Respiration rate counter, counting method, computer program, and recording medium

Abstract

To count a breathing rate properly.SOLUTION: A breathing rate counting device (10) includes: acquisition means (11) for acquiring biological sound information including a breathing sound; generation means (12) for generating a frequency spectrum from the biological sound information; and calculation means (12) for calculating a breathing rate by specifying a peak corresponding to breathing from the frequency spectrum on the basis of an intensity threshold that is determined in accordance with one or more peak frequencies included in the frequency spectrum.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the technical field of a respiratory rate counter, a counting method, a computer program, and a recording medium for counting the respiratory rate of a living body. [Background technology]

[0002] As an example of this type of device, a device has been proposed that determines a subject's acoustic heart rate and acoustic respiratory rate from acoustic signals acquired by body resonation attached to the tip of a rod-shaped body that can be inserted under the subject's armpit (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-126511 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in Patent Document 1 mentioned above has a technical problem in that when detecting respiratory rate, the subject's respiratory rate must be confirmed in advance and parameters must be adjusted, making it difficult to use in clinical settings where the patient's condition is constantly changing.

[0005] The present invention has been made in consideration of the above-mentioned problems, and an object of the present invention is to provide a respiration rate counter, a counting method, a computer program, and a recording medium that can appropriately count the respiration rate. [Means for solving the problem]

[0006] The respiration rate counter according to the present invention includes an acquisition unit for acquiring body sound information including respiratory sounds, a generation unit for generating a frequency spectrum from the body sound information, and a frequency spectrum including a respiratory sound. Complex Peak in numbers Bandwidth whereand a calculation means for calculating a respiration rate by identifying a peak corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to the frequency of the signal.

[0007] The counting method according to the present invention is a counting method for a respiration rate counter, and includes an acquisition step in which the respiration rate counter acquires body sound information including respiratory sounds, a generation step in which the respiration rate counter generates a frequency spectrum from the body sound information, and a generation step in which the respiration rate counter generates a frequency spectrum including respiratory sounds. Complex Peak in numbers Bandwidth where and a calculating step of calculating a respiration rate by identifying a peak corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to the frequency of the signal.

[0008] The computer program according to the present invention is a program for causing a computer included in a respiration rate counter to perform a respiration rate counter operation, the program comprising: an acquisition unit for acquiring body sound information including respiratory sounds; a generation unit for generating a frequency spectrum from the body sound information; and a frequency spectrum including respiratory sounds. Complex Peak in numbers Bandwidth where The frequency spectrum is determined based on an intensity threshold value determined in accordance with the frequency of the signal, and a peak corresponding to respiration is identified from the frequency spectrum, thereby calculating the respiration rate.

[0009] A recording medium according to the present invention has the above-described computer program according to the present invention recorded thereon.

[0010] The functions and other advantages of the present invention will become apparent from the following detailed description of the preferred embodiments. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing the configuration of a respiration rate counter according to an embodiment. [Figure 2] FIG. 10 is a diagram illustrating an example of a frequency spectrum. [Figure 3] 4 is a flowchart showing the operation of the respiration rate counter according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] An embodiment of the present invention will be described.

[0013] (Embodiments of a respiration rate counter) The respiration rate counter according to the embodiment includes an acquiring unit, a generating unit, and a calculating unit. The acquiring unit acquires body sound information including breathing sounds of a living body. The acquiring unit may be configured with an acoustic device such as a microphone, or may be configured with an electronic device such as a computer that can receive an electrical signal output from the acoustic device.

[0014] The generating means generates a frequency spectrum relating to frequencies contained in sound waves representing body sounds such as breathing sounds from the body sound information. The calculating means determines an intensity threshold according to the frequencies of one or more peaks contained in the frequency spectrum. The calculating means identifies peaks corresponding to breathing from the frequency spectrum based on the intensity threshold, and calculates the respiratory rate.

[0015] Living organisms breathe in two ways: normal breathing patterns (for example, about 12 to 30 breaths per minute in the case of an adult) and abnormal breathing patterns, such as bradypnea, which has a lower breathing rate than normal breathing patterns, and tachypnea, which has a higher breathing rate than normal breathing patterns. Research by the inventors of the present application has revealed that the noise level of the frequency spectrum changes relatively significantly depending on the breathing pattern. Therefore, unless some countermeasure is taken, there is a possibility that an incorrect breathing rate will be calculated due to the influence of noise.

[0016] Therefore, the inventors of the present application have focused on the fact that the frequency (or frequency band) at which peaks appear in a frequency spectrum varies depending on the breathing pattern, and have configured a calculation means to determine an intensity threshold according to the frequency of one or more peaks. By determining the intensity threshold, it is possible to appropriately identify peaks corresponding to breathing while suppressing the influence of noise that varies depending on the breathing pattern. Therefore, the respiration rate counter can appropriately count the respiration rate.

[0017] The intensity threshold may be determined according to the frequency of the largest peak included in the frequency spectrum. Alternatively, the intensity threshold may be determined according to the frequency of a band in which multiple peaks included in the frequency spectrum are concentrated. The band in which multiple peaks are concentrated may be identified, for example, as follows. That is, based on the number of peaks included in each of multiple predefined frequency bands, the frequency band containing the largest number of peaks may be determined to be the band in which multiple peaks are concentrated.

[0018] In one aspect of the respiration counting device, the generating means may generate a frequency spectrum in the following manner. That is, the generating means first extracts sound components corresponding to a predetermined frequency band from body sound information, and generates a power spectrum indicating a time change in the power of the sound components. The generating means then smooths the power spectrum to generate a smoothed power spectrum. The generating means then differentiates the smoothed power spectrum to generate a differentiated spectrum. The generating means then frequency converts the differentiated spectrum to generate a frequency spectrum. This configuration can reduce the influence of noise in the process of generating a frequency spectrum from body sound information, which is extremely advantageous in practice.

[0019] The "predetermined frequency band" is, for example, 400 to 600 Hz, and may be set as a frequency band that can contain sounds caused by breathing of a living body.

[0020] (Embodiment of Counting Method) A counting method according to the embodiment is a counting method for a respiration rate counter. The counting method includes an acquiring step, a generating step, and a calculating step. In the acquiring step, the respiration rate counter acquires body sound information including respiratory sounds. In the generating step, the respiration rate counter generates a frequency spectrum from the body sound information. In the calculating step, the respiration rate counter identifies peaks corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to the frequency of one or more peaks included in the frequency spectrum, and calculates the respiration rate. According to the counting method, the respiration rate can be appropriately counted, similar to the respiration rate counter according to the above-described embodiment.

[0021] (Embodiment of Computer Program) The computer program according to the embodiment causes a computer included in the respiration rate counting device to function as an acquisition means for acquiring body sound information including respiratory sounds, a generation means for generating a frequency spectrum from the body sound information, and a calculation means for calculating the respiration rate by identifying peaks corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to the frequency of one or more peaks included in the frequency spectrum.

[0022] According to the computer program, the respiration rate counter according to the above-described embodiment can be realized relatively easily by loading the computer program from a recording medium such as a RAM (Random Access Memory), a CD-ROM (Compact Disc Read Only Memory), or a DVD-ROM (DVD Read Only Memory) that stores the computer program into a computer constituting the respiration rate counter and executing the computer program, or by downloading the computer program via a communication means and then executing the computer program. This makes it possible to appropriately count the respiration rate in the same way as the respiration rate counter according to the above-described embodiment.

[0023] A recording medium, such as a CD-ROM or DVD-ROM, on which the computer program is recorded corresponds to an embodiment of the recording medium. [Example]

[0024] An embodiment of a respiration rate counter will be described with reference to Figures 1 to 3. In Figure 1, a respiration rate counter 10 has an acquisition unit 11, a processing unit 12, and an output unit 13, which are logically implemented within the device as a logic block or physically implemented as a processing circuit.

[0025] The acquisition unit 11 acquires body sound information output from the sensor 20. The sensor 20 will now be described. The sensor 20 is, for example, a microphone having a vibration sensor. The sensor 20 may be attached to the neck of a subject (i.e., a living body). In this case, the sensor 20 mainly detects sounds made when the subject breathes, and generates body sound information including breathing sounds. The acquisition unit 11 corresponds to an example of the "acquisition means" according to the above-described embodiment.

[0026] The processing unit 12 generates a frequency spectrum from the body sound information acquired by the acquisition unit 11. The processing unit 12 identifies peaks corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to the frequency of one or more peaks included in the generated frequency spectrum, and calculates the respiration rate.

[0027] A method for generating a frequency spectrum will now be described in detail. The body sound information may be, for example, information indicating a change over time in an output value (such as a voltage value or a current value) of the sensor 20. In this case, the body sound information can be represented as a graph in which the horizontal axis indicates time and the vertical axis indicates the output value. "Time" may be rephrased as "number of frames."

[0028] The processing unit 12 performs, for example, a Fourier transform on such body sound information to generate a plurality of pieces of frequency data indicating the relationship between frequency and power for each unit time (or for each frame). The processing unit 12 extracts data in a predetermined frequency band (for example, 400 Hz to 600 Hz) from each of the plurality of frequency data. The processing unit 12 calculates the average value of power in the predetermined frequency band based on the extracted data. The processing unit 12 generates a power spectrum indicating the time change of the average power by arranging in time series the average power values ​​calculated based on the plurality of data respectively extracted from the plurality of frequency data.

[0029] The processing unit 12 performs a smoothing process on the power spectrum to generate a smoothed power spectrum. The processing unit 12 performs a differentiation process on the smoothed power spectrum to generate a differentiated spectrum. Note that various existing aspects can be applied to the smoothing process and differentiation process, and therefore detailed explanations thereof will be omitted.

[0030] The processing unit 12 performs frequency transformation processing, such as Fourier transform, on the differential spectrum to generate a frequency spectrum that indicates the relationship between frequency and intensity. Examples of the frequency spectrum are shown in Fig. 2. Fig. 2(a) is an example of a frequency spectrum corresponding to bradypnea. Fig. 2(b) is an example of a frequency spectrum corresponding to normal breathing. Fig. 2(c) is an example of a frequency spectrum corresponding to tachypnea.

[0031] The frequency spectrum corresponding to bradypnea (see FIG. 2(a)) has a relatively small number of noise components (i.e., a relatively large S / N ratio). On the other hand, the frequency spectrum corresponding to normal breathing (see FIG. 2(b)) and the frequency spectrum corresponding to tachypnea (see FIG. 2(c)) have a relatively large number of noise components (i.e., a relatively small S / N ratio). In other words, the noise components (or noise level) of the frequency spectrum change depending on the breathing pattern. Therefore, unless any measures are taken, the influence of noise may result in an incorrect calculation of the respiratory rate.

[0032] Therefore, the processing unit 12 performs a peak search on the frequency spectrum. Since various existing methods can be applied to the peak search method, detailed explanations thereof will be omitted. Next, the processing unit 12 determines an intensity threshold value according to, for example, the frequency of the peak with the greatest intensity among one or more peaks detected by the peak search, or according to, for example, the frequency of a band in which multiple peaks are concentrated.

[0033] The "intensity threshold" is a value used to determine whether a frequency spectrum contains a noise component. The "intensity threshold" is a variable value that changes, for example, depending on the frequency of the peak with the maximum intensity, or depending on the frequency of a band in which multiple peaks are concentrated. Such an "intensity threshold" can be set experimentally, empirically, or by simulation, by determining, for example, the relationship between the frequency of the peak with the maximum intensity in the frequency spectrum and the noise level, or by determining, for example, the relationship between the frequency of the band in which multiple peaks are concentrated in the frequency spectrum and the noise level, and then setting the value as a value obtained by adding a predetermined margin to the maximum noise level based on the determined relationship.

[0034] For example, if the frequency of the peak with the highest intensity in the frequency spectrum or the frequency of the band where multiple peaks are concentrated is less than 10, the processing unit 12 determines the intensity threshold to be "Threshold 1." For example, if the frequency of the peak with the highest intensity in the frequency spectrum or the frequency of the band where multiple peaks are concentrated is greater than 18, the processing unit 12 determines the intensity threshold to be "Threshold 3" which is greater than "Threshold 1." For example, if the frequency of the peak with the highest intensity in the frequency spectrum or the frequency of the band where multiple peaks are concentrated is greater than 10 and less than 18, the processing unit 12 determines the intensity threshold to be "Threshold 2" which is greater than "Threshold 1" and less than "Threshold 3."

[0035] For example, the processing unit 12 may identify the frequency band containing the most peaks as the band in which the multiple peaks are concentrated, based on the number of peaks contained in each of a plurality of predefined frequency bands.

[0036] Based on the intensity threshold determined as described above, the processing unit 12 removes, as noise, peaks smaller than the intensity threshold from one or more peaks detected by, for example, peak search. The processing unit 12 identifies peaks corresponding to breathing from the peaks that have not been removed from the one or more peaks. In each of Figures 2(a) to 2(c), the peaks marked with black circles are examples of peaks corresponding to breathing.

[0037] The processing unit 12 calculates the respiratory rate based on the frequency of the identified peak. Note that various existing methods can be applied to the method of identifying the peak corresponding to respiration, and therefore detailed explanations thereof will be omitted. The output unit 13 outputs the respiratory rate calculated by the processing unit 12 to the display device 30. As a result, the respiratory rate is displayed on the screen of the display device 30. Note that the processing unit 12 corresponds to an example of the "generating means" and "calculating means" according to the above-mentioned embodiment.

[0038] The operation of the respiration rate counter 10 will be further described with reference to the flowchart in Fig. 3. In Fig. 3, the processing unit 12 performs, for example, a Fourier transform on the body sound information to generate a plurality of frequency data items indicating the relationship between frequency and power per unit time. That is, the processing unit 12 performs a spectrum conversion process (step S101). The processing unit 12 extracts data in a predetermined frequency band (for example, 400 Hz to 600 Hz) from each of the plurality of frequency data items (step S102).

[0039] The processing unit 12 calculates the average power value of the predetermined frequency band based on the extracted data. The processing unit 12 generates a power spectrum indicating the time change of the average power value by arranging in time series the average power values ​​calculated based on the plurality of data respectively extracted from the plurality of frequency data. The processing unit 12 performs a smoothing process on the power spectrum to generate a smoothed power spectrum (step S103).

[0040] The processing unit 12 performs differentiation on the smoothed power spectrum to generate a differential spectrum (step S104), and performs frequency transformation, such as Fourier transform, on the differential spectrum to generate a frequency spectrum indicating the relationship between frequency and intensity (step S105).

[0041] The processing unit 12 performs a peak search on the frequency spectrum (step S106). The processing unit 12 determines an intensity threshold based on, for example, the frequency of the peak with the greatest intensity among one or more peaks detected by the peak search, or based on, for example, the frequency of a band in which multiple peaks are concentrated. The processing unit 12 identifies a peak corresponding to respiration from the frequency spectrum based on the determined intensity threshold. The processing unit 12 calculates the respiration rate based on the frequency of the identified peak (step S107).

[0042] (Technical Effects) In the respiration rate counter 10, an intensity threshold is determined in accordance with, for example, the frequency of the peak with the greatest intensity in the frequency spectrum, or, for example, the frequency of a band in which multiple peaks are concentrated. Then, a peak corresponding to respiration is identified from the frequency spectrum based on the intensity threshold. Therefore, the respiration rate counter 10 can appropriately identify the peak corresponding to respiration while suppressing the influence of noise that varies depending on the breathing pattern. Therefore, the respiration rate counter 10 can appropriately count the respiration rate.

[0043] In the respiration rate counter 10, the intensity threshold is automatically determined by the processing unit 12. Therefore, the time required to calculate the respiration rate can be shortened compared to a comparative example in which, for example, a user observes the waveform of a frequency spectrum and determines a threshold equivalent to the intensity threshold. In other words, the respiration rate counter 10 can calculate the respiration rate of a living body in real time. Therefore, the respiration rate counter 10 can be applied to, for example, monitoring the respiration rate of a patient (i.e., a living body) in a medical setting.

[0044] In the respiration counting device 10, both inhalation and exhalation are counted as one breath. However, how one breath is defined can be arbitrary. The loudness of the sound associated with inhalation and the loudness of the sound associated with exhalation may be similar or may be significantly different. When determining the intensity threshold, the processing unit 12 may take into consideration, for example, the frequency of the peak with the highest intensity in the frequency spectrum or the frequency band where multiple peaks are concentrated, as well as the relationship between the loudness of the sound associated with inhalation and the loudness of the sound associated with exhalation.

[0045] As described above, the respiration rate counter 10 calculates the respiration rate from the frequency of the peak corresponding to respiration identified from the frequency spectrum based on the intensity threshold. In order to calculate the respiration rate with relatively high reliability, the respiration rate counter 10 typically outputs the first respiration rate after a predetermined time (e.g., several tens of seconds) has elapsed since the sensor 20 started to detect the breathing sounds of the living body, i.e., after the sensor 20 has outputted body sound information for a predetermined period of time.

[0046] When the body sound information is, for example, information indicating a time change in the output value (e.g., voltage value, current value, etc.) of the sensor 20, the fluctuation period of the output value of the sensor 20 changes depending on the breathing pattern. For example, in the case of tachypnea, the fluctuation period is relatively short. On the other hand, for example, in the case of bradypnea, the fluctuation period is relatively long. Therefore, the respiration rate counter 10 may be configured to output the first respiratory rate (i.e., calculate the first respiratory rate) when the number of times the output value of the sensor 20 indicated by the body sound information exceeds a predetermined value reaches a predetermined number. With this configuration, for example, in the case of tachypnea, the time required from when the sensor 20 starts detecting respiratory sounds of the living body until the first respiratory rate is output can be shortened.

[0047] The present invention is not limited to the above-described embodiments, and can be modified as appropriate within the scope of the claims and the gist or idea of ​​the invention as can be read from the entire specification, and respiration rate counting devices, counting methods, computer programs, and recording media that involve such modifications are also included in the technical scope of the present invention. [Explanation of symbols]

[0048] 10... respiratory rate counter, 11... acquisition unit, 12... processing unit, 13... output unit, 20... sensor, 30... display device

Claims

1. an acquisition means for acquiring body sound information including respiratory sounds; a generation means for generating a frequency spectrum from the body sound information; a calculation means for identifying a peak corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to a frequency band in which a plurality of peaks included in the frequency spectrum are concentrated, and calculating a respiration rate; A respiratory rate counting device comprising:

2. 2. The respiration rate counter according to claim 1, wherein the generating means (i) extracts sound components corresponding to a predetermined frequency band from the body sound information and generates a power spectrum indicating a time change in the power of the sound components, (ii) performs a smoothing process on the power spectrum to generate a smoothed power spectrum, (iii) performs a differentiation process on the smoothed power spectrum to generate a differential spectrum, and (iv) performs a frequency conversion process on the differential spectrum to generate the frequency spectrum.

3. A counting method for a respiration counting device, comprising: an acquiring step of acquiring body sound information including respiratory sounds by the respiratory rate counter; a generating step in which the respiration rate counter generates a frequency spectrum from the body sound information; a calculation step in which the respiration rate counter identifies a peak corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to a frequency band in which a plurality of peaks included in the frequency spectrum are concentrated, and calculates the respiration rate; A counting method comprising:

4. A computer equipped in the respiration rate counter, an acquisition means for acquiring body sound information including respiratory sounds; a generation means for generating a frequency spectrum from the body sound information; a calculation means for identifying a peak corresponding to respiration from the frequency spectrum based on an intensity threshold determined according to a frequency band in which a plurality of peaks included in the frequency spectrum are concentrated, and calculating a respiration rate; A computer program characterized by causing the computer to function as follows.

5. A recording medium on which the computer program according to claim 4 is recorded.

Images