CN109674474B - Sleep apnea recognition method, device and computer readable medium - Google Patents
Sleep apnea recognition method, device and computer readable medium Download PDFInfo
- Publication number
- CN109674474B CN109674474B CN201811455042.9A CN201811455042A CN109674474B CN 109674474 B CN109674474 B CN 109674474B CN 201811455042 A CN201811455042 A CN 201811455042A CN 109674474 B CN109674474 B CN 109674474B
- Authority
- CN
- China
- Prior art keywords
- sleep apnea
- level
- target
- preset
- recognition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0826—Detecting or evaluating apnoea events
Abstract
The embodiment of the invention discloses a sleep apnea identification method, equipment and a computer readable medium, wherein the method comprises the following steps: acquiring target sleep apnea data of a target user in first unit time, inputting the target sleep apnea data into a preset sleep apnea identification model to obtain a predicted value, acquiring a sleep apnea grade of the target user, and adjusting an identification threshold value based on the sleep apnea grade; and acquiring a sleep apnea identification result output by the preset sleep apnea identification model, wherein the sleep apnea identification result is determined based on the adjusted identification threshold. By adopting the embodiment of the invention, the sleep apnea identification result of the target sleep apnea data of the target user can be more accurately output, so that the final evaluation on the sleep apnea of the target user is more accurate and effective.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to a sleep apnea identification method, equipment and a computer readable medium.
Background
Sleep Apnea Syndrome (SAS) is a symptom with unclear etiology and pathogenesis at present, and the clinical manifestations mainly include: the night sleep snoring is accompanied by symptoms such as apnea and daytime sleepiness. SAS is a serious threat to human health and is liable to cause various complications and even cause sudden death. How to accurately diagnose sleep apnea syndrome is an important part of nighttime medicine.
In order to solve the above problems, in the prior art, a Polysolnogram (PSG) is mainly used for detecting an SAS, but the PSG detection method has many monitoring signals and needs a professional to diagnose, which causes consumption of a large amount of labor cost and equipment cost, and is not easy to popularize. Therefore, an easy sleep apnea recognition scheme needs to be designed.
Disclosure of Invention
The embodiment of the invention provides a sleep apnea identification method, equipment and a computer readable medium, which can accurately identify whether a target user has sleep apnea and improve the identification accuracy rate of the sleep apnea.
In a first aspect, an embodiment of the present invention provides a sleep apnea identification method, where the method includes:
acquiring target sleep apnea data of a target user in first unit time, and inputting the target sleep apnea data into a preset sleep apnea identification model to obtain a predicted value, wherein the preset sleep apnea identification model comprises an identification threshold, the identification threshold is used for determining a sleep apnea identification result corresponding to the target sleep apnea data according to the predicted value, and the sleep apnea identification result is used for indicating whether the target user has sleep apnea within the first unit time;
acquiring a sleep apnea level of the target user, and adjusting the identification threshold based on the sleep apnea level;
and acquiring a sleep apnea identification result output by the preset sleep apnea identification model, wherein the sleep apnea identification result is determined based on the adjusted identification threshold.
In some possible embodiments, the sleep apnea level comprises a health level and a severity level, and the adjusting the identification threshold based on the sleep apnea level comprises:
if the sleep apnea grade comprises a health grade, increasing the identification threshold value by a first preset adjustment value;
if the sleep apnea level comprises a severe level, the identification threshold is adjusted to be lower by a second preset adjustment value.
In some possible embodiments, the obtaining the sleep apnea level of the target user comprises:
acquiring target sleep apnea sample data of N target users in the first unit time, wherein N is a positive integer;
determining a sleep apnea level of the target user based on the N target sleep apnea sample data.
In some possible embodiments, said determining a sleep apnea level of said target user based on said N sleep apnea sample data comprises:
acquiring N predicted values of target sleep apnea sample data of N target users output by the sleep apnea recognition model;
and acquiring the distribution trend of the N predicted values, and determining the sleep apnea level of the target user based on the distribution trend of the N predicted values.
In some possible embodiments, the determining the sleep apnea level of the target user based on the distribution trend of the N prediction values includes:
if the proportion of the N predicted values which are larger than or equal to a first threshold value is larger than or equal to a first preset value, determining that the sleep apnea grade is the severe grade;
and if the proportion of the N predicted values smaller than a first threshold value is smaller than a first preset value, the ratio of the proportion of the N predicted values smaller than a second threshold value to the proportion of the N predicted values smaller than the first threshold value is larger than a second preset value, and the ratio of the proportion of the N predicted values larger than or equal to a third threshold value to the proportion of the N predicted values larger than or equal to the first threshold value is smaller than a third preset value, determining that the sleep apnea grade is the health grade.
In some possible embodiments, the sleep apnea recognition model comprises any one of:
gradient lifting tree algorithm, random forest algorithm, conditional random field and neural network.
In some possible embodiments, the sleep apnea levels further include a mild level and a moderate level, and after obtaining the sleep apnea recognition result output by the sleep apnea recognition model, the method further includes:
obtaining a plurality of the sleep apnea recognition results in a second unit time to determine the sleep apnea level to which the target user belongs.
In a second aspect, embodiments of the present invention provide a sleep apnea identifying apparatus, including means for performing the method of the first aspect.
In a third aspect, an embodiment of the present invention provides another sleep apnea identification apparatus, including: a processor, a memory, a communication interface, and a bus; the processor, the memory and the communication interface are connected through the bus and complete mutual communication; the memory stores executable program code; the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory to perform the method of the first aspect.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a computer program is stored, the computer program comprising program instructions, which, when executed by a processor, cause the processor to perform the method of the first aspect.
According to the embodiment of the invention, the sleep apnea identifying device firstly acquires the target sleep apnea data of the target user in first unit time, inputs the target sleep apnea data into the preset sleep apnea identifying model to obtain the predicted value, then determines the sleep apnea grade of the target user, and adjusts the identification threshold value of the preset sleep apnea identifying model based on the sleep apnea grade, so that the adjusted predicted sleep apnea identifying model can be suitable for the target user, the sleep apnea identifying result of the target sleep apnea data of the target user can be more accurately output, and the final estimation on the sleep apnea of the target user is more accurate and effective.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic structural diagram of a sleep apnea evaluation system according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a method for obtaining a recognition model of a preset sleep apnea according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a method for sleep apnea recognition according to an embodiment of the present invention;
FIG. 4-a is a diagram of a health level sleep apnea data prediction profile according to an embodiment of the present invention;
4-b are graphs of predicted values of severe-grade sleep apnea data provided in accordance with embodiments of the present invention;
FIG. 5 is a schematic block diagram of a sleep apnea recognition apparatus provided in an embodiment of the present invention;
fig. 6 is a schematic block diagram of another sleep apnea recognition apparatus provided in an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.
In order to better understand the present solution, first, an application scenario based on the embodiment of the present invention is described below.
Referring first to fig. 1, fig. 1 is a schematic structural diagram of a sleep apnea evaluation system 100 according to an embodiment of the present invention. As shown in fig. 1, the sleep apnea evaluation system includes a user 110 and a sleep apnea recognition device 120, where the user 110 may be a healthy user or a user with sleep apnea phenomenon, and the sleep apnea recognition device 120 is configured to determine sleep apnea based on a physiological signal (sleep apnea data) of the user, and specifically, the sleep apnea recognition device 120 acquires a physiological signal, such as an Electrocardiogram (ECG) signal, of the user 110 and analyzes the physiological signal to evaluate sleep apnea of the user 110.
In the embodiment of the present invention, the sleep apnea recognition device 120 may be an electronic device integrated with a controller and a memory, so as to store and process the physiological signal of the user 110, and further, the sleep apnea recognition device 120 may further include a display, so as to display the sleep apnea condition of the user 110, for example, the sleep apnea recognition device 120 may be an Internet device such as a user device, a smart phone (e.g., an Android phone, an IOS phone, etc.), a personal computer, a tablet computer, a palm computer, a Mobile Internet device (MID, Mobile Internet Devices), or a wearable smart device, and may also be a professional medical device; the sleep apnea evaluation system 100 may be used in hospital wards for monitoring sleep apnea patients, or in the home for monitoring sleep apnea of the user 110 in the home environment. The embodiments of the present invention are not limited.
In the embodiment of the present invention, in the process of evaluating sleep apnea of the user 110, the sleep apnea identifying device 120 firstly performs R-wave monitoring on the acquired physiological signals of the user 110 in units of minute (or other short time periods) to calculate the Heart Rate Variability (HRV), and extracts parameters capable of characterizing sleep apnea based on the HRV; inputting the parameter capable of representing sleep apnea into a trained preset sleep apnea recognition model to obtain a predicted value, and then performing sleep apnea recognition by using a general recognition threshold (for example, 0.5) to obtain a sleep apnea recognition result, wherein in the embodiment of the invention, the sleep apnea recognition result comprises the occurrence of sleep apnea and the absence of sleep apnea, for example, the sleep apnea recognition result corresponding to the predicted value with the predicted value being greater than or equal to 0.5 is determined as the occurrence of sleep apnea, and the sleep apnea recognition result corresponding to the predicted value with the predicted value being less than 0.5 is determined as the absence of sleep apnea; finally, counting the number of sleep apnea occurrences of the user 110 within a certain time period (for example, 8 hours at night) to determine the sleep apnea level of the user 110, wherein the greater the number of sleep apnea occurrences of the user 110 within the certain time period is counted, the higher the sleep apnea level is counted, and conversely, the less the number of sleep apnea occurrences of the user 110 within the certain time period is counted, the lower the sleep apnea level is counted, wherein the sleep apnea level includes four categories, namely a severe level, a moderate level, a mild level and a healthy level; the preset sleep apnea recognition model comprises a Support Vector Machine (SVM), a Neural Network (NN), a Decision Tree (DT), and other mode recognition models.
However, as the distribution of predicted value data corresponding to the sleep apnea data of the sleep apnea patient is generally larger, namely far larger than 0.5, but points partially deviating from the distribution exist, namely partial predicted values are smaller than 0.5, and if a general identification threshold is adopted to determine the sleep apnea identification result, the points with partial predicted values smaller than 0.5 are judged as not having sleep apnea, at the moment, the judgment is wrong, and when the points with wrong judgment are accumulated more, the final evaluation on the sleep apnea grade of the sleep apnea patient is influenced to a certain extent; similarly, the distribution of the predicted value data corresponding to the sleep apnea data of the healthy person is generally small, that is, much smaller than 0.5, but there may be a point where a part of the predicted value deviates from the distribution, that is, a point where a part of the predicted value is larger than 0.5, at this time, if the sleep apnea recognition result is determined by using the general recognition threshold, the point where the part of the predicted value is larger than 0.5 is determined as no sleep apnea, at this time, the determination is made to be wrong, and when the points where the determination is made to be wrong are accumulated more, the healthy person may be determined as a patient with sleep apnea, so that the determination is made to be wrong.
In order to solve the above problem, the present invention provides a sleep apnea identification method, which can be used in, but is not limited to, the sleep apnea assessment system 100 shown in fig. 1. The method comprises the steps of determining the distribution of prediction values of a user by counting sleep apnea data of the user within a period of time to preliminarily determine the sleep apnea grade of the user, then adjusting the recognition threshold of a preset sleep apnea recognition model based on the sleep apnea grade of the user, and finally recognizing the sleep apnea of the user by using the preset sleep apnea recognition model after the recognition threshold is adjusted, so that the sleep apnea recognition result is more accurate.
The following provides a detailed description of embodiments of the invention. The sleep apnea identification method mainly comprises two processes of obtaining a preset sleep apnea identification model and identifying sleep apnea, namely, the preset sleep apnea identification model is firstly obtained, then the preset sleep apnea identification model is used for identifying sleep apnea, and the two processes are introduced below. Referring to fig. 2, fig. 2 is a schematic flowchart of a method for obtaining a preset sleep apnea recognition model according to an embodiment of the present invention, including the following steps:
step S201, obtaining a preset number of sleep apnea sample data in a first unit time.
In an embodiment of the present invention, the sleep apnea recognition device obtains a preset number of sleep apnea sample data of a first unit time, where the sleep apnea sample data is a physiological signal that can be used to determine whether a user has sleep apnea, and the physiological signal includes, but is not limited to, an Electrocardiogram (ECG) signal, photoplethysmography (PPG) data, a Ballistocardiogram (BCG) signal, a Seismogram (SCG) signal, an Impedance Cardiogram (ICG) signal, a Pulse (PW) signal, a Blood Pressure (Blood Pressure, BP) signal, and the like.
In the embodiment of the present invention, the first unit time is a short time, for example, 1min (minute), 2min, etc., so that each sleep apnea sample data may be, for example, an ECG signal of 1 min.
The preset number of sleep apnea sample data is a sleep apnea sample data set which can fully train a sleep apnea evaluation model. For example, if the first unit time is 1min, the preset number of sleep apnea sample data may be 240 sleep apnea sample data obtained by obtaining the sleep apnea sample data for 4 hours of the user; for another example, in order to train the sleep apnea assessment model more fully, the preset number of sleep apnea sample data may be 480 sleep apnea sample data, that is, the sleep apnea sample data acquired by the user for 8 hours at one night. The preset number may also be a positive integer of 500, 1000, etc.
In the embodiment of the present invention, when the sleep apnea identification device acquires the sleep apnea sample data, the sleep apnea sample data of one or more users may be directly acquired, or acquired from a server or other devices through a network, or extracted from a local database, and the like.
Optionally, in an embodiment, when acquiring the sleep apnea sample data, the sleep apnea identifying device needs to remove the acquired abnormal sleep apnea data, for example, the sleep apnea data when the heart rate of the user is greater than 100 per unit time, the sleep apnea data when the heart rate of the user is 0 per unit time, or the acquired sleep apnea data in other abnormal situations, which will be considered unreasonable. The data are removed, and the reasonability of the sleep apnea sample data acquired by the sleep apnea identification equipment is ensured, so that a preset sleep apnea evaluation model obtained by subsequently training by using the sleep apnea sample data is more accurate.
In the embodiment of the present invention, since the proportion of the sleep apnea samples corresponding to the users with different sleep apnea levels in the set is different, in order to make the data samples included in the preset number of sleep apnea sample data complete and ensure the balance between the positive and negative samples, the preset number of sleep apnea sample data needs to include the sleep apnea data of the users with different sleep apnea levels, for example, the preset number of sleep apnea sample data simultaneously includes the sleep apnea sample data of the users with the heavy level, the medium level, the light level and the healthy level, and the number of the sleep apnea sample data of the users with each sleep apnea level keeps balanced. In one embodiment, if the sleep apnea recognition device directly collects sleep apnea data of a user at various sleep apnea levels, the determination of the sleep apnea levels may be determined by Polysolnogram (PSG).
Further, in an embodiment, the sleep apnea identifying apparatus may perform feature extraction on the sleep apnea sample data to obtain a sleep apnea assessment parameter in the first unit time, where the sleep apnea assessment parameter may be used to indicate whether the user has sleep apnea in the first unit time. Specifically, the sleep apnea identification device firstly performs R wave monitoring on sleep apnea sample data (such as ECG signals) to obtain RR intervals of the sleep apnea sample data, and then performs linear and nonlinear data feature extraction on the RR intervals to acquire HRV data. When the sleep apnea identification equipment carries out linear feature extraction processing on the sleep apnea sample data, linear domain HRV feature data are obtained. When the sleep apnea identification equipment carries out nonlinear feature extraction processing on the sleep apnea sample data, obtaining the HRV feature data of a nonlinear domain.
The linear domain HRV signature data may include time domain HRV signature data and frequency domain signature data, among others. Several time-domain and frequency-domain HRV profiles are given below by way of example. For example, the temporal HRV feature data may include one or more of: mean RR, MSD, Mean SD, SDNN, SDANN, r MSSD, PNN50, SDSD, NN50, and the like. Wherein Mean RR is interval Mean value, which is used to reflect average level of heart rate variability HSV, and is called Mean of RR intervals in english. MSD is the average of the absolute values of the differences between adjacent RR intervals, and is called mean fractional differences in English. Mean SD is the Mean value of standard deviation of RR interphase, and is called RR interphase SD means in English. SDNN is the standard deviation of RR intervals of sinus heart beats and is known as standard deviation of normal to normal intervals in English. SDANN is used to represent the Standard deviation of the mean value of RR intervals over a 5 minute period, and is generally referred to in English as Standard definition of the average of NN intervals in minutes of the entry recording. r MSSD is used to represent The root mean square of The difference between adjacent RR intervals, and is called The root mean square of difference between two adjacent ad jacent NNintervals in English. PNN50 represents the ratio of the number of heart beats differing by more than 50 milliseconds between sinus adjacent RR intervals to the total number of heart beats in the RR intervals, which is known collectively as Percent of NN50 in the total number of RR intervals. SDSD is used to represent the Standard deviation of all RR interval differences, all called Standard definition of statistical Difference between adjacent cycles. NN50 is the number of heart beats in all RR intervals, the difference between adjacent RR intervals is greater than 50ms, and is called number of pairs of ad jacent normal to normal differentiation by more than 50 ms.
As another example, the frequency domain HRV feature data includes one or more of: ULF, VLF, HF, LF, nULF, nVLF, nHF, nLF, etc. Among them, ULF is Ultra Low Frequency, used to reflect the influence of circadian rhythm and neuroendocrine rhythm, and is called Ultra Low Frequency in english. VLF is a Very Low Frequency, associated with caloric regulation and humoral regulation, and is known throughout english as Very Low Frequency. Correspondingly, HF is High Frequency, and is called High Frequency throughout english. LF is Low Frequency, and English is called Low Frequency. nULF is normalized Ultra Low Frequency, and is called normal Ultra Low Frequency in English. nVLF is normalized to Very Low Frequency, and is called normal Very Low Frequency in English. nHF is normalized High Frequency, and is called normal High Frequency throughout English. nLF is normalized Low Frequency, and is called normal Low Frequency in English.
Optionally, the non-linear domain HRV characterization data is not limiting of the invention. For example, sleep apnea recognition devices can generally explore sleep apnea sample data (i.e., ECG signals) through nonlinear system theory and methods, and derive nonlinear domain data by processing poincare scattergrams. Detailed description of the non-linear domain feature extraction embodiments of the present invention are not described in detail or limited.
Step S202, adding a sleep apnea label to the preset number of sleep apnea sample data in the first unit time.
In the embodiment of the present invention, the sleep apnea identification device adds a sleep apnea label to each sleep apnea sample data, so that the sleep apnea identification device can be used to train a sleep apnea identification model, the sleep apnea label is used to indicate whether a user has sleep apnea in a first unit time, the sleep apnea label can be added when the sleep apnea identification device acquires each sleep apnea sample data, or can be added uniformly after the sleep apnea identification device acquires a preset number of sleep apnea sample data, and a specific method for adding the sleep apnea label can be any method for labeling sleep apnea data in the prior art, which is not described in detail in the embodiment of the present invention.
Optionally, in an embodiment, after a preset number of sleep apnea sample data are acquired, a part of the preset number of sleep apnea sample data is used to train the sleep apnea recognition model to obtain the preset sleep apnea recognition model, and another part of the preset number of sleep apnea sample data is used to test whether the accuracy of the preset sleep apnea recognition model meets the requirement, for example, half of the preset number of sleep apnea sample data is used to train the sleep apnea recognition model to obtain the preset sleep apnea recognition model, and the other half of the preset number of sleep apnea sample data is used to test whether the accuracy of the preset sleep apnea recognition model meets the requirement.
Furthermore, in an embodiment, the preset number of sleep apnea sample data includes sleep apnea sample data of users at different sleep apnea levels, half of the sleep apnea sample data of the users at different sleep apnea levels are respectively used for training the sleep apnea recognition model to obtain a preset sleep apnea recognition model (training sample), and the other half of the sleep apnea sample data are used for testing whether the accuracy of the preset sleep apnea recognition model meets the requirement (test sample), so that the balance of the positive and negative samples is ensured, and the training of the sleep apnea recognition model is more effective.
Step S203, training a sleep apnea evaluation model by using the preset number of sleep apnea sample data to obtain a preset sleep apnea identification model.
In an embodiment of the present invention, the sleep apnea evaluation model includes a Gradient Boosting Tree (GBDT) algorithm, a Random Forest (RF) algorithm, a conditional Random field, and a neural network, and optionally, the sleep apnea evaluation model may also be another model identification model, which is not limited in the embodiment of the present invention.
In the embodiment of the present invention, after the sleep apnea identification device obtains the preset number of sleep apnea sample data including the sleep apnea labels, the preset number of sleep apnea sample data is input into the sleep apnea identification model for training, so as to obtain the preset sleep apnea identification model, and a specific method for training the sleep apnea identification model is a method in the prior art, which is not described in detail in the embodiment of the present invention. Optionally, if a part of the preset number of sleep apnea sample data is used for testing, inputting the sleep apnea sample data for training into a sleep apnea recognition model for training.
It is worth to be noted that the sleep apnea recognition model may be obtained by training through the methods of steps S201 to S203, or may be configured by other methods, after the preset sleep apnea recognition model is obtained through training, the sleep apnea may be recognized based on the preset sleep apnea recognition model, further, after training, the preset sleep apnea recognition model may be tested by using a test sample selected from the predicted sleep apnea sample data to evaluate whether the recognition accuracy of the preset sleep apnea recognition model is qualified, if so, the preset sleep apnea recognition model may be used to perform subsequent sleep apnea recognition, and if not, the sleep apnea recognition model may be further trained to make the accuracy of the preset sleep apnea recognition model qualified, or, in the case of failure, the training using the method from step S201 to step S205 is not continued. Optionally, if the accuracy rate of the preset sleep apnea recognition model for recognizing sleep apnea is not qualified during testing due to scene change or other reasons, the preset sleep apnea recognition model may be reconfigured or trained.
Referring to fig. 3 again, fig. 3 is a schematic flowchart of a sleep apnea identification method according to an embodiment of the present invention, including the following steps:
step S301, target sleep apnea data of a target user in a first unit time is obtained, and the target sleep apnea data is input into a preset sleep apnea recognition model to obtain a predicted value.
In the embodiment of the present invention, the sleep apnea recognition device continuously acquires target sleep apnea data of the target user, the duration of which is a first unit time.
In the embodiment of the present invention, the target sleep apnea data is a physiological signal that can be used to determine whether a user has sleep apnea, and the physiological signal is specifically described in step S201, and is not described herein again.
Optionally, in an embodiment, the sleep apnea identifying device may perform feature extraction on the target sleep apnea data to obtain a sleep apnea evaluation parameter in a first unit time, and a specific manner of the feature extraction and a specific description of the sleep apnea evaluation parameter are the same as those in step S201, and are not described herein again.
In an embodiment of the present invention, the preset sleep apnea recognition model includes a recognition threshold, and the recognition threshold is used for determining a sleep apnea recognition result corresponding to the target sleep apnea data according to the predicted value, where the sleep apnea recognition result is used for indicating whether the target user has sleep apnea within a first unit time.
Specifically, target sleep apnea data is input into a sleep apnea recognition model to obtain a predicted value, the predicted value is compared with a recognition threshold, if the predicted value is larger than or equal to the recognition threshold, the sleep apnea recognition result is determined to be sleep apnea, if the predicted value is smaller than the recognition threshold, the sleep apnea recognition result is determined not to be sleep apnea, and therefore the sleep apnea recognition result output by the sleep apnea recognition model is determined to be preset through selection of the recognition threshold.
Optionally, the sleep apnea identifying device may first initialize the identification threshold, generally, the identification threshold is initially selected to be 0.5 (initial identification threshold), and of course, the identification threshold may be other values, which is not limited in the embodiment of the present invention.
Step S302, obtaining the sleep apnea level of the target user, and adjusting the identification threshold value based on the sleep apnea level.
In the embodiment of the invention, the sleep apnea identifying equipment acquires the sleep apnea grade of the target user, wherein the sleep apnea levels comprise at least one of a healthy level, a mild level, a moderate level, and a severe level, as a result of the foregoing analysis, for sleep apnea patients of different sleep apnea levels, the distribution of the predicted values corresponding to the sleep apnea data has the point of deviation distribution, so that the finally output sleep apnea identification result is inconsistent with the actual result, the recognition threshold can be adjusted appropriately at this time according to the sleep apnea level and the condition of the point of the shift distribution, the point of the deviation distribution is correctly identified again by the preset sleep apnea identification model, so that the accuracy of the sleep apnea identification result is improved, and the accuracy of the sleep apnea grade evaluation of the target user is finally improved.
Alternatively, in one embodiment, the sleep apnea level of the target user may be determined and determined by the physician via the PSG.
Optionally, in another embodiment, since the sleep apnea data of the users with different sleep apnea levels are different, the sleep apnea level of the target user may be determined by acquiring target sleep apnea sample data of the target user, specifically, the sleep apnea identifying device acquires the target sleep apnea sample data of N first unit times of the target user, for example, the target sleep apnea data of the target user in the previous 4 hours may be acquired as the sleep apnea sample data, where N is a positive integer, and the acquisition and processing of the target sleep apnea sample data of N first unit times are the same as in step S201, and are not described herein again. Since the N target sleep apnea sample data may reflect the target user's sleep apnea level, the target user's sleep apnea level may be determined based on the N target sleep apnea sample data.
Further, in an embodiment, the sleep apnea identifying device determines the sleep apnea level of the target user based on the target sleep apnea sample data of the target user by: the sleep apnea identification equipment firstly obtains N predicted values of N target sleep apnea sample data of a target user output by a sleep apnea identification model, then obtains the distribution trend of the N predicted values, and determines a first sleep apnea level of the target user based on the distribution trend of the N predicted values. It can be understood that, as described above, since the distributions of the prediction values corresponding to the sleep apnea data of different sleep apnea levels are different, the sleep apnea level of the target user can be preliminarily determined according to the distribution of the prediction values of the target sleep apnea sample data of the target user. Referring to fig. 4-a and 4-b, fig. 4-a is a diagram illustrating a health grade sleep apnea data prediction value distribution diagram according to an embodiment of the present invention, and fig. 4-b is a diagram illustrating a severe grade sleep apnea data prediction value distribution diagram according to an embodiment of the present invention. As shown in fig. 4-a and 4-b, it can be seen that the distribution of the prediction values corresponding to the sleep apnea data for different sleep apnea levels is significantly different.
Further, in one embodiment, the sleep apnea identifying apparatus further determines the sleep apnea level of the target user by obtaining a proportion of N predicted values of the target sleep apnea sample data of the N target users in each threshold interval, thereby determining a distribution of the N predicted values by the proportion, or determining which threshold interval the N predicted values are mainly concentrated in. Specifically, referring to fig. 4 again, for the predicted values, the range of the predicted values is 0-1, in order to determine the distribution of the predicted values, the predicted values are first divided into threshold intervals, and then the ratios of N predicted values in each threshold interval are counted, for example, the threshold intervals of the predicted values are divided into 4 threshold intervals, which are [ 0-first threshold ], [ first threshold-1 ], [ 0-second threshold ], and [ third threshold-1 ], respectively, if the first threshold is 0.5, the second threshold is 0.25, and the third threshold is 0.75, then the 4 threshold intervals are [0-0.5], [0.5-1], [0-0.25], and [0.75-1], and then the ratios of N predicted values in each threshold interval are counted, in the case that the N predicted values are mainly concentrated in the threshold intervals of [ first threshold-1 ] and [ third threshold-1 ], the sleep apnea grade of the target user can be preliminarily judged to be a severe grade, if the N predicted values are mainly concentrated in [ 0-second threshold value ], the sleep apnea grade of the target user can be preliminarily judged to be a healthy grade, and if the distribution of the N predicted values belongs to other conditions, for example, if the N predicted values are mainly between 0.3 and 0.6, the sleep apnea grade of the target user can be preliminarily judged to be a moderate grade or a mild grade.
Further, specifically, in an embodiment, under the division of the threshold interval, if a ratio of the N prediction values greater than or equal to the first threshold is greater than or equal to a first preset value, it is determined that the first sleep apnea level is a severe level; and if the proportion of the N predicted values smaller than the first threshold value is smaller than a first preset value, the ratio of the proportion of the N predicted values smaller than the second threshold value to the proportion of the N predicted values smaller than the first threshold value is larger than a second preset value, and the ratio of the proportion of the N predicted values larger than or equal to a third threshold value to the proportion of the N predicted values larger than or equal to the first threshold value is smaller than a third preset value, determining that the sleep apnea grade is a health grade. The first preset value, the second preset value, and the third preset value may all be obtained based on a statistical method, specifically, in an embodiment, the first predicted value is 0.5, the second preset value is 0.7, and the third preset value is 0.3.
For example, in one embodiment, if 4 threshold intervals are [0.5-1], [0-0.5], [0-0.25] and [0.75-1], respectively, the first predicted value is 0.5, the second preset value is 0.7, the third preset value is 0.3, and the ratio of N predicted values in each threshold interval is a, b, c and d, respectively, then e ═ c/a represents the ratio of [0.75-1] in the [0.5-1] threshold interval, and f ═ d/b represents the ratio of [0-0.25] in the [0-0.5] interval. When a is greater than b, namely the proportion of a is greater than 0.5, the fact that most of the N predicted values are distributed above the initial recognition threshold value of 0.5 indicates that the sleep apnea grade of the target user is statistically regarded as a probable partial weight grade; when a < b, the probability of the sleep apnea grade of the target user is considered to be in a non-severe grade statistically, and when a < b, f >0.7 and e <0.3, the N prediction value distributions can be considered to be mostly below 0.5 statistically, and the prediction values are concentrated below 0.25, which indicates that the sleep apnea grade sample of the target user is more likely to be in a healthy grade. Thus, the sleep apnea level of the target user may be preliminarily determined.
Optionally, in other embodiments, the first threshold, the second threshold, and the third threshold may also take other values, and the number of threshold intervals may also be set to other values, for example, may be 6 threshold intervals, and the calculation method for specifically determining the distribution of the N prediction values may also be other manners, which is not limited in the embodiment of the present invention.
In an embodiment of the present invention, after the sleep apnea level of the target user is preliminarily determined, the recognition threshold may be adjusted based on the sleep apnea level.
Optionally, in an embodiment, if the sleep apnea level of the target user is a health level, as described above, the distribution of the predicted value data corresponding to the target sleep apnea data is generally smaller than the initial recognition threshold, and if the initial recognition threshold is assumed to be 0.5, that is, much smaller than 0.5, but there may be a point where a part of the predicted value deviates from the distribution, that is, a point where a part of the predicted value is larger than 0.5, in order to correctly recognize the part of the predicted value, the recognition threshold may be appropriately increased, that is, the recognition threshold is increased by the first preset adjustment value.
Optionally, in an embodiment, if the sleep apnea level of the target user is a heavy level, as described above, the distribution of the predicted value data corresponding to the target sleep apnea data is generally greater than the initial recognition threshold, and if the initial recognition threshold is assumed to be 0.5, that is, the initial recognition threshold is much greater than 0.5, but there may be a point where a part of the predicted value deviates from the distribution, that is, there may be a part of the predicted value smaller than 0.5, in order to correctly recognize the part of the predicted value, the recognition threshold may be appropriately adjusted downward, that is, the recognition threshold is adjusted downward by the second preset adjustment value.
Optionally, in other embodiments, if the sleep apnea level of the target user is a mild level or a moderate level, it is found through a certain amount of sleep apnea sample data that there are no points with more deviation from the distribution in the predicted value data distribution corresponding to the sleep apnea data of the mild level or the moderate level, so that the accuracy of the sleep apnea recognition result can be ensured without adjusting the recognition threshold at this time.
Optionally, in the embodiment of the present invention, the first preset adjustment value and the second preset adjustment value may also be other values, and may also be determined in other manners, for example, by repeated experiments.
Step S303, obtaining a sleep apnea identification result output by a preset sleep apnea identification model, wherein the sleep apnea identification result is determined based on the adjusted identification threshold.
In the embodiment of the invention, after the recognition threshold is adjusted, the sleep apnea recognition equipment recognizes the predicted value corresponding to the target sleep apnea data by using the preset sleep apnea recognition model after the recognition threshold is adjusted, so that the output sleep apnea recognition result is more accurate.
Optionally, in an embodiment, the sleep apnea identifying apparatus first acquires N pieces of sleep apnea sample data of the target user for a period of time (e.g., 4 hours), adjusts the identification threshold of the preset sleep apnea identifying model by using the method in step S302 based on the N pieces of sleep apnea sample data of the target user, and then identifies the target sleep apnea data of the target user acquired subsequently by using the preset sleep apnea identifying model after the identification threshold is adjusted, so as to further determine the sleep apnea level of the target user.
Optionally, in another embodiment, the sleep apnea identifying apparatus first acquires N pieces of sleep apnea sample data of the target user for a period of time (e.g. 4 hours), adjusts the identification threshold of the preset sleep apnea identifying model by using the method in step S302 based on the N pieces of sleep apnea sample data of the target user, and then identifies again the N pieces of sleep apnea sample data of the target user and the subsequently acquired target sleep apnea data of the target user by using the preset sleep apnea identifying model after the identification threshold is adjusted, so as to further determine the sleep apnea level of the target user.
Optionally, in an embodiment, after the sleep apnea identification device obtains the preset sleep apnea identification model after the identification threshold is adjusted, the preset sleep apnea identification model may be tested again by using the test samples in the preset number of sleep apnea sample data obtained in steps S201 to S202, so as to evaluate the identification accuracy of the preset sleep apnea identification model after the identification threshold is adjusted.
Furthermore, in the embodiment of the present invention, the sleep apnea identifying apparatus may obtain a plurality of sleep apnea evaluation results in a second unit time, and determine the number of sleep apnea occurring in the sleep apnea identification result corresponding to the target sleep apnea data in the second unit time of the target user; and then determining the sleep apnea grade of the target user according to the result threshold interval of the number. For example, if the number is within a first resultant threshold interval (greater than 10), the sleep apnea level of the target user is determined to be a severe level, and if the number is within a second resultant threshold interval (e.g., greater than 15, less than 30), the target user is determined to have sleep apnea and the target sleep apnea level is a moderate level; if the number is within a third result threshold interval (e.g., greater than 5, less than 15), then the target user may be determined to have sleep apnea and the sleep apnea level is mild; if the number is within a fourth resulting threshold interval (e.g., less than 5), it may be determined that the target user does not have sleep apnea and the target user's sleep apnea level is a health level.
The second unit time refers to a collection time length of target sleep apnea data of the target user, which is collected for evaluating a sleep apnea level of the target user, and may be, for example, 4 hours or 8 hours, and the target sleep apnea data of the second unit time may not include target sleep apnea sample data of N target users or may include target sleep apnea sample data of N target users.
It can be seen that, in the embodiment of the present invention, the sleep apnea level of the target user is determined first, and the recognition threshold of the preset sleep apnea recognition model is adjusted based on the sleep apnea level, so that the adjusted predicted sleep apnea recognition model can be applicable to the target user, and thus the sleep apnea recognition result of the target sleep apnea data of the target user can be output more accurately, so as to ensure that the final sleep apnea evaluation of the target user is more accurate and effective.
The embodiment of the invention also provides a sleep apnea identification device, which is used for executing the method in any one of the preceding claims. Specifically, referring to fig. 5, fig. 5 is a schematic block diagram of a sleep apnea recognition apparatus according to an embodiment of the present invention. The sleep apnea recognition apparatus 500 of the present embodiment includes: an acquisition unit 501, an adjustment unit 502, and an output unit 503; wherein:
the acquiring unit 501 is configured to acquire target sleep apnea data of a target user in a first unit time, and input the target sleep apnea data into a preset sleep apnea recognition model to obtain a predicted value, where the preset sleep apnea recognition model includes a recognition threshold, the recognition threshold is used to determine a sleep apnea recognition result corresponding to the target sleep apnea data according to the predicted value, and the sleep apnea recognition result is used to indicate whether sleep apnea occurs in the first unit time for the target user;
the adjusting unit 502 is configured to obtain a sleep apnea level of the target user, and adjust the identification threshold based on the sleep apnea level;
the output unit 503 is configured to obtain a sleep apnea identification result output by the preset sleep apnea identification model, where the sleep apnea identification result is determined based on the adjusted identification threshold.
Optionally, in an embodiment, the sleep apnea level includes a health level and a severity level, and the adjusting module 502 is specifically configured to:
if the sleep apnea grade comprises a health grade, increasing the identification threshold value by a first preset adjustment value;
if the sleep apnea level comprises a severe level, the identification threshold is adjusted to be lower by a second preset adjustment value.
Optionally, in an embodiment, the adjusting module 502 acquires the sleep apnea level of the target user, including:
acquiring target sleep apnea sample data of N target users in the first unit time, wherein N is a positive integer;
determining a sleep apnea level of the target user based on the N target sleep apnea sample data.
Optionally, in one embodiment, the adjusting module 502 determines the sleep apnea level of the target user based on the N sleep apnea sample data, including:
acquiring N predicted values of target sleep apnea sample data of N target users output by the sleep apnea recognition model;
and acquiring the distribution trend of the N predicted values, and determining the sleep apnea level of the target user based on the distribution trend of the N predicted values.
Optionally, in an embodiment, the adjusting module 502 determines the sleep apnea level of the target user based on the distribution trend of the N prediction values, including:
if the proportion of the N predicted values which are larger than or equal to a first threshold value is larger than or equal to a first preset value, determining that the sleep apnea grade is the severe grade;
and if the proportion of the N predicted values smaller than a first threshold value is smaller than a first preset value, the ratio of the proportion of the N predicted values smaller than a second threshold value to the proportion of the N predicted values smaller than the first threshold value is larger than a second preset value, and the ratio of the proportion of the N predicted values larger than or equal to a third threshold value to the proportion of the N predicted values larger than or equal to the first threshold value is smaller than a third preset value, determining that the sleep apnea grade is the health grade.
Optionally, in one embodiment, the sleep apnea recognition model comprises any one of:
gradient lifting tree algorithm, random forest algorithm, conditional random field and neural network.
Optionally, in an embodiment, the sleep apnea levels further include a mild level and a moderate level, the sleep apnea recognition device 500 is further configured to:
obtaining a plurality of the sleep apnea recognition results in a second unit time to determine the sleep apnea level to which the target user belongs.
It can be seen that, in the embodiment of the present invention, the sleep apnea level of the target user is determined first, and the recognition threshold of the preset sleep apnea recognition model is adjusted based on the sleep apnea level, so that the adjusted predicted sleep apnea recognition model can be applicable to the target user, and thus the sleep apnea recognition result of the target sleep apnea data of the target user can be output more accurately, so as to ensure that the final sleep apnea evaluation of the target user is more accurate and effective.
Referring to fig. 6, fig. 6 is a schematic structural diagram of another sleep apnea recognition apparatus disclosed in the embodiment of the present invention. The terminal device of the embodiment includes: at least one processor 601, a communication interface 602, a user interface 603 and a memory 604, wherein the processor 601, the communication interface 602, the user interface 603 and the memory 604 can be connected by a bus or other means, and the embodiment of the present invention is exemplified by being connected by the bus 605. Wherein the content of the first and second substances,
The communication interface 602 may be a wired interface (e.g., an ethernet interface) or a wireless interface (e.g., a cellular network interface or using a wireless local area network interface) for communicating with other electronic devices or websites. In this embodiment of the present invention, the communication interface 602 may be specifically configured to acquire sleep apnea data from a memory, a server, or other devices, and to send a sleep apnea evaluation result of the target user processed by the sleep apnea identifying device to other user terminals.
The user interface 603 may specifically be a touch panel, including a touch screen and a touch screen, for detecting an operation instruction on the touch panel, and the user interface 603 may also be a physical button or a mouse. The user interface 603 may also be a display screen for outputting, displaying images or data.
acquiring target sleep apnea data of a target user in first unit time, and inputting the target sleep apnea data into a preset sleep apnea identification model to obtain a predicted value, wherein the preset sleep apnea identification model comprises an identification threshold, the identification threshold is used for determining a sleep apnea identification result corresponding to the target sleep apnea data according to the predicted value, and the sleep apnea identification result is used for indicating whether the target user has sleep apnea within the first unit time;
acquiring a sleep apnea level of the target user, and adjusting the identification threshold based on the sleep apnea level;
and acquiring a sleep apnea identification result output by the preset sleep apnea identification model, wherein the sleep apnea identification result is determined based on the adjusted identification threshold.
In some possible embodiments, the sleep apnea levels include a health level and a severity level, and the processor 601 adjusting the identification threshold based on the sleep apnea levels includes:
if the sleep apnea grade comprises a health grade, increasing the identification threshold value by a first preset adjustment value;
if the sleep apnea level comprises a severe level, the identification threshold is adjusted to be lower by a second preset adjustment value.
In some possible embodiments, the processor 601 obtaining the sleep apnea level of the target user includes:
acquiring target sleep apnea sample data of N target users in the first unit time, wherein N is a positive integer;
determining a sleep apnea level of the target user based on the N target sleep apnea sample data.
In some possible embodiments, the processor 601 determines a sleep apnea level of the target user based on the N sleep apnea sample data, including:
acquiring N predicted values of target sleep apnea sample data of N target users output by the sleep apnea recognition model;
and acquiring the distribution trend of the N predicted values, and determining the sleep apnea level of the target user based on the distribution trend of the N predicted values.
In some possible embodiments, the processor 601, based on the distribution trend of the N prediction values, determines the sleep apnea level of the target user, including:
if the proportion of the N predicted values which are larger than or equal to a first threshold value is larger than or equal to a first preset value, determining that the sleep apnea grade is the severe grade;
and if the proportion of the N predicted values smaller than a first threshold value is smaller than a first preset value, the ratio of the proportion of the N predicted values smaller than a second threshold value to the proportion of the N predicted values smaller than the first threshold value is larger than a second preset value, and the ratio of the proportion of the N predicted values larger than or equal to a third threshold value to the proportion of the N predicted values larger than or equal to the first threshold value is smaller than a third preset value, determining that the sleep apnea grade is the health grade.
In some possible embodiments, the sleep apnea recognition model comprises any one of:
gradient lifting tree algorithm, random forest algorithm, conditional random field and neural network.
In some possible embodiments, the sleep apnea levels further include a mild level and a moderate level, and the processor 601, after obtaining the sleep apnea recognition result output by the sleep apnea recognition model, is further configured to:
obtaining a plurality of the sleep apnea recognition results in a second unit time to determine the sleep apnea level to which the target user belongs.
It can be seen that, in the embodiment of the present invention, the sleep apnea level of the target user is determined first, and the recognition threshold of the preset sleep apnea recognition model is adjusted based on the sleep apnea level, so that the adjusted predicted sleep apnea recognition model can be applicable to the target user, and thus the sleep apnea recognition result of the target sleep apnea data of the target user can be output more accurately, so as to ensure that the final sleep apnea evaluation of the target user is more accurate and effective.
In a further embodiment of the invention, a computer-readable storage medium is provided, which stores a computer program comprising program instructions, which when executed by a processor, implement all or part of the implementation or implementation steps of the method embodiments described above.
The computer readable storage medium may be an internal storage unit of the terminal according to any of the foregoing embodiments, for example, a hard disk or a memory of the terminal. The computer readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the terminal. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of the terminal. The computer-readable storage medium is used for storing the computer program and other programs and data required by the terminal. The computer readable storage medium may also be used to temporarily store data that has been output or is to be output.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the terminal and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed terminal and method can be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may also be an electric, mechanical or other form of connection.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment of the present invention.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A sleep apnea recognition device, wherein said sleep apnea recognition device comprises:
the device comprises an acquisition unit, a prediction unit and a processing unit, wherein the acquisition unit is used for acquiring target sleep apnea data of a target user in first unit time and inputting the target sleep apnea data into a preset sleep apnea recognition model to obtain a predicted value, the preset sleep apnea recognition model comprises a recognition threshold, the recognition threshold is used for determining a sleep apnea recognition result corresponding to the target sleep apnea data according to the predicted value, and the sleep apnea recognition result is used for indicating whether the target user has sleep apnea within the first unit time;
the adjusting unit is used for acquiring the sleep apnea grade of the target user and adjusting the identification threshold value based on the sleep apnea grade and the point of offset distribution existing in the distribution of predicted values corresponding to the sleep apnea data;
the output unit is used for acquiring a sleep apnea identification result output by the preset sleep apnea identification model, and the sleep apnea identification result is determined based on the adjusted identification threshold;
the adjusting unit acquires the sleep apnea level of the target user, and comprises the following steps:
acquiring target sleep apnea sample data of N target users in the first unit time, wherein N is a positive integer;
acquiring N predicted values of target sleep apnea sample data of N target users output by the sleep apnea recognition model;
acquiring the distribution trend of the N predicted values, and determining the sleep apnea level of the target user based on the distribution trend of the N predicted values;
the adjusting unit determines the sleep apnea level of the target user based on the distribution trend of the N predicted values, and comprises the following steps:
if the proportion of the N predicted values which are larger than or equal to the first threshold value is larger than or equal to the first preset value, determining that the sleep apnea grade is a severe grade;
and if the proportion of the N predicted values smaller than the first threshold value is smaller than a first preset value, the ratio of the proportion of the N predicted values smaller than the second threshold value to the proportion of the N predicted values smaller than the first threshold value is larger than a second preset value, and the ratio of the proportion of the N predicted values larger than or equal to a third threshold value to the proportion of the N predicted values larger than or equal to the first threshold value is smaller than a third preset value, determining that the sleep apnea grade is a health grade.
2. The sleep apnea recognition device of claim 1, wherein said sleep apnea level comprises a health level and a severity level, and said adjustment unit is specifically configured to:
if the sleep apnea grade comprises a health grade, increasing the identification threshold value by a first preset adjustment value;
if the sleep apnea level comprises a severe level, the identification threshold is adjusted to be lower by a second preset adjustment value.
3. A sleep apnea recognition device according to claim 1 or 2, said sleep apnea recognition model comprising any of:
gradient lifting tree algorithm, random forest algorithm, conditional random field and neural network.
4. A sleep apnea recognition device according to claim 1 or 2, said sleep apnea levels further comprising a mild level and a moderate level, said sleep apnea recognition device further being configured to:
obtaining a plurality of the sleep apnea recognition results in a second unit time to determine the sleep apnea level to which the target user belongs.
5. A sleep apnea recognition device, comprising: a processor, a memory, a communication interface, and a bus; the processor, the memory and the communication interface are connected through the bus and complete mutual communication; the memory stores executable program code; the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory to perform the steps performed by the sleep apnea recognition apparatus as recited in any of claims 1-4.
6. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program comprising program instructions which, when executed by a processor, cause the processor to carry out the steps performed by the sleep apnea recognition device of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811455042.9A CN109674474B (en) | 2018-11-30 | 2018-11-30 | Sleep apnea recognition method, device and computer readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811455042.9A CN109674474B (en) | 2018-11-30 | 2018-11-30 | Sleep apnea recognition method, device and computer readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109674474A CN109674474A (en) | 2019-04-26 |
| CN109674474B true CN109674474B (en) | 2021-12-03 |
Family
ID=66185990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811455042.9A Active CN109674474B (en) | 2018-11-30 | 2018-11-30 | Sleep apnea recognition method, device and computer readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109674474B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111166294B (en) * | 2020-01-29 | 2021-09-14 | 北京交通大学 | Automatic sleep apnea detection method and device based on inter-heartbeat period |
| CN112089413A (en) * | 2020-09-08 | 2020-12-18 | 长春理工大学 | Blocking type sleep apnea syndrome screening system |
| CN115486833B (en) * | 2022-08-22 | 2023-06-06 | 华南师范大学 | Respiratory state detection method, respiratory state detection device, computer equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108305688A (en) * | 2017-12-27 | 2018-07-20 | 深圳和而泰数据资源与云技术有限公司 | Illness appraisal procedure, terminal device and computer-readable medium |
| CN108882867A (en) * | 2016-04-15 | 2018-11-23 | 欧姆龙株式会社 | Biont information analytical equipment, system and program |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10869619B2 (en) * | 2016-08-19 | 2020-12-22 | Nox Medical | Method, apparatus, and system for measuring respiratory effort of a subject |
-
2018
- 2018-11-30 CN CN201811455042.9A patent/CN109674474B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882867A (en) * | 2016-04-15 | 2018-11-23 | 欧姆龙株式会社 | Biont information analytical equipment, system and program |
| CN108305688A (en) * | 2017-12-27 | 2018-07-20 | 深圳和而泰数据资源与云技术有限公司 | Illness appraisal procedure, terminal device and computer-readable medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109674474A (en) | 2019-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11517212B2 (en) | Electrocardiogram information dynamic monitoring method and dynamic monitoring system | |
| Gierałtowski et al. | Multiscale multifractal analysis of heart rate variability recordings with a large number of occurrences of arrhythmia | |
| US11617528B2 (en) | Systems and methods for reduced lead electrocardiogram diagnosis using deep neural networks and rule-based systems | |
| CN107456227B (en) | Full-lead electrocardiogram clustering template system and method | |
| CN109674474B (en) | Sleep apnea recognition method, device and computer readable medium | |
| WO2015200750A1 (en) | Early detection of hemodynamic decompensation using taut-string transformation | |
| US20190328251A1 (en) | Arrhythmia detection method, arrhythmia detection device and arrhythmia detection system | |
| CN108201435A (en) | Sleep stage determines method, relevant device and computer-readable medium | |
| CN113995419B (en) | Atrial fibrillation risk prediction system based on heartbeat rhythm signal and application thereof | |
| Forrest et al. | The effect of signal acquisition and processing choices on ApEn values: towards a “gold standard” for distinguishing effort levels from isometric force records | |
| JP6468637B2 (en) | Sleepiness estimation apparatus and sleepiness estimation program | |
| EP4041073A1 (en) | Systems and methods for electrocardiogram diagnosis using deep neural networks and rule-based systems | |
| CN114366060A (en) | Health early warning method and device based on heart rate variability and electronic equipment | |
| CN111684540A (en) | Statistical display method of physiological parameters in monitoring equipment and monitoring equipment | |
| CN111839494A (en) | Heart rate monitoring method and system | |
| CN108182974B (en) | Disease evaluation method, terminal device and computer readable medium | |
| EP4216232A1 (en) | Methods and system for cardiac arrhythmia prediction using transformer-based neural networks | |
| US11179046B2 (en) | Method and system for detection of atrial fibrillation | |
| CN110960207A (en) | Tree model-based atrial fibrillation detection method, device, equipment and storage medium | |
| Fathail et al. | Ecg paper digitization and r peaks detection using fft | |
| CN104921720A (en) | Heart rate variability based evaluation method for hypnotherapy effect of depression | |
| Vasyltsov et al. | Statistical approach for lightweight detection of anomalies in ECG | |
| TW201909840A (en) | System, method and machine-readable medium for detecting atrial fibrillation | |
| Zhi-Min et al. | Fast ECG anomaly detection on Android platform | |
| CN116392136A (en) | Disease early warning method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |