JP7774265B2 - Evaluation device, evaluation system, evaluation method, and evaluation program for evaluating intestinal condition - Google Patents

Description

This disclosure relates to an evaluation device, evaluation system, evaluation method, and evaluation program for evaluating intestinal conditions.

Technologies for evaluating intestinal condition based on bowel sounds have been proposed (for example, Patent Document 1). Specifically, Patent Document 1 discloses a method for calculating the amount of gas in the intestines based on time-domain features such as the duration and frequency of bowel sounds.

Japanese Patent Application Laid-Open No. 2021-74238

However, even if the condition of the intestines is similar, there are individual differences in the duration and frequency of bowel sounds. For example, among subjects with healthy bowel conditions, there are subjects who have a relatively high frequency of bowel sounds and subjects who have a relatively low frequency of bowel sounds. As a result, the conventional technology described in Patent Document 1 had the problem of being unable to accurately evaluate the condition of the intestines.

The objective of this disclosure is to accurately evaluate the condition of the intestines.

In order to solve the above problems, the present disclosure includes the following aspects.
Item 1.
An evaluation device for evaluating the intestinal condition of a subject,
an extraction unit that extracts bowel sounds from acoustic data obtained from the subject;
a first feature amount calculation unit that calculates a first feature amount, which is a feature amount of the bowel sound before the bowel is stimulated, from among the extracted bowel sounds;
a second feature amount calculation unit that calculates a second feature amount, which is a feature amount of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation unit that evaluates an intestinal condition of the subject based on the first feature amount and the second feature amount;
An evaluation device comprising:
Item 2.
Item 2. The evaluation device according to Item 1, wherein the feature amount is a feature amount related to a time domain of the bowel sound.
Item 3.
Item 3. The evaluation device according to Item 2, wherein the feature amount is the frequency of occurrence of the bowel sounds.
Item 4.
Item 3. The evaluation device according to Item 2, wherein the feature amount is the duration of the bowel sound.
Item 5.
Item 3. The evaluation device according to Item 2, wherein the feature amount is a signal level of the bowel sound.
Item 6.
Item 2. The evaluation device according to Item 1, wherein the feature amount is a feature amount related to a frequency domain of the bowel sound.
Section 7.
Item 7. The evaluation device according to Item 6, wherein the feature amount is a spectral bandwidth of the bowel sound.
Section 8.
Item 2. The evaluation device according to Item 1, wherein the evaluation unit evaluates the state of the intestine based on the logarithm of the first feature amount and the logarithm of the second feature amount.
Item 9.
Item 2. The evaluation device according to Item 1, wherein the evaluation unit evaluates the state of the intestine based on the logarithm of the first feature amount and the second feature amount.
Item 10.
Item 2. The evaluation device according to Item 1, wherein the evaluation unit evaluates the state of the intestine based on logarithms of the first feature amount and the second feature amount.
Item 11.
Item 8. The evaluation device according to any one of Items 1 to 7, wherein the evaluation unit evaluates the state of the intestine based on a ratio between the first feature amount and the second feature amount.
Item 12.
Item 12. The evaluation device according to Item 11, wherein the evaluation unit evaluates the state of the intestine based on the first feature amount and the ratio.
Item 13.
Item 13. The evaluation device according to Item 12, wherein the evaluation unit evaluates the state of the intestines based on a distance between a coordinate corresponding to the first feature amount and the ratio and an approximation curve obtained from a distribution of coordinates corresponding to the first feature amount and the ratio of the first feature amount and the second feature amount of bowel sounds extracted from acoustic data obtained from a plurality of healthy subjects.
Section 14.
Item 12. The evaluation device according to Item 11, wherein the evaluation unit evaluates the state of the intestine based on the logarithm of the first feature amount and the logarithm of the ratio.
Item 15.
Item 15. The evaluation device according to Item 14, wherein the evaluation unit evaluates the state of the intestine based on a distance between coordinates corresponding to the logarithm of the first feature amount and the logarithm of the ratio and a regression line obtained from a distribution of coordinates corresponding to the logarithm of the first feature amount and the logarithm of the ratio between the first feature amount and the second feature amount of bowel sounds extracted from acoustic data obtained from a plurality of healthy subjects.
Section 16.
Item 12. The evaluation device according to Item 11, wherein the evaluation unit evaluates the state of the intestine based on the logarithm of the first feature amount and the ratio.
Section 17.
Item 12. The evaluation device according to Item 11, wherein the evaluation unit evaluates the state of the intestine based on the first feature amount and the logarithm of the ratio.
Section 18.
Item 8. The evaluation device according to any one of Items 1 to 7, wherein the evaluation unit evaluates the state of the intestine based on a difference between the first feature amount and the second feature amount.
Item 19.
Item 19. The evaluation device according to Item 18, wherein the evaluation unit evaluates the state of the intestine based on the first feature amount and the difference.
Section 20.
Item 8. The evaluation device according to any one of Items 1 to 7, wherein the evaluation unit evaluates the state of the intestine based on the sum of the first feature amount and the second feature amount.
Section 21.
Item 21. The evaluation device according to Item 20, wherein the evaluation unit evaluates the state of the intestine based on the first feature amount and the sum.
Section 22.
Item 8. The evaluation device according to any one of Items 1 to 7, wherein the evaluation unit evaluates the state of the intestine based on a product of the first feature amount and the second feature amount.
Section 23.
Item 23. The evaluation device according to Item 22, wherein the evaluation unit evaluates the state of the intestine based on the first feature amount and the product.
Section 24.
Item 24. The evaluation device according to any one of Items 1 to 23, wherein the evaluation unit evaluates the state of the intestines based on a feature of a bowel sound having a signal level equal to or higher than a threshold, out of the first feature and the second feature.
Section 25.
Item 25. The evaluation device according to any one of Items 1 to 24, wherein the lengths of the bowel sounds corresponding to the first feature amount and the second feature amount are 5 to 10 minutes.
Section 26.
An evaluation device according to any one of items 1 to 25,
a sound collection device for obtaining the acoustic data from the subject;
An evaluation system comprising:
Section 27.
1. A method for evaluating the intestinal condition of a subject, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is a feature of the bowel sound before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is a feature of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
An evaluation method having the following characteristics.
Section 28.
1. An assessment program for assessing a subject's intestinal status, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is a feature of the bowel sound before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is a feature of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
An evaluation program that causes a computer to execute the above.

This disclosure makes it possible to accurately evaluate the condition of the intestines.

1 is a block diagram illustrating a configuration of an evaluation system according to an embodiment of the present disclosure. 1 is a flowchart illustrating a processing procedure of an evaluation method according to an embodiment of the present disclosure. 10 is a flowchart showing more specific processing steps of the step of collecting acoustic data. 1 is an example of a waveform of acoustic data obtained from a subject. 10A is a scatter plot of the first feature and the ratio obtained from the DB of the carbonated water intake test, and FIG. 10B is a scatter plot of the logarithm of the first feature and the logarithm of the ratio. FIG. 5(b) is a scatter plot of the logarithm of the first feature amount calculated from the feature amount of bowel sounds whose signal levels are equal to or greater than a threshold, and the logarithm of the ratio. FIG. 10 is a scatter plot of first feature amounts and differences obtained from a DB of a carbonated water intake test. 10A is a scatter plot of the first feature and the relative change obtained from the DB of the carbonated water intake test, and FIG. 10B is a scatter plot of the logarithm of the first feature and the logarithm of the relative change. FIG. 10 is a scatter plot of first feature values and sums obtained from a DB of carbonated water intake tests. FIG. 10 is a scatter plot of first feature values and products obtained from the DB of the carbonated water intake test. 10A is a scatter plot of the first feature and the ratio obtained from the DB of the carbonated water intake test, and FIG. 10B is a scatter plot of the logarithm of the first feature and the logarithm of the ratio. (a) is a scatter plot of the logarithm of the first feature value and the logarithm of the ratio of bowel sounds obtained from subjects with healthy bowel conditions, and (b) is a scatter plot plotting the logarithm of the first feature value and the logarithm of the ratio of bowel sounds obtained from subjects with unhealthy bowel conditions.

Embodiments of the present disclosure will be described below with reference to the accompanying drawings. Note that the present disclosure is not limited to the following embodiments, and various modifications are possible without departing from the spirit of the disclosure.

(System configuration)
1 is a block diagram showing the configuration of an evaluation system 1 according to an embodiment of the present disclosure. The evaluation system 1 includes a sound collection device 2 and an evaluation device 3.

The sound collection device 2 is a device for obtaining acoustic data from the subject. When collecting acoustic data at a medical institution, the sound collection device 2 is, for example, a stethoscope, and when collecting acoustic data outside a medical institution, the sound collection device 2 is, for example, a microphone. When the sound collection device 2 is a microphone, the sound collection device 2 is attached to the subject's abdomen directly or via clothing, and obtains acoustic data including the subject's bowel sounds.

The sound collection device 2 is connected to the evaluation device 3 via a wired or wireless connection and transfers the acquired acoustic data to the evaluation device 3. Note that the acoustic data may also be transferred from the sound collection device 2 to the evaluation device 3 via a recording medium or communication device.

The evaluation device 3 is a device that evaluates the intestinal condition of a subject. The evaluation device 3 can be configured as a general-purpose computer or a portable computer such as a smartphone or tablet terminal. The hardware configuration of the evaluation device 3 includes a processor such as a CPU or GPU, a main memory device such as DRAM or SRAM (not shown), and an auxiliary memory device 30 such as an HDD or SSD. The auxiliary memory device 30 stores the evaluation program P, extraction model M, correlation data R, etc.

The auxiliary storage device 30 may be external to the evaluation device 3. The evaluation device 3 may also be provided on the cloud.

The evaluation device 3 includes functional blocks: an acoustic data acquisition unit 31, an extraction unit 32, a feature calculation unit 33, and an evaluation unit 34. Furthermore, the feature calculation unit 33 includes a first feature calculation unit 331 and a second feature calculation unit 332. These functional blocks are realized by the processor of the evaluation device 3 reading and executing the evaluation program P into the main storage device. The evaluation program P may be downloaded to the evaluation device 3 via a communications network such as the Internet, or may be recorded on a computer-readable, non-transitory recording medium such as a CD-ROM or SD card and installed on the evaluation device 3 via that storage medium. The functions of each functional block will be described below.

(Evaluation method processing procedure)
The functions of the evaluation system 1 will be described with reference to Fig. 2. Fig. 2 is a flowchart showing the processing procedure of the evaluation method according to this embodiment. The evaluation method has steps S1 to S5, with step S1 being executed by the sound collection device 2 and steps S2 to S5 being executed by the above-mentioned respective functional blocks of the evaluation device 3. That is, the evaluation program P causes the evaluation device 3 to execute steps S2 to S5.

In step S1, the sound collection device 2 collects acoustic data including the subject's bowel sounds. In this embodiment, a stimulus is applied to the subject's intestines, and the sound collection device 2 collects acoustic data before and after the stimulus is applied to the intestines.

Figure 3 is a flowchart showing more specific processing steps of step S1.

In step S11, sound collection by the sound collection device 2 begins. After a predetermined time (e.g., 5 minutes) has elapsed (YES in step S12), in step S13, the subject is prompted to consume a beverage to stimulate the subject's intestines.

The beverage is not particularly limited as long as it can stimulate the intestines, but examples include carbonated water and coffee. Furthermore, methods of stimulating the intestines are not limited to ingesting beverages. Examples of methods of stimulating the intestines include ingesting solid foods such as laxatives, abdominal exercises that stimulate the intestines such as twisting the abdomen, abdominal pressure, stimulating acupressure points related to the intestines, waking up and going to bed, etc.

After a predetermined time (e.g., 10 minutes) has elapsed since the intestines were stimulated (YES in step S14), sound collection by the sound collection device 2 is terminated in step S15.

As described above, the sound collection device 2 collects acoustic data before and after stimulating the subject's intestines.

In step S2 shown in FIG. 2, the acoustic data acquisition unit 31 of the evaluation device 3 acquires acoustic data obtained from the subject by the sound collection device 2.

In step S3 (extraction step), the extraction unit 32 extracts bowel sounds from the acoustic data acquired in step S2. The acoustic data contains noise such as breathing sounds and body movement sounds in addition to bowel sounds, but bowel sounds can be extracted using known techniques. In this embodiment, the extraction unit 32 extracts bowel sounds using the technique described in WO 2019/216320.

Specifically, the extraction unit 32 detects multiple segments from the acoustic data, extracts PNCCs (power normalized cepstral coefficients), which are frequency-related features, from each segment, and inputs the extracted PNCCs into the extraction model M. The extraction model M is a neural network model that has undergone machine learning to learn the relationship between bowel sounds in the acoustic data and PNCCs, and outputs a prediction score indicating the likelihood that each segment contains a bowel sound based on the input PNCCs. The extraction unit 32 extracts segments with prediction scores greater than a predetermined threshold as bowel sounds. A bowel sound is a continuous sound, and its length (duration) is indefinite.

Figure 4 shows an example of the waveform of acoustic data obtained from a subject. The dashed frames represent segments, and the extraction unit 32 extracts segments that are likely to contain bowel sounds.

Note that the frequency of normal bowel sounds is mainly between approximately 100 and 500 Hz. Therefore, it is preferable that the extraction unit 32 apply a bandpass filter to the acoustic data acquired in step S2 that passes a frequency band including 100 to 500 Hz (e.g., 80 to 1000 Hz) before extracting the bowel sounds. It is also preferable that the extraction unit 32 extract bowel sounds whose signal level is equal to or greater than a threshold.

In step S4 (first feature calculation step, second feature calculation step) shown in FIG. 2, the feature calculation unit 33 calculates the feature amounts of the bowel sounds extracted in step S3. Specifically, the first feature calculation unit 331 of the feature calculation unit 33 calculates a first feature amount, which is a feature amount of the extracted bowel sounds before the intestines are stimulated, and the second feature calculation unit 332 of the feature calculation unit 33 calculates a second feature amount, which is a feature amount of the extracted bowel sounds after the intestines are stimulated. The duration of each bowel sound corresponding to the first feature amount and the second feature amount (the measurement time of the bowel sounds before and after the stimulation) is not particularly limited, but is preferably 5 to 10 minutes, for example.

In this embodiment, the feature quantities are preferably those related to the time domain of bowel sounds, and more specifically, the frequency of bowel sounds, the duration of bowel sounds, or the signal level of bowel sounds. Other features related to the time domain include, for example, the maximum amplitude of bowel sounds (BS Max. amplitude), bowel sound power (BS power), maximum amplitude spectral density of bowel sounds (Maximum Amplitude Spectral Density), zero crossing rate of bowel sounds, kurtosis of bowel sounds, and the Ratio of largest absolute to root mean squared value of bowel sounds.

The features are not particularly limited as long as they are acoustic features of bowel sounds, and may be acoustic features such as features related to the frequency domain. Examples of features related to the frequency domain include the spectral bandwidth of bowel sounds (the width from the peak to an arbitrary attenuation point), the peak frequency of bowel sounds, the spectral centroid of bowel sounds, the spectral flatness of bowel sounds, the median frequency of bowel sounds, the mean frequency of bowel sounds, the first formant of bowel sounds, the second formant of bowel sounds, the spectral entropy of bowel sounds, the spectral spread of bowel sounds, the spectral rolloff of bowel sounds, the spectral slope of bowel sounds, the spectral kurtosis of bowel sounds, and the spectral skewness of bowel sounds. Additionally, the number of segments per unit time that are likely to contain bowel sounds (BS segments) or the number of non-BS segments per unit time (also known as Bowel-Sound Duration or Silence Duration) may also be used as features.

In step S5 (evaluation step), the evaluation unit 34 evaluates the intestinal condition of the subject based on the first feature amount and the second feature amount. In this embodiment, the evaluation unit 34 evaluates the intestinal condition based on the ratio or difference between the first feature amount and the second feature amount. At this time, the logarithm may be taken for at least one of the first feature amount and the second feature amount (ratio or difference).

As will be described in the examples below, it has been discovered that when the intestinal condition is healthy, there is a correlation between the first feature amount and the second feature amount. The auxiliary storage device 30 of the evaluation device 3 stores correlation data R relating to the correlation between the first feature amount and the second feature amount obtained from subjects with healthy intestinal conditions through prior verification experiments, etc.

In contrast, when the intestinal condition is poor, the response motility to intestinal stimuli is different compared to when the intestinal condition is healthy, and the relationship between the first feature amount and the second feature amount will deviate from the correlation between the first feature amount and the second feature amount when the intestinal condition is healthy. Therefore, by comparing the first feature amount and the second feature amount calculated in step S4 with the correlation data R, the evaluation unit 34 can accurately evaluate the intestinal condition of the subject.

Furthermore, in the embodiment,
It has been found that the following correlations are particularly high: a correlation between the first feature amount and the ratio of the first feature amount to the second feature amount; a correlation between the logarithm of the first feature amount and the logarithm of the ratio; a correlation between the first feature amount and the difference between the first feature amount and the second feature amount; a correlation between the first feature amount and the sum of the first feature amount and the second feature amount; and a correlation between the first feature amount and the product of the first feature amount and the second feature amount.
It is preferable to evaluate the intestinal condition based on the first feature amount and the ratio, the logarithm of the first feature amount and the logarithm of the ratio, the first feature amount and the difference, the first feature amount and the sum, and the first feature amount and the product. This allows the intestinal condition to be evaluated with higher accuracy.

Examples of the present disclosure are described below, but the present disclosure is not limited to the examples below.

[Examples 1 to 4]
In Examples 1 to 4, a beverage ingestion test was conducted on subjects with healthy intestines to verify whether there was a correlation between the features of bowel sounds before and after intestinal stimulation. The subjects were 20 people (12 men, 8 women, age: 32.90±7.89 years, height: 167.28±8.52 cm, weight: 59.04±8.62 kg, BMI: 21.02±1.96) who had been diagnosed with healthy intestines according to the Rome III diagnostic criteria. Each subject underwent the beverage ingestion test twice.

Specifically, in the first test, each subject, who had fasted for approximately 12 hours, ingested carbonated water at 10°C or below, and acoustic data was collected from each subject's body sounds while they were in a resting state for 5 minutes before and 10 minutes after ingestion. The acoustic data collected from each subject in the first test is designated database (DB) 1.

The second test was conducted on a different day from the first test. In the second test, each subject drank coffee at approximately 45°C, and acoustic data was collected from each subject's biological sounds while they were in a resting state for 5 minutes before and 10 minutes after drank the coffee. The acoustic data collected for each subject in the second test is referred to as database (DB) 2.

The equipment used to collect acoustic data in these tests was an electronic stethoscope (E-scope2, Cardionics Inc., Houston, TX, USA) and a multi-track recorder (R16 Zoom Co., Ltd., Tokyo, Japan). During acoustic data collection, the subject was placed in a supine position, and the electronic stethoscope was fixed in a cross shape using masking tape, 9 cm to the right of the navel. The electronic stethoscope has two sound collection modes: heart and respiratory. The respiratory mode was selected, which allows recording over a wider frequency range. The sampling frequency of the acoustic data during collection was 44,100 Hz, and the digital resolution was 16 bits/sample. Taking into account the frequency characteristics of the electronic stethoscope and bowel sounds, the frequency of the collected acoustic data was downsampled to 4,000 Hz.

Evaluation device 3 used a trained neural network model as extraction model M to extract multiple bowel sounds from each acoustic data set. Specifically, the acoustic data was segmented with a segment length of 64 ms and a shift size of 16 ms, and PNCC and LPC coefficients, which are frequency-related features, were extracted from each segment. The extracted PNCC and LPC coefficients were then input into the neural network model, and a determination was made based on the output of the neural network as to whether each segment contained bowel sounds. When consecutive segments containing bowel sounds (BS segments) were found, the consecutive BS segments were considered to be a single BS episode. The performance of extraction model M was as follows: sensitivity: 90.58 ± 4.46, specificity: 91.76 ± 4.61, PPV: 66.92 ± 8.47, NPV: 98.45 ± 0.55, accuracy: 92.07 ± 3.34, and F1 score: 76.60 ± 5.89.

Next, a first feature quantity, which is a feature quantity before ingestion of a beverage, and a second feature quantity, which is a feature quantity after ingestion of a beverage, were calculated from the extracted bowel sounds.
Four feature quantities were used: frequency of bowel sounds, duration of bowel sounds, signal level of bowel sounds, and spectral bandwidth of bowel sounds. Below, the correlation between the first feature quantity and the second feature quantity was verified for each feature quantity.

Example 1
In Example 1, the correlation between the first feature amount and the second feature amount was verified when the feature amount was the frequency of bowel sounds. The first feature amount was defined as the average number of bowel sounds per minute ( _xb ) in the five minutes immediately before ingestion of a beverage, and the second feature amount was defined as the average number of bowel sounds per minute ( _xa ) in the ten minutes immediately after ingestion of a beverage. The first feature amount _xb and the second feature amount _xa were calculated for each subject.

First, the correlation between the ratio of the first feature amount _xb to the second feature amount _xa and the first feature amount _xb was verified. The ratio was defined as in Equation (1).
ratio=x _a /x _b ...(1)

FIG. 5(a) is a scatter plot of the first feature _xb and the ratio obtained from DB1 of the carbonated water intake test. From FIG. 5(a), it was confirmed that the larger the first feature _xb , the lower the ratio tends to be. The correlation coefficient for each point in FIG. 5(a) was R = -0.642, and the significance level was p = 0.002. The significance of the correlation coefficient was evaluated using a Student's t-test, and a correlation was considered to exist when the significance level p was p < 0.05.

As described above, it was found that there is a high nonlinear correlation between the first feature amount _xb and the ratio ratio when a subject with a healthy intestinal condition ingests carbonated water. In other words, it was found that if the intestinal condition is healthy, the coordinates of the first feature amount _xb and the ratio ratio tend to concentrate in a predetermined region.

In contrast, when the intestinal condition is poor, the response motility to stimuli is different from that when the intestinal condition is healthy, and the coordinates of the first feature amount _xb and the ratio "ratio" fall outside the predetermined range. Therefore, for example, the intestinal condition of an unknown subject can be evaluated by obtaining an approximate curve for each point shown in FIG. 5A and calculating the distance between the approximate curve and the coordinates of the first feature amount _xb and the ratio "ratio" calculated in the same manner as above from bowel sounds obtained from an unknown subject who has ingested carbonated water. That is, the evaluation unit 34 can evaluate the intestinal condition based on the distance between the coordinates corresponding to the first feature amount _xb and the ratio "ratio _" of the first feature amount _xb and the second feature amount xa of bowel sounds extracted from acoustic data obtained from the subject and the approximate curve obtained from the distribution of coordinates corresponding to the first feature amount _xb and the ratio "ratio" of the first feature amount _{xb and} the second feature amount xa of bowel sounds extracted from acoustic data obtained from a plurality of healthy subjects.

Furthermore, the logarithm of the first feature quantity _xb and the logarithm of the ratio (ratio) were calculated for each subject. FIG. 5B is a scatter plot of the logarithm of the first feature quantity _xb and the logarithm of the ratio (ratio) obtained from DB1. The correlation coefficient for each point was R = -0.943, and the significance level was p = 0.000. Therefore, it was found that there is a very high linear correlation between the logarithm of the first feature quantity _xb and the logarithm of the ratio (ratio) when subjects with healthy intestinal conditions ingested carbonated water.

5(b) , and calculating the distance between the regression line and the coordinates of the first feature quantity _xb and the ratio (ratio) calculated in the same manner as above from bowel sounds obtained from an unknown subject who ingested carbonated water, the intestinal condition of the unknown subject can be evaluated with higher accuracy. That is, the evaluation unit 34 can evaluate the intestinal condition based on the distance between the coordinates corresponding to the logarithm of the first feature quantity _xb of bowel sounds extracted from acoustic data obtained from the subject and the logarithm of the ratio (ratio) of the first feature quantity _xb and the second feature quantity _xa and the regression line calculated from the distribution of coordinates corresponding to the logarithm of the first feature quantity _xb of bowel sounds extracted from acoustic data obtained from a plurality of healthy subjects and the logarithm of the ratio (ratio) of the first feature quantity _xb and the second feature quantity _xa .

In this embodiment, the logarithm is a natural logarithm, but the base of the logarithm is not particularly limited. In this embodiment, logarithms are taken for both the first feature quantity _xb and the ratio ratio, but logarithms may be taken for only one of them. That is, the evaluation unit 34 may evaluate the intestinal condition based on the logarithm of the first feature quantity _xb and the ratio ratio, or based on the logarithm of the first feature quantity _xb and the ratio ratio. Furthermore, instead of taking the logarithm of at least one of the first feature quantity _xb and the ratio ratio, Yeo-Johnson transformation, Box-Cox transformation, arcsin transformation, Hilbert transform, square root transformation, dynamic range compression, exponential scaling, or custom transformations thereof may be performed.

The correlation coefficient between the logarithm of the first feature amount x _b and the logarithm of the ratio obtained from DB2 (coffee intake test) was R=-0.768, and the significance level was p=0.000.

Furthermore, from the first feature amount _xb and the second feature amount _xa , the feature amount of the bowel sound whose signal level was equal to or greater than the threshold value _Ath was extracted, and the ratio ratio was calculated in the same manner as above. In this example, the threshold value _Ath = 5 × ^10-4 .

6 is a scatter plot of the logarithm of the first feature quantity _xb and the logarithm of the ratio calculated from the feature quantities of bowel sounds whose signal levels are equal _{to or} greater than the threshold Ath in FIG. 5(b). The correlation coefficient for each point was R = -0.952, and the significance level was p = 0.000. Therefore, it was found that the correlation between the _first feature quantity _xb and the second feature quantity _xa was further increased by limiting the feature quantities to those whose signal levels are equal to or greater than the threshold Ath.

Next, the correlation between the difference (x _a −x _b ) between the first feature amount and the second feature amount and the first feature amount was verified.

7 is a scatter plot of the absolute values of the first feature _xb and the difference ( _xa - _xb ) obtained from DB1. The correlation coefficient for each point was R = -0.637, and the significance level was p = 0.003. Therefore, it was found that there is a high negative correlation between the first feature _xb and the difference ( _xa - _xb ) when a subject with healthy intestinal conditions ingested carbonated water.

Furthermore, the relative change in the first feature amount and the second feature amount (( _xa - _xb )/ _xb ) was calculated for each subject. Figure 8(a) is a scatter plot of the first feature amount _xb and the relative change amount obtained from DB1. The correlation coefficient for each point was R = -0.642, and the significance level was p = 0.002. Therefore, it was found that there is a high nonlinear correlation between the first feature amount _xb and the relative change amount when subjects with healthy intestinal conditions ingested carbonated water.

Furthermore, the logarithm of the relative change was calculated for each subject. Figure 8(b) is a scatter plot of the logarithm of the first feature _xb and the logarithm of the relative change obtained from DB1. The correlation coefficient for each point was R = -0.957, and the significance level was p = 0.000. Therefore, it was found that there is a very high linear correlation between the logarithm of the first feature _xb and the logarithm of the relative change when subjects with healthy intestinal conditions ingested carbonated water.

Next, the correlation between the sum of the first feature amount and the second feature amount (x _a +x _b ) and the first feature amount was verified.

9 is a scatter plot of the first feature _xb and the sum ( _xa + _xb ) obtained from DB1. The correlation coefficient for each point was R = 0.960, and the significance level was p = 0.000. Therefore, it was found that there is a high positive correlation between the first feature _xb and the sum ( _xa + _xb ) when a subject with healthy intestinal conditions ingested carbonated water.

Next, the correlation between the product (x _a ×x _b ) of the first feature amount and the second feature amount and the first feature amount was verified.

10 is a scatter plot of the first feature _xb and the product ( _xa × _xb ) obtained from DB1. The correlation coefficient for each point was R = 0.984, and the significance level was p = 0.000. Therefore, it was found that there is a high positive correlation between the first feature _xb and the product ( _xa × _xb ) when a subject with healthy intestinal conditions ingested carbonated water.

Example 2
In Example 2, the correlation between the first feature and the second feature was verified when the feature was the duration of bowel sounds. The first feature was defined as the average duration (t _b ) of each bowel sound occurring within 5 minutes immediately before ingestion of a beverage, and the second feature was defined as the average duration (t _a ) of each bowel sound occurring within 10 minutes immediately after ingestion of a beverage. The first feature t _b and the second feature t _a were calculated for each subject.

Similar to Example 1, the correlation between the ratio of the first feature amount t _b to the second feature amount t _a and the first feature amount t _b was verified. The ratio ratio was defined as in Equation (2).
ratio=t _a /t _b ...(2)

Furthermore, the logarithm of the first feature quantity _tb and the logarithm of the ratio (ratio) were calculated for each subject. As a result, the correlation coefficient between the logarithm of the first feature quantity _xb obtained from DB1 (carbonated water intake test) and the logarithm of the ratio (ratio) was R = -0.707, with a significance level of p = 0.000. Furthermore, the correlation coefficient between the logarithm of the first feature quantity _xb obtained from DB2 (coffee intake test) and the logarithm of the ratio (ratio) was R = -0.537, with a significance level of p = 0.015.

As described above, it was found that there is a high correlation between the logarithm of the first feature amount _tb and the logarithm of the ratio ratio when subjects with healthy intestinal conditions ingest carbonated water or coffee.

Example 3
In Example 3, the correlation between the first feature and the second feature was verified when the feature was the signal level of the bowel sounds. The first feature was defined as the average signal level (SNR _b ) of each bowel sound generated in the 5 minutes immediately before ingestion of a beverage, and the second feature was defined as the average signal level (SNR _a ) of each bowel sound generated in the 10 minutes immediately after ingestion of a beverage. The first feature SNR _b and the second feature SNR _a were calculated for each subject.

Similar to Examples 1 and 2, the correlation between the ratio of the first feature amount SNR _b to the second feature amount SNR _a and the first feature amount SNR _b was verified. The ratio was defined as in Equation (3).
ratio=SNR _a /SNR _b ...(3)

Furthermore, the logarithm of the first feature SNR _b and the logarithm of the ratio ratio were calculated for each subject. As a result, the correlation coefficient between the logarithm of the first feature SNR _b and the logarithm of the ratio ratio obtained from DB1 (carbonated water intake test) was R = -0.538, with a significance level of p = 0.014. Furthermore, the correlation coefficient between the logarithm of the first feature SNR _b and the logarithm of the ratio ratio obtained from DB2 (coffee intake test) was R = -0.477, with a significance level of p = 0.034.

As described above, it was found that there is a high correlation between the logarithm of the first feature amount SNR _b and the logarithm of the ratio ratio when subjects with healthy intestinal conditions ingest carbonated water or coffee.

Example 4
In Example 4, the correlation between the first feature and the second feature was verified when the feature was the spectral bandwidth of bowel sounds. The spectral bandwidth was defined as the frequency range in which the amplitude spectrum is higher than half the maximum amplitude. The first feature was defined as the average spectral bandwidth per minute ( _wb ) of bowel sounds in the five minutes immediately before ingestion of a beverage, and the second feature was defined as the average spectral bandwidth per minute ( _wa ) of bowel sounds in the ten minutes immediately after ingestion of a beverage. The first feature _wb and the second feature _wa were calculated for each subject.

First, the correlation between the ratio of the first feature amount _wb to the second feature amount _wa and the first feature amount _wb was verified. The ratio was defined as in Equation (4).
ratio= _wa / _wb ...(4)

11A is a scatter plot of the first feature value _wb and the ratio obtained from DB1, where the correlation coefficient for each point was R = -0.599 and the significance level was p = 0.005. Therefore, it was found that there is a high correlation between the first feature value _wb and the ratio when subjects with healthy intestinal conditions ingested carbonated water or coffee.

Furthermore, the logarithm of the first feature quantity _wb and the logarithm of the ratio (ratio) were calculated for each subject. Figure 11(b) is a scatter plot of the logarithm of the first feature quantity _wb and the logarithm of the ratio (ratio) obtained from DB1, where the correlation coefficient for each point was R = -0.708 and the significance level was p = 0.000. Therefore, it was found that there is a high correlation between the logarithm of the first feature quantity _wb and the logarithm of the ratio (ratio) when subjects with healthy intestinal conditions ingested carbonated water or coffee.

[Example 5]
In Example 5, 30 female subjects were given a questionnaire reflecting the Rome IV diagnostic criteria, and based on their responses, the subjects were classified into functional diarrhea (FD), functional constipation (FC), and healthy individuals. The classification results are shown in Table 1.

Next, a beverage ingestion test was conducted on each subject. Specifically, each subject, who had fasted for approximately 12 hours, ingested 200 mL of strongly carbonated water at approximately 10°C or below. Each subject's body sounds were collected as acoustic data while they were in a resting state for 5 minutes before ingestion and 10 minutes after ingestion. The equipment and methods used to collect the acoustic data were the same as those used in Examples 1 to 4 above. Taking into account the frequency characteristics of the electronic stethoscope and the frequency characteristics of bowel sounds, the frequency of the collected acoustic data was downsampled to 4000 Hz.

In the evaluation device 3, multiple bowel sounds were extracted from each acoustic data set using a trained neural network model as the extraction model M. Specifically, the acoustic data was segmented with a segment length of 64 ms and a shift size of 16 ms, and the signals within the segments were normalized to a mean value of 0 and a standard deviation of 1. A total of 52 features, consisting of 10-dimensional linear prediction cepstral (LPC) coefficients, their Δ and ΔΔ, and a 22-dimensional MFCC, were extracted from the normalized segments and used as input to the neural network. Each feature was labeled with a binary value indicating whether it was a bowel sound (BS) segment or not to generate training data. The generated training data was used to train the neural network and generate the extraction model M for two-class classification of whether it was a bowel sound or not. The number of units in the intermediate layer of the neural network was 40. The performance of extraction model M was evaluated using five-fold cross-validation, resulting in a sensitivity of 80.47±2.83%, specificity of 97.61±0.74%, PPV of 86.17±1.88%, NPV of 96.52±0.74%, accuracy of 95.02±0.87%, and F1 score of 83.17±0.80%.

In addition, taking into account changes in acoustic features due to environmental noise, noise subtraction was performed using the average of the spectra obtained from segments determined not to be bowel sounds.

Next, a first feature, which is a feature before ingestion of the beverage, and a second feature, which is a feature after ingestion of the beverage, were calculated from the extracted bowel sounds. More specifically, the first feature was defined as the average ( _xb ) of the first feature calculated from the bowel sounds in the five minutes immediately before ingestion of the beverage, and the second feature was defined as the average ( _xa ) of the second feature calculated from the bowel sounds in the ten minutes immediately after ingestion of the beverage, and the first feature _xb and the second feature _xa were calculated for each subject. The features were as follows:
・Bow sound maximum amplitude (BS maximum amplitude)
・Bow sound power (BS power)
- Frequency of bowel sounds - Maximum Amplitude Spectral Density (MASD) of bowel sounds
・Peak frequency of bowel sounds ・Spectral bandwidth of bowel sounds
The maximum amplitude spectral density is the maximum value of the amplitude spectrum obtained by performing a fast Fourier transform (FFT) on the time signal.

Furthermore, a regression line was calculated from the distribution of coordinates corresponding to the logarithm of the first feature quantity _xb of bowel sounds obtained from healthy subjects and the logarithm of the ratio (ratio) between the first feature quantity _xb and the second feature quantity _xa , and it was verified whether the intestinal condition of each subject could be evaluated based on the distance between the regression line and the coordinates corresponding to the logarithm of the first feature quantity _xb of bowel sounds obtained from each subject and the logarithm of the ratio (ratio).

The following describes a case where the feature is the spectral bandwidth of bowel sounds. First, bowel sounds obtained from 30 subjects were analyzed, and the average spectral bandwidth corresponding to the bowel sounds for 5 minutes immediately before ingestion (first feature _xb ) and the average spectral bandwidth corresponding to the bowel sounds for 10 minutes immediately after ingestion (second feature _xa ) were calculated. Furthermore, the logarithm of the first feature _xb and the logarithm of the ratio (ratio) of the first feature _xb and the second feature _xa were calculated for each subject.

Next, the logarithms of the first feature xb and the ratio ratio of bowel sounds obtained from 10 subjects with healthy bowel conditions were extracted and plotted on a graph with the logarithm of the first feature _xb on the x-axis and the logarithm of the ratio ratio on the y-axis (green circles ₎ as shown in Figure 12(a). Furthermore, using the plotted coordinates, a linear regression model was constructed and trained to estimate the linear regression coefficients (slope and intercept) based on the least squares method. Specifically, the coefficients (model parameters) were estimated by minimizing the sum of the squared errors between the plotted coordinates and the model predictions, and a regression line was obtained.

Furthermore, a threshold line was set that was a threshold distance away from the regression line. The red line shown in Figure 12(a) is the regression line, and the two blue lines are the threshold lines.

Next, the logarithm of the first feature value _xb and the logarithm of the ratio (ratio) of bowel sounds obtained from 20 subjects whose bowel conditions were not healthy (FD or FC) were extracted and plotted on a graph with a regression line and a threshold line (red circles) as shown in Figure 12(b). It can be seen that the healthy group (green circles) is distributed near the regression line, while the FD and FC groups (red circles) are distributed in areas away from the regression line. Therefore, when the coordinates corresponding to the logarithm of the first feature value _xb and the logarithm of the ratio (ratio) of bowel sounds obtained from a subject are located in the area between the two blue lines, i.e., when the distance between the coordinates and the regression line is within the threshold, the bowel condition can be evaluated as healthy.

The healthy subject group (green circles) was labeled as 0, and the FD and FC groups (red circles) as 1. The distance between each circle and the regression line was calculated as a score, and a threshold was used to classify the score into two classes (1 if greater than the threshold, 0 if less than the threshold). Two-class classification was also performed in the same way as above for the five feature quantities other than the spectral bandwidth of bowel sounds. The classification performance of each feature quantity is shown in Table 2. The threshold was adjusted to maximize classification performance.

The results in Table 2 show that regardless of which feature is used, regression analysis can accurately assess whether the intestinal condition is healthy, FD, or FC. Keeping in mind the bias in the dataset, it was found that, in terms of time-domain features, the frequency of bowel sounds in particular contributed to classification, and, in terms of frequency-domain features, the spectral bandwidth of bowel sounds in particular contributed to classification.

In the results of Table 2, the length of the bowel sounds corresponding to the first feature was 5 minutes, and the length of each bowel sound corresponding to the second feature was 10 minutes. However, by changing the length of each bowel sound corresponding to the first and second feature, two-class classification was performed in the same manner as above. Table 3 shows the classification performance of each feature when the length of the bowel sounds corresponding to the first feature was 10 minutes, and the length of the bowel sounds corresponding to the second feature was 10 minutes.

Table 4 shows the classification performance of each feature when the length of the bowel sound corresponding to the first feature is 5 minutes and the length of the bowel sound corresponding to the second feature is 5 minutes.

Tables 3 and 4 show that the condition of the intestines can be accurately evaluated by setting the length of each bowel sound corresponding to the first feature and second feature to 5 to 10 minutes.

(Additional Notes)
In the above examples, the average value at a predetermined time before or after stimulation was used as the feature value, but statistics such as the median and variance can also be used. Furthermore, in each graph, the horizontal axis represents the measured value, but it may represent a normalized or standardized value, and the vertical axis may also be normalized or standardized. Furthermore, values may be incremented by +1 or the like before logarithmic transformation. Furthermore, in each graph, the horizontal axis represents the measured value before stimulation, but it may represent the measured value after stimulation.

REFERENCE SIGNS LIST 1 Evaluation system 2 Sound collection device 3 Evaluation device 30 Auxiliary storage device 31 Acoustic data acquisition unit 32 Extraction unit 33 Feature amount calculation unit 331 First feature amount calculation unit 332 Second feature amount calculation unit 34 Evaluation unit M Extraction model P Evaluation program R Correlation data

Claims (26)

An evaluation device for evaluating the intestinal condition of a subject,
an extraction unit that extracts bowel sounds from acoustic data obtained from the subject;
a first feature amount calculation unit that calculates a first feature amount, which is one feature amount of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature amount calculation unit that calculates a second feature amount, which is the one feature amount of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation unit that evaluates an intestinal condition of the subject based on the first feature amount and the second feature amount;
Equipped with
the stimulus is a stimulus caused by ingestion of a drink or solid food;
The evaluation unit evaluates the state of the intestine based on the first feature amount and a ratio between the first feature amount and the second feature amount . 2. The evaluation device according to claim 1, wherein the evaluation unit evaluates the state of the intestine based on a distance between a coordinate corresponding to the first feature amount and a ratio between the first feature amount and the second feature amount and an approximation curve obtained from a distribution of coordinates corresponding to the first feature amount and the ratio between the first feature amount and the second feature amount of bowel sounds extracted from acoustic data obtained from a plurality of healthy subjects. An evaluation device for evaluating the intestinal condition of a subject,
an extraction unit that extracts bowel sounds from acoustic data obtained from the subject;
a first feature amount calculation unit that calculates a first feature amount, which is one feature amount of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature amount calculation unit that calculates a second feature amount, which is the one feature amount of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation unit that evaluates an intestinal condition of the subject based on the first feature amount and the second feature amount;
Equipped with
the stimulus is a stimulus caused by ingestion of a drink or solid food;
The evaluation unit evaluates the state of the intestine based on the logarithm of the first feature amount and the logarithm of the ratio between the first feature amount and the second feature amount. 4. The evaluation device according to claim 3, wherein the evaluation unit evaluates the state of the intestine based on a distance between coordinates corresponding to the logarithm of the first feature amount and the logarithm of a ratio between the first feature amount and the second feature amount and a regression line obtained from a distribution of coordinates corresponding to the logarithm of the first feature amount and the logarithm of a ratio between the first feature amount and the second feature amount of bowel sounds extracted from acoustic data obtained from a plurality of healthy subjects. An evaluation device for evaluating the intestinal condition of a subject,
an extraction unit that extracts bowel sounds from acoustic data obtained from the subject;
a first feature amount calculation unit that calculates a first feature amount, which is one feature amount of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature amount calculation unit that calculates a second feature amount, which is the one feature amount of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation unit that evaluates an intestinal condition of the subject based on the first feature amount and the second feature amount;
Equipped with
the stimulus is a stimulus caused by ingestion of a drink or solid food;
The evaluation unit evaluates the state of the intestine based on the first feature amount and a difference between the first feature amount and the second feature amount. An evaluation device for evaluating the intestinal condition of a subject,
an extraction unit that extracts bowel sounds from acoustic data obtained from the subject;
a first feature amount calculation unit that calculates a first feature amount, which is one feature amount of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature amount calculation unit that calculates a second feature amount, which is the one feature amount of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation unit that evaluates an intestinal condition of the subject based on the first feature amount and the second feature amount;
Equipped with
the stimulus is a stimulus caused by ingestion of a drink or solid food;
The evaluation unit evaluates the state of the intestine based on the first feature amount and the sum of the first feature amount and the second feature amount. An evaluation device for evaluating the intestinal condition of a subject,
an extraction unit that extracts bowel sounds from acoustic data obtained from the subject;
a first feature amount calculation unit that calculates a first feature amount, which is one feature amount of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature amount calculation unit that calculates a second feature amount, which is the one feature amount of the bowel sound after giving a stimulus to the intestine, from among the extracted bowel sounds;
an evaluation unit that evaluates an intestinal condition of the subject based on the first feature amount and the second feature amount;
Equipped with
the stimulus is a stimulus caused by ingestion of a drink or solid food;
The evaluation unit evaluates the state of the intestine based on the first feature amount and a product of the first feature amount and the second feature amount. The evaluation device according to claim 1 , wherein the feature amount is a feature amount related to a time domain of the bowel sound. The evaluation device according to claim 8 , wherein the feature amount is an occurrence frequency of the bowel sounds. The evaluation device according to claim 8 , wherein the feature amount is a duration of the bowel sound. The evaluation device according to claim 8 , wherein the feature amount is a signal level of the bowel sound. The evaluation device according to claim 1 , wherein the feature amount is a feature amount related to a frequency domain of the bowel sounds. The evaluation device according to claim 12 , wherein the feature amount is a spectral bandwidth of the bowel sound. The evaluation device according to claim 1 , wherein the evaluation unit evaluates the state of the intestines based on a feature of bowel sounds having a signal level equal to or higher than a threshold, out of the first feature and the second feature. The evaluation device according to claim 1 , wherein the duration of each of the bowel sounds corresponding to the first feature amount and the second feature amount is 5 to 10 minutes. The evaluation device according to any one of claims 1 to 7 ,
a sound collection device for obtaining the acoustic data from the subject;
An evaluation system comprising: 1. An evaluation method in which a computer evaluates a subject's intestinal condition, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
and
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and a ratio between the first feature amount and the second feature amount . 1. An evaluation method in which a computer evaluates a subject's intestinal condition, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
and
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the logarithm of the first feature amount and the logarithm of the ratio between the first feature amount and the second feature amount. 1. An evaluation method in which a computer evaluates a subject's intestinal condition, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
and
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and a difference between the first feature amount and the second feature amount. 1. An evaluation method in which a computer evaluates a subject's intestinal condition, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
and
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and the sum of the first feature amount and the second feature amount. 1. An evaluation method in which a computer evaluates a subject's intestinal condition, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
and
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and a product of the first feature amount and the second feature amount. 1. An assessment program for assessing a subject's intestinal status, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
on the computer,
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and a ratio between the first feature amount and the second feature amount . 1. An assessment program for assessing a subject's intestinal status, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
on the computer,
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the logarithm of the first feature amount and the logarithm of the ratio between the first feature amount and the second feature amount. 1. An assessment program for assessing a subject's intestinal status, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
on the computer,
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and a difference between the first feature amount and the second feature amount. 1. An assessment program for assessing a subject's intestinal status, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
on the computer,
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and the sum of the first feature amount and the second feature amount. 1. An assessment program for assessing a subject's intestinal status, comprising:
an extraction step of extracting bowel sounds from acoustic data obtained from the subject;
a first feature calculation step of calculating a first feature, which is one feature of the bowel sounds before the bowel is stimulated, from among the extracted bowel sounds;
a second feature calculation step of calculating a second feature, which is the one feature of the bowel sound after the bowel has been stimulated, from among the extracted bowel sounds;
an evaluation step of evaluating an intestinal condition of the subject based on the first feature amount and the second feature amount;
on the computer,
the stimulus is a stimulus caused by ingestion of a drink or solid food;
In the evaluation step, the state of the intestine is evaluated based on the first feature amount and the product of the first feature amount and the second feature amount.