WO2012057406A1

WO2012057406A1 - Non-invasive device and method for measuring bowel motility using bowel sound analysis

Info

Publication number: WO2012057406A1
Application number: PCT/KR2010/009614
Authority: WO
Inventors: 송철규; 서정환; 김거식; 유상훈; 김민호; 서자영
Original assignee: 전북대학교산학협력단
Priority date: 2010-10-26
Filing date: 2010-12-31
Publication date: 2012-05-03
Also published as: KR20120061143A; KR101183709B1

Abstract

The present invention relates to a non-invasive device and to a non-invasive method for measuring bowel motility using bowel sound analysis. The non-invasive device for measuring bowel motility through bowel sound analysis includes: a signal-collecting unit that measures and outputs sounds from the bowel, i.e. a bowel sound; a noise-canceling unit that removes noise from the bowel signal outputted from the unit for collecting signals and which outputs the signal; a parameter-extracting unit that extracts feature parameters from the bowel sound from which the noise was removed; a relationship-deriving unit that derives the relationship between the extracted feature parameters and the colon transit time (CTT) measured in advance; and a unit for diagnosing bowel motility that diagnoses the motility state of the bowel on the basis of the derived relationship.

Description

Apparatus and method for measuring noninvasive bowel motility through bowel analysis

The present invention relates to an apparatus and method for measuring intestinal motility, and more particularly, to an apparatus and method for non-invasive diagnosis of intestinal motility using bowel sound signals generated by the movement of intestinal digestive substances and gases. will be.

In general, the intestinal motility is diagnosed using Barr index, Blethyn index, colon transit time (CTT), and the most widely used method of diagnosing intestinal motility using CTT.

The method of diagnosing intestinal motility using CTT is performed by a subject swallowing a capsule containing a radiopaque marker, and then radiating the large intestine through X-ray or MRI after 1, 3, or 7 days. It is a method of diagnosing intestinal motility by checking the number of markers remaining.

However, such a method for diagnosing motility of the intestine based on the radiation apparatus has a long measurement time of one week, expensive test, and side effects due to frequent radiation exposure (at least three times).

On the other hand, the existing long sound measurement technology has been filed and registered in addition to US Patent Nos. 6,776,776 (02.10.09.) And 6,228,040 (01.05.08) filed by Sandler et al.

However, in the related art, only a stethoscope-type sound collection device and a method for deriving an acoustic variable have a problem in that a method for quantitatively diagnosing motility using the derived variable has not been described.

In addition, in the conventional technology, there is no process of removing noise generated by friction between the microphone and skin in the process of collecting long sound, and thus there is a problem that such a noise component is reflected in a feature variable.

Accordingly, the present invention is to solve the above problems, by using the characteristic parameters derived by analyzing the long-sound signal that is generated in the intestine using a feature variable to estimate the transit time (CTT) based on the quantitative motility It is an object of the present invention to provide an apparatus and method for measuring non-invasive bowel motility through diagnosing bowel sound.

In addition, another object of the present invention is to provide an apparatus and method for measuring non-invasive bowel motility through bowel analysis, which can reduce noise inevitably introduced to collect sound generated in the intestine.

In order to achieve the above object, the intestinal motility measuring apparatus according to the present invention, in the non-invasive intestinal motility measurement device through the long sound analysis, the signal collection unit for measuring and outputting the long sound signal that is the sound generated in the intestine; A noise removing unit for removing noise from the long sound signal output from the signal collecting unit and outputting the noise; A variable extraction unit for extracting feature variables from the noise-removed long sound signal; A relation expression derivation unit for deriving a relation between the extracted feature variables and a previously measured colon transit time (CTT); And an intestinal motility diagnosis unit for diagnosing the exercise state of the intestine based on the derived relational expression.

In addition, the method for measuring intestinal motility according to the present invention, in the non-invasive method of diagnosing intestinal motility through the sound analysis of the bowel, comprising the steps of: measuring a long sound signal which is a sound generated in the intestine; Removing noise from the measured long sound signal; Extracting feature variables from the noise-free long sound signal; Deriving a relationship between the extracted feature variable and the colon transit time through a regression analysis; And diagnosing bowel motility using the derived relational expression.

The apparatus for removing noise from a long sound signal according to the present invention calculates a kurtosis vector of a long sound signal, which is a sound generated in a field, and sets the standard deviation of the vector as a critical point, Eliminating the signal is characterized in that to remove the time invariant signal.

The present invention has the effect of measuring the intestinal motility non-invasive and quantitatively by measuring and analyzing a long sound signal, which is a sound generated in the intestine, from the outside of the subject.

In addition, the present invention can measure the motility of the intestine in a short period of time through the long sound signal, there is an effect that helps in the early diagnosis and prognosis of the disease.

In addition, the present invention can measure the motility of the intestine within a short period of time without exposure to the radiation, there is an effect that can replace the use of conventional radiation equipment.

1 is a block diagram showing the configuration of a non-invasive bowel motility measurement device through the sound analysis according to an embodiment of the present invention.

2 is an exemplary view showing a position for collecting a long sound signal according to an embodiment of the present invention.

3 is an example of a graph showing a long sound signal measured in the present invention.

4 is a flowchart illustrating a method for measuring non-invasive bowel motility through bowel analysis according to an embodiment of the present invention.

FIG. 5 is a flowchart specifically showing noise removing from a long sound signal in the flowchart of FIG. 4.

Hereinafter, with reference to the accompanying drawings, a preferred embodiment of the apparatus and method for measuring the bowel motility using the sound analysis according to the present invention configured as described above in detail.

However, the scope of the present invention is not limited to the following examples, and various modifications can be made by those skilled in the art without departing from the technical gist of the present invention. to be. In addition, the terms or words used in the specification and claims are not to be construed as limiting in their usual or dictionary meanings, and the inventors appropriately define the concept of terms in order to best explain their invention in the best way possible. It should be interpreted as meaning and concept corresponding to the technical idea of the present invention based on the principle to be possible.

In addition, when it is determined that the detailed description of the known function and the configuration related to the present invention may unnecessarily obscure the subject matter of the present invention, it should be noted that the detailed description is omitted.

1 is a block diagram showing the configuration of a non-invasive bowel motility measurement device through the sound analysis according to an embodiment of the present invention, Figure 2 is an illustration showing a position for collecting the long sound signal according to an embodiment of the present invention 3 is an example of a graph showing a long sound signal measured in the present invention, FIG. 4 is a flowchart illustrating a method for measuring non-invasive field motility through long sound analysis according to an embodiment of the present invention, and FIG. 5 FIG. 4 is a flowchart illustrating the steps of removing noise from the long sound signal in the flowchart of FIG. 4.

As shown in Figure 1, the non-invasive bowel motility measurement device (A) through the long sound analysis according to an embodiment of the present invention, in the non-invasive bowel motility measurement device through the long sound analysis, the sound generated in the intestine A signal collector 100 for measuring and outputting a long sound signal; A noise removing unit 200 which removes and outputs noise from the long sound signal output from the signal collecting unit 100; A variable extracting unit (300) for extracting feature variables from the noise-removed long sound signal; A relational expression deriving unit 400 for deriving a relational expression between the extracted feature variables and a previously measured colon transit time (CTT); And an intestinal motility diagnosis unit 500 for diagnosing the exercise state of the intestine based on the derived relational expression.

And, as shown in Figure 4, the non-invasive bowel motility measurement method through the long sound analysis according to an embodiment of the present invention, the step of collecting and measuring the long sound signal which is a sound generated in the intestine (S100); Removing invariant signals such as noise, especially heart sounds or respiratory sounds, from the collected and measured long sound (S200); Extracting feature variables from the noise-free long sound signal (S300); Deriving a relationship between the extracted feature variable and colon transit time (CTT) through regression analysis (S400); A method of diagnosing intestinal motility is performed by converting the extracted feature variable into an intestinal motility index, for example, a large intestine transit time, using the derived relational expression (S500).

The signal collection unit 100 collects and measures a long sound signal through at least one or more electronic stethoscopes or microphones (S100).

In this case, the signal collection unit 100 in a state in which the subject maintains an empty stomach for at least 8 hours or more, after ingesting a predetermined amount of food, for 10 minutes every 10 hours after 1 hour, 4 hours, and 8 hours Measure

Here, the signal collector 100 may measure the long sound at a specific position of the abdomen as shown in FIG. 2 using three electronic stethoscopes (or microphones).

As shown in FIG. 2, the signal collection unit 100 includes an ascending colon (CH1), a descending colon (CH2), and a sigmoid colon (CH3) located in the abdomen of the subject. It is desirable to measure the long sound signal in place.

The signal collector 100 includes the measured long sound signal analog / digital converter 110, and converts the measured long sound signal, which is the measured analog signal, into a digital signal.

In this case, the analog / digital converter 110 preferably converts the digital signal to a sampling frequency of 8 KHz in consideration of the frequency band of the long sound signal.

The signal collection unit 100 includes a fourth Butterworth bandpass filter 120 to remove or minimize signal fluctuation noise caused by breathing noise or unnecessary movement caused by the subject's breathing. Here, the frequency band of the fourth-order Butterworth bandpass filter 120 is preferably 5 ~ 600 Hz.

On the other hand, the sampling frequency, the type of filter, the order and the frequency band of the signal collector 100 can be adjusted according to the characteristics of the device used when measuring the long sound.

The noise removing unit 200 removes noise of the long sound signal output from the signal collecting unit 100 by using an kurtosis-based noise detection method (IKD) (S200).

That is, the noise removing unit 200 calculates the kurtosis vector of the long sound signal, removes the time-invariant signal by removing a signal in the time domain corresponding to a point below the threshold, using the standard deviation of the vector as a critical point. This removes the noise from the long sound signal.

In this case, the IKD algorithm is for selectively extracting only non-stationary signals from the entire signal, and is calculated by calculating the kurtosis of the signals. The kurtosis of a time-varying signal has a larger characteristic than the kurtosis of a stationary signal. Since the long sound signal corresponds to a time-varying signal, this algorithm can effectively remove the respiratory sound and the heart sound, which are a kind of time invariant signal mixed with the long sound.

The noise removal using the IKD algorithm will now be described in detail with reference to FIG. 5.

As shown in FIG. 5, the noise removing unit 200 includes removing the noise by calculating a kurtosis vector of the long sound signal (S210); Calculating a standard deviation (SD) of the calculated kurtosis vector (S220); Removing only a long sound signal of a region in which a value less than the standard deviation (SD) exists among the calculated kurtosis vectors (S230); Calculating a sum of squares of noise which is the removed long sound signal (S240); And repeating the above step until the calculated sum of squares is smaller than a preset reference value (S250).

First, the noise removing unit 200 calculates the kurtosis (K = {K _j }) of the current long sound signal using Equation 1 below (S210). Here, the kurtosis (K = {K _j }) is a vector composed of kurtosis (K _j ) measured in a j-th sliding window having a size M. m _j and σ _j are mean and standard deviation, respectively, measured in long sound.

Equation 1

The noise removing unit 200 calculates the standard deviation of the derived K in a second step (S220), so that a signal within the j th sliding window region in which a value less than the calculated standard deviation exists among the derived K is present. Only remove (S230).

In operation S240, the noise removing unit 200 calculates a sum of squares of the total noise, which is a signal in a j-th sliding window region in which a value less than the calculated standard deviation is present, in operation S240. If the sum of squares is larger than the preset reference value, it is determined that extra noise is mixed in the long sound signal, and the

steps

1 and 2 are repeated. If the sum is smaller than the preset reference value, the noise is almost eliminated to determine that the noise is almost eliminated. The process of removing ends (S250).

Here, the preset reference value is a value set arbitrarily by the user as a single kurtosis value such as 10 and 20, for example. At this time, the smaller the predetermined reference value, the more noise is removed, but the long sound signal may also be inevitably removed.

The variable extractor 300 extracts a feature variable from the long sound signal from which the noise is removed (S300). The characteristic variable is jitter (J _{ch, t} ) _, which is the period variation rate between pitches of the long sound signal, shimmer (S _{ch, t} ), which is the magnitude variation rate, and trace (T _ch ), which is the change amount of the jitter and shimmer. Equation 2] to [Equation 4].

Equation 2

Equation 3

Equation 4

Pi, Ai, and N are the periods between the pitches of the long sound signals, the magnitudes between the peaks of the pitches, and the total number of pitches, respectively. And ch and t represent channels (CH1, CH2, CH3) and time (after 1, 4, 8 hours) for measuring long sound, respectively.

The Pi, Ai, N are obtained through the long sound signal graph as shown in FIG.

As shown in FIG. 3, the period and amplitude between pitches of the long sound signal according to the present invention can be obtained through a graph.

Feature variable which is obtained by the above method is, for a period of a degree of variability in the collected prolonged sound pitch (jitter), the size change rate (shimmer) and measuring time represents the amount of change (trace) of the two variables, total of 21 (J _{ch, t} 9 gae , S _{ch, t} 9 and 3 T _ch ).

In addition to the 21 variables thus extracted in the present invention, it is natural that other variables (center frequency, average frequency, etc.) obtained through the pitch analysis of the acoustic signal may be used.

The relational expression deriving unit 400 derives a relational expression for quantitatively diagnosing intestinal motility using the feature variable extracted through the variable extracting unit 300 (S400). In this case, the relational expression obtains the colon pass time as an output value using the extracted feature variable as an input value. Here, the relationship is derived through regression analysis between the colon transit time (CTT) and the extracted variables. That is, the CTT of the examinee may be estimated by inputting the measured feature variables in the form of a generalized CTT estimation relation "y = a1x1 + a2x2 + a3x3 + ...". Where y is CTT, a? Is a coefficient, and x? Is a feature variable.

In addition, the intestinal motility diagnosis unit 500 converts a feature variable into an intestinal motility index, for example, a large intestine transit time (CTT), based on the derived relationship to diagnose the intestinal motility. That is, the motility of the intestine of the subject is diagnosed based on the estimated CTT value. For example, if the CTT value is in the normal range (in the paper, it is generally known that the normal range of CTT is about 20 to 40 hours), it is judged as normal intestinal motility, and outside this range, it is diagnosed as fast or delayed motility. .

As described above and described with reference to a preferred embodiment for illustrating the technical idea of the present invention, the present invention is not limited to the configuration and operation as shown and described as described above, it is a deviation from the scope of the technical idea It will be apparent to those skilled in the art that many changes and modifications can be made to the invention without departing from the scope of the invention. Accordingly, all such suitable changes, modifications, and equivalents should be considered to be within the scope of the present invention.

The present invention estimates colonic transit time (CTT) using characteristic variables derived by analyzing long-term signals, which are sounds generated in the intestine, and based on this, non-invasive field through long-term analysis that can quantitatively diagnose intestinal motility. TECHNICAL FIELD This invention relates to an apparatus and method for measuring mobility.

Claims

In the non-invasive intestinal motility measurement device through the analysis of long sound,

A signal collector which measures and outputs a long sound signal, which is a sound generated in the intestine;

A noise removing unit for removing noise from the long sound signal output from the signal collecting unit and outputting the noise;

A variable extraction unit for extracting feature variables from the noise-removed long sound signal;

A relation expression derivation unit for deriving a relation between the extracted feature variables and a previously measured colon transit time (CTT); And

Non-invasive intestinal motility measurement device through the sound analysis, including a bowel motility diagnostic unit for diagnosing the motion state of the intestine based on the derived relational expression.
The method of claim 1,

The signal collection unit, non-invasive bowel motility through the long sound analysis, characterized in that to measure the long bowel signal in the ascending colon (descending colon) and sigmoid colon (sigmoid colon) located in the abdomen Measuring device.
The method of claim 1,

The signal collector, the non-invasive field motility measurement device through the long sound analysis, characterized in that it comprises an analog / digital converter for converting the long sound signal from an analog signal to a digital signal.
The method of claim 3, wherein

The analog / digital converter, the non-invasive field motility measurement device through the long sound analysis, characterized in that for converting the long sound signal into a digital signal at a sampling frequency of 8 KHz.
The method of claim 1,

The signal collection unit, the non-invasive field motility measurement device through the long sound analysis, characterized in that it comprises a band pass filter for removing noise due to respiratory noise or unnecessary movement of the subject from the long sound signal.
The method of claim 5,

The band pass filter is a non-invasive field motility measurement device through the long sound analysis, characterized in that the fourth-order Butterworth band pass filter.
The method of claim 1,

The noise removing unit,

The non-invasive analysis of the long sound signal is performed by calculating a kurtosis vector of the long sound signal, and removing time-varying signals by removing a signal in a time domain corresponding to a point below the threshold, using the standard deviation of the vector as a critical point. Intestinal motility measurement device.
The method of claim 1,

The variable extraction unit,

The apparatus for measuring non-invasive field motility through long sound analysis, characterized by extracting the jitter, which is the period variation rate between pitches of the noise-removed long sound signal, the shimmer which is the magnitude variation rate, and the trace, which is a change amount of the jitter and shimmer, as feature variables.
The method of claim 8,

The variable extracting unit, after ingesting the food in a fasting state, non-invasive through the analysis of long-term sound, characterized in that for extracting the feature variable based on the sound generated in the intestine measured after 1 hour, 4 hours, 8 hours respectively Intestinal motility measurement device.
The method of claim 1,

The relational expression derivation unit,

Non-invasive bowel motility measurement device through the long sound analysis, characterized in that to derive a relationship through the regression analysis between the extracted feature variable and colon transit time.
The method of claim 1,

The bowel motility diagnostic unit,

Non-invasive bowel motility measurement device through the sound analysis, characterized in that for converting the feature variable to the intestinal motility index based on the derived relational expression.
In the method of diagnosing non-invasive bowel motility through the analysis of bowel sound,

Measuring a long sound signal, which is a sound generated in the intestine;

Removing noise from the measured long sound signal;

Extracting feature variables from the noise-free long sound signal;

Deriving a relation between the extracted feature variable and the colon transit time through a regression analysis;

Non-invasive bowel motility measurement device through the analysis of the bowel sound comprising the step of diagnosing bowel motility using the derived equation.
The method of claim 12,

Removing the noise,

Calculating a kurtosis vector of the long sound signal;

Calculating a standard deviation of the calculated kurtosis vector;

Removing only a long sound signal of a region having a value less than the standard deviation among the calculated kurtosis vectors;

Calculating a sum of squares of noise that is the removed long sound signal; And

And repeating the above steps until the calculated sum of squares is smaller than a predetermined reference value.
The method of claim 12,

And converting the measured sound into a digital signal and passing a band pass filter.
The method of claim 12,

The noise is a non-invasive bowel motility measurement method through the analysis of the long sound, characterized in that in the constant signal, such as the heart sound or breathing sound.
In the non-invasive intestinal motility measurement device through the analysis of long sound,

A signal collector which measures and outputs a long sound signal, which is a sound generated in the intestine;

A variable extraction unit for extracting feature variables from the long sound signal;

A relation expression derivation unit for deriving a relation between the extracted feature variables and a previously measured colon transit time (CTT); And

Non-invasive intestinal motility measurement device through the sound analysis, including a bowel motility diagnostic unit for diagnosing the motion state of the intestine based on the derived relational expression.
A long sound characterized in that a time-invariant signal is removed by calculating a kurtosis vector of a long sound signal, which is a sound generated in the field, and removing a signal in a time domain corresponding to a point below the threshold, using the standard deviation of the vector as a critical point. A device that removes noise from a signal.