Sleep apnea fragment detection method and device based on multi-scale fuzzy entropy

Abstract

The sleep apnea fragment detection method based on the multi-scale fuzzy entropy comprises the following steps: collecting physiological signal data of a user when the user is in a sleep state, and dividing the physiological signal data into a plurality of fragments; extracting an interval sequence from the physiological signal data, and performing coarse graining processing on the interval sequence to obtain a coarse grain sequence; the fuzzy entropy of the coarse grain sequence is solved, and the characteristic parameters of the fuzzy entropy are calculated; and determining the sleep apnea fragment of the user based on the difference degree between the characteristic parameter of the fuzzy entropy and the standard sample. According to the sleep apnea fragment detection method and device based on the multi-scale fuzzy entropy, through calculation of the fuzzy entropy on different scales, the method and device are more sensitive to small changes of interval sequences, and therefore the accuracy of sleep apnea detection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep apnea detection, and in particular to a sleep apnea segment detection method and device based on multi-scale fuzzy entropy. Background Art

[0002] For a long time, sleep apnea has attracted widespread attention due to its widespread incidence and harmfulness to human health. Sleep apnea mainly refers to the phenomenon of disappearance or reduction of respiratory airflow that occurs more than 30 times during a whole night of sleep, each time lasting more than 10 seconds. Obstructive Sleep Apnea (OSA) accounts for a large proportion of sleep apnea, and OSA also has the highest mortality rate. It is characterized by complete or partial intermittent obstruction of the upper airway during sleep, resulting in frequent episodes of apnea and hypopnea.

[0003] In the prior art, sleep apnea is usually evaluated using the ApneaHypopnea Index (AHI), which is the total number of apneas and hypopneas per hour of sleep. The severity of OSA can be divided into the following categories based on AHI: when AHI is less than 5, it is a normal person; when AHI is between 5-15 (not including 15), it is a mild OSA patient; when AHI is between 15-30 (not including 30), it is a moderate OSA patient; when AHI is greater than or equal to 30, it is a severe OSA patient.

[0004] Timely diagnosis of sleep apnea is of great significance. It can effectively assist patients in determining the occurrence of sleep apnea and formulate targeted treatment measures to effectively improve the patient's quality of life and ensure the patient's life safety. In the prior art, sleep apnea detection mainly uses a polysomnography monitor, which can monitor a variety of physiological data during the patient's sleep and can accurately determine the occurrence of sleep apnea. However, this monitor interferes too much with the subject's sleep, affects the patient's sleep quality, and is costly. Summary of the invention

[0005] In order to address the deficiencies in the prior art, the present invention provides a sleep apnea segment detection method and device based on multi-scale fuzzy entropy, which is more sensitive to slight changes in the interval sequence by calculating fuzzy entropy at different scales, thereby improving the accuracy of sleep apnea detection.

[0006] In order to achieve the above object, the specific scheme adopted by the present invention is: a sleep apnea segment detection method based on multi-scale fuzzy entropy, comprising the following steps:

[0007] When the user is in a sleeping state, physiological signal data of the user is collected and divided into a plurality of segments;

[0008] Extracting an interval sequence from the physiological signal data, and performing coarse-graining processing on the interval sequence to obtain a coarse-grained sequence;

[0009] Solve the fuzzy entropy of the coarse-grained sequence and calculate the characteristic parameters of the fuzzy entropy;

[0010] The sleep apnea segment of the user is determined based on the difference between the characteristic parameters of the fuzzy entropy and the standard sample.

[0011] Preferably, the physiological signal data includes electrocardiogram data and pulse wave data, and the physiological signal data is divided into a plurality of segments based on preset segment durations.

[0012] Preferably, the specific method for extracting the interval sequence from the physiological signal data is the Pan-Tompkins algorithm, and the extracted interval sequence includes the RR interval sequence.

[0013] Preferably, the specific method for coarse-graining the interphase sequence includes:

[0014] Constructing original sequence and coarse-grained sequence;

[0015] The original sequence is processed by mean coarse-graining according to the coarse-grained sequence to obtain a coarse-grained sequence.

[0017] The original sequence is X = {X 1 ,X 2 ,…,X N}, where N is the total time length;

[0018] The coarse-grained sequence is Where τ is the scale parameter.

[0019] Preferably, the method for solving the fuzzy entropy of the coarse-grained sequence includes:

[0020] Reconstruct the coarse-grained sequence into an n-dimensional vector in

[0021] Calculate two n-dimensional vectors X i and X j The distance between

[0022]

[0023] Using fuzzy function to calculate two n-dimensional vectors X i and X j The similarity between

[0024] Among them, m is the gradient of the exponential function boundary, and r is the similarity tolerance;

[0025] Constructing a similarity ratio function And represented as an n+1-dimensional vector;

[0026] Calculating fuzzy entropy

[0027] Preferably, the gradient of the exponential function boundary is m=1.8-2.2, and the similarity tolerance is r=0.23-0.26SD, where SD is the standard deviation of the coarse-grained sequence.

[0028] Preferably, when the length of N is finite, the fuzzy entropy is FuzzyEn(n,m,r,N)=lnO m (m,r)-lnO m+1 (m,r).

[0029] Preferably, the characteristic parameters of the fuzzy entropy include a mean MFE_Mean, a standard deviation MFE_Std and a range MFE_Max-Min.

[0030] A sleep apnea segment detection device based on multi-scale fuzzy entropy is used to implement the above-mentioned sleep apnea segment detection method based on multi-scale fuzzy entropy, and the device includes:

[0031] A collection module, used to collect physiological signal data of the user when the user is in a sleeping state;

[0032] A signal processing module, used for dividing the physiological signal data into a plurality of segments, extracting an interval sequence from the physiological signal data, and performing coarse-graining processing on the interval sequence to obtain a coarse-grained sequence;

[0033] The detection module is used to solve the fuzzy entropy of the coarse-grained sequence, calculate the characteristic parameters of the fuzzy entropy, and determine the user's sleep apnea segment based on the difference between the characteristic parameters of the fuzzy entropy and the standard sample.

[0034] The present invention is based on multi-scale fuzzy entropy and is more sensitive to small changes in the interval sequence by calculating fuzzy entropy at different scales, thereby improving the accuracy of sleep apnea detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 is a flow chart of the detection method of the present invention;

[0037] Figure 2 It is a flowchart for solving the fuzzy entropy of coarse-grained sequences;

[0038] Figure 3 It is a schematic diagram of the correlation degree between the first type of fuzzy entropy and physiological signal data;

[0039] Figure 4 It is a schematic diagram of the correlation degree between the first fuzzy entropy and physiological signal data. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] like Figure 1 As shown, a sleep apnea segment detection method based on multi-scale fuzzy entropy includes S1 to S4.

[0042] S1. Collect the user's physiological signal data when the user is in a sleeping state, and divide it into multiple segments. Physiological signal data includes electrocardiogram data and pulse wave data. What is directly obtained during the collection process is an analog signal. It is necessary to first perform analog-to-digital conversion to convert the analog signal into a digital signal that is easier to process, and then perform subsequent processing. Before performing the analog-to-digital conversion, the analog signal needs to be preprocessed. The preprocessing method may include bandpass filtering to remove artificial artifacts, power frequency interference and other noise in the analog signal. When the physiological signal data is divided into multiple segments, it is divided based on a preset segment length. The specific segment length can be determined according to actual needs. For example, the segment length can be set to 1 minute, and the physiological signal data is divided into multiple segments with a length of 1 minute accordingly.

[0043] S2. Extract the interval sequence from the physiological signal data, and perform coarse-graining on the interval sequence to obtain a coarse-grained sequence. The specific method for extracting the interval sequence from the physiological signal data is the Pan-Tompkins algorithm, and the extracted interval sequence includes the RR interval sequence. Specifically, the Pan-Tompkins algorithm is first used to extract the ECG R wave from the segment of the physiological signal data, and then the RR interval sequence is calculated. The Pan-Tompkins algorithm used and the obtained RR interval sequence are both prior arts in the field and will not be described in detail here. The specific method for coarse-graining the interval sequence includes S21 to S22.

[0044] S21, construct the original sequence and the coarse-grained sequence. The original sequence is X = {X 1 ,X 2 ,…,X N}, where N is the total time length, and the coarse-grained sequence is Where τ is the scale parameter.

[0045] S22. Perform mean coarse-graining processing on the original sequence according to the coarse-grained sequence to obtain a coarse-grained sequence.

[0046] S3, solve the fuzzy entropy of the coarse-grained sequence and calculate the characteristic parameters of the fuzzy entropy. Figure 2 As shown, the method for solving the fuzzy entropy of the coarse-grained sequence includes S31 to S35.

[0047] S31. Reconstruct the coarse-grained sequence into an n-dimensional vector in

[0048] S32. Calculate two n-dimensional vectors X i and X j The distance between

[0049] S33, using fuzzy function to calculate two n-dimensional vectors X i and X j The similarity between Where m is the gradient of the exponential function boundary and r is the similarity tolerance.

[0050] S34. Construct similarity ratio function And it is represented as an n+1 dimensional vector.

[0051] S35. Calculate fuzzy entropy When the length of N is finite, the fuzzy entropy is FuzzyEn(n,m,r,N)=lnO m (m,r)-lnO m+1 (m,r).

[0052] Figure 3 and 4 It is a schematic diagram of the correlation degree between fuzzy entropy and physiological signal data at two different scales of the present invention. It can be seen that the present invention can well reflect the distribution of physiological signal data by calculating the fuzzy entropy, which is beneficial to the subsequent detection of sleep apnea segments.

[0053] In the present invention, the gradient of the exponential function boundary is m=1.8-2.2, and the similarity tolerance is r=0.23-0.26SD, where SD is the standard deviation of the coarse-grained sequence. In one embodiment of the present invention, m=1.8, r=0.23SD; in another embodiment of the present invention, m=2.2, r=0.26SD; in a preferred embodiment of the present invention, m=2, r=0.25SD.

[0054] S4. Determine the user's sleep apnea segment based on the degree of difference between the characteristic parameters of fuzzy entropy and the standard sample. The characteristic parameters of fuzzy entropy include mean MFE_Mean, standard deviation MFE_Std and range MFE_Max-Min. It should be noted that the specific calculation method of the three characteristic parameters of fuzzy entropy belongs to the conventional technology in this field and will not be repeated here. The specific comparison method of the degree of difference is: use one-way analysis of variance and minimum significant difference test to test the significant difference between normal people and apnea patients, perform OSA screening based on SPSS's Fisher linear discriminant method, calculate the binary classification accuracy, and use this for detection.

[0055] The detection method of the present invention is a nonlinear analysis method, which improves the accuracy of sleep apnea detection compared with the traditional time-frequency domain analysis method. Specifically, after the time series is coarse-grained, the fuzzy entropy feature parameters of the sequence are calculated respectively. By calculating the fuzzy entropy of different samples at different scales, the feature information between different samples can be effectively selected. Furthermore, based on the mean MFE_Mean, standard deviation MFE_Std and the difference between the maximum and minimum values ​​MFE_Max-Min of the fuzzy entropy sequence of different samples, it is more sensitive to the slight changes in the interval sequence and can reflect the nonlinear interaction between different states of the sample data. Therefore, the extraction of fuzzy entropy features at different scales also has a significant effect on the identification of sleep apnea. At the same time, the present invention does not limit the state of the subject during the data collection process. As long as the electrocardiogram / pulse wave signal collected under non-violent exercise state, the method of the present invention can be applied. In addition, the operation of the present invention is relatively simple, and the statistical method is common, so it is easy to implement.

[0056] A sleep apnea segment detection device based on multi-scale fuzzy entropy is used to implement the above-mentioned sleep apnea segment detection method based on multi-scale fuzzy entropy, and the device includes:

[0057] A collection module, used to collect physiological signal data of the user when the user is in a sleeping state;

[0058] A signal processing module, used for dividing the physiological signal data into a plurality of segments, extracting an interval sequence from the physiological signal data, and performing coarse-graining processing on the interval sequence to obtain a coarse-grained sequence;

[0059] The detection module is used to solve the fuzzy entropy of the coarse-grained sequence, calculate the characteristic parameters of the fuzzy entropy, and determine the user's sleep apnea segment based on the difference between the characteristic parameters of the fuzzy entropy and the standard sample.

[0061] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0062] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A sleep apnea segment detection method based on multi-scale fuzzy entropy, characterized in that: The steps include: When the user is in a sleeping state, physiological signal data of the user is collected and divided into a plurality of segments; Extracting an interval sequence from the physiological signal data, and performing coarse-graining processing on the interval sequence to obtain a coarse-grained sequence; Solve the fuzzy entropy of the coarse-grained sequence and calculate the characteristic parameters of the fuzzy entropy; The sleep apnea segment of the user is determined based on the difference between the characteristic parameters of the fuzzy entropy and the standard sample.

2. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 1, characterized in that: The physiological signal data includes electrocardiogram data and pulse wave data. When the physiological signal data is divided into a plurality of segments, the segmentation is performed based on a preset segment duration.

3. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 1, characterized in that: The specific method for extracting the interval sequence from the physiological signal data is the Pan-Tompkins algorithm, and the extracted interval sequence includes the RR interval sequence.

4. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 1, characterized in that: The specific methods for coarse-graining the interphase sequence include: Constructing original sequence and coarse-grained sequence; The original sequence is processed by mean coarse-graining according to the coarse-grained sequence to obtain a coarse-grained sequence.

5. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 4, characterized in that: The original sequence is X = {X 1 ,X 2 ,…,X N }, where N is the total time length; The coarse-grained sequence is Where τ is the scale parameter.

6. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 1, characterized in that: The methods for solving the fuzzy entropy of coarse-grained sequences include: Reconstruct the coarse-grained sequence into an n-dimensional vector in Calculate two n-dimensional vectors X i and X j The distance between Using fuzzy function to calculate two n-dimensional vectors X i and X j The similarity between Where m is the gradient of the exponential function boundary, r is the similarity tolerance; construct a similarity ratio function And represented as an n+1-dimensional vector; Calculating fuzzy entropy 7. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 6, characterized in that: The gradient of the exponential function boundary is m=1.8~2.2, and the similarity tolerance is r=0.23~0.26SD, where SD is the standard deviation of the coarse-grained sequence.

8. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 6, characterized in that: When the length of N is finite, the fuzzy entropy is FuzzyEn(n,m,r,N)=lnO m (m,r)-lnO m+1 (m,r).

9. The sleep apnea segment detection method based on multi-scale fuzzy entropy according to claim 1, characterized in that: The characteristic parameters of fuzzy entropy include mean MFE_Mean, standard deviation MFE_Std and range MFE_Max-Min.

10. A sleep apnea segment detection device based on multi-scale fuzzy entropy, characterized in that: The device is used to implement the sleep apnea segment detection method based on multi-scale fuzzy entropy as described in any one of claims 1 to 9, the device comprising: A collection module, used to collect physiological signal data of the user when the user is in a sleeping state; A signal processing module, used for dividing the physiological signal data into a plurality of segments, extracting an interval sequence from the physiological signal data, and performing coarse-graining processing on the interval sequence to obtain a coarse-grained sequence; The detection module is used to solve the fuzzy entropy of the coarse-grained sequence, calculate the characteristic parameters of the fuzzy entropy, and determine the user's sleep apnea segment based on the difference between the characteristic parameters of the fuzzy entropy and the standard sample.

Images