TW201400088A - Snore detecting equipment - Google Patents

Abstract

A snore detecting equipment for a sleeper and it includes a storage module and a control device. The storage module records the sleeper's audio signal during his sleep. The control device electrically connects to the storage module and has a pre-processing unit, a cutting unit, a conversion unit and a graphic generating unit. The pre-processing unit removes the background noise of the audio signal. The cutting unit cut the signal into several snoring events that occur in different time. The conversion unit calculates different frequency spectrum and their strength of the snoring events. The graphic generating unit defines a reference coordinate system and integrates the different frequency spectrum and their strength of the snoring events in the reference coordinate system and depicts a snoring map.

Description

Sleep snoring equipment

The invention relates to a sleep snoring device, in particular to a method for detecting and recording the snoring of a Sleep-Disordered Breathing patient and estimating the risk of Obstructive Sleep Apnea. A sleep snoring device that accurately and objectively evaluates the effectiveness of snoring treatment.

Sleep-disordered breathing is a common disease worldwide, with age groups including infants, young people, and the elderly. The Chang Gung Hospital Sleep Center conducted a survey of patients over 15 years old and found that 51.9% of the respondents had snoring and 2.6% had been seen Sleep apnea.

Referring to Fig. 1, the part where the click is generated mainly occurs in the soft palate portion 91, the tongue root portion 92 or the epiglottis portion 93 of the throat, and the vibration frequency range of the buzzing sound is different in different portions.

Known sleep snoring equipment includes: a neck microphone, a vibration sensor, and a pressure oscillation detector; among them, the disadvantages of the neck microphone include Disadvantages of falling off, being disturbed by breath sounds or speech, and narrow income bandwidth; shortcomings of neck-mounted vibration detectors include easy falling off, easy to be disturbed by breathing or talking vibrations, and inability to detect click spectrum and volume; Disadvantages of the device include the inability to measure the actual amount of noise, the inability to detect the click spectrum, and the inability to use it alone. Disadvantages of current sleep snoring equipment include:

First, it is not easy to obtain continuous snoring data for continuous nights, resulting in reduced credibility; standard multi-functional sleep physiology (Polysomnography) must be carried out in the sleep examination room, and must be obtained through the professional machine of the hospital. However, the restricted examination room site and the health care fund, the routine multi-function sleep physiology check is a one-night (six-hour) night sleep test. A study of nighttime humatory variability found that even a 6-hour nighttime sleep test, compared to a continuous four-night continuous snoring test, had a relatively low confidence in the measurement of one-night snoring data.

Second, it is not designed to detect the intensity of snoring or snoring associated with the severity of obstructive sleep apnea; multi-functional sleep physiology is often measured by other respiratory airflow or temperature changes, electromyography, brain, and blood oxygenation. To monitor the occurrence of sleep apnea and its classification (obstructive, central, or mixed), so usually only the snoring analysis such as snoring time and snoring index, but not the severity of obstructive sleep apnea The intensity of the hum or the spectrum of the hum.

3. Accurate and objective assessment of treatment outcomes cannot be performed; these detection techniques are not related to the subjective perception of snoring patients or family members, nor can they be used as a reference for snoring treatment.

As a result, a snoring detection device in the form of a home has emerged, allowing patients to continuously monitor the actual situation of snoring at home.

However, the shortcomings of these systems include the use of microphones with different sensitivities, unfixed measurement distances, inability to analyze spectral features, inability to record for long periods of time, and inability to extend to standard sleep examination rooms, and cannot be a standard for snoring assessment or treatment. .

Accordingly, it is an object of the present invention to provide a sleep that overcomes the deficiencies of the prior art, detects and records the snoring of a patient with sleep disordered breathing, predicts the risk of obstructive sleep apnea, and accurately and objectively evaluates the effectiveness of the snoring treatment. Beep the device.

Thus, the sleep snoring inspection apparatus of the present invention is for use by a subject, and includes a storage module and a control device.

The storage module records an audio signal generated by the subject during the sleep period; the control device is electrically connected to the storage module, and has a pre-processing unit, a cutting unit, a converting unit and a graphic generating unit. The pre-processing unit removes the background noise from the audio signal, and the cutting unit cuts the pre-processed signal into a plurality of signals respectively representing the buzzing events occurring at different times, and the converting unit calculates the signals according to the signals of the buzzing events. The frequency distribution and signal strength of the click event are converted into a spectrum. The graphics generation unit defines a reference coordinate system, and the reference coordinate system integrates the spectrum of each of the click events and depicts a click map.

In order to overcome the problem that the existing system can only display a single click event and cannot observe more details of the click event in the single-platform system, the reference coordinate system of the graphic generating unit has a first axis, a second axis and a third axis. The first axis defines a frequency interval, the second axis defines a signal strength interval, the third axis defines a plurality of event indexes, and each event index represents each of the click events, and the click coordinates overlap each of the clicks The spectrum of the event to depict the beep map.

In order to overcome the problem of not being able to accurately classify the buzzing event, the control The device further includes a classification unit, the classification unit defining a first threshold value and a second threshold value, and performing low frequency sound on the frequency distribution of each of the click events according to the first threshold value and the second threshold value, The judgment of medium frequency sound and high frequency sound. Preferably, the classification unit can also determine the sum of the humming events, and if it is mainly low frequency sound, it is classified into the first type, and the main low frequency sound and the middle frequency sound are classified into the second type, mainly the low frequency sound and The high-frequency sound is classified into the third type, and the low-frequency sound, the intermediate frequency sound, and the high-frequency sound are classified into the fourth type.

In order to overcome the problem of the subject being used at home, the sleep snoring device further includes a recording device and a host, the recording device including a near-side audio sensor, a processing module, a memory module and a transmission interface. The proximity audio sensor is disposed adjacent to the subject to generate a near-side sensing signal, and the processing module controls the recording of the near-side sensing signal as a first memory data in the memory module. And controlling the first memory data to be outputted to the host by the transmission interface; the host includes the storage module, the control device, a transmission module and a display device, wherein the transmission module is electrically connected to the control device and the a transmission interface, the control device controls the transmission module to receive the first memory data transmitted by the transmission interface, and dumps the data to the storage module as a source of the audio signal, the display device is configured to display the generated by the control device The buzz map.

In order to overcome the traditional lack of subjective assessment of the snoring by the family member, the recording device also has a side that is remote from the subject to instantly generate a side sensing signal for the side (family side) of the subject. The audio sensor controls the side sensing signal to be recorded as a second memory data in the memory module, and controls the second memory data to be output to the host through the transmission interface.

In order to overcome the lack of positioning, the recording device further has a base and two fixing components, and the fixing components respectively fix the proximal audio sensor and the distal audio sensor to the base. Different positions on the seat at a fixed distance. The pedestal has a plurality of spaced apart bumps, and each of the fixing components has a plurality of accommodating holes for limiting the bumps; or the pedestal has a plurality of spaced apart accommodating holes, and each of the pedestals The fixing component has a bump that can be limited by each of the receiving holes. The base has a ruler and a slide rail for measuring a distance between the fixing components, wherein the slide rails are movably fixed in the slide rails of the fixing components, thereby measuring The distance between the tester and the family.

The utility model of the sleep snoring inspection device of the invention has the advantages of overcoming the lack of the existing system, configuring the spectrum of a plurality of snoring events in the reference coordinate system and drawing a snoring map, so that the problem of the subject and the family can be quickly observed, and The single-label system observes more details of the details of the snoring event, and can accurately classify the snoring events; in addition, the subject can be used at home without affecting the accuracy and easy to track before and after treatment.

The foregoing and other objects, features, and advantages of the invention are set forth in the <RTIgt;

Referring to Figure 2, in a preferred embodiment of the present invention, the sleep snoring apparatus 100 includes a recording device 1 and a host 5, wherein the recording device 1 is for use by a subject 8, the recording device 1 including a processing module The group 10, a sound module 11, a power supply module 12, a memory module 14, and a transmission interface 13.

The subject 8 is, for example, a patient with snoring symptoms or obstructive sleep apnea, and the radio module 11 is a microphone capable of recording natural nighttime sleep snoring, and the recording time is continuous for six hours, after the completion of the recording. Data analysis, and the second snoring recorded six months after treatment, can be used as an objective examination of changes in the severity of snoring.

The power supply module 12 is powered by a battery or a commercial power supply. In addition, it is necessary to measure the background noise of the day before the inspection, the time is 10 minutes, as a correction basis after the click analysis, and recording at a sampling rate of 44100 Hz. The files are stored in a designated folder of the memory module 14.

The background noise is referred to the Environmental Protection Agency's environmental noise correction method. The noise noise (L _S ) must be 10 dB higher than the background noise (L _B ). If the difference (L _S -L _B ) is below 10 dB, Table 1 is corrected; if the difference is below 3dB, another measurement scheme is found.

Referring to FIG. 3, in order to synchronously measure the snoring information of the patient position and the family position, thereby effectively improving the reliability of the snoring information, the recording apparatus 1 of FIG. 2 further has a near-side audio sensor 111 and a side audio. The sensor 112, a base 31, and two fixing components 321, 322, the base 31 has a ruler 313 and a slide rail 310, and the ruler 313 is convenient for the user to observe the fixed component 321, The distance between the 322, the slide rail 310 facilitates the movably fixing of the respective fixing components 321, 322 within the slide rail 310. Each of the fixing components 321 and 322 fixes the proximal audio sensor 111 and the distal audio sensor 112 on different positions of the fixed distance of the base 31. In this embodiment, the base 31 has a plurality of spaced intervals. The receiving hole 312 and each of the fixing components 321 and 322 have protrusions 331 and 332 that can be restricted by the receiving holes 312.

In other embodiments, the pedestal 31 has a plurality of spaced apart bumps, and each of the fixing components 321 and 322 has a plurality of accommodating holes for limiting the bumps, and is also within the scope of the present invention.

The proximity audio sensor 111 is disposed adjacent to the subject 8, such as, but not limited to, placed next to the pillow of the subject 8 to detect the click actually heard by the family member, to generate a near side by instantaneous radio reception. The sensing module 10 controls the recording of the near-side sensing signal as a first memory data in the memory module 13, and controls the first memory data to be output to the host 5 through the transmission interface 14.

The side audio sensor 112 is disposed away from the subject 8 , for example, but may be, but not limited to, placed next to the pillow of the family member to immediately generate a side to the side of the subject of the subject 8 . The processing module 10 controls the side sensing signal to be recorded as a second memory data in the memory module 13, and controls the second memory data to be output to the host 5 through the transmission interface 14. Subsequent analysis processing can be performed by the host 5 to further accurately measure the actual condition of the subject 8 and the family member for the occurrence of the snoring to perform appropriate treatment and post-treatment follow-up after treatment.

Referring to FIG. 2, the host 5 includes a control device 2, a display device 3, a storage module 4, and a transmission module 211. The control device 2 controls the display device. 3. Operation of the storage module 4 and the transmission module 211.

In this embodiment, the transmission interface 14 of the recording device 1 and the transmission module 211 of the host 5 are all standard transmission interfaces, for example, a USB standard transmission interface, and the control device 2 can control the transmission module 211 to receive the transmission interface 14 A memory data is transferred to the storage module 4 as a source of the audio signal to facilitate the subsequent processing analysis by the control device 2, and the display device 3 is configured to display a click map generated by the control device 2 after processing the analysis, as shown in the figure. The vocal maps 501-504 of FIG. 12 to FIG. 12 are examples (described later), but it is known to those skilled in the art that the manner of variation is not limited to the disclosure of the specification and the drawings.

The host device 5 can be used in conjunction with the recording device 1 or can be used offline with the recording device 1. The audio module generated by the subject 8 during the sleep period is pre-recorded in the storage module 5; the control device 2 has a pre-processing The unit 21, a cutting unit 22, a converting unit 23, a graphic generating unit 24 and a sorting unit 25 are mainly described with reference to Fig. 2 and with the aid of the respective figures.

Referring to FIG. 4, the time domain waveform 401 of a plurality of click events obtained from the sound file for a period of time, since the background noise covers 40 Hz of noise, in order to avoid background noise interference, the pre-processing unit 21 firstly analyzes the click sound to be analyzed. After the file is read in, the time domain waveform 401 of the plurality of click events is appropriately filtered, and the high-pass filter is used to filter the background noise, and the cutoff frequency is 40 Hz.

Referring to FIG. 5, the time domain waveform 402 of the energy type is removed. In this embodiment, the signal of the background noise is removed for the cutting unit 22 and a plurality of generations are cut out. The main purpose of the signal of the buzzing event occurring at different times is to convert the analog audio signal of multiple buzzing events into the energy-mode time domain waveform 402, and then set a threshold according to the energy value of the background noise, which is used as the sound zone. Judgment basis for the segment/unvoiced segment.

Referring to FIG. 6, the cutting unit 22 is mainly based on the temporal characteristics of the voiced/unvoiced section. In this embodiment, the duration of the click (T _duration , referred to as T _du ) is about 0.6~3 seconds on average, so According to this characteristic, the duration of each sound (T _du ) is judged. If the duration (T _du ) is within the range, the signal is retained, and if the duration (T _du ) is not within the range, the signal is signaled. Filtered out, the result of determining the voiced segment/unvoiced segment can be represented by a square wave 601, so that each voiced segment can be treated as a single click event and a plurality of click events can be cut.

Referring to FIG. 7, the conversion unit 23 calculates the frequency distribution and signal strength of each click event according to the signal of each click event and converts it into a spectrum. In this embodiment, the conversion formula uses Fast Fourier Transform (FFT). The formula can convert a time domain signal into a frequency domain signal. The frequency spectrum waveform 403 in the frequency domain signal is a frequency distribution representing a certain sound event, and the spectrum waveform 404 is a frequency distribution of the background noise.

Referring to FIG. 8, if a plurality of click events are overlapped and the spectrum waveform 405 of the outer profile is taken, the main frequency of occurrence is more prominent, so that it is easy to detect.

Referring to Figure 9, the graphics generation unit 24 defines a reference coordinate system, and configures the spectrum of each click event (represented by the index S(1)-S(9)) in the reference coordinate system and depicts the click map 501 at each event. Bar graph height generation Table intensity, the bar graph of each beep event can be expressed in different color depths in different color depths, such as high frequency, intermediate frequency and low frequency.

Referring to FIG. 10, in order to be able to observe more detailed changes of each click event in a single reference coordinate system, the reference coordinate system may have a first axis, a second axis, and a third axis, the first axis defining a frequency interval, The two axes define a signal strength interval, the third axis defines a plurality of event indices S(1)-S(9) and each event index S(1)-S(9) represents each click event and overlaps with the reference coordinate system. The spectrum of each click event is depicted to depict a click map 502, which is presented in a three-axis reference coordinate system.

Referring to Figures 11 and 12, in addition to being presented in a three-axis reference coordinate system, a two-axis reference coordinate system can also be presented, for example; the click map 503 represents different click events with different color lines, and the click map 504 is represented by different color areas. Different humming events, well known to those skilled in the art, and other similar graphical representations are also within the scope of the present invention.

In addition, it has been found that the distribution of the snoring sound can be roughly divided into three regions below which are defined as low frequency, intermediate frequency, and high frequency. The low frequency arpeggio is equivalent to the soft 颚 91 as shown in FIG. As produced by the epiglottis 93 of Figure 1, the high frequency arpeggio is equivalent to the tongue root 92 of Figure 1.

Referring to FIG. 13, the classification unit 25 defines a first threshold value and a second threshold value, and performs frequency distribution and low frequency sound, medium frequency sound, and high frequency sound for each click event according to the first threshold value and the second threshold value. In this embodiment, the first threshold is set to 300 Hz and the second threshold is set to 850 Hz. The main reason is that most of the low frequencies are mostly within 300 Hz during the development process, and the intermediate frequency mostly falls within 300~. Between 850 Hz, the last high frequency part is 850 Between ~2000 Hz, the first threshold is set to 300 Hz and the second threshold is set to 850 Hz, and the number of clicks, sound pressure level (A-weighting) and frequency distribution are calculated separately.

Referring to FIG. 14 to FIG. 17, the classification unit 25 can also determine the sum of the humming events. If the sum of the humming events is mainly low frequency sounds, as shown in FIG. 14, the first type is classified; if the humming events are The sum is mainly low frequency sound and medium frequency sound, as shown in Fig. 15, which is classified into the second type; if the sum of the sound events is mainly low frequency sound and high frequency sound, as shown in Fig. 16, it is classified into the third type; The sum of the buzzing events has both low frequency, medium frequency and high frequency sounds, as shown in Figure 17 as the fourth type.

In summary, the function of the sleep snoring device of the present invention is to overcome the lack of the existing system, and to configure a spectrum of a plurality of snoring events in the reference coordinate system and to draw a snoring map, so that the problem of the subject can be quickly observed. It can also observe more details of the details of the snoring event in a single standard system, and can accurately classify the snoring events. In addition, the subject can be used at home without affecting the accuracy, and it is easy to track before and after treatment, so it can indeed The object of the invention is achieved.

The above is only the preferred embodiment of the present invention, and the scope of the invention is not limited thereto, that is, the simple equivalent changes and modifications made by the scope of the invention and the description of the invention are All remain within the scope of the invention patent.

100‧‧‧Sleeping snoring equipment

1‧‧‧recording device

10‧‧‧Processing module

11‧‧‧ Radio Module

111‧‧‧Near-side audio sensor

112‧‧‧side audio sensor

12‧‧‧Power supply module

13‧‧‧Transport interface

14‧‧‧Memory Module

211‧‧‧Transmission module

2‧‧‧Control device

21‧‧‧Pre-processing unit

22‧‧‧Cutting unit

23‧‧‧Transition unit

24‧‧‧Graphic generation unit

25‧‧‧Classification unit

3‧‧‧Display device

31‧‧‧ Pedestal

310‧‧‧Slide rails

312‧‧‧ accommodating holes

313‧‧‧ rule

321, 322‧‧‧ fixed components

331, 332‧‧ ‧ bumps

4‧‧‧Storage module

401-402‧‧‧Time domain waveform

403-405‧‧‧Spectral waveform

5‧‧‧Host

501-504‧‧‧鼾声地图

601‧‧‧Fangbo

8‧‧‧ Subjects

1 is a schematic view showing a portion where a click is generated; and FIG. 2 is a system block diagram showing the sleep click sound detecting device of the present invention. The preferred embodiment of the present invention; FIG. 3 is a perspective view showing that the near-side audio sensor and the side audio sensor of the recording device can be assembled at different positions of the base by a fixing component; FIG. 4 is a waveform diagram. A time-domain waveform illustrating a plurality of click events over a period of time; FIG. 5 is a waveform diagram illustrating that the time-domain waveform of the energy pattern cuts a plurality of click events respectively representing different times; FIG. 6 is a waveform diagram illustrating The time characteristic of the voiced section/unvoiced section is filtered; FIG. 7 is a waveform diagram illustrating the frequency distribution of a certain click event and background noise; FIG. 8 is a waveform diagram illustrating the adjusted overlap of a plurality of click events. Spectrum waveform; Figure 9 is a long bar diagram illustrating an example of a humming map in which the distribution of different frequency ranges is indicated by different shades of color in the bar graph of each humming event; Figure 10 is a three-dimensional waveform diagram illustrating the humming map Another example is to superimpose the spectrum of each beeping event; FIG. 11 is a two-dimensional waveform diagram illustrating another example of a beating map with different color lines representing different click events; FIG. 12 is a two-dimensional waveform Described another example of a map in a different color snoring areas represent different snoring events; FIG. 13 is a waveform diagram illustrating the definition of two at the snoring threshold map for determining the low frequency sound, the sound of the midrange and high frequency; 14 is an energy waveform diagram illustrating a first type mainly of low frequency sounds; FIG. 15 is an energy waveform diagram illustrating a second type in which low frequency sounds and medium frequency sounds are mixed; FIG. 16 is an energy waveform diagram illustrating mixing A third type having a low frequency sound and a high frequency sound; and Fig. 17 is an energy waveform diagram illustrating a fourth type having both low frequency sound, medium frequency sound and high frequency sound.

100‧‧‧Sleeping snoring inspection Prepare

1‧‧‧recording device

10‧‧‧Processing module

11‧‧‧ Radio Module

12‧‧‧Power supply module

13‧‧‧Transport interface

14‧‧‧Memory Module

2‧‧‧Control device

211‧‧‧Transmission module

21‧‧‧Pre-processing unit

22‧‧‧Cutting unit

23‧‧‧Transition unit

24‧‧‧Graphic generation unit

25‧‧‧Classification unit

3‧‧‧Display device

4‧‧‧Storage module

5‧‧‧Host

8‧‧‧ Subjects

Claims (10)

A sleep snoring device for use by a subject, comprising: a storage module for recording an audio signal generated by the subject during sleep; and a control device electrically connected to the storage module, having: a pre-processing unit that removes background noise from the audio signal, and a cutting unit that cuts the pre-processed signal into a plurality of signals respectively representing the buzzing events occurring at different times, and a conversion unit calculates the signals according to the respective buzzing events The frequency distribution and signal intensity of each of the click events are converted into a spectrum, and a graphics generating unit defines a reference coordinate system, and the reference coordinate system integrates the spectrum of each of the click events and depicts a click map. The sleep sounding inspection apparatus according to claim 1, wherein the reference coordinate system of the graphic generation unit has a first axis, a second axis, and a third axis, the first axis defining a frequency interval The second axis defines a signal strength interval, the third axis defines a plurality of event indexes, and each event index represents each of the click events, and the reference coordinate system overlaps the spectrum of each of the click events to depict the click map. The sleep snoring inspection apparatus according to claim 1, wherein the control device further comprises: a classification unit, the classification unit defining a first threshold value and a second threshold value, and according to the first threshold Value and the second threshold for each The frequency distribution of the buzzing event is judged by low frequency sound, medium frequency sound, and high frequency sound. The sleep snoring device according to claim 3, wherein the classification unit determines the sum of the snoring events, and if the frequency is mainly low frequency, the first type is mainly classified into a low frequency sound and a medium frequency sound. It is classified into the second type, mainly the low frequency sound and the high frequency sound are classified into the third type, and the low frequency sound, the intermediate frequency sound and the high frequency sound are classified into the fourth type. The sleep snoring inspection apparatus according to claim 1, 2, 3 or 4, further comprising a recording device and a host, the recording device comprising a near-side audio sensor, a processing module, and a memory module And a transmission interface, the proximity audio sensor is disposed adjacent to the subject to generate a near-side sensing signal, and the processing module controls the recording of the near-side sensing signal into the memory module. a first memory data, and controlling the first memory data to be output to the host through the transmission interface; the host includes the storage module, the control device, a transmission module and a display device, and the transmission module is electrically Connecting the control device and the transmission interface, the control device controls the transmission module to receive the first memory data transmitted from the transmission interface, and deposit the same to the storage module as a source of the audio signal, the display device is used for The click map generated by the control device is displayed. The sleep snoring device according to claim 5, wherein the recording device further has a side side that is remote from the subject to immediately generate a side sensing signal for the side of the subject. An audio sensor, the processing module controls the side sensing signal to be recorded as a second memory data in the memory module, and controls the second memory data by using the transmission medium The output is output to the host. The sleep snoring device according to the sixth aspect of the invention, wherein the recording device further has a base and two fixing components, each of the fixing components respectively respectively the proximal audio sensor and the distal audio sense The detector is fixed at a fixed distance from the base. The sleep snoring inspection apparatus according to claim 7, wherein the pedestal has a ruler and a slide rail for measuring a distance between each of the fixing components, the slide rail is provided for each The stationary component is movably secured within the slide rail. The sleep snoring inspection apparatus according to claim 8, wherein the pedestal has a plurality of spaced apart accommodating holes, and each of the fixing components has a bump that can be restricted by each of the accommodating holes. The sleep snoring inspection apparatus according to claim 8, wherein the pedestal has a plurality of spaced apart bumps, and each of the fixing components has a plurality of accommodating holes for each of the bumps.