Sleep staging analysis method and system based on electroencephalogram signals and medium

Abstract

The embodiment of the invention provides a sleep staging analysis method and system based on electroencephalogram signals and a medium. The method comprises the steps that the electroencephalogram signals are obtained, feature extraction is conducted on the electroencephalogram signals, electroencephalogram features are obtained, and the electroencephalogram features are preprocessed; inputting a sleep analysis model based on the preprocessing result, outputting sleep stage data, and performing stage division on sleep based on the sleep stage data to obtain a plurality of sleep stage analysis results; comparing the sleep period analysis result with set standard information to obtain a staging deviation ratio, and judging whether the staging deviation ratio is greater than or equal to a set deviation ratio threshold value or not; if yes, adjusting model parameters of the sleep analysis model based on the correction information; if yes, evaluating the sleep effect based on the sleep evaluation information; the sleep state is subjected to staging processing by analyzing the electroencephalogram signal, so that the sleep effect is analyzed according to different staging data, and the analysis precision of the sleep state is improved.

Description

Technical Field

[0001] This application relates to the technical field of sleep analysis. Specifically, it relates to a sleep stage analysis method, system, and medium based on electroencephalogram (EEG) signals. Background Art

[0002] Currently, deep learning models such as CNN and LSTM models are used for EEG sleep staging. CNN is used for feature extraction, and LSTM is used for sleep cycle prediction. CNN and LSTM are trained as a hybrid model. Due to the limitation of the number of training samples, in actual applications, new data will be used to retrain the model, and the trained model will be replaced with a new one. Since the model is sometimes large, a large amount of data needs to be downloaded online during the upgrade, which takes a long time. Summary of the Invention

[0003] The purpose of the embodiments of this application is to provide a sleep stage analysis method, system, and medium based on EEG signals, which perform stage processing on the sleep state by analyzing EEG signals, so as to analyze the sleep effect according to different stage data and improve the analysis accuracy of the sleep state.

[0004] The embodiments of this application also provide a sleep stage analysis method based on EEG signals, including:

[0005] Obtain EEG signals, extract features from the EEG signals to obtain EEG features, and preprocess the EEG features to obtain a preprocessing result;

[0006] Input the preprocessing result into a sleep analysis model to output sleep stage data, and perform stage division on the sleep based on the sleep stage data to obtain multiple sleep stage analysis results;

[0007] Compare the sleep stage analysis results with the set standard information to obtain a stage deviation rate, and determine whether the stage deviation rate is greater than or equal to the set deviation rate threshold;

[0008] If it is greater than or equal to, generate correction information, and adjust the model parameters of the sleep analysis model based on the correction information;

[0009] If it is less than, generate sleep evaluation information based on the sleep stage analysis results, and evaluate the sleep effect based on the sleep evaluation information.

[0010] Optionally, in the sleep stage analysis method based on EEG signals described in the embodiments of this application, obtaining EEG signals, extracting features from the EEG signals to obtain EEG features, and preprocessing the EEG features to obtain a preprocessing result specifically include:

[0011] Obtain EEG signals, extract EEG signal features to obtain EEG features;

[0012] Normalize the EEG features and analyze whether the EEG features are within the set feature interval;

[0013] If it is within the set feature interval, obtain the preprocessing result;

[0014] If it is not within the set feature interval, generate optimization information and optimize the EEG features based on the optimization information.

[0015] Optionally, in the sleep stage analysis method based on EEG signals described in the embodiments of the present application, after obtaining the EEG signals, extracting features from the EEG signals to obtain EEG features, and preprocessing the EEG features to obtain a preprocessing result, it further includes:

[0016] Obtain the EEG signals, perform wavelet transform on the EEG signals to obtain time-domain signals and frequency-domain signals;

[0017] Compare the time-domain signals with a set first signal to obtain first noise signals, and eliminate the first noise signals to obtain first optimized signals;

[0018] Compare the frequency-domain signals with a set second signal to obtain second noise signals, and eliminate the second noise signals to obtain second optimized signals;

[0019] Obtain the optimized EEG signals based on the first optimized signals and the second optimized signals;

[0020] Obtain the preprocessing result according to the optimized EEG signals.

[0021] Optionally, in the sleep stage analysis method based on EEG signals described in the embodiments of the present application, input the preprocessing result into a sleep analysis model, output sleep stage data, and perform stage division on sleep based on the sleep stage data to obtain multiple sleep stage analysis results, specifically including:

[0022] Obtain historical analysis data based on big data and establish a data set according to the historical analysis data;

[0023] Construct an initial model, input the training set into the initial model for iterative training to obtain a training result;

[0024] Judge whether the training result converges;

[0025] If it converges, generate a sleep analysis model and output sleep stage data based on the sleep analysis model;

[0026] If it does not converge, adjust the number of iterations and perform secondary training on the initial model until the initial model converges.

[0027] Optionally, in the sleep stage analysis method based on EEG signals according to the embodiments of the present application, adjusting the model parameters of the sleep analysis model based on the correction information specifically includes:

[0028] Obtain the sleep analysis result, compare the sleep analysis result with the set stage condition information to obtain the stage deviation rate;

[0029] Compare the stage deviation rate with the set deviation rate threshold, where the set deviation rate threshold includes a first deviation rate threshold and a second deviation rate threshold, and the first deviation rate threshold is less than the second deviation rate threshold;

[0030] If the stage deviation rate is greater than the first deviation rate threshold and less than the second deviation rate threshold, generate the first correction information, generate the first correction coefficient based on the first correction information, and correct the model parameters in the first way according to the first correction coefficient;

[0031] If the stage deviation rate is greater than or equal to the second deviation rate threshold, generate the second correction information, generate the second correction coefficient based on the second correction information, and correct the model parameters in the second way according to the second correction coefficient.

[0032] Optionally, in the sleep stage analysis method based on EEG signals according to the embodiments of the present application, generating sleep evaluation information based on the sleep stage analysis result and evaluating the sleep effect based on the sleep evaluation information specifically includes:

[0033] Obtain the sleep analysis result, compare the sleep analysis result with the analysis evaluation form to obtain the sleep evaluation information;

[0034] Compare the sleep evaluation information with the set evaluation information to obtain the evaluation deviation rate;

[0035] Judge whether the evaluation deviation rate is greater than or equal to the set deviation rate threshold;

[0036] If it is greater than or equal to, generate an adjustment coefficient and adjust the sleep evaluation information based on the adjustment coefficient;

[0037] If it is less than, evaluate the sleep effect based on the sleep evaluation information to obtain the evaluation result.

[0038] In a second aspect, the embodiments of the present application provide a sleep stage analysis system based on EEG signals, and the system includes: a memory and a processor. The memory includes a program of the sleep stage analysis method based on EEG signals. When the program of the sleep stage analysis method based on EEG signals is executed by the processor, the following steps are implemented:

[0039] Obtain the EEG signal, extract the features of the EEG signal to obtain the EEG features, and preprocess the EEG features to obtain the preprocessing result;

[0040] Input the preprocessing result into the sleep analysis model to output sleep stage data, and based on the sleep stage data, divide the sleep into stages to obtain multiple sleep stage analysis results;

[0041] Compare the sleep stage analysis results with the set standard information to obtain the stage deviation rate, and determine whether the stage deviation rate is greater than or equal to the set deviation rate threshold;

[0042] If it is greater than or equal to, generate correction information and adjust the model parameters of the sleep analysis model based on the correction information;

[0043] If it is less than, generate sleep evaluation information based on the sleep stage analysis results and evaluate the sleep effect based on the sleep evaluation information.

[0044] Optionally, in the sleep stage analysis system based on EEG signals described in the embodiments of the present application, obtain EEG signals, extract features of the EEG signals to obtain EEG features, and preprocess the EEG features to obtain a preprocessing result, specifically including:

[0045] Obtain EEG signals, extract EEG signal features to obtain EEG features;

[0046] Normalize the EEG features and analyze whether the EEG features are in the set feature interval;

[0047] If it is in the set feature interval, obtain the preprocessing result;

[0048] If it is not in the set feature interval, generate optimization information and optimize the EEG features based on the optimization information.

[0049] Optionally, in the sleep stage analysis system based on EEG signals described in the embodiments of the present application, obtain EEG signals, extract features of the EEG signals to obtain EEG features, and preprocess the EEG features to obtain a preprocessing result, further including:

[0050] Obtain EEG signals, perform wavelet transform on the EEG signals to obtain time domain signals and frequency domain signals;

[0051] Compare the time domain signals with the set first signal to obtain the first noise signal, and remove the first noise signal to obtain the first optimized signal;

[0052] Compare the frequency domain signals with the set second signal to obtain the second noise signal, and remove the second noise signal to obtain the second optimized signal;

[0053] Obtain the optimized EEG signals based on the first optimized signal and the second optimized signal;

[0054] Obtain a preprocessing result based on the optimized electroencephalogram (EEG) signal.

[0055] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which includes a sleep staging analysis method program based on an EEG signal. When the sleep staging analysis method program based on the EEG signal is executed by a processor, the steps of the sleep staging analysis method based on the EEG signal as described in any one of the above are implemented.

[0056] As can be seen from the above, a sleep staging analysis method, system, and medium based on an EEG signal provided by an embodiment of the present application obtain an EEG signal, extract features from the EEG signal to obtain EEG features, preprocess the EEG features to obtain a preprocessing result; input the preprocessing result into a sleep analysis model to output sleep staging data, and perform stage division on sleep based on the sleep staging data to obtain multiple sleep stage analysis results; compare the sleep stage analysis results with set standard information to obtain a staging deviation rate, and determine whether the staging deviation rate is greater than or equal to a set deviation rate threshold; if it is greater than or equal to, generate correction information and adjust the model parameters of the sleep analysis model based on the correction information; if it is less than, generate sleep evaluation information based on the sleep stage analysis results, and evaluate the sleep effect based on the sleep evaluation information; perform stage processing on the sleep state by analyzing the EEG signal, so as to analyze the sleep effect according to different stage data and improve the analysis accuracy of the sleep state. Description of the Drawings

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0058] Figure 1 It is a flowchart of the sleep staging analysis method based on the EEG signal provided by the embodiment of the present application;

[0059] Figure 2 It is a flowchart of the EEG signal preprocessing method of the sleep staging analysis method based on the EEG signal provided by the embodiment of the present application;

[0060] Figure 3 It is a flowchart of the preprocessing result acquisition method of the sleep staging analysis method based on the EEG signal provided by the embodiment of the present application. Detailed Embodiments

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0062] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0063] Please refer to Figure 1 , Figure 1 which is a flowchart of a sleep stage analysis method based on electroencephalogram (EEG) signals in some embodiments of the present application. This sleep stage analysis method based on EEG signals is used in a terminal device. This sleep stage analysis method based on EEG signals includes the following steps:

[0064] S101, acquire EEG signals, perform feature extraction on the EEG signals to obtain EEG features, and perform preprocessing on the EEG features to obtain a preprocessing result;

[0065] S102, input the preprocessing result into a sleep analysis model, output sleep stage data, and perform stage division on sleep based on the sleep stage data to obtain multiple sleep stage analysis results;

[0066] S103, compare the sleep stage analysis results with the set standard information to obtain a stage deviation rate, and determine whether the stage deviation rate is greater than or equal to the set deviation rate threshold;

[0067] S104, if it is greater than or equal to, generate correction information, and adjust the model parameters of the sleep analysis model based on the correction information;

[0068] S105, if it is less than, generate sleep evaluation information based on the sleep stage analysis results, and evaluate the sleep effect based on the sleep evaluation information.

[0069] It should be noted that by analyzing EEG signals, the sleep is divided into stages, and then the sleep effect is evaluated, thereby improving the accuracy of sleep analysis.

[0070] Please refer toFigure 2 , Figure 2 This is a flowchart of an electroencephalogram (EEG) signal preprocessing method for a sleep stage analysis method based on EEG signals in some embodiments of the present application. According to the embodiments of the present invention, an EEG signal is acquired, features of the EEG signal are extracted to obtain EEG features, and the EEG features are preprocessed to obtain a preprocessing result, which specifically includes:

[0071] S201: Acquire an EEG signal, extract features of the EEG signal to obtain EEG features;

[0072] S202: Normalize the EEG features and analyze whether the EEG features are within a set feature interval;

[0073] S203: If it is within the set feature interval, obtain the preprocessing result;

[0074] S204: If it is not within the set feature interval, generate optimization information and optimize the EEG features based on the optimization information.

[0075] It should be noted that by analyzing the EEG signal, extracting and analyzing the features of the EEG signal, and optimizing the EEG features, the reflection accuracy of the EEG features is improved.

[0076] Please refer to Figure 3 , Figure 3 This is a flowchart of a method for obtaining a preprocessing result of a sleep stage analysis method based on EEG signals in some embodiments of the present application. According to the embodiments of the present invention, an EEG signal is acquired, features of the EEG signal are extracted to obtain EEG features, and the EEG features are preprocessed to obtain a preprocessing result. It further includes:

[0077] S301: Acquire an EEG signal, perform wavelet transform on the EEG signal to obtain a time-domain signal and a frequency-domain signal;

[0078] S302: Compare the time-domain signal with a set first signal to obtain a first noise signal, remove the first noise signal to obtain a first optimized signal;

[0079] S303: Compare the frequency-domain signal with a set second signal to obtain a second noise signal, remove the second noise signal to obtain a second optimized signal;

[0080] S304: Obtain an optimized EEG signal based on the first optimized signal and the second optimized signal;

[0081] S305: Obtain a preprocessing result according to the optimized EEG signal.

[0082] It should be noted that by dividing the EEG signals in the time domain and frequency domain, and then separately processing and analyzing them in the time domain and frequency domain, the analysis accuracy of the EEG signals can be improved, and the optimization effect of the EEG signals can be enhanced.

[0083] According to the embodiments of the present invention, based on the preprocessing results, input them into the sleep analysis model, output the sleep staging data, and based on the sleep staging data, conduct a stage division of sleep to obtain multiple sleep stage analysis results, specifically including:

[0084] Obtain historical analysis data based on big data, and establish a data set according to the historical analysis data;

[0085] Construct an initial model, input the training set into the initial model for iterative training, and obtain the training results;

[0086] Judge whether the training results converge;

[0087] If it converges, generate a sleep analysis model, and output the sleep staging data based on the sleep analysis model;

[0088] If it does not converge, adjust the number of iterations, and conduct secondary training on the initial model until the initial model converges.

[0089] It should be noted that by continuously training the model with historical analysis data, the learning ability of the model can be realized, the model can be continuously optimized, the output accuracy of the model can be improved, and the output result of the model can be closer to the actual result.

[0090] According to the embodiments of the present invention, adjust the model parameters of the sleep analysis model based on the correction information, specifically including:

[0091] Obtain the sleep analysis results, compare the sleep analysis results with the set staging condition information, and obtain the staging deviation rate;

[0092] Compare the staging deviation rate with the set deviation rate thresholds. The set deviation rate thresholds include a first deviation rate threshold and a second deviation rate threshold, and the first deviation rate threshold is less than the second deviation rate threshold;

[0093] If the staging deviation rate is greater than the first deviation rate threshold and less than the second deviation rate threshold, generate the first correction information, generate the first correction coefficient based on the first correction information, and correct the model parameters in the first way according to the first correction coefficient;

[0094] If the staging deviation rate is greater than or equal to the second deviation rate threshold, generate the second correction information, generate the second correction coefficient based on the second correction information, and correct the model parameters in the second way according to the second correction coefficient.

[0095] It should be noted that by analyzing the sleep analysis results, comparing the sleep analysis results with the set staging condition information, and then accurately adjusting the model parameters to improve the output accuracy of the model.

[0096] According to an embodiment of the present invention, sleep evaluation information is generated based on the sleep stage analysis results, and the sleep effect is evaluated based on the sleep evaluation information, specifically including:

[0097] Obtain the sleep analysis results, compare the sleep analysis results with the analysis evaluation form to obtain sleep evaluation information;

[0098] Compare the sleep evaluation information with the set evaluation information to obtain an evaluation deviation rate;

[0099] Judge whether the evaluation deviation rate is greater than or equal to the set deviation rate threshold;

[0100] If it is greater than or equal to, generate an adjustment coefficient and adjust the sleep evaluation information based on the adjustment coefficient;

[0101] If it is less than, evaluate the sleep effect based on the sleep evaluation information to obtain an evaluation result.

[0102] It should be noted that by comparing the sleep analysis results with the analysis evaluation form, the sleep effect can be accurately analyzed, and then the sleep evaluation result can be accurately adjusted according to the sleep evaluation information to improve the adjustment accuracy.

[0103] According to an embodiment of the present invention, it further includes: obtaining sleep analysis data and analyzing the sleep state information from the sleep analysis data;

[0104] Compare the sleep state information with the set state information to obtain sleep quality information;

[0105] Analyze the human health state information based on the sleep quality information;

[0106] Analyze the relationship between the sleep quality information and human health based on the human health state information to obtain sleep effect information; generate a sleep processing strategy based on the sleep effect information, generate a sleep assistance method based on the sleep processing strategy, and interfere with the electroencephalogram signal according to the sleep assistance method.

[0107] It should be noted that the sleep quality is judged by analyzing the sleep state information, different sleep assistance methods are selected according to the sleep quality, and the electroencephalogram signal is adjusted by interference to improve the sleep effect.

[0108] Second aspect, embodiments of the present application provide a sleep staging analysis system based on electroencephalogram (EEG) signals. The system includes: a memory and a processor. The memory includes a program for the sleep staging analysis method based on EEG signals. When the program for the sleep staging analysis method based on EEG signals is executed by the processor, the following steps are implemented:

[0109] Obtain EEG signals, extract features from the EEG signals to obtain EEG features, preprocess the EEG features to obtain a preprocessing result;

[0110] Input the preprocessing result into a sleep analysis model, output sleep staging data, and perform stage division on sleep based on the sleep staging data to obtain multiple sleep stage analysis results;

[0111] Compare the sleep stage analysis results with the set standard information to obtain a staging deviation rate, and determine whether the staging deviation rate is greater than or equal to the set deviation rate threshold;

[0112] If it is greater than or equal to, generate correction information and adjust the model parameters of the sleep analysis model based on the correction information;

[0113] If it is less than, generate sleep evaluation information based on the sleep stage analysis results and evaluate the sleep effect based on the sleep evaluation information.

[0114] It should be noted that by analyzing EEG signals, the sleep is stage-divided, and then the sleep effect is evaluated, thereby improving the accuracy of sleep analysis.

[0115] According to the embodiments of the present invention, obtaining EEG signals, extracting features from the EEG signals to obtain EEG features, and preprocessing the EEG features to obtain a preprocessing result specifically include:

[0116] Obtain EEG signals, extract EEG signal features to obtain EEG features;

[0117] Perform normalization processing on the EEG features and analyze whether the EEG features are within the set feature interval;

[0118] If it is within the set feature interval, obtain the preprocessing result;

[0119] If it is not within the set feature interval, generate optimization information and optimize the EEG features based on the optimization information.

[0120] It should be noted that by analyzing EEG signals, extracting and analyzing features of the EEG signals, and optimizing the EEG features, the reflection accuracy of the EEG features is improved.

[0121] According to an embodiment of the present invention, an electroencephalogram (EEG) signal is acquired, feature extraction is performed on the EEG signal to obtain EEG features, and preprocessing is performed on the EEG features to obtain a preprocessing result. It further includes:

[0122] Acquire an EEG signal, perform wavelet transform on the EEG signal to obtain a time-domain signal and a frequency-domain signal;

[0123] Compare the time-domain signal with a set first signal to obtain a first noise signal, and eliminate the first noise signal to obtain a first optimized signal;

[0124] Compare the frequency-domain signal with a set second signal to obtain a second noise signal, and eliminate the second noise signal to obtain a second optimized signal;

[0125] Obtain an optimized EEG signal based on the first optimized signal and the second optimized signal;

[0126] Obtain a preprocessing result according to the optimized EEG signal.

[0127] It should be noted that by dividing the EEG signal into the time domain and the frequency domain, and then separately processing and analyzing in the time domain and the frequency domain, the analysis accuracy of the EEG signal is improved, and the optimization effect of the EEG signal is enhanced.

[0128] According to an embodiment of the present invention, based on the preprocessing result, input it into a sleep analysis model, output sleep staging data, and perform stage division on sleep based on the sleep staging data to obtain multiple sleep stage analysis results. Specifically, it includes:

[0129] Obtain historical analysis data based on big data, and establish a data set according to the historical analysis data;

[0130] Construct an initial model, input the training set into the initial model for iterative training to obtain a training result;

[0131] Determine whether the training result converges;

[0132] If it converges, generate a sleep analysis model, and output sleep staging data based on the sleep analysis model;

[0133] If it does not converge, adjust the number of iterative times, and perform secondary training on the initial model until the initial model converges.

[0134] It should be noted that through continuous training of the model with historical analysis data, the learning ability of the model is realized, the model is continuously optimized, the output accuracy of the model is improved, and the output result of the model is closer to the actual result.

[0135] According to an embodiment of the present invention, adjust the model parameters of the sleep analysis model based on correction information. Specifically, it includes:

[0136] Obtain the sleep analysis result, compare the sleep analysis result with the set staging condition information, and obtain the staging deviation rate;

[0137] Compare the staging deviation rate with the set deviation rate thresholds. The set deviation rate thresholds include a first deviation rate threshold and a second deviation rate threshold, and the first deviation rate threshold is less than the second deviation rate threshold;

[0138] If the staging deviation rate is greater than the first deviation rate threshold and less than the second deviation rate threshold, generate first correction information, generate a first correction coefficient based on the first correction information, and correct the model parameters in a first manner according to the first correction coefficient;

[0139] If the staging deviation rate is greater than or equal to the second deviation rate threshold, generate second correction information, generate a second correction coefficient based on the second correction information, and correct the model parameters in a second manner according to the second correction coefficient.

[0140] It should be noted that by analyzing the sleep analysis result, the sleep analysis result is compared with the set staging condition information, and then the model parameters are accurately adjusted to improve the output accuracy of the model.

[0141] According to an embodiment of the present invention, sleep evaluation information is generated based on the sleep stage analysis result, and the sleep effect is evaluated based on the sleep evaluation information, specifically including:

[0142] Obtain the sleep analysis result, compare the sleep analysis result with the analysis evaluation form, and obtain the sleep evaluation information;

[0143] Compare the sleep evaluation information with the set evaluation information to obtain an evaluation deviation rate;

[0144] Judge whether the evaluation deviation rate is greater than or equal to the set deviation rate threshold;

[0145] If it is greater than or equal to, generate an adjustment coefficient and adjust the sleep evaluation information based on the adjustment coefficient;

[0146] If it is less than, evaluate the sleep effect based on the sleep evaluation information to obtain an evaluation result.

[0147] It should be noted that by comparing the sleep analysis result with the analysis evaluation form, the sleep effect is accurately analyzed, and then the sleep evaluation result is accurately adjusted according to the sleep evaluation information to improve the adjustment accuracy.

[0148] According to an embodiment of the present invention, it further includes: obtaining sleep analysis data and analyzing the sleep state information from the sleep analysis data;

[0149] Compare the sleep state information with the set state information to obtain sleep quality information;

[0150] Analyze the human body health status information based on the sleep quality information;

[0151] Analyze the relationship between the sleep quality information and the human body health based on the human body health status information to obtain the sleep effect information; generate a sleep processing strategy based on the sleep effect information, generate a sleep assistance method based on the sleep processing strategy, and interfere with the electroencephalogram (EEG) signals according to the sleep assistance method.

[0152] It should be noted that by analyzing the sleep state information to judge the sleep quality, different sleep assistance methods are selected according to the sleep quality, and the EEG signals are adjusted by interference to improve the sleep effect.

[0153] The third aspect of the present invention provides a computer-readable storage medium, which includes a sleep stage analysis method program based on EEG signals. When the sleep stage analysis method program based on EEG signals is executed by a processor, the steps of the sleep stage analysis method based on EEG signals as described in any one of the above are implemented.

[0154] A sleep stage analysis method, system and medium based on EEG signals disclosed by the present invention obtain EEG signals, extract features of the EEG signals to obtain EEG features, preprocess the EEG features to obtain a preprocessing result; input the preprocessing result into a sleep analysis model, output sleep stage data, and perform a phased division of sleep based on the sleep stage data to obtain multiple sleep stage analysis results; compare the sleep stage analysis results with the set standard information to obtain a stage deviation rate, and judge whether the stage deviation rate is greater than or equal to the set deviation rate threshold; if it is greater than or equal to, generate correction information and adjust the model parameters of the sleep analysis model based on the correction information; if it is less than, generate sleep evaluation information based on the sleep stage analysis results, and evaluate the sleep effect based on the sleep evaluation information; perform a phased processing of the sleep state by analyzing the EEG signals, so as to analyze the sleep effect according to different stage data and improve the analysis accuracy of the sleep state.

[0155] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed with each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0156] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they may be located in one place or distributed over multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] In addition, each functional unit in the embodiments of the present invention may be all integrated in one processing unit, or each unit may be separately taken as one unit, or two or more units may be integrated in one unit; the above-mentioned integrated unit may be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0158] Those of ordinary skill in the art can understand that all or part of the steps to implement the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a readable storage medium. When the program is executed, it performs the steps including the above method embodiments; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical discs and other various media that can store program codes.

[0159] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention essentially or the part that contributes to the prior art can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the foregoing storage medium includes: removable storage devices, ROM, RAM, magnetic disks or optical discs and other various media that can store program codes.

Claims

1. A sleep stage analysis method based on EEG signals, characterized in that: include: Acquire an EEG signal, perform feature extraction on the EEG signal to obtain EEG features, and preprocess the EEG features to obtain a preprocessing result; Based on the preprocessing results, a sleep analysis model is input, sleep stage data is output, and sleep is divided into stages based on the sleep stage data to obtain multiple sleep stage analysis results; Compare the sleep stage analysis result with the set standard information to obtain the stage deviation rate, and determine whether the stage deviation rate is greater than or equal to the set deviation rate threshold; If it is greater than or equal to, generating correction information, and adjusting the model parameters of the sleep analysis model based on the correction information; If it is less than, sleep evaluation information is generated based on the sleep period analysis result, and the sleep effect is evaluated based on the sleep evaluation information.

2. The sleep staging analysis method based on EEG signals according to claim 1, characterized in that: Acquire EEG signals, extract features of the EEG signals to obtain EEG features, and preprocess the EEG features to obtain preprocessing results, specifically including: Acquire EEG signals, extract EEG signal features, and obtain EEG features; Normalize the EEG features and analyze whether the EEG features are within the set feature range; If it is within the set characteristic interval, the preprocessing result is obtained; If it is not in the set feature interval, optimization information is generated, and the EEG features are optimized based on the optimization information.

3. The sleep staging analysis method based on EEG signals according to claim 2, characterized in that: Acquiring EEG signals, extracting features of the EEG signals to obtain EEG features, preprocessing the EEG features to obtain preprocessing results, and also including: Acquire EEG signals, perform wavelet transform on the EEG signals, and obtain time domain signals and frequency domain signals; Compare the time domain signal with a set first signal to obtain a first noise signal, and remove the first noise signal to obtain a first optimized signal; The frequency domain signal is compared with a set second signal to obtain a second noise signal, and the second noise signal is removed to obtain a second optimized signal; Obtaining an optimized EEG signal based on the first optimized signal and the second optimized signal; The preprocessing result is obtained according to the optimized EEG signal.

4. The sleep staging analysis method based on EEG signals according to claim 3, characterized in that: Based on the preprocessing results, the sleep analysis model is input, and the sleep stage data is output. The sleep is divided into stages based on the sleep stage data to obtain multiple sleep stage analysis results, including: Obtain historical analysis data based on big data and establish a data set based on the historical analysis data; Build an initial model, input the training set into the initial model for iterative training, and obtain the training results; Determining whether the training result converges; If converged, a sleep analysis model is generated, and sleep stage data is output based on the sleep analysis model; If it does not converge, adjust the number of iterations and train the initial model a second time until the initial model converges.

5. The sleep staging analysis method based on EEG signals according to claim 1, characterized in that: Adjust the model parameters of the sleep analysis model based on the correction information, including: Obtaining sleep analysis results, comparing the sleep analysis results with the set staging condition information, and obtaining the staging deviation rate; Comparing the installment deviation rate with a set deviation rate threshold, wherein the set deviation rate threshold includes a first deviation rate threshold and a second deviation rate threshold, and the first deviation rate threshold is less than the second deviation rate threshold; If the phase deviation rate is greater than the first deviation rate threshold and less than the second deviation rate threshold, first correction information is generated, a first correction coefficient is generated based on the first correction information, and the model parameters are corrected in a first manner according to the first correction coefficient; If the phase deviation rate is greater than or equal to the second deviation rate threshold, second correction information is generated, a second correction coefficient is generated based on the second correction information, and the model parameters are corrected in a second manner according to the second correction coefficient.

6. The sleep staging analysis method based on EEG signals according to claim 5, characterized in that: The sleep evaluation information is generated based on the sleep period analysis results, and the sleep effect is evaluated based on the sleep evaluation information, specifically including: Obtaining sleep analysis results, comparing the sleep analysis results with the analysis evaluation table, and obtaining sleep evaluation information; Compare the sleep evaluation information with the set evaluation information to obtain the evaluation deviation rate; Determining whether the evaluation deviation rate is greater than or equal to a set deviation rate threshold; If it is greater than or equal to, an adjustment coefficient is generated, and the sleep evaluation information is adjusted based on the adjustment coefficient; If it is less than, the sleep effect is evaluated based on the sleep evaluation information to obtain an evaluation result.

7. A sleep stage analysis system based on EEG signals, characterized in that: The system includes: a memory and a processor, wherein the memory includes a program of a sleep staging analysis method based on electroencephalogram signals, and when the program of the sleep staging analysis method based on electroencephalogram signals is executed by the processor, the following steps are implemented: Acquire an EEG signal, perform feature extraction on the EEG signal to obtain EEG features, and preprocess the EEG features to obtain a preprocessing result; Based on the preprocessing results, a sleep analysis model is input, sleep stage data is output, and sleep is divided into stages based on the sleep stage data to obtain multiple sleep stage analysis results; Compare the sleep stage analysis result with the set standard information to obtain the stage deviation rate, and determine whether the stage deviation rate is greater than or equal to the set deviation rate threshold; If it is greater than or equal to, generating correction information, and adjusting the model parameters of the sleep analysis model based on the correction information; If it is less than, sleep evaluation information is generated based on the sleep period analysis result, and the sleep effect is evaluated based on the sleep evaluation information.

8. The sleep staging analysis system based on EEG signals according to claim 7, characterized in that: Acquire EEG signals, extract features of the EEG signals to obtain EEG features, and preprocess the EEG features to obtain preprocessing results, specifically including: Acquire EEG signals, extract EEG signal features, and obtain EEG features; Normalize the EEG features and analyze whether the EEG features are within the set feature range; If it is within the set characteristic interval, the preprocessing result is obtained; If it is not in the set feature interval, optimization information is generated, and the EEG features are optimized based on the optimization information.

9. The sleep staging analysis system based on EEG signals according to claim 8, characterized in that: Acquiring EEG signals, extracting features of the EEG signals to obtain EEG features, preprocessing the EEG features to obtain preprocessing results, and also including: Acquire EEG signals, perform wavelet transform on the EEG signals, and obtain time domain signals and frequency domain signals; Compare the time domain signal with a set first signal to obtain a first noise signal, and remove the first noise signal to obtain a first optimized signal; The frequency domain signal is compared with a set second signal to obtain a second noise signal, and the second noise signal is removed to obtain a second optimized signal; Obtaining an optimized EEG signal based on the first optimized signal and the second optimized signal; The preprocessing result is obtained according to the optimized EEG signal.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a sleep staging analysis method program based on EEG signals. When the sleep staging analysis method program based on EEG signals is executed by a processor, the steps of the sleep staging analysis method based on EEG signals as described in any one of claims 1 to 6 are implemented.

Images