WO2007046504A1

WO2007046504A1 - Stethoscope heart sound signal processing method and stethoscope device

Info

Publication number: WO2007046504A1
Application number: PCT/JP2006/320964
Authority: WO
Inventors: Zhongwei Jiang; Samjin Choi
Original assignee: Yamaguchi University
Priority date: 2005-10-21
Filing date: 2006-10-20
Publication date: 2007-04-26
Also published as: JP4848524B2; JPWO2007046504A1; CN101291628B; CN101291628A

Abstract

[PROBLEMS] The detected heart sound signal is adjusted to an adequate volume and quantitavely analyzed with high accuracy, and a processing to improve the sound quality of the heart sound signal to make the heart sound signal easy to listen to. [MEANS FOR SOLVING PROBLEMS] In analyzing a heart sound, an evaluation index of a heart sound feature value waveform generated by using heart sound data and a vibration model is determined, the center of a data set is so determined that the evaluation function expressing the dispersion of the evaluation index takes on a minimum value, and the evaluation index for a threshold and the distribution of the center of the data set are determined. To improve the quality of the heart sound, after substantially eliminating the phase difference and multiply the heart sound signal and the heart sound feature value waveform, the product is used as output heart sound data. In adjusting the volume of the heart sound signal, the up gain is determined using a reference table according to the average intensity for a time Ta of the detected heart sound signal, and the heart sound signal for a time Tb after the time Ta is amplified and recorded.

Description

Specification

Auscultation heart sound signal processing method and auscultation apparatus

Technical field

The present invention relates to an auscultation heart sound signal processing method and an auscultation apparatus.

Background art

[0002] Mortality from heart disease is high in Japan. In 1985, it surpassed stroke and was ranked second. Among heart diseases, myocardial infarction, which is said to be heart failure or ischemic heart disease, is common, and acute death of unknown cause accounts for 30%. Life-style related diseases such as cardiovascular disorders are said to have a gradual change in disease state, and it is said that there are many cases in which they are not diagnosed accurately and cannot be discovered without quantitative follow-up over a long period of time.

[0003] In recent years, the development of a system for performing health management 'diagnosis at home and in the company has been promoted, and it is widely used as a stethoscope to auscultate heart sounds and breathing sounds together with a weight scale, thermometer, blood pressure monitor, etc. The power of the stethoscope It cannot be said that the stethoscope is used so much. The reason is that diagnosis of auscultation sounds requires skill and is difficult. The part to which the stethoscope is applied greatly affects auscultation, and the skilled doctor determines whether the heart sound is normal or abnormal at the part where the heart sound is best heard while changing the part to which the stethoscope is applied. As described above, the auscultation technique requiring skill is low in skill and difficult for general users to learn.

[0004] A simple stethoscope listens to detected heart sounds with earphones to determine whether the heart sounds are normal or abnormal. As a stethoscope for more precise examination of heart sounds, a probe is Sound is detected, and the detected heart sound signal is recorded as heart sound data. The detected heart sounds or further recorded heart sounds are listened to by the earphones, and the recorded heart sound data is analyzed and used to determine whether the heart sounds are normal or abnormal. The recorded heart sound signal is also transmitted over a line and used in such a form that a specialist at a remote location inspects the heart sound.

[0005] In order to listen to the recorded heart sounds and to accurately analyze the heart sound signal, it is important to set the recording sound volume level appropriately. From the aspect of playback sound quality, it is advantageous to increase the recording volume level. However, it is often difficult for a specialist to auscultate because the audio signal contains noise. Even if general noise removal or reduction measures are taken, the sound that is usually used by doctors may be different or useful information may be deleted.

[0006] In addition to listening to the recorded heart sound signal, a computer-assisted method is used in which heart sound analysis is performed on heart sound data to assist in diagnosis of heart disease. As a heart disease diagnosis system using such heart sound analysis, a dedicated system for professionals is large-scale and difficult to use for general users. In addition, the simpler and smaller ones for general users have made it difficult to accurately identify abnormal heart sounds.

A stethoscope or auscultation device is disclosed in the following patent documents.

[0007] Patent Document 1: Japanese Patent Laid-Open No. 2005-52521

Patent Document 2: Japanese Patent Publication No. 10-504748

Patent Document 3: JP-A 61-290936

Patent Document 4: JP-A-5-309075

[0008] Patent Document 1 converts sound acquired by a microphone into an electrical signal, and selectively enhances a signal in a frequency range corresponding to a heart sound and a signal in a frequency range corresponding to a breathing sound. However, it is described on electronic stethoscopes that set the frequency characteristics of the equalizer to selectively attenuate signals in other frequency ranges.

In this electronic stethoscope, since the characteristics of heart sound data vary to some extent, what should be enhanced as heart sounds may be attenuated in some cases, and noise may not be effectively reduced. It was possible.

Patent Document 2 includes an electronic auscultation that includes a digital filter for pre-emphasis, hearing loss compensation, and the like, and has a pattern recognition means for suppressing repetitive signals in the observed signal and removing noise. It is written on the vessel.

However, since these electronic stethoscopes are equipped with elements such as a filter means for probing the impulse transfer function and a pattern recognition means for pre-emphasis, the apparatus becomes complicated, and individual differences in heart sounds occur. As a result, noise may not be effectively reduced by the filter means and pattern recognition means. [0010] In Patent Document 3, gain adjustment of a heart sound waveform signal detected by a heart sound detection probe is performed in a signal conversion device, and AD conversion is performed as an input signal of a general-purpose personal computer. The phonocardiograph that performs this is disclosed. The heart sound meter disclosed in Patent Document 3 is a heart sound meter for a general user such as a household user, and can use a heart sound waveform signal with a simple configuration using a general-purpose personal computer. By the way, the heart sound detected by the heart sound probe includes noise, and when listening to the heart sound transmitted depending on how much gain adjustment is performed in the heart sound conversion device, or transmitting and receiving heart sound data. However, in Patent Document 3, it is better to adjust the gain of the detected heart sound waveform signal in performing the heart sound analysis. However, it was not enough for accurate analysis of heart sounds.

[0011] In Patent Document 4, in order to accurately determine heart sound data without setting a determination criterion, the amplitude of the heart sound is stored together with the elapsed time, and the characteristics of a predetermined portion of the stored amplitude are depicted. Based on the above, there is disclosed a heart sound analysis apparatus that performs predetermined recognition by a Yural network, outputs the degree of recognition, and displays the degree of abnormality of heart sounds.

This heart sound analyzer uses a force neural network that sets the borderline in determining the necessity of precision screening required in the primary screening of heart screening. It is not easy to use o

[0012] Further, the present inventor has made a prior patent application (Japanese Patent Application No. 2005-80720) concerning the invention of a digital auscultation analysis system capable of heart sound diagnosis with a simple configuration used by general users. I applied. The heart sound analysis method in the preceding invention is based on the heart sound data force measured using the tympanic membrane vibration model to obtain the vibration response and analyze and evaluate the time widths of the peak I sound and stuttering sound. This is to detect abnormalities. In this heart sound analysis method, abnormalities of heart sounds can be confirmed by analyzing and evaluating the time span obtained by vibration response force, but the shape of heart sound data and vibration response is the type and characteristics of heart sound abnormalities. This analysis method may not be able to accurately grasp abnormal heart sounds for various characteristics of abnormal heart sounds.

Disclosure of the invention

Problems to be solved by the invention

[0013] As described above, in the conventional auscultation apparatus, the heart sound signal that can have various characteristics is efficiently removed, noise is effectively improved, the sound quality is improved, and the heart sound recording level is improved. It is inadequate to listen to and discriminate heart sounds recorded with proper sound quality, and to transmit heart sound signals to remote locations for use, making the device complicated and expensive. There was a difficulty of becoming a thing.

As a heart disease diagnosis system using analysis processing that is not effective for analyzing the heartbeat signals and analyzing abnormalities of various heart sounds accurately and quantitatively for the normality / abnormality of heart sounds. The scale was large and it was difficult for general users to use.

[0014] Therefore, by using inexpensive methods and devices, the sound quality of the heart sound auscultation sound is improved, the sound is low and the heart sound auscultation sound is easy to hear, the recording level of the heart sound signal is set appropriately, and the normal / abnormal heart sound is also detected. It was desired to analyze the heart sound signal so that it could be accurately and quantitatively determined. It was also desired to adjust the detected heart sound signal to an appropriate volume for listening to the heart sound, processing and analyzing the heart sound signal.

Means for solving the problem

[0015] The present invention has been made to solve the above-mentioned problems, and the auscultatory heart sound signal processing method for performing heart sound analysis for detecting abnormal heart sounds according to the present invention sets model parameters of a vibration model. Detecting heart sounds, thereby obtaining heart sound data, generating feature value waveform data for the obtained heart sound data under set model parameters, and threshold (THV) For the evaluation value indicating the time width and time interval of the peak of the feature value waveform data, and the evaluation index power is also defined by using the fuzzy member function (w), the center of the data set (V) And the evaluation function J (W, V) representing the distribution of the evaluation index from the center of the evaluation index and data set, and the center of the data set so that the evaluation function is minimized by iterative calculation. To decide And the predetermined Minimum evaluation function value Ci for range THV)

Dependency of m is calculated and J

Each step force is to select the THV that minimizes m, and to display the evaluation index obtained for the selected THV and the distribution state of the center of the data set.

[0016] The evaluation index is the time width (T1, T2) and time interval (T11, T12) of sound I and sound II in the characteristic value waveform data for THV, and W = {w}, V = {v} , D = || v—z

j j k, j

Given that II is the Euclidean distance between the center of the data set and the data position, the evaluation function is

[Equation 7]

'• -1 Zo-1 as shown.

[0017] The auscultation device according to the present invention includes means for setting model parameters of a vibration model, heart sound detecting means for detecting heart sounds and thereby obtaining heart sound data, and setting for the obtained heart sound data. Means for generating feature value waveform data under model parameters, means for obtaining an evaluation index indicating a time width and a time interval of a peak of the feature value waveform data with respect to a threshold value (THV), and the evaluation index power Using fuzzy member function (w),

Means for obtaining the center (V) of the data set to be determined, means for obtaining an evaluation function ¾J (W, V) representing the distribution of the evaluation index from the center of the evaluation index and the data set, and the evaluation function A means for determining the center of a data set by iterative calculation so as to minimize, a means for obtaining a minimum evaluation function town where the evaluation function is minimum, and a dependence of J on a predetermined range of THV are obtained, and within that range The means to select the THV that minimizes (CF), and the selection mm

And a means for displaying the evaluation index obtained for the obtained THV and the distribution state of the center of the data set, and a heart sound analysis processing unit for performing heart sound signal analysis for detecting abnormal heart sounds.

[0018] The evaluation index is the time width (T1, T2) and time interval (T11, T12) of the I sound and the II sound in the characteristic value waveform data for THV, and W = {w}, V = {v} , D = || v— z

j j k, j

Given that II is the Euclidean distance between the center of the data set and the data position, the evaluation function is , (= ∑∑,.,) '') ² (?)

[0019] Further, the auscultatory heart sound signal processing method for performing heart sound signal processing for sound quality improvement in auscultation according to the present invention includes setting a model parameter of a vibration model to form a vibration model and detecting a heart sound. To obtain the heartbeat signal, to obtain the feature value waveform data output by applying the obtained heart sound signal to the vibration model, and to obtain the heart sound signal or the signal obtained by removing high-frequency component noise as heart sound data. The phase lag is calculated by taking the cross-correlation between the heart sound data and the feature value waveform data, and the phase lag is divided so that there is substantially no phase difference between the heart sound data and the feature value waveform data. The output heart sound data is obtained as a product of shifting the phase of the feature value waveform data only by the product of the heart sound data and the feature value waveform data shifted in phase by the phase delay. , But also made steps force.

[0020] The heart sound signal may be normalized before the heart sound signal is applied to the vibration model.

The auscultation apparatus according to the present invention inputs a heart sound signal detected by the heart sound detecting means or a signal obtained by removing noise of a high frequency component from the heart sound data as a heart sound data, thereby inputting characteristic value waveform data corresponding to the heart sound data. A vibration model to be output; a phase lag calculation unit that calculates a phase lag of the feature value waveform data with respect to the heart sound data by cross-correlating the feature value waveform data output by the vibration model and the heart sound data; A multiplication conversion unit that takes a product of the feature value waveform data and the heart sound data, the phase of which is shifted by the phase delay so that there is substantially no phase difference between the heart sound data and the feature value waveform data; It also has a heart sound signal processing unit that performs heart sound signal processing to detect abnormal heart sounds.

You may make it further have a normalization means for normalizing a heart sound signal before inputting into the said vibration model.

[0021] Further, the auscultation apparatus according to the present invention includes a probe for converting a heart sound into an electric signal, and the probe. An automatic volume adjustment recording unit that records and outputs a heart sound by adjusting and amplifying the signal of the heart sound obtained by the audio signal, and the time of the signal of the heart sound obtained by the probe by the automatic volume adjustment recording unit τ Suitable for amplifying heart sound signals based on average intensity between

a

A microcomputer control unit that obtains an up gain so as to obtain a proper volume and controls amplification of a heart sound signal, and a time T following the time T according to the obtained up gain under the control of the microcomputer control unit. Amplifies the heart sound signal and records the amplified signal ab

It is also possible to provide a means for automatically adjusting the sound recording volume comprising an amplification adjusting unit that can be extracted as a power sound.

[0022] The microcomputer control unit holds a relationship of the magnitude of the up gain for amplifying the sound intensity to an appropriate sound volume with respect to the calculated average intensity of the heart sound signal, and refers to the table. Try to amplify the heart sound signal to an appropriate volume.

The time T force ^ is a time within a range of 3 seconds, and the time T force is within a range of 12 seconds.

a b

The time T may be in the range of 8 to: LO seconds.

b

It may be.

The heart sound recording sound volume automatic adjusting means may be configured as a transmission side unit so that the amplified heart sound signal can be transmitted to the reception side unit.

The invention's effect

[0023] According to the present invention, the heart sound feature value waveform and the heart sound signal obtained using the heart sound data and the vibration model are substantially eliminated from the phase difference, and the impulse is obtained as output heart sound data. Therefore, the heart sound of the peak part is emphasized and the noise part is weakened with respect to the original heart sound data, and the sound quality is easy to hear when played back, and it is efficient by inexpensive methods and devices. The sound quality of the heart sound auscultation sound can be improved accurately, and it can contribute to the accurate determination of normal / abnormal heart sounds.

[0024] Further, according to the present invention, the evaluation index and the center of the data set are defined for the feature value waveform generated from the heart sound data using the vibration model so that the evaluation function expressed thereby is minimized. The center of the data set is determined in order to obtain the evaluation index and the processing result for the center of the data set, which is easy for general users to use and takes various forms. Quantify the normality / abnormality of heart sounds with respect to the characteristics of abnormal heart sounds Can be distinguished

[0025] Further, regarding the adjustment of the volume of the heart sound, the time T (1 to 3 seconds) of the detected heart sound signal

a

The up-gain is calculated using a reference table based on the average intensity between

a Record the heart sound signal by amplifying the heart sound signal for the following time T (4-12 seconds)

b

By doing so, it is possible to amplify the heart sound so as to always have an appropriate intensity, making it easier to listen to the heart sound on the monitor, and also to accurately analyze the heart sound.

BEST MODE FOR CARRYING OUT THE INVENTION

[0026] In the present invention, the recorded heart sound signal is used in the form of analyzing the heart sound signal so that normality / abnormality of the heart sound can be discriminated, and further, the recorded heart sound signal is listened to, or When a heart sound signal transmitted to a remote location is listened to, it is considered to handle the heart sound signal in a heart disease diagnosis system such as that used in a rectangular shape. At that time, the recorded heart sound signal is processed! Based on the eardrum vibration model, the heart sound signal is processed using the characteristic value waveform. Therefore, the features of the present invention are as follows: (A) feature value waveform based on vibration model, (B) heart sound signal analysis processing for abnormal heart sound detection, (C) heart sound signal processing for sound quality improvement in auscultation, (D The explanation is divided into the aspects of) automatic adjustment of the volume of heart sound recording, and (E) auscultation device configuration.

[0027] (A) Feature value waveform by vibration mode

The heart is divided into four parts: the left atrium, the left ventricle, the right atrium, and the right ventricle. As a whole, it plays the role of a pump that circulates blood throughout the body by repeating contraction and relaxation. There is a mitral valve at the entrance of the left atrium, an aortic valve at the entrance of the left ventricle, a tricuspid valve at the entrance of the right ventricle, and a pulmonary valve at the entrance of the right atrium. These valve membranes prevent blood backflow. Is the sound emitted when these valve membranes close.

[0028] Heart sound by auscultation is normal force. Accurately distinguishing abnormal force requires specialized knowledge and experience, but some people can recognize it relatively easily. This is thought to be because it is easy for the human ear to sense the low-order mode vibration that occurs when the eardrum is pulsated by the sound waves from the stethoscope. Based on this idea, it is considered effective to approximate the vibration of the eardrum and to process the heart sound signal as the relationship between the heart sound collected from the stethoscope and the vibration response of the vibration model. FIG. 1 is a conceptual diagram showing an eardrum vibration model. In FIG. 1, 1 is an object having an equivalent mass of m corresponding to the eardrum, 2 is an object having one end corresponding to the eardrum 1 and the other end being fixed to the fixed part, and 3 is one end. Is a damper in which the other end is attached to the fixed part to the object corresponding to the object 1 corresponding to the eardrum. The equivalent mass of object 1 corresponding to the eardrum is m, the spring constant of spring 2 is K, the viscous damping coefficient of damper 3 is C, the heart from the stethoscope

h h

When the sound is S, the vibration response X of the eardrum is calculated from the equation (1).

[Equation 1] mx + cx + L · = ύ (1) Divide both sides of equation (1) by m, and further calculate the input value of the heart sound signal.

If [Equation 2] = ± | 57m |, the natural frequency is p, and the damping ratio coefficient is ξ, Equation (1) is

[Equation 3]

X + 2ξρχ + p ² x ^ S (2) Therefore, when the natural frequency p and damping ratio coefficient ξ are set, the vibration response X of the eardrum can be obtained from Eq. (2), and this natural frequency ρ and damping ratio coefficient ξ are used as model parameters for the vibration model. It shows the characteristics. Instead of a mechanical vibration system with equivalent mass m, mechanical damper C, spring constant K force,

h h

It may be an electric vibration system having a scale and electric capacity.

[0030] (B) Analysis processing of heart sound signal for detection of abnormal heart sound

Fig. 2 (a) shows the results of the normal heart sound force vibration response X based on the tympanic membrane vibration model described in (A), and Fig. 2 (b) shows the mitral valve closure based on the tympanic membrane vibration model. Incomplete heart sound force Shows the results of vibration response. The heart sound data is recorded in a commercially available heart sound auscultation training material. The parameter p in Eq. (3) is 10 Hz, and is 0.707. In Fig. 2 (a) and (b), the gray waveform is the original waveform of heart sound S. The solid line waveform shows the vibration response, and the waveform showing this vibration response is called the feature value waveform. Yeah. The waveform in Fig. 2 (a) represents a normal heart sound that can be heard as "dock, dock", and Fig. 2 (b) represents a heart sound in which the mitral regurgitation sounded as "go, goo". The waveforms in Fig. 2 (a) and (b) are characteristic value waveforms with positive and negative amplitudes, but it is only necessary to analyze either the positive or negative waveform part.

[0031] Fig. 3 (a) shows a feature value waveform portion having a positive amplitude when the feature value waveform is analyzed. In general, it is known that peaks called I sound and stuttering appear repeatedly in a normal heart sound waveform. This I sound is caused by the closure of the mitral and tricuspid valves, and the II sound is caused by the closure and tension of the aortic and pulmonary valves. To determine whether the heart sound is normal or abnormal, the I sound and II It is considered effective to analyze and evaluate the sound duration (peak duration).

[0032] In Fig. 3 (a), the vertical axis indicates the intensity of the heart sound, the horizontal axis indicates the time, and the measurement time is shown for 2 seconds. Two I sound peaks 48 and two stuttering peaks 49 each. It has been extracted. In Fig. 3 (a), with 50% of the maximum intensity as a threshold, the duration (time width) of the I sound peak 48 and the stuttering peak 49 is obtained from the point where this threshold line intersects the feature value waveform, and each The evaluation index is Tl and Τ2. In addition, in order to take into account the appearance of continuous noise between the I and II sounds due to heart valve insufficiency, etc., the starting force of the I sound peak 48 followed by the end of the II sound peak 49 In order to consider that the interval between I sounds changes due to arrhythmia and heartbeat disturbance, the time for the starting force of II sound peak 48 is also set to the duration until the start of the next I sound peak 48. The evaluation index is T11. For the heart sound analysis, these evaluation indices Tl, T2, T12, Til are used in appropriate combination.

[0033] One set of evaluation indices Tl, T2, T12, Ti l is defined for each pair of the I sound peak 48 and the II sound peak 49, and the evaluation indices for a plurality of groups are plotted. Figure 3 (b) plots the points indicated by (Tl, T2) and (Ti l, T12) with Tl and Tl l on the horizontal axis and Τ2 and Τ12 on the vertical axis. Normal or abnormal heart sounds are determined. In other words, with normal heart sounds, the points represented by the evaluation index tend to gather in the categories surrounded by dotted lines as shown in Fig. 3 (b). From this, it can be determined that the heart sound is normal when a point represented by the evaluation index enters the region of the normal value range, and the heart sound is abnormal when the point does not fall within this region. However, since there are individual differences in heart sound data, the range of normal values is the default for many healthy individuals. It is desirable to obtain data and determine statistically.

[0034] In Fig. 3 (c), the horizontal axis is the type of evaluation index, and the vertical axis is the frequency of each evaluation index during the measurement time of 10 seconds. You can compare the frequency of the index. For example, if T2 is clearly less than evaluation index T1, there is a possibility of an arrhythmia that is not observed due to the loss of II sound peak 49, and this evaluation index is used when T11 is represented by multiple bars. It is visually determined that there is a variation in the value of.

[0035] In the invention of the prior patent application, such an evaluation index is used to judge abnormalities in heart sounds, and to change model parameters ρ and ξ as necessary. Fig. 2 (a) The waveforms shown in (b) have various shapes depending on the type of heart sound abnormality, and depending on the type and degree of abnormality, it may be difficult to accurately determine heart sound abnormality.

[0036] Considering this, in the invention of the prior patent application (Japanese Patent Application No. 2005-80720), the threshold (THV) for defining the evaluation indices Tl, T2, Tl l, and T12 is set to 50%. This THV was a force not considered otherwise. However, as a result of further investigation, when determining abnormalities in heart sounds, the results greatly depend on the threshold setting, and THV setting is an important factor for heart sound analysis. It was surprising that it was greatly affected by the recording conditions and individual differences among the measurement subjects. In principle, the threshold is a force that can be set in the range of 0 to: LOO%. From the actual situation, the range of 10 to 70% is reasonable.

[0037] Figure 4 (a) shows the relationship between the heart sound feature value waveform and T HV when THV is 15%, 30%, and 60%, respectively. Figure 4 (b) and (c) 5 is a distribution diagram of points indicated by evaluation indices (T 1, T2) and (Ti l, Tl 2) obtained for THV of the present invention. Fig. 4 (b) and (c) When THV = 15% (indicated by mouth) and THV = 60% (indicated by ○), distribution of (Tl, T2) or (Ti l, T12) Is concentrated when THV = 30% (indicated by ▲). As seen in Fig. 4 (a), the threshold line of THV = 30% intersects with the peaks of all heart sound feature waveform waveforms, whereas at THV = 60%, it shows some peak waveforms. There is a part that does not intersect, and when THV = 15%, it is considered that there is a partial force S that intersects the noise part below the feature value waveform.

[0038] As described above, when analyzing the characteristic value waveform of the same heart sound, it is evaluated by setting the THV. The degree of concentration of the distribution of points indicated by the price index is different. In this example, THV = 30% is better than THV = 15% and THV = 60%. However, in this situation, depending on the type of heart sound abnormality, the feature value waveform has various shapes. Therefore, it is recommended that a good THV for heart sound analysis be set according to these different situations.

[0039] The distribution of the evaluation index differs as shown in Fig. 4 (b) and (c) depending on the THV setting for the characteristic value waveform shown in Fig. 4 (a). In order to make a good judgment, it is better that the evaluation feature index is less scattered for the same feature value waveform. Therefore, in the present invention, the fuzzy C means (FCM) data clustering method is used as the data grouping method. FCM has been proposed in various ways, and is one of the data clustering methods.

[0040] For example, a set of data

Nif, ..., ...

v ^{1 2;} "1 (3)

Z _J = [∑,, ..., z _k , ..., z _c ]. Suppose we have groups of sc forces. In this case, the center position v. Of the i-th group

[Equation 5]

U = I, 2,..., C). Where w is

,

[Equation 6]

> w, _y = 1, j = 1, 2, ..., n (5)

A fuzzy member function between 0 and 1 that satisfies Me (l, ∞) is a waitin This is called an etasponent, and in general it is better to set m = 2. The Euclidean distance d between the i th cluster center position V and the j th data position z

k, j

d = II v -z

j II (6)

j k,

It is defined as Evaluation function for FCM clustering is

[Equation 7]

'• -1 Zo-1 is represented. Here, w = {w} and v = {v}. Evaluation function # α (w, v) is scattered data,

This represents the degree of scatter, and it can be said that the smaller the evaluation function ¾J (W, V), the less the dispersion. In order to minimize the data dispersion, the cluster center position {v} is determined so that the evaluation function is minimized by iterative calculation. Specifically, first, the initial value of the member function matrix {w} is set by fuzzy theory, and the cluster center position {v},

Calculate Obtain Euclidean distance d from Equation (6) and substitute it into Equation (7). The evaluation function is,

If it is not the minimum, W is recalculated as follows using the Euclidean distance d obtained in the previous calculation.

[Equation 8]

= -—— ¹ ——-(8)

∑, d, Jd ^

Since the FCM clustering method depends on the initial value of the member function w,

It is recommended to execute the above algorithm using a member function.

[0041] The force that applies this FCM clustering method to heart sound analysis. At that time, the data set [Tl, T2, Til, T12] obtained from the heart sound feature value waveform is used as the data in Equation (3). Apply FCM clustering algorithm as set [z, z, z, z].

1 2 3 4]

[0042] Divide [Tl, T2, Ti l, T12] into (Tl, T2) and (Ti l, T1)

Figure 5 (a) and (b) are shown. The center (V, V) of the distribution of the point represented by (Tl, T2) and the center (V, V) of the distribution of the point represented by (Ti l, T12) are obtained from Equation (4).

2 3 4

The centers of are denoted as A> and <B>, respectively. From the data obtained for normal heart sounds Calculated threshold value (THV), evaluation function ¾J (W, V) minimum table function city, inside the data set mm

The values of the heart [v, v, v, v] are shown in Table 1, and the minimum table function value J in Table 1 depends on THV.

1 2 3 4 m

It changes as shown in Fig. 5 (c). When THV increases by 10% in the range of 10% to 70%, the minimum evaluation function town becomes particularly small when THV is in the range of 30% to 60%, and this range is shown in Fig. 5 (c). It is in a valley bottom state. The distributions of (Tl, T2) and (Ti l, T12) obtained when the THV increases by 10% in the range of 10% to 70% are shown in Figs. 6 (a) and (b). Figures 6 (c) and 6 (d) show the data obtained in the range of 30% to 60%. It can be seen that when the THV is in the range of 10% to 70%, the data varies considerably, but in the range where the THV is in the range of 30% to 60%, there is little variation in the data. Thus, judging from Table 1 and Figures 6 (a) to (d), it can be said that THV (30% to 60%) in which J is smaller than 0.01 in Table 1 is an effective value.

[0043] The dependence of the data center V, V, V, V and the minimum evaluation function city on THV is

1 2 3 4 m

Normal heart sounds differ to some extent, and the effective threshold range of THV also varies slightly. Generally, in the case of normal heart sounds, the value of J becomes minimum, or the THV that reaches the valley state as shown in Fig. 7 There is a range, and in that range J can be said to be a very small value of about 0.01.

Next, let us consider the case where the FCM clustering method is applied to abnormal heart sound data. Figure 7 shows the heart rate data for atrial fibrillation and atrial flutter (AF, arrhythmia), mitral stenosis (MS), and aortic regurgitation (AR). Function value (J) dependency (a to c) and center of data set for THV [V, V, V m 1 2 3

, V] shows the dependence (d ~; f). In this figure, the minimum evaluation function city force

4 m

As shown in Fig. 7 (a), the effective threshold range is 16% to 46%, and for MS in Fig. 7 (b), the effective threshold range is 45% to 66%. For AR, the effective threshold range is 10% to 22%.

[0045] Based on the results in Fig. 7, points represented by evaluation indices (Tl, T2) and (Ti l, T12) obtained by setting THV to a value within the effective threshold range for AF, MS, and AR Is plotted as shown in Figure 8. In Figure 8, (a) and (b) are for AF, (c) and (d) are for MS, (e) and (f) are for (Tl, T2), (Ti The distribution of l, Tl 2) is shown. Fig 8 The distribution of the points represented by the evaluation index for AF, MS, and AR is clearly different from that for normal heart sounds as shown in Figs. 6 (c) and (d). From the distribution of the points represented by the index, normal / abnormal heart sounds can be accurately identified.

[0046] Thus, the minimum evaluation function city, the center [V, V, V, V] of the data set, the effective threshold range m 1 2 3 4

Based on the distribution of points represented by the evaluation index [Tl, T2, Til, T12] obtained by setting a threshold in the box, it is possible to discriminate between normal and abnormal heart sounds. In the case of normal heart sounds, J is lower than 0.02 in the effective threshold, and the distribution of the points represented by the center of the data set and the evaluation index is within a certain range. Among the values, at least one or more values are extremely high compared to normal heart sounds. For example, in the case of AF and MS, the center of the data set is about the same as in the case of normal heart sounds, and the minimum evaluation function city within the range of the force effective threshold is 0.4 compared to the case of normal heart sounds. In this case, the value of J within the effective threshold range is small, but the value indicating the center of the data set is much larger than that of normal heart sounds.

[0047] For both normal heart sounds and abnormal heart sounds, there is a range of effective thresholds where the minimum evaluation function city is minimized or in a valley bottom state. One THV value within the range should be selected appropriately.

The heart sound analysis using the FCM clustering method according to the present invention described above is performed in the form shown in the flow of FIG.

(1) Set the model parameters (, p).

(2) Heart sound is detected and heart sound data is obtained.

(3) Generate feature value waveform data under the set model parameters for the obtained heart sound data.

(4) Obtain evaluation indices Tl, T2, Ti l, and T12 for THV.

(5) Find the centers V, V, V, V of the data set defined by the member function from the evaluation indices Tl, T2, Til, T12.

1 2 3 4

(6) Evaluation function that shows the distribution of evaluation index from the center of evaluation index and data set ¾J

Find (w, v).

(7) The center of the data set is determined so that the evaluation function is minimized by iterative calculation. (8) Minimum evaluation function value Ci for a predetermined range of THV)

Dependency of m is calculated and J

Select the THV that minimizes m.

(9) Display the evaluation index obtained for the selected THV and the distribution state of the center of the data set.

Here, in (8), J is not necessarily minimized at one point of THV, but may be substantially minimized within a certain range (when it becomes valley-shaped within a certain range of THV). In such a case, select the THV range and select the range internal force THV as appropriate.

In this way, based on the results of heart sound analysis using the FCM clustering method, normal and abnormal heart sounds can be discriminated accurately with a simple configuration.

[0049] (C) Heart sound signal processing for sound quality improvement in auscultation

Listening to the heart sound signal to determine whether the heart sound is normal or abnormal, to reduce the noise of the recorded signal and make it easier to hear, and to transmit the recorded heart sound signal to a remote location In order to reduce the deterioration of the heart sound signal, the signal processing using the feature value waveform is performed based on the vibration model described in (A). It is advantageous to use a digital circuit for the vibration model type conversion circuit, and the auscultation heart sound is preferably converted into a digital heart sound signal by first performing AZD conversion before being input to the conversion circuit. Further, if the efficiency of signal processing is important, it is preferable to perform normality representing the signal value as a ratio to the maximum value.

FIG. 10 shows the result of obtaining the vibration response X by giving a heart sound signal to the vibration model, which is the same as FIG. The horizontal axis represents time, but for time t, the sampling period At is used as a reference, and t = i At, and the time is indicated by a discrete variable i, the heart sound signal is Y (i), vibration Express the response as x (i). The vertical axis indicates the signal intensity.

[0051] Although the heart sound signal Y (i) has positive and negative values, S (i) (S (i) = IY (i) I) takes the absolute value of Y (i) when looking at the action of vibration. Can be considered as heart sound data. Also, from the relationship with equations (1) and (2), S (i) Zm is changed to S below. In Fig. 10, the gray waveform is the original waveform of the input heart sound data S (i). The solid line waveform shows the vibration response x (i). This waveform data is the characteristic value waveform data.

[0052] In the heart sound data waveform S (i) and feature value waveform data x (i) in FIG. The portion that corresponds to the heart sound, and the portion with a low value between the peaks contains noise. In particular, since the heart sound signal obtained initially often contains high-frequency component noise, the heart sound data Y (i) force is also extracted by wavelet analysis to extract high-frequency components of 5 kHz or more or 2.5 kHz or more. ) Is preferably used as heart sound data.

W

[0053] In the present invention, the heart sound data S (i) and the feature value waveform data x (i) are obtained by multiplying the peaks by matching the peaks, thereby reducing the noise in the heart sound data and reducing the heart sound itself. By emphasizing, it is considered that the heart sound data is finally reduced in noise.

[0054] By the way, the feature value waveform data x (i) in FIG.

W

There is a delay (k). For this purpose, it is necessary to obtain the phase delay k and shift the phase of the waveform of x (i) by this phase delay, that is, take the product with the phase of S (i). Phase lag

W

k is

[Equation 9]

∑ ( ^S w U) ― S _Warg ) W (x (j) ― X _ayg )

7-0

Therefore, the phase difference k is obtained as the value of i when (i) is maximized. Here, p (i) is a sequence of cross-correlation functions, and S and X represent average values. This k is a sequence p (i)

Wavg avg

Used as the value of i at the maximum. In computation, the phase difference can be substantially eliminated by using k.

[0055] When the waveform of the feature value waveform data X (i) is shifted by the phase delay k, as shown in Fig. 11, the heart sound data waveform S (i) and the feature value waveform x (i) have a peak portion. Substantially overlap

W

. Then, the phase lag is shifted from the heart sound data waveform S (i) and the feature value waveform 7? (I) x (k

W

Perform a transformation that takes the product of + i). Let S (i) and x (i) be

W

S (i) = [S (1) S (2) ---- S (N-k)]

W W W w

x (k + i) = [x (k + l) x (k + 2) ---- x (N)]

When expressed in a matrix as shown in Fig. 2, feature value waveform data with a phase shifted from the heart sound data S (i)

W

Conversion of product with x (k + i) is a matrix operation TS (i) = S (i) -x (k + i) ^T

w

It is represented by Where x (k + i) ^T is a transposed matrix.

[0056] The output heart sound data of TS (i) obtained in this way was reproduced by enhancing the heart sound of the peak portion and weakening the noise portion of the original heart sound data. In this case, the sound quality is easy to hear, and it is also accurate for analyzing heart sounds. In this way, the heart sound signal processing method for enhancing the heart sound and weakening the noise according to the present invention is shown as a flow chart in FIG.

[0057] (D) Automatic adjustment of sound recording volume

In the auscultation device, the heart sound is converted into a heart sound signal with a microphone that converts it into an electrical signal, and the heart sound signal is listened to with a earphone to determine whether the heart sound is normal or abnormal, and as described in (B) and (C). As the processing of a heartbeat signal, there is a form of performing a heartbeat analysis or processing for improving the sound quality of the heartbeat. When using a heartbeat signal in this way, the heartbeat signal is first recorded at the stage of recording. It is important to adjust the volume to an appropriate level

[0058] In order to adjust the recording volume of the heart sound in an auscultation system in which the heart sound signal recorded by the probe is recorded in the automatic volume adjustment recording section by adjusting the volume and then transmitting it.

(i) Maximize the amplification volume of the microphone built into the probe's chest piece and attenuate it to an appropriate level using a digitally controlled audio attenuator

(ii) Detects the volume of the microphone built into the probe chest piece and amplifies it to an appropriate level by a gain-adjustable amplifier

There is a method like this. Although method (i) is suitable for heart sound feature numerical analysis, it cannot be amplified because the volume is set to an appropriate level by lowering the volume. In the method (ii), the volume can be increased or decreased. However, if the input signal is too small, it becomes difficult to determine the data force volume level after AZD conversion.

[0059] In the present invention, the amplification degree of the heart sound signal is controlled from the data after the AZD conversion, and sound analysis is considered with respect to the obtained heart sound data. This method is used. In other words, the amplification level at the probe is maximized and sent to the volume control section. Set it to come out.

[0060] The basic idea of adjusting the recording volume of heart sounds according to the present invention is that the first relatively short !, data for gain adjustment is acquired during time T, and the subsequent time T is acquired. The ab heart sound signal should be properly amplified and used. Thus, appropriate gain adjustment is made by setting how much the heart sound signal is amplified in a short time before amplifying the heart sound to be used. Time to acquire data for gain adjustment T

a is about 1 to 3 seconds, and about 2 seconds is most appropriate. This is because the normal heartbeat period is 0.8 to 1 second, and if heart sounds of about 2 periods can be collected, it can be used as data for gain adjustment. Time to amplify heart sound to use T is about 4-12 seconds

b

It is. This is because, in heart sound feature value waveform analysis, if there are about 10 cycles of data, a reasonable analysis result can be obtained.

[0061] Based on the basic concept described above, the recording volume of the heart sound is adjusted according to the following procedure. This volume adjustment is performed according to the following procedure according to the operation of the microcomputer on the AD signal of the detected heart sound signal. At this time, the time T (1 to 3 seconds) for acquiring data and the heart sound to be used are amplified for gain adjustment as described above

a

Set the time T (4 to 12 seconds) to be used and the data acquired for gain adjustment.

b

The up gain necessary to amplify the heart sound to be used is set as a gain reference table.

[0062] <Procedure for adjusting the recording volume of heart sounds>

(1) Signal adjustment is performed on the detected heart sound signal and converted to 8-bit (10-bit or 12-bit) digital data by the AZD converter.

Import into a.

(2) Average value of x (i) indicating the magnitude of N heart sounds during time T Y (Y = [∑x (i)]

a

/ N).

(3) Compare the Y value with the gain reference table to determine the up gain.

(4) Set the gain to the control amplifier and output the heart sound signal during the time Τ following the time Τ.

a b

Amplified by the set up gain and recorded as an appropriate volume.

However, in (2), it is appropriate to exclude the noise part when calculating the average value Y. Threshold value x is defined for heart sound volume x (i), and x (i)> x during time T

Take the average of the number of x (i) that become O a 0 as N! /.

[0063] The above steps (1) to (4) are carried out at least once, but it is usually recommended to repeat several times (2 to 5) times. The automatic adjustment of the heartbeat recording volume in steps (1) to (4) is performed by the microcomputer 23. The microcomputer 23 holds a program necessary for such control, a gain reference table, and the like.

FIG. 13 is a diagram showing a state of a heart sound signal when a heart sound is recorded by the procedures (1) to (4). In Fig. 13, (a) shows the original heart waveform signal, (b) shows the heart waveform signal during time T for calculating the gain, and (c) shows the up-state obtained by the signal in (b). By gain (aa

) Of heart sound signals recorded in amplified during time T following each time T

A waveform signal is shown. Hearts captured during time τ before recording heart sounds like this a

Based on the sound signal, the up-gain is determined, and the heart sound signal is amplified to an appropriate volume accordingly, so that the heart sound recorded in the stage (c) of FIG. 13 has an appropriate volume.

[0065] Figure 14 shows the NZS ratio when the heartbeat recording volume is automatically adjusted according to steps (1) to (4) and when the heartbeat is recorded without such automatic adjustment of the heartbeat recording volume. The horizontal axis indicates how many heart sounds are recorded. In Fig. 14, ○ indicates that the automatic adjustment of the recording sound volume of the heart sound is not performed at all, the mouth is T b = 10 seconds, and the automatic adjustment of the heart sound is performed according to the procedures (1) to (4). 12 steps to (1) to (4)

Each of the cases where the automatic adjustment of heart sounds is shown. In this way, automatic adjustment of heart sounds according to steps (1) to (4) improves the NZS ratio, which is advantageous for monitoring auscultatory sounds, and is also accurate for analyzing heart sounds. Will be.

[0066] (E) Form of auscultation device

FIG. 15 schematically shows the form of the auscultation device according to the present invention. As a whole, the auscultation device detects the heart sound, and the volume of the heart sound signal detected by the probe a is appropriately increased. The sound volume processing and recording section b, which adjusts the sound volume, and the sound volume processing section c, which performs sound volume enhancement and noise reduction processing for the heart sound signals recorded and adjusted by the automatic volume adjustment recording section b, Automatic sound volume adjustment recording unit b Adjusts the sound volume by the heart sound signal recorded by S, and performs heart sound analysis to determine whether the heart sound is normal or abnormal. It has a heart sound analysis processing unit d and a receiving unit e that generate data to be provided separately.

[0067] The heart sound signal recorded by adjusting the volume at the automatic volume adjustment recording unit b, or the heart sound signal subjected to the heart sound enhancement and noise reduction processing by the heart sound signal processing unit c can be heard by the monitoring means. In addition, it is transmitted to a remote location by the transmission means. The result of the analysis of the heart sound signal recorded by the automatic sound volume control unit b with the volume adjusted is displayed on the monitor or the result of the analysis process is displayed. It is transmitted by the transmission means. The receiver e can receive the transmitted heart sound signal and analyzed data and use it to determine whether the heart sound is normal or abnormal, or to process the received signal and data. . FIG. 15 shows an example of the configuration, and in addition to the probe a, it is possible to combine it as necessary. Hereinafter, the element parts in the apparatus configuration of FIG. 15 will be described separately.

[0068] <Automatic volume control recording unit>

FIG. 16 shows elements of the auscultation probe a, the automatic volume control recording unit b, and the receiving unit e in the auscultation apparatus of FIG. In this case, the recorded heart sound signal is transmitted with the volume adjusted, and is received by the receiving unit e. However, the heart sound signal processing unit c in FIG. A form or a form in which the transmitted heart sound is received by the receiving unit e and the heart sound signal processing unit c performs enhancement of the heart sound and noise reduction processing can be considered.

[0069] In FIG. 16, an auscultation probe a includes a chest piece 11 that includes a microphone that detects a heart sound and converts it into an electrical signal, an amplification unit 12 that amplifies the detected heart sound signal, and an amplified heart sound. Has a headset to listen to. The amplifying unit 12 includes a preamplifier, a filter, and a power amplifier, and is amplified to an appropriate level and a heart sound signal is sent to the volume automatic adjustment transmission module 10.

The automatic volume adjustment recording unit b includes a signal adjustment unit 21, an AZD conversion unit 22, a microcomputer 23, a high-pass filter 24, an amplification adjustment unit 25, and a transmitter 26. The signal adjustment unit 21 includes a high-nos filter and a signal adjustment circuit. The signal adjustment unit 21 adjusts the heart sound signal received from the amplification unit 12 of the auscultation probe a and sends it to the AZD conversion unit 22 and the no-pass filter. 24 to amplifying unit 25 via 24. Amplification adjustment unit 25 is also a high-pass signal conditioner of signal adjustment unit 21 It includes a control amplifier, signal conditioner, and high-pass filter that amplifies the heart sound signal sent through the filter 24 to an appropriate gain in response to a command from the microcomputer.

[0071] The heart sound signal sent to the AZD conversion unit 22 is converted into 8-bit (10-bit or 12-bit) digital data and processed by the microcomputer 23. Thus, the amplification of the heart sound signal is controlled. In other words, the time Ta (1 to 3 seconds) for acquiring data for gain adjustment and the time Tb (4 to 12 seconds) for amplifying the heart sound to be used are set in advance. Keep a reference table that gives the up-gain. Based on the average intensity of the heart sound signal obtained with the auscultation probe a during time T

The reference table power is calculated so that the up gain is obtained and the microcomputer control unit performs amplification control of the heart sound signal, and the time T following the time T according to the obtained up gain under the control of the microcomputer control unit. Amplifying the heart sound signal between

a b

In addition to recording, the heart sound signal is taken out and listened to by the earphone, and the amplification adjustment unit 25 amplifies and adjusts it so that it can be transmitted by the transmitter 26.

[0072] The receiving unit e has a receiver 31, an adjusting unit 32, and an amplifying unit 33 for receiving a heart sound signal transmitted from the transmitter 26 of the automatic volume adjustment recording unit b. The signal is monitored by earphones or used for heart sound analysis by a computer. Further, it can be amplified as an analog output by an amplifier. Automatic volume adjustment and recording unit b side force receiving unit The transmission to the e side may be performed by wireless transmission via an antenna or via a cable.

[0073] <Heart sound signal processor>

17 (a) and 17 (b) show the configuration of the heart sound signal processing unit c. Fig. 17 (a) shows the whole heart sound signal processing unit, and the heart sound signal processing unit c includes the AZD conversion unit of the automatic volume adjustment recording unit b obtained after the heart sound detection unit of the auscultation probe a. The signal Y (i) converted into a digital signal at is input. The signal adjustment unit 43 that takes the absolute value of the AZD-converted heart sound signal Y (i) and performs processing such as normal 匕, and the signal S (i) from the signal adjustment unit 43 is input and the feature value waveform X (i ) Forming the vibration model 44 and the signal from the signal adjustment unit 43. Filter part 45 that outputs noise data S (i) and heart sound from filter part 45

W

The data and the characteristic value waveform data from the vibration model 44 are input and the signal processing is performed, and the heartbeat is emphasized and the noise is weakened. Yes. Here, the signal adjustment unit S (i) the force signal S (i) is input to the vibration model 44. The vibration model 14 is converted to the heart sound data S (i) from which high-frequency component noise has been removed in the force filter unit 45. In that case, the same heart sound data

W

The data SW (i) is also input to the conversion circuit unit 46. 47 is a parameter setting unit for setting parameters in the vibration model 44.

FIG. 17 (b) shows the conversion circuit unit 46 in FIG. 17 (a) in more detail. The conversion circuit unit 46 includes a phase lag calculation unit 51 and an integration conversion unit 22. It is configured. The heart sound data S (i) and the feature value waveform data x (i) are multiplied by the phase delay calculation unit 51, respectively.

W

The value of the phase delay k input to both of the conversion units 52 and obtained by the phase delay calculation unit 51 is input to the multiplication conversion unit 52. The multiplication conversion unit 52 shifts x (i) by the phase delay k to x (k + i), calculates the product with S (i), and outputs the result.

W

[0075] The heart sound signal processing unit shown in FIGS. 17 (a) and 17 (b) includes a part for processing the digitized signal in the A / D conversion unit 12 and a memory for recording the heart sound signal. It can be configured as a digital circuit and built into the auscultation device, or it can be formed as a unique device that processes the signals using the recorded heart sound signals.

[0076] <Heart sound analysis processing unit>

FIG. 18 shows the configuration of the heart sound analysis processing unit d. a is a heart sound detection unit, b is an automatic volume adjustment recording unit that performs AZD conversion of the heart sound signal detected by the heart sound detection unit a and adjusts the volume, and the heart sound data that has been AZD converted and volume adjusted is the heart sound processing analysis processing unit d Is input. The heart sound analysis processing unit d includes a parameter setting unit 61 that sets model parameters of the vibration model, a feature value waveform generating unit 62 that generates feature value waveform data of heart sound data under the set model parameters, and a threshold value (THV). Means for obtaining the evaluation index 63, Means for obtaining the center of the data set defined by using the member function from the evaluation index 64, Means for obtaining the evaluation function from the evaluation index and the center of the data set 65, Evaluation by iterative calculation Means to determine the center of the data set so that the function is minimized 66, for a given range of THV Means 67 for determining the THV that minimizes the minimum evaluation function value Ci), and means 68 for sending display data such as an evaluation index for the selected THV and the center of the data set to the display unit 70. . The parts 62 to 67 perform arithmetic processing on the input data according to the set parameters, and these arithmetic processes may be performed to form a dedicated circuit including a memory or as shown in FIG. It may be configured to be executed by a personal computer equipped with a program for performing arithmetic processing according to the above flow.

[0077] The display unit 70 displays data obtained as a result of heart sound analysis, and a display having a screen such as a liquid crystal panel may be used. The display contents are displayed as a graph without numerical values of the evaluation index and the distribution status at the center of the data set. By displaying such heart sound data, normality and abnormality of heart sounds can be quantitatively grasped with accuracy. The result obtained by the arithmetic processing in the heart sound analysis processing unit d may be transmitted to the receiving unit at a remote location by the transmission means in addition to being displayed on the display unit 70 as display data.

The heart sound analysis processing unit d requires a means for accumulating and holding necessary items such as definition formulas necessary for the analysis of heart sound data, and further, actual data actually obtained by heart sound analysis is stored in a database. By providing a means for storing and storing the data, the data can be used as comparison data when performing a new heart sound analysis.

Industrial applicability

[0078] The present invention can be used to perform automatic adjustment of heart sound recording volume, improvement of sound quality of heart sound signals, and analysis processing of heart sound signals, respectively, and as an auscultation device in a combination of them. It can also be used.

Brief Description of Drawings

FIG. 1 is a conceptual diagram showing an eardrum vibration model.

[Fig. 2] (a) shows the results of obtaining the vibration response X of the normal heart sound force based on the tympanic membrane vibration model, and (b) shows the mitral regurgitation heart sound vibration based on the tympanic membrane vibration model. It is a figure which shows the result which calculated | required the response.

[Fig. 3] (a) is a diagram showing how to obtain the evaluation index in the case of normal heart sounds, (b) is a diagram showing the correlation of the evaluation index, and (c) is a diagram showing the frequency of the evaluation index. It is.

(4) (a) is a diagram showing the relationship between the characteristic value waveform representing the vibration response and the threshold, and (b) and (c) It is a figure which shows the condition where the correlation of an evaluation index changes with threshold values.

[Fig. 5] (a) and (b) are diagrams showing examples of defining the center of data collection by the PCM clustering method for the correlation of evaluation indices, and (c) is the minimum evaluation obtained by the PCM clustering method. It is a figure which shows the dependence with respect to the threshold value of a function value.

[Fig. 6] (a) and (b) are graphs showing the correlation of evaluation indices with threshold values ranging from 10% to 70%. (C) and (d) are threshold values ranging from 30% to 60%. It is a figure which shows the correlation of the table | surface or index | exponent shown in the range.

7) (a) for arrhythmia (AF), (b) for mitral stenosis (MS), (c) for aortic insufficiency (AR) (D) for arrhythmia (AF), (e) for mitral stenosis (MS), and (f) for aortic atresia (AR) ), It is a diagram showing the dependence of the center of the data set on the threshold value.

[Fig. 8] (a), (b) for arrhythmia (AF), (c), (d) for mitral stenosis (MS), (e), (f) for aortic regurgitation It is a figure which shows the correlation of an evaluation index, respectively in the case of (AR).

FIG. 9 is a diagram showing a flow of heart sound analysis according to the present invention.

[10] It is a diagram showing a characteristic value waveform as a vibration response by heart sound data and a vibration model.

FIG. 11 is a diagram showing the characteristic value waveform in FIG. 10 shifted by the phase delay.

FIG. 12 is a diagram showing a flow of processing of a heart sound signal.

[13] FIG. 13 is a diagram showing a state of a heart sound signal when a heart sound is recorded.

FIG. 14 is a diagram showing a comparison of NZS ratios when the heartbeat recording volume is automatically adjusted according to the present invention and when heartbeats are recorded without automatic adjustment.

[15] FIG. 15 is a diagram showing a schematic configuration of an auscultation apparatus according to the present invention.

FIG. 16 is a diagram showing an automatic volume adjustment recording unit and a consultation unit in the configuration of FIG.

FIG. 17 (a) is a diagram showing a heart sound signal processing unit in the configuration of FIG. (b) It is a figure which shows especially about the conversion circuit part among (a).

[18] FIG. 18 is a diagram showing a heart sound analysis processing unit in the configuration of FIG. Explanation of symbols

a auscultation probe

b Automatic volume control recording section

c Heart sound signal processor

d Heart sound analysis processor

e Consultation Department

1 object

2 Spring

3 Damper

11 Chestpiece

12 Amplifier

21 Signal adjustment section

22 AZD converter

23 Microcomputer

25 Amplifier

26 Transmitter

43 Signal adjustment section

44 Vibration model

45 Huinoleta 咅 ^

46 Conversion circuit

47 Parameter setting section

51 Calculation unit

52 Multiplication converter

61 Parameter setting section

62 Feature value waveform generation means

63 Means for obtaining the evaluation index

64 Means of finding the center of a data set

65 Means for obtaining the evaluation function Evaluation function minimization means

Means for determining the smallest evaluation function value range Display section

Claims

The scope of the claims

[1] Setting model parameters for the vibration model;

Detecting heart sounds and thereby obtaining heart sound data;

Generating characteristic value waveform data under the set model parameters for the obtained heart sound data;

Obtaining an evaluation index indicating a time width and a time interval of a peak of the feature value waveform data with respect to a threshold value (THV);

The evaluation index power is the center of the data set defined by the fuzzy member function (w),

Seeking (V),

Obtaining an evaluation function ¾J (w, v) representing the distribution of the evaluation index from the center of the evaluation index and the data set;

The center of the data set is determined so that the evaluation function is minimized by iterative calculation, and the minimum evaluation function value CF for a predetermined range of THV)

Find the dependency of m, select the THV that minimizes J m within that range,

Display the evaluation index obtained for the selected THV and the distribution state of the center of the data set;

A method of processing an auscultatory heart sound signal, comprising performing heart sound analysis for detecting an abnormal heart sound.

[2] The evaluation index is the time width (T1, T2) and time interval (T11, T12) of sound I and sound II in the characteristic value waveform data for THV, and W = {w} and V = {v}. , D = || v— zjjk, j

[Equation 7]

J _m (W, V) -∑∑ {id (7)

, ′ — 1—1 The method for processing an auscultatory heart sound signal according to claim 1, wherein:

[3] means for setting model parameters of the vibration model; Heart sound detecting means for detecting heart sounds and thereby obtaining heart sound data; means for generating feature value waveform data for the obtained heart sound data under set model parameters;

Means for obtaining an evaluation index indicating a time width and a time interval of a peak of the feature value waveform data with respect to a threshold value (THV);

The evaluation index power is also in a data set defined by using a fuzzy member function (w),

Means to find the heart ( _Vi ),

Evaluation function representing the distribution of evaluation index from the center of the evaluation index and data set

Means for obtaining J (W, V);

Means for determining the center of the data set by iterative calculation so that the evaluation function is minimized;

Means for obtaining a minimum evaluation function »that minimizes the evaluation function;

J for a given range of THV

Dependency of m is calculated and within that range (J)

means to select the THV that minimizes m,

Means for displaying the evaluation index obtained for the selected THV and the distribution state of the center of the data set;

An auscultation apparatus comprising a heart sound analysis processing unit for performing heart sound signal analysis for detecting abnormal abnormal heart sounds.

[4] The evaluation index is the time width (T1, T2) and time interval (T11, T12) of sound I and sound II in the characteristic value waveform data for THV, and W = {w} and V = {v}. , D = || v— z

j j k, j

[Equation 7]

^ (^ n = ∑∑ (,,) "k, _/ ) ² (7)

4. The auscultation device according to claim 3, characterized in that it is represented by '• 1-1.

[5] Setting the model parameters of the vibration model to form the vibration model;

Detecting a heart sound and obtaining a heart sound signal; Applying the obtained heart sound signal to the vibration model to obtain output characteristic value waveform data;

The heart sound signal or a signal obtained by removing high-frequency noise from the heart sound signal is used as heart sound data, a phase lag is calculated by cross-correlating the heart sound data and the feature value waveform data, and the heart sound data and the feature value waveform data are calculated. Shifting the phase of the feature value waveform data by the phase delay so that there is substantially no phase difference between

Obtaining output heart sound data as a product of the heart sound data and feature value waveform data shifted in phase by the phase delay;

A processing method for an auscultatory heart sound signal, characterized by performing a heart sound signal processing for sound quality improvement in the auscultation of each step.

6. The auscultatory heart sound signal processing method according to claim 5, wherein the heart sound signal is normalized before the heart sound signal is applied to the vibration model.

[7] A vibration model for outputting feature value waveform data corresponding to the heart sound data by inputting a heart sound signal detected by the heart sound detecting means or a signal obtained by removing the high frequency component noise from the heart sound data as the heart sound data. When,

A phase lag calculation unit for calculating a phase lag of the feature value waveform data with respect to the heart sound data by taking a cross-correlation between the feature value waveform data output by the vibration model and the heart sound data;

Multiplication conversion unit that takes the product of the feature value waveform data and the heart sound data, the phase of which is shifted by the amount of the phase delay so that there is substantially no phase difference between the heart sound data and the feature value waveform data When,

An auscultation apparatus comprising a heart sound signal processing unit that performs heart sound signal processing for detecting a powerful abnormal heart sound.

8. The auscultation apparatus according to claim 7, further comprising normalization means for normalizing a heart sound signal before inputting to the vibration model.

[9] a probe that converts heart sounds into electrical signals;

An automatic volume control recording unit for recording the heart sound by adjusting and amplifying the signal of the heart sound obtained by the probe; When the automatic sound volume adjustment recording unit amplifies the heart sound signal based on the average intensity during the time T of the heart sound signal obtained by the probe a

A microcomputer control unit for obtaining an up gain and performing amplification control of a heart sound signal, and following the time T according to the obtained up gain under the control of the microcomputer control unit

a Amplify the heart sound signal during time T and use the amplified signal as the heart sound to be recorded b

An auscultation device comprising an automatic adjustment means for recording sound of heart sounds, comprising an amplification adjustment section adapted to be removed.

[10] The microcomputer control unit holds a relationship of the magnitude of the up gain for amplifying to an appropriate sound volume with respect to the average intensity of the calculated heart sound signal, and refers to the table. 10. The auscultation apparatus according to claim 9, wherein a heart sound signal is amplified to an appropriate volume.

[11] The time T-force ^ is a time within a range of 3 seconds, and the time T-force is within a range of 12 seconds.

a b

The auscultation device according to claim 9, wherein the auscultation device is time.

12. The time T is a time within a range of 8 to 10 seconds.

b

Auscultation device.

[13] The means for automatically adjusting the heart sound recording volume according to any one of claims 9 to 12 is configured as a transmitting unit so that an amplified heart sound signal can be transmitted to the receiving unit. Auscultation device characterized by.