optimum exercise intensity of the person to be measured.

However, the CPX test imposes a heavy physical burden on the person to be measured, and there are problems such as a limited number of facilities where the test device can be carried out at a high price. In addition, wearing a mask necessary for breath gas analysis, which is essential for CPX examination, is unpleasant for the measurement subject, and a simple method for measuring optimal exercise intensity that does not require breath gas analysis is required. Therefore, some devices have been proposed to realize the measurement of the optimum exercise intensity more easily and inexpensively without exhaled gas analysis. For example, conventionally, the double product of the heart rate data and the heart sound (specifically, the first heart sound) amplitude data of the measurement target person is calculated as the double product data, and an approximation line that approximates the distribution of the double product data. An apparatus has been proposed for detecting the bending exercise intensity of the above as the optimum exercise intensity (see, for example, Patent Document 1).

Japanese Patent No. 5526421

However, while the optimal exercise intensity detected by such a device can detect a substantially accurate optimal exercise intensity for many measurement subjects, some measurement subjects (for example, patients with heart failure). In patients with heart failure such as), the optimal exercise intensity detected by such a device may be far from the actual optimal exercise intensity. In addition, it is not possible to detect information useful for determining exercise intensity other than the anaerobic metabolism threshold (AT).
The present disclosure focuses on the fact that multiple biological marker data show flexion at changes in cardiopulmonary function states such as anaerobic metabolic threshold (AT) and respiratory compensation initiation point (RC). Cardiopulmonary function state change estimation system, cardiopulmonary function state change estimation device, cardiopulmonary function state change estimation method, and cardiopulmonary function state change estimation program that estimate cardiopulmonary function state with high accuracy for any subject. The purpose is to provide.

In order to solve the above problems, the cardiopulmonary function state change estimation system according to the present disclosure includes a load device that gives a load to a person to be measured who is exercising.
A data acquisition unit that acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of a plurality of time points.
Based on the multidimensional vector at the plurality of time points composed of the biological marker data, a multidimensional movement vector between two points at the plurality of time points is constructed, and two consecutive multidimensional movements are formed. A cardiopulmonary function state change estimation unit that estimates the local maximum point of the vector flexion degree as a change point of the cardiopulmonary function state of the measurement subject,
Cardiopulmonary function state change estimator with
A cardiopulmonary function state change estimation system equipped with.

The cardiopulmonary function state change estimator according to another aspect of the present disclosure acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of a plurality of time points. Based on a multidimensional vector at a plurality of time points composed of a data acquisition unit and biological marker data, a multidimensional movement vector between two points at a plurality of time points is constructed, and two consecutive multiples are formed. It is characterized by including a cardiopulmonary function state change estimation unit that estimates a local maximum point of the degree of flexion of the dimensional movement vector as a change point of the cardiopulmonary function state of the measurement subject.

The cardiopulmonary function state change estimation method according to another aspect of the present disclosure acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of a plurality of time points. Then, based on the multidimensional vector at a plurality of time points composed of biological marker data, a multidimensional movement vector between two points at the plurality of time points is constructed, and two consecutive multidimensional movement vectors are formed. The feature is that the local maximum point of the degree of flexion is estimated as the change point of the cardiopulmonary function state of the measurement subject.

The cardiopulmonary function state estimation program according to another aspect of the present disclosure acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of a plurality of time points. That, and based on the multidimensional vector at multiple time points composed of biological marker data, each of the multidimensional movement vectors between two points at multiple time points is constructed, and two consecutive multidimensional ones. It is characterized in that the local maximum point of the degree of flexion of the movement vector is estimated as the change point of the cardiopulmonary function state of the measurement subject, and the computer is made to execute.

The present disclosure can provide a highly accurate cardiopulmonary function state change estimation system, a cardiopulmonary function state change estimation device, a cardiopulmonary function state change estimation method, and a cardiopulmonary function state change estimation program.

It is a block diagram which shows one configuration example of the cardiopulmonary function state change estimation system and the cardiopulmonary function state change estimation device which concerns on 1st Embodiment. It is a schematic diagram explaining the electrocardiographic data and the cardiac sound data used in the cardiopulmonary function state change estimation apparatus which concerns on 1st Embodiment. It is a block diagram which shows one structural example of the cardiopulmonary function state change estimation part of the cardiopulmonary function state change estimation apparatus which concerns on 1st Embodiment. It is a schematic diagram explaining the outline of the multidimensional movement vector configured in the multidimensional movement vector component of the cardiopulmonary function state change estimation apparatus which concerns on 1st Embodiment. This is an example of a graph shown in the output unit of the cardiopulmonary function state change estimation device according to the first embodiment. It is the schematic explaining the anaerobic metabolism threshold AT and the respiratory compensation start point RC. It is a flowchart explaining the cardiopulmonary function state change estimation method which concerns on 1st Embodiment. It is a block diagram which shows one configuration example of the cardiopulmonary function state change estimation apparatus which concerns on 2nd Embodiment. It is a block diagram which shows one configuration example of the cardiopulmonary function state change estimation apparatus which concerns on 3rd Embodiment. It is a block diagram which shows an example of the hardware configuration of the cardiopulmonary function state change estimation apparatus which concerns on 1st to 3rd Embodiment. It is a block diagram which shows an example of the internal structure of the signal processing terminal included in the cardiopulmonary function state change estimation apparatus which concerns on 1st to 3rd Embodiment. It is a block diagram which shows another example of the hardware configuration of the cardiopulmonary function state change estimation apparatus which concerns on 1st to 3rd Embodiment. It is a block diagram which shows another example of the internal structure of the signal processing terminal included in the cardiopulmonary function state change estimation apparatus which concerns on 1st to 3rd Embodiment. It is a graph which shows an example of the measurement result in an Example. It is a graph which shows an example of the measurement result in an Example. It is a graph which shows an example of the measurement result in an Example.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. However, the embodiments described below are merely examples, and there is no intention of excluding the application of various modifications and techniques not specified below. The present disclosure can be implemented in various modifications (for example, by combining each embodiment) without departing from the spirit of the present disclosure. Further, in the description of the following drawings, the same or similar parts are represented by the same or similar reference numerals.

1. 1. First Embodiment (1.1) Configuration of Cardiopulmonary Function State Change Estimating Device Hereinafter, the cardiopulmonary function state change estimation system 1000 according to the first embodiment will be described with reference to FIGS. 1 to 5.
FIG. 1 is a block diagram showing a configuration example of a cardiopulmonary function state change estimation system 1000. As shown in FIG. 1, the cardiopulmonary function state change estimation system 1000 includes a load device 60 that applies a load to a person to be measured who is exercising, and a cardiopulmonary function state change estimation device 1.
As the load device 60, for example, an ergometer capable of controlling the magnitude of the exercise load (exercise intensity) given to the person to be measured is used.
The cardiopulmonary function state change estimation device 1 includes an exercise load control unit 10, a data acquisition unit 20, a storage unit 30, a cardiopulmonary function state change estimation unit 40, and an output unit 50. The cardiopulmonary function state change estimation device 1 acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of a plurality of time points, and the biological marker data. It has a function of estimating the change of cardiopulmonary function state of the measurement subject based on the above and outputting the estimation result.

The cardiopulmonary function state change estimation device 1 is realized as, for example, a small device. Such devices include, for example, stick-on (patch-type), cable-type (holter-type), wear-type (belt-type, vest-type) non-invasive (wearable) medical devices, or implantable medical devices. It will be realized.
Only a part of the cardiopulmonary function state change estimation device 1 (for example, a function of acquiring the electrocardiographic data and the heartbeat data of the measurement target person) is realized as a small device, and the other part is realized on a host computer or the cloud. You may.
Hereinafter, each part of the cardiopulmonary function state change estimation device 1 will be described in detail.

<Exercise load control unit>
In the load test, the exercise load control unit 10 controls the load device 60 so as to increase the exercise load (hereinafter, also referred to as “exercise intensity”) at a constant rate with time. The exercise load control unit 10 controls the load device 60 so as to give a predetermined constant load in the exercise therapy of cardiac rehabilitation or the gradual increase load (ramp load) in the cardiopulmonary exercise load test to the measurement subject. As the rate of increase in exercise load, for example, 5 Watts / min or more and 20 Watts / min or less is used. The exercise load control unit 10 outputs exercise intensity data (hereinafter referred to as “exercise intensity data”) at each time point (for example, every 10 [seconds]) to the storage unit 30 and the cardiopulmonary function state change estimation unit 40. The exercise intensity data is, for example, in a digital data format.

<Data acquisition department>
The data acquisition unit 20 acquires data of a plurality of biological markers related to the cardiopulmonary function of the measurement target person (hereinafter, also referred to as “biological marker data”) from the measurement target person who is exercising. The data acquisition unit 20 acquires biological marker data related to the cardiopulmonary function of the measurement target at a plurality of time points from the electrocardiographic data and the heartbeat data of the measurement target measured at the same time. Hereinafter, the biological marker data acquired from the electrocardiographic data and the biological marker data acquired from the cardiac sound data may be referred to as electrocardiographic biological marker data and cardiac sound biological marker data, respectively. The measurement subjects to be measured by the cardiopulmonary function state change estimation device 1 are, for example, heart disease patients, heart failure patients, healthy people who do not have these diseases, and animals such as pets and racehorses. is there.
The data acquisition unit 20 includes, for example, an electrocardiographic collection unit 21, an electrocardiographic processing unit 22, a heart sound collection unit 23, and a heart sound processing unit 24.

(Electrocardiographic collection section)
The electrocardiographic collection unit 21 is composed of two or more electrodes. The electrocardiographic sampling unit 21 is formed so as to be wearable to the measurement target person, and when an exercise load is applied to the measurement target person, the electric potential at each time point of the measurement target person's heart is an electric signal (hereinafter, hereinafter, electric signal) for each electrode. Also called "electrode signal"). Each time the electrode signal is taken out, the electrocardiographic collecting unit 21 outputs the taken out electrode signal to the electrocardiographic processing unit 22.

(Electrocardiographic processing unit)
The electrocardiographic processing unit 22 generates electrocardiographic data at each time point based on the potential difference between the plurality of electrode signals output by the electrocardiographic sampling unit 21. The electrocardiographic data is, for example, in a digital data format.
The electrocardiographic processing unit 22 generates biological marker data related to the cardiopulmonary function of the measurement subject based on the electrocardiographic data. Biological markers derived from electrocardiographic data calculate, for example, heart rate (HR), QR width (QRI), QRS width (QRSD), or at least one of these divided by heart rate length. Biological markers derived from electrocardiographic data will be described in detail later. The electrocardiographic processing unit 22 outputs the generated electrocardiographic data and biological marker data to the storage unit 30. In addition, the electrocardiographic processing unit 22 outputs the generated biological marker data to the cardiopulmonary function state change estimation unit 40.

(Heart sound collection department)
The heart sound collecting unit 23 is composed of a microphone, an acceleration sensor, a pressure sensor, and the like. The heart sound collecting unit 23 is formed so as to be wearable to the measurement target person, and when an exercise load is applied to the measurement target person, the heart sound at each time point of the measurement target person's heart is taken out as an electric signal. As the electric signal, for example, a heart sound signal corresponding to the heart sound vibration propagating in the chest of the person to be measured is used. Each time the heart sound signal is taken out, the heart sound collecting unit 23 outputs the taken out heart sound signal to the heart sound processing unit 24.

(Heart sound processing unit)
The heart sound processing unit 24 generates heart sound data at each time point based on the heart sound signal output by the heart sound collecting unit 23. The heartbeat data is, for example, in a digital data format.
The heart sound processing unit 24 generates biological marker data related to the cardiopulmonary function of the measurement subject based on the heart sound data. The biological marker derived from the heart sound data is, for example, a parameter indicating the length of a section such as systole or diastole of the heart (hereinafter referred to as a heart section length parameter). The heart sound processing unit 24, for example, has electromechanical activity time (EMAT), left ventricular contraction time (LVST), left ventricular diastolic perfusion time (LDPT), left ventricular dilatation time (LVDT), and QoS2 as heart section length parameters. Calculate at least one of. Alternatively, the biological marker derived from the heart sound data is, for example, a parameter indicating the intensity of the heart sound (hereinafter referred to as a heart sound intensity parameter). The heart sound processing unit 24 has, for example, as heart sound intensity parameters, S1 intensity (S1 Intensity), S2 intensity (S2 Intensity), S3 intensity (S3 Intensity) and S4 intensity (S4 Intensity), S1 intensity, S2 intensity, S3 intensity and The natural logarithmic value of S4 intensity and at least one of S1 certainty (S1 Strength), S2 certainty (S2 Strongth), S3 certainty (S3 Strongth) and S4 certainty (S4 Strength) are calculated. Biological markers derived from heartbeat data will be described in detail later. The heart sound processing unit 24 outputs the generated electrocardiographic data and biological marker data to the storage unit 30. In addition, the heart sound processing unit 24 outputs the generated biological marker data to the cardiopulmonary function state change estimation unit 40.

The data acquisition unit 20 may have, for example, a function as a medical device capable of simultaneously measuring the electrocardiographic data and the cardiac sound data of the measurement target person. Further, the data acquisition unit 20 is a biology calculated based on the electrocardiographic data and the electrocardiographic data, or the electrocardiographic data and the electrocardiographic data from an external medical device capable of simultaneously measuring the electrocardiographic data and the electrocardiographic data of the measurement target person. The configuration may be such that the target marker data is acquired.
The data acquisition unit 20 outputs the acquired biological marker to the multidimensional vector configuration unit 41.

(Biological marker)
Hereinafter, the biological markers generated by the electrocardiographic processing unit 22 and the heart sound processing unit 24 will be described with reference to FIGS. 2 (A) and 2 (B).
Biological markers representing cardiopulmonary function states that can be calculated from electrocardiographic data and heart sound data are collectively called "CABs (Cardiac Acoustic Biomarkers)". CABs are, for example, either the above-mentioned heart section length parameter or heart sound intensity parameter. Further, a test for measuring electrocardiographic data and electrocardiographic data and calculating CABs from the electrocardiographic data or cardiac sound data is called a "CABs test".

(Heart rate)
The heart rate (hereinafter, may be referred to as HR (Heart Rate)) is the number of heartbeats per minute. As shown in FIG. 2A, the heart rate can be measured by counting the number of heartbeats per minute, with the interval from the R point of the electrocardiographic data to the next R point as one heart rate. .. Further, as shown in FIG. 2B, the heart rate can be measured by counting the number of heartbeats per minute, with the interval from the normal heart sound S1 of the heart sound data to the next S1 as one heartbeat. it can.

(Heart rate)
The heartbeat length is the interval from the R point of the electrocardiographic data to the next R point, and is the length of one heartbeat. Hereinafter, for convenience, the heart rate may be referred to as RR.

(CABs)
As described above, specific examples of biological markers derived from electrocardiographic data include, for example, electromechanical activity time (EMAT), left ventricular contraction time (LVST), left ventricular diastolic perfusion time (LDPT), and so on. Left ventricular expansion time (LVDT), total time of EMAT and LVST (QoS2), QR width (QRI: QR Interval) which is the time interval from Q point to R point of electrocardiographic data, Q point of electrocardiographic data The QRS width (QRSD: QRS Duration), which is the time interval from the point R to the point S, can be mentioned.

Further, as a specific example of CABs, for example, EMATc, LVSTc, LDPTc, LVDTc, QoS2c, which are values obtained by dividing the above-mentioned EMAT, LVST, LDPT, LVDT, QoS2, PRI, and QRSD by the heart rate (RR), respectively. , QRIc, QRSDc and the like.

The Electro Mechanical Activation Time (EMAT) is the interval from the Q point of the electrocardiographic data to the point where the S1 intensity of the normal heart sound S1 of the heart sound data is the strongest. EMAT represents the interval from the electrical firing of electrocardiographic data due to the contraction of the myocardium of the left ventricle to the onset of contraction of the left ventricle. It is said that when EMAT exceeds 120 ms, it suggests that the contractility of the heart is reduced.
The left ventricular systolic time (LVST: Left Ventricular Systolic Time) is the interval from the point where the S1 intensity of the normal heart sound S1 is the strongest to the point where the S2 intensity of the normal heart sound S2 is the strongest. LVST represents the contraction time interval of the heart.
The left ventricular diastolic perfusion time (LDPT) is the interval from the point where the S2 intensity of the normal heart sound S2 is the strongest to the point Q of the electrocardiographic data. LDPT represents the time interval obtained by subtracting the electromechanical activity time (EMAT) from the expansion time of the left heart.

The left ventricular diastolic time (LVDT) is the interval from the point where the S2 intensity of the normal heart sound S2 is the strongest to the point where the S1 intensity of the subsequent normal heart sound S1 is the strongest. LVDT represents the expansion time interval of the heart. LVDT is indicated by the sum of LDPT and EMAT.

The total time (QoS2) of EMAT and LVST is the interval from the Q point of the electrocardiographic data to the point where the S2 intensity of the normal heart sound S2 is the strongest. QoS2 represents the interval from the electrical firing of the electrocardiographic data for the contraction of the myocardium of the left ventricle to the end of the contraction of the left heart.
The QR width (QRI: QR Interval) is the time interval from the Q point to the R point of the electrocardiographic data.
The QRS width (QRSD: QRS Duration) is the time interval from the Q point of the electrocardiographic data to the S point via the R point. A wide QRS complex means that the excitatory conduction of the electrocardiogram in the left heart takes time, and the contraction start timing of the myocardium is shifted between the location of the fast excited myocardium and the location of the last excited myocardium.

Further, as a specific example of the biological marker derived from the heart sound data, for example, S1 intensity (S1 Intensity) which is the maximum value of the peak-to-peak amplitude in the section of the normal heart sound S1 of the heart sound data. , S2 intensity (S2 Intensity) which is the maximum value of the peak peak amplitude of the normal heart sound S2 section of the heart sound data, S3 intensity (S3 Intensity) which is the maximum value of the peak peak amplitude of the abnormal heart sound S3 section of the heart sound data, and heart sound. The S4 intensity, which is the maximum value of the peak peak amplitude in the section of the abnormal heart sound S4 of the data, can be mentioned.

Further, specific examples of CABs include, for example, S1 probability (S1 Strength) and S2 probability (S2 Strength), which are the certainty (or likelihood) of the normal heart sounds S1 and S2 and the abnormal heart sounds S3 and S4 of the heart sound data. S3 accuracy (S3 Strength) and S4 accuracy (S4 Strength), as well as the above-mentioned S1 intensity, S2 intensity, S3 intensity, and S4 intensity, which are the natural logarithmic values of log (S1 Integrity), log (S2 Integrity), and log (S3 Integrity). ) And log (S4 Integrity).

Each index (biological marker) described above is a specific example of CABs that can be calculated from electrocardiographic data or heartbeat data, but if it is an index that can be calculated from electrocardiographic data or heartbeat data, each of the above-mentioned indexes Not limited to.

<Memory>
The storage unit 30 stores each of the exercise intensity data output by the exercise load control unit 10, the electrocardiographic data output by the electrocardiographic processing unit 22, the cardiac sound data output by the cardiac sound processing unit 24, and the biological marker data. The storage unit 30 stores exercise intensity data, electrocardiographic data, cardiac sound data, and biological marker data at each time point when an exercise load is applied to the measurement target person.
As the storage unit 30, for example, a recording medium such as a hard disk or an SD (Secure Digital) card is used.

<Cardiopulmonary function state change estimation unit>
The cardiopulmonary function state change estimation unit 40 estimates the change point of the cardiopulmonary function state of the measurement subject based on the biological marker data input from the electrocardiographic processing unit 22 and the heart sound processing unit 24.
As shown in FIG. 3, the cardiopulmonary function state change estimation unit 40 calculates the angle between the multidimensional movement vector constituent unit 41, the multidimensional movement vector constituent unit 42, the multidimensional movement vector length calculation unit 43, and the multidimensional movement vector. A unit 44, a flexion degree calculation unit 45, a flexion degree local maximum point extraction unit 46, and a cardiopulmonary function state change point estimation unit 47 are provided.
Hereinafter, the details of the cardiopulmonary function state change estimation unit 40 will be described with reference to FIG. 3 and FIG.

(Multidimensional vector component)
The multidimensional vector constituent unit 41 constructs a multidimensional vector from biological marker data at a plurality of time points. That is, the multidimensional vector constituent unit 41 constructs a multidimensional vector indicating a measurement point in which a plurality of biological marker data obtained at the same timing is reflected in the multidimensional space. The multidimensional vector configuration unit 41 constructs a multidimensional vector for each biological marker data measured at a plurality of time points (for example, every 10 seconds).
The multidimensional vector configuration unit 41 outputs data of a plurality of multidimensional vectors (hereinafter, referred to as “multidimensional vector data”) to the multidimensional movement vector configuration unit 42.

FIG. 4 is a diagram schematically showing a measurement point and a multidimensional vector reflected in a multidimensional space. The multidimensional vector component 41 is the biological marker data measured at N time points (first time point, second time point, ..., Nth time point) from which the biological marker data was acquired. The first multidimensional vector a1, the second multidimensional vector a2, ..., The Nth multidimensional vector aN indicating the measurement points m1, m2, ... .. Note that FIG. 4 shows the first multidimensional vector a1 to the third multidimensional vector a3 as the multidimensional vector. FIG. 4 shows the origin O of the multidimensional space.
Hereinafter, the kth multidimensional vector ak indicating the measurement point mk in which the biological marker data measured at the kth time point is reflected in the multidimensional space is referred to as "kth multidimensional vector ak at the kth time point" (. k is an integer of 1 or more and N or less).

In the multidimensional vector component 41, for example, nine biological marker data of HR, EMAT, LVST, LDPT, QRSD, S1 Intensity, S2 Intensity, S3 Intensity, and S4 Intensity are converted into a multidimensional vector. Alternatively, in the multidimensional vector component 41, for example, nine organisms of HR, EMAT, LVST, LDPT, QRSD, log (S1 Integrity), log (S2 Integrity), log (S3 Integrity), and log (S4 Integrity). The scientific marker data may be converted into a multidimensional vector. EMAT, LVST, LDPT, and QRSD are cardiac section length parameters related to the contraction of the heart as a blood pump and the time length of diastolic exercise. S1 Intensity, S2 Intensity, S3 Intensity, and S4 Intensity are heart sound intensity parameters relating to the intensity of heart sounds related to hemodynamics within the heart as a blood pump. The multidimensional vector configuration unit 41 may appropriately change the number and combination of biological markers used in the configuration of the multidimensional vector according to the state of the cardiopulmonary function of the measurement target person.

When constructing a multidimensional vector, these biological marker data have different dynamic ranges. Therefore, the multidimensional vector component 41 converts each biological marker data into a [0.0, 1.0] interval normalization value by a method such as linear normalization. If there is a missing data of biological marker data, the missing data can be filled in by linear interpolation, polynomial approximation, etc. using the values of the biological marker data before and after, or each biological marker data. The method of filling in the blanks with the median value of is used.

(Multidimensional movement vector component)
The multidimensional movement vector configuration unit 42 selects two consecutive time points from a plurality of time points, and constructs a multidimensional movement vector between the two multidimensional movement vectors from the two multidimensional vectors at the two time points. That is, the multidimensional movement vector configuration unit 42 uses the multidimensional vector data at two time points selected from the multidimensional vector data received from the multidimensional vector configuration unit 41, and two from the difference between the two multidimensional vectors. Construct a multidimensional movement vector between time points.
The multidimensional movement vector configuration unit 42 transfers the data of a plurality of multidimensional movement vectors (hereinafter referred to as “multidimensional movement vector data”) to the multidimensional movement vector length calculation unit 43 and the multidimensional movement vector inter-angle calculation unit 44. Output each. Further, the multidimensional movement vector configuration unit 42 may output a plurality of multidimensional movement vector data to the output unit 50.

When constructing one multidimensional movement vector, the multidimensional movement vector component 42 selects two consecutive time points in the time series of acquiring biological marker data to form the multidimensional movement vector. The multidimensional movement vector component 42 is continuous with the first time point and the first time point among the plurality of time points from which the biological marker data was acquired (that is, the biological marker data after the first time point). Select the second time point (which was obtained).
As shown in FIG. 4, the multidimensional movement vector component 42 receives from the first multidimensional vector a1 and the second multidimensional vector a2 between the first time point and the second time point (that is, measurement points m1 and m2). The first multidimensional movement vector b1 of (between) is constructed. When the acquisition cycle of biological marker data is 10 seconds, the second time point is a time point 10 seconds after the first time point. Further, the acquisition cycle of the biological marker data is set to be shorter than, for example, 10 seconds, and the moving average of the biological markers obtained within the interval of every 10 seconds is calculated to be the first time point, the second time point, and so on.・ ・ It may be the value at the Nth time point.

Similarly, the multidimensional movement vector component 42 selects a second time point and a third time point continuous with the second time point.
As shown in FIG. 4, the multidimensional movement vector component 42 has a second time point and a third time point from the second multidimensional vector a2 at the second time point and the third multidimensional vector a3 at the third time point. The second multidimensional movement vector b2 between the time points (that is, between the measurement points m2 and m3) is constructed. In this way, the multidimensional movement vector component 42 constructs a plurality of multidimensional movement vectors between two consecutive time points among the plurality of time points for which the biological marker data has been acquired. The multidimensional movement vector component 42 obtains the first (N-1) multidimensional vector a (N-1) from the first multidimensional movement vector b1 based on the first multidimensional vector a1 to the Nth multidimensional vector aN. Configure.

(Multidimensional movement vector length calculation unit)
The multidimensional movement vector length calculation unit 43 calculates the lengths of the plurality of multidimensional movement vectors received from the multidimensional movement vector configuration unit 42, respectively. The multidimensional movement vector length calculation unit 43 can calculate existing distances such as the Euclidean distance, the Mahalanobis distance, the Manhattan distance, and the inner product as the length of the multidimensional movement vector (distance between continuous measurement points).
The multidimensional movement vector length calculation unit 43 outputs the calculated multidimensional movement vector length data (hereinafter, referred to as “multidimensional movement vector length data”) to the tortuosity calculation unit 45.

(Multidimensional movement vector angle calculation unit)
The multidimensional movement vector inter-angle angle calculation unit 44 calculates the angle θ of two consecutive multidimensional movement vectors from the plurality of multidimensional movement vectors received from the multidimensional movement vector constituent unit 42. Here, the "angle θ of two continuous multidimensional movement vectors" is because one multidimensional movement vector points in the same direction as the other multidimensional movement vector continuous with the one multidimensional movement vector. The angle of rotation. More specifically, as shown in FIG. 4, for example, the first multidimensional movement vector b1 between the first time point and the second time point is the second multidimensional movement between the second time point and the third time point. The angle θ2 that rotates to face the same direction as the vector b2 is an example of “the angle θ of two continuous multidimensional movement vectors”. The angle θ is different from the second multidimensional movement vector b2 when the second multidimensional movement vector b2 is moved so that the start point of the second multidimensional movement vector b2 coincides with the start point of the first multidimensional movement vector b1. This is the smaller angle formed by the first multidimensional movement vector b1.
The multidimensional movement vector-to-vector angle calculation unit 44 outputs the angle data of the multi-dimensional movement vector (hereinafter, referred to as “multidimensional movement vector-to-vector angle data”) to the bending degree calculation unit 45.

The multidimensional movement vector inter-angle angle calculation unit 44 selects two consecutive multidimensional movement vectors from the first multidimensional movement vector b1 to the (N-1) multidimensional movement vector b (N-1), and selects two consecutive multidimensional movement vectors. Calculate the angle between two consecutive multidimensional movement vectors. For example, as shown in FIG. 4, the multidimensional movement vector-to-angle angle calculation unit 44 selects the first multidimensional movement vector b1 and the second multidimensional movement vector b2 as two consecutive multidimensional movement vectors.

When the angle between the first multidimensional movement vector b1 and the second multidimensional movement vector b2 is the second angle θ2, and the first multidimensional movement vector b1 and the second multidimensional movement vector b2 are n-dimensional movement vectors. To do. In this case, the first multidimensional movement vector b1 is indicated by (b1 ₁ , b1 ₂ , ..., b1 _n ), and the second multidimensional movement vector b2 is indicated by (b2 ₁ , b2 ₂ , ..., b2 _n ). Is done. The angle calculation unit 44 between the multidimensional movement vectors sets the inner product of the first multidimensional movement vector b1 and the second multidimensional movement vector b2 as the length of the first multidimensional movement vector b1 and the length of the second multidimensional movement vector b2. The second angle θ2 [rad] can be calculated by dividing by the product of. Further, the multidimensional movement vector-to-vector angle calculation unit 44 may convert the second angle θ2 [rad] indicated by the radian angle into the second angle θ2'[°] indicated by the angle. Further, the multidimensional movement vector-to-vector angle calculation unit 44 can calculate the third (N-1) angle θ (N-1) from the third angle θ3 by the same calculation.

(Tortuosity calculation unit)
The tortuosity calculation unit 45 calculates the tortuosity between two consecutive multidimensional movement vectors based on the length and angle of each of the two consecutive multidimensional movement vectors. In the present disclosure, "tortuosity" is defined as, for example, a value obtained by subtracting the inner product of the two multidimensional movement vectors from the product of the lengths of two consecutive multidimensional movement vectors.

The case where the two multidimensional movement vectors are the first multidimensional movement vector b1 and the second multidimensional movement vector b2 will be described. In this case, it is the angle (second angle θ2) between the first multidimensional movement vector b1 and the second multidimensional movement vector b2. In this case, the tortuosity BP (second tortuosity BP2) can be expressed by the following equation (1). Here, "BP" means Bending Power.
Tortuosity BP = | b1 | ・ | b2 |-| b1 |
= | B1 | ・ | b2 | − | b1 | ・ (| b2 | ・ cosθ2)
= | B1 | · | b2 | (1-cosθ2) ... Equation (1)
Here, | b2'| is the length of the mapping vector of the second multidimensional movement vector b2 onto the first multidimensional movement vector b1, and is | b2'| = | b2 | · cosθ2. When the second angle θ2 is 0 ° or more and 90 ° or less, 0 ≦ cos θ2 ≦ 1 and the tortuosity BP is 0 or more and | b1 | ・ | b2 | or less. When the second angle θ2 is 90 ° or more and 180 ° or less, -1 ≦ cos θ2 <0, and the tortuosity BP is | b1 | ・ | b2 | or more and 2 | b1 | ・ | b2 | or less.
The tortuosity calculation unit 45 can calculate the (N-1) tortuosity BP (N-1) from the third tortuosity BP3 by the same calculation.

As another definition of "tortuosity", the tortuosity BP (second tortuosity BP2) represented by the following formula (2) and the tortuosity BA (second tortuosity BA2) represented by the formula (3) are also defined. Conceivable.
Tortuosity BS = | b2 |-| b2'|
= | B2 |-| b2 | ・ cosθ2
= | B2 | (1-cosθ2) ... Equation (2)
Tortuosity BA = 1-cos θ2 ・ ・ ・ Equation (3)
Here, "BS" means Bending Strength. Also, "BA" means Bending Angle.
In the present disclosure, the invention is disclosed in the case of the tortuosity BP, but the same can be applied to the cases of the tortuosity BS and the tortuosity BA.

As can be seen from the equation (1), the tortuosity BP has a large angle θ (close to 180 °) between two consecutive vectors when the product of the lengths of two consecutive multidimensional movement vectors is a constant value. ) The larger the value. Further, when the angle between two continuous vectors is a constant value, the tortuosity BP becomes a larger value as the product of the lengths of the two continuous multidimensional movement vectors is larger. This indicates that the larger the angle θ or the larger the product of the lengths of two continuous vectors, the greater the change in the state of cardiopulmonary function. That is, such tortuosity is considered to be one phenotype of the change in the cardiopulmonary function state of the measurement subject.
The tortuosity calculation unit 45 outputs the tortuosity data (hereinafter referred to as “tortuosity data”) between two consecutive multidimensional movement vectors to the tortuosity local maximum point extraction unit 46.

(Tortuosity local maximum point extraction part)
The tortuosity local maximum point extraction unit 46 extracts one or more tortuosity local maximum points from a plurality of tortuosity degrees based on a plurality of tortuosity data input from the tortuosity calculation unit 45. The tortuosity is the local maximum point. The tortuosity local maximum point extraction unit 46 compares, for example, two tortuosities input after the start of the load test, sets the larger tortuosity as the local maximum tortuosity, and at that time until the load test is completed thereafter. Compares the tortuosity local maximum point of Update as.
The tortuosity local maximum point extraction unit 46 stores the tortuosity local maximum point in a memory (not shown) or the like, and reads out the tortuosity local maximum point at that time from the memory at the time of extracting the tortuosity maximum point. Further, the tortuosity local maximum point extraction unit 46 outputs the extracted tortuosity local maximum point to the cardiopulmonary function state change point estimation unit 47.

(Cardiopulmonary function state change point estimation part)
The cardiopulmonary function state change point estimation unit 47 extracts and extracts the tortuosity local maximum point input from the tortuosity local maximum point extraction unit 46 with reference to the exercise intensity data input from the exercise load control unit 10. The local maximum point of tortuosity is estimated as the change point of the cardiopulmonary function state of the measurement subject. As an example of the change point of the cardiopulmonary function state estimated by the cardiopulmonary function state change point estimation unit 47, for example, an anaerobic threshold (AT) or a respiratory compensation start point (RC) can be mentioned.

When the cardiopulmonary function state change estimation device 1 has information indicating the relationship between the exercise load and the time from the start of the load test, the cardiopulmonary function state change point estimation unit 47 corresponds to the local maximum point of tortuosity. The time from the start of the stress test may be extracted and used as the change point of the cardiopulmonary function state. When the storage unit 30 stores information indicating the relationship between the exercise load generated based on the exercise intensity data from the exercise load control unit 10 and the time from the start of the load test, the cardiopulmonary function state change point estimation unit For 47, the time from the start of the load test corresponding to the local maximum flexion point can be set as the change point of the cardiopulmonary function state.
The cardiopulmonary function state change point estimation unit 47 outputs data indicating the estimated change point of the cardiopulmonary function state (hereinafter, also referred to as “cardiopulmonary function state change point data”) to the output unit 50. The cardiopulmonary function state change point data is, for example, exercise intensity data or data of the time from the start of the stress test.

<Output section>
The output unit 50 is, for example, a display unit of a liquid crystal display, an organic electro-luminescence (OEL) display, or the like. The display unit displays information indicating the change point or change point of the cardiopulmonary function state such as the anaerobic metabolism threshold value or the respiratory compensation start point estimated by the cardiopulmonary function state change point estimation unit 47.
For example, as shown in FIG. 5, the display unit is a change point of the cardiopulmonary function state on a graph showing the degree of flexion with respect to the exercise intensity or the information indicating the exercise intensity based on the exercise load data input from the exercise load control unit 10. (Indicated by ⊚ in FIG. 5) may be indicated. In this case, the display unit inputs the cardiopulmonary function state change point data input from the cardiopulmonary function state change point estimation unit 47, the flexion degree data input from the flexion degree calculation unit 45, and the exercise load control unit 10 or the storage unit 30. The graph shown in FIG. 5 is generated and displayed based on the exercise intensity data obtained.

In FIG. 5, the time from the start of the load test is displayed as information indicating the exercise intensity. The storage unit 30 stores data showing the relationship between the time from the start of the load test and the load intensity based on the data transmitted from the exercise load control unit 10. Therefore, the cardiopulmonary function state change estimation device 1 can obtain an exercise load corresponding to the change point of the cardiopulmonary function state from the relationship between the time from the start of the load test and the tortuosity.
As information indicating the change point of the cardiopulmonary function state, the display unit displays the exercise intensity corresponding to the change point of the cardiopulmonary function state (for example, the exercise intensity indicating the anaerobic metabolic threshold and the respiratory compensation start point) by character / numerical information or the like. It may be displayed.

Figure 6 shows, for example, "Journal of the Japanese Society of Internal Medicine, Vol. 101, No. 6, June 10, 2012, Special Feature COPD: Advances in Diagnosis and Treatment Topics II. Diagnosis and Examination 3. Cardiopulmonary Exercise Load Examination (CPET)" I drew the figure published in. 6, together with the degree of curvature, the amount of exercise, oxygen uptake _V O2 obtained by CXP test [ml / min], carbon dioxide emissions _V CO2 [ml / min], minute ventilation _V E [l / min ], And the behavior of lactic acid, hydrogen ion index pH, etc. obtained by blood test. Here, in FIG. 6, oxygen uptake V _O2 , carbon dioxide emission V _CO2 , minute ventilation, lactic acid, and hydrogen ion index pH show the behavior of each index with respect to exercise load, and the amount of each index and It does not indicate the concentration.

As shown in FIG. 6, by estimating the change point of the cardiopulmonary function state by the cardiopulmonary function state change estimation device 1, the conventionally known change of the cardiopulmonary function state such as the anaerobic metabolism threshold AT and the respiratory compensation start point RC Change points other than points can also be detected.
For example, as shown in FIG. 6, _{the change points of the minute ventilation volume VE} , the carbon dioxide emission amount V _CO2, and the slope of lactic acid are the anaerobic metabolism threshold AT. The cardiopulmonary function state change estimation device 1 detects the peak LP2 of the tortuosity at the exercise load corresponding to the change points of the minute ventilation volume _VE , the carbon dioxide emission amount V _{CO2, and the inclination of lactic acid.} The cardiopulmonary function state change estimation device 1 determines that the exercise load corresponding to the peak is the anaerobic metabolism threshold AT.

Further, as shown in FIG. 6, the point at which the slopes of only the _{minute ventilation volume VE} and the carbon dioxide emission amount V _{CO2 change is the respiratory compensation start point RC.} The cardiopulmonary function state change estimation device 1 detects the peak LP4 of the tortuosity at the exercise load corresponding to the change point of the slope of the minute ventilation volume _VE and the carbon dioxide emission amount V _CO2. The cardiopulmonary function state change estimation device 1 determines that the exercise load corresponding to the peak is the respiratory compensation start point RC.

On the other hand, as shown in FIG. 6, peak LP0 and peak LP1 may occur during exercise load before reaching the anaerobic metabolism threshold AT. Peak LP0 and peak LP1 are minor compensatory mechanisms, and peak LP0 often does not appear. Further, in the region between the anaerobic metabolism threshold AT and the respiratory compensation start point RC, peak LP3 may occur. Even at such a peak LP3 position, a compensation mechanism different from that of, for example, peak LP0 and peak LP1 occurs. It is thought that it has occurred.
Such tortuosity peaks often apply to the behavior of tortuosity in patients with heart failure. In addition, the behavior of the tortuosity of healthy subjects tends to be observed to be more dynamic.

(1.2) Cardiopulmonary function state estimation method and cardiopulmonary function state estimation program The cardiopulmonary function state change estimation device 1 according to the present embodiment includes at least one processor, and the processor includes an exercise load control unit 10 and data acquisition. It functions as an electrocardiographic processing unit 22, a heart sound processing unit 24, a storage unit 30, and a cardiopulmonary function state change estimation unit 40 of the unit 20.

The cardiopulmonary function state change estimation device 1 according to the present embodiment acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement target person from the measurement target person who is exercising at each of the plurality of time points. In the cardiopulmonary function state change estimation device 1, a multidimensional movement vector between two points at a plurality of time points is constructed and continuous based on a multidimensional vector at a plurality of time points composed of biological marker data. A cardiopulmonary function state estimation method is executed in which the exercise load corresponding to the local maximum point of the degree of flexion of the two multidimensional movement vectors is estimated as the change point of the cardiopulmonary function state of the measurement subject.

Further, the cardiopulmonary function state change estimation device 1 includes at least one processor that executes an instruction and a storage unit that stores a cardiopulmonary function state estimation program. The cardiopulmonary function state estimation program acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of multiple time points, and is composed of the biological marker data. Based on the multidimensional vectors at multiple time points, a multidimensional movement vector between two points at multiple time points is constructed, and corresponds to the local maximum point of the degree of bending of two consecutive multidimensional movement vectors. The processor is made to execute an instruction to estimate the exercise load as a change point of the cardiopulmonary function state of the person to be measured.

Hereinafter, the cardiopulmonary function state estimation method will be described in more detail with reference to FIG. 1 and FIG. 7.
As shown in FIG. 7, in step S11, the cardiopulmonary function state change estimation device 1 starts a load test. In the cardiopulmonary function state change estimation device 1, the exercise load control unit 10 controls the load device 60 so as to increase the exercise intensity at a constant rate with time. The exercise load control unit 10 applies an exercise load that allows the person to be measured to rest for several minutes after the start of the load test, and sets the load device 60 so that the exercise intensity increases at a constant rate over time from several minutes after the start of the load test. You may control it. In this way, the cardiopulmonary function state change estimation device 1 applies an exercise load to the measurement target person who operates the load device 60 such as an ergometer.

In step S12, the cardiopulmonary function state change estimation device 1 acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of the plurality of time points. In the cardiopulmonary function state change estimation device 1, at least one of the electrocardiographic data or the heartbeat data of the measurement subject is generated by at least one of the electrocardiographic collection unit 21 or the heartbeat collection unit 23 attached to the measurement subject who is exercising. To do. The electrocardiographic collection unit 21 and the cardiac sound collection unit 23 acquire electrocardiographic data or cardiac sound data at regular intervals (for example, every 10 [seconds]) from the start of the load test. In the cardiopulmonary function state change estimation device 1, the electrocardiographic processing unit 22 generates biological marker data derived from the electrocardiographic data, or the cardiac sound processing unit 24 generates biological marker data derived from the cardiac sound data. After that, the electrocardiographic processing unit 22 and the heart sound processing unit 24 output each of the generated biological marker data to the storage unit 30 and the cardiopulmonary function state change estimation unit 40.

In step S13, the cardiopulmonary function state change estimation device 1 constructs a multidimensional vector from the acquired biological marker data. In the cardiopulmonary function state change estimation device 1, the multidimensional vector component 41 reflects a plurality of (n types) biological marker data obtained from the measurement target person at the same timing in the multidimensional space (n-dimensional space). A multidimensional (n-dimensional) vector indicating the measured points is constructed. The cardiopulmonary function state change estimation device 1 constructs a plurality of multidimensional vectors based on biological marker data acquired at a plurality of time points.

In step S14, the cardiopulmonary function state change estimation device 1 selects two consecutive time points from a plurality of time points and constructs a multidimensional movement vector between the two multidimensional vectors at the two selected time points. In the cardiopulmonary function state change estimation device 1, the multidimensional movement vector component 42 constitutes a multidimensional movement vector between two multidimensional vectors at all two consecutive time points.

In step S15, the cardiopulmonary function state change estimation device 1 calculates the tortuosity between two consecutive multidimensional movement vectors. In the cardiopulmonary function state change estimation device 1, the multidimensional movement vector length calculation unit 43 calculates the length of each multidimensional movement vector, and the multidimensional movement vector angle calculation unit 44 is between two continuous multidimensional movement vectors. Calculate the angle. The bending degree calculation unit 45 calculates the inner product of the two consecutive multidimensional movement vectors from the product of the lengths of the two consecutive multidimensional movement vectors based on the length and angle of each of the two consecutive multidimensional movement vectors. The subtracted value is calculated and used as the degree of bending.

In step S16, the cardiopulmonary function state change estimation device 1 estimates the cardiopulmonary function change point based on a plurality of tortuosities. In the cardiopulmonary function state change estimation device 1, the tortuosity local maximum point extraction unit 46 extracts one or more tortuosity, which is the maximum point, out of a plurality of tortuosities, and sets it as the maximum tortuosity point. The part 47 estimates the exercise load corresponding to the maximum tortuosity point as the change point of the cardiopulmonary function state of the measurement subject.
In step S17, the cardiopulmonary function state change estimation device 1 presents the estimated change point of the cardiopulmonary function state to the output unit. For example, the cardiopulmonary function state change estimation device 1 causes the display unit to display the change point of the cardiopulmonary function state. As a result, the cardiopulmonary function state change estimation device 1 can present the change point of the cardiopulmonary function state to the user or the like of the cardiopulmonary function state change estimation device.

(1.3) Use of cardiopulmonary function state change estimation device As described above, the user of the cardiopulmonary function state change estimation device can use the cardiopulmonary function of the measurement target person based on the change point of the cardiopulmonary function state displayed on the output unit 50. You can check the status.
When the measurement target is a heart disease patient or a heart failure patient, the cardiopulmonary function state change estimation device 1 determines the anaerobic metabolic threshold (AT) or the respiratory compensation start point as the change point of the cardiopulmonary function state of the heart disease patient or the heart failure patient. (RC) can be determined. The cardiopulmonary function state change estimation device 1 determines the optimum exercise intensity when performing exercise therapy as a treatment for lifestyle-related diseases, heart diseases, etc., based on the anaerobic metabolic threshold (AT) and the respiratory compensation start point (RC). To do. A medical worker such as a doctor who is a user of the cardiopulmonary function state change estimation device 1 can give guidance on exercise therapy based on the change point of the cardiopulmonary function state and the optimum exercise intensity of the measurement target person.
Similarly, when the measurement target is a healthy person who has not yet reached heart disease or heart failure, the cardiopulmonary function state change estimation device 1 is based on the anaerobic metabolic threshold (AT) and the respiratory compensation start point (RC). , Determine the optimal exercise intensity when exercising for the prevention of lifestyle-related diseases.

When the measurement target is a healthy person such as an athlete, the cardiopulmonary function state change estimation device 1 is the optimum exercise for training based on the anaerobic metabolic threshold (AT) and the respiratory compensation start point (RC). The strength can be determined. A sports instructor or the like who is a user of the cardiopulmonary function state change estimation device 1 can provide training guidance based on the change point of the cardiopulmonary function state and the optimum exercise intensity of the measurement target person.
Similarly, when the measurement target is an animal (for example, a racehorse), the cardiopulmonary function state change estimation device 1 can determine the optimum exercise intensity for training and perform training.

<Modification example>
(1) First Modified Example In the present embodiment, an example in which the output unit 50 of the cardiopulmonary function state change estimation device 1 is a display unit has been described, but the present invention is not limited to such a configuration. For example, the cardiopulmonary function state change estimation device 1 may have a communication unit as an output unit 50 for connecting to an external terminal via a network such as the Internet. In this case, the cardiopulmonary function state change estimation unit 40 transmits cardiopulmonary function state change point data to an external terminal (not shown) via the output unit 50 which is a communication unit.

The external terminal may be, for example, a terminal having a display unit such as a personal computer, a smartphone, or a tablet computer, a terminal having a printing function such as a printer, or a terminal having an audio output function such as a speaker.
When the external terminal is a terminal having a printing function, information indicating a change point or a change point of the cardiopulmonary function state is printed on paper or the like. When the external terminal is a terminal having a voice output function, the speaker of the external terminal outputs the change point of the cardiopulmonary function state or the information indicating the change point by voice.

When the external terminal is a terminal having a display unit, the cardiopulmonary function state change estimation device 1 causes the display unit of the external terminal to display a change point or information indicating the change point of the cardiopulmonary function state. Further, the cardiopulmonary function state change estimation device 1 may transmit the exercise load data output from the exercise load control unit 10 and the tortuosity data output from the tortuosity calculation unit 45 to the external terminal. In this case, the cardiopulmonary function state change estimation device 1 displays a change in the cardiopulmonary function state on the display unit of the external terminal on a graph showing the degree of flexion with respect to the information indicating the exercise intensity based on the exercise load data and the degree of flexion data. Points can be displayed.

(2) Second Modified Example In the present embodiment, the output unit 50 of the cardiopulmonary function state change estimation device 1 is a display unit, and information indicating a change point or a change point of the cardiopulmonary function state estimated based on the tortuosity is displayed. An example of display has been described, but the configuration is not limited to this. For example, the output unit 50, which is a display unit, may display a graph depicting each of the plurality of biological marker data used for constructing the multidimensional vector and the multidimensional movement vector. That is, the output unit 50, which is a display unit, may draw a plurality of biological marker data on a graph of tortuosity or on a graph different from the graph of tortuosity. Further, the output unit 50, which is a display unit, has a biological marker value derived from electrocardiographic data such as heart rate (HR) and electromechanical activity time (EMAT), and S1 intensity, S2 intensity, and S3 intensity. And the values of biological markers derived from heartbeat data such as S4 intensity may be depicted on different graphs.

(3) Third Modified Example In the present embodiment, an example in which the cardiopulmonary function state change estimation device 1 has both an electrocardiographic collecting unit 21 and a heart sound collecting unit 23 has been described, but the configuration is not limited to this. When it is preferable to use only a plurality of biological marker data that can be obtained from only either electrocardiographic data or cardiac sound data as the biological marker data used to obtain the tortuosity, the cardiopulmonary function state change estimation device 1 The data acquisition unit 20 may include only one of the electrocardiographic collection unit 21 and the heart sound collection unit 23.

(4) Fourth Modified Example In the present embodiment, an example in which the calculation formula (formula (1)) is used to obtain the bending degree BP has been described, but the present invention is not limited to such a configuration. For example, the cardiopulmonary function state change estimation device 1 has data indicating the length and angle of two consecutive multidimensional movement vectors in order to obtain the tortuosity BP, and the change point of the cardiopulmonary function state obtained by the CPX test (the change point of the cardiopulmonary function state (CPX test). It may have a trained model generated by machine learning using the anaerobic metabolism threshold AT) as teacher data. In addition, in order to obtain the tortuosity BP, a trained model generated by machine learning the biological marker data acquired at a plurality of time points and the change point of the cardiopulmonary function state obtained by the CPX test as teacher data is used. You may use it. In addition, the cardiopulmonary function state change estimation device 1 may have such a trained model in order to obtain an anaerobic metabolism threshold AT and a respiratory compensation start point (RC).

<Effect of the first embodiment>
The cardiopulmonary function state change estimation device 1 according to the first embodiment has the following effects.
(1) The cardiopulmonary function state change estimation device 1 constructs a multidimensional vector and a multidimensional movement vector based on a plurality of biological marker data acquired from a measurement subject, and between two consecutive multidimensional movement vectors. The degree of change in cardiopulmonary function is estimated using the degree of flexion. The tortuosity between two consecutive multidimensional movement vectors is an index that is totally influenced by a plurality of biological marker data. Therefore, the cardiopulmonary function state change estimation device 1 can estimate the change point of the cardiopulmonary function state with high accuracy regardless of the degree of the heart disease of the measurement target person and the cardiopulmonary function state peculiar to the measurement target person.

(2) The cardiopulmonary function state change estimation device 1 is less likely to cause variation in the estimation result of the change point of the cardiopulmonary function state as compared with the case where a person judges. Therefore, the difference in diagnosis between doctors and the like is unlikely to occur, and the cardiopulmonary function state can be estimated with high accuracy.
(3) The cardiopulmonary function state change estimation device 1 can detect changes other than the conventionally known change points of the cardiopulmonary function state such as the anaerobic metabolism threshold (AT) and the respiratory compensation start point (RC). .. For example, the cardiopulmonary function state change estimation device 1 has a peak (for example, peak LP0 and peak LP1) showing a minor compensatory mechanism that occurs before reaching the anaerobic metabolic threshold AT, or an anaerobic metabolic threshold (AT) and respiratory compensation initiation. The peak (for example, peak LP3) generated in the region between the point (RC) can be detected accurately.

2. Second Embodiment (2.1) Configuration of Cardiopulmonary Function State Change Estimating Device Hereinafter, the cardiopulmonary function state change estimating device 2 according to the second embodiment will be described with reference to FIGS. 1 to 7 and FIG. To do.
The cardiopulmonary function state change estimation device 2 includes an exercise load control unit 10, a data acquisition unit 20, a storage unit 30, a cardiopulmonary function state change estimation unit 140, and an output unit 50. That is, the cardiopulmonary function state change estimation device 2 is different from the cardiopulmonary function state change estimation device 1 in that the cardiopulmonary function state change estimation unit 140 is provided in place of the cardiopulmonary function state change estimation unit 40.
Hereinafter, the memory unit 30 and the cardiopulmonary function state change estimation unit 140 will be described. Since the exercise load control unit 10, the data acquisition unit 20, and the output unit 50 have the same configurations as those described in the first embodiment, description thereof will be omitted.

<Memory>
Similar to the first embodiment, the storage unit 30 outputs the exercise intensity data output by the exercise load control unit 10, the electrocardiographic data and biological markers output by the electrocardiographic processing unit 22, and the cardiac sound processing unit 24. Store each of the heartbeat data and biological markers.
In addition, the storage unit 30 has data (hereinafter, "cardiopulmonary function change point") indicating a predetermined area of exercise load estimated to include the cardiopulmonary function change point (hereinafter, referred to as "cardiopulmonary function change point existence region"). "Existence area data") is stored. Cardiopulmonary function changes include, for example, the anaerobic metabolic threshold (AT) or the respiratory compensation initiation point (RC). The storage unit 30 stores at least one of the anaerobic metabolism threshold existence region data indicating the anaerobic metabolism threshold existence region and the respiratory compensation start point region data indicating the respiratory compensation start point region. In addition, the storage unit 30 may store the cardiopulmonary function change point existence region other than the anaerobic metabolism threshold existence region and the respiratory compensation start point region as data.

The anaerobic metabolism threshold presence region is a region in which the anaerobic metabolism threshold (AT) is presumed to be present in the range of exercise load to be increased during the stress test. In addition, the respiratory compensation start point region is a region of exercise load in which the respiratory compensation start point (RC) is presumed to exist in the range of exercise load to be increased during the load test. The anaerobic metabolism threshold existence region and the respiratory compensation starting point region may be estimated, for example, by a prior CPX test or the like, or may be estimated based on conventional knowledge.

For example, when a cardiopulmonary exercise load test is performed as a load test, a local peak of tortuosity may occur in the initial section of the gradual increase load (ramp load) even though it is not a change point of the cardiopulmonary function state. The region where the cardiopulmonary function change point exists is set in order to exclude the local peak of the tortuosity that hinders the detection of the change point of the cardiopulmonary function state and detect the change point of the cardiopulmonary function state with high accuracy.

<Cardiopulmonary function state change estimation unit>
FIG. 8 is a block diagram showing a configuration example of a cardiopulmonary function state change estimation unit 140 included in the cardiopulmonary function state change estimation device 2. As shown in FIG. 8, the cardiopulmonary function state change estimation unit 140 calculates the angle between the multidimensional movement vector constituent unit 41, the multidimensional movement vector constituent unit 42, the multidimensional movement vector length calculation unit 43, and the multidimensional movement vector. A unit 44, a flexion degree calculation unit 45, a flexion degree local maximum point extraction unit 146, and a cardiopulmonary function state change point estimation unit 147 are provided. That is, the cardiopulmonary function state change estimation unit 140 includes a tortuosity local maximum point extraction unit 146 and a cardiopulmonary function state change point estimation unit 147 in place of the tortuosity local maximum point extraction unit 46 and the cardiopulmonary function state change point estimation unit 47. In that respect, it differs from the cardiopulmonary function state change estimation unit 40 of the first embodiment.

Hereinafter, the tortuosity local maximum point extraction unit 146 and the cardiopulmonary function state change point estimation unit 147 will be described. The multidimensional vector configuration unit 41, the multidimensional movement vector configuration unit 42, the multidimensional movement vector length calculation unit 43, the multidimensional movement vector angle calculation unit 44, and the bending degree calculation unit 45 have the same configurations as described in the first embodiment. Since the same is true, the description thereof will be omitted.

(Tortuosity local maximum point extraction part)
The tortuosity local maximum point extraction unit 146 determines the tortuosity local maximum point based on the cardiopulmonary function change point existence region data input from the storage unit 30 and a plurality of tortuosity local maximum points input from the tortuosity calculation unit 45. Extract. The tortuosity local maximum point extraction unit 146 compares the two tortuosities input in the range of the load intensity of the cardiopulmonary function change point existence region based on the cardiopulmonary function change point existence region data, and bends the larger tortuosity. Degree Local maximum point. The tortuosity local maximum point extraction unit 146 compares the tortuosity local maximum point at that time with the input tortuosity until the range of the load intensity in the region where the cardiopulmonary function change point exists is completed, and the input tortuosity local maximum point extraction unit 146. If is greater than the current tortuosity local maximum point, the input tortuosity is updated as a new tortuosity local maximum point.
The tortuosity local maximum point extraction unit 146 stores the tortuosity local maximum point in a memory (not shown) or the like, and reads out the tortuosity local maximum point at that time from the memory at the time of extracting the tortuosity local maximum point. Further, the tortuosity local maximum point extraction unit 146 outputs the extracted tortuosity local maximum point to the cardiopulmonary function state change point estimation unit 147.

(Cardiopulmonary function state change point estimation part)
The cardiopulmonary function state change point estimation unit 147 refers to the exercise intensity data input from the exercise load control unit 10 and calculates the exercise load corresponding to the tortuosity local maximum point input from the tortuosity local maximum point extraction unit 146. It is extracted, and the extracted exercise load is estimated as the change point of the cardiopulmonary function state of the measurement subject.
The cardiopulmonary function state change point estimation unit 147 outputs cardiopulmonary function state change point data indicating the change point of the cardiopulmonary function state to the output unit 50.

(2.2) Cardiopulmonary function state estimation method and cardiopulmonary function state estimation program The cardiopulmonary function state change estimation device 2 according to the present embodiment includes at least one processor, and the processor includes an exercise load control unit 10 and data acquisition. It functions as an electrocardiographic processing unit 22, a heart sound processing unit 24, a storage unit 30, and a cardiopulmonary function state change estimation unit 140 of the unit 20.

The cardiopulmonary function state change estimation device 2 according to the present embodiment acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement target from the measurement target who is exercising at each of the plurality of time points. Based on the multidimensional vector at multiple time points composed of biological marker data, a multidimensional movement vector between two points at multiple time points is constructed respectively, and the degree of bending is formed from two consecutive multidimensional movement vectors. Is calculated, the local maximum point of the degree of flexion is extracted from the region where the cardiopulmonary function change point exists, and the cardiopulmonary function state estimation method is executed in which the local maximum point is estimated as the change point of the cardiopulmonary function state of the measurement subject.

Further, the cardiopulmonary function state change estimation device 1 includes at least one processor that executes an instruction and a storage unit that stores a cardiopulmonary function state change estimation program. The cardiopulmonary function state change estimation program acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of multiple time points, and obtains multiple biological marker data related to the cardiopulmonary function of the measurement subject from the biological marker data. Based on the constructed multidimensional movement vectors at a plurality of time points, a multidimensional movement vector between two points at a plurality of time points is constructed, the degree of bending is calculated from two consecutive multidimensional movement vectors, and the degree of bending is calculated. The local maximum point is extracted from the region where the cardiopulmonary function change point exists, and the processor is made to execute an instruction to estimate the local maximum point as the change point of the cardiopulmonary function state of the measurement subject.

In the cardiopulmonary function state change estimation method of the present embodiment, in step S16 shown in FIG. 7, the local maximum point of the tortuosity of two consecutive multidimensional movement vectors is extracted from the cardiopulmonary function change point existence region and used as the local maximum point. It differs from the cardiopulmonary function state estimation method of the first embodiment in that the local maximum point is estimated as the change point of the cardiopulmonary function state of the measurement subject. Steps S11 to S15 and S17 are the same as the cardiopulmonary function state change estimation method of the first embodiment.

<Modification example>
(1) First Modification Example In the present embodiment, the local maximum tortuosity extraction unit 146 extracts the local maximum tortuosity from a predetermined region where a cardiopulmonary function change point is presumed to exist, and changes in the cardiopulmonary function state. The example of the point has been described, but the configuration is not limited to this. For example, the local maximum tortuosity extraction unit 146 may use the local maximum tortuosity as the change point of the cardiopulmonary function state in a predetermined region where the cardiopulmonary function change point is presumed to exist. Here, the "local maximum point of the first tortuosity" means that among the local maximum points existing in the region where the cardiopulmonary function change point exists, the exercise load when the local maximum point occurs is the smallest, or from the start of the load test. The local maximum point with a short time.
(2) Second Modified Example In the first modified example of the present embodiment, an example in which the local maximum point of the first tortuosity in a predetermined region where the cardiopulmonary function change point is presumed to exist is set as the change point of the cardiopulmonary function state will be described. However, it is not limited to such a configuration. For example, when the tortuosity local maximum point extraction unit 146 detects a local maximum point with a tortuosity greater than or equal to a predetermined value based on the rate of change or inclination of the tortuosity, the local maximum point is set to the cardiopulmonary function state. It may be a change point of.

<Effect of the second embodiment>
The cardiopulmonary function state change estimation device 2 according to the second embodiment has the following effects in addition to the effects described in the first embodiment.
(4) In the cardiopulmonary function state change estimation device 2, the local maximum point of the tortuosity of two consecutive multidimensional movement vectors is extracted from a predetermined cardiopulmonary function change point existence region and used as the local maximum point. The point is estimated as the change point of the cardiopulmonary function state of the measurement subject. Therefore, the cardiopulmonary function state change estimation device 2 excludes the local maximum point of tortuosity that occurs in a region other than the change point of the cardiopulmonary function state, which hinders the detection of the change point of the cardiopulmonary function state, and the cardiopulmonary function state. The change point of can be detected with high accuracy.

3. 3. Third Embodiment (3.1) Configuration of Cardiopulmonary Function State Change Estimating Device Hereinafter, the cardiopulmonary function state change estimating device 3 according to the third embodiment will be described with reference to FIGS. 1 to 8 with reference to FIG. To do.
The cardiopulmonary function state change estimation device 3 includes an exercise load control unit 10, a data acquisition unit 20, a storage unit 30, a cardiopulmonary function state change estimation unit 240, and an output unit 50. That is, the cardiopulmonary function state change estimation device 3 is different from the cardiopulmonary function state change estimation device 1 in that the cardiopulmonary function state change estimation unit 240 is provided in place of the cardiopulmonary function state change estimation unit 40.
Hereinafter, the memory unit 30 and the cardiopulmonary function state change estimation unit 240 will be described. Since the exercise load control unit 10, the data acquisition unit 20, and the output unit 50 have the same configurations as those described in the first embodiment, description thereof will be omitted.

<Cardiopulmonary function state change estimation unit>
FIG. 9 is a block diagram showing a configuration example of a cardiopulmonary function state change estimation unit 240 included in the cardiopulmonary function state change estimation device 3. As shown in FIG. 9, the cardiopulmonary function state change estimation unit 240 calculates the angle between the multidimensional movement vector constituent unit 41, the multidimensional movement vector constituent unit 42, the multidimensional movement vector length calculation unit 43, and the multidimensional movement vector. A unit 44 and a bending degree calculation unit 45 are provided. Further, the cardiopulmonary function state change estimation unit 240 includes a tortuosity local maximum point extraction unit 146, a cardiopulmonary function state change point estimation unit 147, a two-dimensional coordinate calculation unit 248, and a cardiopulmonary function state classification unit 249. There is. That is, the cardiopulmonary function state change estimation unit 240 includes a flexion local maximum point extraction unit 146 and a cardiopulmonary function state change point estimation unit 147 in place of the flexion local maximum point extraction unit 46 and the cardiopulmonary function state change point estimation unit 47. Further, it is different from the cardiopulmonary function state change estimation unit 40 of the first embodiment in that it includes a two-dimensional coordinate calculation unit 248 and a cardiopulmonary function state classification unit 249.

The following will be described. The multidimensional vector configuration unit 41, the multidimensional movement vector configuration unit 42, the multidimensional movement vector length calculation unit 43, the multidimensional movement vector angle calculation unit 44, and the bending degree calculation unit 45 have the same configurations as described in the first embodiment. Since the same is true, the description thereof will be omitted. Further, since the tortuosity local maximum point extraction unit 146 and the cardiopulmonary function state change point estimation unit 147 have the same configurations as those described in the second embodiment, the description thereof will be omitted.

(Two-dimensional coordinate calculation unit)
The two-dimensional coordinate calculation unit 248 calculates the coordinates on the two-dimensional plane corresponding to the plurality of multidimensional vectors configured by the multidimensional vector configuration unit 41 by using a mathematical method.
As a mathematical method used in the two-dimensional coordinate calculation unit 248, for example, the document "JW Sammon," A nonlinear mapping for data structure analysis ", IEEE Trans. Computers, vol.C-18, no.5, pp.401" -409, May 1969. ”, the Sammon method, the projection tracking method, the self-organizing map method, and other multi-dimensional scale construction methods (MDS: Multi-Dimensional Scaling), and the two-dimensional system using the principal components of the principal component analysis method. Examples include algorithms such as visualization.

(Cardiopulmonary function status classification department)
The cardiopulmonary function state classification unit 249 classifies or estimates the cardiopulmonary function state of the measurement target person based on the two-dimensional plane coordinate data input from the two-dimensional coordinate calculation unit 248 and the cardiopulmonary function state-specific model.
The cardiopulmonary function state classification unit 249 has a cardiopulmonary function state-specific model (not shown) showing the cardiopulmonary function state of the measurement subject, which is generated in advance based on a plurality of two-dimensional plane coordinate data. Here, the cardiopulmonary functional state of the person to be measured means, for example, the degree of heart disease. The cardiopulmonary function state-specific model is, for example, a model for classifying the degree of heart disease. The cardiopulmonary function state-specific model is, for example, a trained model generated by machine learning the trajectories of a plurality of multidimensional vectors, which will be described later, as teacher data. The cardiopulmonary function state-specific model can be generated by collecting a large amount of data showing the locus of a multidimensional vector, storing it in a storage unit 30 or the like, and using the stored data as teacher data.

Further, the cardiopulmonary function state-specific model may be, for example, a state transition model indicating the degree of heart disease. When multiple state transition models are generated according to the degree of heart disease, the coordinate data on the two-dimensional plane based on the biological marker data of the measurement subject transitions from one state transition model to another. Accordingly, it is also possible to estimate the change point of the cardiopulmonary function state.

<Display unit>
The output unit 50 uses a point corresponding to a plurality of multidimensional vectors as a cardiopulmonary function state point representing a plurality of cardiopulmonary function states based on the coordinate data on the two-dimensional plane input from the two-dimensional coordinate calculation unit 248. Display above. The output unit 50 connects a plurality of cardiopulmonary function state points with a line segment and draws a locus of a plurality of multidimensional vectors.
Further, the output unit 50 may display the cardiopulmonary function state of the measurement target person classified by the cardiopulmonary function state classification unit 249.

(3.2) Cardiopulmonary function state estimation method and cardiopulmonary function state estimation program The cardiopulmonary function state change estimation device 2 according to the present embodiment includes at least one processor, and the processor includes an exercise load control unit 10 and data acquisition. It functions as an electrocardiographic processing unit 22, a heart sound processing unit 24, a storage unit 30, and a cardiopulmonary function state change estimation unit 240 of the unit 20.

The cardiopulmonary function state change estimation device 2 according to the present embodiment acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement target from the measurement target who is exercising at each of the plurality of time points. Based on the multidimensional vector at multiple time points composed of biological marker data, a multidimensional movement vector between two points at multiple time points is constructed respectively, and the degree of bending is formed from two consecutive multidimensional movement vectors. Is calculated, the local maximum point of flexion is extracted from the region where the cardiopulmonary function change point exists, and the exercise load corresponding to the local maximum point is estimated as the change point of the cardiopulmonary function state of the measurement subject. Will be executed. Further, in the cardiopulmonary function state change estimation device 2, the coordinates on the two-dimensional plane corresponding to a plurality of multidimensional vectors are calculated by using a mathematical method, and the coordinates data on the two-dimensional plane and the cardiopulmonary function state generated in advance are calculated. Based on another model, a cardiopulmonary function state estimation method for classifying the cardiopulmonary function state of the measurement subject and estimating the change in the cardiopulmonary function state is executed.

In addition, the cardiopulmonary function state change estimation device 1 acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement target from the measurement target who is exercising at each of the plurality of time points, and is biological. Based on the multidimensional vector at a plurality of time points composed of marker data, a multidimensional movement vector between two points at a plurality of time points is constructed, and the degree of bending is calculated from two consecutive multidimensional movement vectors. , The local maximum point of the degree of flexion is extracted from the region where the cardiopulmonary function change point exists, the exercise load corresponding to the local maximum point is estimated as the change point of the cardiopulmonary function state of the measurement subject, and it corresponds to multiple multidimensional vectors. The coordinates on the two-dimensional plane are calculated using a mathematical method, and the cardiopulmonary functional states of the measurement subjects are classified and cardiopulmonary based on the coordinate data on the two-dimensional plane and the cardiopulmonary function state-specific model generated in advance. Causes the processor to execute an instruction to estimate a change in functional state.

<Effect of the third embodiment>
The cardiopulmonary function state change estimation device 3 according to the third embodiment has the following effects in addition to the effects described in the first embodiment and the second embodiment.
(5) In the cardiopulmonary function state change estimation device 3, the local maximum point of the tortuosity of two consecutive multidimensional movement vectors is extracted from a predetermined cardiopulmonary function change point existence region and used as the local maximum point. The exercise load corresponding to the point is estimated as the change point of the cardiopulmonary function state of the measurement subject. Therefore, the cardiopulmonary function state change estimation device 2 excludes the local maximum point of tortuosity that occurs in a region other than the change point of the cardiopulmonary function state, which hinders the detection of the change point of the cardiopulmonary function state, and the cardiopulmonary function state. The change point of can be detected with high accuracy.

4. Hardware Configuration of Cardiopulmonary Function State Change Estimator The following, more specific for realizing the cardiopulmonary function state change estimation device 1 according to the first embodiment of the present disclosure and the cardiopulmonary function state change estimation device 2 according to the second embodiment. An example of a typical hardware configuration will be described.

FIG. 10 is an example of the hardware configuration of the cardiopulmonary function state change estimation system 1000. The cardiopulmonary function state change estimation devices 1 and 2 of the cardiopulmonary function state change estimation system 1000 include an electrocardiographic electrode 321, an acceleration sensor 323, a signal processing terminal 324, an exercise load control device 325, and a computer 326. Further, the exercise load control device 325 of the cardiopulmonary function state change estimation devices 1 and 2 is connected to the ergometer 327 which is a load device.

In FIG. 10, the electrocardiographic electrode 321 corresponds to the electrocardiographic sampling unit 21 of FIG. 1, the acceleration sensor 323 corresponds to the heart sound collecting unit 23 of FIG. 1, and the signal processing terminal 324 corresponds to the electrocardiographic processing unit 22 of FIG. And the heart sound processing unit 24, and the exercise load control device 325 corresponds to the exercise load control unit 10 of FIG. Further, in FIG. 10, the computer 326 corresponds to the storage unit 30, the cardiopulmonary function state change estimation unit 40 or 140, and the output unit 50 of FIG.

The electrocardiographic electrode 321 is attached to the chest of the measurement target 330, and extracts the electrical state of the heart of the measurement target 330 at each time as an electric signal (electrode signal).
The acceleration sensor 323 is attached to the chest of the measurement target 330, and extracts the heart sounds of the measurement target 330's heart at each time as an electric signal (heart sound signal).
As shown in FIG. 11, the signal processing terminal 324 includes a differential amplifier 324A for processing an electrode signal, a filter circuit 324B and an A / D converter 324C, and a filter circuit 324D and an amplifier 324E for processing a heartbeat signal. And an A / D converter 324F. The signal processing terminal 324 transmits the digitized electrocardiographic data and cardiac sound data to the computer 326.

The exercise load control device 325 controls the ergometer 327 so as to give a predetermined exercise load (exercise intensity) to the measurement target person 330, and transmits the exercise intensity data at each time to the computer 326.
The computer 326 estimates the change point of the cardiopulmonary function state of the measurement subject 330 based on the electrocardiographic data and the cardiac sound data output by the signal processing terminal 324 and the exercise intensity data output by the exercise load control device 325.
The computer 326 is a measurement target based on at least one of the heartbeat data and the electrocardiographic data output by the signal processing terminal 324, the exercise intensity data output by the exercise load control device 325, and the selection conditions set by the computer 326. Estimate the change point of the cardiopulmonary function state of the person. The computer 326 displays the change point of the cardiopulmonary function state of the measurement subject 330 on the display 326A, or outputs the change point of the cardiopulmonary function state of the measurement subject 330 to the outside.

In FIG. 10, the communication between the signal processing terminal 324 and the computer 326 and the communication between the exercise load control device 325 and the computer 326 are wired communication, but wireless communication may also be used. Further, as shown in FIG. 12, at least one of the digitized electrocardiographic data and the cardiac sound data and the exercise intensity data are input to the computer 326 using a recording medium 328 or 329 such as an SD card. You may do so. As shown in FIG. 13, when at least one of the electrocardiographic data and the cardiac sound data is input to the computer 326 using the recording medium 328, the signal processing terminal 324 includes a storage unit 324G.

5. Examples Hereinafter, examples of the cardiopulmonary function state estimation method in the cardiopulmonary function state change estimation device according to the present disclosure will be described.
In this example, 26 patients with heart failure are measured, and each of the 26 measurement subjects is subjected to a load test in which a ramp load is applied, and anaerobic metabolism is performed by the following methods (a) to (c). The threshold (AT) was estimated.

<Estimation of anaerobic metabolism threshold (AT)>
(A) Estimating the anaerobic metabolic threshold (AT) by the CPX test Changes in cardiopulmonary function by a doctor based on the oxygen uptake and carbon dioxide emissions of the measurement subject obtained by the exhaled gas analysis by the conventional CPX test. A point was estimated and the change point was defined as the anaerobic metabolism threshold (AT). The anaerobic metabolism threshold obtained by this method was defined as the anaerobic metabolism threshold AT1.

(B) Estimating the anaerobic metabolic threshold (AT) based on one type of biological marker data Exercise intensity of load test using the same device as the cardiopulmonary function state change estimation device according to the second embodiment of the present disclosure. The data and the S1 intensity obtained from the heartbeat data of the measurement subject obtained by the load test were acquired. Subsequently, the maximum change point of the gradient of the S1 intensity of the heart sound with respect to the exercise intensity was detected, and the exercise intensity of the load test corresponding to the maximum change point of the gradient was set as the anaerobic metabolism threshold (AT). The anaerobic metabolism threshold obtained by this method was defined as the anaerobic metabolism threshold AT2.

(C) Estimating the anaerobic metabolic threshold (AT) by the cardiopulmonary function state estimation method of the present disclosure Using the cardiopulmonary function state change estimation device according to the second embodiment of the present disclosure, the exercise intensity data of the load test and the load test A plurality of biological marker data of the measurement subject obtained by the above were obtained. Here, as a plurality of biological marker data, heart rate (HR), electromechanical activity time (EMAT), left ventricular contraction time (LVST), left ventricular diastolic perfusion time (LDPT), QRS width (QRSD). , S1 Intensity, S2 Intensity, S3 Intensity and S4 Intensity Natural Logistics (log (S1 Intensity), log (S2 Intensity), log (S3 Intensity), log (S4 Intensity)) Obtained.

After that, a cardiopulmonary function state change estimator is used to construct a multidimensional vector at a plurality of time points composed of nine biological marker data, and based on the multidimensional vector, a multipoint between two points at the plurality of time points. A dimensional movement vector was constructed. Finally, the local maximum point of tortuosity between two consecutive multidimensional movement vectors is extracted from the region where the cardiopulmonary function change point exists, and the exercise load corresponding to the local maximum point is taken as the change point of the cardiopulmonary function state of the measurement subject. Estimated, the exercise intensity of the load test corresponding to the change point was defined as the anaerobic metabolism threshold (AT).
In the load test (c), 60 ± 20 seconds or later of the anaerobic metabolism threshold (AT) existence time estimated by the doctor in the load test (a) is defined as the anaerobic metabolism threshold existence region, and 60 ± 20 seconds or later. The first tortuosity peak was defined as the anaerobic metabolic threshold (AT). The anaerobic metabolism threshold obtained by this method was designated as the anaerobic metabolism threshold AT3.

<Evaluation of cardiopulmonary function state change point detection accuracy>
For each of the 26 measurement subjects, the accuracy of detecting the change point of the cardiopulmonary function state when the method (c) for estimating the anaerobic metabolism threshold (AT) by the cardiopulmonary function state change estimation method of the present disclosure was used was evaluated. The evaluation compares the absolute value of the difference between the anaerobic metabolism threshold AT1 and the anaerobic metabolism threshold AT2 and the absolute value of the difference between the anaerobic metabolism threshold AT1 and the anaerobic metabolism threshold AT3 for each of the 26 measurement subjects. However, it was determined that the accuracy of the estimation method (c) of the present disclosure is improved when the absolute value of the difference between the anaerobic metabolism threshold AT1 and the anaerobic metabolism threshold AT3 is small.

<Evaluation result>
Table 1 below shows the measurement results of the anaerobic metabolism threshold obtained by each measurement. Table 1 shows the difference between the anaerobic metabolism threshold AT1 and the anaerobic metabolism threshold AT2 or the anaerobic metabolism threshold AT3, and the absolute value of the difference. Further, Table 1 shows the accuracy improvement rate when the estimation method (c) of the present disclosure is used. The improvement rate was such that the absolute value of the difference between the anaerobic metabolism threshold AT1 and the anaerobic metabolism threshold AT3 was smaller than the absolute value of the difference between the anaerobic metabolism threshold AT1 and the anaerobic metabolism threshold AT2 among the 26 measurement subjects. It is the ratio of the number of people.

The graphs of FIGS. 14 (A), 15 (A) and 16 (A) show the patient numbers shown in Table 1. 1, No. 2 and No. The tortuosity obtained by the estimation method (c) of each of the three measurement subjects is shown. The graphs of FIGS. 14 (B), 15 (B) and 16 (B) show the heart rate (HR) and cardiac interval length parameters with respect to the time from the start of the stress test. The graphs of FIGS. 14 (C), 15 (C) and 16 (C) show the heartbeat intensity parameters with respect to the time from the start of the load test. These heart rate, heart section length parameters, and heart sound intensity parameters are biological markers used to calculate the tortuosity shown in FIGS. 14 (A), 15 (A), and 16 (A).
In addition, in FIG. 14A to FIG. 14C, FIG. 15A to FIG. 15C, and FIGS. 16A to 16C, each value with respect to the time from the start of the load test is shown. Has been done.

As shown in Table 1, the absolute value AB3 of the difference between the anaerobic metabolism threshold AT3 and the anaerobic metabolism threshold AT1 is compared with the absolute value AB2 of the difference between the anaerobic metabolism threshold AT2 and the anaerobic metabolism threshold AT1. Overall, the absolute value AB2 tends to be smaller. For example, as shown in FIGS. 14 to 16, the anaerobic metabolism threshold AT3 has a smaller dissociation from the anaerobic metabolism threshold AT1 than the anaerobic metabolism threshold AT2.

As shown in FIG. 14 (A), a local peak of tortuosity (indicated by ⊚ in FIG. 14 (A)) occurs at a position where the test time is 500 seconds. The anaerobic metabolism threshold AT3 estimated based on this local peak is consistent with the anaerobic metabolism threshold AT1 estimated based on the CPX test.
On the other hand, in FIGS. 14 (B) and 14 (C), none of the local peaks of the heart rate, the heart section length parameter, and the heart sound intensity parameter appeared at the position of the test time of 500 seconds. Therefore, the anaerobic metabolism threshold AT2 estimated based on the estimation method (b) deviates from the anaerobic metabolism threshold AT1.
Similar trends can be seen in FIGS. 15 (A) to 15 (C) and FIGS. 16 (A) to 16 (C). That is, in the test time in which the local peaks of tortuosity (indicated by ⊚ in FIGS. 15A and 16A) appear, all the local peaks of the heart rate, the heart section length parameter, and the heart sound intensity parameter appear. Absent.

Of the 26 measurement subjects, 20 have an absolute value AB3 smaller than the absolute value AB2. That is, the accuracy improvement rate when the estimation method (c) of the present disclosure is used is 20/26 × 100 = 76.9%, and the estimation method (c) of the present disclosure is more accurate than the conventional method. Was found to improve.

Further, as shown in Table 1, the maximum value 13.3 of the absolute value AB3 is smaller than the maximum value 20.0 of the absolute value AB2, and the average value 4.2 of the absolute value AB3 is from the average value 8.8 of the absolute value AB2. small. Further, the variance 8.5 of the absolute value AB3 is sufficiently smaller than the variance 40.1 of the absolute value AB2, and the standard deviation 2.9 of the absolute value AB3 is smaller than the standard deviation 6.3 of the absolute value AB2.
As described above, when the estimation method (c) of the present disclosure is used, the deviation from the anaerobic metabolism threshold AT1 estimated by the CPX test is smaller than that when the estimation method (b) is used, and the cardiopulmonary function state change point is small. It was found that the detection accuracy was improved overall.

Although the specific configuration of the present disclosure has been described above by each embodiment, the scope of the present disclosure is not limited to the illustrated and described exemplary embodiments, and the present disclosure is intended. It also includes all embodiments that provide equal effect. Furthermore, the scope of the present disclosure is not limited to the combination of technical features defined by the claims, but may be defined by any desired combination of specific features among all disclosed features.

For example, the scope of the present disclosure is not limited to heart failure, but the primary diseases of heart failure are hypertension, ischemic heart disease (myocardial infarction), arrhythmia (atrial fibrillation, etc.), valvular disease, cardiomyopathy. Also includes application to heart diseases such as congenital heart disease.
Further, the cardiopulmonary function state change estimation device may be configured by combining the configurations described in each embodiment.

1, 2, 3 Cardiopulmonary function state change estimation device 10 Exercise load control unit 20 Data acquisition unit 21 Electrocardiographic collection unit 22 Electrocardiographic processing unit 23 Heart sound collection unit 24 Heart sound processing unit 30 Storage unit 40, 140, 240 Cardiopulmonary function state change Estimating unit 41 Multidimensional vector configuration unit 42 Multidimensional movement vector configuration unit 43 Multidimensional movement vector length calculation unit 44 Multidimensional movement vector inter-angle angle calculation unit 45 Flexibility calculation unit 46,146 Flexibility local maximum point extraction unit 47,147 Cardiopulmonary function state change point estimation unit 50 Output unit 60 Load device 248 Two-dimensional coordinate calculation unit 249 Cardiopulmonary function state classification unit 321 Electrocardiographic electrode 323 Accelerometer 324 Signal processing terminal 325 Exercise load control device 326 Computer 326A Display 327 Ergometer 328, 329 Recording media 330 Measurement target

Claims (21)

A load device that gives a load to the person to be measured who is exercising,
A data acquisition unit that acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of a plurality of time points.
Based on the multidimensional vector at the plurality of time points composed of the biological marker data, a multidimensional movement vector between two points at the plurality of time points is constructed, and two consecutive multidimensional movements are formed. A cardiopulmonary function state change estimation unit that estimates the local maximum point of the vector flexion degree as a change point of the cardiopulmonary function state of the measurement subject,
Cardiopulmonary function state change estimator with
A cardiopulmonary function state change estimation system equipped with. The cardiopulmonary function state change estimation unit
From the biological marker data, a multidimensional vector component that constitutes a multidimensional vector at each of the plurality of time points,
A multidimensional movement vector generation unit that selects two consecutive time points from the plurality of time points and constitutes a multidimensional movement vector between the two multidimensional movement vectors from the two multidimensional vectors at the two time points.
An angle calculation unit between multidimensional movement vectors that calculates the angles of two consecutive multidimensional movement vectors,
A tortuosity calculation unit that calculates the tortuosity from the angles of two consecutive multidimensional movement vectors,
Of the above-mentioned tortuosity, one or more tortuosity local maximum points that are the local maximum points are extracted and used as the tortuosity local maximum point.
A cardiopulmonary function state change point estimation unit that estimates the local maximum point of tortuosity as a change point of the cardiopulmonary function state,
The cardiopulmonary function state change estimation system according to claim 1. A multidimensional movement vector length calculation unit for calculating the lengths of two consecutive multidimensional movement vectors is provided.
The cardiopulmonary function state change estimation system according to claim 2, wherein the tortuosity calculation unit calculates the tortuosity based on the length and angle of each of the two consecutive multidimensional movement vectors. The bending degree calculation unit according to claim 3, wherein the bending degree calculation unit calculates the bending degree from the difference between the product of the lengths of two continuous multidimensional movement vectors and the inner product of two continuous multidimensional movement vectors. Cardiopulmonary function state change estimation system. The data acquisition unit
An electrocardiographic sampling unit that has multiple electrodes and extracts the potential of the heart of the person to be measured as an electrode signal for each electrode.
An electrocardiographic processing unit that generates electrocardiographic data based on a potential difference between a plurality of electrode signals taken out by the electrocardiographic sampling unit and generates the biological marker data from the electrocardiographic data.
The cardiopulmonary function state change estimation system according to any one of claims 2 to 4. The electrocardiographic processing unit, as the biological marker data, is at least one of the heart rate (HR), QR width (QRI), QRS width (QRSD), or a value obtained by dividing these by the heart rate length from the electrocardiographic data. The cardiopulmonary function state change estimation system according to claim 5, which calculates data indicating the above. The data acquisition unit
A heart sound collection unit that extracts the heart sounds of the person to be measured as a heart sound signal,
The data acquisition unit includes a heart sound processing unit that generates heart sound data based on the heart sound signal extracted by the heart sound collection unit and generates the biological marker data from the heart sound data.
The cardiopulmonary function state change estimation system according to any one of claims 2 to 6. The heart sound processing unit uses the heart sound data as the biological marker data to obtain S1 intensity (S1 Intensity), S2 intensity (S2 Intensity), S3 intensity (S3 Intensity), S4 intensity (S4 Intensity), S1 intensity, and S2. Intensity, natural logarithmic values of S3 and S4 intensities, as well as S1 intensities (S1 Strength), S2 intensities (S2 Strength), S3 intensities (S3 Strength) and S4 intensities (S4 Strength), electromechanical activity time (EMAT). The cardiopulmonary function state change estimation system according to claim 7, which calculates data indicating at least one of left ventricular contraction time (LVST), left ventricular diastolic perfusion time (LDPT), left ventricular dilatation time (LVDT), and QoS2. The cardiopulmonary function state change estimation system according to any one of claims 2 to 8, further comprising an exercise load control unit that controls the load device so as to give a gradual increase load in the cardiopulmonary exercise load test to the measurement target person. The cardiopulmonary function state change estimation system according to any one of claims 2 to 8, further comprising an exercise load control unit that controls the load device so as to give a constant load in exercise therapy for cardiac rehabilitation to the measurement subject. A storage unit for storing cardiopulmonary function change point existence area data indicating a cardiopulmonary function change point existence area, which is an area of exercise load presumed to include the change point of the cardiopulmonary function state, is provided.
The claim that the cardiopulmonary function state change point estimation unit estimates the change point of the cardiopulmonary function state existing in the cardiopulmonary function change point existence region based on the cardiopulmonary function change point existence region data acquired from the storage unit. The cardiopulmonary function state change estimation system according to any one of 2 to 10. Based on the biological marker data, the cardiopulmonary function state change point existence region, which is a region of exercise load presumed to include the cardiopulmonary function state change point, is determined, and the cardiopulmonary function state change point existence region is shown. It is provided with a cardiopulmonary function state change point existence area estimation unit that outputs cardiopulmonary function state change point existence area data to the cardiopulmonary function state change point estimation unit.
The cardiopulmonary function state change point estimation unit is based on the cardiopulmonary function state change point existence area data acquired from the cardiopulmonary function state change point existence area estimation unit, and the cardiopulmonary function present in the cardiopulmonary function state change point existence area. The cardiopulmonary function state change estimation system according to any one of claims 2 to 10 for estimating a state change point. The change point of the cardiopulmonary function state is the anaerobic metabolism threshold.
The cardiopulmonary function state change point estimation unit is located in the anaerobic metabolism threshold existence region where the anaerobic metabolism threshold is presumed to exist based on the acquired anaerobic metabolism threshold data indicating the anaerobic threshold. The cardiopulmonary function state change estimation system according to claim 11 or 12, wherein the change point of the functional state is estimated as the anaerobic threshold value. The cardiopulmonary function state change estimation system according to claim 13, wherein the cardiopulmonary function state change estimation unit determines the exercise load corresponding to the anaerobic metabolism threshold as the optimum exercise intensity of the person to be measured. The change point of the cardiopulmonary function state is the respiratory compensation starting point.
The cardiopulmonary function state change point estimation unit is set in the respiratory compensation start point region where the respiratory compensation start point is presumed to exist, based on the acquired respiratory compensation start point region data indicating the respiratory compensation start point. The cardiopulmonary function state change estimation system according to claim 11 or 12, wherein the existing change point of the cardiopulmonary function state is estimated as the respiratory compensation start point. Any of claims 2 to 15, wherein the multidimensional vector component normalizes each of the biological marker data and constructs the multidimensional vector based on the normalized biological marker data. The cardiopulmonary function state change estimation system described in item 1. Equipped with a display
The cardiopulmonary function state change estimation unit
It is provided with a two-dimensional coordinate calculation unit that calculates coordinate data on the two-dimensional plane indicating coordinates on the two-dimensional plane corresponding to the plurality of multidimensional vectors composed of the multi-dimensional vector configuration unit by using a mathematical method.
Based on the coordinate data on the two-dimensional plane, the display unit displays points corresponding to the plurality of multidimensional vectors on the two-dimensional plane as cardiopulmonary function state change points representing a plurality of cardiopulmonary function state changes. The cardiopulmonary function state change estimation system according to any one of claims 2 to 16, wherein a plurality of cardiopulmonary function state change points are connected by a line segment to draw a locus of the plurality of multidimensional vectors. The cardiopulmonary function state change estimation unit
A cardiopulmonary function state-specific model showing changes in the cardiopulmonary function state of the measurement subject, which was generated in advance based on the plurality of coordinate data on the two-dimensional plane,
A cardiopulmonary function state classification unit that classifies the cardiopulmonary function state of the measurement target person from the two-dimensional plane coordinate data input from the two-dimensional coordinate calculation unit and the cardiopulmonary function state-specific model.
The cardiopulmonary function state change estimation system according to any one of claims 17. A data acquisition unit that acquires a plurality of biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of the plurality of time points.
Based on the multidimensional vector at the plurality of time points composed of the biological marker data, a multidimensional movement vector between two points at the plurality of time points is constructed, and two consecutive multidimensional movements are formed. A cardiopulmonary function state change estimation unit that estimates the local maximum point of the vector flexion degree as a change point of the cardiopulmonary function state of the measurement subject,
A cardiopulmonary function state change estimator equipped with. At each of the plurality of time points, a plurality of biological marker data related to the cardiopulmonary function of the measurement subject were obtained from the measurement subject who was exercising.
Based on the multidimensional vector at the plurality of time points composed of the biological marker data, a multidimensional movement vector between two points at the plurality of time points is constructed.
The local maximum point of the tortuosity of the two consecutive multidimensional movement vectors is estimated as the change point of the cardiopulmonary function state of the measurement subject.
Cardiopulmonary function state change estimation method. Obtaining multiple biological marker data related to the cardiopulmonary function of the measurement subject from the measurement subject who is exercising at each of the plurality of time points, and
Based on the multidimensional vector at the plurality of time points composed of the biological marker data, the multidimensional movement vector between two points at the plurality of time points is constructed, and
Estimating the local maximum point of the tortuosity of the two consecutive multidimensional movement vectors as the change point of the cardiopulmonary function state of the measurement subject,
A cardiopulmonary function state change estimation program that causes a computer to execute.

Images