CN111345782A - Snore type identification method and electronic equipment - Google Patents

Abstract

The invention relates to the technical field of sound identification, and discloses a snore type identification method and electronic equipment. The snore type identification method comprises the following steps: acquiring a snore signal set, wherein the snore signal set comprises a plurality of snore signals, and the snore signals are signals obtained by extracting snore characteristics of physiological signals; determining a first snore type judgment result according to the time domain characteristics of the snore signals; determining a second snore type judgment result according to the frequency domain characteristics of the snore signals; and identifying the snore type of the snore corresponding to the snore signal set according to the first snore type judgment result and the second snore type judgment result. Therefore, the snore type is identified by synthesizing the snore judging results obtained by the time domain characteristics and the frequency domain characteristics, and the method is high in accuracy and reliable.

Description

Snore type identification method and electronic equipment

Technical Field

The invention relates to the technical field of sound identification, in particular to a snore type identification method and electronic equipment.

Background

Snoring is a rough breath sound accompanying respiration after falling asleep. In a sleeping state of a human body, factors such as relaxation of muscles of the throat, collapse of tissues of the throat and the like can cause unsmooth ventilation or obstruction of the upper respiratory tract. When respiratory airflow is obstructed, the soft tissues of the respiratory tract vibrate and then make sounds.

The most objective data for measuring the severity of sleep Apnea is the AHI Index (Apnea-Hypopnea Index), which is the average number of apneas + hypopneas per hour during sleep, with normal values of < 5, 5-15 for mild sleep Apnea syndrome, 15-30 for moderate sleep Apnea syndrome, and > 30 for severe sleep Apnea syndrome. However, the snoring detection apparatus supporting the AHI index cannot accurately and reliably distinguish simple snoring from apnea snoring, and is susceptible to drinking, fatigue, and other conditions.

Disclosure of Invention

An object of the embodiments of the present invention is to provide a snore type identifying method and an electronic device, which can accurately and reliably identify a snore type.

In a first aspect, an embodiment of the present invention provides a snore type identifying method, including: acquiring a snore signal set, wherein the snore signal set comprises a plurality of snore signals, and the snore signals are signals obtained by extracting snore characteristics of physiological signals; determining a first snore type judgment result according to the time domain characteristics of the snore signals; determining a second snore type judgment result according to the frequency domain characteristics of the snore signals; and identifying the snore type of the snore corresponding to the snore signal set according to the first snore type judgment result and the second snore type judgment result.

In a second aspect, embodiments of the present invention provide a non-transitory computer-readable storage medium storing computer-executable instructions for causing a robot to perform any one of the snore type identification methods.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any one of the snore type identification methods.

Compared with the prior art, in the snore type identification method provided by each embodiment of the invention, firstly, a snore signal set is obtained, the snore signal set comprises a plurality of snore signals, and the snore signals are signals obtained by extracting snore characteristics of physiological signals. Secondly, determining a first snore type judgment result according to the time domain characteristics of the snore signals. And thirdly, determining a second snore type judgment result according to the frequency domain characteristics of the snore signals. And finally, identifying the snore type of the snore corresponding to the snore signal set according to the first snore type judgment result and the second snore type judgment result. Therefore, the snore type is identified by integrating the snore judging results obtained by the time domain characteristics and the frequency domain characteristics, and the method is high in accuracy and reliable.

Drawings

One or more embodiments are illustrated by way of example in the accompanying drawings, which correspond to the figures in which like reference numerals refer to similar elements and which are not to scale unless otherwise specified.

Fig. 1 is a schematic circuit structure diagram of a snore type identifying device according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of a snore type identification method according to an embodiment of the present invention;

FIG. 3a is a time domain waveform of a simple snoring condition provided by an embodiment of the present invention;

FIG. 3b is a time domain waveform of an apnea snore provided by an embodiment of the present invention;

FIG. 4a is a histogram corresponding to FIG. 3 a;

FIG. 4b is a histogram corresponding to FIG. 3 b;

FIG. 5a is a power spectrum corresponding to FIG. 3 a;

FIG. 5b is a power spectrum corresponding to FIG. 3 b;

fig. 6a is a schematic structural diagram of a snore type identifying device according to an embodiment of the present invention;

FIG. 6b is a schematic diagram of the structure of the time domain feature block of FIG. 6 a;

FIG. 6c is a schematic diagram of the structure of a frequency domain feature module in FIG. 6 a;

fig. 6d is a schematic structural diagram of a snore type identifying device according to another embodiment of the present invention;

fig. 7 is a schematic circuit structure diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that, if not conflicted, the various features of the embodiments of the invention may be combined with each other within the scope of protection of the invention. Additionally, while functional block divisions are performed in apparatus schematics, with logical sequences shown in flowcharts, in some cases, steps shown or described may be performed in sequences other than block divisions in apparatus or flowcharts. The terms "first", "second", "third", and the like used in the present invention do not limit data and execution order, but distinguish the same items or similar items having substantially the same function and action.

The embodiment of the invention provides snore type identification equipment. In this embodiment, the snore type identifying device may be configured to facilitate detection of snore, for example, the snore type identifying device may be configured as a pillow structure or the like.

Referring to fig. 1, the snore type identifying device 100 includes a data collecting unit 11, a data processing unit 12, a clock unit 13, an indicating unit 14, a communication unit 15, a main control unit 16 and a power management unit 17.

The data acquisition unit 11 is used for acquiring signals of a human body in a sleep state, and the data acquisition unit 11 can select any suitable sensor to acquire snore signals, for example, the data acquisition unit 11 can select a piezoelectric film sensor which is arranged below the chest cavity, the head or the neck of the human body and used for acquiring micro-motion signals generated by the human body, wherein the micro-motion signals comprise snore signals, heartbeat signals, breathing signals, body movement signals and the like. As another example, the data acquisition unit 11 may select a microphone, which is disposed at a position far away from the human body and is used for remotely acquiring the voice signal of the user, wherein the voice signal includes a snore signal, environmental noise, and the like.

It is understood that the data acquisition unit 11 may employ a piezoresistive sensor, an acceleration sensor, a bone conduction sensor, etc. in order to acquire a vibration signal generated when snoring.

The data processing unit 12 is configured to amplify the signal acquired by the data acquisition unit 11, convert the amplified analog signal into a digital signal, and send the digital signal to the main control unit 16. In some embodiments, the data processing unit 12 includes a charge amplifier 121 and an ADC collecting module 122, where the charge amplifier 121 is configured to amplify the signal collected by the data collecting unit 11 and output the amplified analog signal. The ADC acquisition module 122 is configured to convert the amplified analog signal into a digital signal, and then send the digital signal to the main control unit 16.

The clock unit 13 is used for providing a working master clock, a power-down memory clock and a real-time clock for the main control unit 16. In some embodiments, the clock unit 13 includes an RTC module 131, an RTC backup battery 132, and a master clock passive crystal 133. The RTC module 131 provides a real-time clock for the main control unit 16, the RTC backup battery 132 provides a backup power supply for the RTC module 131, and the master clock passive crystal 133 provides a working master clock for the main control unit 16.

The indication unit 14 is used to generate a prompt signal to convey information of a particular meaning to the user. In some embodiments, the indication unit 14 may select any indication device, for example, the indication unit 14 includes a first LED lamp 141 and a second LED lamp 142, the first LED lamp 141 is used for providing information about whether power supply is normal, and when the power supply is normal, the first LED lamp 141 is turned off. In the case of an abnormality, the first LED lamp 141 displays red. The second LED lamp 142 is used to provide information on whether the communication unit 15 is networked or not, and when the communication unit is networked, the second LED lamp 142 displays green. In the event of an abnormality, the second LED lamp 142 is turned off.

The communication unit 15 is used for communicating with an external communication device, and the snore type identifying device 100 can send data to the external communication device through the communication unit 15 or receive a control instruction from the external communication device. In some embodiments, the communication unit 15 may select any suitable communication module, including a WIFI module, a bluetooth module, a 4G module, or a 5G module, among others. In some embodiments, the communication unit 15 may be configured with a tap key that a user presses for a long time of 3 seconds to cause the communication unit 15 to enter a binding state.

The main control unit 16 is used as a control core of the snore type identifying device 100, and can control the working logic of each circuit unit. In some embodiments, the master control unit 16 may be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), a single chip, an arm (acorn RISC machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components. Also, the master control unit 16 may be any conventional processor, controller, microcontroller, or state machine. The master control unit 16 may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, and/or any other such configuration.

The power management unit 17 supplies power to the above circuit units. In some embodiments, power management unit 17 includes a USB interface 171, a charge management chip 172, a battery 173, a control switch 174, a first LDO chip 175, an LDO reference voltage source 176, and a second LDO chip 177.

During charging, the external adapter is electrically connected to the power management unit 17 through the USB interface 171. The charging management chip 172 is responsible for power management to provide stable power for the above-mentioned circuit units. The battery 173 is used for storing power, and the control switch 174 is used for controlling the power supply loop of the battery 173 and the first LDO chip 175, and the battery 173 and the second LDO chip 177. The first LDO chip 175 is used to generate a stable power and provide it to the data processing unit 12, the indicating unit 14, and the main control unit 16. LDO reference voltage source 176 is used to provide a reference voltage for data processing unit 12. The second LDO chip 177 is used to provide power to the communication unit 15.

It can be understood that the snore type identifying device provided by the embodiment of the invention can effectively identify the snore type by adopting the snore type identifying method explained below.

Referring to fig. 2, fig. 2 is a schematic flow chart of a snore type identification method according to an embodiment of the present invention. As shown in fig. 2, the snore type identifying method S200 includes:

s21, acquiring a snore signal set, wherein the snore signal set comprises a plurality of snore signals;

the snore signal is a signal obtained by extracting snore characteristics from a physiological signal, wherein the physiological signal comprises a breathing signal or a pressure signal and the like.

When extracting the snore characteristics from the physiological signals, in a possible embodiment, the data acquisition unit is a piezoelectric film sensor and is arranged under the chest cavity of the user, wherein the data acquisition unit is set to a sampling frequency of 2000 Hz. The data acquisition unit sends the physiological signal to the main control unit, and the main control unit filters the body movement signal from the physiological signal, for example, filters the signal with the amplitude higher than the preset amplitude.

And then, the main control unit processes the filtered physiological signals by using a band-pass filter, and respectively extracts snore signals, heartbeat signals and respiratory signals from the physiological signals. Since the vibration signals generated by snoring, heartbeat and respiration are respectively in different frequency bands, wherein the vibration signals generated by snoring are in the range of 20-800Hz, the signals in the frequency band of 20-800Hz can be extracted as the snoring sound signals by using the band-pass filter. In some embodiments, the band pass filter may be a butterworth band pass filter.

Then, because the fundamental frequency 50Hz and its harmonic of the power frequency signal are both in the frequency band of the snore signal, a comb trap is designed in the main control unit, which can filter the 50Hz power frequency signal and its frequency-doubling harmonics.

Finally, after the filtering processing, the main control unit performs snore detection on the filtered signals to obtain final snore signals, for example, the main control unit performs snore detection by adopting a self-adaptive zero-crossing rate and short-time energy endpoint detection algorithm. Since the duration of each snore is generally within 1 second due to the respiration of the snore, when windowing, a window of 0.5 second is set as a frame of data, when calculating the next frame of data, the next window is slid forward by 0.1 second relative to the previous window, and so on, and the short-time zero-crossing rate and the short-time energy in each window are calculated respectively.

The short-time zero crossing rate refers to the number of times that a signal value passes through a zero value in each second, and a symbol obtained by multiplying two points is smaller than zero and is recorded as a zero crossing point.

The formula for calculating the short-time energy is as follows:

where i is the number of sample points in each frame, and x (i) is the magnitude of the sample points.

In one possible embodiment, two detection thresholds TH1 and TH2 are set. When the short-term energy of the current window is greater than the detection threshold TH1, the current window is marked as a suspected snore starting point, and then whether the short-term zero-crossing rate of a continuous specified number of frames is greater than the detection threshold TH2 is judged, for example, the short-term zero-crossing rate of continuous 3 frames is greater than the detection threshold TH2, and the window is marked as the snore starting point. And after the master control unit determines the snore starting point, continuing searching backwards. And once the current short-time zero-crossing rate is less than the detection threshold TH2, marking as the snore termination point. By analogy, the main control unit can obtain a plurality of snore signals. In addition, the skilled person can freely select the two detection thresholds TH1 and TH2 according to the actual needs, and the disclosure is not limited herein.

The embodiment adopts the self-adaptive zero-crossing rate and short-time energy endpoint detection algorithm to implement snore detection, on one hand, the detection precision of the snore characteristics is high, and on the other hand, the data volume processing is less than that of other snore detection algorithms, so the extraction speed of the snore characteristics is high.

In a possible embodiment, the master control unit may select the snore detecting algorithm to perform snore detection, and may also select another suitable snore detecting algorithm to perform snore detection.

S22, determining a first snore type judgment result according to the time domain characteristics of the snore signals;

in this embodiment, the time domain feature is used to indicate the amplitude change of each snore signal in time, and the first snore determination result includes a simple snore symptom result or an apnea snore result.

Generally, after falling asleep at night, muscles are relaxed, and the hypertrophic soft tissue part around the pharynx blocks the airway, causing local stenosis of the airway. When the airflow passes through this narrow part, a vortex is generated and causes local soft tissue to vibrate, thereby generating snoring. The sleep snoring does not have apnea, and is called simple snoring or benign snoring, short for snore disease. Severe snorers not only affect well-being but can also develop sleep apnea or ventricular rest episodes, or even frequent sudden awakenings, which is also medically known as obstructive sleep apnea syndrome.

In some embodiments, the user can distinguish the simple snore and the apnea snore by combining the time domain waveform diagrams of the simple snore and the apnea snore in the time domain state.

In some embodiments, the snore type identifying device may be configured with a control logic of the time domain feature model, and the first snore determining result may be obtained only by inputting corresponding parameter data or time domain features into the time domain feature model.

Referring to fig. 3a and fig. 3b, it can be seen that: from a time domain perspective, pure snoring presents a continuous uniform characteristic, with the amplitude and interval of each snore being substantially the same, because pure snoring occurs with breathing, because a collapsed airway flutters with restricted airflow. When the apnea of the obstructive sleep apnea patient is finished, the closed upper airway is suddenly opened, so that the upper airway cavity is opened and closed for many times in a short time to generate sound like blasting, therefore, the snore of the apnea is related to the apnea time, the waveform is complex, and the snore can continuously appear for many times after one-time apnea usually.

In view of the above distinction between the two symptoms, in one embodiment, the first snore determination may be made in the following manner.

Firstly, the main control unit may select continuous N snore signals from the snore signal set according to the time sequence, and then calculate the interval duration of every two adjacent snore signals in the N snore signals to obtain a plurality of interval durations, or may directly calculate the interval duration of every two adjacent snore signals in the snore signal set. For example, the snore signal set is { a }₁,a₂,a₃,a₄,a₅......a_nAnd (4) calculating:

T_a2-T_a1＝ΔT1；

T_a3-T_a2＝ΔT2；

T_a4-T_a3＝ΔT3；

T_a5-T_a4＝ΔT4......。

and so on to obtain N-1 interval duration, wherein T_anThe snore starting point of each snore signal can be the snore ending point of each snore signal.

In a possible embodiment, the main control unit may be configured with a coordinate system in advance, when constructing the coordinate system, the X axis of the coordinate system is divided into a plurality of equal-length time zones according to the time increasing sequence, and the vertical coordinate of the Y axis in the coordinate system is preset as the frequency of the interval duration falling into the same equal-length time zones.

In one possible embodiment, referring to fig. 4a and 4b, considering that the normal human breath interval is between 10-24 times/min, corresponding to each breath interval being 2.5-6S, the X-axis of the coordinate system can be divided into 15 equal-length time zones in time increasing order, the abscissa range is set to-3500 ms-3500ms, and each equal-length time zone is 500 ms.

Secondly, after obtaining a plurality of interval durations, the main control unit counts the plurality of interval durations to generate a histogram, that is, the histogram is generated according to the equal-length time zone corresponding to the X axis of each interval duration falling into the preset coordinate system and the frequency of the interval durations falling into the corresponding equal-length time zone. For example, Δ T1 is 200ms, Δ T2 is 150ms, the frequency his (Δ T1) of the interval duration falling in the range of [ -250,250] is 2, and so on, the frequency his (Δ T2) of the interval duration falling in the range of [ -250, -750] is 1.

As can be seen from fig. 4a and 4 b: because the simple snore symptom has continuous and uniform characteristics, the characteristics are concentrated in the same interval on the histogram and are high and sharp. The snore interval of the apnea snore is influenced by the apnea time, is disordered and irregular, can be scattered in different intervals on the histogram and is in a short and wide shape.

And thirdly, the main control unit calculates a form index of the histogram, wherein the form index is used for describing the change state of the histogram, and the change states of the pure snore and the apnea snore in the histogram are different, so that the pure snore and the apnea snore can be preliminarily identified according to the form index. In one possible embodiment, the morphology index may be any suitable index, such as a rate of change, etc.

In a possible embodiment, the main control unit may determine the maximum frequency count of the histogram in the process of calculating the morphological index of the histogram, that is, may traverse the maximum frequency count from a plurality of frequency counts. Then, the main control unit accumulates all the interval durations to obtain the total interval duration. Then, the main control unit calculates the average value of the total interval duration to obtain the average duration. Finally, the main control unit uses the quotient of the maximum frequency and the average time length as the form index of the histogram.

In one embodiment, the master control unit is slave frequency [ his (Δ T1), his (Δ T2), his (Δ T3.).. his (Δ Tn)]The maximum frequency his (Δ T3) is traversed, i.e., his (Δ T3) is max (his (Δ T)). Then, Δ T₁+ΔT₂+.....+ΔT_n-1＝ΔT_{General assembly}I.e. total interval duration Δ T_{General assembly}. Then, Δ T_{General assembly}/n-1＝ΔT₀I.e. average duration of Δ T₀. Finally, the morphological index

And finally, after the morphological index of the histogram is calculated, the main control unit can generate a first snore judgment result according to the morphological index.

As mentioned above, the intervals of the snoring disease snores are continuous and uniform, and the interval duration mostly falls in the same interval, so compared with the apnea snore, the max (his (Δ T)) of the snoring disease snore is obviously increased, and the average duration Δ T is obviously increased₀This is significantly reduced and the division of the two further increases the difference between the features.

Therefore, when the first snore determination result is generated, in a specific embodiment, the main control unit determines whether the form index is greater than a preset index threshold value, and if so, a simple snore symptom result is generated. If not, generating an apnea snore result.

By constructing the histogram, the distribution condition of each snore signal is described by visual dimensions, the difference between the simple snore and the apnea snore is accurately reflected, and the snore type judgment result is accurately and reliably obtained.

S23, determining a second snore type judgment result according to the frequency domain characteristics of the snore signals;

the frequency domain feature is used for indicating amplitude variation of each snore signal in frequency, and the second snore judgment result comprises a simple snore symptom result or an apnea snore result.

The snore type identifying device can be configured with control logic of the frequency domain characteristic model, and a second snore judging result can be obtained as long as corresponding parameter data or time domain characteristics are input into the frequency domain characteristic model.

In one possible embodiment, the user can distinguish the simple snore and the apnea snore by combining the frequency domain oscillograms of the simple snore and the apnea snore in the frequency domain state.

For example, the power spectrum between 20Hz and 200Hz, please refer to FIG. 5a and FIG. 5b, which show that: the snore of the pure snore symptom is continuous and uniform, and is shown as an obvious main peak form on a frequency domain, and the main peak and most of energy are mainly located between 20Hz and 200 Hz. The snore of the apnea is caused by the sudden opening and closing of the airway when the apnea is finished, the snore presents a form that the snore is high in frequency and large in amplitude and then gradually goes to low frequency and small in amplitude, the peak value and the energy distribution are scattered on a frequency domain, and the snore is distributed in high and low frequency areas. Generally, the pure snore has single frequency and concentrated energy, the snore of apnea has a plurality of peaks, the energy is dispersed, and the main peak frequency is higher.

In view of the above two symptoms difference, in a possible embodiment, the frequency domain features may be extracted in the following manner to obtain the second snore determination.

First, the master control unit calculates the power spectrum of each snore signal.

In a possible embodiment, the master control unit intercepts snore fragments of a preset duration from each snore signal, wherein the starting point of each snore fragment is the snore starting point of each snore signal, for example, generally, the range from the starting point to the ending point of each snore is 0.5-5S, and therefore, for each snore, the snore fragment from the starting point to the ending point of 0.5S is selected for power spectrum analysis.

Further, the master control unit averagely divides each snore fragment into a plurality of data segments, and each two adjacent data segments have repeated fragments with preset proportion, for example, 0.5S snore fragments are averagely divided into 8 data segments, and each two adjacent data segments have 50% of repeated fragments.

Further, the main control unit performs windowing on each data segment to obtain a plurality of windowed data segments, for example, the main control unit performs hamming windowing on each data segment to obtain 8 windowed data segments.

Further, the main control unit performs Fourier transform on each windowed data segment to obtain a plurality of power spectrums.

Further, the main control unit calculates the average value of a plurality of power spectrums to obtain an average power spectrum, and the average power spectrum is used as the power spectrum of each snore signal.

In addition to calculating the power spectrum of each snore signal in the manner described above, the master control unit may calculate the power spectrum of each snore signal according to any suitable power spectrum algorithm, such as an average periodogram method, a window function method, a modified periodogram averaging method, or a weighted overlap averaging method (Welch method).

In this embodiment, the frequency of each power spectrum is divided into a plurality of frequency intervals in the order from low to high, for example, the frequency of the power spectrum is divided into a first frequency interval, a second frequency interval and a third frequency interval, wherein the first frequency interval is 20Hz to 200Hz, the second frequency interval is 220Hz to 400H, and the third frequency interval is 620Hz to 800 Hz.

Secondly, after the power spectrum of each snore signal is obtained, the main control unit obtains the weighting result of each snore signal according to the power spectrum of each snore signal, and the weighting result is used for quantitatively indicating the frequency characteristic of the snore signal.

In a possible implementation manner, the main control unit calculates the total power of the target snore signal in a plurality of frequency intervals according to the power spectrum of the target snore signal, wherein the target snore signal is any one of a plurality of snore signals.

Then, the main control unit sequentially divides the total power corresponding to the high frequency interval in the two adjacent frequency intervals by the total power of the low frequency interval in the two adjacent frequency intervals to obtain a plurality of weighting indexes corresponding to the target snore signal, for example, the main control unit obtains the following equation: and calculating the weighting indexes, namely K1, E2/E1, and K2, namely E3/E2. Wherein, K1 and K2 are weighting indicators, E1 is the total power corresponding to the first frequency interval, E2 is the total power corresponding to the second frequency interval, and E3 is the total power corresponding to the third frequency interval.

Then, the main control unit performs weighted summation on a plurality of weighted indexes corresponding to the target snore to obtain a weighted result of the target snore signal, for example, according to the following weighted equation: m_iP1K 1+ P2K 2, the weighted result of the ith snore signal, M_iThe weighting result for the ith snore signal is P1 and P2. Since the energy of the first half is more concentrated, the weight P1 may be set to be greater than the weight P2, e.g., 0.75 for P1 and 0.25 for P2.

In general, when the master control unit obtains the weighted result of each snore signal, the master control unit accumulates the weighted results of the snore signals to obtain a total weighted result, for example, M₁+M₂+M₃+......+M_N＝M_{General assembly}Wherein M is_{General assembly}Is the overall weighted result.

According to the embodiment, a weighting mode is adopted to carry out weighting operation on each snore signal, on one hand, the characteristics of each snore signal belonging to different snore types in a designated frequency band can be evaluated in an all-around manner through the weighting mode in consideration of the difference of the total power of the simple snore and the apnea snore in different frequency bands, so that the snore type of each snore signal can be accurately and reliably judged in a frequency domain angle. On the other hand, the method also accumulates the weighting results of a plurality of snore signals and then continues the judgment, and can effectively eliminate the interference of noise data caused by some emergency situations, thereby more accurately and reliably judging the snore type of each snore signal in a frequency domain angle.

And finally, the main control unit generates a second snore judgment result according to the total weighting result.

In one embodiment, the master unit averages the total weighted result, e.g., M_{General assembly}Mp. Then, the main control unit judges whether the total weighting result is largeAnd if so, generating a simple snore symptom result. If not, generating an apnea snore result.

In another specific embodiment, the main control unit may also directly compare the total weighting result with a preset threshold, and if the total weighting result is greater than the preset threshold, a simple snoring result is generated. Otherwise, an apnea snore result is generated.

S24, identifying the snore type of the snore corresponding to the snore signal set according to the first snore type judgment result and the second snore type judgment result.

In a possible embodiment, the main control unit determines whether both the first snore type determination result and the second snore type determination result are the simple snore syndrome results, and if so, identifies that the snore type of the snore is the simple snore syndrome type. If not, identifying that the snore type of the snore is an apnea snore type.

In this embodiment, because the percentage of the population with simple snoring can be small, the determination condition of simple snoring can be more strictly limited, so that when the first snoring type determination result and the second snoring type determination result are both simple snoring results at the same time, the snoring type of snoring can be identified as a simple snoring type, otherwise, the snoring type of snoring is an apnea snoring type.

Therefore, the snore type is identified by synthesizing the snore type judgment results obtained by the time domain characteristics and the frequency domain characteristics, and the method is high in accuracy and reliable.

With the development of apnea snore symptoms, the main frequency and energy of snore move to high frequency, so that the frequency index can be used for representing the snore frequency change condition.

In a possible implementation manner, the median frequency can be selected to represent the variation trend of the frequency, the median frequency is the median of the total power of the selected frequency band, in the power spectrum, the areas on both sides of the median frequency are mutually equal, the main peak and the energy of the snore sound of the simple snore are mainly located in the low frequency band, and the energy of the apnea snore sound is scattered in the whole frequency band, so that the median frequency of the snore sound of the simple snore is less than the apnea snore sound.

In a possible implementation manner, the master control unit calculates the median frequency of each snore signal according to the power spectrum of each snore signal, and generates a snore frequency change report according to the median frequency of each snore signal.

Because the snore frequency change condition is closely related to pure snore and apnea snore, the snore frequency change condition of the user is indicated by using the median frequency, and therefore the snore frequency change condition can simply, rapidly and scientifically reflect the snore change progress of the user in time.

In a specific embodiment, the main control unit calculates a power spectrum of each snore signal in a designated frequency interval, wherein the designated frequency interval is a 20-800Hz frequency band, and the main control unit calculates the power spectrum in the 20-800Hz frequency band according to a power spectrum algorithm.

The user can determine the development trend of the snore disease condition according to the variation trend of the median frequency, and when the disease condition is worsened, the median frequency moves towards the right, that is, the median frequency is in an increasing trend. When the disease condition is better, the median frequency is shifted to the left, i.e., the median frequency is decreasing.

In a specific embodiment, the snore frequency change report includes a snore frequency left shift report or a snore frequency right shift report, and when the snore frequency change report is generated, the master control unit accumulates the median frequency of each snore signal to obtain the total frequency. Then, the main control unit calculates the average value of the total frequency to obtain the average frequency, and if the average frequency is greater than or equal to the preset frequency threshold, a snore frequency right shift report is generated. And if the average frequency is smaller than a preset frequency threshold, generating a snore frequency left shift report. Therefore, the embodiment can simply, rapidly and scientifically reflect the snore change progress of the user in time.

When calculating the median frequency of each snore signal, in one embodiment, the master control unit integrates the power spectrum within the specified frequency interval to obtain the power area of the specified frequency interval, e.g., the power area

Next, the master control unit follows the equation:

and calculating the median frequency of each snore signal, wherein the lower limit f1 is the starting point frequency of the specified frequency interval, the upper limit MDF is the median frequency, P (k) is a power spectral density function, and S is a power area.

By analogy, the master control unit can calculate the median frequency of different snore signals.

By means of an integration method, it is simple and fast to find the median frequency in order to generate a snore frequency change report quickly.

It should be noted that, in the foregoing embodiments, a certain order does not necessarily exist between the foregoing steps, and it can be understood by those skilled in the art from the description of the embodiments of the present invention that, in different embodiments, the foregoing steps may have different execution orders, that is, may be executed in parallel, may also be executed in an exchange manner, and the like.

As another aspect of the embodiments of the present invention, an embodiment of the present invention provides a snore type identifying apparatus. The snore type identifying device can be a software module, the software module comprises a plurality of instructions which are stored in a memory, and the processor can access the memory and call the instructions to execute the instructions so as to complete the snore type identifying method explained in each embodiment.

In some embodiments, the snore type identifying device may also be built by hardware components, for example, the snore type identifying device may be built by one or more than two chips, and each chip may work in coordination with each other to complete the snore type identifying method described in each embodiment. For another example, the snore type identifying device may also be constructed by various logic devices, such as a general processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), a single chip, an arm (acorn RISC machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components.

Referring to fig. 6a, the snore type identifying device 600 includes a snore obtaining module 61, a time domain feature module 62, a frequency domain feature module 63, and a snore type identifying module 64.

The snore acquiring module 61 is configured to acquire a snore signal set, where the snore signal set includes a plurality of snore signals, and the snore signals are signals obtained by extracting snore features of a physiological signal. The time domain feature module 62 is configured to determine a first snore type determination result according to time domain features of the plurality of snore signals. The frequency domain characteristic module 63 is configured to determine a second snore type determination result according to the frequency domain characteristics of the plurality of snore signals. The snore type identifying module 64 is configured to identify a snore type of the snore corresponding to the snore signal set according to the first snore type determination result and the second snore type determination result.

Therefore, the snore type is identified by synthesizing the snore judging results obtained by the time domain characteristics and the frequency domain characteristics, and the snore type identification device is high in accuracy and reliable.

In some embodiments, the snore signal set includes a plurality of snore signals, and referring to fig. 6b, the time domain feature module 62 includes a duration calculation unit 621, a histogram unit 622, and a generation unit 623.

The duration calculating unit 621 is configured to calculate an interval duration of two adjacent snore signals to obtain a plurality of interval durations. The histogram unit 622 is configured to count the interval durations to generate a histogram, where an X axis of the histogram is the duration, the X axis includes multiple equal-length time zones, and a Y axis of the histogram is the frequency of the interval durations falling into the same equal-length time zone. The generating unit 623 is configured to generate a first snore type determination result according to the morphological index of the histogram.

In some embodiments, the generating unit 623 is specifically configured to determine the maximum frequency of the histogram, obtain an average value of a plurality of interval durations, obtain an average duration, and use a quotient of the maximum frequency and the average duration as a morphological index of the histogram, determine whether the morphological index is greater than a preset index threshold, if so, generate a simple snore result, and if not, generate an apnea snore result.

In some embodiments, referring to fig. 6c, the frequency domain characteristic module 63 includes a power spectrum calculating unit 631, a weighting unit 632, and a judging unit 633, wherein the power spectrum calculating unit 631 is configured to calculate a power spectrum of each snore signal. The weighting unit 632 is configured to obtain a weighting result of each snore signal according to the power spectrum of each snore signal, where the weighting result is used to quantify a frequency characteristic of the snore signal. The judging unit 633 is configured to accumulate the weighted results of the plurality of snore signals to obtain a total weighted result, and judge whether an average value of the total weighted result is greater than a preset weighted threshold, if so, generate a simple snore symptom result, and if not, generate an apnea snore result.

In some embodiments, the weighting unit 632 is specifically configured to calculate, according to a power spectrum of the target snore signal, total powers of the target snore signal in multiple frequency intervals, respectively, where the target snore signal is any one of the multiple snore signals, and after the target snore signal is arranged in a sequence from low to high in the frequency intervals, the total power corresponding to a high-frequency interval in two adjacent frequency intervals is sequentially divided by the total power of a low-frequency interval in the two adjacent frequency intervals to obtain multiple weighting indexes corresponding to the target snore signal, and the multiple weighting indexes corresponding to the target snore signal are subjected to weighted summation to obtain a weighting result of the target snore signal.

In some embodiments, the plurality of frequency intervals include a first frequency interval of 20Hz to 200Hz, a second frequency interval of 220Hz to 400H, and a third frequency interval of 620Hz to 800 Hz.

In some embodiments, referring to fig. 6d, the snore type identifying device 600 further includes a frequency calculating module 65 and a frequency varying module 66. The frequency calculation module 65 is configured to calculate a median frequency of each snore signal according to the power spectrum of each snore signal, and the frequency change module 66 is configured to generate a snore frequency change report according to the median frequency of each snore signal.

In some embodiments, the snore frequency change report includes a snore frequency left shift report or a snore frequency right shift report, and the frequency change module 66 is specifically configured to accumulate median frequencies of the snore signals to obtain a total frequency, obtain an average value of the total frequency to obtain an average frequency, generate a snore frequency right shift report if the average frequency is greater than or equal to a preset frequency threshold, and generate a snore frequency left shift report if the average frequency is less than the preset frequency threshold.

In some embodiments, the power spectrum calculating unit 631 is specifically configured to intercept snore fragments with a preset duration from each snore signal, where a starting point of each snore fragment is a snoring starting point of each snore signal, averagely divide each snore fragment into a plurality of data segments, where each two adjacent data segments have a repeated fragment with a preset proportion, perform windowing on each data segment to obtain a plurality of windowed data segments, perform fourier transform on each windowed data segment to obtain a plurality of power spectrums, obtain an average value of the plurality of power spectrums, obtain an average power spectrum, and use the average power spectrum as the power spectrum of each snore signal.

In some embodiments, the snore type identifying module 64 is specifically configured to determine whether the first snore type determination result and the second snore type determination result are both simple snore syndrome results, if so, identify that the snore type of the snore is the simple snore syndrome type, and if not, identify that the snore type of the snore is the apnea snore type.

It should be noted that the snore type identifying device can execute the snore type identifying method provided by the embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method. For technical details that are not described in detail in the embodiment of the snore type identifying device, reference may be made to the snore type identifying method provided by the embodiment of the present invention.

Fig. 7 is a schematic circuit structure diagram of an electronic device according to an embodiment of the present invention. As shown in fig. 7, the electronic device includes one or more processors 71 and a memory 72. Fig. 7 illustrates an example of one processor 71.

The processor 71 and the memory 72 may be connected by a bus or other means, such as the bus connection in fig. 7.

The memory 72 is a non-volatile computer readable storage medium, and can be used for storing non-volatile software programs, non-volatile computer executable programs, and modules, such as program instructions/modules corresponding to the snore type identification method in the embodiment of the present invention. The processor 71 executes various functional applications and data processing of the snore type identifying device by running the non-volatile software program, instructions and modules stored in the memory 72, that is, the snore type identifying method provided by the above method embodiment and the functions of the various modules or units of the above device embodiment are realized.

The memory 72 may include high speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory 72 may optionally include memory located remotely from the processor 71, and such remote memory may be connected to the processor 71 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The program instructions/modules are stored in the memory 72 and, when executed by the one or more processors 71, perform the snore type identification method of any of the method embodiments described above.

Embodiments of the present invention further provide a non-volatile computer storage medium, where the computer storage medium stores computer-executable instructions, which are executed by one or more processors, for example, one processor 71 in fig. 7, so that the one or more processors can execute the snore type identification method in any of the above method embodiments.

An embodiment of the present invention further provides a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, where the computer program includes program instructions, and when the program instructions are executed by an electronic device, the electronic device is caused to execute any one of the snore type identification methods.

Therefore, the electronic equipment integrates the snore judging results obtained by the time domain characteristic and the frequency domain characteristic respectively to identify the type of the snore, and is high in accuracy and reliable.

The above-described embodiments of the apparatus or device are merely illustrative, wherein the unit modules described as separate parts may or may not be physically separate, and the parts displayed as module units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network module units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the above technical solutions substantially or contributing to the related art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; within the idea of the invention, also technical features in the above embodiments or in different embodiments may be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

Claims (12)

1. A snore type identification method is characterized by comprising the following steps:

acquiring a snore signal set, wherein the snore signal set comprises a plurality of snore signals, and the snore signals are signals obtained by extracting snore characteristics of physiological signals;

determining a first snore type judgment result according to the time domain characteristics of the snore signals;

determining a second snore type judgment result according to the frequency domain characteristics of the snore signals;

and identifying the snore type of the snore corresponding to the snore signal set according to the first snore type judgment result and the second snore type judgment result.

2. The method of claim 1, wherein said determining a first snore type determination based on time domain characteristics of said plurality of snore signals comprises:

calculating the interval duration of two adjacent snore signals to obtain a plurality of interval durations;

counting the interval durations to generate a histogram, wherein an X axis of the histogram is the duration, the X axis comprises a plurality of equal-length time zones, and a Y axis of the histogram is the frequency of the interval durations falling into the same equal-length time zone;

and generating a first snore type judgment result according to the form index of the histogram.

3. The method of claim 2, wherein the generating the first snore category determination based on the morphology metric of the histogram comprises:

determining a maximum frequency of the histogram;

calculating the average value of the interval durations to obtain the average duration, and taking the quotient of the maximum frequency and the average duration as the form index of the histogram;

judging whether the morphological index is larger than a preset index threshold value or not;

if so, generating a simple snore symptom result;

if not, generating an apnea snore result.

4. The method of any of claims 1 to 3, wherein determining a second snore type determination based on the frequency domain characteristics of the plurality of snore signals comprises:

calculating the power spectrum of each snore signal;

obtaining a weighting result of each snore signal according to the power spectrum of each snore signal, wherein the weighting result is used for quantitatively indicating the frequency characteristic of the snore signal;

accumulating the weighted results of the plurality of snore signals to obtain a total weighted result;

judging whether the average value of the total weighting results is greater than a preset weighting threshold value or not;

if so, generating a simple snore symptom result;

and if not, generating the apnea snore result.

5. The method of claim 4, wherein said deriving a weighted result for each of said snore signals based on a power spectrum of each of said snore signals comprises:

calculating the total power of the target snore signals on a plurality of frequency intervals respectively according to the power spectrum of the target snore signals, wherein the target snore signals are any snore signals in the plurality of snore signals;

after the frequency intervals are arranged in the sequence from low to high, the total power corresponding to the high frequency interval in the two adjacent frequency intervals is divided by the total power of the low frequency interval in the two adjacent frequency intervals in sequence to obtain a plurality of weighting indexes corresponding to the target snore signal;

and carrying out weighted summation on a plurality of weighted indexes corresponding to the target snore to obtain a weighted result of the target snore signal.

6. The method of claim 5, wherein the plurality of frequency bins comprises a first frequency bin, a second frequency bin, and a third frequency bin;

the first frequency interval is 20Hz-200 Hz;

the second frequency interval is 220 Hz-400H;

the third frequency interval is 620Hz-800 Hz.

7. The method of claim 5, further comprising:

calculating the median frequency of each snore signal according to the power spectrum of each snore signal;

and generating a snore frequency change report according to the median frequency of each snore signal.

8. The method of claim 7, wherein the report of snore frequency change comprises a report of snore frequency left shift or a report of snore frequency right shift, and wherein generating the report of snore frequency change based on the median frequency of each of the snore signals comprises:

accumulating the median frequency of each snore signal to obtain a total frequency;

calculating the average value of the total frequency to obtain an average frequency;

if the average frequency is greater than or equal to a preset frequency threshold, generating a snore frequency right shift report;

and if the average frequency is smaller than a preset frequency threshold, generating a snore frequency left shift report.

9. The method of claim 4, wherein said calculating a power spectrum of each of said snore signals comprises:

intercepting snore fragments with preset time length from each snore signal, wherein the starting point of each snore fragment is the snore starting point of each snore signal;

averagely dividing each snore fragment into a plurality of data sections, wherein repeated fragments with preset proportions exist in every two adjacent data sections;

windowing each data segment to obtain a plurality of windowed data segments;

performing Fourier transform on each windowed data segment to obtain a plurality of power spectrums;

and calculating the average value of the plurality of power spectrums to obtain an average power spectrum, and taking the average power spectrum as the power spectrum of each snore signal.

10. The method according to any one of claims 1 to 4, wherein the identifying, according to the first snore type determination result and the second snore type determination result, the snore type of the snore corresponding to the snore signal set comprises:

judging whether the first snore type judgment result and the second snore type judgment result are both simple snore symptom results;

if so, identifying that the snore type of the snore is a simple snore symptom type;

if not, identifying that the snore type of the snore is an apnea snore type.

11. A non-transitory computer-readable storage medium storing computer-executable instructions for causing a robot to perform the snore type identifying method of any one of claims 1 to 10.

12. An electronic device, comprising:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a snore type identification method as claimed in any one of claims 1 to 10.

Images