Obstructive sleep apnea syndrome monitoring device and storage medium

Abstract

The invention provides an obstructive sleep apnea syndrome monitoring device and a storage medium, and relates to the technical field of sleep monitoring. The device comprises an electrocardio acquisition assembly used for acquiring electrocardio signals of a user in the sleep period; and the processor is used for determining a waveform feature set of the electrocardiosignal in a target time period, determining sleep apnea features of the target time period based on the waveform feature set, and determining an obstructive sleep apnea syndrome monitoring result of the target time period based on the sleep apnea features. According to the invention, the operand of the processor can be reduced, so that the monitoring efficiency of the obstructive sleep apnea syndrome is improved, and the power consumption of the processor is reduced, so that equipment with relatively low electric quantity or operand such as wearable equipment can comprise the processor, whether a user is in an obstructive sleep apnea state or not can be monitored in real time, and the user experience is improved. And the timeliness and the efficiency of monitoring the obstructive sleep apnea syndrome are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep monitoring, and in particular to an obstructive sleep apnea syndrome monitoring device and a storage medium. Background Art

[0002] Obstructive Sleep Apnea Syndrome (OSAS) refers to a sleep-related respiratory disorder in which breathing frequently stops due to partial or complete obstruction of the upper airway during sleep. Since apnea is usually accompanied by a decrease in blood oxygen saturation and may lead to complications such as hypertension, coronary heart disease, diabetes, and cerebrovascular disease, as well as waking up from sleep to breathe and decreased daytime cognitive ability due to lack of sleep, there is an urgent need for timely and effective monitoring of OSAS.

[0003] At present, most of the monitoring results of obstructive sleep apnea syndrome are determined by collecting the user's ECG signal and performing post-analysis of the ECG signal, that is, transmitting the ECG signal to the server for analysis using the artificial intelligence model in the server. However, during the transmission of the ECG signal, there may be network delays or network failures, which may result in the inability to timely monitor whether the user is in obstructive sleep apnea syndrome; in addition, the currently commonly used artificial intelligence model has a large amount of computation, which increases the power consumption of the computing equipment and reduces the monitoring efficiency of obstructive sleep apnea syndrome. Summary of the invention

[0004] The present invention provides an obstructive sleep apnea syndrome monitoring device and a storage medium, which are used to solve the defects of obstructive sleep apnea syndrome monitoring in the prior art, such as large calculation, high power consumption and low efficiency.

[0005] The present invention provides an obstructive sleep apnea syndrome monitoring device, comprising:

[0006] An ECG acquisition component, which is used to acquire ECG signals of the user during sleep;

[0007] A processor is used to determine a waveform feature set of the electrocardiogram signal within a target time period, and based on the waveform feature set, determine a sleep apnea feature of the target time period, and based on the sleep apnea feature, determine an obstructive sleep apnea syndrome monitoring result of the target time period.

[0008] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the step of determining the sleep apnea characteristics of the target time period based on the waveform feature set includes:

[0009] Removing the first feature sequence and the second feature sequence from the waveform feature set to obtain a target waveform feature set, wherein the waveform features in the first feature sequence are all smaller than any waveform feature in the target waveform feature set, and the waveform features in the second feature sequence are all larger than any waveform feature in the target waveform feature set;

[0010] The sleep apnea feature of the target time period is determined based on the mean value of the target waveform feature set.

[0011] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the target waveform feature set includes an R peak feature set, an S trough feature set and an RR interval feature set, and the sleep apnea feature includes a first sleep apnea feature, a second sleep apnea feature and a third sleep apnea feature;

[0012] The step of determining the sleep apnea feature of the target time period based on the mean of the target waveform feature set includes:

[0013] Determining a first standard deviation of the R-peak feature set based on a mean of the R-peak feature set, and determining the first sleep apnea feature based on the first standard deviation;

[0014] Determining a second standard deviation of the S wave trough feature set based on a mean value of the S wave trough feature set, and determining the second sleep apnea feature based on the second standard deviation;

[0015] Based on the mean of the RR interval feature set, a third standard deviation of the RR interval feature set is determined, and based on the coefficient of variation of the RR interval feature set, the third sleep apnea feature is determined, wherein the coefficient of variation is determined based on a ratio of the third standard deviation to the mean of the RR interval feature set.

[0016] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the obstructive sleep apnea syndrome monitoring result of the target time period is determined based on the sleep apnea characteristics, including:

[0017] If the first sleep apnea characteristic is less than a first preset threshold, the second sleep apnea characteristic is less than a second preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then determining that the obstructive sleep apnea syndrome monitoring result is that the user is in an obstructive sleep apnea state;

[0018] If the first sleep apnea characteristic is greater than or equal to a first preset threshold, or the second sleep apnea characteristic is greater than or equal to a second preset threshold, or the third sleep apnea characteristic is less than or equal to a third preset threshold, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is not in an obstructive sleep apnea state.

[0019] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the sleep apnea feature further includes a fourth sleep apnea feature, and the fourth sleep apnea feature is determined based on the mean value of the RR interval feature set;

[0020] If the first sleep apnea characteristic is less than a first preset threshold, the second sleep apnea characteristic is less than a second preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then determining that the obstructive sleep apnea syndrome monitoring result is that the user is in an obstructive sleep apnea state includes:

[0021] If the first sleep apnea characteristic is less than a first preset threshold, and the second sleep apnea characteristic is less than a second preset threshold, and the fourth sleep apnea characteristic is greater than a fourth preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is in an obstructive sleep apnea state.

[0022] According to the obstructive sleep apnea syndrome monitoring device provided by the present invention, the ECG acquisition component is also used to collect the user's reference ECG signal within a preset time period before the sleep period;

[0023] The processor is further configured to determine a baseline first sleep apnea feature, a baseline second sleep apnea feature, and a baseline fourth sleep apnea feature of the baseline ECG signal in different time periods, and determine the first preset threshold based on an average of each of the baseline first sleep apnea features, determine the second preset threshold based on an average of each of the baseline second sleep apnea features, and determine the fourth preset threshold based on an average of each of the baseline fourth sleep apnea features.

[0024] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the first feature sequence and the second feature sequence in the waveform feature set are removed to obtain a target waveform feature set, including:

[0025] Sorting the waveform feature set to obtain a sorted waveform feature set;

[0026] Based on a preset screening ratio, determining the first feature sequence and the second feature sequence from the sorted waveform feature set;

[0027] The first feature sequence and the second feature sequence in the waveform feature set are removed to obtain a target waveform feature set. According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the step of determining the waveform feature set of the electrocardiogram signal within a target time period includes:

[0028] Performing filtering processing on the ECG signal to obtain a target ECG signal, wherein the filtering processing includes filtering processing and / or noise reduction processing;

[0029] Dividing the target ECG signal according to time to obtain ECG signal segments in different time periods;

[0030] A target electrocardiogram signal segment of the target time period is determined from each of the electrocardiogram signal segments, and a waveform feature set of the target electrocardiogram signal segment is determined.

[0031] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the filtering process of the electrocardiogram signal to obtain a target electrocardiogram signal includes:

[0032] Based on the baseline value of each signal data point in the electrocardiogram signal, performing baseline filtering processing on the electrocardiogram signal to obtain a filtered electrocardiogram signal;

[0033] The filtered ECG signal is subjected to high-frequency noise reduction processing to obtain the target ECG signal.

[0034] According to an obstructive sleep apnea syndrome monitoring device provided by the present invention, the obstructive sleep apnea syndrome monitoring device is a wearable device.

[0035] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the method executed by the processor in any one of the above items is implemented.

[0036] The obstructive sleep apnea syndrome monitoring device and storage medium provided by the present invention include an electrocardiogram acquisition component for collecting the electrocardiogram signal of the user during sleep, so that the processor can determine the waveform feature set within the target time period based on the electrocardiogram signal, and determine the sleep apnea feature based on the waveform feature set, and finally determine the obstructive sleep apnea syndrome monitoring result based on the sleep apnea feature. Through the above method, the processor only needs to perform the above simple data processing to determine the obstructive sleep apnea syndrome monitoring result. Compared with the obstructive sleep apnea detection by the algorithm model in the prior art, the present invention can reduce the amount of calculation of the processor, thereby improving the monitoring efficiency of obstructive sleep apnea syndrome, and reducing the power consumption of the processor, so that the wearable device and other devices with low power or calculation amount can include the processor. In other words, the obstructive sleep apnea syndrome monitoring method executed by the above processor can be executed on the wearable device and other devices with low power or calculation amount, thereby ensuring that the user can be monitored in real time to determine whether he is in the state of obstructive sleep apnea, and improving the timeliness and efficiency of obstructive sleep apnea syndrome monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are 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.

[0038] Figure 1 A schematic diagram of the structure of the obstructive sleep apnea syndrome monitoring device provided by the present invention;

[0039] Figure 2 One of the flowcharts of the method executed by the processor provided by the present invention;

[0040] Figure 3 A second flowchart diagram of a method executed by a processor provided by the present invention;

[0041] Figure 4 A third flowchart diagram of a method executed by a processor provided by the present invention;

[0042] Figure 5 This is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0044] Obstructive Sleep Apnea Syndrome (OSAS) refers to a sleep breathing disorder in which breathing frequently stops due to partial or complete obstruction of the upper airway during sleep. Apnea is the condition of complete obstruction of the upper airway, and hypopnea is the condition of partial obstruction of the upper airway. Apnea and hypopnea are two types of respiratory intensities that can be used to distinguish different types of respiratory states. The standard method for detecting and quantifying sleep apnea is based on respiratory monitoring. Traditionally, several techniques can be used to obtain respiratory signals. A popular technology used in bedside patient monitors involves sharing ECG (ElectroCardioGram) electrodes to measure transthoracic impedance, that is, multiple electrodes share signal transmission lines to measure transthoracic impedance. This method does not obtain respiratory signals from ECG signals. It requires dedicated hardware to sense impedance at specific frequencies and is not suitable for recording ECG and respiratory signals at the same time, but it has the significant advantage of being able to obtain continuous respiratory-related waveforms. Other ways to obtain respiratory signals, such as airflow sensing respiratory monitoring, usually have problems with environmental interference or affecting sleep comfort, and the monitoring equipment is usually expensive.

[0045] OSAS can be treated with nasal continuous positive airway pressure (CPAP) to prevent upper airway collapse. If patients are detected and treated in the early stages of OSAS, adverse health effects can be reduced. Therefore, early identification of subjects at risk for OSAS is crucial. Currently, the screening method for sleep-related breathing disorders is mainly diagnosed using polysomnography detection systems. For this sleep test, clinician evaluation and supervision are crucial. In polysomnography, some physiological parameters such as heart rate and breathing are recorded, which are severely affected during apnea episodes and are calibrated by clinical professionals. Although polysomnography is the most important diagnostic tool in sleep medicine, sleep studies for OSAS diagnosis are expensive for patients because they require overnight evaluation polysomnography (PSG) in a sleep laboratory with specialized systems and caregivers. Accurate identification of apnea events requires direct measurement of upper airway airflow and active breathing. The high cost and low comfort significantly reduce its feasibility as a wearable device. Due to the lack of sleep laboratories mentioned above, the vast majority of OSAS patients are hidden and undiagnosed. Therefore, many researchers use as few signals as possible or a single physiological signal to facilitate the detection of sleep apnea.

[0046] Since apnea is often accompanied by a decrease in blood oxygen saturation, it may lead to complications such as hypertension, coronary heart disease, diabetes, and cerebrovascular disease, as well as waking from sleep to breathe, and insufficient sleep leading to decreased daytime cognitive ability, increased risk of motor vehicle and workplace accidents, depression, sexual dysfunction, and memory loss. In addition, due to its long-term effects on the cardiovascular system, sleep apnea is considered a risk factor for cardiovascular morbidity and mortality. This effect is associated with different physiological mechanisms, such as systemic hypertension and increased sympathetic nervous regulation, which in the long term impairs the normal function of the heart. Therefore, the academic community currently believes that undiagnosed OSAS is considered an important risk factor for the development of cardiovascular disease. There is an urgent need for timely and effective monitoring of OSAS.

[0047] At present, most of the monitoring results of obstructive sleep apnea syndrome are determined by collecting the user's ECG signal and performing post-analysis of the ECG signal, that is, transmitting the ECG signal to the server for analysis using the artificial intelligence model in the server. However, during the transmission of the ECG signal, there may be network delays or network failures, which may result in the inability to timely monitor whether the user is in obstructive sleep apnea syndrome; in addition, the currently commonly used artificial intelligence model has a large amount of computation, which increases the power consumption of the computing equipment and reduces the monitoring efficiency of obstructive sleep apnea syndrome.

[0048] In view of the above problems, the present invention proposes the following embodiments. Figure 1The structural diagram of the obstructive sleep apnea syndrome monitoring device provided by the present invention is as follows: Figure 1 As shown, the obstructive sleep apnea syndrome monitoring device includes: an electrocardiogram acquisition component 110 and a processor 120.

[0049] The ECG acquisition component is used to acquire the ECG signal of the user during sleep.

[0050] Here, the ECG signal is a bioelectric signal generated when the user's heart moves. In one embodiment, the ECG signal is a single-lead ECG signal, and accordingly, the ECG acquisition component is a single-lead ECG measurement component.

[0051] In one embodiment, the sampling frequency of the ECG signal is within a preset frequency range, which can be set according to actual needs. For example, the preset frequency range is 500 data points per second to 1000 data points per second, that is, the preset frequency range is 500 Hz to 1000 Hz.

[0052] The processor is used to determine a waveform feature set of the electrocardiogram signal within a target time period, and based on the waveform feature set, determine a sleep apnea feature of the target time period, and based on the sleep apnea feature, determine an obstructive sleep apnea syndrome monitoring result of the target time period.

[0053] Here, the processor may execute an obstructive sleep apnea syndrome monitoring method. Figure 2 One of the flowcharts of the method executed by the processor provided by the present invention is as follows: Figure 2 As shown, the method includes:

[0054] Step 210, determining a waveform feature set of the electrocardiogram signal within a target time period.

[0055] Here, any waveform feature in the waveform feature set is used to characterize the change characteristics of the ECG signal. The waveform feature set may include but is not limited to at least one of the following: an R wave peak feature set, an S wave trough feature set, and an RR interval feature set, etc. The waveform feature set may be extracted using the Pantompkins method.

[0056] The R peak feature set includes a plurality of R peak features, and the R peak feature is a voltage value corresponding to an R peak position.

[0057] Exemplarily, the positions of all R wave peaks within the target time period and the voltage values ​​corresponding to the positions are extracted from the electrocardiogram signal.

[0058] The S-trough feature set includes a plurality of S-trough features, and the S-trough feature is a voltage value corresponding to the S-trough position.

[0059] Exemplarily, based on the R wave peak position, the minimum value is found in the interval of k data points after the R wave peak position to determine the voltage values ​​corresponding to all S wave trough positions in the target time period in the ECG signal. Wherein, k is a preset value, and the preset value k can be set according to actual needs, for example, k is 15.

[0060] For ease of understanding, the formula for determining the S wave trough position is as follows:

[0061] Sloc i =Min(Y[Rloc i :Rloc i+k ]);

[0062] Where Sloc i is the position of the i-th S trough; Rloc i is the position of the ith R peak; Rloc i+k is the position of the i+kth R peak; Y[Rloc i :Rloc i+k ] is the ECG signal segment of the interval of k data points after the i-th R wave peak position; Min() is used to determine the minimum value position in the ECG signal segment.

[0063] For ease of understanding, the formula for determining the S-wave trough characteristics is as follows:

[0064] Svalue i =location(Y[Sloc i ]);

[0065] In the formula, Svalue i is the voltage value corresponding to the i-th S wave valley position; location(Y[Sloc i ]) is used to determine the voltage value corresponding to the i-th S wave valley position.

[0066] The set A2 of S trough features, taking the target time period of 60 seconds as an example, is expressed as follows:

[0067] A2={Svalue1,Svalue2,Svalue3,....} 60sec ;

[0068] In the formula, {} 60sec It is the collection of features of each S wave valley within 60 seconds.

[0069] The RR interval feature set includes a plurality of RR interval features, and the RR interval feature is the time interval between two adjacent R wave peak features. The time interval can be represented by the number of data points.

[0070] Exemplarily, the time interval between the R peak features of two adjacent points is obtained by calculating the R peak positions of two adjacent points.

[0071] The set A3 of RR interval features, taking the target time period of 60 seconds as an example, is expressed as follows:

[0072] A3={RR1,RR2,RR3,…} 60sec ;

[0073] In the formula, {} 60sec Represents the set of features of each RR interval within 60 seconds.

[0074] It should be noted that the ECG signal includes waveform feature sets of multiple time periods, and the implementation methods of the waveform feature sets of other time periods are basically the same as the implementation methods of the waveform feature sets in the target time period, and the embodiments of the present invention will not be described one by one.

[0075] In some embodiments, the ECG signal is divided according to time to obtain ECG signal segments of different time periods; a target ECG signal segment of a target time period is determined from each ECG signal segment, and a waveform feature set of the target ECG signal segment is determined.

[0076] In one embodiment, the ECG signal is divided according to a preset fixed time interval to obtain a plurality of target ECG signal segments of different time periods. Each target ECG signal segment includes a plurality of signal data points. In other words, according to a preset fixed time interval, a plurality of target ECG signal segments are obtained by segmenting the ECG signal one by one, that is, a target ECG signal segment is intercepted from the ECG signal at each fixed time interval, that is, the time length of each target ECG signal segment is a fixed time interval.

[0077] The preset fixed time interval can be set according to actual needs, for example, the fixed time interval ranges from 30 seconds to 90 seconds.

[0078] Step 220: Determine the sleep apnea characteristics of the target time period based on the waveform feature set.

[0079] Here, the sleep apnea feature is used to characterize the breathing feature corresponding to sleep apnea. The sleep apnea feature may include but is not limited to at least one of the following: a first sleep apnea feature, a second sleep apnea feature, a third sleep apnea feature, a fourth sleep apnea feature, and the like.

[0080] Specifically, based on the mean value of the waveform feature set, the sleep apnea feature of the target time period is determined.

[0081] In a specific embodiment, the waveform feature set includes an R peak feature set, a first standard deviation of the R peak feature set is determined based on the mean of the R peak feature set, and a first sleep apnea feature is determined based on the first standard deviation. The specific execution process of this embodiment is referred to the following embodiment and will not be repeated here.

[0082] In a specific embodiment, the waveform feature set includes an S-trough feature set, a second standard deviation of the S-trough feature set is determined based on the mean of the S-trough feature set, and a second sleep apnea feature is determined based on the second standard deviation. The specific execution process of this embodiment is referred to the following embodiment and will not be repeated here.

[0083] In a specific embodiment, the waveform feature set includes an RR interval feature set, and based on the mean of the RR interval feature set, a third standard deviation of the RR interval feature set is determined; based on the third standard deviation of the RR interval feature set, a coefficient of variation of the RR interval feature set is determined, and a third sleep apnea feature is determined based on the coefficient of variation of the RR interval feature set, wherein the coefficient of variation is determined based on the ratio of the third standard deviation to the mean of the RR interval feature set. The specific execution process of this embodiment refers to the following embodiment and will not be repeated here.

[0084] In a specific embodiment, the waveform feature set includes an RR interval feature set, and the fourth sleep apnea feature is determined based on the mean of the RR interval feature set. The mean of the RR interval feature set can be directly used as the fourth sleep apnea feature, or the mean of the RR interval feature set can be further processed to obtain the fourth sleep apnea feature.

[0085] For example, the fourth sleep apnea feature is the average value of all RR interval features in the RR interval feature set.

[0086] For ease of understanding, the calculation formula for the fourth sleep apnea feature is as follows:

[0087]

[0088] Wherein, Apnea4 is the fourth sleep apnea feature; RR avg is the mean of all RR interval features in the RR interval feature set; RR i is the i-th RR interval feature in the RR interval feature set; N3 is the number of RR interval features in the RR interval feature set.

[0089] Step 230: Determine the obstructive sleep apnea syndrome monitoring result of the target time period based on the sleep apnea characteristics.

[0090] Here, the obstructive sleep apnea syndrome monitoring result may include but is not limited to: the user is in a normal sleeping state, the user is in an obstructive sleep apnea state, and the like.

[0091] Specifically, based on the comparison result of the sleep apnea characteristic with the preset threshold, the obstructive sleep apnea syndrome monitoring result is determined. The number of the preset thresholds may be 1, 2 or more.

[0092] The obstructive sleep apnea syndrome monitoring device provided by the embodiment of the present invention includes an ECG acquisition component for collecting ECG signals of the user during sleep, so that the processor can determine the waveform feature set within the target time period based on the ECG signal, and determine the sleep apnea feature based on the waveform feature set, and finally determine the obstructive sleep apnea syndrome monitoring result based on the sleep apnea feature. In the above manner, the processor only needs to perform the above simple data processing to determine the obstructive sleep apnea syndrome monitoring result. Compared with the obstructive sleep apnea detection by the algorithm model in the prior art, the embodiment of the present invention can reduce the amount of calculation of the processor, thereby improving the monitoring efficiency of obstructive sleep apnea syndrome, and reducing the power consumption of the processor, so that a device with low power or calculation amount such as a wearable device can include the processor. In other words, the obstructive sleep apnea syndrome monitoring method executed by the above processor can be executed on a device with low power or calculation amount such as a wearable device, thereby ensuring that it can monitor in real time whether the user is in an obstructive sleep apnea state, and improve the timeliness and efficiency of obstructive sleep apnea syndrome monitoring.

[0093] Based on the above embodiments, Figure 3 The second flowchart of the method executed by the processor provided by the present invention is as follows: Figure 3 As shown, the above step 220 includes:

[0094] Step 221, removing the first feature sequence and the second feature sequence from the waveform feature set to obtain a target waveform feature set, wherein the waveform features in the first feature sequence are all smaller than any waveform feature in the target waveform feature set, and the waveform features in the second feature sequence are all larger than any waveform feature in the target waveform feature set.

[0095] Here, the first feature sequence includes several waveform features, and the second feature sequence includes several waveform features. Considering that there is noise data in the electrocardiogram signal, which leads to noise in the waveform feature set, the first feature sequence and the second feature sequence can be removed to remove noise interference, thereby suppressing the influence of changes such as motion noise in long-term monitoring on the calculation of sleep apnea features, thereby improving the accuracy of sleep apnea features, and further improving the accuracy of obstructive sleep apnea syndrome monitoring results.

[0096] In some embodiments, based on a preset screening ratio, a first feature sequence and a second feature sequence are determined from the sorted waveform feature set.

[0097] Step 222: Determine the sleep apnea characteristics of the target time period based on the mean of the target waveform feature set.

[0098] Here, the mean of the target waveform feature set may include but is not limited to at least one of the following: the mean of the R peak feature set, the mean of the S trough feature set, the mean of the RR interval feature set, and the like.

[0099] The calculation formula of the mean value of the R peak feature set is as follows:

[0100]

[0101] In the formula, Rvalue avg Rvalue is the average value of all R peak features in the R peak feature set; i is the i-th R-peak feature in the R-peak feature set; N1 is the number of all R-peak features in the R-peak feature set.

[0102] The calculation formula of the mean of the S trough feature set is as follows:

[0103]

[0104] In the formula, Svalue avg Svalue is the average value of all S-trough features in the S-trough feature set; i is the i-th S-trough feature in the S-trough feature set; N2 is the number of all S-trough features in the S-trough feature set.

[0105] In a specific embodiment, the waveform feature set includes an R peak feature set, a first standard deviation of the R peak feature set is determined based on the mean of the R peak feature set, and a first sleep apnea feature is determined based on the first standard deviation. The specific execution process of this embodiment is referred to the following embodiment and will not be repeated here.

[0106] In a specific embodiment, the waveform feature set includes an S-trough feature set, a second standard deviation of the S-trough feature set is determined based on the mean of the S-trough feature set, and a second sleep apnea feature is determined based on the second standard deviation. The specific execution process of this embodiment is referred to the following embodiment and will not be repeated here.

[0107] In a specific embodiment, the waveform feature set includes an RR interval feature set, and based on the mean of the RR interval feature set, a third standard deviation of the RR interval feature set is determined; based on the third standard deviation of the RR interval feature set, a coefficient of variation of the RR interval feature set is determined, and a third sleep apnea feature is determined based on the coefficient of variation of the RR interval feature set, wherein the coefficient of variation is determined based on the ratio of the third standard deviation to the mean of the RR interval feature set. The specific execution process of this embodiment refers to the following embodiment and will not be repeated here.

[0108] In a specific embodiment, the waveform feature set includes an RR interval feature set, and the fourth sleep apnea feature is determined based on the mean of the RR interval feature set. The mean of the RR interval feature set can be directly used as the fourth sleep apnea feature, or the mean of the RR interval feature set can be further processed to obtain the fourth sleep apnea feature.

[0109] The obstructive sleep apnea syndrome monitoring device provided by the embodiment of the present invention removes the first feature sequence and the second feature sequence in the waveform feature set within the target time period to obtain a more accurate target waveform feature set; thereby, based on the mean of the more accurate target waveform feature set, a more accurate sleep apnea feature is determined, thereby further improving the accuracy of the obstructive sleep apnea syndrome monitoring results.

[0110] Based on any of the above embodiments, in the method, the target waveform feature set includes an R peak feature set, an S trough feature set, and an RR interval feature set, the sleep apnea feature includes a first sleep apnea feature, a second sleep apnea feature, and a third sleep apnea feature, and the above step 222 includes:

[0111] Determining a first standard deviation of the R-peak feature set based on a mean of the R-peak feature set, and determining the first sleep apnea feature based on the first standard deviation;

[0112] Determining a second standard deviation of the S wave trough feature set based on a mean value of the S wave trough feature set, and determining the second sleep apnea feature based on the second standard deviation;

[0113] Based on the mean of the RR interval feature set, a third standard deviation of the RR interval feature set is determined, and based on the coefficient of variation of the RR interval feature set, the third sleep apnea feature is determined, wherein the coefficient of variation is determined based on a ratio of the third standard deviation to the mean of the RR interval feature set.

[0114] Here, the first standard deviation may be directly used as the first sleep apnea feature, or the first standard deviation may be further processed to obtain the first sleep apnea feature.

[0115] Exemplarily, the calculation formula of the first sleep apnea feature is as follows:

[0116]

[0117] In the formula, Apnea1 is the first sleep apnea feature; std(A1) is the first standard deviation of the R peak feature set A1; Rvalue i is the i-th R peak feature in the R peak feature set A1; Rvalue avg is the mean of all R-peak features in the R-peak feature set A1; N1 is the number of R-peak features in the R-peak feature set A1.

[0118] Here, the second standard deviation may be directly used as the second sleep apnea feature, or the second standard deviation may be further processed to obtain the second sleep apnea feature.

[0119] Exemplarily, the calculation formula of the second sleep apnea feature is as follows:

[0120]

[0121] In the formula, Apnea2 is the second sleep apnea feature; std(A2) is the second standard deviation of the S-wave trough feature set A2; Svalue i is the i-th S-trough feature in the S-trough feature set A2; Svalue avg is the mean of all S-trough features in the S-trough feature set A2; N2 is the number of S-trough features in the S-trough feature set A2.

[0122] Here, the coefficient of variation may be directly used as the third sleep apnea feature, or the coefficient of variation may be further processed to obtain the third sleep apnea feature.

[0123] Exemplarily, the calculation formula of the third sleep apnea feature is as follows:

[0124]

[0125] Wherein, Apnea3 is the third sleep apnea feature, std(A3) is the third standard deviation of the RR interval feature set A3; avg is the mean of all RR interval features in RR interval feature set A3; is the coefficient of variation of the RR interval feature set A3; RR i is the i-th RR interval feature in the RR interval feature set A3; N3 is the number of RR interval features in the RR interval feature set A3.

[0126] The obstructive sleep apnea syndrome monitoring device provided by the embodiment of the present invention determines the first standard deviation of the R wave peak feature set based on the mean of the R wave peak feature set, and determines the first sleep apnea feature based on the first standard deviation, thereby providing support for determining the first sleep apnea feature; determines the second standard deviation of the S wave trough feature set based on the mean of the S wave trough feature set, and determines the second sleep apnea feature based on the second standard deviation, thereby providing support for determining the second sleep apnea feature; determines the third standard deviation of the RR interval feature set based on the mean of the RR interval feature set, and determines the third sleep apnea feature based on the coefficient of variation of the RR interval feature set, wherein the coefficient of variation is determined based on the ratio of the third standard deviation to the mean of the RR interval feature set, thereby providing support for determining the third sleep apnea feature.

[0127] Based on any of the above embodiments, the above step 230 includes:

[0128] If the first sleep apnea characteristic is less than a first preset threshold, the second sleep apnea characteristic is less than a second preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then determining that the obstructive sleep apnea syndrome monitoring result is that the user is in an obstructive sleep apnea state;

[0129] If the first sleep apnea characteristic is greater than or equal to a first preset threshold, or the second sleep apnea characteristic is greater than or equal to a second preset threshold, or the third sleep apnea characteristic is less than or equal to a third preset threshold, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is not in an obstructive sleep apnea state.

[0130] Here, the first preset threshold value can be set based on the first sleep apnea characteristics in the preset time period before the sleep period, or can be set according to actual needs. The average of the first sleep apnea characteristics in the preset time period before the sleep period can be directly used as the first preset threshold value, or the average value can be further processed to obtain the first preset threshold value.

[0131] Exemplarily, the first preset threshold is obtained by multiplying the mean of the first sleep apnea feature by the first preset parameter. For example, the first preset parameter is 110%, and the calculation formula of the first preset threshold is as follows:

[0132]

[0133] Wherein, α1 is the first preset threshold; M is the number of time periods included in the preset time period before the sleep period, that is, the number of first sleep apnea features in the preset time period before the sleep period; Apnea1 jis the first sleep apnea feature of the jth ECG signal segment in the preset time period before the sleep period. For example, the preset time period before the sleep period is 10 minutes, and the length of each time period is 1 minute. In other words, the preset time period before the sleep period includes 10 time periods, that is, M is equal to 10, and the calculation formula of the first preset threshold is as follows:

[0134]

[0135] It should be noted that, since the first sleep apnea feature is used to characterize the magnitude of the breathing intensity, the first sleep apnea feature is compared with the first preset threshold value. When the first sleep apnea feature is less than the first preset threshold value, that is, when the user's breathing intensity is low, the user may be in an obstructive sleep apnea state. Further, the first sleep apnea feature is the standard deviation of the R wave peak feature in the electrocardiogram signal, and the smaller the standard deviation of the R wave peak feature in the electrocardiogram signal, the lower the breathing intensity, that is, the breathing intensity reflected by the first sleep apnea feature is less than the breathing intensity reflected by the first preset threshold value. Here, the second preset threshold value can be set based on the second sleep apnea feature in the preset time period before the sleep period, or it can be set according to actual needs. The mean of each second sleep apnea feature in the preset time period before the sleep period can be directly used as the second preset threshold value, or the mean value can be further processed to obtain the second preset threshold value.

[0136] Exemplarily, the second preset threshold is obtained by multiplying the mean of the second sleep apnea feature by the second preset parameter. For example, the second preset parameter is 110%, and the calculation formula of the second preset threshold is as follows:

[0137]

[0138] Wherein, α2 is the second preset threshold; M is the number of time periods included in the preset time period before the sleep period, that is, the number of second sleep apnea features in the preset time period before the sleep period; Apnea2 j is the second sleep apnea feature of the jth ECG signal segment in the preset time period before the sleep period. For example, the preset time period before the sleep period is 10 minutes, and the length of each time period is 1 minute. In other words, the preset time period before the sleep period includes 10 time periods, that is, M is equal to 10, and the calculation formula of the second preset threshold is as follows:

[0139]

[0140] It should be noted that, since the second sleep apnea feature is used to characterize the magnitude of the breathing intensity, the second sleep apnea feature is compared with the second preset threshold. When the second sleep apnea feature is less than the second preset threshold, that is, when the user's breathing intensity is low, the user may be in an obstructive sleep apnea state. Further, the second sleep apnea feature is the standard deviation of the S wave trough feature in the electrocardiogram signal, and the smaller the standard deviation of the S wave trough feature in the electrocardiogram signal, the lower the breathing intensity, that is, the breathing intensity reflected by the second sleep apnea feature is less than the breathing intensity reflected by the second preset threshold. Here, the third preset threshold can be set according to actual needs. In one embodiment, during sleep, the cessation of oral and nasal airflow for more than 10 seconds is used as a reference label for obstructive sleep apnea, and the optimal preset threshold is determined based on the ROC (Receiver Operating Characteristic) curve. Specifically, the third sleep apnea feature is used as a judgment feature of obstructive sleep apnea, and the third preset threshold is continuously changed to obtain the optimal third preset threshold. For example, when the third preset threshold is 0.08, the sensitivity can reach 94.7%, the specificity can reach 89.2%, and the area under the ROC curve (AUC) can reach 0.960.

[0141] It should be noted that, because in the state of obstructive sleep apnea, the user's heart rate variation should be large. Based on this, since the third sleep apnea feature is used to characterize the magnitude of the heart rate variation, the third sleep apnea feature is compared with the third preset threshold. When the third sleep apnea feature is greater than the third preset threshold, that is, when the user's heart rate variation is large, the user may be in an obstructive sleep apnea state. Further, the third sleep apnea feature is the normalized standard deviation of the RR interval feature in the electrocardiogram signal, and the larger the normalized standard deviation of the RR interval feature in the electrocardiogram signal, the larger the heart rate variation, that is, the heart rate variation reflected by the third sleep apnea feature is greater than the heart rate variation reflected by the third preset threshold.

[0142] The obstructive sleep apnea syndrome monitoring device provided in an embodiment of the present invention determines the obstructive sleep apnea syndrome monitoring results based on the comparison results of the first preset threshold, the second preset threshold, the third preset threshold and the first sleep apnea characteristic, the second sleep apnea characteristic, and the third sleep apnea characteristic, thereby improving the accuracy of the obstructive sleep apnea syndrome monitoring results and ultimately improving the monitoring accuracy of obstructive sleep apnea syndrome.

[0143] Based on any of the above embodiments, in the method, the sleep apnea feature further includes a fourth sleep apnea feature, the fourth sleep apnea feature is determined based on the mean of the RR interval feature set, and if the first sleep apnea feature is less than a first preset threshold, and the second sleep apnea feature is less than a second preset threshold, and the third sleep apnea feature is greater than a third preset threshold, then determining that the obstructive sleep apnea syndrome monitoring result is that the user is in an obstructive sleep apnea state includes:

[0144] If the first sleep apnea characteristic is less than a first preset threshold, and the second sleep apnea characteristic is less than a second preset threshold, and the fourth sleep apnea characteristic is greater than a fourth preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is in an obstructive sleep apnea state.

[0145] In a specific embodiment, the waveform feature set includes an RR interval feature set, and the fourth sleep apnea feature is determined based on the mean of the RR interval feature set. The mean of the RR interval feature set can be directly used as the fourth sleep apnea feature, or the mean of the RR interval feature set can be further processed to obtain the fourth sleep apnea feature.

[0146] For example, the fourth sleep apnea feature is the average value of all RR interval features in the RR interval feature set.

[0147] For ease of understanding, the calculation formula for the fourth sleep apnea feature is as follows:

[0148]

[0149] Wherein, Apnea4 is the fourth sleep apnea feature; RR avg is the mean of all RR interval features in the RR interval feature set; RR i is the i-th RR interval feature in the RR interval feature set; N3 is the number of RR interval features in the RR interval feature set.

[0150] Here, the fourth preset threshold value can be set based on the fourth sleep apnea characteristic in the preset time period before the sleep period, or can be set according to actual needs. The average value of each fourth sleep apnea characteristic in the preset time period before the sleep period can be directly used as the fourth preset threshold value, or the average value can be further processed to obtain the fourth preset threshold value.

[0151] Exemplarily, the fourth preset threshold is obtained by multiplying the mean of the fourth sleep apnea feature by the fourth preset parameter. For example, the fourth preset parameter is 105%, and the calculation formula of the fourth preset threshold is as follows:

[0152]

[0153] Wherein, α4 ​​is the fourth preset threshold; M is the number of time periods included in the preset time period before the sleep period, that is, the number of the fourth sleep apnea characteristics in the preset time period before the sleep period; Apnea4 j is the fourth sleep apnea feature of the jth ECG signal segment in the preset time period before the sleep period. For example, the preset time period before the sleep period is 10 minutes, and the length of each time period is 1 minute. In other words, the preset time period before the sleep period includes 10 time periods, that is, M is equal to 10, and the calculation formula of the fourth preset threshold is as follows:

[0154]

[0155] It should be noted that, because in the state of obstructive sleep apnea, the user's heart rate should be slower. Based on this, since the fourth sleep apnea feature is used to characterize the speed of the heart rate, the fourth sleep apnea feature is compared with the fourth preset threshold. When the fourth sleep apnea feature is greater than the fourth preset threshold, that is, when the user's heart rate is slower, the user may be in the state of obstructive sleep apnea. Further, the fourth sleep apnea feature is the mean of the RR interval feature in the electrocardiogram signal, and the larger the mean of the RR interval feature in the electrocardiogram signal, the slower the heart rate, that is, the heart rate reflected by the fourth sleep apnea feature is slower than the heart rate reflected by the third preset threshold. In one embodiment, the first sleep apnea feature, the second sleep apnea feature, and the fourth sleep apnea feature are first jointly compared with the first preset threshold, the second preset threshold, and the fourth preset threshold; and then the third sleep apnea feature is compared with the third preset threshold. For example, if the conditions of Apnea1<α1, Apnea2<α2, Apnea4>α4 are first met, and then the condition of Apnea3>α3 is met, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is in an obstructive sleep apnea state.

[0156] The obstructive sleep apnea syndrome monitoring device provided by an embodiment of the present invention determines the obstructive sleep apnea syndrome monitoring results based on the comparison results of the first preset threshold, the second preset threshold, the third preset threshold, the fourth preset threshold and the first sleep apnea characteristic, the second sleep apnea characteristic, the third sleep apnea characteristic, and the fourth sleep apnea characteristic, thereby further improving the accuracy of the obstructive sleep apnea syndrome monitoring results, and ultimately further improving the monitoring accuracy of obstructive sleep apnea syndrome.

[0157] Based on any of the above embodiments, in the method, the ECG acquisition component is further used to acquire a reference ECG signal of the user within a preset time period before the sleep period;

[0158] The processor is further configured to determine a baseline first sleep apnea feature, a baseline second sleep apnea feature, and a baseline fourth sleep apnea feature of the baseline ECG signal in different time periods, and determine the first preset threshold based on an average of each of the baseline first sleep apnea features, determine the second preset threshold based on an average of each of the baseline second sleep apnea features, and determine the fourth preset threshold based on an average of each of the baseline fourth sleep apnea features.

[0159] Here, the first reference sleep apnea feature may be the first standard deviation of the R wave peak feature set, or the first standard deviation may be further processed to obtain the first reference sleep apnea feature. The determination method of the first reference sleep apnea feature is substantially the same as the determination method of the first sleep apnea feature described above, and will not be described in detail here.

[0160] Here, the reference second sleep apnea feature may be the second standard deviation of the S wave trough feature set, or the second standard deviation may be further processed to obtain the reference second sleep apnea feature. The method for determining the reference second sleep apnea feature is substantially the same as the method for determining the second sleep apnea feature described above, and will not be described in detail here.

[0161] Here, the fourth baseline sleep apnea feature may be the mean of the RR interval feature set, or the mean of the RR interval feature set may be further processed to obtain the fourth baseline sleep apnea feature. The method for determining the fourth baseline sleep apnea feature is substantially the same as the method for determining the fourth sleep apnea feature described above, and will not be described in detail here.

[0162] Here, the first preset threshold value can be set based on the baseline first sleep apnea feature in the preset time period before the sleep period, or can be set according to actual needs. The average of the baseline first sleep apnea features in the preset time period before the sleep period can be directly used as the first preset threshold value, or the average value can be further processed to obtain the first preset threshold value.

[0163] Exemplarily, the first preset threshold is obtained by multiplying the mean of the first baseline sleep apnea feature by the first preset parameter. For example, the first preset parameter is 110%, and the calculation formula of the first preset threshold is as follows:

[0164]

[0165] Wherein, α1 is the first preset threshold; M is the number of time periods included in the preset time period before the sleep period, that is, the number of the first baseline sleep apnea features in the preset time period before the sleep period; Apnea1 j is the benchmark first sleep apnea feature of the jth ECG signal segment in the preset time period before the sleep period. For example, the preset time period before the sleep period is 10 minutes, and the length of each time period is 1 minute, that is, the preset time period before the sleep period includes 10 time periods, that is, M is equal to 10, and the calculation formula of the first preset threshold is as follows:

[0166]

[0167] Here, the second preset threshold value can be set based on the benchmark second sleep apnea feature in the preset time period before the sleep period, or can be set according to actual needs. The average of the benchmark second sleep apnea features in the preset time period before the sleep period can be directly used as the second preset threshold value, or the average value can be further processed to obtain the second preset threshold value.

[0168] Exemplarily, the second preset threshold is obtained by multiplying the mean of the second baseline sleep apnea feature by the second preset parameter. For example, the second preset parameter is 110%. The calculation formula of the second preset threshold is as follows:

[0169]

[0170] Wherein, α2 is the second preset threshold; M is the number of time periods included in the preset time period before the sleep period, that is, the number of the benchmark second sleep apnea features in the preset time period before the sleep period; Apnea2 j is the reference second sleep apnea feature of the jth ECG signal segment in the preset time period before the sleep period. For example, the preset time period before the sleep period is 10 minutes, and the length of each time period is 1 minute, that is, the preset time period before the sleep period includes 10 time periods, that is, M is equal to 10, and the calculation formula of the second preset threshold is as follows:

[0171]

[0172] Here, the fourth preset threshold value can be set based on the fourth baseline sleep apnea feature in the preset time period before the sleep period, or can be set according to actual needs. The average of the fourth baseline sleep apnea features in the preset time period before the sleep period can be directly used as the fourth preset threshold value, or the average value can be further processed to obtain the fourth preset threshold value.

[0173] Exemplarily, the fourth preset threshold is obtained by multiplying the mean of the fourth baseline sleep apnea feature by the fourth preset parameter. For example, the fourth preset parameter is 105%. The calculation formula of the fourth preset threshold is as follows:

[0174]

[0175] Wherein, α4 ​​is the fourth preset threshold; M is the number of time periods included in the preset time period before the sleep period, that is, the number of the fourth baseline sleep apnea features in the preset time period before the sleep period; Apnea4 j is the fourth baseline sleep apnea feature of the jth ECG signal segment in the preset time period before the sleep period. For example, the preset time period before the sleep period is 10 minutes, and the length of each time period is 1 minute, that is, the preset time period before the sleep period includes 10 time periods, that is, M is equal to 10, and the calculation formula of the fourth preset threshold is as follows:

[0176]

[0177] In the obstructive sleep apnea syndrome monitoring device provided by the embodiment of the present invention, the ECG acquisition component is also used to collect the reference ECG signal of the user in a preset time period before the sleep period, so as to provide support for the acquisition of the reference ECG signal; the processor is also used to determine the reference first sleep apnea feature of the reference ECG signal in different time periods, and determine the first preset threshold value based on the average of each reference first sleep apnea feature, thereby improving the accuracy of the first preset threshold value, and further improving the accuracy of the obstructive sleep apnea syndrome monitoring result; the processor is also used to determine the reference second sleep apnea feature of the reference ECG signal in different time periods, and determine the second preset threshold value based on the average of each reference second sleep apnea feature, thereby improving the accuracy of the second preset threshold value, and further improving the accuracy of the obstructive sleep apnea syndrome monitoring result; the processor is also used to determine the reference fourth sleep apnea feature of the reference ECG signal in different time periods, and determine the fourth preset threshold value based on the average of each reference fourth sleep apnea feature, thereby improving the accuracy of the fourth preset threshold value, and further improving the accuracy of the obstructive sleep apnea syndrome monitoring result.

[0178] Based on any of the above embodiments, the above step 221 includes:

[0179] Sorting the waveform feature set to obtain a sorted waveform feature set;

[0180] Based on a preset screening ratio, determining the first feature sequence and the second feature sequence from the sorted waveform feature set;

[0181] The first feature sequence and the second feature sequence in the waveform feature set are removed to obtain a target waveform feature set.

[0182] Specifically, the waveform feature set is sorted according to the size of the waveform feature values ​​in the waveform feature set to obtain a sorted waveform feature set. The sorting method can be an ascending method or a descending method, which is not specifically limited in the embodiment of the present invention.

[0183] Here, the screening ratio can be limited according to actual needs, and the screening ratios corresponding to the first feature sequence and the second feature sequence can be the same or different. For example, the screening ratio corresponding to the first feature sequence can be 10%, and the screening ratio corresponding to the second feature sequence can be 10%.

[0184] Specifically, based on a preset screening ratio and a preset selection position, a first feature sequence and a second feature sequence are determined from the sorted waveform feature set. The selection position can be set according to actual needs, for example, the first feature sequence can come from the last 10% of the waveform feature set in descending order, or from the first 10% of the waveform feature set in ascending order; the second feature sequence can come from the last 10% of the waveform feature set in ascending order, or from the first 10% of the waveform feature set in descending order.

[0185] For ease of understanding, a specific embodiment is used for illustration. Taking the R peak feature set in the waveform feature set as an example, as shown below:

[0186] A1={Rvalue1,Rvalue2,…,Rvalue 30}.

[0187] Afterwards, the R peak feature set A1 is sorted in descending order to obtain the R peak feature set A1S in the sorted waveform feature set. The R peak feature set A1S in the sorted waveform feature set is as follows:

[0188] A1S={Rvalue 1′ ,Rvalue 2′ ,…,Rvalue 30′};

[0189] Among them, Rvalue 1′ >Rvalue 2′ >…>Rvalue 30′ .

[0190] Afterwards, the first feature sequence and the second feature sequence are obtained by screening according to the preset screening ratio. Exemplarily, the first feature sequence is as follows:

[0191] {Rvalue 26′ ,Rvalue27′ ,Rvalue 28′ ,Rvalue 29′ ,Rvalue 30′};

[0192] The second signature sequence is as follows:

[0193] {Rvalue 1′ ,Rvalue 2′ ,Rvalue 3′ ,Rvalue 4′ ,Rvalue 5′}.

[0194] It should be noted that the method for determining the first feature sequence and the second feature sequence by other feature sets such as the S trough feature set and the RR interval feature set in the waveform feature set is the same as the method for determining the first feature sequence and the second feature sequence by the R peak feature set in the waveform feature set, and will not be repeated here.

[0195] Specifically, the first feature sequence and the second feature sequence in the waveform feature set are removed to obtain the target waveform feature set.

[0196] Exemplarily, the first feature sequence and the second feature sequence in the R peak feature set A1 are removed to obtain the target waveform feature set. ′ As shown below:

[0197] A1 ′ = {Rvalue 6′ ,Rvalue 7′ ,…,Rvalue 24′}.

[0198] It can be understood that removing the first feature sequence and the second feature sequence can effectively suppress the impact of changes such as motion noise during long-term monitoring on the calculation of sleep apnea characteristics, thereby improving the accuracy of sleep apnea characteristics, and then improving the accuracy of obstructive sleep apnea syndrome monitoring results, and ultimately improving the monitoring accuracy of obstructive sleep apnea syndrome.

[0199] The obstructive sleep apnea syndrome monitoring device provided by the embodiment of the present invention sorts the waveform feature set to obtain a sorted waveform feature set, and then determines the first feature sequence and the second feature sequence from the sorted waveform feature set based on a preset screening ratio, thereby removing the first feature sequence and the second feature sequence in the waveform feature set to obtain a more accurate target waveform feature set, and then better calculate the mean of the target waveform feature set to improve the accuracy of sleep apnea features, thereby improving the accuracy of obstructive sleep apnea syndrome monitoring results, and ultimately improving the monitoring accuracy of obstructive sleep apnea syndrome.

[0200] Based on any of the above embodiments, Figure 4 The third flowchart of the method performed by the processor provided by the present invention, the above step 210 includes:

[0201] Step 211, filtering the ECG signal to obtain a target ECG signal, wherein the filtering process includes filtering process and / or noise reduction process.

[0202] Here, the target ECG signal is a high-quality ECG signal, thereby improving the accuracy of the waveform feature set, and further improving the accuracy of the obstructive sleep apnea syndrome monitoring results.

[0203] In one embodiment, the filtering process may be baseline filtering, and further, the baseline filtering may be median filtering.

[0204] In some embodiments, a baseline value of each data point in the ECG signal is determined; based on the baseline value, the ECG signal is subjected to baseline filtering to obtain a target ECG signal.

[0205] Here, the baseline value may include but is not limited to: the median value of each data point in the ECG signal, the average value of each data point in the ECG signal, the root mean square value of each data point in the ECG signal, and the like.

[0206] In a specific embodiment, a target data point to be filtered in the ECG signal is determined; based on the target data point, a target data point sequence is selected from each data point in the ECG signal; and the median of the target data point sequence is determined as the baseline value of the target data point.

[0207] The number of data points in the target data point sequence can be determined according to the sliding window length, and the sliding window length can be set according to actual needs. For example, the sliding window length is 201, that is, the number of data points in the target data point sequence is 201. Based on this, the median of the target data point sequence is the 101th data point in the sorted target data point sequence. For ease of understanding, assuming that the target data point sequence is {X[n], X[n+1], X[n+2], ..., X[n+200]}, the baseline value of the target data point sequence is X[n+100].

[0208] It should be noted that the baseline value is the drift baseline of the ECG signal, and the sliding window length of 201 points is selected to effectively suppress the low-frequency baseline drift.

[0209] In one embodiment, the length of the sliding window is 0.1 to 1.5 times the sampling frequency of the ECG signal.

[0210] Specifically, the target ECG signal is determined based on the difference between each data point in the ECG signal and the corresponding baseline value. Exemplarily, the calculation formula of the target ECG signal is as follows:

[0211] Y[n]=X[n]-Q[n];

[0212] Where Y[n] is the nth data point in the target ECG signal, X[n] is the nth data point in the ECG signal, and Q[n] is the baseline value corresponding to the nth data point in the target ECG signal.

[0213] It is understandable that based on the difference between each data point in the ECG signal and the corresponding baseline value, the baseline value can be removed to obtain a stable target ECG signal, thereby facilitating the subsequent determination of the waveform feature set and thereby improving the accuracy of the waveform feature set.

[0214] In some other embodiments, the ECG signal is subjected to noise reduction processing to obtain the target ECG signal. Further, the ECG signal is subjected to high frequency noise reduction processing to obtain the target ECG signal.

[0215] In one embodiment, the high frequency noise reduction process uses a sliding average method, and the time interval of the sliding average method can be 0.02 times to 0.1 times the sampling frequency of the electrocardiogram signal.

[0216] Specifically, according to a preset time interval, sliding selection is performed in the ECG signal to obtain a number of signal data point sequences, and the signal data point mean corresponding to all signal data points in each signal data point sequence is determined; based on the signal data point mean corresponding to all signal data points in each signal data point sequence, each signal data point in the ECG signal is denoised to obtain a target ECG signal.

[0217] For ease of understanding, the calculation formula for high-frequency noise reduction processing is as follows:

[0218]

[0219] Where Y[n] is the nth data point in the target ECG signal, X[n] is the nth data point in the ECG signal, and L is the sliding average window length. For example, if the sliding average window length L is 3, the calculation formula for high-frequency noise reduction processing is as follows:

[0220]

[0221] It can be understood that by performing noise reduction processing on each signal data point in the ECG signal, a stable target ECG signal can be obtained, thereby facilitating the subsequent determination of the waveform feature set, thereby improving the accuracy of the waveform feature set, and playing a key role in the resolution of obstructive sleep apnea syndrome monitoring.

[0222] Step 212, dividing the target ECG signal according to time to obtain ECG signal segments in different time periods.

[0223] In a specific embodiment, according to a preset time window length, a sliding selection is performed in the target ECG signal to obtain ECG signal segments of several different time periods. Each ECG signal segment includes multiple signal data points. In other words, according to a preset time window length, a plurality of ECG signal segments are obtained by segmenting the target ECG signal one by one, that is, one ECG signal segment is intercepted from the target ECG signal per time window length, that is, the time length of each ECG signal segment is the time window length.

[0224] The preset time window length may be set according to actual needs, for example, the time window length ranges from 30 seconds to 90 seconds.

[0225] For ease of understanding, assuming that the time window length is 60 seconds, a number of ECG signal segments are obtained from the target ECG signal one by one, that is, one ECG signal segment is intercepted from the target ECG signal every 60 seconds, that is, the time length of each ECG signal segment is 60 seconds.

[0226] Step 213, determining a target ECG signal segment of the target time period from each of the ECG signal segments, and determining a waveform feature set of the target ECG signal segment.

[0227] In a specific embodiment, a target ECG signal segment is determined from ECG signal segments of different time periods according to a preset target time period, and a waveform feature set of the target ECG signal segment is determined.

[0228] The obstructive sleep apnea syndrome monitoring device provided in the embodiment of the present invention can obtain a stable target ECG signal by filtering the ECG signal, thereby facilitating the subsequent extraction of a waveform feature set, thereby improving the accuracy of the waveform feature set, thereby improving the accuracy of the sleep apnea feature, and ultimately improving the monitoring accuracy of obstructive sleep apnea syndrome.

[0229] Based on any of the above embodiments, the above step 211 includes:

[0230] Based on the baseline value of each signal data point in the electrocardiogram signal, performing baseline filtering processing on the electrocardiogram signal to obtain a filtered electrocardiogram signal;

[0231] The filtered ECG signal is subjected to high-frequency noise reduction processing to obtain the target ECG signal.

[0232] It should be noted that the ECG signal is obtained by sampling based on a sampling frequency, so the ECG signal includes a plurality of signal data points, and the number of the signal data points is determined according to the sampling frequency and the sampling duration.

[0233] Here, the baseline value may include but is not limited to: the median value of each signal data point in the ECG signal, the average value of each signal data point in the ECG signal, the root mean square value of each signal data point in the ECG signal, and the like.

[0234] In a specific embodiment, a target signal data point to be filtered in the ECG signal is determined; based on the target signal data point, a target data point sequence is selected from each signal data point in the ECG signal; and the median of the target data point sequence is determined as the baseline value of the target signal data point.

[0235] The number of signal data points in the target data point sequence can be determined according to the sliding window length, and the sliding window length can be set according to actual needs. For example, the sliding window length is 101, that is, the number of signal data points in the target data point sequence is 101. Based on this, the median of the target data point sequence is the 51st signal data point in the sorted target data point sequence. For ease of understanding, assuming that the target data point sequence is {X[n], X[n+1], X[n+2], ..., X[n+100]}, the baseline value of the target data point sequence is X[n+50].

[0236] It should be noted that the baseline value is the drift baseline of the ECG signal, and the sliding window length of 101 points is selected to effectively suppress the low-frequency baseline drift.

[0237] In one embodiment, the length of the sliding window is 0.1 to 1.5 times the sampling frequency of the ECG signal.

[0238] Specifically, the baseline filtering process is as follows: based on the difference between each signal data point in the ECG signal and the corresponding baseline value, a filtered ECG signal is determined. Exemplarily, the calculation formula of the filtered ECG signal is as follows:

[0239] Z[n]=X[n]-Q[n];

[0240] Where Z[n] is the nth signal data point in the filtered ECG signal, X[n] is the nth signal data point in the ECG signal, and Q[n] is the baseline value corresponding to the nth signal data point in the filtered ECG signal. Z[n] is the signal data point obtained by splitting X[n] after median filtering.

[0241] It is understandable that based on the difference between each signal data point in the ECG signal and the corresponding baseline value, the baseline value can be removed to obtain a stable filtered ECG signal, thereby facilitating subsequent extraction of a waveform feature set and thereby improving the accuracy of the waveform feature set.

[0242] In one embodiment, the high frequency noise reduction process uses a sliding average method, and the time interval of the sliding average method can be 0.02 times to 0.1 times the sampling frequency of the electrocardiogram signal.

[0243] Specifically, the high-frequency noise reduction processing is as follows: Based on the mean value of each signal data point, the mean value of each signal data point can be directly determined as the target ECG signal, or the mean value of each signal data point can be further processed to obtain the target ECG signal. Exemplarily, each signal data point in the ECG signal is subjected to noise reduction processing to obtain the target ECG signal. The calculation formula of the high-frequency noise reduction processing is as follows:

[0244]

[0245] Where Y[n] is the nth signal data point in the target ECG signal, Z[n] is the nth signal data point in the filtered ECG signal, and L is the time interval.

[0246] It can be understood that by performing noise reduction processing on each signal data point in the ECG signal, a stable target ECG signal can be obtained, thereby facilitating the subsequent determination of the waveform feature set, thereby improving the accuracy of the waveform feature set, and playing a key role in the resolution of obstructive sleep apnea syndrome monitoring.

[0247] The obstructive sleep apnea syndrome monitoring device provided in the embodiment of the present invention provides support for filtering processing of electrocardiogram signals in the above manner. Based on the baseline value of each signal data point in the electrocardiogram signal, the electrocardiogram signal is baseline filtered to obtain a filtered electrocardiogram signal, and then the filtered electrocardiogram signal is subjected to high-frequency noise reduction processing to obtain a stable target electrocardiogram signal, thereby facilitating subsequent extraction to obtain a waveform feature set, thereby improving the accuracy of the waveform feature set, thereby improving the accuracy of the sleep apnea feature, and ultimately improving the monitoring accuracy of obstructive sleep apnea syndrome.

[0248] Based on any of the above embodiments, the obstructive sleep apnea syndrome monitoring device is a wearable device.

[0249] Here, the wearable device may include but is not limited to at least one of the following: a smart watch, a smart bracelet, a wristband device, a smart ring, etc.

[0250] It should be noted that using wearable devices such as smart watches and smart bracelets to monitor health is becoming a new lifestyle. Hundreds of millions of wearable devices are used every day, and the number is increasing dramatically. Since sleep is a key determinant of personal health and happiness, sleep monitoring is one of the most popular features of such wearable devices.

[0251] The obstructive sleep apnea syndrome monitoring device provided in the embodiment of the present invention can be a wearable device, thereby ensuring that it is possible to monitor in real time whether the user is in an obstructive sleep apnea state, thereby improving the timeliness and efficiency of obstructive sleep apnea syndrome monitoring, and improving the convenience of obstructive sleep apnea syndrome monitoring.

[0252] In actual application scenarios, the use of low-cost and easy-to-use wearable devices to diagnose and screen OSAS at home and evaluate treatment effects will greatly contribute to public health. One of the most critical technologies for monitoring OSAS with wearable devices is the OSAS monitoring method. In addition, low-power wearable devices also provide convenience and feasibility for continuous sleep monitoring at night, which increases the availability of ECG signals during sleep, facilitates the accumulation of sufficient ECG signal data for auxiliary screening, and serves as an auxiliary access method for real-time feedback to other physical therapy systems to achieve better therapeutic effects. Based on the above embodiments, the sleep apnea feature based on the ECG signal can monitor the sleep period, and can effectively determine whether obstructive sleep apnea occurs, with very high accuracy; and the sleep apnea feature used in this application has an algorithm complexity of O(n), which is O(n) compared to machine learning or deep learning, wavelet analysis, and other algorithms with a complexity of O(n). 2) The above algorithms or technical solutions, this application has the characteristics of low cost and low power consumption, and at the same time has high-precision sleep apnea detection capabilities, which are suitable for use in wearable devices. In other words, the method executed by the processor in this application can make full use of the convenient characteristics of wearable devices, and use lower computing power and electricity in the process of dynamic ECG monitoring of sleep breathing, and efficiently analyze and screen standard hypopnea events in sleep breathing. It can be used as one of the key parameters for sleep quality assessment, which is of great help to the implementation of ECG home health detection scenarios and optimization of sleep quality related industry scenarios.

[0253] Based on the above embodiments, this application proposes to judge the breathing intensity and duration based on the electrocardiogram characteristics, and then accurately and efficiently monitor and identify obstructive sleep apnea with low power consumption, which helps to develop the quality indicators of sleep breathing on smart health wearable devices, and becomes a very useful function on health wearable devices. At the same time, it also has the characteristics of low computing power and low power consumption, and has the feasibility of a wide range of clinical, health care, and home sleep health monitoring application scenarios.

[0254] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 120, a communication interface 510, a memory 520 and a communication bus 530, wherein the processor 120, the communication interface 510 and the memory 520 communicate with each other through the communication bus 530. The processor 120 may call the logic instructions in the memory 520 to execute the obstructive sleep apnea syndrome monitoring method, which includes: determining a waveform feature set of the electrocardiogram signal in a target time period; determining the sleep apnea feature of the target time period based on the waveform feature set; and determining the obstructive sleep apnea syndrome monitoring result of the target time period based on the sleep apnea feature.

[0255] In addition, the logic instructions in the above-mentioned memory 520 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0256] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the obstructive sleep apnea syndrome monitoring method provided by the above-mentioned methods, and the method includes: determining a waveform feature set of the electrocardiogram signal within a target time period; based on the waveform feature set, determining the sleep apnea characteristics of the target time period; based on the sleep apnea characteristics, determining the obstructive sleep apnea syndrome monitoring results of the target time period.

[0257] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the obstructive sleep apnea syndrome monitoring method provided by the above-mentioned methods, the method comprising: determining a waveform feature set of the electrocardiogram signal within a target time period; based on the waveform feature set, determining the sleep apnea characteristics of the target time period; based on the sleep apnea characteristics, determining the obstructive sleep apnea syndrome monitoring results of the target time period.

[0258] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0259] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0260] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An obstructive sleep apnea syndrome monitoring device, characterized in that: include: An ECG acquisition component, which is used to acquire ECG signals of the user during sleep; A processor is used to determine a waveform feature set of the electrocardiogram signal within a target time period, and based on the waveform feature set, determine a sleep apnea feature of the target time period, and based on the sleep apnea feature, determine an obstructive sleep apnea syndrome monitoring result of the target time period.

2. The obstructive sleep apnea syndrome monitoring device according to claim 1, characterized in that: The step of determining the sleep apnea feature of the target time period based on the waveform feature set includes: Removing the first feature sequence and the second feature sequence from the waveform feature set to obtain a target waveform feature set, wherein the waveform features in the first feature sequence are all smaller than any waveform feature in the target waveform feature set, and the waveform features in the second feature sequence are all larger than any waveform feature in the target waveform feature set; The sleep apnea feature of the target time period is determined based on the mean value of the target waveform feature set.

3. The obstructive sleep apnea syndrome monitoring device according to claim 2, characterized in that: The target waveform feature set includes an R peak feature set, an S trough feature set, and an RR interval feature set, and the sleep apnea feature includes a first sleep apnea feature, a second sleep apnea feature, and a third sleep apnea feature; The step of determining the sleep apnea feature of the target time period based on the mean of the target waveform feature set includes: Determining a first standard deviation of the R-peak feature set based on a mean of the R-peak feature set, and determining the first sleep apnea feature based on the first standard deviation; Determining a second standard deviation of the S wave trough feature set based on a mean value of the S wave trough feature set, and determining the second sleep apnea feature based on the second standard deviation; Based on the mean of the RR interval feature set, a third standard deviation of the RR interval feature set is determined, and based on the coefficient of variation of the RR interval feature set, the third sleep apnea feature is determined, wherein the coefficient of variation is determined based on a ratio of the third standard deviation to the mean of the RR interval feature set.

4. The obstructive sleep apnea syndrome monitoring device according to claim 3, characterized in that: The step of determining the obstructive sleep apnea syndrome monitoring result of the target time period based on the sleep apnea characteristic includes: If the first sleep apnea characteristic is less than a first preset threshold, the second sleep apnea characteristic is less than a second preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then determining that the obstructive sleep apnea syndrome monitoring result is that the user is in an obstructive sleep apnea state; If the first sleep apnea characteristic is greater than or equal to a first preset threshold, or the second sleep apnea characteristic is greater than or equal to a second preset threshold, or the third sleep apnea characteristic is less than or equal to a third preset threshold, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is not in an obstructive sleep apnea state.

5. The obstructive sleep apnea syndrome monitoring device according to claim 4, characterized in that: The sleep apnea feature further includes a fourth sleep apnea feature, and the fourth sleep apnea feature is determined based on the mean value of the RR interval feature set; If the first sleep apnea characteristic is less than a first preset threshold, the second sleep apnea characteristic is less than a second preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then determining that the obstructive sleep apnea syndrome monitoring result is that the user is in an obstructive sleep apnea state includes: If the first sleep apnea characteristic is less than a first preset threshold, and the second sleep apnea characteristic is less than a second preset threshold, and the fourth sleep apnea characteristic is greater than a fourth preset threshold, and the third sleep apnea characteristic is greater than a third preset threshold, then the obstructive sleep apnea syndrome monitoring result is determined to be that the user is in an obstructive sleep apnea state.

6. The obstructive sleep apnea syndrome monitoring device according to claim 5, characterized in that: The ECG acquisition component is also used to acquire a reference ECG signal of the user within a preset time period before the sleep period; The processor is further configured to determine a baseline first sleep apnea feature, a baseline second sleep apnea feature, and a baseline fourth sleep apnea feature of the baseline ECG signal in different time periods, and determine the first preset threshold based on an average of each of the baseline first sleep apnea features, determine the second preset threshold based on an average of each of the baseline second sleep apnea features, and determine the fourth preset threshold based on an average of each of the baseline fourth sleep apnea features.

7. The obstructive sleep apnea syndrome monitoring device according to claim 2, characterized in that: The removing the first feature sequence and the second feature sequence from the waveform feature set to obtain a target waveform feature set comprises: Sorting the waveform feature set to obtain a sorted waveform feature set; Based on a preset screening ratio, determining the first feature sequence and the second feature sequence from the sorted waveform feature set; The first feature sequence and the second feature sequence in the waveform feature set are removed to obtain a target waveform feature set.

8. The obstructive sleep apnea syndrome monitoring device according to claim 1, characterized in that: The step of determining a waveform feature set of the electrocardiogram signal within a target time period includes: Performing filtering processing on the ECG signal to obtain a target ECG signal, wherein the filtering processing includes filtering processing and / or noise reduction processing; Dividing the target ECG signal according to time to obtain ECG signal segments in different time periods; A target electrocardiogram signal segment of the target time period is determined from each of the electrocardiogram signal segments, and a waveform feature set of the target electrocardiogram signal segment is determined.

9. The obstructive sleep apnea syndrome monitoring device according to claim 8, characterized in that: The filtering process of the electrocardiogram signal to obtain a target electrocardiogram signal includes: Based on the baseline value of each signal data point in the electrocardiogram signal, performing baseline filtering processing on the electrocardiogram signal to obtain a filtered electrocardiogram signal; The filtered ECG signal is subjected to high-frequency noise reduction processing to obtain the target ECG signal.

10. The obstructive sleep apnea syndrome monitoring device according to any one of claims 1 to 9, characterized in that: The obstructive sleep apnea syndrome monitoring device is a wearable device.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the method executed by the processor as claimed in any one of claims 1 to 9 is implemented.

Images