WO2022114858A1 - Respiratory disease sensing patch and system, and method for providing respiratory disease information - Google Patents

Abstract

A respiratory disease sensing patch is provided. A respiratory disease sensing patch according to an embodiment of the present invention may comprise: an adhesive part (10) attached to the skin of a subject to be analyzed, and having, formed therein, a first sound collection passage (11) for collecting sound projected through the attached skin; a sensor unit (30) which is provided to be positioned above the first sound collection passage (11) and which outputs an acoustic signal by collecting the sound having passed through the first sound collection passage (11); a conversion unit (41) for converting, through a preset method, the acoustic signal, outputted from the sensor unit (30), into a converted acoustic signal; a feature extraction unit (42) for extracting, by means of a preset method, respiratory-disease-related features from the converted acoustic signal converted by means of the conversion unit (41); an information computation unit (43) for computing respiratory-disease-related information from the collected sound by using the respiratory-disease-related features extracted by means of the feature extraction unit (42); and a medication unit (80) in which medicine is stored, and which is opened to allow the medicine to be administered to the skin according to the information computed by means of the information computation unit (43).

Description

Respiratory disease sensing patch and system, and method of providing respiratory disease information

The present invention relates to a patch and system capable of sensing respiratory diseases, particularly asthma and chronic obstructive pulmonary disease, by being attached to an object to be analyzed, and a method for providing information on respiratory diseases.

Worldwide, it is estimated that 339 million people suffer from asthma, and the majority of the patients are children, and the number of chronic obstructive pulmonary disease (COPD) patients is estimated at 251 million, most of them being children. occupied by adults.

Asthma patients are about 10 times more likely to develop bronchitis and emphysema than those without asthma, and chronic obstructive pulmonary disease patients are reported to be about 12 times more likely to develop bronchitis and emphysema compared to those who do not. have.

Therefore, although early diagnosis of asthma or chronic obstructive pulmonary disease is very important, the problem is that there are few initial symptoms to specify the diagnosis of the disease.

While the prevalence of chronic diseases other than respiratory diseases has steadily decreased due to advances in early diagnosis and treatment techniques, the prevalence of asthma and chronic obstructive pulmonary disease has increased worldwide, making it the third leading cause of death in the United States.

On the other hand, the conventional method for performing the diagnosis and examination method of respiratory diseases required expensive equipment and skilled medical staff, but the conventional method was impossible to monitor for a long period of time, and in particular, the examination itself was impossible for pediatric patients under the age of 4 years. There were downsides.

The prior art is as follows.

Korean Patent Registration No. 10-1760449 relates to a patch for diagnosis and treatment of asthma and chronic obstructive pulmonary disease, which is attached to the body to acquire an acoustic signal, and compares the acquired acoustic signal with a reference signal stored in advance to compare asthma or chronic obstructive pulmonary disease A technique for diagnosing lung disease and administering a drug to a patient by opening the drug site cover when diagnosed with a respiratory disease is presented. Although different patterns of signals for each patient are detected, asthma or chronic obstructive pulmonary disease is diagnosed through simple comparison with pre-stored reference signals, so the accuracy of diagnosis is very low. It only presents the simple concept of diagnosing respiratory diseases such as asthma and chronic obstructive pulmonary disease by analyzing the

U.S. Patent No. 9237862 relates to a device for diagnosing asthma, and it is a technology capable of distinguishing mild and severe asthma by measuring changes according to respiration of the pharynx, airways, and intercostal muscles, and analyzing image patterns of respiration rates and frequency bands. present. It is to classify the mild and severe asthma by using the change in physical deformation of the subject to be analyzed. It is only effective for the subject to be analyzed whose asthma has already progressed considerably, and it is difficult to make an accurate diagnosis in the early stage of asthma. .

Accordingly, there is a need for an apparatus and method capable of long-term monitoring in daily life and capable of accurately sensing the presence or absence of a respiratory disease even in the initial stage of a respiratory disease.

(Patent Document 1) Korean Patent Publication No. 10-1760449 (2017.07.21)

(Patent Document 2) US Registered Patent Publication No. 9237862 (2016.01.19)

The present invention collects the sound generated from the subject to be analyzed using a patch-type sensor attached to the subject to be analyzed, and analyzes the collected sound to determine whether the subject to be analyzed has respiratory diseases, particularly asthma and chronic obstructive pulmonary disease. An object of the present invention is to provide a respiratory disease sensing patch and system capable of providing calculation information and administering according to the calculation, and a method for providing information on respiratory diseases.

An embodiment of the present invention for solving the above problems is attached to the skin of the subject to be analyzed, and the adhesive portion 10 through which the first sound collection passage 11 for collecting the sound coming out through the attached skin is formed. ), the sensor unit 30 that is installed to be positioned above the sound collection passage 11, and outputs a sound signal by collecting the sound that has passed through the sound collection passage 11, the output from the sensor unit 30 A conversion unit 41 that converts an acoustic signal into a converted sound signal through a preset method, and a feature extractor 42 that extracts respiratory disease-related features from the converted sound signal converted by the conversion unit 41 in a preset method , an information calculating unit 43 that calculates respiratory disease related information for the collected sound using the respiratory disease related features extracted by the feature extracting unit 42, and a drug is stored therein, and the information calculating unit 43 ) provides a respiratory disease sensing patch, including an administration unit 80 that is opened so that the medicine is administered to the skin according to the information calculated by.

In an embodiment, the first sound collection passage 11 may be reduced in diameter toward the sensor unit 30 .

In one embodiment, formed between the adhesive part 10 and the sensor part 30, the substrate part ( 20), and the inner diameter of the second sound collection passage 21 may be the same even if it faces the sensor unit 30 , but may be the same as the inner diameter of the upper portion of the first sound collection passage 11 .

In one embodiment, the sensor unit 30 is located above the absorption unit 50 made of a porous material, the sound insulating unit 60 located above the absorption unit 50, and the sound insulating unit 60 upper portion It may further include a cover portion 70 to cover the.

In one embodiment, the conversion unit 41 converts the sound signal output from the sensor unit 30 into a first converted sound signal in the form of an intensity of the sound signal over time, and the The first transformed sound signal is Fourier transformed to convert the second transformed sound signal in the form of intensity of the sound signal according to the frequency band, and the sound signal output from the sensor unit 30 is converted into a spectrogram form Converted to a third converted sound signal of , wherein the third converted sound signal may represent the frequency and intensity of the sound signal according to the passage of time.

In an embodiment, the feature extraction unit 42 extracts a signal having a first peak from the second converted sound signal as a fundamental signal, and the basic signal from the second converted sound signal If there is a peak having a center frequency of n times the center frequency of n (n is an integer greater than or equal to 2), it can be extracted as the nth harmonic signal.

In one embodiment, the information calculating unit 43, when the n-th harmonic signal is extracted by the feature extracting unit 42, calculates that the sound signal includes a wheezing sound, or the basic Whether or not the sound signal includes wheezing may be calculated using a ratio of a signal to the nth harmonic signals.

In an embodiment, the information calculating unit 43 may calculate that the sound signal includes wheezing when a ratio calculated through the following equation is greater than a preset first ratio.

[Equation]

In an embodiment, the feature extracting unit 42 extracts an inspiration duration and an expiration duration from the first converted sound signal, and the information calculating unit 43 includes the inspiration duration and the expiration duration. And it is possible to calculate whether the sound signal includes a wheezing or crackle sound by using the ratio of the expiration duration to the inspiration duration.

In an embodiment, the information calculating unit 43 is configured to, when the expiration duration is longer than a preset first time, and the ratio of the expiration duration to the inspiration duration is greater than a preset second ratio, the It can be calculated that the acoustic signal includes wheezing.

In an embodiment, the information calculating unit 43 may calculate that the sound signal includes a crackling sound when the inspiration duration is within a preset second time range.

In an embodiment, when the sound signal is calculated to include the wheezing sound or the crackling sound, the information calculating unit 43 may calculate that there is a respiratory disease.

In one embodiment, when it is calculated that there is a respiratory disease by the information calculating unit 43, the administering unit 60 is opened so that the medicament can be administered to the skin of the subject to be analyzed.

In addition, the present invention is a method using the above-described patch, (a) the sensor unit 30 collects the sound coming from the outside to output a sound signal, (b) the conversion unit 41 is the ( A step of converting the sound signal output in step a) into a converted sound signal by a preset method, (c) the feature extracting unit 42 using a preset method from the converted sound signal converted in step (b) to cause respiratory diseases extracting related features and (d) the information calculating unit 43 obtains respiratory disease related information for the acoustic signal output in step (a) using the respiratory disease related features extracted in step (c). It provides a method of providing respiratory disease information, including the step of calculating.

In one embodiment, after the step (a) and before the step (b), (a) the filter (f) further comprises the step of filtering the acoustic signal other than the 50 to 1000 Hz band, the step (d) may further include the step of calculating, by the information calculating unit 43, whether the sound signal includes wheezing and crackling sounds by using the respiratory disease-related features extracted in step (c).

In an embodiment, in the step (b), the conversion unit 41 converts the sound signal output from the sensor unit 30 into a first converted sound in the form of intensity of the sound signal over time. converting the signal into a signal, the transforming unit 41 performing a Fourier transform on the first transformed sound signal to convert the first transformed sound signal into a second transformed sound signal in the form of intensity of the sound signal according to a frequency band, and the transforming unit (41) converts the sound signal output from the sensor unit 30 into a third converted sound signal in the form of a spectrogram, wherein the third converted sound signal is the frequency and intensity of the sound signal over time may include a step in which is expressed.

In an embodiment, in the step (c), the feature extraction unit 42 extracts a signal having a first peak from the second converted sound signal as a fundamental signal, and the second conversion The method may further include extracting a peak having a center frequency n times the center frequency of the basic signal (n is an integer greater than or equal to 2) in the acoustic signal as an nth harmonic signal.

In an embodiment, the step (c) may further include the step of extracting, by the feature extraction unit 42, an inspiration duration and an expiration duration from the first converted sound signal.

In an embodiment, in step (d), when the nth harmonic signal is extracted in step (c), the information calculating unit 43 calculates that the sound signal includes wheezing, or The method may further include calculating that the sound signal includes wheezing when the ratio between the fundamental signal and the nth harmonic signal is greater than a preset first ratio.

In an embodiment, in the step (d), when the ratio calculated by the information calculating unit 43 through the following equation is greater than a preset first ratio, the sound signal includes wheezing It may further include the step of calculating the

[Equation]

In an embodiment, in the step (d), the information calculating unit 43 determines that the expiratory duration is longer than a preset first time and the ratio of the expiratory duration to the inspiration duration is preset in a second second When the ratio is greater than the ratio, calculating that the sound signal includes wheezing, and when the information calculating unit 43 determines that the inspiratory duration is within a preset second time range, the sound signal includes a whistling sound It may further include the step of calculating.

In addition, the present invention provides a program stored in a computer-readable recording medium for executing the above-described method.

According to the present invention, convenience is improved because the respiratory disease of the subject to be analyzed can be calculated just by attaching the patch to the skin of the subject to be analyzed without requiring skilled medical staff.

In addition, a horn-shaped sound collecting passage is formed through the patch, so that the sensitivity of the collected sound is improved.

In addition, long-term monitoring of respiratory disease is possible while the patch is attached to the subject being analyzed.

In addition, since features related to respiratory diseases are extracted from various types of converted acoustic signals, and respiratory diseases are calculated using the extracted features, the accuracy of the calculation is improved.

In addition, it is also possible to provide an administration unit capable of administering a drug to the subject to be analyzed, and to calculate whether or not the respiratory disease is improved by administering the drug to the administration unit, and accordingly, it is possible to adjust the drug dosage.

In addition, since the cover part is provided, the influence of the sound introduced from the outside rather than the analysis target object is minimized.

1 is a schematic view for explaining a respiratory disease sensing patch according to an embodiment of the present invention.

FIG. 2 is a schematic block diagram illustrating the respiratory disease sensing patch of FIG. 1 .

3 is a schematic block diagram illustrating a respiratory disease sensing system according to an embodiment of the present invention.

4 is an exploded perspective view illustrating the respiratory disease sensing patch of FIG. 1 .

5 is a view showing an actual appearance of the respiratory disease sensing patch of FIG. 3 , (a) the respiratory disease sensing patch to which the cover part is not applied (b) the respiratory disease sensing patch to which the cover part is applied.

FIG. 6 is a view for explaining the structure of an absorption part of the respiratory disease sensing patch of FIG. 1 .

FIG. 7 is a view for explaining first to third converted sound signals obtained by converting sounds collected using the respiratory disease sensing patch of FIG. 4 .

8 is a flowchart illustrating a method for providing information on respiratory diseases according to an embodiment of the present invention.

9 and 10 are diagrams of results according to verification experiment 1 performed using the respiratory disease sensing patch according to an embodiment of the present invention.

11 is a view for explaining the shape of the converted acoustic signal appearing in normal people and asthma patients, and the shape of the converted acoustic signal appearing in a crackling sound and a wheezing sound.

12 is a view of the results of verification experiment 2 for explaining the shape of the converted acoustic signal appearing in normal persons, asthmatics, and chronic obstructive pulmonary disease using the respiratory disease sensing patch according to an embodiment of the present invention.

FIG. 13 is a view showing results according to verification experiment 3 for explaining the shape of a converted acoustic signal appearing in a normal asthma patient using the respiratory disease sensing patch according to an embodiment of the present invention.

14 is a view for explaining the shape of an acoustic signal generated in a normal person, an acoustic signal including a wheezing sound, and an acoustic signal including a crackling sound.

15 to 17 are views according to verification experiment 4 for explaining the level of external noise when the respiratory disease sensing patch according to an embodiment of the present invention is compared with a respiratory disease sensing patch without a cover part.

18 is a view according to verification experiment 5 comparing the respiratory disease sensing patch according to an embodiment of the present invention with a conventionally used Littmann electronic stethoscope.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1. Respiratory disease sensing patch

1 and 2, a respiratory disease sensing patch according to an embodiment of the present invention will be described in more detail.

1 is a schematic diagram illustrating a respiratory disease sensing patch according to an embodiment of the present invention, and FIG. 2 is a block diagram of the respiratory disease sensing patch of FIG. 1 .

1 and 2 , the respiratory disease sensing patch according to an embodiment of the present invention includes an adhesive unit 10 , a substrate unit 20 , a sensor unit 30 , an operation unit 40 , an absorption unit 50 , and a sound insulation unit. (60), it may include a cover portion (70).

The adhesive part 10 is a part attached to the analysis target object. More specifically, it is attached to the skin of the subject to be analyzed, and the location where it is attached is not particularly limited, but it is preferably attached to the chest or back in order to collect lung sounds, breathing sounds, and heart sounds with high sensitivity.

A material of a hydrocolloid component may be applied to the surface of the adhesive part 10 to be attached to the object to be analyzed.

The first sound collecting passage 11 is formed through the adhesive portion 10 in the vertical direction. The first sound collection passage 11 may have a shape in which its inner diameter decreases as it moves away from the object to be analyzed, that is, it may have a shape in which its diameter decreases as it moves away from the object to be analyzed. As the first sound collecting passage 11 has the above shape, it is possible to solve the echo and noise problems that occur in the sensor unit 30 .

The substrate unit 20 is formed on the bonding unit 10 , and a sensor unit 30 and a calculation unit 40 to be described later are installed. That is, it is a part that serves as a base for the operation of the sensor.

A second sound collecting passage 21 is formed through the substrate 20 in the vertical direction. The second sound collection passage 21 communicates with the first sound collection passage 11 and extends vertically so that the inner diameter is the same even if it is farther away from the object to be analyzed. Also, the inner diameter of the second sound collection passage 21 may be the same as the inner diameter of the uppermost portion of the first sound collection passage 11 .

As the first sound collection passage 11 and the second sound collection passage 21 are not formed of material, but are formed as empty spaces, sounds such as breathing sound, lung sound, and heart sound generated inside the object to be analyzed are produced. It is possible to directly face the sensor unit 30 through the first sound collection passage 11 and the second sound collection passage 21 . Thus, the performance of sound collection is improved.

The sensor unit 30 is installed on the substrate unit 20, and the sound introduced from the outside, specifically, comes out of the skin of the subject to be analyzed and passes through the first sound collection passage 11 and the second sound collection passage 21. Outputs a sound signal according to the sound.

For example, the sensor unit 30 may be a MEMS microphone (Micro Electro-Mechanical System Microphone) including a membrane that moves in response to a change in atmospheric pressure caused by sound waves. However, if it is a sensor capable of outputting an acoustic signal according to the sound generated by the object to be analyzed, it is not particularly limited thereto.

In order to improve sound collection performance, the sensor unit 30 is preferably installed above the first sound collection passage 11 and the second sound collection passage 21 , and in FIG. 1 , one sensor unit 30 is formed. Although an embodiment is shown, an embodiment in which two or more sensor units are installed is also possible.

As the MEMS microphone is used as the sensor unit 30, distortion can be minimized in a band of 50 Hz to 1000 Hz, which is a region corresponding to heart sounds and breathing sounds, and it is possible to obtain a flat frequency response in the corresponding band.

The sensor unit 30 collects the sound that has passed through the first sound collection passage 11 and the second sound collection passage 21 (that is, the sound generated inside the object to be analyzed) and outputs it as a sound signal, In response to the sound coming from the outside, it is output as a sound signal. For example, external noise caused by friction between clothes and the sensor unit 30, external living environment noise (vehicle noise, etc.) may also be detected by the sensor unit 30, and since the noise corresponds to noise, the present invention The respiratory disease sensing patch according to the embodiment further includes an absorption unit 50 , a sound insulating unit 60 , and a cover unit 70 .

The absorber 50 is located above the sensor unit 30 and is made of a porous material (see FIG. 6 ). As shown in FIG. 6 , various types of porous structures may be applied, and it is possible to attenuate the vibration of sound introduced from the outside through the porous structure. Accordingly, it is minimized that the sound introduced from the outside reaches the sensor unit 30 , so that the sensor unit 30 can collect only the sound emitted from the object to be analyzed.

The sound insulating unit 60 is located above the absorption unit 50 , and serves to minimize the sound introduced from the outside together with the absorption unit 50 to reach the sensor unit 30 . Since the sound insulating part 60 is formed, the efficiency of external noise reduction can be maximized compared to the use of the absorbing part 50 alone.

The cover part 70 is located above the sound insulation part 60 and is configured to minimize noise caused by friction between an external object such as clothes and the patch according to the present invention.

For example, as shown in FIG. 1 , it may have a hemispherical dome shape, and a contact area with an external object may be minimized and a friction coefficient may be minimized through micro-nano-scale patterning.

The acoustic signal output from the sensor unit 30 is transmitted to the operation unit 40 , and the operation unit 40 calculates whether the object to be analyzed has a respiratory disease by using the acoustic signal.

The operation unit 40 may be a microcontroller unit (MCU) or a computer having a data operation and processing function including the conversion unit 41 , the feature extraction unit 42 , and the information operation unit 43 .

The conversion unit 41 converts the acoustic signal (electrical signal) output from the sensor unit 30 by a predetermined method. The sound signal converted by the conversion unit 41 is referred to as a converted sound signal for convenience.

The operation unit 40 according to an embodiment of the present invention may further include a filter f, and the filter f filters the remaining sound signals except for the sound signals in a preset band, specifically, in the 50 Hz to 1000 Hz band. do. When the filtering process by the filter (f) is performed, only the sound signal of the preset band remains, and the converter 41 converts the sound signal that has undergone the filtering process.

The conversion unit 41 converts the acoustic signal output from the sensor unit 30 into three data types.

As a first data format, the conversion unit 41 converts the acoustic signal into the shape of the intensity of the acoustic signal over time. That is, the sound signal may be converted in a graph form with time as the x-axis and the intensity of the sound signal as the y-axis (refer to the middle diagram of FIG. 7 ). This is referred to as a first converted acoustic signal.

As the second data form, the transform unit 41 performs Fourier transform on the first transformed sound signal, and converts it into the form of intensity of the sound signal according to the frequency band. That is, the acoustic signal may be converted in the form of a graph with the frequency as the x-axis and the intensity of the acoustic signal as the y-axis (refer to the upper drawing of FIG. 9 ). This is referred to as a second converted sound signal.

As a third data format, the converter 41 converts the acoustic signal into a spectrogram format. That is, the acoustic signal may be converted into a graph in which time is the x-axis, frequency is the y-axis, and the intensity of the acoustic signal is expressed in color (refer to the lower drawing of FIG. 9 ). This is referred to as a third converted acoustic signal.

The above-described data conversion may be performed using a conventional technique, and a detailed description of the conversion process will be omitted.

The feature extraction unit 42 extracts respiratory disease-related features by a predetermined method using the data converted by the transformation unit 41 .

First, the feature extraction unit 42 calculates the signal strength during inspiration/expiration, inhalation duration, and exhalation duration in the first converted acoustic signal (intensity of the acoustic signal over time). Extract the features of (duration).

In general, in the acoustic signal, the inspiration signal appears before the expiration signal (see FIG. 9 ), and the feature extraction unit 42 extracts the characteristics of the duration of the inspiration signal and the duration of the expiration signal.

In another embodiment, the feature extraction unit 42 may extract the features of the inspiration duration and the expiration duration from the third converted sound signal in the form of a spectrogram, and the inspiration duration and expiration duration from the two types of data, respectively By extracting the temporal features, the accuracy of the extraction can be improved.

In addition, the feature extraction unit 42 extracts a signal having a first peak from the second converted acoustic signal (intensity of the acoustic signal according to frequency) as a fundamental signal, and the central frequency of the basic signal. frequency), when a peak having a center frequency n times (n is an integer of 2 or more) exists, it is extracted as an nth harmonic signal and the center frequency of the nth harmonic signal is extracted.

The information calculating unit 43 calculates the respiratory disease-related features extracted by the feature extraction unit 42, more specifically, the signal strength during inspiration/expiration, the inspiratory duration, the expiratory duration, the center frequency of the fundamental signal, and the nth harmonic. Using the center frequency of the signal, it is calculated whether the acoustic signal output from the sensor unit 30 includes a wheezing sound and a crackle sound, and whether or not a respiratory disease is present.

In one embodiment, when the n-th harmonic signal is extracted by the feature extraction unit 42, the information calculating unit 43 may calculate that the acoustic signal includes wheezing, and may calculate that there is a respiratory disease, More specifically, it can be calculated as having asthma.

In another embodiment, the information operation unit 43 is configured to generate a wheezing sound when the center frequency of the n-th harmonic signal to the center frequency of the fundamental signal extracted by the feature extraction unit 42 is greater than a preset first ratio. It can be calculated by including, it can be calculated as having a respiratory disease, and more specifically, it can be calculated as having asthma.

The ratio of the center frequency of the nth harmonic signal to the center frequency of the fundamental signal may be calculated through Equation 1 below.

[Formula 1]

Here, p _fundamental is the center frequency of the fundamental signal, and p _harmonics is the center frequency of the nth harmonic signal.

The information calculating unit 43 may calculate whether the acoustic signal includes a wheezing or a crackling sound using the signal strength during inspiration/expiration extracted by the feature extraction unit 42, the inspiration duration, and the expiration duration.

Specifically, when the intensity of the inspiratory signal included in one respiratory cycle consisting of inspiration/expiration is less than the intensity of the expiration signal, the information calculating unit 43 may calculate that the acoustic signal includes wheezing, and If the intensity is greater than the intensity of the exhalation signal, it can be calculated that the acoustic signal includes a crackling sound (see FIG. 14 ).

In addition, the information calculating unit 43 calculates that the sound signal includes wheezing when the expiration duration is longer than the first preset time and the ratio of the expiration duration to the inspiration duration is greater than the preset second ratio. can Here, the preset first time may be 72 ms to 88 ms, more specifically, 80 ms. In addition, when the sound signal is continuous, the expiratory duration is longer than the first preset time, and the ratio of the expiratory duration to the inspiratory duration is greater than the preset second ratio, the sound signal is continuous. It can also be calculated that the signal contains wheezing (see FIG. 14).

Also, when the inspiration duration is within the preset second time range, the information calculating unit 43 may calculate that the acoustic signal includes a crackling sound. Here, the preset second time range may be 5 ms to 15 ms. Also, when the acoustic signal is discontinuous and the inhalation duration is within the preset second time range, the information calculating unit 43 may calculate that the acoustic signal includes a crackling sound (refer to FIG. 14 ).

The information calculating unit 43 may calculate that the subject to be analyzed has asthma if the acoustic signal includes wheezing, and may calculate that the subject to be analyzed has chronic obstructive pulmonary disease if the acoustic signal includes a whistling sound. . In addition, when it is calculated that the acoustic signal does not include any wheezing or crackling sounds, it may be calculated that the subject to be analyzed does not have a respiratory disease (normal).

The respiratory disease sensing patch according to an embodiment of the present invention may include a control unit 44 , a storage unit 45 , and a communication unit 46 .

The control unit 44 controls the administration unit 80 provided in the patch. Medicines are stored in the dispensing unit 80 , and when the cover of the lower portion of the dispensing unit is opened, the medicaments in the dispensing unit 80 may be administered to the subject to be analyzed. It is preferable that a plurality of dosage units are provided in one patch. In another embodiment, since the amount of the drug stored in each medication unit is different from each other, it is also possible to adjust the amount of the drug administered to the subject to be analyzed later according to the calculation result of the information calculating unit 43 .

When it is calculated by the information calculating unit 43 that the subject to be analyzed has a respiratory disease such as asthma or chronic obstructive pulmonary disease, the control unit 44 transmits a dosing signal to open the administration unit 80 to the administration unit 60 ) will be sent to When the dosing signal is transmitted to the dispensing unit 80, the lower portion of the dispensing unit 80 is opened so that the medicament stored in the dispensing unit 80 can be administered to the subject to be analyzed.

According to the information calculated by the information calculating unit 43, the amount of the drug administered to the subject to be analyzed through the administering unit 80 may be adjusted. For example, it may be calculated that there is a mild or severe respiratory disease based on the frequency of occurrence of wheezing, the strength of the wheezing signal, and the like. When it is calculated that there is no respiratory disease, the control unit 44 may control the administration unit 80 not to be opened. In addition, when it is calculated that there is a mild respiratory disease, the control unit 44 may control the degree of opening of the dispensing unit 80 so that 50% of the medicine stored in the dispensing unit 80 is administered. In addition, when it is calculated that there is a severe respiratory disease, the control unit 44 may control the degree of opening of the dispensing unit 80 so that all (ie, 100%) of the drugs stored in the dispensing unit 80 are administered. However, this is only an example, and it will be said that the opening degree of the administration unit 80, time, etc. can be controlled in various ways.

The storage unit 45 stores acoustic signals collected from the analysis target entity, respiratory disease-related features extracted by the feature extraction unit 42 , and information calculated by the information operation unit 43 .

In an environment in which communication with the external computer 90 through the communication unit 46 to be described later is impossible, the information calculated by the operation unit 40 may be temporarily stored in the storage unit 45 , and communication with the external computer 90 . Even when this is possible, information calculated by the operation unit 40 may be stored in the storage unit 45 according to user settings.

The communication unit 46 communicates with the external computer 90 . In order to perform wired/wireless communication with the external computer 90 , for example, a configuration having a Bluetooth function may be applied. However, if a communication protocol such as WiFi or Zigbee is applied to perform communication with the external computer 90 , it will not be particularly limited thereto. The information stored in the storage unit 45 through the communication unit 46 may be transmitted to the external computer 90, and when the administration signal is transmitted from the external computer 90, the control unit 44 controls the administration unit 80 ) to transmit a dosing signal.

The respiratory disease sensing patch according to the present invention further includes a temperature sensor and a position sensor.

The temperature sensor is configured to measure a temperature of an analysis target object to which the respiratory disease sensing patch is attached, and the position sensor is configured to calculate a position of the respiratory disease sensing patch. As an example, the location sensor may be a GPS sensor that calculates a location using a GPS method, but any sensor capable of calculating a location of a configuration in which the location sensor is installed may be applied.

The temperature information measured by the temperature sensor and the position information calculated by the position sensor are transmitted to the calculating unit 40, and the calculating unit 40 uses the temperature information transmitted from the temperature sensor to sense a respiratory disease, and is transmitted from the position sensor. Real-time location monitoring of the corresponding analysis target object can be performed using the location information.

2. Respiratory disease sensing system

3, a respiratory disease sensing system according to another embodiment of the present invention will be described in more detail.

A respiratory disease sensing system according to an embodiment of the present invention includes a respiratory disease sensing patch 100 and an external processing device 200 .

Compared to the above-described respiratory disease sensing patch performing all functions of sound collection, sound signal preprocessing, feature extraction, and respiratory disease-related information calculation, the respiratory disease sensing patch 100 in the respiratory disease sensing system collects sound. , there is a difference in that only pre-processing of the acoustic signal is performed, and feature extraction from the pre-processed acoustic signal in the external processing device 200 and operation of respiratory disease-related information are performed.

The respiratory disease sensing patch 100 includes the above-described adhesive part 10 , the substrate part 20 , the sensor part 30 , the communication part 46 , the absorption part 50 , the sound insulation part 60 , the cover part 70 , It includes a dosage unit 80, a temperature sensor and a position sensor, and further includes a filter (f).

The filter f filters the remaining sound signals except for the sound signals in the preset band, specifically, in the 50 Hz to 1000 Hz band.

The filtered acoustic signal is transmitted to the external processing device 200 through the communication unit 46 , and the external processing device 200 receives the acoustic signal, extracts respiratory disease-related features, and uses the extracted features for respiratory diseases Calculate related information.

In order to perform wired/wireless communication with the external processing device 200, the communication unit 46 may have, for example, a configuration having a Bluetooth function, and is not particularly limited thereto, and communication protocols such as WiFi and Zigbee are applied to the external processing device Wired/wireless communication with the 200 may be performed.

The external processing device 200 is a device having a communication function, and for example, a device having an arithmetic function, such as a smart phone, a tablet PC, or a computer, may be applied.

The external processing device 200 includes a conversion unit 210 , a feature extraction unit 220 , an information operation unit 230 , a control unit 240 , a storage unit 250 , and a communication unit 260 .

The transformation unit 210 , the feature extraction unit 220 , the information operation unit 230 , and the storage unit 250 are the transformation unit 41 , the feature extraction unit 42 , and the information operation unit 43 in the respiratory disease sensing patch described above. ) and performs the same function as the storage unit 45 , and a detailed description thereof will be omitted.

By the information calculating unit 230, when the acoustic signal is calculated to include a crackling sound or a wheezing sound, and it is calculated that the subject to be analyzed has a respiratory disease, the control unit 240 sends the medication signal to the respiratory disease sensing patch 100 send.

When the administration signal is transmitted, the administration unit 80 is opened to administer the medicine to the subject to be analyzed.

When the respiratory disease sensing system is compared with the respiratory disease sensing patch, data calculations such as feature extraction and information calculation are all performed by the external processing device 200 . Compared with the embodiment in which data calculations such as feature extraction and information calculation are performed within the respiratory disease sensing patch, the patch itself has the advantage of minimizing power consumption and heat generation.

3. How to provide information on respiratory diseases

Hereinafter, a method for providing respiratory disease information according to an embodiment of the present invention will be described in detail with reference to FIG. 8 .

First, the sensor unit 30 collects sound introduced from the outside and outputs a sound signal (S71).

Next, the converters 41 and 210 convert the sound signal output from the sensor unit 30 into a converted sound signal by a predetermined method (S72). The converted sound signals converted by the converters 41 and 210 are shown in FIGS. 9 to 14 .

Next, the feature extraction units 42 and 220 extract respiratory disease-related features by using the converted acoustic signal (S73). Respiratory disease-related features extracted by the feature extraction units 42 and 220 are signal strength during inspiration/expiration, inspiration duration, expiration duration, fundamental signal, center frequency of the fundamental signal, the nth harmonic signal, and the nth harmonic It may include the center frequency of the signal. Since a detailed description thereof has been described above, it will be omitted.

Next, the information calculating units 43 and 230 use the respiratory disease-related features extracted by the feature extracting units 42 and 220 to determine whether the acoustic signal output from the sensor unit 30 includes wheezing and crackling sounds, and asthma and asthma symptoms. It is calculated whether or not a respiratory disease including chronic obstructive pulmonary disease (S74).

When the n-th harmonic signal is extracted from the acoustic signal by the feature extracting units 42 and 220 , the information calculating units 43 and 230 may calculate that the acoustic signal includes wheezing, and may calculate that there is a respiratory disease. It can be, and more specifically, it can be calculated as having asthma.

In addition, when the center frequency of the n-th harmonic signal with respect to the center frequency of the basic signal extracted by the feature extraction units 42 and 220 is greater than a preset first ratio, the information calculating units 43 and 230 determine that the sound signal is wheezing. It can be calculated as including , it can be calculated as having a respiratory disease, and more specifically, it can be calculated as having asthma.

The ratio of the center frequency of the nth harmonic signal to the center frequency of the fundamental signal may be calculated through Equation 1 below.

[Formula 1]

Here, p _fundamental is the center frequency of the fundamental signal, and p _harmonics is the center frequency of the nth harmonic signal.

In addition, in the signal strength during inspiration/expiration extracted by the feature extraction units 42 and 220, when the strength of the inspiration signal included in one breathing cycle is less than the strength of the expiration signal, the acoustic signal includes wheezing. It can be calculated as that, and when the strength of the inspiration signal is greater than the strength of the expiration signal, it can be calculated that the acoustic signal includes a crackling sound.

In addition, when the expiratory duration extracted by the feature extraction units 42 and 220 is longer than the first preset time and the ratio of the expiratory duration to the inspiration duration is greater than the preset second ratio, the information calculating unit 43 , 230) may be calculated as including the wheezing sound, may be calculated as having a respiratory disease, and more specifically, may be calculated as having asthma. Here, the preset first time may be 72 ms to 88 ms, more specifically, 80 ms.

In addition, when the inspiratory duration extracted by the feature extracting units 42 and 220 is within a preset second time range, the information calculating units 43 and 230 may calculate that the acoustic signal includes a crackling sound, and the respirator It can be calculated as having a disease, and more specifically, it can be calculated as having chronic obstructive pulmonary disease. Here, the preset second time range may be 5 ms to 15 ms.

When the acoustic signal is calculated as having a respiratory disease by the information calculating unit 43, 230, the control unit 44, 240 transmits a medication signal for opening the medication unit 60 provided in the patch to the medication unit 60 do. Accordingly, the drug stored in the administering unit 60 may be administered to the subject to be analyzed to which the patch is attached.

4. Validation Experiment

(1) Verification Experiment 1

The respiratory disease sensing patch according to an embodiment of the present invention was attached to a human model to which a Lung Sound Simulator was attached, which generates the sound observed in a normal person, the sound observed in an asthma patient, and the sound observed in a chronic obstructive pulmonary disease patient, A verification experiment was conducted to determine whether it is possible to extract only the breathing signal depending on the presence of sounds generated from the Lung Sound Simulator and additional sounds (singing, walking, etc.).

As a result of the experiment, even though normal (normal sound signal), wheeze (wheeze), normal+HB (normal sound signal + heart sound signal), and wheeze+HB (wheeze + heart sound signal) are generated, each signal can be clearly distinguished. was confirmed (FIG. 9).

Also, although normal+Song (normal sound signal + song signal), wheeze+Song (wheeze + song signal), normal+walk (normal sound signal + gait), and wheeze+walk (wheeze + gait) are generated, each It was confirmed that the signals of can be clearly distinguished (FIG. 10), thus confirming that respiratory disease-related features can be stably extracted even in an environment in which external noise is introduced.

(2) Verification Experiment 2

The respiratory disease sensing patch according to an embodiment of the present invention was attached to a human body model equipped with a Lung sound simulator that generates the sound observed in normal people, the sound observed in asthma patients, and the sound observed in chronic obstructive pulmonary disease patients, The sound generated in the lung sound simulator was collected by the sensor unit 30 and the sound signal outputted accordingly was converted.

As a result of the experiment, it was confirmed that the acoustics observed in asthma patients and the acoustics observed in patients with chronic obstructive pulmonary disease were irregular compared to the acoustics observed in normal persons. It was confirmed that they were distinguished from normal persons in the characteristics (see FIG. 12 ).

(3) Verification Experiment 3

The respiratory disease sensing patch according to an embodiment of the present invention was attached to a human body model equipped with a Lung sound simulator that generates the sound observed in normal people, the sound observed in asthma patients, and the sound observed in chronic obstructive pulmonary disease patients, The sensor unit 30 collects the sound and Fourier transforms the sound signal output accordingly to obtain the converted sound signal as shown in FIG. 13 .

In the case of the sound observed in normal persons, only one fundamental signal is observed in the converted sound signal, but in the case of the sound observed in asthma patients, in addition to the fundamental signal in the converted sound signal, n (n is 2) of the center frequency of the fundamental signal It was confirmed that the nth harmonic signal (the second harmonic signal and the third harmonic signal are observed in FIG. 13) having a center frequency multiple of the integer) was observed.

Accordingly, when the n-th harmonic signal has a center frequency n (n is an integer of 2 or more) times the center frequency of the basic signal, it can be confirmed that the sound signal includes a wheezing sound, and the sound signal is a wheezing sound. It could be confirmed that the inclusion of asthma becomes a strong basis for judging that a person has asthma.

(4) Verification Experiment 4

When the respiratory disease sensing patch according to the embodiment of the present invention and the respiratory disease sensing patch in which the absorption unit 50, the sound insulating unit 60, and the cover unit 70 do not exist at different distances, only wheezing is generated , an experiment was performed to measure the effect of wheezing + external noise (pre-recorded frequencies 256Hz, 512Hz, 1024Hz sound).

As a result of the experiment, the respiratory disease sensing patch (right diagram of FIG. 15, W/_COVER) according to an embodiment of the present invention compared with the control group (left diagram of FIG. 15, W/O_cover), external noise (512Hz, 1024Hz) It could be confirmed that about 25% of the was attenuated (FIG. 16).

That is, in the case of the respiratory disease sensing patch according to the embodiment of the present invention, the sound to be detected (wheezing, wheezing, It was confirmed that it was possible to effectively reduce external noise generated in a band other than the band in which the noise was generated (refer to FIG. 17 ).

Therefore, it was confirmed that the influence of external sound was reduced, and thus the sensitivity of sound generated in the body of the subject to be analyzed could be improved.

(5) Verification Experiment 5

A respiratory disease sensing patch according to an embodiment of the present invention and a conventional Littmann stethoscope were attached to each chronic obstructive pulmonary disease patient to collect sound.

Acoustic signals were collected by attaching a respiratory disease sensing patch to the same site or by bringing a Littmann stethoscope to the same site, and the results shown in FIG. 18 were obtained.

The signal corresponding to the band in which wheezing was not observed was removed, and the spectrogram as in FIG. 18 was obtained through the above-described process such as Fourier transform (the spectrogram obtained when the upper side of FIG. 18 is a spectrogram obtained when using the patch according to the present invention), it was confirmed that wheezing that occurs periodically in patients with chronic obstructive pulmonary disease can be detected with higher accuracy than the conventional Littmann stethoscope ( arrow in Fig. 18).

The method according to an embodiment of the present invention described above may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to carry out the operations of the present invention, and vice versa.

In the above, the present specification has been described with reference to the embodiments shown in the drawings so that those skilled in the art can easily understand and reproduce the present application, but these are merely exemplary, and those skilled in the art can make various modifications and equivalents from the embodiments of the present application. It will be appreciated that embodiments are possible. Therefore, the protection scope of the present application should be defined by the claims.

(Explanation of symbols)

10: adhesive part

11: first sound collection passageway

20: substrate part

21: second sound collection passageway

30: sensor unit

40: arithmetic unit

41: conversion unit

42: feature extraction unit

43: information calculation unit

44: control unit

45: storage

46: communication department

50: absorption part

60: sound insulation

70: cover part

80: dosage unit

90: external computer

100: patch

200: external processing device

210: conversion unit

220: feature extraction unit

230: information calculation unit

240: control unit

250: storage

260: communication unit

Claims (22)

Adhesive part 10 attached to the skin of the subject to be analyzed and through which the first sound collection passage 11 for collecting the sound coming out through the attached skin is formed; a sensor unit 30 installed to be positioned above the sound collection passage 11 and configured to collect the sound passing through the sound collection passage 11 and output a sound signal; a conversion unit 41 for converting the sound signal output from the sensor unit 30 into a converted sound signal through a preset method; a feature extracting unit 42 for extracting respiratory disease-related features from the converted sound signal converted by the conversion unit 41 by a preset method; an information calculating unit 43 for calculating respiratory disease-related information on the collected sound by using the respiratory disease-related features extracted by the feature extracting unit 42; and The medicine is stored therein, and the administration unit 80 is opened so that the medicine is administered to the skin according to the information calculated by the information operation unit 43; including; Respiratory disease sensing patch. According to claim 1, The first sound collection passage 11 is reduced toward the sensor unit 30, Respiratory disease sensing patch. 3. The method of claim 2, The substrate part 20 is formed between the adhesive part 10 and the sensor part 30 and through which the second sound collection passage 21 connected to the first sound collection passage 11 is formed through; do, The inner diameter of the second sound collection passage 21 is the same even as it faces the sensor unit 30, but is the same as the inner diameter of the upper portion of the first sound collection passage 11, Respiratory disease sensing patch. 3. The method of claim 2, an absorption part 50 which is located on the sensor part 30 and made of a porous material; a sound insulating part (60) positioned above the absorption part (50); and Further comprising a cover portion 70 that covers the upper portion of the sound insulation portion 60, Respiratory disease sensing patch. According to claim 1, The conversion unit 41, Converting the sound signal output from the sensor unit 30 into a first converted sound signal in the form of intensity of the sound signal over time, Fourier transforms the first transformed sound signal into a second transformed sound signal in the form of intensity of the sound signal according to a frequency band, Converts the sound signal output from the sensor unit 30 into a third converted sound signal in the form of a spectrogram, wherein the third converted sound signal expresses the frequency and intensity of the sound signal over time, Respiratory disease sensing patch. 6. The method of claim 5, The feature extraction unit 42, A signal having a first peak in the second converted acoustic signal is extracted as a fundamental signal, and in the second converted acoustic signal, a center frequency of n (n is an integer greater than or equal to 2) times the center frequency of the basic signal If there is a peak having a frequency, it is extracted as an nth harmonic signal, Respiratory disease sensing patch. 7. The method of claim 6, The information calculating unit 43, When the nth harmonic signal is extracted by the feature extraction unit 42, it is calculated that the sound signal includes a wheezing sound, or the sound using the ratio of the basic signal and the nth harmonic signals. Calculate whether a signal contains wheezing, Respiratory disease sensing patch. 8. The method of claim 7, The information calculating unit 43 calculates that the sound signal includes wheezing when the ratio calculated through the following equation is greater than a preset first ratio, [Equation]

Respiratory disease sensing patch. 6. The method of claim 5, The feature extraction unit 42, extracting an inspiration duration and an expiration duration from the first converted acoustic signal; The information calculating unit 43, calculating whether the acoustic signal includes a wheezing or crackle sound by using the inspiration duration, the expiration duration, and the ratio of the expiration duration to the inspiration duration, Respiratory disease sensing patch. 10. The method of claim 9, The information calculating unit 43, Calculating that the acoustic signal includes wheezing when the expiration duration is longer than a first preset time and the ratio of the expiration duration to the inspiration duration is greater than a second ratio, Respiratory disease sensing patch. 10. The method of claim 9, The information calculating unit 43, Calculating that the acoustic signal includes a crackling sound when the inspiration duration is within a preset second time range, Respiratory disease sensing patch. 12. The method of claim 10 or 11, The information calculating unit 43 calculates that there is a respiratory disease when it is calculated that the sound signal includes a wheezing sound or a crackling sound, Respiratory disease sensing patch. 13. The method of claim 12, When it is calculated that there is a respiratory disease by the information calculating unit 43, the administering unit 60 is opened and the drug is administered to the skin of the subject to be analyzed, Respiratory disease sensing system. As a method using the patch according to claim 2, (a) outputting a sound signal by the sensor unit 30 collecting sound introduced from the outside; (b) converting, by the converter 41, the sound signal output in step (a) into a converted sound signal by a preset method; (c) extracting, by the feature extraction unit 42, respiratory disease-related features from the converted acoustic signal converted in step (b) by a preset method; and (d) calculating, by the information calculating unit 43, respiratory disease-related information on the acoustic signal output in step (a) using the respiratory disease-related features extracted in step (c); , How to provide information on respiratory diseases. 15. The method of claim 14, After step (a) and before step (b), (a) the filter (f) further comprising the step of filtering the acoustic signal other than the 50 to 1000 Hz band, Step (d) is, Calculating, by the information calculating unit 43, whether the acoustic signal includes wheezing and crackling sounds by using the respiratory disease-related features extracted in step (c); How to provide information on respiratory diseases. 16. The method of claim 15, The step (b) is, converting, by the conversion unit 41, the sound signal output from the sensor unit 30 into a first converted sound signal in the form of an intensity of the sound signal over time; converting the first transformed sound signal into a second transformed sound signal in the form of intensity of the sound signal according to a frequency band by the transforming unit 41 performing a Fourier transform; and The conversion unit 41 converts the sound signal output from the sensor unit 30 into a third converted sound signal in the form of a spectrogram, wherein the third converted sound signal has frequency and sound according to the passage of time. The signal strength is expressed, comprising; How to provide information on respiratory diseases. 17. The method of claim 16, Step (c) is, The feature extraction unit 42 extracts a signal having a first peak in the second converted sound signal as a fundamental signal, and n(n) of the center frequency of the fundamental signal in the second converted sound signal is an integer greater than or equal to 2) when there is a peak having a multiple of the center frequency, further comprising the step of extracting it as an n-th harmonic signal, How to provide information on respiratory diseases. 17. The method of claim 16, Step (c) is, The step of extracting, by the feature extraction unit 42, an inspiration duration and an expiration duration from the first converted acoustic signal; further comprising How to provide information on respiratory diseases. 18. The method of claim 17, Step (d) is, When the nth harmonic signal is extracted in the step (c), the information calculating unit 43 calculates that the sound signal includes wheezing, or a ratio between the basic signal and the nth harmonic signal is preset When it is greater than the first ratio, calculating that the sound signal includes wheezing; further comprising, How to provide information on respiratory diseases. 20. The method of claim 19, Step (d) is, Calculating, by the information calculating unit 43, that the sound signal includes wheezing when the ratio calculated through the following equation is greater than a preset first ratio; further comprising [Equation]

How to provide information on respiratory diseases. 18. The method of claim 17, Step (d) is, When the information calculating unit 43 determines that the expiratory duration is longer than a preset first time and the ratio of the expiratory duration to the inspiration duration is greater than a preset second ratio, the sound signal includes wheezing arithmetic; and Calculating, by the information calculating unit 43, that the sound signal includes a crackling sound when the inspiration duration is within a preset second time range; How to provide information on respiratory diseases. stored in a computer-readable recording medium to execute the method according to any one of claims 14 to 21; program.

Images