Wheezing detection device

Abstract

Wheezing detection device comprising: a breath sound detection unit (111, 112) configured to detect a breath sound of a measurement subject and to obtain a breath sound signal in a time series expressing the breath sound; and a determination processing unit (115) configured to convert the respiratory sound signal into a frequency domain in each predetermined processing unit period to obtain a frequency spectrum of the respiratory sound and to determine, based on a height and a width of a peak in the frequency spectrum, whether the peak indicates wheezing or not, wherein the determination processing unit (115) obtains a ratio between the height and the width of the peak and determines whether or not the wheezing is displayed based on whether the ratio is greater than a predetermined first threshold value, wherein the determination processing unit (115) determines whether or not the wheezing is displayed based only on a dominant peak having the largest area among a plurality of peaks in the frequency spectrum in a graph of frequency with respect to sound pressure.

Description

Technical area

[0001] The present invention relates to a wheezing detection device, and more particularly, to a wheezing detection device that detects whether or not wheezing is included in a breath sound of a measurement subject. Background technology

[0002] For example, US 2011 / 0125044 A1 (Patent Document 1) discloses an automated system for monitoring respiratory diseases, such as asthma. Based on data from a microphone and an accelerometer, the system provides a summary of data and an alert when the severity of symptoms reaches a threshold. Specifically, for wheezing, peaks of a frequency spectrum in a frequency range of approximately 200 to 800 Hz are measured. The peaks of the frequency spectrum and a predetermined value associated with the wheezing and stored in memory are compared, and the result of the comparison is used as an element for determining the severity.

[0003] US 6 168 568 B1 discloses a phonopneumograph system for analyzing respiratory sounds, comprising a plurality of respiratory sensors mounted around the respiratory system of a patient to measure respiratory activities, and a respiratory analyzer.

[0004] US 2011 / 0230777 A1 discloses lightweight methods and systems for wheezing detection for wearable respiratory health monitoring devices by using an algorithm that calculates a partial STFT image of a respiratory signal that includes all data points required for wheezing detection but excludes many data points unnecessary for wheezing detection.

[0005] US 2008 / 30636 A1 discloses a method for detecting and monitoring body sounds in humans and animals, using bioacoustic sensors and downstream analyzers for stationary or mobile long-term monitoring of the respiration of intensive care patients. Summary of the inventionTechnical problem

[0006] However, since the system described above only compares the magnitudes of the peaks of the frequency spectrum and the reference value (value stored in the memory) for the wheeze, there is a problem that the detection accuracy of the wheeze is not good.

[0007] In view of the above, the present invention aims to provide a wheezing detection device that can accurately detect whether or not wheezing is included in a breath sound of a measurement subject. Solution to the problem

[0008] The present invention is defined by independent claims 1 and 7; the dependent claims describe embodiments of the invention.

[0009] To solve the problem described above, the wheezing detection device of the present invention comprises: a breath sound detection unit configured to detect a breath sound of a measurement subject and obtain a breath sound signal in a time series expressing the breath sound; and a determination processing unit configured to convert the breath sound signal into a frequency domain at each predetermined processing unit time period to obtain a frequency spectrum of the breath sound, and to determine whether the peak indicates wheezing or not based on a height and a width of a peak in the frequency spectrum.

[0010] Here, the "peaks" in the frequency spectrum described above refer to peaks in noise intensity (sound pressure). The "height" and "width" of a peak in the frequency spectrum refer to a peak in a graph of frequency versus sound pressure. Also, if background noise is present in the graph of frequency versus sound pressure, the "height" and "width" refer to the essential "height" and "width" of the peak from which the background noise has been removed.

[0011] In the analysis conducted by the inventor, a wheezing sound is characterized by the relatively narrow (almost monotonous) peak widths in its frequency spectrum. Also, a wheezing sound is characterized by the fact that it includes multiple peaks with relatively narrow widths. Therefore, to accurately detect wheezing, the peak widths in the frequency spectrum should be somehow incorporated into the determination. Here, the breath sound detection unit in the wheezing detection device of the present invention detects the breath sound of the measurement subject and obtains the breath sound signal in a time series that expresses the breath sound.At each predetermined processing unit time period, the determination processing unit converts the breath sound signal into a frequency domain to obtain a frequency spectrum of the breath sound. Based on the heights and widths of the peaks in the frequency spectrum, it determines whether the peaks indicate wheezing or not. As a result, it is possible to accurately determine whether the breath sound of the test subject contains wheezing or not.

[0012] With the wheezing detection device according to an embodiment, the determination processing unit obtains a ratio between the height and the width of the peak and determines whether or not the wheezing is displayed based on whether or not the ratio is greater than a predetermined first threshold.

[0013] With the wheezing detection device according to the embodiment, the determination processing unit obtains a ratio between the height and width of the peak and determines whether wheezing is indicated or not based on whether the ratio is greater than a predetermined first threshold. As a result, it is possible to more accurately detect whether wheezing is included in the breath sound of the measurement subject.

[0014] With the wheezing detection device according to the invention, the determination processing unit determines whether the wheezing is displayed or not based only on a dominant peak having the largest area among a plurality of peaks in the frequency spectrum in a graph of frequency with respect to sound pressure.

[0015] Given this, the “area” on the frequency versus sound pressure graph represents the significant area from which the background noise has been removed.

[0016] A dominant peak having the largest area in the frequency vs. sound pressure graph among the plurality of peaks in the frequency spectrum corresponds to the peak with the highest energy. Therefore, the dominant peak determines whether wheezing is included in the above-described unit processing time period. In view of this, in the wheezing detection device according to the embodiment, the determination processing unit determines whether wheezing is displayed or not based only on a dominant peak having the largest area among the plurality of peaks in the frequency spectrum in a frequency vs. sound pressure graph. As a result, it is possible to more accurately detect whether wheezing is included in the breath sound of the measurement subject.

[0017] With the wheezing detection device according to an embodiment, the determination processing unit determines whether wheezing is indicated or not based only on the peak at a frequency ranging from 200 Hz to 1500 Hz in the frequency spectrum.

[0018] In the analysis conducted by the inventor, the wheezing sound frequently observed in cases of childhood asthma is a sound having peaks with relatively narrow widths (a sound that is almost monotonic) in the frequency range of approximately 900 Hz to 1200 Hz. In view of this, in the wheezing detection device according to the embodiment, the determination processing unit determines whether wheezing is indicated or not based only on a peak in a frequency range of 200 Hz to 1500 Hz in the frequency spectrum. Consequently, it is possible to detect whether wheezing, including a wheezing sound frequently observed in cases of childhood asthma, is included in the breath sound of the measurement subject, in addition to the wheezing sound. On the other hand, sounds outside the range of 200 Hz to 1500 Hz are not considered wheezing and are therefore not considered in the determination.As a result, it is possible to more accurately detect whether wheezing is included in the breath sound of the test subject or not.

[0019] A wheezing detection device according to an embodiment comprises: an addition processing unit configured to set an addition unit period including a plurality of processing unit periods, add the lengths of processing unit periods in which it was determined that the wheezing was included in each addition unit period, and obtain the result as a wheezing period, wherein the addition processing unit, in each of the processing unit periods, converts the breath sound signal into a frequency domain to obtain the frequency spectrum of the breath sound, classifies the strength of the wheezing sound into a plurality of levels based on the area of ​​the dominant peak having the largest area in the frequency-to-sound-pressure diagram among the plurality of peaks in the frequency spectrum, and adds the lengths of the processing unit periods determined to include the wheezing,added in each of the classified levels; and, a warning generation unit configured to generate a warning when a percentage of time for which a wheezing sound intensity reaches a specific level in the addition unit period based on the addition performed by the addition processing unit exceeds a predetermined second threshold.

[0020] Here, the generation of an “alert” broadly includes the generation of an alarm sound, the alarm display on a display screen, the wireless transmission of an alarm signal, and the like.

[0021] In the wheezing detection device according to the embodiment, the addition processing unit sets the addition unit period including a plurality of processing unit periods, adds the lengths of the processing unit periods in the addition unit period in which wheezing was determined to be included, and obtains the result as the wheezing period. In each of the processing unit periods, the addition processing unit converts the breath sound signal into a frequency domain to obtain the frequency spectrum of the breath sound, classifies the intensity of the wheezing sound into a plurality of levels based on the area of ​​the dominant peak that has the largest area in the frequency-to-sound pressure graph among the plurality of peaks in the frequency spectrum, and adds the lengths of the processing unit periods for which wheezing was determined to be included in each of the classified levels.If the percentage of time during the summation unit period that the wheezing intensity has reached a specific level exceeds a predefined second threshold, the alert generation unit generates an alert. This alert can be used to alert the user (typically the subject, a caregiver or support worker caring for the subject, a medical professional such as a nurse, or the like) to the fact that the subject's symptoms have worsened, and thus to take countermeasures such as administering medication to the subject. This alert is particularly beneficial in the case where the subject is a young child, a seriously ill patient, or the like who has difficulty expressing their intentions.

[0022] The wheezing detection device according to an embodiment includes a sound recording unit configured to record the breathing sound signal when the warning generating unit generates the warning.

[0023] In the wheezing detection device according to the embodiment, a sound detection unit records the breath sound signal when the warning generation unit generates a warning. Consequently, it is possible to automatically record the breath sound of the measurement subject when the wheezing is relatively strong (ie, when the wheezing is severe). Consequently, by playing the recorded content, the next time the measurement subject has a medical examination, for example, a doctor can listen to the breath sound of the measurement subject when the wheezing is severe. As a result, it is easier for the doctor to diagnose whether the measurement subject has asthma or not and the severity of the asthma, and the doctor can easily generate a treatment plan.

[0024] The wheezing detection device of one embodiment further comprises: a phase identification unit configured to identify the breathing cycle of the measurement subject by dividing it into an exhalation phase and an inhalation phase based on the breath sound signal obtained by the breath sound detection unit, a phase instruction input unit configured to input an instruction to select one or both of the exhalation phase and the inhalation phase of the breath sound signal; and a sound recording unit configured to record the phase of the breath sound signal instructed by the phase instruction input unit.

[0025] With the wheezing detection device according to the embodiment, the phase identification unit identifies the breathing cycle of the measurement subject, distinguishing between the exhalation phase and the inspiration phase based on the breath sound signal acquired by the breath sound detection unit. For example, in response to a request from the physician, the user uses the phase instruction input unit to input an instruction to select one or both of the exhalation phase and the inspiration phase of the breath sound signal. Thereby, the sound recording unit records the phase of the breath sound signal instructed by the phase instruction input unit. Consequently, when the physician listens to the recorded content of the wheezing during the next medical examination, the user can have the physician listen to the recorded content of the phase of the breathing cycle requested by the physician.

[0026] In the wheezing detection device of one embodiment, the respiratory sound detection unit comprises: a first microphone in the form of a stethoscope, which is attached to the skin of the chest of the test subject; and a second microphone attached to clothing or the skin of a part remote from the chest and respiratory organ of the test subject, and the breath sound detection unit outputs a difference obtained by subtracting the output of the second microphone from the output of the first microphone.

[0027] Here, the "attachment" of the first microphone and the second microphone includes, for example, a method of attachment via an adhesive layer. The "clothing" to which the second microphone is attached refers to clothing, such as underwear. The "part away from the subject's chest and respiratory organ" to which the second microphone is attached refers to a part, such as a shoulder, where breathing noise has little influence.

[0028] In the wheezing detection device of the embodiment, the breath sound detection unit includes a first microphone in the form of a stethoscope that is attached to the skin of the subject's chest, and a second microphone that is attached to clothing or the skin of a part remote from the subject's chest and respiratory organ. Also, a difference obtained by subtracting the output of the second microphone from the output of the first microphone is output. Consequently, noise components in the subject's environment can be removed from the breath sound signal. As a result, it is possible to more accurately detect whether or not wheezing is included in the subject's breath sound. Advantageous results of the invention

[0029] As is obvious from the above description, with the wheezing detecting device of the present invention, it is possible to accurately detect whether or not wheezing is included in the breath sound of the measurement subject. Short description of the drawings Fig. 1 is a drawing showing a schematic block structure of a wheezing detection system according to an embodiment of the invention. Fig. 2(A) is a diagram showing the exterior of a wheeze detecting device included in the above-mentioned wheeze detecting system. Fig. 2(B) is a diagram showing an enlarged view of the exterior of a main body of the wheezing detecting device. Fig. 3 is a diagram showing a block structure of the main body of the wheezing detecting device. Fig. 4 is a diagram showing a block structure of a smartphone included in the wheeze detection system. Fig. 5(A) is a diagram showing an operation mode in which the wheeze detecting device is attached to an infant as a measurement subject. Fig. 5(B) is a diagram showing an operation mode in which the wheeze detection device is operated via the smartphone. Fig. 6 is a diagram showing both a breath sound signal detected by a microphone of the wheeze detection device and a breath flow signal output from a breath flow sensor. Fig. Figure 7 is a diagram showing a frequency spectrum obtained by converting the breath sound signal into a frequency space. Fig. Figure 8 is a diagram showing a frequency spectrum of a breath sound obtained in a certain processing unit time period. Fig. Figure 9 is a graph showing both the temporal change in an L / D value of a dominant peak contained in the frequency spectrum of breath sounds and the temporal change in the area of ​​the dominant peak for a certain asthmatic patient. Fig. Figure 10 is a histogram showing the data frequency of L / D values ​​of a normal breath sound actually observed without wheezing and the data frequency of L / D values ​​for a breath sound actually observed with wheezing. Fig. Figure 11 is a diagram showing methods for expressing the temporal change in wheezing frequency. Fig. 12 is a diagram showing an example in which a bar chart expressing the temporal change in the frequency of wheezing is displayed on a display screen of a smartphone. Fig. 13 is a diagram showing both a breath sound signal detected by a microphone of the wheeze detector and an envelope calculated for the breath sound signal. Fig. 14 is a diagram showing both the Fig. 13 as well as the envelope shown in Fig. 6 shows the respiratory flow rate signal. Fig. 15(A) is a diagram showing an initialization menu screen for a “wheeze checker” program installed on the smartphone. Fig. 15(B) is a diagram showing a screen displayed when a “medicine administration time” switch in Fig. 15(A) is pressed. Fig. 16(A) is a diagram showing a screen displayed when a “Drug Administration A” switch and a “Drug Administration B” switch are Fig. 15(B) must be pressed. Fig. 16(B) is a diagram showing an example in which both a bar chart showing the temporal change in the frequency of wheezing and information related to drug administration are displayed on the display screen of the smartphone. Fig. 17 is a flowchart showing a method with operations to be performed by a user in the case of displaying the Fig. 12 shown display example on the display screen of the smartphone. Fig. Fig. 18 is a flowchart showing a method with operations to be performed by a user in the case of displaying the Fig. 16(B) shown display example on the display screen of the smartphone. Fig. 19 is a flowchart showing a method including operations performed by a user in the case of recording and playing back the breath sound using the smartphone. Description of embodiments

[0030] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0031] Fig. Figure 1 shows a block diagram of a wheeze detection system (indicated overall by reference numeral 1) that is one embodiment of the wheeze detection device of the present invention. The wheeze detection system 1 includes a wheeze detection device 100 and a smartphone 200. The wheeze detection device 100 and the smartphone 200 can communicate with each other via wireless communication.

[0032] As in Fig. As shown in Figure 2(A), the wheeze detector 100 includes a main body 100M and a first microphone 111 and a second microphone 112 connected to the main body 111 via a microphone connector 113. In this example, the first microphone 111 and the second microphone 112 are in the form of stethoscopes, both of which have circular plate shapes (with adhesive layers 119 provided on their recessed surfaces). The first microphone 111 is intended to adhere to the skin of the chest of the measurement subject, and the second microphone 112 is intended to adhere to the clothing of the measurement subject.

[0033] As in Fig. 2(B) enlarged, the main body 100M is provided with a bracket 100C, a microphone jack 114, an operation unit 130, a headphone jack 116, a power source switch 191, a charging jack 192, and a communication status display LED (light-emitting diode) 193.

[0034] The clamp 100C is used to attach the main body 100M to the clothing of the measurement test subject.

[0035] The microphone terminal 114 is used to receive the output of the first microphone 111 and the second microphone 112 in a state where the microphone plug 113 is inserted therein.

[0036] The control unit 130 includes a volume increase push-button switch 131, a volume decrease push-button switch 132, and a communication switch 133. The volume increase push-button switch 131 is used to increase the volume of sound output to headphones (not shown) via the headphone jack 116. Conversely, the volume decrease push-button switch 132 is used to decrease the volume of sound output to the headphones. The communication switch 133 is used to establish a connection for wireless near-field communication between the main body 100M and the smartphone 200. In other words, when the communication switch 133 is pressed, a communication link is established between the wheeze detector 100 and the smartphone 200 through a known protocol, and wireless near-field communication becomes possible.

[0037] The power source switch 191 is used to turn the power source of the wheezing detector 100 on and off.

[0038] The charging port 192 is used to charge a battery built into the main body 100M.

[0039] The communication status indicator LED 193 indicates the communication status between the wheeze detection device 100 and the smartphone 200. Specifically, if the wireless near-field communication between the wheeze detection device 100 and the smartphone 200 has not been established, the LED 193 illuminates red. If the wireless near-field communication connection between the wheeze detection device 100 and the smartphone is in the process of being established, the LED 193 flashes green.

[0040] When the wireless near-field communication connection between the detector 100 and the smartphone 200 has been established, the LED 193 illuminates green. When the wireless near-field communication connection is established, the wheeze detector 100 enters a state capable of operating according to an instruction from the smartphone 200 (a standby state).

[0041] As in Fig. 3, the main body 100M of the wheezing detection device 100 is provided with a control unit 110, a sound signal processing circuit 115, a memory 120, a near-field wireless communication unit 180, and a power source unit 190, in addition to the above-mentioned headphone jack 116, the operation unit 130, a power source switch 191, and the charging port 192.

[0042] In this example, the sound signal processing circuit 115 consists of a CODEC IC (CODEC integrated circuit), receives the output of the first microphone 111 and the output of the second microphone 112, subtracts the output of the second microphone 112 from the output of the first microphone 111, and outputs a breath sound signal indicating the obtained difference to the control unit 110 and the headphone jack 116. The user can confirm that the breath sound signal has been received by connecting headphones (not shown) to the headphone jack 116 and listening. Note that the first microphone 111, the second microphone 112, and the sound signal processing circuit 115 constitute a breath sound detection unit.

[0043] The memory 120 includes a ROM (read-only memory) and a RAM (random access memory). The ROM stores data from programs for controlling the wheeze detector 100. Likewise, the RAM stores setting data for setting various functions of the wheeze detector 110, data for a calculation result, and the like.

[0044] The control unit 110 includes a CPU (central processing unit) and controls the units (including the memory 120 and the near-field wireless communication unit 180) of the wheeze detection unit 100 according to a program for controlling the wheeze detection unit 100 stored in the memory 120. Specifically, the control unit 110 determines whether or not wheeze is included in the breath sound of the measurement subject and generates image data indicating a temporal change in the frequency of wheeze (this will be described in detail later).

[0045] In this example, the power source unit 190 includes a lithium-ion battery (secondary battery) and supplies power or stops the power supply to the units of the wheezing detector 100 according to the on / off of the power source switch 191. The lithium-ion battery can be charged via the charging port 192.

[0046] The wireless near-field communication unit 180 performs wireless communication, and in this example, performs wireless near-field communication (BT (Bluetooth (registered trademark)) and BLE (Bluetooth Low Energy Communication)) with the smartphone 200 according to the control performed by the control unit 110. For example, information expressing a calculation result and the like is transmitted to the smartphone 200. An operation instruction is also received from the smartphone 200.

[0047] As in Fig. 4, the smartphone 200 includes a main body 200M, a control unit 210 mounted in the main body 200M, a memory 220, an operation unit 230, a display unit 240, a speaker 260, a near-field wireless communication unit 280, and a network communication unit 290. The smartphone 200 has application software (referred to as a "wheeze checker" program) installed thereon for causing a commercially available smartphone to perform processing of wheeze-related information.

[0048] The control unit 210 includes a CPU and its auxiliary circuits, controls the units of the smartphone 200, and executes processing according to a program and data stored in the memory 220. For example, the data input from the communication units 280 and 290 is processed based on an instruction input via the operation unit 230, and the processed data is stored in the memory 220, displayed on the display unit 240, and output via the communication units 280 and 290.

[0049] The memory 220 includes a RAM used as a work area required for the control unit 210 to execute a program, and a ROM for storing basic programs to be executed by the control unit 210. Likewise, a semiconductor memory (memory card, SSD (solid state drive)), or the like, can be used as the storage medium of the auxiliary storage device to support the storage area of ​​the memory 220.

[0050] In this example, the control unit 230 consists of a touch panel provided on the display unit 240. Note that a keyboard or a hardware control device may be included.

[0051] In this example, the display unit 240 includes a display screen consisting of an LCD (liquid crystal display element) or an organic EL (electroluminescent) display. The display unit 240 displays various images on the display screen according to the control performed by the control unit 210.

[0052] The speaker 260 generates various sounds, such as an audio sound and an alarm sound serving as a warning, according to the control performed by the control unit 210.

[0053] The wireless near-field communication unit 280 performs wireless communication, in this example, wireless near-field communication (BT communication and BLE communication), with the wheezing detector 100 according to the control performed by the control unit 210. For example, an operation instruction is transmitted to the wheezing detector 100. Likewise, information or the like indicating the calculation result is received from the wheezing detector 100.

[0054] The network communication unit 290 may transmit the information from the control unit 210 to another device via the network 900, receive information transmitted from another device via the network 900, and transmit the received information to the control unit 210.

[0055] Fig. 5(A) illustrates an operation mode in which the wheeze detector 100 is attached to an infant 90 serving as the measurement subject. In this example, the infant 90 lies down in a child's room 98, and the main body 100M of the wheeze detector 100 is secured to the infant's body via the bracket 100C (see Fig. 2) attached to a sleeve of the clothing (in this example the pajamas) of the toddler 90.

[0056] The first microphone 111 is attached to the skin of the chest of the infant 90 with an adhesive layer 119 provided on a circular plate. Similarly, the second microphone 112 is attached to the clothing (in this case, the pajamas) of the infant 90. Note that the second microphone 112 may be attached to the skin of a part (a part with little influence on the breathing sound, such as a shoulder) that is distant from the chest and respiratory organ of the infant 90. First operational example

[0057] Fig. 17 shows an operating method by which the user (in this example, the mother of the toddler 90) 91 uses the wheezing detection system 1 to cause the display screen of the smartphone 200 to display a temporal change in the frequency of wheezing of the toddler 90 (in this example, the Fig. 12 shows the bar chart AT). (1) The user 91 attaches, as in Fig. 5(A), the wheezing detection device 100 is applied to the infant 90 (step S1 in Fig. 17), presses the power source switch 191 and the communication switch 133 of the wheezing detection device 100 and puts the wheezing detection device 100 into the standby state (step S2 in Fig. 17). (2) Next, the user activates 91 in a living room 99, which is different from the children's room 98, as in Fig. 5(B), for example, a “wheeze checker” program installed in the smartphone 200 (step S3 in Fig. 17). Then the user 91 presses a button “Start wheezing test” (in the mode described later Fig. 15(A) by reference number 23) displayed on the display screen of the smartphone 200, thereby instructing the wheezing detection device 100 to start the measurement (step S4 in Fig. 17). (3) Thus, the wheezing detection device 100 detects the breathing sound of the infant 90, the breathing sound signal is processed, and image data expressing the temporal change in the frequency of wheezing is generated (step S5 in Fig. 17). i) Specifically, the first microphone 111 mainly detects the breathing sound passing through the bronchi of the infant 90, and the second microphone 112 mainly detects ambient noise in the environment of the infant 90. The sound signal processing circuit 115 receives the output from the first microphone 111 and the output from the second microphone 112, subtracts the output of the second microphone 112 from the output of the first microphone 111, and outputs a breathing sound signal (indicated by reference symbol BS) to the control unit 110 in a time series expressing the obtained difference. Consequently, noise components in the environment of the infant 90 are removed from the breathing sound signal BS.

[0058] Fig. Figure 6 illustrates the respiratory sound signal BS obtained using the sound signal processing circuit 115 of the wheezing detector 100. Note that in Fig. 6, a respiratory flow rate signal BF output from a respiratory flow rate sensor (not included in the wheezing detection system 1) is also shown. The positive side of the respiratory flow rate signal BF indicates the exhalation flow rate, and the negative side indicates the inspiration flow rate.

[0059] ii) Next, the control unit 110 operates as a determination processing unit and determines whether or not wheezing is included in the breath sound based on the breath sound signal BS at every predetermined processing unit time period (indicated by reference symbol tu, in this example tc = 0.05 seconds).

[0060] Here Fig. 7 shows a frequency spectrum PS obtained by the control unit 110 by converting the respiratory sound signal BS into a frequency domain in each processing unit time period tc. Likewise, Fig. 8 shows a frequency spectrum PS for a breathing sound which in a certain processing unit time period tc (corresponds in this example to a processing unit time period with a time of 0.05 seconds in Fig. 7). In the analysis carried out by the inventor, the whistling sound Pw of the wheezing (see Fig. 7) characterized in that, as in Fig. 8, the width D of the peak of its frequency spectrum PS is relatively narrow (almost monotonous). Also, the wheezing sound is characterized by the fact that it comprises several peaks with relatively narrow widths D (e.g., see Fig. 6 in US 2011 / 0125044 A1). Consequently, in order to accurately detect wheezing, the widths D of the peaks in the frequency spectrum PS should be used in some way in the determination. In view of this, the control unit 110 in the wheezing detector 100 determines, based on the height L and the width D of the peak, whether a peak in the frequency spectrum PS indicates wheezing or not. Specifically, the control unit 110 obtains the ratios between the heights L and the widths D of the peaks (L / D indicates the steepness of the peaks) and determines whether wheezing is indicated or not based on whether the ratios (L / D) are greater than a predetermined first threshold (indicated by a reference symbol α; in this example, α = 0.35).

[0061] Note that when background noise is present in the frequency versus sound pressure graph, the heights L and widths D of the peaks indicate the true heights L and widths D of the peaks from which the background noise has been removed. For example, in the graph shown in Fig. 8, the heights L and widths D are the true heights L and widths D of the peaks P1, P2, P3, P4, and P5, which are the sections exceeding the line segments connecting the local minima m0, m1, m2, m3, m4, m5, ... obtained in the frequency versus tone pressure diagram. Similarly, for the peak areas described later, the true areas S1, S2, S3, S4, S5, ... of the sections exceeding the line segments connecting the local minima are given.

[0062] In the analysis carried out by the inventor, the whistling sound Pw (see Fig. 7) Wheezing, which is frequently observed in the case of childhood asthma, is a sound having peaks with relatively narrow widths D (almost monotonous) with a frequency in a range of approximately 900 Hz to 1200 Hz. In view of this, the control unit 110 with the wheezing detector 100 determines whether wheezing is indicated only for peaks with frequencies within the range of 200 Hz to 1500 Hz in the frequency spectrum PS. Consequently, it is possible to detect whether wheezing, including whistling wheezing, which is frequently observed in the case of childhood asthma, is included in the breath sound of a measurement subject in addition to the wheezing sound. On the other hand, sounds outside the range of 200 Hz to 1500 Hz are not considered wheezing and are therefore not considered in the determination.

[0063] In addition, the dominant peak Pd (in this example the peak P5) corresponds to the frequency in relation to the sound pressure ( Fig. 8) of the plurality of peaks P1, P2, P3, ... in the frequency spectrum PS, the peak with the largest area is the peak with the greatest energy. Consequently, the dominant peak Pd determines whether wheezing is included in the processing unit time period tc or not. In view of this, the control unit 110 in the wheezing detection device 100 determines only based on the dominant peak Pd having the largest area in the frequency versus sound pressure diagram ( Fig. 8) of the multiple peaks P1, P2, P3, ... in the frequency spectrum PS, whether wheezing is indicated.

[0064] For example, Fig. 9 both a temporal change CPd in ​​the L / D value of the dominant peak Pd contained in the frequency spectrum of the breath sound of a certain asthmatic patient, as well as a temporal change Spd in the area of ​​the dominant peak Pd. Periods without wheezing and periods with wheezing that are actually observed are specified on the time axis (horizontal axis). As can be seen from Fig. 9, it is understood that the L / D value of the dominant peak Pd is less than or equal to the threshold α in the periods in which no wheezing was actually observed, while the L / D value of the dominant peak Pd in ​​the periods with wheezing that were actually observed approximately exceeds the threshold α. Concomitantly, the area of ​​the dominant peak Pd is also larger in the periods with wheezing.

[0065] Also shows Fig. 10 shows as a histogram the data frequency of the L / D values ​​for a normal breath sound that was actually observed to contain no wheezing, and the data frequency of the L / D values ​​for a breath sound that was actually observed to contain wheezing. As can be seen from Fig. 10, it is understood that the data group H0 of L / D values ​​for the normal breath sound is less than or equal to the threshold α, and the data group H1 of L / D values ​​for the breath sound actually observed to contain wheezing exceeds the threshold α.

[0066] According to the Fig. 9 and Fig. From the results shown in Fig. 10, it can be said that with the wheezing detection device 100, it is possible to accurately determine whether wheezing is included in the breathing sound of the measurement subject or not.

[0067] The determination results for each processing unit period tc, or in other words, the results of determining whether or not wheezing is included in the breath sound of the measurement subject, are sequentially stored and accumulated as binary data in the memory 120. For example, if wheezing is included in the breath sound, 1 is stored, and if no wheezing is included in the breath sound, 0 is stored.

[0068] iii) Next, based on the determination results described above, the control unit 110 operates as an addition processing unit, sets an addition unit period (e.g., 30 seconds) with a plurality of processing unit periods tc, and sequentially adds the lengths of the processing unit periods tc in each addition unit period in which it was determined that wheezing was included.

[0069] As in Fig. Specifically, as shown schematically in Figure 11, in each addition unit time period T1, T2, T3, T4, ...., the lengths of the processing unit time periods tc in each processing unit time period in which wheezing was determined to be included are added, and the results are obtained as wheezing time periods Tw1, Tw2, Tw3, Tw4, .... In addition, information indicating a temporal change in the frequency of wheezing is generated as a graph indicating percentages occupied by the wheezing time periods Tw1, Tw2, Tw3, Tw4, ... in bars AT1, AT2, AT3, AT4, ... with a certain length corresponding to the addition unit time period. Note that the percentages of the time periods (normal breathing time periods) in which wheezing is not present in the addition unit time periods are indicated by O1, O2, O4, ....

[0070] Specifically, in this example, the intensity of the wheezing sound in each addition unit time period T1, T2, T3, T4, ... is calculated based on the area (S5 in the Fig. 8 example) of the dominant peak Pd in ​​the frequency spectrum Ps of the breath sound was classified into five levels A, B, C, D and E, and the lengths of the processing periods tc in which wheezing was determined to be included were added for each of the classified levels A, B, C, D and E. Level A is set such that the area of ​​the dominant peak Pd is 0 or more and less than 250, level B is set such that the area of ​​the dominant peak Pd is 250 or more and less than 500, level C is set such that the area of ​​the dominant peak Pd is 500 or more and less than 750, level D is set such that the area of ​​the dominant peak Pd is 750 or more and less than 1000, and level E is set such that the area of ​​the dominant peak Pd is 1000 or more and less than 1250. Note that for ease of understanding, in the Fig. In the example shown in Figure 11, the percentages of the five levels A, B, C, D and E are shown shifted to the right one after the other, but they can also be shown in a straight line.

[0071] The levels for classifying wheezing intensity (i.e., wheezing severity) are not limited to the five levels A, B, C, D, and E. For example, it may be easier for an average user who is not a medical professional to intuitively understand that wheezing intensity is classified into three levels.

[0072] iv) In view of this, in the case of generating image data to be actually transmitted to the smartphone 200, the control unit 110 functions as a display processing unit and sets the percentages obtained by combining the normal breathing time periods and the time periods in which the area of ​​the dominant peak Pd is in a range of 0 or more and less than 250 as green G. The percentages of the time periods in which the area of ​​the dominant peak Pd is in a range of 250 or more and less than 750 are made yellow Y. Also, the percentages of the time periods in which the area of ​​the dominant peak Pd is in the range of 750 or more are made red R. In response, the bars with certain lengths corresponding to the addition unit time periods are displayed divided into three colors, namely green G, yellow Y, and red R. The control unit 110 also sets up the plurality of bars AT1, AT2, ...according to the addition unit time period in parallel to generate image data representing a bar chart (in this example, the one shown in . Fig. 12, which indicates the temporal change in the frequency of wheezing.

[0073] Note that the addition unit time span in the Fig. In the example shown in Figure 11, the time period is set to 30 seconds, but there is no limitation. The addition unit period can be set in various ways, such as one minute, two minutes, five minutes, 10 minutes, 30 minutes, one hour or more, and less than 24 hours, one day, one week, or one month. In the following example, the addition unit period is one hour.

[0074] Even if the percentage of time for which the intensity of the wheezing sound has reached red R in an addition unit period of time exceeds a predetermined second threshold (indicated by reference symbol β; in this example, β = 5[%]), the control unit 110 functions as a warning generation unit that transmits an alarm signal as a warning to the smartphone 200 via the wireless near-field communication unit 180.

[0075] (4) Next, the user 91 presses a button “End measurement results” (in the manner described later Fig. 15(A) by reference numeral 28) displayed on the display screen of the smartphone 200, and receives the image data from the wheezing detection device 100 via the wireless near-field communication unit 280 (specifically, BLE communication) (step S6 in Fig. 17). The received data is automatically stored and recorded in the memory 220, which serves as the storage unit.

[0076] The Fig. 12, which indicates the temporal change in the frequency of wheezing, is displayed on the display screen 10 of the smartphone 200 (step S7 in Fig. 17).

[0077] Here, a remaining battery amount 11 and a current time 12 are displayed at the top level of the display screen 10. Also provided below these are an "Asthma Checker" indicator 13 as a name of the application software, and a "Wheeze Checker" indicator 14. A "Cancel" switch 17 for inputting an instruction to terminate the "Wheeze Checker" program and a "Back" switch 18 for inputting an instruction to return to the screen displayed immediately before the content of the display screen was displayed are provided to the left and right of the "Wheeze Checker" indicator 14. Furthermore, a wheeze detection result display panel 50 for displaying image data received from the wheeze detector 100 is provided below.

[0078] A measurement time indicator (in this example, "December 25, 2014, 11:24") 51 indicating the final measurement date and time, a field name indicator 52 reading "Pulse Detection Results," and an image data display area 56 are provided in the pulse detection result display area 50. An order (in this example, "1, 2, 3, ...") 55 of the addition unit time periods is displayed on the horizontal axis in the image data display area 56. Also, directly below the horizontal axis, it is indicated that the measurement date and time data indicated in the measurement time indicator 51 is included in the sixth data item (bar AT 4), since the number 6 indicates the result obtained at the displayed time. Furthermore, in the image data area 56, the vertical axis 54 indicates "Frequency of Occurrence (%)" in increments of 2%; e.g. 90%, 92%, 94%, ..., 100% is displayed.Also, the bar chart AT indicating the temporal change in the frequency of the wheezing described above is displayed within the image data display area 56.

[0079] The bar chart AT includes multiple bars AT1, AT2, ... of a certain height, each corresponding to the addition unit time period (in this example, one hour), in order of change over time. The bars are displayed in three colors: green G, yellow Y, and red R. As described above, green G indicates a percentage obtained by combining a normal breathing period and a period in which the area of ​​the dominant peak Pd is 0 or more and less than 250, or in other words, a percentage of time in which there is no or approximately no wheezing in the addition unit time period corresponding to the bar. Yellow Y indicates a percentage of time for which the area of ​​the dominant peak Pd is 250 or more and less than 750, or in other words, a percentage of time in which wheezing is relatively light.Red R indicates a percentage of time for which the area of ​​the dominant peak Pd is 750 or more, or in other words a percentage of time in which the wheezing is relatively strong.

[0080] By viewing the graph AT, the user 91 can intuitively visually find out the temporal change of the severity of the wheezing in each addition unit period, along with the temporal change in each addition unit period of the frequency of the wheezing.

[0081] For example, in the example of Fig. 12, the percentage of time obtained by combining yellow Y and red R, next to the unit time period number "3," is about 2%, and therefore, it can be assumed that there was wheezing for about 1.2 minutes in one hour. Since the percentage of time for yellow Y and the percentage of time for red are also each about 1%, it can be assumed that there was relatively weak wheezing and relatively strong wheezing for about the same percentage of time. Also, the percentage of time obtained by combining yellow Y and red R in the addition unit time period "4" is about 4%, and therefore, it can be assumed that there was wheezing for about 2.4 minutes in one hour.Also, the percentage of time spent on yellow Y is about 3%, and the percentage of time spent on red R is about 1%, and therefore, it can be assumed that the percentage of time spent on relatively weak wheezing increased to about three times the percentage of time spent on relatively strong wheezing. Also, in addition unit time periods "5" and "6," it can be assumed that there was an approximate return to the state of addition unit time period "3."

[0082] Thus, by viewing the bar chart AT, the user 91 can intuitively visually determine the temporal change in each addition unit period of the wheezing frequency and the severity of the wheezing contained in the breath sound of the infant 90 serving as the measurement subject. Also, the information expressed by the bar chart AT is automatically stored in the memory 220. Consequently, by loading the information from the memory 220 and causing the bar chart AT to be displayed on the display screen 10 the next time the infant 90 has a medical examination, the user 91 can show the physician the temporal change in the wheezing frequency and the severity of the wheezing contained in the breath sound of the infant 90. As a result, the physician can more easily diagnose whether the infant 90 has asthma and the severity of the asthma, and can easily generate a treatment plan.

[0083] The user 91 can also understand whether asthma is getting worse or better by observing the state of the temporal change for each addition unit period of the wheezing frequency and wheezing severity in the bar chart AT. For example, if asthma is getting worse, preventive measures such as administering medication are possible, resulting in preventing the worsening of asthma.

[0084] On the display screen 10 of the smartphone 200, a direction button 61, according to which a target time period displayed in the wheezing detection result display area 50 is moved up, and a direction button 62, according to which a target time period is moved down, are provided below the wheezing detection result display area 50. The user can select a target time period displayed on the display screen 10 as the wheezing detection result by pressing the direction buttons 61 and 62. Also, when the user presses the "Edit Note" switch 63, a note screen (not shown) opens, and the user can manually enter and record what he or she felt while viewing the wheezing detection results. A "Play Wheezing Sound" switch 64 will also be described later.

[0085] When the percentage of time the wheezing sound intensity reaches R in the addition unit time period exceeds the predetermined threshold β (= 5 [%]), the smartphone 200 receives an alarm signal from the wheezing detector 100 via the near-field wireless communication unit 280. Upon receiving the alarm signal, the control unit 210 of the smartphone 200 uses the speaker 260 to generate an alarm sound serving as a warning. With this alarm sound, the user 91 can be made aware of the fact that the symptoms of the infant 90 serving as the measurement subject have worsened, even if the user 91 is in a living room 99 separate from the child's room 98 (where the infant 90 is lying). Consequently, it is possible to take a countermeasure, such as administering medicine to the infant 90.The warning is particularly beneficial in the case where the measurement subject is a toddler 90, a critically ill patient or similar who has difficulty expressing his / her intentions.

[0086] Note that the warning is not limited to generating an alarm sound using the speaker 260, and it is also possible to perform an alarm display (not shown) on the display screen 10 or to vibrate using a vibrator (not shown) that performs the notification of signal reception. Second operating example

[0087] Fig. 18 shows an operation method that displays, on the display screen of the smartphone 200, a bar chart AT indicating the temporal change in the frequency of wheezing of the infant 90 and information related to medication administration according to the wheezing detection system 1. (1) Steps S11 to S15 in Fig. 18 are carried out similarly to steps S1 to S5 in Fig. 17 executed.

[0088] Here Fig. 15(A) shows an initialization menu screen displayed on the display screen 10 when the user 91 in step S13 of Fig. 18 activates the “wheeze checker” program installed in the smartphone 200.

[0089] On the Fig. In the initialization menu screen shown in Fig. 15(A), a state of the BT communication 16 or the BLE communication 15 with the wheeze detector 100 performed by the near-field wireless communication unit 280 is shown as "disconnected" or "connected" between the "asthma checker" display 13 and the "wheeze checker" display 14. Also provided thereunder are the "wheeze checker" display 14, a "reserve" switch 21, a "setting" switch 22, a "start wheeze test" switch 23, a "stop wheeze test" switch 24, a "start recording" switch 25, a "stop recording" switch 26, a "setting screen" switch 27, a "stop measurement results" switch 28, a "drug administration time" switch 30, and a phase selection switch 40 serving as a phase instruction input.

[0090] The "Reserve" switch 21 is used to allow the user 91 to reserve a time period, such as from 9:00 PM on December 26, 2014, to 7:00 AM on December 27, 2014, during which the measurement is to be performed by the wheeze detector 100. The "Set" switch 22 is used to set the condition (in this example, the threshold β) under which the wheeze detector 100 generates the above-described alarm signal, and to set whether or not to generate the alarm sound, whether or not to perform alarm display, whether or not to perform automatic recording, and the like when the smartphone 200 receives the alarm signal. The "Start Wheeze Test" switch 23 is used to instruct the wheeze detector 100 to start the measurement.The "Stop Wheezing Test" switch 24 is used to instruct the wheezing detector 100 to stop the measurement. The "Start Recording" switch 25 is used to instruct the wheezing detector 100 to transmit the breath sound signal BS. The "Stop Recording" switch 26 is used to instruct the wheezing detector 100 to stop transmitting the breath sound signal BS. The "Setting Screen" switch 27 is used to set an SSID (Service Set Identifier) ​​and an encryption key (password) between the wireless near-field communication unit 280 of the smartphone 200 and the wireless near-field communication unit 180 of the wheezing detector 100. The "Drug Administration Time" switch 30 is used to enter a drug administration time (year, month, day, hour, minute). Note that the phase selection switch 40 is described later.

[0091] (2) In this example, the user presses 91 in the state where the Fig. 15(A) is displayed on the display screen 10 of the smartphone 200, the “medicine administration time” switch 30 (step S16 in Fig. 18). Subsequently, the time at which the "Medicine Administration Time" switch 30 was pressed is stored as the medication administration time (year, month, time, hour, minute) in the memory 220. Note that it is also possible to use a configuration in which, when the user 91 presses the "Medicine Administration Time" switch 30, a screen for entering the medication administration time opens, the user 91 enters the medication administration time (year, month, day, hour, minute) in this screen, and when the user 91 presses the "Medicine Administration Time" switch 30 again, the medication administration time is saved. When the medication administration time is saved, as shown in Fig. 15(B), the drug administration information input screen for entering the type of drugs administered is displayed on the display screen 10. In this example, three types of switches, namely, a "Drug Administration A" switch 31, a "Drug Administration B" switch 32, and a "Drug Administration C" switch 33, are displayed. Note that, in fact, specific drug administration names designated in a prescription by a doctor are registered in advance as the "Drug Administration A," "Drug Administration B," and "Drug Administration C" using a drug administration name registration screen (not shown). The specific drug administration names that have been registered will be displayed at the display locations for “Drug Administration A”, “Drug Administration B” and “Drug Administration C” in Fig. 15(B). Switches 30, 31, 32, and 33 constitute drug administration information input units.

[0092] (3) In a state where the Fig. 15(B) is displayed, the user 91 presses one of the drug administration switches and inputs the type of drug administration (step S17 in Fig. 18). For example, if the user 91 presses the “Medication Administration A” switch 31, as shown in Fig. 16(A), a switch for checking containing drug administration information 34 stating “December 25, 2014, 11:22, Drug Administration A” is displayed on the display screen 10 of the smartphone 200 (step S18 in Fig. 18).

[0093] (4) If the user 91 also presses the “Back” switch 18 twice to return to the Fig. 15(A), press the “Drug Administration Time” switch 30 and press the “Drug Administration B” switch 32 on the screen shown in Fig. 15(B), the medication administration information input screen shown in Fig. 16(A), a review screen containing medication administration information 35 stating "December 15, 2014, 11:24, Medication Administration B" is displayed along with the preceding medication administration information 34. In other words, steps S16 to S18 in Fig. 18 repeated for each drug administration.

[0094] (5) The user 91 then presses the “Back” button 18 twice to return to the Fig. 15(A), presses the “End Measurement Results” switch 28 to receive image data from the wheezing detection device 100 via the wireless near-field communication unit 280 (specifically, BLE communication) (step S19 in Fig. 18). The received data is automatically stored and recorded in the memory 220, which serves as the storage unit.

[0095] Then, the control unit 210 of the smartphone 200 operates as the display processing unit, and, as in Fig. 16(B), the drug administration information for each addition unit period is displayed on the display screen 10 together with the bar graph AT indicating the temporal change in the frequency of wheezing (step S20 in Fig. 18). In the Fig. 16(B), “Medicine Administration A” 57 is displayed above the bar chart AT1 for the addition unit period number 5, and “Medicine Administration B” 58 is displayed above the bar chart AT1 for the addition unit period number 6. The user 91 can visually and intuitively find out the frequency of wheezing in the addition unit period, information related to the medicine administration, and in this example, find out that the toddler 90 was given the medicine administration A in the addition unit period number 5, and that the toddler was given the medicine administration B in the addition unit period number 6. Consequently, the user 91 or the doctor to whom the Fig. 16(B), it is easy to determine whether the drug administration had an effect (a reduction in the frequency of wheezing) on ​​the infant 90 or not.

[0096] For example, in the Fig. 16(B), despite the fact that there was "Medicine Administration A" in the addition unit period number 5 (in which the percentage of red time R was about 3%), the percentage of red time R in the addition unit period number 6 decreased to about 5%, and the symptoms of toddler 90 worsened. For this reason, there is a high probability that "Medicine Administration A" was not effective. It might be better to determine the effect of "Medicine Administration B" after considering the addition unit period number 7 and beyond. Third operating example

[0097] Fig. 19 shows a method of operations according to which the user 91 records the breathing sound of the infant 90 in the memory 220 of the smartphone 200 using the wheeze detection system 1. (1) Steps S21 to S23 in Fig. 19 move similarly to steps S1 to S3 in Fig. 17 before.

[0098] Here, in step S23, Fig. 19 of the Fig. The initialization menu screen shown in Figure 15(A) is displayed on the display screen 10 of the smartphone 200. A phase selection switch 40 serving as a phase instruction input unit is included on the initialization menu screen. When the breath sound signal BS is to be recorded, the phase selection switch 40 includes an "exhalation" switch 41 for selecting only the exhalation phase, an "exhalation / inhalation" switch 42 for selecting both the exhalation phase and the inhalation phase, and an "inhalation" switch 43 for selecting the inhalation phase.

[0099] (2) In a state where the phase selection switch 40 is displayed on the display screen 10 of the smartphone 200 according to a request made by the doctor, the user 91 presses, for example, one of the “exhalation” switch 41, the “exhalation / inhalation” switch 42, and the “inhalation” switch 43 according to whether only the exhalation phase, both the exhalation phase and the inhalation phase, and only the inhalation phase should be recorded (step S24 in Fig. 19). Consequently, the phase to be recorded is selected.

[0100] (3) Next, the user 91 determines whether to record manually or automatically (step S25 in Fig. 19). For example, if toddler 90's current wheezing symptoms are severe and the user wants to record the wheezing immediately, it is desirable to select manual recording. On the other hand, if toddler 90's current wheezing symptoms are favorable and the user wants to record 91 when the wheezing symptoms become severe, it is desirable to select automatic recording.

[0101] (4) In the case of performing manual recording (Yes in step S25 of Fig. 19) the user 91 presses the “Start recording” button 25 on the Fig. 15(A) shown initialization menu screen (step S26 of Fig. 19). The control unit 210 of the smartphone 200 then instructs the wheezing detection device 100 to transmit the respiratory sound signal BS via the wireless near-field communication unit 280. On the other hand, in the case of performing the automatic recording, the user 91 uses the Fig. 15(A) to determine that the "automatic recording" is to be performed. In the "automatic recording" mode, the control unit 210 of the smartphone 200 waits for the above-described alarm signal (indicating that the percentage of time in which the addition unit period reaches red R has exceeded the threshold β) from the wheezing detection device 100 (step S29 in Fig. 19) and instructs the wheezing detection device 100 at the time of receiving the alarm signal (Yes in step S29 of Fig. 19) to transmit the breath sound control signal BS via the wireless near-field communication unit 280. In both the case of manual recording and the case of automatic recording, the wheeze detection device 100 transmits the breath sound signal BS to the smartphone 200 via the wireless near-field communication unit 180 (specifically, BT communication) when the wheeze detection device 100 receives the instruction to transmit the breath sound signal BS from the smartphone 200.

[0102] (5) After receiving the respiratory sound signal BS in the smartphone 200, the control unit 210 operates as a sound recording unit and performs the recording by storing the phase of the respiratory sound signal BS, which is selected with the phase selection switch 40, in the memory 220 (step S27 in Fig. 19).

[0103] Specifically, the control unit 210 of the smartphone 200 operates as the phase identification unit and detects the phases of the breathing sound signal BS as follows.

[0104] First, the breath sound signal BS, as in Fig. 6, local minima synchronized with zero-crossing points at which the respiratory flow rate signal BF transitions from negative (inhalation) to positive (exhalation). Consequently, the control unit 210 can obtain a respiratory cycle tc of the measurement subject (in this example, the infant 90) by detecting the local minima BS0, BS1, ... of the respiratory sound signal BS.

[0105] Next, the control unit 210 generates, as shown in Fig. 13, an envelope BE for the respiratory sound signal BS (if the respiratory sound signal BS in this example is 3000 or less, it is replaced by 0 for simplicity). If the envelope BE for the respiratory sound signal BS and the respiratory flow rate signal BF shown in Fig. 6 are combined, are peaks BEp of the envelope BE, as shown in Fig. 14, synchronized with zero-crossing points at which the respiratory flow rate signal BF transitions from positive (inhalation) to negative (exhalation). Consequently, the control unit 210 can distinguish and identify the exhalation phase te and the inhalation phase ti in the respiratory cycle tc by detecting the peaks BEp of the envelope BE. Note that a peak BEn that is misaligned from the cycle of an original peak BEp in the envelope is ignored as noise based on the average cycle of the original peaks BEp.

[0106] Consequently, after distinguishing between and identifying the exhalation phase te and the inhalation phase ti in the respiratory cycle tc, the control unit 210 performs the recording by storing the phase of the respiratory sound signal BS in the memory 220 using the phase selection switch 40. Note that the phases of the respiratory sound signal BS in the memory 220 are Fig. 16(B) indicates that the exhalation phase has been recorded.

[0107] Note that if the user 91 in the case of manual recording presses the “Stop Recording” button 26 on the Fig. 15(A), the recording performed by the smartphone 200 stops. The wheeze detector 100 receives a signal indicating "stop recording" via the near-field wireless communication unit 180 and stops the transmission of the breath sound signal BS. In the automatic recording mode, the period during which the wheeze detector 100 transmits the breath sound signal BS and the period during which the smartphone 200 performs the recording are set as a default to 30 seconds starting from the recording start time (this can be changed and set by the user 91).

[0108] When the wheezing of the toddler 90 serving as the measurement subject is relatively large in the automatic recording mode (i.e., when the wheezing is severe), the breathing sound of the toddler 90 can be automatically recorded. Consequently, the user 91 can, for example, by playing the recorded content the next time the toddler 90 has a medical examination, have a doctor listen to the breathing sound of the toddler 90 when the wheezing is severe. As a result, the doctor can more easily diagnose whether the toddler 90 has asthma or not, the severity of the asthma, and can easily generate a treatment plan. Note that if the smartphone 200 does not receive the alarm signal from the wheezing detection device 100 (No in step S29 of Fig. 19), the recording is not carried out.

[0109] (6) After that, the user can, for example, press the “Play wheezing sound” button 64 on the Fig. 12, play the breathing sound signal BS stored in the memory 220, for example, with the loudspeaker 260 (step S28 in Fig. 19).

[0110] According to this operational example, the phase of the respiratory sound signal BS selected using the phase selection switch 40 can be recorded. Therefore, if the phase requested by the physician during the previous medical examination is selected, the user 91 can, for example, have the physician listen to the recorded content of the phase requested by the physician of the respiratory cycle tc when the user 91 has the physician listen to the recorded content of the wheezing of the infant 90 during the next medical examination.

[0111] The wheezing detection result (bar chart AT, which is shown in Fig. 12 and Fig. 16(B)) displayed on the display screen 10 of the smartphone 200, and the recorded content of the respiratory sound signal BS are transmitted to a doctor's computer (terminal in a hospital) via the network communication unit 290. Consequently, the user can obtain the diagnosis of a doctor at a remote location away from the hospital.

[0112] In the above-described embodiment, the wheezing-related information display device of the present invention is configured as a wheezing detection system including the wheezing detector 100 and the smartphone 200, but it is not limited to this. For example, it is possible to incorporate a commercially available computer (personal computer, etc.) instead of the smartphone 200. In this case, the above-described "Wheezing Checker" program is installed on the commercially available computer.

[0113] Likewise, the wheezing detection device of the present invention may be constituted solely by the smartphone 200, for example. In this case, the first microphone 111 and the second microphone 112 are connected to the smartphone 200, and the noise signal processing circuit 115 is mounted in the smartphone 200. (The function of the noise signal processing circuit 115 may be implemented by software and executed by the control unit 210.) In this case, the wheezing detection device of the present invention can be made small and compact. This configuration is advantageous in the case where the measurement subject is the user of the smartphone 200.

[0114] In the embodiment described above, the temporal change in wheezing frequency was displayed as a bar chart AT, but there is no limitation to this. The temporal change in wheezing frequency may be displayed as a different form of chart. For example, the temporal change in only the percentage (%) of time obtained by combining yellow Y and red R may be displayed as a bar chart.

[0115] The above-described embodiment is merely an example, and various modifications are possible without departing from the scope of the invention. The above-described several embodiments can be implemented independently, but it is also possible to combine embodiments. Various characteristics of the various embodiments can also be implemented independently, but it is also possible to combine characteristics of different embodiments. List of reference symbols 1 wheezing detection system 10 Display screen 50 Display field for wheezing detection result 100 Wheezing detection device 111 first microphone 112 second microphone 200 smartphones AT bar chart AT1, AT2, AT3, AT4 bars T1, T2, T3, T4 addition unit period Tw1, Tw2, Tw3, Tw4 panting period

Claims

[1] Wheezing detection device comprising: a breath sound detection unit (111, 112) configured to detect a breath sound of a measurement subject and to obtain a breath sound signal in a time series expressing the breath sound; and a determination processing unit (115) configured to convert the respiratory sound signal into a frequency domain in each predetermined processing unit period to obtain a frequency spectrum of the respiratory sound and to determine, based on a height and a width of a peak in the frequency spectrum, whether the peak indicates wheezing or not, wherein the determination processing unit (115) obtains a ratio between the height and the width of the peak and determines whether or not the wheezing is displayed based on whether the ratio is greater than a predetermined first threshold value, wherein the determination processing unit (115) determines whether or not the wheezing is displayed based only on a dominant peak having the largest area among a plurality of peaks in the frequency spectrum in a graph of frequency with respect to sound pressure. [2] The wheezing detection device according to claim 1, wherein the determination processing unit (115) determines whether wheezing is indicated or not based only on the peak at a frequency ranging from 200 Hz to 1500 Hz in the frequency spectrum. [3] A wheezing detection device according to any one of claims 1 or 2, comprising: an addition processing unit configured to set an addition unit period comprising a plurality of processing unit periods, add the lengths of processing unit periods in which the wheezing was determined to be included in each addition unit period, and obtain the result as a wheezing period; wherein, in each of the processing unit periods, the addition processing unit converts the breath sound signal into a frequency domain to obtain the frequency spectrum of the breath sound, classifies the strength of the wheezing sound into a plurality of levels based on the area of ​​the dominant peak having the largest area in the frequency versus sound pressure graph among the plurality of peaks in the frequency spectrum, and adds the lengths of the processing unit periods determined to include the wheezing in each of the classified levels; and a warning generation unit configured to generate a warning when a percentage of time for which a wheezing sound intensity reaches a specific level in the addition unit period based on the addition performed by the addition processing unit exceeds a predetermined second threshold. [4] A wheezing detection device according to claim 3, comprising: a sound recording unit configured to record the breathing sound signal when the warning generating unit generates the warning. [5] A wheezing detection device according to any one of claims 1 to 4, further comprising: a phase identification unit configured to identify the breathing cycle of the measurement subject by dividing it into an exhalation phase and an inhalation phase based on the breath sound signal obtained by the breath sound detection unit, a phase instruction input unit configured to input an instruction to select one or both of the exhalation phase and the inhalation phase of the breath sound signal; and a sound recording unit configured to record the phase of the breath sound signal instructed by the phase instruction input unit. [6] A wheezing detection device according to any one of claims 1 to 5, wherein the respiratory sound detection unit (111, 112) comprises: a first microphone (111) in the form of a stethoscope, which is attached to the skin of the chest of the measurement subject; and a second microphone (112) attached to clothing or the skin of a part remote from the chest and respiratory organ of the measurement subject, and the breath sound detection unit outputs a difference obtained by subtracting the output of the second microphone (112) from the output of the first microphone (111). [7] Wheezing detection device comprising: a breath sound detecting unit (111, 112) configured to detect a breath sound of a measurement subject and obtain a breath sound signal in a time series expressing the breath sound; a determination processing unit (115) configured to convert the respiratory sound signal into a frequency domain in each predetermined processing unit period to obtain a frequency spectrum of the respiratory sound and to determine, based on a height and a width of a peak in the frequency spectrum, whether the peak indicates wheezing or not, an addition processing unit configured to set an addition unit period comprising a plurality of processing unit periods, add the lengths of processing unit periods in which the wheezing was determined to be included in each addition unit period, and obtain the result as a wheezing period; wherein, in each of the processing unit periods, the addition processing unit converts the breath sound signal into a frequency domain to obtain the frequency spectrum of the breath sound, classifies the strength of the wheezing sound into a plurality of levels based on the area of ​​the dominant peak having the largest area in the frequency versus sound pressure graph among the plurality of peaks in the frequency spectrum, and adds the lengths of the processing unit periods determined to include the wheezing in each of the classified levels; and a warning generation unit configured to generate a warning when a percentage of time for which a wheezing sound intensity reaches a specific level in the addition unit period based on the addition performed by the addition processing unit exceeds a predetermined second threshold.

Images