close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JPH06161470

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JPH06161470
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a
body sound measuring apparatus for measuring body sound, and more particularly to a body
sound measuring apparatus for measuring body sound in contact with a body surface.
[0002]
Normally, it is ideal to measure body sounds in a sufficiently quiet environment such as a
soundproof room, but in environments such as a doctor's office or a bedside, various noises such
as human voice and equipment sounds can be used as a microphone. It is difficult to mix and
measure accurate body sounds.
[0003]
2. Description of the Related Art In the conventional measurement of body sound, the
microphone was closely attached to the body surface, but there was a drawback that the ambient
noise was also picked up at the same time.
[0004]
Therefore, the applicant conducted an experiment to investigate the sound insulation effect by
covering the periphery of the microphone with a metal cover.
10-04-2019
1
FIG. 5 is a view showing an experimental method for examining the sound insulation effect by
the metal cover.
Here, we conducted experiments on lung sounds. The microphone 52 is brought into contact
with the chest wall 54, and the periphery of the microphone 52 is covered with a cylindrical
metal cover 53 having an outer diameter of 10 cm and a height of 4 cm. In this state, the lung
sound s of the lung 55 is heard, and this is recorded and analyzed by the lung sound
measurement device 51.
[0005]
As a result, for the noise x, a sound insulation effect of 15 to 30 dB was obtained in the band of
100 Hz to 1 kHz. On the other hand, the chest wall 54 did not overturn the metal cover 53, and
only by strengthening the sound insulation of the microphone 52 itself, only a few dB of sound
insulation effect was obtained.
[0006]
From the above results, it can be seen that the path entering the microphone 52 through the
chest wall 54 around the measurement point is dominant as the mixed path of the noise x.
[0007]
However, as described above, it is inconvenient to use a metal cover every time lung sound is
measured, and in order to be in close contact with the chest wall 54, various shapes are required.
There is a problem that it can not do.
[0008]
The present invention has been made in view of these points, and it is an object of the present
invention to provide a lung sound measurement device capable of removing surrounding noise
and having a simple measurement operation.
[0009]
10-04-2019
2
FIG. 1 is a principle view of a living body sound measuring apparatus according to the present
invention for achieving the above object.
In the present invention, the first microphone 1 for listening to the body sound s in contact with
the body surface 4, the second microphone 2 for listening to the ambient noise x provided apart
from the body surface 4, and the noise x A noise removal means 3 for removing noise
components from the body sound s is provided.
[0010]
The first microphone 1 comes in contact with the body surface 4 to listen to the body sound s
from the lung 5 etc., and the second microphone 2 provided apart from the body surface 4 hears
the ambient noise x .
The noise removal means 3 removes noise components from the body sound s heard by the first
microphone 1 based on the noise x heard by the second microphone 2.
[0011]
As a result, an accurate biological sound from which noise has been removed is measured.
[0012]
An embodiment of the present invention will be described below with reference to the drawings.
FIG. 2 is a view showing a lung sound measurement method by the lung sound measurement
apparatus of the present embodiment.
A microphone 12 is connected to the lung sound measurement apparatus main body 11. The
microphone 12 is composed of a main microphone 12a and an auxiliary microphone 12b. The
main microphone 12 a contacts the chest wall 14 of the human body 13 to listen to the lung
sound s from the lung 15. However, the surrounding noise x is altered via the chest wall 14 and
input as noise ax to the main microphone 12a. On the other hand, the auxiliary microphone 12b
10-04-2019
3
is integrally provided with the main microphone 12a so as to be separated from the chest wall 14
by a predetermined distance, and listens for the ambient noise x.
[0013]
The lung sound s + noise ax and noise x heard by the main microphone 12 a and the auxiliary
microphone 12 b are sent to the lung sound measurement device 11. The lung sound
measurement device 11 is a computer such as a personal computer or a work station, and is
provided with an adaptive noise canceller 11a and a storage unit 11b. The noise canceller 11a
has a function of adding a process to be described later to the signal component of the noise x
input from the auxiliary microphone 12b and subtracting the value from the signal component of
the lung sound s + noise ax input from the main microphone 12a. . The noise canceller 11a may
be configured by hardware using a DSP (Digital Signal Processor) or the like.
[0014]
The lung sound data thus processed by the noise canceller 11a is stored in the storage means
11b and displayed on the display device 17 as required. The data processing operation is
performed by the keyboard 16. The storage unit 11 b may be a RAM or an external storage
device such as a floppy disk or a hard disk.
[0015]
FIG. 3 is a block diagram showing a functional configuration of the noise canceller 11a. The
ambient noise x is sent to the noise canceller 11a via the auxiliary microphone 12b. Further, the
noise x is converted by passing through the chest wall 14 to be ax, mixed with the lung sound s,
heard by the main microphone 12a, and sent to the subtractor 22 of the noise canceller 11a.
[0016]
The noise x (n) input to the noise canceller 11 a at time n is sent to the multiplier 231 and the
delay circuit 241. The multiplier 231 multiplies the coefficient h (1, n) sent from the adaptive
algorithm 21 by the noise x (n) and sends the value to the adder 25. The delay circuit 241 delays
10-04-2019
4
the noise x (n) by the unit delay Z-1 and sends it to the multiplier 232 and the delay circuit 242
of the next stage.
[0017]
In the noise canceller 11a, a total of M multipliers 231 to 23M and (M-1) delay circuits 241 to
24 (M-1) are provided as shown in the figure. Assuming that 1 <i <M, the ith multiplier 23i
multiplies the coefficient h (i, n) sent from the adaptive algorithm 21 by the output from the
delay circuit 24 (i-1) of the previous stage, and adds the adder 25 Send to
[0018]
The adder 25 adds up the outputs from the multipliers 231 to 23 M, and sends the output y (n)
to the subtractor 22. このとき、y(n)は、
[0019]
y (n) = XT (n) H (n) (1) Where X (n) is the tap input vector of the noise canceller 11a as a
transversal filter, XT (n) is the transpose of X (n), H (n) is the filter coefficient vector, and X (n)
and H (n) n) respectively
[0022] 【0022】である。 The subtractor 22 subtracts the output y (n) from the output s (n)
+ ax (n) from the microphone 12a, and outputs an error signal e (n). Here, the error signal e (n) is
[0023] And its mean square error E [e2 (n)] is
[0024] となる。 Here, the first term of equation (5) is irrelevant to the adaptive operation. The
second term is 0 if the signal and the noise may be uncorrelated. Therefore, equation (5) is
[0025] となる。 Therefore, if E [e2 (n)] is minimized, the error signal e (n) approximates the pure
10-04-2019
5
signal s (t) in the sense of least mean square error.
[0026] In order to optimize the filter coefficient vector H (n) of the transversal filter in the sense
of least mean square error, it is sufficient to solve the filter's normal equation, but it is necessary
to calculate the correlation coefficient and the inverse matrix and the efficiency is poor . For this
reason, in this embodiment, the LMS (Least Mean Square) algorithm proposed by Widrow and
Hoff was used. This LMS algorithm is widely used in the implementation of adaptive filters, and
only the results will be described here. That is, in the LMS algorithm, the filter coefficient vector
H (n) is updated as follows.
[0027] Here, μ is a parameter that affects the convergence speed and stability of the filter
coefficient and the deviation from the post-convergence optimum value referred to as
misadjustment.
[0028] FIG. 4 is a diagram showing an example of an experimental result of the lung sound
measurement apparatus 11 of the present embodiment, (A) is a graph of recording data of lung
sound mixed with circumferential noise by only the main microphone 12a, (B) It is a graph of the
result of having performed processing by adding the auxiliary microphone 12b and the noise
canceller 11a. The horizontal axis of each graph is time (seconds), and the vertical axis is an
output level. Also, the numbers on the vertical axis represent normalized ones with the maximum
level of the recording data.
[0029] The lung sound recorded in this experiment is one that can be heard strongly during
inspiration, which is called alveolar respiratory sound. Waveforms appear twice each in the
ranges A1 and A2 in FIG. 4A and in the ranges B1 and B2 in FIG. 4B. Also, in this experiment,
voice of radio news is used as ambient noise.
[0030] As can be seen by comparing FIGS. 4A and 4B, in FIG. 4B, the noise is largely removed
and the lung sound is clearly measured. Further, in this experiment, although both FIGS. (A) and
(B) are applied from 0 seconds, it can be seen that the adaptation processing of the noise
canceller 11a is completed in about 3 seconds in FIG. (B).
[0031] In this embodiment, the lung sound measurement device 11 is applied to a work station
or the like, but the noise canceller 11a is realized by a DSP (Digital Signal Processor) to perform
real-time processing and provide a D / A converter and headphones. As a result, an electronic
10-04-2019
6
stethoscope with high noise resistance can be realized.
[0032] As described above, according to the present invention, the noise removal means removes
noise components from the body sound heard by the first microphone based on the ambient
noises heard by the second microphone. As a result, the accurate biological sound with noise
removed is measured.
10-04-2019
7
Документ
Категория
Без категории
Просмотров
0
Размер файла
15 Кб
Теги
description, jph06161470
1/--страниц
Пожаловаться на содержимое документа