JP7122225B2 - Processing device, system, processing method, and program - Google Patents

Description

The present invention relates to a processing device, system, processing method, and program.

In the medical field, it is important to measure the respiratory sound volume of a patient. For example, if attenuation of the respiratory sound volume can be detected, it is useful for early detection of abnormalities such as pneumothorax. Also, in recent years, a method has been proposed for estimating a respiratory condition or the like by analyzing a signal obtained using a sensor such as an electronic stethoscope.

Patent Document 1 describes a technique for calculating body sound characteristics using the power ratio of the body sound at the first part of the body and the body sound at the second part, the power of the body signal in a specific frequency band, and the like. It is

Patent Literature 2 describes extracting expiratory sounds from continuous breath sounds and detecting abnormal expiratory sounds and suspected pseudo-expiratory sounds using a value representing the sound pressure of the expiratory sounds. Further, Patent Document 2 describes that a breathing band sensor is attached so as to wrap around the chest of the subject, changes in expansion and contraction of the chest during breathing are measured, and exhalation sounds are extracted using the measurement results. there is

Patent Literature 3 describes digitizing an output signal of a sensor having a piezoelectric element and filtering it with a high-pass filter to obtain a respiratory airflow sound signal of a living body. Further, Patent Document 3 describes specifying a time zone in which an inspiratory sound and an expiratory sound are generated based on the magnitude of the amplitude of a respiratory airflow sound signal.

Patent Document 4 describes extraction of breathing sounds from body sounds using a band-pass filter. It also describes estimating breathing intervals based on the power pattern of breath sounds. Note that a preset threshold value is used for estimating the breathing period.

International Publication No. 2012/060107 JP 2012-205693 A JP 2017-169647 A International Publication No. 2014/103107

However, the signal obtained by the sensor includes the effects of body sounds other than breathing sounds, and the effects vary from person to person.

One example of the problem to be solved by the present invention is to provide a technique for calculating a breathing volume that is close to the volume perceived by human hearing from body sound data.

The invention according to claim 1,
an acquisition unit that acquires one or more sound data including breath sounds;
an interval identification unit that identifies at least one of a first interval in which breathing is estimated and a second interval between the plurality of first intervals;
using a first portion of the target sound data determined based on the first interval and a second portion of the target sound data determined based on the second interval to generate the target sound data , and a calculating unit for calculating volume information indicating breathing volume.

The invention according to claim 9,
A processing apparatus according to claim 1;
a sensor;
The acquisition unit is a system that acquires the sound data representing the sound detected by the sensor.

The invention according to claim 10,
an acquiring step of acquiring one or more sound data including breath sounds;
an interval identification step of identifying at least one of a first interval in which breathing is estimated and a second interval between the plurality of first intervals;
using a first portion of the target sound data determined based on the first interval and a second portion of the target sound data determined based on the second interval to generate the target sound data , and a calculating step of calculating volume information indicative of breathing volume.

The invention according to claim 11,
11. A program that causes a computer to execute each step of the processing method according to claim 10.

It is a figure which illustrates the structure of the processing apparatus which concerns on 1st Embodiment. 4 is a flowchart illustrating a processing method according to the first embodiment; It is the figure which image-displayed an example of sound data. It is a figure which illustrates the structure of the processing apparatus and system which concern on 2nd Embodiment. 9 is a flowchart illustrating a processing method executed by a processing device according to the second embodiment; It is a flowchart which illustrates in detail the content of processing of area identification step S130. (a) to (d) are diagrams for explaining an example of processing contents of a section identification step S130 according to the second embodiment. FIG. 11 is a diagram for explaining an example of processing contents of a section identifying step S130 according to the second embodiment; FIG. (a) and (b) are diagrams for explaining an example of the processing content of a section identification step S130 according to the second embodiment. FIG. 3 is a diagram illustrating a computer for realizing a processing device; It is a figure which illustrates the structure of the processing apparatus and system which concern on 3rd Embodiment. FIG. 4 is a diagram illustrating attachment positions of a plurality of sensors; FIG. 10 is a diagram showing a display example of volume information of a plurality of parts; It is a figure which illustrates the structure of the processing apparatus and system which concern on 5th Embodiment. FIG. 10 is a diagram showing a display example of volume information of a plurality of parts; FIG. 10 is a diagram showing a display example of volume information of a plurality of parts; FIG. 10 is a boxplot showing the relationship between the value indicating the volume calculated in the comparative example and the result of hearing evaluation. FIG. 10 is a box and whisker diagram showing the relationship between the value indicating the volume calculated in the example and the result of auditory evaluation. FIG. 10 is a histogram showing the relationship between the hearing evaluation results and the volume values calculated in the comparative example. FIG. 10 is a histogram showing the relationship between the hearing evaluation results and the volume values calculated in the example.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, in all the drawings, the same constituent elements are denoted by the same reference numerals, and the description thereof will be omitted as appropriate.

In the following description, each component of the processing device 10 indicates a functional block rather than a hardware configuration, unless otherwise specified. Each component of the processing device 10 includes a CPU of an arbitrary computer, a memory, a program loaded in the memory, a storage medium such as a hard disk for storing the program, an interface for network connection, and an arbitrary combination of hardware and software. Realized by combination. There are various modifications of the implementation method and device.

(First embodiment)
FIG. 1 is a diagram illustrating the configuration of a processing device 10 according to the first embodiment. The processing device 10 according to this embodiment includes an acquisition unit 110 , a section identification unit 130 and a calculation unit 150 . Acquisition unit 110 acquires one or more sound data including breath sounds. The section identification unit 130 identifies at least one of the first section and the second section. The first interval is an interval in which breathing is estimated, and the second interval is an interval between multiple first intervals. The calculation unit 150 uses the first part determined based on the first section of the target sound data and the second part determined based on the second section of the target sound data to calculate the target sound data. Volume information indicating breathing volume is calculated.

FIG. 2 is a flow chart illustrating a processing method according to the first embodiment. The method includes an acquisition step S110, an interval identification step S130, and a calculation step S150. In acquisition step S110, one or more sound data including breath sounds are acquired. At least one of the first section and the second section is identified in the section identifying step S130. The first interval is an interval in which breathing is estimated, and the second interval is an interval between multiple first intervals. In the calculation step S150, the first portion of the target sound data determined based on the first section and the second portion of the target sound data determined based on the second section are used to calculate the target sound data. , volume information indicating breathing volume is calculated.

This processing method can be executed by the processing device 10 .

As a method of calculating the breathing sound volume, for example, there is a method of performing filter processing to remove a specific frequency component from the biological signal and obtaining the signal power after extracting the breathing sound component. However, it is difficult to uniquely determine a cutoff frequency that separates respiratory sound components from other body sound components (pulsation, heart sound, blood flow sound, etc.). This is because the cut-off frequency that is appropriate when trying to extract breath sounds by filtering alone differs depending on the part where the body sound is acquired and the individual difference of the person to be measured.

In addition, the frequency band of the respiratory sound component and the frequency band of the other body sound components overlap each other in the body sounds detected at any part. In one measurement example, in body sounds obtained by a sensor attached to the neck, breathing sound components appeared in the band from 0 Hz to 1500 Hz, and pulsation and blood flow sound components appeared in the bands from 0 Hz to 200 Hz. In addition, in body sounds obtained by the sensor attached to the upper right part of the chest, respiratory sound components appeared in the band from 0 Hz to 300 Hz, and heart sound components appeared in the band from 0 Hz to 500 Hz. In body sounds obtained by the sensor attached to the upper left of the chest, respiratory sound components appeared in the band from 0 Hz to 300 Hz, and heart sound components appeared in the band from 0 Hz to 400 Hz. In body sounds obtained by the sensor attached to the lower right part of the chest, respiratory sound components appeared in the band from 0 Hz to 300 Hz, and heart sound components appeared in the band from 0 Hz to 300 Hz. Therefore, even if the signal power is obtained by simply limiting the band, the influence of other components cannot be avoided.

In the processing device 10 according to this embodiment, the acquisition unit 110 acquires sound data from a sensor attached to a living body, for example. The section identification unit 130 identifies the first section and the second section. The first section is a section in which the living body is estimated to be inhaling or exhaling. A second section is a section between the first sections. More specifically, the second section is a section other than the first section. That is, the second interval is an interval in which no breathing is estimated, and an interval in which apnea is estimated to occur temporarily between respirations. Note that the second section does not necessarily have to be between the first sections. For example, the end of sound data may be identified as the second segment.

FIG. 3 is a diagram showing an image display of an example of sound data. In this figure, the display area 501 shows the time waveform of the sound data, and the display area 502 shows the spectrogram of the sound data. In the spectrogram, the horizontal axis is time (time), the vertical axis is frequency, and the intensity of each frequency component is indicated by luminance. Note that the horizontal axis of the time waveform and the horizontal axis of the spectrogram are aligned. Also, the section specified as the first section is indicated by an arrow in the display area 502 . On the other hand, the section with no arrow is the second section. Thus, an interval indicates a time range. Further, the sound data has a plurality of first intervals separated from each other and a plurality of second intervals separated from each other.

Here, it is assumed that the portion estimated to be breathing includes breath sound components and other body sound components, and the portion estimated to be apnea includes only other body sound components. be done. Therefore, by comparing the data of the portion where it is estimated that breathing is performed and the data of the other portion, it is possible to obtain volume information in which the influence of other body sound components is reduced. The volume information calculated in this manner has a high correlation with, for example, the volume of breathing sounds perceived by human hearing. The use of such sound volume information makes it easier to detect, for example, biological abnormalities by data processing, and is useful, for example, in assisting diagnosis and monitoring the patient's condition.

Details of the processing performed by the processing device 10 will be described in detail in the second embodiment and thereafter.

As described above, according to the present embodiment, the calculation unit 150 calculates the first part determined based on the first section of the target sound data and the second part determined based on the second section of the target sound data. is used to calculate volume information indicating the breathing volume of the target sound data. Therefore, from body sound data, it is possible to calculate a breathing volume that is close to the volume sensed by human hearing.

(Second embodiment)
FIG. 4 is a diagram illustrating configurations of the processing device 10 and the system 20 according to the second embodiment. The processing apparatus 10 according to this embodiment has the configuration of the processing apparatus 10 according to the first embodiment. FIG. 5 is a flowchart illustrating a processing method executed by the processing device 10 according to the second embodiment. The processing method according to this embodiment has the configuration of the processing method according to the first embodiment.

A system 20 according to this embodiment includes a processing device 10 and a sensor 210 . Acquisition unit 110 then acquires sound data representing the sound detected by sensor 210 .

Each component of processing device 10 and system 20 is described in detail below.

Sensor 210 detects body sounds including breathing sounds. The sensor 210 generates an electrical signal representing the body sound and outputs it as sound data. Sensor 210 is, for example, a microphone or a vibration sensor. Vibration sensors are for example displacement sensors, velocity sensors or acceleration sensors. Microphones convert air vibrations caused by body sounds into electrical signals. The signal level value of this electrical signal indicates the sound pressure of the air vibration. On the other hand, the vibration sensor converts the vibration of a medium (for example, the subject's body surface) caused by the body sound into an electrical signal. The signal level value of this electrical signal directly or indirectly indicates the vibrational displacement of the medium. For example, if the vibration sensor has a diaphragm, the vibration of the medium is transmitted to the diaphragm and the vibration of the diaphragm is converted into an electrical signal. Note that the electric signal may be an analog signal or a digital signal. Further, the sensor 210 may be configured including a circuit or the like for processing electrical signals. Circuits that process electrical signals include, for example, A/D conversion circuits and filter circuits. However, A/D conversion and the like may be performed by the processing device 10 .

The sound data is data indicating an electrical signal, and is data indicating signal level values based on the electrical signal obtained by the sensor 210 in time series. That is, the sound data represents the waveform of sound waves. Note that one piece of sound data means a series of sound data temporally.

Also, sensor 210 is, for example, an electronic stethoscope. For example, the sensor 210 is pressed or attached to a part of the subject's body where the body sound is to be measured by the subject. In this embodiment, an example in which the acquisition unit 110 acquires only one continuous sound data will be described.

Acquisition unit 110 acquires, for example, sound data from sensor 210 in acquisition step S110. In this case, the acquisition unit 110 can acquire sound data detected by the sensor 210 in real time. However, the sound data measured by the sensor 210 in advance and held in the storage device may be read and acquired by the acquisition unit 110 . The storage device may be provided inside the processing device 10 or may be provided outside. A storage device provided inside the processing device 10 is, for example, a storage device 1080 of the computer 1000, which will be described later. Alternatively, the acquisition unit 110 may acquire sound data that has been output from the sensor 210 and has undergone conversion processing or the like in the processing device 10 or in a device other than the processing device 10 . Examples of conversion processing include amplification processing and A/D conversion processing.

Acquisition unit 110 continuously acquires sound data including body sounds from sensor 210 , for example. Each signal level value of the sound data is associated with the recording time. The sound data may be associated with the time in the sensor 210, or when the sound data is obtained from the sensor 210 in real time, the obtaining unit 110 may associate the acquisition time of the sound data with the sound data.

Of the sound data acquired by the acquisition unit 110, the sound data targeted for calculation of the volume information is particularly referred to as target sound data. In the present embodiment, the acquisition unit 110 acquires only one sound data, so the acquired sound data is the target sound data. Acquisition unit 110 can continue to acquire sound data while subsequent step S120, section identification step S130, and calculation step S150 are being performed. The acquired sound data is subjected to the following processing in order from the beginning.

In the example of this figure, the processing device 10 further includes a filtering unit 120 . When acquisition of sound data by the acquisition unit 110 is started, step S120 is started. In step S120, the filtering unit 120 performs a first filtering process on at least target sound data. Specifically, the filter processing unit 120 performs band-pass filtering with the cutoff frequency on the low frequency side set to f _L1 [Hz] and the cutoff frequency on the high frequency side set to f _H1 [Hz]. In the first filtering process, for example, Fourier transform is performed on the sound data to remove a band below f _L1 [Hz] and a band above f _H1 [Hz] in the frequency space. After that, the time domain waveform is restored by inverse Fourier transform. The Fourier transform is for example the Fast Fourier Transform (FFT). However, the first filtering process is not limited to the above example, and may be, for example, a process using an FIR (Finite Impulse Response) filter or an IIR (Infinite Impulse Response) filter.

In the example of this figure, the calculation unit 150 calculates the volume information using the first part and the second part of the target sound data after the first filtering process. Here, it is preferable that 50≦f _L1 ≦150 holds, and 500≦f _H1 ≦1500 holds. Noise contained in the sound data can be removed by the first filtering process. Note that it is not necessary to extract only the breath sound component by the first filtering process. The sound data after the first filter processing may include components other than breath sounds.

The section identification unit 130 identifies at least one of the first section and the second section. In the examples of FIGS. 4 and 5 , the section identifying section 130 identifies at least one of the first section and the second section based on the sound data acquired by the acquiring section 110 . However, the section identification unit 130 may identify at least one of the first section and the second section without using sound data. A method for identifying a section by the section identification unit 130 will be described later in detail.

Further, the section identification unit 130 generates at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section based on the section identification result. For example, when section identification unit 130 identifies a first section, section identification section 130 generates first time information, and when section identification section 130 identifies a second section, section identification section 130 generates second time information. do. Note that when a plurality of discontinuous first intervals are identified by the interval identification unit 130, a plurality of pieces of first time information are generated. Also, when a plurality of discontinuous second intervals are identified by the interval identification unit 130, a plurality of pieces of second time information are generated.

In calculation step S<b>150 , the calculation unit 150 calculates volume information using the data in the first area among the target sound data acquired by the acquisition unit 110 . The _first region is, for example, a region indicating the sound during the time T1, which is the most recent time when the calculation unit 150 performs this step. Although T1 is not particularly limited, it is, for example, ₂ seconds or more and 30 seconds or less.

As described above, the calculation unit 150 uses at least one of the first time information and the second time information generated by the section identification unit 130 to identify the first part and the second part in the first region of the target sound data. do.

When the section identification unit 130 generates the first time information and the second time information, the calculation unit 150 sets the time range portion indicated by the first time information in the first region of the target sound data as the first portion, A part of the time range indicated by the second time information in the first area is referred to as a second part. Further, when the section identification unit 130 generates only the first time information, the calculation unit 150 sets the part of the time range indicated by the first time information in the first area as the first part, and the other part of the first area. Let the part be the second part. Further, when the section identification unit 130 generates only the second time information, the calculation unit 150 sets the part of the time range indicated by the second time information in the first area as the second part, Let the part other than be the 1st part.

The calculator 150 then calculates a first signal strength that is the strength of the first portion of the target sound data and a second signal strength that is the strength of the second portion of the target sound data. Specifically, for example, the calculator 150 calculates the RMS (root mean square) of the first portion of the target sound data as the first signal strength. Also, the calculation unit 150 calculates the RMS of the second portion of the target sound data as the second signal strength. Note that the calculator 150 may calculate another index such as a peak to peak value as the signal strength instead of the RMS. However, the method for calculating the first signal strength and the method for calculating the second signal strength are the same.

When two or more first portions are included in the processing range, the first signal strength is a value indicating the signal strength when all the first portions are viewed as one continuous signal, for example. Moreover, when two or more second portions are included in the processing range, the second signal strength is a value indicating the signal strength when all the second portions are viewed as one continuous signal, for example.

The calculator 150 then calculates the volume information of the target sound data using the first signal strength and the second signal strength. Volume information does not need to be absolute volume (eg, dB SPL, etc.) measured by other equipment. The volume information may be a relative value that can be compared at least between the volume information obtained by the processing device 10 .

Specifically, the calculator 150 calculates at least one of the information specifying the ratio of the first signal strength to the second signal strength and the information specifying the difference between the first signal strength and the second signal strength as the target sound data. volume information. However, it is preferable that the calculation unit 150 calculates at least information specifying the ratio of the first signal strength to the second signal strength as volume information. By doing so, the volume information can be expressed in units of dB like the normal volume, and the information can be closer to the volume sensed by human hearing. The information specifying the ratio of the first signal strength to the second signal strength is, for example, the value obtained by dividing the first signal strength by the second signal strength, the value obtained by dividing the second signal strength by the first signal strength, and these values is one of the values expressed in decibels.

Calculation unit 150 similarly calculates volume information, for example, _every time T2. Although T2 is not particularly limited, it is, for example, 1 _second or more and 10 seconds or less.

The volume information calculated by the calculator 150 is displayed, for example, on a display device. Here, the calculation unit 150 may calculate a plurality of pieces of volume information of the target sound data in time series, and a graph showing the plurality of pieces of volume information in time series may be displayed on the display device. However, the volume information may be displayed numerically. Also, the volume information calculated by the calculation unit 150 may be stored in a storage device, or may be output to a device other than the processing device 10 .

FIG. 6 is a flowchart illustrating in detail the processing contents of the section identification step S130. 7(a) to 9(b) are diagrams for explaining an example of the processing contents of the section identification step S130 according to this embodiment. 7(a) to 7(d), 9(a), and 9(b), the horizontal axis represents the elapsed time from the reference time. An example of how the section identification unit 130 identifies sections will be described in detail below with reference to FIGS. 6 to 9B.

In a section identification step S130, the section identification unit 130 identifies at least one of the first section and the second section using a threshold for the value indicating the amplitude of the first sound data. The value indicating amplitude is a value indicating the magnitude of vibration of the first sound data at each time. The section identification unit 130 then determines the threshold based on the first sound data.

The sound data acquired by the acquisition unit 110 includes first sound data and target sound data. Here, the first sound data is sound data used for identifying the section, and the target sound data is sound data for which volume information is to be calculated. When the acquisition unit 110 acquires only one sound data, both the first sound data and the target sound data are this one sound data and are the same when acquired by the acquisition unit 110 . In this embodiment, the sound data acquired by the acquisition unit 110 is preferably body sound data acquired at the neck. Then, the section can be specified more accurately. This is because the breath sound component can be detected at a high ratio in the neck and its vicinity.

The section identifying unit 130 performs second filtering on the first sound data acquired by the acquiring unit 110 in step S131. In the second filtering process, for example, Fourier transform is performed on the sound data, and a band below f _L2 [Hz] and a band above f _H2 [Hz] are removed in the frequency space. After that, the time domain waveform is restored by inverse Fourier transform. The Fourier transform is for example the Fast Fourier Transform (FFT). However, the second filter processing is not limited to the above example, and may be processing using an FIR filter or an IIR filter, for example.

The interval specifying unit 130 obtains a mode value, which will be described later, based on the first sound data after performing the second filtering process. The second filtering process is a bandpass filtering process in which the cutoff frequency on the low frequency side is f _L2 [Hz] and the cutoff frequency on the high frequency side is f _H2 [Hz]. Here, 150≦f _L2 ≦250 is preferably satisfied, and 550≦f _H2 ≦650 is preferably satisfied. By the second filtering process, it is possible to mainly extract breath sound components included in the first sound data. Note that it is not necessary to extract only breath sound components by the second filtering process. Also, the first sound data after the second filter processing may include components other than breath sounds.

FIG. 7A is a diagram illustrating the waveform of the first sound data when acquired by the acquisition unit 110, and FIG. FIG. 11 is a diagram illustrating waveforms after the . As indicated by arrows in FIG. 7(b), breath sound components are mainly extracted by the second filtering process.

Note that the first sound data acquired by the acquiring unit 110 undergoes both the first filtering process in step S120 by the filtering process part 120 and the second filtering process in step S131 by the section specifying part 130. It's okay to be broken. However, if the passband of the second filter process is narrower than the passband of the first filter process, the presence or absence of the first filter process does not affect the section identification result.

Next, in step S132, the section identifying unit 130 calculates the absolute value of the data obtained in step S131. That is, each signal level value in time-series data is converted into an absolute value. FIG. 7(c) is a diagram showing the result of calculating the absolute value of the data in FIG. 7(b).

Next, in step S133, the section identifying unit 130 down-samples the data obtained in step S132. By doing so, the contour of the data waveform is obtained as shown in FIG. 7(d). Each data point of the data obtained by the downsampling process corresponds to a value indicating the amplitude of the first sound data at each time. The interval specifying unit 130 obtains the mode value, which will be described later, based on the data obtained by performing at least the downsampling process on the first sound data. Here, in FIG. 7(d), downward arrows indicate portions where it is estimated that breathing is taking place, and upward arrows indicate portions where it is estimated that breathing is not occurring. Ultimately, it is sufficient if the section identification unit 130 can identify these.

Next, in step S134, the section identification unit 130 determines whether or not to update (determine) the threshold based on predetermined update conditions. Specifically, for example, if the interval identification unit 130 has never determined a threshold for the first sound data, the interval identification unit 130 determines that the update condition is satisfied. On the other hand, if the interval identification unit 130 has determined the threshold value for the first sound data at least once, the interval identification unit 130 determines that the update condition is not satisfied.

If it is determined that the update condition is satisfied (Yes in step S134), the section identification unit 130 proceeds to step S135 and performs processing for determining a threshold value. On the other hand, if it is determined that the update condition is not satisfied (No in step S134), the section identifying unit 130 performs step S137 using the threshold already determined.

Processing for determining the threshold value by the section identification unit 130 will be described below. The interval identification unit 130 obtains the mode of values indicating amplitude in the first data. Specifically, in step S135, the section identifying unit 130 obtains the mode of the value indicating the amplitude of the first sound data. For this purpose, the section identification unit 130 counts the number of appearances of values indicating each amplitude within a predetermined time range in the first sound data. Then, the section identification unit 130 obtains the value indicating the amplitude with the highest number of appearances as the mode value. The predetermined time range may be, for example, a range from when acquisition section 110 starts acquiring the first sound data to when section identification section 130 performs this step. It may be time T3 _, which is the most recent time at the time of performing Although T3 is not particularly limited _, it is, for example, 2 seconds or more and 30 seconds or less. Also, for example _, T3 may be the same as _T1 . When the first sound data is the target sound data, the area of the predetermined time range may be matched with the above-described first area.

As described above, the acquisition unit 110 may read and acquire the sound data held in the storage device. In that case, for example, the section identification unit 130 may use the entire sound data to identify the mode and determine the threshold value.

FIG. 8 is a histogram illustrating the number of occurrences of each amplitude. Such a histogram corresponds to a graph in which the first sound data is represented by a value indicating amplitude on the horizontal axis and the number of appearances on the vertical axis. Note that the value indicating the amplitude is not limited to the value obtained by the above processing, and may be, for example, a peak-to-peak value or a normalized value.

Then, in step S<b>136 , the section identification unit 130 determines a threshold that is greater than the mode. Specifically, in the histogram, among the values representing one or more amplitudes having the local minimum value, the value representing the amplitude closest to the mode is set as the threshold value. Note that the mode value in the histogram is the value that indicates the smallest amplitude among the values that indicate the maximum amplitude. In the graph of FIG. 8, a point 505 with the maximum number of appearances and a plurality of local minimum values are circled. In the example of this figure, the value indicating the amplitude at this point 505 is the mode. Then, among the minimum values, the value indicating the amplitude that takes the minimum value 506 on the lowest amplitude side is determined as the threshold value. The mode value is considered to be a value that mainly indicates the amplitude corresponding to the apnea interval. Therefore, by determining the threshold value in this way, a threshold value that can distinguish between an apnea interval and other intervals can be obtained. Moreover, since this threshold value is obtained using the sound data obtained by the obtaining unit 110, it is possible to specify the breathing timing in body sounds with high accuracy regardless of individual differences. Note that the threshold is not limited to the above example. For example, the threshold may be a value indicating the second or third minimum amplitude from the low amplitude side in the histogram, or may be a value indicating the maximum amplitude next to the point 505 .

If the threshold is determined in step S136, or if it is determined that the update condition is not satisfied in step S134, then in step S137 the section identification unit 130 performs at least one of the following first processing and second processing. The first process is a process of specifying, as a first section, a section in the first sound data in which at least a value indicating amplitude exceeds a threshold value. On the other hand, the second process is a process of specifying, as a second segment, a segment in the first sound data in which at least the value indicating the amplitude is less than the threshold value. Note that the interval that matches the threshold may be included in the first interval or may be included in the second interval.

Specifically, the section identification unit 130 applies a threshold value to the first sound data processed in step S133, and divides the sections in which the value indicating the amplitude continuously exceeds the threshold value into one first section. and On the other hand, a section in which the value indicating the amplitude is continuously less than the threshold value is defined as one second section. Note that the section identification unit 130 may identify only one of the first section and the second section. In that case, the remaining section becomes the other section.

FIG. 9(a) is obtained by marking the points below the threshold in the graph of FIG. 7(d) with circles. FIG. 9(b) shows an enlarged view of the small amplitude side of FIG. 9(a). Further, in FIG. 9B, a straight line indicating the threshold is attached.

The section identifying unit 130 identifies a section of newly acquired sound data, for example, every time _T4 . Although _T4 is not particularly limited, it is, for example, 1 second or more and 10 seconds or less. Note that _T4 is preferably equal to or less _than T2 described above.

Note that the section identifying unit 130 may determine a threshold each time a section is identified. Also, each time the calculation unit 150 calculates the volume information, the section identification unit 130 may identify the section. For example, determination of a threshold value, specification of a section, and calculation of volume information may be performed on sound data in the same time range as processing targets.

Also, the section identification unit 130 may identify the section by another method. For example, a predetermined threshold may be used to similarly determine the section. Also, the section identification unit 130 may identify the section based on the output of a band sensor attached to the chest of the target person. For example, a band sensor can detect chest expansion and movement during breathing.

Each functional configuration unit of the processing device 10 may be implemented by hardware (eg, hardwired electronic circuit, etc.) that implements each functional configuration unit, or may be implemented by a combination of hardware and software (eg, electronic A combination of a circuit and a program that controls it, etc.). A case in which each functional configuration unit of the processing device 10 is implemented by a combination of hardware and software will be further described below.

FIG. 10 is a diagram illustrating a computer 1000 for realizing the processing device 10. As shown in FIG. Computer 1000 is any computer. For example, the computer 1000 is an SoC (System On Chip), a Personal Computer (PC), a server machine, a tablet terminal, a smart phone, or the like. The computer 1000 may be a dedicated computer designed to implement the processing device 10, or may be a general-purpose computer.

Computer 1000 has bus 1020 , processor 1040 , memory 1060 , storage device 1080 , input/output interface 1100 and network interface 1120 . A bus 1020 is a data transmission path through which the processor 1040, memory 1060, storage device 1080, input/output interface 1100, and network interface 1120 mutually transmit and receive data. However, the method of connecting processors 1040 and the like to each other is not limited to bus connection. The processor 1040 is various processors such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an FPGA (Field-Programmable Gate Array). The memory 1060 is a main memory implemented using a RAM (Random Access Memory) or the like. The storage device 1080 is an auxiliary storage device implemented using a hard disk, SSD (Solid State Drive), memory card, ROM (Read Only Memory), or the like.

The input/output interface 1100 is an interface for connecting the computer 1000 and input/output devices. For example, the input/output interface 1100 is connected to an input device such as a keyboard and a mouse, and an output device such as a display device. Also, the input/output interface 1100 may be connected to a touch panel or the like that serves as both a display device and an input device.

A network interface 1120 is an interface for connecting the computer 1000 to a network. This communication network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network). A method for connecting the network interface 1120 to the network may be a wireless connection or a wired connection.

The storage device 1080 stores program modules that implement each functional component of the processing device 10 . The processor 1040 reads each program module into the memory 1060 and executes it, thereby realizing the function corresponding to each program module.

Sensor 210 is connected, for example, to input/output interface 1100 of computer 1000 or to network interface 1120 through a network.

As described above, according to the present embodiment, as in the first embodiment, the calculation unit 150 calculates the first portion determined based on the first section of the target sound data and the second section of the target sound data. Volume information indicating the breathing volume of the target sound data is calculated using the second part determined based on the above. Therefore, from body sound data, it is possible to calculate a breathing volume that is close to the volume sensed by human hearing.

(Third embodiment)
FIG. 11 is a diagram illustrating the configurations of the processing device 10 and the system 20 according to the third embodiment. The processing apparatus 10 and system 20 according to this embodiment are the same as the processing apparatus 10 and system 20 according to the second embodiment, respectively, except for the points described below.

A system 20 according to this embodiment includes a plurality of sensors 210 , and an acquisition unit 110 according to this embodiment acquires a plurality of sound data representing sounds detected by the plurality of sensors 210 . A plurality of sensors 210 detect body sounds at, for example, a plurality of body parts of the same person. The acquisition unit 110 can acquire sound data of body sounds simultaneously detected at a plurality of parts of the living body of the same person. In addition, the acquiring unit 110 can acquire a plurality of sound data whose recording times are at least partly overlapping with each other. Although the acquisition unit 110 may further acquire body sound data of a plurality of persons, at least the processing by the section identification unit 130 and the calculation unit 150 is performed for each person. In addition, the acquisition unit 110 may acquire a plurality of sound data whose recording times do not overlap each other, but at least the processing by the section identification unit 130 and the calculation unit 150 is performed for each piece of sound data or for at least part of the recorded sound data. This is performed for each of a plurality of sound data whose times overlap each other.

FIG. 12 is a diagram illustrating mounting positions of the plurality of sensors 210. As shown in FIG. Sensors 210 are attached to sites A through D in the example of this figure. For example, the processing device 10 receives input of information indicating the attachment site of the sensor 210 by the user prior to acquisition of the sound data. Specifically, a diagram showing the living body is displayed on the display device together with candidates for the attachment sites of the sensors 210, and the user designates the attachment positions of the respective sensors 210 among the candidates using an input device such as a mouse, a keyboard, or a touch panel. do. The sound data acquired by each sensor 210 is associated with information indicating the site.

In the present embodiment, the section identification unit 130 identifies at least one of the first section and the second section based on the first sound data, as described in the second embodiment. That is, the plurality of sound data acquired by the acquisition unit 110 includes the target sound data and the first sound data. Here, the target sound data is not limited to one. In the following, an example in which target sound data includes at least second sound data different from the first sound data will be described. That is, the volume information of the second sound data is calculated based on the section specified by the first sound data. A plurality of second sound data may exist.

For example, the first sound data indicates sound detected by a first sensor 210 provided at a first location on or inside the human body. The second sound data indicates the sound detected by the second sensor 210 provided at a second position on or inside the human body. Here, it is preferred that the first position is located at the neck or the first position is closer to the neck than the second position. Then, the section can be specified more accurately. For example, in the example of FIG. 12, it is preferable that the sound data obtained at the site A is the first sound data. For example, the section identification unit 130 selects which of the plurality of sound data acquired by the acquisition unit 110 is to be the first sound data based on the information indicating the part associated with each sound data.

Acquisition step S110 and step S120 according to the present embodiment are performed by the acquisition unit 110 and the filter processing unit 120, respectively, in the same manner as in the second embodiment.

Then, in the section identification step S130, the section identification unit 130 according to the present embodiment identifies at least one of the first section and the second section in the first sound data, as in the second embodiment. Then, at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section is generated based on the first sound data.

Further, in the calculation step S150, the calculation unit 150 according to the present embodiment uses at least one of the first time information and the second time information to calculate the first portion of each target sound data as in the second embodiment. and a second portion. By doing so, even in the second sound data included in the target sound data, it is possible to specify the first section in which breathing is estimated and the other second section.

Further, the calculation unit 150 calculates volume information for each target sound data, as in the second embodiment. By doing so, volume information for each part can be obtained. Note that the calculation unit 150 may set all the sound data acquired by the acquisition unit 110 as the target sound data, or may set only the sound data corresponding to the part designated by the user as the target sound data.

FIG. 13 is a diagram showing a display example of volume information of a plurality of parts. In the example of this figure, the volume information of a plurality of parts is displayed in a state in which the correspondence to each part can be understood. Specifically, a numerical value indicating the volume is displayed based on the volume information on the map indicating the site. In addition, numerical values indicating the volume are shown in the graph in chronological order. In the graph of this figure, the horizontal axis indicates the elapsed time from the reference time, and the plural parts are synchronized. It should be noted that the scale of the axis of the graph may be enlarged or reduced or moved in parallel as needed by the user.

Note that the target sound data may further include the first sound data. Also, in the present embodiment, an example in which the target sound data includes at least the second sound data different from the first sound data has been described, but the target sound data may include only the first sound data. Volume information of the first sound data can also be calculated in the same manner as described above.

As described above, according to the present embodiment, as in the first embodiment, the calculation unit 150 calculates the first portion determined based on the first section of the target sound data and the second section of the target sound data. Volume information indicating the breathing volume of the target sound data is calculated using the second part determined based on the above. Therefore, from body sound data, it is possible to calculate a breathing volume that is close to the volume sensed by human hearing.

(Fourth embodiment)
The processing device 10 and the system 20 according to the fourth embodiment are the same as the processing device 10 and the system 20 according to the third embodiment, respectively, except for the processing contents of the section identification unit 130 and the calculation unit 150 .

The acquiring unit 110 according to the present embodiment acquires a plurality of sound data representing sounds detected by the plurality of sensors 210, like the acquiring unit 110 according to the third embodiment. Then, the section identification unit 130 identifies the first section and the second section in each of two or more pieces of sound data among the acquired plurality of sound data. That is, in the present embodiment, the section identification unit 130 uses two or more pieces of sound data among the plurality of pieces of sound data acquired by the acquisition unit 110 as the first sound data. By specifying a section using two or more pieces of first sound data, the accuracy of section specification can be improved.

Note that when the threshold is determined based on the first sound data as in the section identification step S130 according to the second embodiment, the section identification unit 130 sets the threshold for each first sound data that identifies the section. decide.

The section identification unit 130 determines the time range designated as the first section in all the first sound data whose sections are identified, or the time range designated as the second section in all the first sound data whose sections are identified. The third time information shown is generated.

Here, the two or more first sound data preferably include sound data detected at the neck or sound data detected at a position closest to the neck among a plurality of sound data. Then, the section can be specified more accurately.

Then, the calculation unit 150 identifies the first part and the second part of each target sound data using the third time information. Specifically, when the third time information is information indicating the time range defined as the first section in all the first sound data, the calculation unit 150 determines that the third time information in each target sound data indicates Let the time range portion be the first portion, and the other portion be the second portion. On the other hand, if the third time information is information indicating the time range defined as the second section in all the first sound data, the calculation unit 150 determines the time range indicated by the third time information in each target sound data. Let the part be the second part, and let the other part be the first part.

The target sound data may or may not include the first sound data. Further, the target sound data may or may not include second sound data different from the first sound data.

Further, the calculation unit 150 calculates volume information for each target sound data, as in the second embodiment. By doing so, volume information for each part can be obtained.

As described above, according to the present embodiment, as in the first embodiment, the calculation unit 150 calculates the first portion determined based on the first section of the target sound data and the second section of the target sound data. Volume information indicating the breathing volume of the target sound data is calculated using the second part determined based on the above. Therefore, from body sound data, it is possible to calculate a breathing volume that is close to the volume sensed by human hearing.

In addition, according to this embodiment, the section identification unit 130 identifies the first section and the second section in each of the two or more pieces of sound data. Therefore, it is possible to improve the accuracy of section identification.

(Fifth embodiment)
FIG. 14 is a diagram illustrating configurations of the processing device 10 and the system 20 according to the fifth embodiment. The processing apparatus 10 and system 20 according to this embodiment are similar to at least any one of the second to fourth embodiments except for the points described below.

In this embodiment, the processing device 10 further includes an estimator 170 . Based on the volume information calculated by the calculation unit 150, the estimation unit 170 estimates the state of the person whose body sound is detected. A detailed description is given below.

For example, the calculator 150 calculates volume information of a plurality of target sound data in the same manner as at least one of the second to fourth embodiments.

15 and 16 are diagrams showing display examples of volume information of a plurality of parts, respectively.

When the calculator 150 calculates the volume information, the estimator 170 acquires the calculated volume information from the calculator 150 . Each piece of volume information is associated with information indicating a part. Estimating unit 170 calculates, for example, the rate of decrease in volume indicated by the volume information of each part per hour. Then, if the rate of decrease exceeds a predetermined reference value, it is estimated that breathing is weakened. In this case, the estimating unit 170 displays or notifies that breathing is weak, as shown in FIG. 15, for example. It should be noted that the estimation unit 170 may estimate that breathing is weakened when the rate of volume reduction is high in a predetermined number or more of parts. In addition, the estimating unit 170 may estimate that breathing is weakened when the rate of decrease in volume increases over a predetermined length.

Also, the estimation unit 170 calculates, for example, the difference between two sound data representing body sounds detected at mutually symmetrical positions in the living body. Then, the estimation unit 170 estimates that pneumothorax is suspected when the magnitude of the difference exceeds a predetermined reference value. Also, based on whether the difference is positive or negative, it estimates which lung is suspected of having a pneumothorax. In this manner, the estimation unit 170 may estimate the position of the sound source of the abnormal breath sound based on the calculated plurality of volume information. Then, the estimating unit 170 displays or notifies that pneumothorax is suspected and the estimated position as shown in FIG. 16, for example. Note that the estimating unit 170 may estimate that breathing is weak when the magnitude of the difference exceeds the reference value over a predetermined length.

The processing device 10 according to this embodiment can also be implemented using a computer 1000 as shown in FIG.

As described above, according to the present embodiment, as in the first embodiment, the calculation unit 150 calculates the first portion determined based on the first section of the target sound data and the second section of the target sound data. Volume information indicating the breathing volume of the target sound data is calculated using the second part determined based on the above. Therefore, from body sound data, it is possible to calculate a breathing volume that is close to the volume sensed by human hearing.

In addition, according to the present embodiment, the processing device 10 includes the estimation unit 170 that estimates the state of the person from whom body sounds are detected based on the volume information calculated by the calculation unit 150 . Therefore, it is possible to monitor the condition of the patient or the like.

Hereinafter, this embodiment will be described in detail with reference to examples. It should be noted that the present embodiment is not limited to the description of these examples.

Sound data was obtained by measuring body sounds at 4 sites (neck, upper right chest, upper left chest, and lower right chest) in 22 subjects. Then, the obtained sound data was reproduced, and an auditory evaluation was performed to determine whether or not breathing sounds could be heard. In the auditory evaluation, "0" was given when no breathing sound was heard, "1" when barely heard, and "2" when well heard.

On the other hand, the sound data was processed by each method of the example and the comparative example, and the value indicating the volume was calculated. Then, the calculated values were compared with the results of auditory evaluation.

In the comparative example, high-pass filter processing with a cutoff frequency of 400 Hz was performed on sound data. Then, the RMS was calculated, and the value obtained by expressing the RMS in decibels was used as the value indicating the volume. Note that 0 dB=1 digit.

In the example, the value indicating the volume was calculated as described in the second embodiment. Specifically, a threshold was determined based on each sound data, and the interval was specified. Also, the RMS of each section of the sound data subjected to the first filter processing with f _L1 set to 100 Hz and f _H1 set to 1000 Hz was calculated. Then, the value obtained by dividing the RMS of the first interval by the RMS of the second interval and expressed in decibels was used as the value indicating the volume. That is, the RMS of the second section was set to 0 dB. The determination of the threshold value, the identification of the section, and the calculation of the value indicating the volume were performed independently for each sound data.

17 and 18 are boxplots showing the relationship between the values indicating the volume calculated in the comparative example and the example and the auditory evaluation results, respectively. FIG. 19 is a histogram showing the relationship between the hearing evaluation results and the volume values calculated in the comparative example. FIG. 20 is a histogram showing the relationship between the auditory evaluation results and the volume values calculated in the example. As shown in FIGS. 17 and 19, in the comparative example, the difference between the sound data of evaluation 1 and the sound data of evaluation 0 did not appear appropriately as a difference in volume. On the other hand, as shown in FIGS. 18 and 20, in the example, the magnitude of the value indicating the volume is well correlated with the evaluation 0, the evaluation 1, and the evaluation 2, and the sound data with the evaluation 0 and the sound with the evaluation 1 The data were also clearly identifiable with values indicating loudness. As described above, the method of the embodiment could calculate a value highly correlated with the result of the auditory evaluation as the volume. Therefore, it was confirmed that the method of the embodiment can calculate a breathing volume close to the volume sensed by human hearing from the sound data.

Although the embodiments of the present invention have been described above with reference to the drawings, these are examples of the present invention, and various configurations other than those described above can also be adopted. For example, in the flowcharts used in the above description, a plurality of steps (processes) are described in order, but the execution order of the steps executed in each embodiment is not limited to the described order. In each embodiment, the order of the illustrated steps can be changed within a range that does not interfere with the content. Moreover, each of the above-described embodiments can be combined as long as the contents do not contradict each other.

Examples of reference forms are added below.
1-1. an acquisition unit that acquires one or more sound data including breath sounds;
an interval identification unit that identifies at least one of a first interval in which breathing is estimated and a second interval between the plurality of first intervals;
using a first portion of the target sound data determined based on the first interval and a second portion of the target sound data determined based on the second interval to generate the target sound data and a calculating unit for calculating volume information indicating breathing volume.
1-2. 1-1. In the processing device according to
The calculation unit
calculating a first signal strength that is the strength of the first portion of the target sound data and a second signal strength that is the strength of the second portion of the target sound data;
A processing device for calculating the volume information of the target sound data using the first signal strength and the second signal strength.
1-3. 1-2. In the processing device according to
The calculation unit is a processing device that calculates information specifying a ratio of the first signal strength to the second signal strength as the volume information of the target sound data.
1-4. 1-1. to 1-3. In the processing device according to any one of
further comprising a filtering unit that performs filtering on at least the target sound data,
The calculation unit calculates the volume information using the first part and the second part of the target sound data after the filtering process,
The filter processing unit performs band-pass filtering with f _L1 [Hz] as the cutoff frequency on the low frequency side and f _H1 [Hz] as the cutoff frequency on the high frequency side,
50≦f _L1 ≦150 holds,
A processor satisfying 500≦f _H1 ≦1500.
1-5. 1-1. to 1-4. In the processing device according to any one of
The acquisition unit acquires a plurality of sound data representing sounds detected by a plurality of sensors,
The section specifying unit generates at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section based on the first sound data,
The calculation unit uses at least one of the first time information and the second time information to identify the first part and the second part of the target sound data,
The processing device, wherein the target sound data includes at least the second sound data different from the first sound data.
1-6. 1-5. In the processing device according to
the first sound data indicates a sound detected by the first sensor provided at a first position on or inside a human body;
the second sound data indicates a sound detected by a second sensor provided at a second position on or inside the human body;
The processing device, wherein the first position is located at the neck or the first position is closer to the neck than the second position.
1-7. 1-1. to 1-4. In the processing device according to any one of
The acquisition unit acquires a plurality of sound data representing sounds detected by a plurality of sensors,
The section identification unit
identifying the first section and the second section in each of the two or more sound data;
generating third time information indicating a time range defined as the first section in all of the two or more sound data or a time range defined as the second section in all of the two or more sound data;
The calculation unit is a processing device that identifies the first portion and the second portion of the target sound data using the third time information.
1-8. 1-5. to 1-7. In the processing device according to any one of
The calculation unit calculates the volume information of the plurality of target sound data,
The processing device further comprising an estimating unit that estimates the position of the sound source of the abnormal breath sound based on the calculated plurality of pieces of volume information.
1-9. 1-1. The processing device according to any one of 1 to 8;
a sensor;
The acquisition unit acquires the sound data representing the sound detected by the sensor.
2-1. an acquiring step of acquiring one or more sound data including breath sounds;
an interval identification step of identifying at least one of a first interval in which breathing is estimated and a second interval between the plurality of first intervals;
using a first portion of the target sound data determined based on the first interval and a second portion of the target sound data determined based on the second interval to generate the target sound data , and a calculating step of calculating volume information indicative of breathing volume.
2-2. 2-1. In the processing method described in
In the calculation step,
calculating a first signal strength that is the strength of the first portion of the target sound data and a second signal strength that is the strength of the second portion of the target sound data;
A processing method for calculating the volume information of the target sound data using the first signal strength and the second signal strength.
2-3. 2-2. In the processing method described in
In the calculating step, information specifying a ratio of the first signal strength to the second signal strength is calculated as the volume information of the target sound data.
2-4. 2-1. to 2-3. In the processing method according to any one of
further comprising a filtering step of filtering at least the target sound data;
In the calculating step, the volume information is calculated using the first portion and the second portion of the filtered sound data of interest;
In the filtering step, band-pass filtering is performed with a cutoff frequency on the low frequency side of f _L1 [Hz] and a cutoff frequency on the high frequency side of f _H1 [Hz],
50≦f _L1 ≦150 holds,
A processing method that satisfies 500≦f _H1 ≦1500.
2-5. 2-1. to 2-4. In the processing method according to any one of
In the acquiring step, acquiring a plurality of sound data representing sounds detected by a plurality of sensors;
In the section specifying step, at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section is generated based on the first sound data,
In the calculating step, using at least one of the first time information and the second time information, the first portion and the second portion of the target sound data are specified;
A processing method in which the target sound data includes at least a second sound data different from the first sound data.
2-6. 2-5. In the processing method described in
the first sound data indicates a sound detected by the first sensor provided at a first position on or inside a human body;
the second sound data indicates a sound detected by a second sensor provided at a second position on or inside the human body;
The method of claim 1, wherein the first location is located at the neck or the first location is closer to the neck than the second location.
2-7. 2-1. to 2-4. In the processing method according to any one of
In the acquiring step, acquiring a plurality of sound data representing sounds detected by a plurality of sensors;
In the section identification step,
identifying the first section and the second section in each of the two or more sound data;
generating third time information indicating a time range defined as the first section in all of the two or more sound data or a time range defined as the second section in all of the two or more sound data;
In the calculating step, the processing method specifies the first portion and the second portion of the target sound data using the third time information.
2-8. 2-5. to 2-7. In the processing method according to any one of
In the calculating step, calculating the volume information of the plurality of target sound data,
The processing method further comprising an estimating step of estimating the position of the sound source of the abnormal breath sound based on the calculated plurality of volume information.
3-1. 2-1. to 2-8. A program that causes a computer to execute each step of the processing method according to any one of .
4-1. an acquisition unit that acquires sound data including breath sounds;
At least one of a first section in the sound data in which breathing is estimated and a second section between the plurality of the first sections is determined using a threshold for a value indicating the amplitude of the sound data. and an interval identification unit to identify,
The value indicating the amplitude is a value indicating the magnitude of vibration at each time of the sound data,
The section specifying unit is a processing device that determines the threshold value based on the sound data.
4-2. 4-1. In the processing device according to
The section identification unit
Finding the mode of the value indicating the amplitude in the sound data,
Determining the threshold greater than the mode,
A first process of identifying, as the first section, a section in the sound data in which at least the value indicating the amplitude exceeds the threshold, and identifying a section in which at least the value indicating the amplitude is less than the threshold as the first section A processing device that performs at least one of the second processes specified as two sections.
4-3. 4-2. In the processing device according to
When the sound data is expressed in a graph in which the horizontal axis represents the amplitude and the vertical axis represents the number of occurrences, the threshold is one or more values representing the amplitude that take the minimum value, and the mode value is is the value indicative of said amplitude closest to the processor.
4-4. 4-3. In the processing device according to
The processing device, wherein, when the sound data is represented on the graph, the mode value is a value indicating the smallest amplitude among a plurality of values indicating the amplitude having a maximum value.
4-5. 4-2. to 4-4. In the processing device according to any one of
The section identifying unit obtains the mode based on the sound data after filtering,
The filtering process is a band-pass filtering process with a cutoff frequency on the low frequency side of f _L2 [Hz] and a cutoff frequency on the high frequency side of f _H2 [Hz],
150≦f _L2 ≦250 holds,
A processor where 550≦f _H2 ≦650.
4-6. 4-2. to 4-5. In the processing device according to any one of
The section identifying unit is a processing device that obtains the mode based on data obtained by at least down-sampling the sound data.
4-7. 4-1. to 4-6. In the processing device according to any one of
The acquisition unit acquires a plurality of sound data representing sounds detected by a plurality of sensors,
The section identification unit
identifying at least one of the first section and the second section in the first sound data included in the plurality of sound data;
generating at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section;
the first time interval and the second interval of the second sound data included in the plurality of sound data and different from the first sound data; A processing device specified based on at least one of time information.
4-8. 4-7. In the processing device according to
the first sound data indicates a sound detected by the first sensor provided at a first position on or inside a human body;
the second sound data indicates a sound detected by a second sensor provided at a second position on or inside the human body;
The processing device, wherein the first position is located at the neck or the first position is closer to the neck than the second position.
4-9. 4-1. to 4-6. In the processing device according to any one of
The acquisition unit acquires a plurality of sound data representing sounds detected by a plurality of sensors,
The section identification unit
identifying the first section and the second section in each of the two or more sound data;
A process of generating third time information indicating the time range identified as the first section in all of the two or more sound data or the time range identified as the second section in all of the two or more sound data Device.
4-10. 4-1. to 4-9. A processing device according to any one of
a sensor;
The acquisition unit acquires the sound data representing the sound detected by the sensor.
5-1. an acquisition step of acquiring sound data including breath sounds;
At least one of a first section in the sound data in which breathing is estimated and a second section between the plurality of the first sections is determined using a threshold for a value indicating the amplitude of the sound data. and a section identification step to identify,
The value indicating the amplitude is a value indicating the magnitude of vibration at each time of the sound data,
The processing method, wherein, in the section identifying step, the threshold is determined based on the sound data.
5-2. 5-1. In the processing method described in
In the section identification step,
Finding the mode of the value indicating the amplitude in the sound data,
Determining the threshold greater than the mode,
A first process of identifying, as the first section, a section in the sound data in which at least the value indicating the amplitude exceeds the threshold, and identifying a section in which at least the value indicating the amplitude is less than the threshold as the first section A processing method for performing at least one of the second processing for specifying two sections.
5-3. 5-2. In the processing method described in
When the sound data is expressed in a graph in which the horizontal axis represents the amplitude and the vertical axis represents the number of occurrences, the threshold is one or more values representing the amplitude that take the minimum value, and the mode value is A processing method that is the value that indicates the amplitude that is closest to .
5-4. 5-3. In the processing method described in
The processing method, wherein, when the sound data is represented on the graph, the mode value is a value indicating the smallest amplitude among a plurality of values indicating the amplitude having a maximum value.
5-5. 5-2. to 5-4. In the processing method according to any one of
In the section identification step, the mode is obtained based on the sound data after filtering,
The filtering process is a band-pass filtering process with a cutoff frequency on the low frequency side of f _L2 [Hz] and a cutoff frequency on the high frequency side of f _H2 [Hz],
150≦f _L2 ≦250 holds,
A processing method that satisfies 550≦f _H2 ≦650.
5-6. 5-2. to 5-5. In the processing method according to any one of
In the section identifying step, the processing method for obtaining the mode based on data obtained by at least down-sampling the sound data.
5-7. 5-1. to 5-6. In the processing method according to any one of
In the acquiring step, acquiring a plurality of sound data representing sounds detected by a plurality of sensors;
In the section identification step,
identifying at least one of the first section and the second section in the first sound data included in the plurality of sound data;
generating at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section;
the first time interval and the second interval of the second sound data included in the plurality of sound data and different from the first sound data; A processing method for specifying based on at least one of time information.
5-8. 5-7. In the processing method described in
the first sound data indicates a sound detected by the first sensor provided at a first position on or inside a human body;
the second sound data indicates a sound detected by a second sensor provided at a second position on or inside the human body;
The method of claim 1, wherein the first location is located at the neck or the first location is closer to the neck than the second location.
5-9. 5-1. to 5-6. In the processing method according to any one of
In the acquiring step, acquiring a plurality of sound data representing sounds detected by a plurality of sensors;
In the section identification step,
identifying the first section and the second section in each of the two or more sound data;
A process of generating third time information indicating the time range identified as the first section in all of the two or more sound data or the time range identified as the second section in all of the two or more sound data Method.
6-1. 5-1. to 5-9. A program that causes a computer to execute each step of the processing method according to any one of .

10 Processing Unit 20 System 110 Acquisition Unit 120 Filter Processing Unit 130 Section Identification Unit 150 Calculation Unit 170 Estimation Unit 210 Sensor 1000 Computer 1020 Bus 1040 Processor 1060 Memory 1080 Storage Device 1100 Input/Output Interface 1120 Network Interface

Claims (11)

an acquisition unit that acquires one or more sound data including breath sounds;
an interval identification unit that identifies at least one of a first interval in which breathing is estimated and a second interval between the plurality of first intervals;
using a first portion of the target sound data determined based on the first interval and a second portion of the target sound data determined based on the second interval to generate the target sound data and a calculating unit for calculating volume information indicating breathing volume. The processing apparatus of claim 1,
The calculation unit
calculating a first signal strength that is the strength of the first portion of the target sound data and a second signal strength that is the strength of the second portion of the target sound data;
A processing device for calculating the volume information of the target sound data using the first signal strength and the second signal strength. The processing apparatus according to claim 2,
The calculation unit is a processing device that calculates information specifying a ratio of the first signal strength to the second signal strength as the volume information of the target sound data. In the processing apparatus according to any one of claims 1 to 3,
further comprising a filtering unit that performs filtering on at least the target sound data,
The calculation unit calculates the volume information using the first part and the second part of the target sound data after the filtering process,
The filter processing unit performs band-pass filtering with a cutoff frequency on the low frequency side of f _L1 [Hz] and a cutoff frequency on the high frequency side of f _H1 [Hz],
50≦f _L1 ≦150 holds,
A processor satisfying 500≦f _H1 ≦1500. In the processing apparatus according to any one of claims 1 to 4,
The acquisition unit acquires a plurality of sound data representing sounds detected by a plurality of sensors,
The section specifying unit generates at least one of first time information indicating the time range of the first section and second time information indicating the time range of the second section based on the first sound data,
The calculation unit uses at least one of the first time information and the second time information to identify the first part and the second part of the target sound data,
The processing device, wherein the target sound data includes at least the second sound data different from the first sound data. In the processing apparatus according to claim 5,
the first sound data indicates a sound detected by the first sensor provided at a first position on or inside a human body;
the second sound data indicates a sound detected by a second sensor provided at a second position on or inside the human body;
The processing device, wherein the first position is located at the neck or the first position is closer to the neck than the second position. In the processing apparatus according to any one of claims 1 to 4,
The acquisition unit acquires a plurality of sound data representing sounds detected by a plurality of sensors,
The section identification unit
identifying the first section and the second section in each of the two or more sound data;
generating third time information indicating a time range defined as the first section in all of the two or more sound data or a time range defined as the second section in all of the two or more sound data;
The calculation unit is a processing device that identifies the first portion and the second portion of the target sound data using the third time information. In the processing apparatus according to any one of claims 5 to 7,
The calculation unit calculates the volume information of the plurality of target sound data,
The processing device further comprising an estimating unit that estimates the position of the sound source of the abnormal breath sound based on the calculated plurality of pieces of volume information. A processing apparatus according to claim 1;
a sensor;
The acquisition unit acquires the sound data representing the sound detected by the sensor. an acquiring step of acquiring one or more sound data including breath sounds;
an interval identification step of identifying at least one of a first interval in which breathing is estimated and a second interval between the plurality of first intervals;
using a first portion of the target sound data determined based on the first interval and a second portion of the target sound data determined based on the second interval to generate the target sound data , and a calculating step of calculating volume information indicative of breathing volume. A program that causes a computer to execute each step of the processing method according to claim 10 .

Images