JPH03133426A - Cardiac sound meter - Google Patents

Abstract

PURPOSE:To separate a I sound, a II sound and cardiac noise with high accuracy and to exactly calculate cardiac noise distribution by adding the I sound and the II sound which can be detected simultaneously by plural cardiac microphones and enlarging the amplitude. CONSTITUTION:Cardiac sound signals detected by each cardiac sound microphone 11-130 are amplified by amplifiers 21-230, and converted from analog signals to digital signals by A/D converters 31-330. These cardiac sound signals are attenuated as to its high frequency components by low-pass filter circuits 41-430, become positive side waveforms by absolute value circuits 51-530, a signal obtained by adding simultaneously the positive side waveforms from each channel by an adder 6 is inputted to a smoothing circuit 8 and a fine undulation is eliminated and a I II sound extracting signal is obtained. By applying a cardiogram signal from a cardiogram sensor 7 to a time division circuit 9 in addition to this I II sound extracting signal, a time division signal for discriminating the I sound and each start time point and end time point of the I sound is outputted, and by applying this time division signal an the cardiac sound signal to an arithmetic circuit 10, cardiac noise intensity distribution is calculated.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a phonocardiograph used for diagnosing heart disease, and more particularly, to a phonocardiograph that facilitates the separation and extraction of heart murmurs from heart sound signals with high precision. This relates to phonocardiography that can be performed at

[Prior art] Heart sounds are the opening and closing of the atrioventricular valves and semilunar valves and the sound of the atria → ventricle → when the heart repeatedly contracts and expands to pump blood to various tissues.
This is the sound caused by blood flow to the large blood vessels. As shown in FIG. 4, the heart sounds consist of the ``■'' sound that occurs at the beginning of systole, the systolic heart murmur that occurs during the systole, the ``■'' sound that occurs during the early diastole, and the diastolic heart murmur that occurs during the diastole. In healthy individuals, systolic murmurs and diastolic murmurs are not audible, but when abnormalities are observed in the heart, heart murmurs with characteristics unique to each case are audible. When diagnosing heart disease, information on heart sounds such as ■sound, ■sound, etc. is also used, but in many cases, temporal changes in heart murmur,
Diagnosis is made based on the intensity, tone, and location of the heart murmur on the chest.

To calculate the intensity of a heart murmur, identify the ■sound and ■sound, and generally define the period between the ■sound and ■sound as the systolic phase, and the period between the ■sound and ■sound as the diastole, and calculate the heart murmur within that interval. This is done by finding the mean square amplitude of .

However, compared to heart murmurs, heart sound components (■ sounds and ■
(sound) has a large amplitude, so for example, if the systolic setting is incorrect and a part of the I sound is captured, part of the ■ sound will be included as a systolic murmur, so the systolic period must be set correctly. There is a problem in that the heart murmur intensity distribution differs greatly compared to the heart murmur intensity distribution. Therefore, in order to measure the heart murmur intensity with sufficient accuracy, it is important to correctly distinguish between the ■ sound, the ■ sound, and the heart murmur.

Conventionally, in order to distinguish ■ sounds and ■ sounds from heart sound waveforms, envelope detection was performed on heart sound waveforms from a single measurement point, or envelope detection was performed on heart sound signals from heart sound sensors placed at multiple points on the chest wall. After performing route detection and adding these multiple envelope route detection signals by matching the detection points or combining these signals to the maximum value, a signal in which ■ sound and ■ sound are particularly noticeable is created. , ■Sound, ■Distinguishes sound from heart murmur.

[Problem to be solved by the invention]

However, if the heart murmur is loud, if the heart murmur occurs following the ■ sound, ■ sound, or if the ■ sound, ■ sound is weak, the heart murmur is accompanied by the ■ sound, ■ sound. The above method cannot be said to be effective for all cases, as the boundary between the two is unclear, making identification difficult.

The present invention focuses on the above-mentioned problems, and even when the boundary between the detected heart sound signal ■, ■, and the heart murmur is unclear, the present invention can easily and accurately separate the ■, ■, and the heart murmur. The object of the present invention is to provide a phonocardiograph that can calculate heart murmur distribution more accurately.

[Means to solve the problem]

The configuration of the phonocardiograph of the present invention to achieve the above objects is as follows:
A plurality of heart sound microphones are placed at multiple positions on the chest wall, and the heart sound signals detected by each of the heart sound microphones are
The absolute value of the output signal from this low-pass filter circuit is taken and applied to an absolute value circuit that creates a positive waveform, and the absolute value signal for each heart sound microphone output from this absolute value circuit is detected at the time of detection. are given to the adder circuit that adds them together,
The signal output from this adder circuit is given to a smoothing circuit to smooth the waveform, and this smoothed signal, or this smoothed signal and electrocardiographic signal, is given to a time division circuit to (2) A time-division signal for identifying the start and end points of each sound is output, and the time-division signal and the heart sound signal are fed to an arithmetic circuit to calculate the heart murmur intensity distribution.

[For production]

The present invention is based on the following points: ① and ② sounds can be heard over a relatively wide range on the chest wall, whereas heart murmurs can be heard only in a very limited range, and only one sound.

■It was developed based on the fact that while sound consists of low-frequency components, heart murmur signals contain many high-frequency valves. That is, the low-pass filter circuit attenuates and outputs the high frequency components of the input signal, and the absolute value circuit inverts the negative wave of the manually input waveform to the positive side and outputs it, and the absolute value for each heart sound microphone is output. By adding the signals in the adding circuit, the ■ sounds and ■ sounds that can be detected simultaneously by multiple heart sound microphones are added to increase the amplitude, and the localized heart murmurs are added to the signals common to each heart sound microphone. In addition, since the high frequency components are attenuated, the amplitude does not increase even when the absolute value signals are added. Therefore, the signal outputted by the adding circuit, that is, the Inn oil extraction signal,
The peaks and troughs of the waveforms of sounds and heart murmurs are separated and emphasized to facilitate identification.

■Sound, systolic murmur, ■Sound and diastolic murmur can be easily and accurately identified by inputting the above-mentioned Inn oil extraction signal and electrocardiographic signal into a time division circuit. As a means for this, the sound oil extraction signal (1) and (2) can be clearly separated from the heart murmur, as will be explained in detail below, so this signal alone can be used to identify the heart murmur.

〔Example〕

Hereinafter, the present invention will be explained in detail by way of an example with reference to the accompanying drawings.

FIG. 1 is a block circuit diagram of the phonocardiograph of this embodiment. In the figure, 11-l. Each of these microphones is a heart sound microphone, that is, a sensor for detecting heart sounds, and in this example, they were attached to 30 locations on the chest wall of the subject (not shown). 21-2
．． . 4. is an amplifier; 31 to 33 degrees is an A/D converter; 4. ~4
．． . is a low-pass filter circuit, 5. ~5. . is an absolute value circuit, 6 is each channel, that is, each heart sound microphone 1, -1,
.

7 is an electrocardiographic sensor; 8 is a smoothing circuit that removes fine undulations from the signal waveform output by the adder 6; 9 is an inn oil extraction signal and an electrocardiographic signal; A time division circuit that identifies sounds and creates time division signals, l
O is an arithmetic circuit that calculates a cardiac murmur intensity distribution, which calculates an intensity distribution for each frequency band and each time from the cardiac waveform of each channel.

In the above circuit configuration, each heart sound microphone 1. ~
The heart sound signal detected by 13° is transmitted to amplifiers 2, -2. . and is amplified by A/D converters 3I to 3. .

Converts an analog signal to a digital signal.

A/D converters 31-3. . The heart sound signal outputted by the low-pass filter circuit 4. ~4. . The high frequency components are attenuated by the absolute value circuits 51 to 5. . The positive side waveforms from each channel are simultaneously added by an adder 6, and the resulting signal is input to a smoothing circuit 8 to remove fine undulations to obtain a sound oil extraction signal. The upper limit of the frequency of one sound in a normal person is approximately 250 Hz, and the upper limit of the sound of ■ is approximately 300) 1 z. However, the low-pass filter circuits 4I to 43° used in this example A device with an upper frequency limit of approximately 100H2 was used.

The Inn oil extraction signal of this embodiment will be explained with reference to FIGS. 2 and 3. FIG. 2 is a graph comparing the heart waveform (upper row) of the left sternal border of the fourth intercostal space of a patient with aortic valve stenosis and the I■ sound oil extraction signal (lower row). As can be understood from FIG. 2, even when a heart murmur occurs following the sound ■, the location where the heart murmur occurs is localized near the left edge of the sternum in the fourth intercostal space, so the heart sound microphone 1. ~l,
. The Inn oil extraction signal added for the above, the r sound, and the ■ sound can be made into a waveform that emphasizes the sound, and can be easily distinguished from a heart murmur.

FIG. 3 is a graph comparing the heart waveform of the right sternal border of the second intercostal space of a patient with ventricular septal defect (upper row) and the Inn oil extraction signal of this patient (lower row). Even when the I sound is weak and it is extremely difficult to distinguish between the ■ sound and the systolic heart murmur, as shown in the upper waveform of Figure 3, the I sound oil extraction signal ■ sound, the ■ sound and the heart murmur can be distinguished. The boundary between them becomes a minimum point and can be clearly distinguished. If there is no minimum point, the boundary between the heart murmur, the I sound, and the ■ sound is the point where the sound becomes sufficiently lower than the height of this peak value within a certain period of time before and after the peak value of the ■ sound. It can be done. In the case of this embodiment, the above-mentioned fixed time is: (1) 70 milliseconds before and after the sound, (2) 60 milliseconds before and after the sound, that is, a width of 140 milliseconds for the I sound, (1)
The boundary was defined as a position where the value was sufficiently low for the sound within a width of 120 milliseconds.

In addition to the Inn oil extraction signal, the time division circuit 9 is supplied with an electrocardiographic signal obtained by amplifying the electrocardiographic signal from the electrocardiographic sensor 7 with an amplifier 2c and converting it into a digital signal with an A/D converter 3c. It will be done. The time division circuit 9 obtains an R wave from the electrocardiogram signal, and based on this R wave, one sound from the sound oil extraction signal,
■After identifying the sound, ■Set the start and end points of the sound; ■Set the start and end points of the sound. The period from the end of the I sound determined in this manner to the start of the ■ sound is the systolic period, and the period from the end of the ■ sound to the start of the first sound is the diastolic period. A time-shared signal is obtained by the systolic and diastolic signals. Further, by further dividing the systolic and diastolic signals into a plurality of periods, it is possible to obtain a time-divided signal for calculating temporal changes in heart murmur intensity distribution, which is useful as diagnostic data. .

The time division signal and each channel's A/D converter 31~
3. . The digitized heart sound signal outputted by the heart sound signal is given to the arithmetic circuit 10 to calculate the heart murmur intensity. Note that the arithmetic circuit IO of this embodiment is provided with a filtering circuit that filters heart sounds into each frequency band. As described above, the heart sound signal was time-divided into each period, and the heart murmur intensity for each frequency band was calculated and used as data for diagnosing heart disease.

As can be understood from the above-mentioned phonocardiograms, the systolic interval is usually shorter than the diastolic interval;
As is clear from the figure, the ■■■ sound output signal of this embodiment is
It is extremely easy to distinguish between sounds and heart murmurs. Therefore, in this embodiment, the sound ■ and the sound ■ can be identified without using the electrocardiographic signal from the electrocardiographic sensor 7 .

That is, in the III sound extraction waveform, the ■ sound and the ■ sound appear as a large mountain. In order to determine which of these mountains is the I sound and which is the ■ sound without using electrocardiographic signals, r
This can be done by taking advantage of the fact that the sounds and ■ sounds appear alternately and that the contraction period of the heart is shorter than the diastole period. That is, when three consecutive peaks are detected from the sound oil extraction waveform,
If the distance between the first and second peaks is shorter than the distance between the second and third peaks, the first and third peaks make an r sound, and the second peak makes an r sound. ■Can be determined as a sound. Conversely, if the distance between the first peak and the second peak is longer than the distance between the second peak and the third peak, the distance between the first peak and the third peak is 4.
゜ and ■ sound, and the mountain in the middle becomes one sound.

However, depending on the symptoms, the length of the contraction period and the diastole period may be approximately the same, or the contraction period may be longer than the diastole period, and in this case, the above method is not effective.

(Effects of the Invention) As explained above, according to the phonocardiograph of the present invention, ■ sounds;
■Since sounds and heart murmurs can be clearly distinguished, it is possible to accurately calculate the intensity distribution of heart murmurs, which is important in diagnosis based on heart sounds.

[Brief explanation of the drawing]

Fig. 1 is a block circuit diagram of a phonocardiograph of the present invention according to an embodiment, and Figs. 2 and 3 compare the heart sound waveform obtained by a heart disease patient using the phonocardiograph shown in Fig. 1 and the sound oil extraction signal. Graph, FIG. 4 is a graph showing an example of a heart sound waveform having a heart murmur. l...Heart sound microphone, 2...Amplifier, 3...
・A/D converter, 4...Low-pass filter circuit, Absolute value circuit, 6...Adder, 7...Electrocardiogram sensor, 8...
Smoothing circuit, 9... Time division circuit, 10... Heart murmur intensity calculation circuit.

Claims (2)

[Claims] (1) A plurality of heart sound microphones are arranged at a plurality of positions on the chest wall, the heart sound signals detected by each of the heart sound microphones are applied to a low-pass filter circuit, and the output signal from the low-pass filter circuit is The absolute value is taken and given to an absolute value circuit that creates a positive waveform, and the absolute value signal for each heart sound microphone output from this absolute value circuit is given to an adder circuit that adds together the detection time points, and this adder circuit outputs The signal is supplied to a smoothing circuit to smooth the waveform, and the smoothed signal and the electrocardiographic signal are supplied to a time division circuit to produce a time division signal that identifies the start and end points of each of sound I and sound II. A phonocardiograph that outputs this time-division signal and the heart sound signal to an arithmetic circuit to calculate a heart murmur intensity distribution. (2) A plurality of heart sound microphones are arranged at a plurality of positions on the chest wall, the heart sound signals detected by each of the heart sound microphones are applied to a low-pass filter circuit, and the output signal from the low-pass filter circuit is converted into the output signal. The absolute value is taken and given to an absolute value circuit that creates a positive waveform, and the absolute value signal for each heart sound microphone output from this absolute value circuit is given to an adder circuit that adds together the detection time points, and this adder circuit outputs The signal is supplied to a smoothing circuit to smooth the waveform, and this smoothed signal is supplied to a time division circuit to output a time division signal that identifies the start and end points of each of sound I and sound II. A phonocardiograph that calculates a heart murmur intensity distribution by applying a time-division signal and the heart sound signal to an arithmetic circuit.