Deep learning model optimization method for sleep health monitoring

Abstract

The present invention discloses a deep learning model optimization method for sleep health monitoring, which belongs to the field of model optimization technology. The method monitors and counts the user's daily exercise data and beverage data, and uses different impact modules in the sleep health monitoring model to implement targeted data processing and analysis, uses the user's daily exercise data and beverage data to verify the reliability of the basic sleep quality data corresponding to his or her night sleep, uses the sleep impact verification data corresponding to several sleep health monitorings of the user to perform reliability analysis on the impact effects of the sleep positive impact module, sleep negative impact module and sleep fusion impact module in the sleep health monitoring model, and dynamically manages the impact processing schemes implemented by different impact modules in the sleep health monitoring model according to the evaluation results. The present invention is used to solve the technical problem that the deep learning model in the existing scheme has poor personalized monitoring, analysis and management effects for different individual users.

Description

Technical Field

[0001] The present invention relates to the field of model optimization technology, and in particular to a deep learning model optimization method for sleep health monitoring. Background Art

[0002] In the field of sleep health monitoring, deep learning models are being used more and more widely. They can automatically extract features from complex sleep data and perform accurate classification and prediction.

[0003] When implementing existing deep learning models for sleep health monitoring, most of them still remain at the level of monitoring statistics and analysis prompts for a single user's sleep data and sleep environment data, without combining the user's daily exercise data and beverage data for supervision and integrated analysis. This results in poor reliability of sleep recommendations and health management prompts for user sleep health monitoring, and the deep learning models for sleep health monitoring are not effective in personalized monitoring, analysis, and autonomous optimization for different individual users. Summary of the Invention

[0004] The purpose of the present invention is to provide a deep learning model optimization method for sleep health monitoring, which is used to solve the technical problem that the deep learning model of sleep health monitoring in the existing solution has poor personalized monitoring analysis and autonomous optimization effects for different individual users.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] Deep learning model optimization methods for sleep health monitoring, including:

[0007] Monitor and compile statistics on the user's daily exercise and beverage data, and use the sleep positive impact module, sleep negative impact module, or sleep fusion impact module in the sleep health monitoring model to process and analyze them to obtain sleep impact processing data corresponding to the user's daily exercise and beverage data;

[0008] The sleep health monitoring model monitors and analyzes the user's sleep data, and outputs the basic sleep quality data corresponding to the user's sleep at night. The sleep impact processing data is used to verify the reliability of the basic sleep quality data, and the sleep impact verification data corresponding to the user's single sleep health monitoring is obtained;

[0009] The sleep impact verification data corresponding to several sleep health monitoring sessions of the user is used to perform a reliability analysis of the impact effects of the sleep positive impact module, sleep negative impact module, and sleep fusion impact module in the sleep health monitoring model. Based on the evaluation results, the impact treatment plans implemented by different impact modules in the sleep health monitoring model are adaptively managed dynamically.

[0010] Preferably, when only motion data exists, the user's motion data for that day is processed and analyzed. If the motion time period in the motion data is not within the user's pre-sleep impact period, a positive impact processing instruction is generated, and according to the positive impact processing instruction, the motion intensity value and exercise duration in the motion data are analyzed through the sleep positive impact processing scheme associated with the sleep positive impact module, and the corresponding sleep positive impact processing data is output.

[0011] Preferably, if the exercise time period in the exercise data is within the user's pre-sleep impact period, a negative impact processing instruction is generated, and according to the negative impact processing instruction, the exercise intensity value and exercise duration in the exercise data are analyzed by the first sleep negative impact processing scheme associated with the sleep negative impact module, and the corresponding first sleep negative impact processing data is output.

[0012] Preferably, when only beverage data exists, the drinking time and drinking amount in the beverage data are analyzed using the second negative sleep impact processing solution associated with the negative sleep impact module, and corresponding second negative sleep impact processing data is output.

[0013] Preferably, when both exercise data and beverage data exist, the exercise intensity value and exercise duration in the exercise data and the drinking time and drinking amount in the beverage data are analyzed using the sleep fusion impact processing scheme associated with the sleep fusion impact module, and the corresponding sleep fusion impact processing data is output.

[0014] Preferably, the total sleep time, deep sleep time and light sleep time of the user in the basic sleep quality data are obtained, and the total sleep time, deep sleep time and light sleep time are respectively combined with the sleep impact processing data to obtain the corresponding sleep index reliability value through calculation.

[0015] Preferably, data analysis is performed on the reliability value of the sleep indicator, and the impact identification flag corresponding to the sleep indicator is set to 0 or 1 according to the analysis result;

[0016] All sleep indicators are sorted and combined with the impact identification identifiers obtained by the corresponding processing to obtain a single impact identification processing sequence.

[0017] Preferably, all elements in a single impact identification processing sequence are summed and analyzed;

[0018] If the sum of all elements is 0, the total number of completely effective treatments of the impact treatment scheme corresponding to the sleep impact treatment data is increased by one;

[0019] If the sum of all elements is greater than 0 and less than 3, the total number of effective processing of the impact processing scheme corresponding to the sleep impact processing data is increased by one;

[0020] If the sum of the elements is equal to 3, the total number of completely ineffective treatments of the impact treatment scheme corresponding to the sleep impact treatment data is increased by one;

[0021] The obtained single impact identification processing sequence and the total number of completely effective processing, partially effective processing or completely invalid processing updated by the corresponding impact processing scheme data are sorted and combined to obtain the sleep impact verification data corresponding to the user's single sleep health monitoring.

[0022] Preferably, based on the sleep impact verification data corresponding to several sleep health monitoring sessions of the user, the total number of completely effective treatments, the total number of partially effective treatments, and the total number of completely ineffective treatments associated with the sleep positive impact module, the sleep negative impact module, and the sleep fusion impact module in the sleep health monitoring model are counted respectively, and the impact effect reliability values corresponding to different impact modules are obtained by calculation;

[0023] Data analysis is performed on the reliability value of the impact effect to determine the sleep impact prediction processing effect of the corresponding impact module.

[0024] Preferably, if the impact effect reliability value is less than or equal to 1, it is determined that the sleep impact prediction processing effect of the corresponding impact module is normal, and the subsequent implementation of the existing impact processing solution of the corresponding impact module is maintained;

[0025] Otherwise, it is determined that the sleep impact prediction processing effect of the corresponding impact module is abnormal, and the existing impact processing plan of the impact module is dynamically managed.

[0026] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0027] The present invention monitors and compiles statistics on the user's daily exercise data and beverage data, and uses different impact modules in the sleep health monitoring model to implement targeted data processing and analysis to obtain sleep impact processing data corresponding to the user's daily exercise data and beverage data. This can provide diverse sleep impact data support for subsequent sleep recommendations and health management tips for the user's sleep health monitoring.

[0028] The present invention uses the user's daily exercise data and beverage data to verify the reliability of the basic sleep quality data corresponding to his or her sleep at night, and determines whether the user's daily exercise and beverage consumption have affected his or her sleep quality at night, so that reliable sleep suggestions can be provided to the user later, thereby improving the reliability and pertinence of sleep health monitoring analysis and suggestions for different users.

[0029] The present invention uses the sleep impact verification data corresponding to several sleep health monitoring sessions of users to perform reliability analysis on the impact effects of the sleep positive impact module, sleep negative impact module and sleep fusion impact module in the sleep health monitoring model, and adaptively and dynamically manages the impact processing schemes implemented by different impact modules in the sleep health monitoring model based on the evaluation results, thereby realizing autonomous supervision and optimization of the implementation effects of the sleep health monitoring models for different users, and improving the personalized monitoring analysis and autonomous optimization effects of the deep learning model of sleep health monitoring for different individual users. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below with reference to the accompanying drawings.

[0031] Figure 1 This is a module block diagram of the deep learning model optimization method for sleep health monitoring in the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] like Figure 1 As shown, the present invention is a deep learning model optimization method for sleep health monitoring, comprising:

[0034] Monitor and compile statistics on the user's daily exercise and beverage data, and use the sleep positive impact module, sleep negative impact module, or sleep fusion impact module in the sleep health monitoring model to process and analyze them to obtain sleep impact processing data corresponding to the user's daily exercise and beverage data;

[0035] The sleep impact processing data specifically includes positive sleep impact processing data, first negative sleep impact processing data, second negative sleep impact processing data, or fusion sleep impact processing data; including:

[0036] When only exercise data exists, the user's exercise data for the day is processed and analyzed. If the exercise time period in the exercise data is not within the user's pre-sleep influence period, a positive influence processing instruction is generated. According to the positive influence processing instruction, the exercise intensity value and exercise duration in the exercise data are analyzed through the sleep positive influence processing solution associated with the sleep positive influence module, and the corresponding sleep positive influence processing data is output;

[0037] It should be noted that the bedtime impact period is specifically one hour before the user goes to bed;

[0038] In addition, exercise intensity is determined based on the median heart rate across all exercise periods. Data from existing models predicting the impact of exercise on sleep show that for every 30 minutes of moderate to high-intensity daytime exercise, the model predicts an increase in deep sleep duration of approximately 8-12 minutes. Analysis of the exercise time window shows that exercise 3-4 hours before bedtime has the greatest effect on promoting deep sleep.

[0039] In addition, the positive impact of exercise on sleep is specifically the total duration of deep sleep; moderate exercise, especially moderate to high-intensity activity during the day, significantly increases the duration of deep sleep by raising core body temperature and promoting subsequent body temperature drop;

[0040] A sleep positive impact processing solution is used to match exercise data with an exercise positive impact experimental dataset and output corresponding sleep positive impact processing data; the exercise positive impact experimental dataset includes several experimental exercise intensity value ranges and experimental exercise duration ranges and the corresponding deep sleep estimated impact duration;

[0041] Sleep is positively impacting processing data, specifically including the estimated duration of deep sleep;

[0042] If the exercise time period in the exercise data falls within the user's pre-sleep impact period, a negative impact processing instruction is generated, and according to the negative impact processing instruction, the exercise intensity value and exercise duration in the exercise data are subjected to data analysis using a first sleep negative impact processing solution associated with the sleep negative impact module, and corresponding first sleep negative impact processing data is output;

[0043] It should be noted that when the exercise period is partly within the sleep-affecting period, the data analysis of the first sleep-affecting treatment plan is also performed on the entire exercise period;

[0044] In addition, the negative impact of exercise on sleep, specifically the total deep sleep duration indicator;

[0045] Specifically, strenuous exercise within 1 hour before bedtime causes sympathetic nerve excitement, temporarily reducing the proportion of deep sleep (by about 5-10%), but the overall impact on the total sleep duration indicator is small;

[0046] A first negative sleep impact processing solution is used to match the motion data with the motion negative impact experimental data set and output corresponding first negative sleep impact processing data;

[0047] The negative impact of exercise experiment dataset contains several experimental exercise intensity value ranges and experimental exercise duration ranges, as well as the corresponding estimated deep sleep impact durations. The experimental data in the negative impact of exercise experiment dataset is different from the experimental data in the positive impact of exercise experiment dataset.

[0048] The first sleep negative impact processing data includes the estimated impact duration of deep sleep, and the value is negative;

[0049] When only beverage data exists, the drinking time and drinking amount in the beverage data are analyzed using the second negative sleep impact processing solution associated with the negative sleep impact module, and the corresponding second negative sleep impact processing data is output;

[0050] It should be noted that the beverage is specifically coffee; the beverage data needs to be filled in by the user through a smart terminal, such as filling in through a smart phone and synchronously transmitting it to the smart bracelet, or filling in directly through the smart bracelet;

[0051] In addition, the negative effects of beverages on sleep are specifically the total sleep time index, the total deep sleep time index, and the total light sleep time index;

[0052] Caffeine directly inhibits deep sleep by antagonizing adenosine receptors. Experimental data shows that for every 100mg of caffeine consumed, deep sleep duration is reduced by 9.2±2.1 minutes, with the greatest effect within 6 hours of ingestion. When the residual caffeine concentration is ≥20mg, deep sleep duration decreases by 18%-25%.

[0053] Caffeine prolongs light sleep and causes sleep fragmentation. Experimental data: consuming 200mg of caffeine increases the proportion of light sleep from 8% to 12%.

[0054] Caffeine prolongs sleep onset latency and increases nighttime awakenings, resulting in a shortened total sleep duration. Experimental data: 200mg of caffeine consumed 3 hours before bedtime: Total sleep duration decreased by 40±12 minutes; the number of awakenings increased from 1.2 to 2.8 times.

[0055] In addition, the second negative sleep impact processing scheme is used to match the beverage data with the beverage negative impact experimental data set and output corresponding second negative sleep impact processing data;

[0056] The beverage negative impact experiment dataset includes several experimental drinking duration ranges and experimental drinking amount ranges, as well as the corresponding estimated total sleep impact duration, estimated deep sleep impact duration, and estimated light sleep impact duration;

[0057] The second sleep negative impact processing data specifically includes the total estimated sleep impact time, the estimated deep sleep impact time, and the estimated light sleep impact time; the corresponding values are negative, negative, and positive respectively;

[0058] When both exercise data and beverage data are present, the exercise intensity and duration in the exercise data, as well as the drinking time and amount in the beverage data, are analyzed using the sleep fusion impact processing solution associated with the sleep fusion impact module, and the corresponding sleep fusion impact processing data is output;

[0059] When users exercise and consume caffeine on the same day, the physiological effects of the two will produce dynamic antagonism and synergy:

[0060] Interaction between body temperature regulation and caffeine metabolism: The increase in body temperature caused by exercise accelerates caffeine metabolism, increasing the rate of decline of residual concentration by about 15%;

[0061] Caffeine safety threshold: When daily exercise volume is ≥60 minutes, caffeine intake ≤150mg can maintain normal deep sleep levels (±5% fluctuation);

[0062] Exercise compensation limit: When caffeine intake is ≥300mg, even after 90 minutes of exercise, deep sleep is still reduced by ≥10 minutes (compensation rate is less than 50%).

[0063] In addition, a sleep fusion impact processing scheme is used to match the exercise data and beverage data together with the fusion negative impact experimental data set and output the corresponding sleep fusion impact processing data;

[0064] The fusion negative impact experimental data set includes several experimental exercise intensity value ranges and experimental exercise duration ranges, several experimental drinking duration ranges and experimental drinking amount ranges, as well as the corresponding estimated total sleep impact duration, deep sleep impact duration, and light sleep impact duration;

[0065] Sleep fusion impact processing data, specifically including the total estimated sleep impact duration, deep sleep estimated impact duration, and light sleep estimated impact duration; the corresponding values are negative, negative, and positive respectively;

[0066] The absolute values of the different indicators in the sleep fusion impact processing data are all smaller than the absolute values of the corresponding indicators in the second sleep negative impact processing data;

[0067] In an embodiment of the present invention, by monitoring and counting the user's daily exercise data and beverage data, and using different impact modules in the sleep health monitoring model to implement targeted data processing and analysis, sleep impact processing data corresponding to the user's daily exercise data and beverage data can be obtained, which can provide diverse sleep impact data support for subsequent sleep recommendations and health management tips for the user's sleep health monitoring.

[0068] The sleep health monitoring model monitors and analyzes the user's sleep data, outputs basic sleep quality data corresponding to the user's sleep at night, and uses sleep impact processing data to verify the reliability of the basic sleep quality data, obtaining sleep impact verification data corresponding to the user's single sleep health monitoring; including:

[0069] Among them, the sleep health monitoring model is an existing deep learning model, such as CNN (convolutional neural network): processing the temporal features in PPG or EEG signals;

[0070] ‌LSTM (Long Short-Term Memory) network‌: Analyzes the continuity of sleep stages (such as the transition from wakefulness to deep sleep); can monitor sleep data based on the smart bracelet worn by the user;

[0071] The sleep health monitoring model monitors and analyzes the user's sleep data using existing conventional technical means. The specific implementation steps are not detailed here.

[0072] Understandably, most existing sleep health monitoring and analysis solutions still rely on monitoring and analyzing user sleep data, combined with environmental data about the user's sleep, to determine the user's sleep quality and provide prompts. However, they lack active monitoring, processing, and analysis of the user's daily exercise and beverage data. This results in unreliable notifications of poor sleep quality and the resulting sleep recommendations.

[0073] Basic sleep quality data includes the user's total sleep time, total deep sleep time, and total light sleep time. The content of basic sleep quality data can also be added, deleted, or modified according to the application requirements of actual application scenarios.

[0074] Obtain the user's total sleep time, deep sleep time, and light sleep time in the basic sleep quality data, all in minutes, and combine the total sleep time, deep sleep time, and light sleep time with the sleep impact processing data through the formula Calculate and obtain the corresponding sleep index reliability value Zk; where k is 1, 2, and 3, representing the total sleep duration index, deep sleep total duration index, and light sleep total duration index, respectively; Tk is T1, T2, and T3, representing the user's total sleep duration, deep sleep total duration, and light sleep total duration in the basic sleep quality data, respectively; t0k is t01, t02, and t03, representing the user's standard total sleep duration, deep sleep standard total duration, and light sleep standard total duration, respectively, and is obtained based on the median values of all basic sleep quality data corresponding to different indicators in the user's history without exercise or beverage consumption; tk is t1, t2, and t3, representing the total sleep duration impact value, deep sleep total duration impact value, and light sleep total duration impact value in the sleep impact processing set, respectively, and is determined based on the sleep impact processing data obtained from the corresponding processing of the user on that day;

[0075] It should be noted that the sleep index reliability value is used to process and calculate the impact prediction data and sleep monitoring data corresponding to different indicators to digitally represent their corresponding impact identification status;

[0076] If Zk∈[w1k, w2k], it indicates that the impact identification corresponding to the sleep indicator is reliable, and the impact identification flag corresponding to the sleep indicator is set to 0; w1k and w2k are the minimum and maximum errors corresponding to different sleep indicators, respectively, and can be determined based on the test error data of several previous sleep experiments in the laboratory;

[0077] Otherwise, it is prompted that the impact identification corresponding to the sleep indicator is unreliable, and the impact identification flag corresponding to the sleep indicator is set to 1;

[0078] Sort and combine the impact identification marks obtained by corresponding processing of all sleep indicators to obtain a single impact identification processing sequence;

[0079] In the embodiments of the present invention, by processing, digitally representing, and combining the impact recognition states corresponding to different sleep indicators, it is possible to monitor and analyze different sleep indicators and provide reliable local indicator data support for the sleep impact processing corresponding to a user's single sleep health monitoring.

[0080] All elements of a single impact identification processing sequence are summed and analyzed;

[0081] If the sum of all elements is 0, the total number of completely effective treatments of the impact treatment scheme corresponding to the sleep impact treatment data is increased by one;

[0082] If the sum of all elements is greater than 0 and less than 3, the total number of effective processing of the impact processing scheme corresponding to the sleep impact processing data is increased by one;

[0083] If the sum of the elements is equal to 3, the total number of completely ineffective treatments of the impact treatment scheme corresponding to the sleep impact treatment data is increased by one;

[0084] Sort and combine the obtained single impact identification processing sequence and the total number of completely effective processing, partially effective processing, or completely invalid processing corresponding to the impact processing solution data update to obtain the sleep impact verification data corresponding to the user's single sleep health monitoring;

[0085] In an embodiment of the present invention, the reliability of the basic sleep quality data corresponding to the user's sleep at night is verified by utilizing the user's exercise data and beverage data for the day, and it is determined whether the user's exercise and beverage consumption on the day have affected the user's sleep quality at night, so that reliable sleep advice can be provided to the user subsequently, thereby improving the reliability and pertinence of sleep health monitoring analysis and advice for different users.

[0086] The sleep impact verification data corresponding to several sleep health monitoring sessions of the user is used to conduct a reliability analysis of the impact effects of the sleep positive impact module, sleep negative impact module, and sleep fusion impact module in the sleep health monitoring model. Based on the evaluation results, the impact treatment solutions implemented by different impact modules in the sleep health monitoring model are adaptively managed dynamically; including:

[0087] According to the sleep impact verification data corresponding to several sleep health monitoring of users, due to the differences in the bodies of different users, the reliability of their sleep health monitoring analysis will be poor. Therefore, the specific value of several times should not be too small, at least 15 times. The total number of completely effective processing, partially effective processing and completely invalid processing associated with the sleep positive impact module, sleep negative impact module and sleep fusion impact module in the sleep health monitoring model are counted respectively, and the total number of completely effective processing, partially effective processing and completely invalid processing are calculated through the formula Calculate the reliability value Kd of the impact effect corresponding to different impact modules; where d is 1, 2, and 3, representing the positive sleep impact module, negative sleep impact module, and fusion sleep impact module, respectively; n1d, n2d, and n3d are the total number of completely effective treatments, partially effective treatments, and completely ineffective treatments corresponding to different impact modules, respectively; α is the error impact factor, which is a real number greater than one and has no specific value limit; Ad is the reliability threshold of the impact effect corresponding to different impact modules, determined based on the test error data of several sleep experiments in the laboratory in the early stage;

[0088] It should be noted that the impact effect reliability value is used to process and calculate all sleep impact data of different impact modules to digitally represent the corresponding sleep impact prediction processing effect;

[0089] It is understandable that the larger the total number of completely invalid processing and the total number of partially effective processing obtained by the impact module, the worse the effect of the sleep impact prediction processing on the user. The corresponding reason may be due to the user's physical condition. Even if the user exercises and consumes coffee that day, it will not have a significant impact on his or her sleep at night. In this case, targeted management of subsequent exercise of the impact module is required.

[0090] Conduct data analysis on the reliability value of the impact effect to determine the sleep impact prediction processing effect of the corresponding impact module;

[0091] If the impact effect reliability value is less than or equal to 1, it is determined that the sleep impact prediction processing effect of the corresponding impact module is normal, and the subsequent implementation of the existing impact processing plan of the corresponding impact module is maintained;

[0092] Otherwise, the sleep impact prediction processing effect of the corresponding impact module is determined to be abnormal, and the existing impact processing plan of the impact module is dynamically managed;

[0093] Dynamic management of the existing impact processing scheme of the impact module can be carried out. Specifically, the subsequent implementation of the existing impact processing scheme of the impact module can be stopped. Alternatively, the experimental data set of the existing impact processing scheme of the impact module can be supplemented and optimized with all the user's historical exercise data, beverage data, and corresponding basic sleep quality data. The optimized and trained impact processing scheme can then be used to perform subsequent sleep impact prediction processing on the user.

[0094] In addition, the experimental data sets of the existing impact treatment solutions are supplemented and optimized for training, which is a conventional technical means, such as optimizing training through the Q-learning model. The specific implementation steps are not repeated here.

[0095] In an embodiment of the present invention, the reliability of the impact effects of the sleep positive impact module, the sleep negative impact module and the sleep fusion impact module in the sleep health monitoring model are analyzed through the sleep impact verification data corresponding to several sleep health monitoring of the user, and the impact processing schemes implemented by different impact modules in the sleep health monitoring model are adaptively managed dynamically according to the evaluation results, thereby realizing autonomous supervision and optimization of the implementation effects of the sleep health monitoring models for different users, and improving the personalized monitoring analysis and autonomous optimization effects of the deep learning model of sleep health monitoring for different individual users.

[0096] In the several embodiments provided by the present invention, it should be understood that the disclosed system can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative. For example, the division of modules is only a logical function division, and other division methods may be used in actual implementation.

[0097] Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, and may be located in one place or distributed across multiple network modules. Some or all of these modules may be selected to achieve the objectives of this embodiment based on actual needs.

[0098] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing module, each module may exist physically separately, or two or more modules may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or hardware plus software functional modules.

[0099] It is obvious to a person skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, but that the present invention can be implemented in other specific forms without departing from the essential characteristics of the present invention.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A deep learning model optimization method for sleep health monitoring, characterized in that: include: Monitor and compile statistics on the user's daily exercise and beverage data, and use the sleep positive impact module, sleep negative impact module, or sleep fusion impact module in the sleep health monitoring model to process and analyze them to obtain sleep impact processing data corresponding to the user's daily exercise and beverage data; The sleep health monitoring model monitors and analyzes the user's sleep data, and outputs the basic sleep quality data corresponding to the user's sleep at night. The sleep impact processing data is used to verify the reliability of the basic sleep quality data, and the sleep impact verification data corresponding to the user's single sleep health monitoring is obtained; The sleep impact verification data corresponding to several sleep health monitoring sessions of the user is used to conduct a reliability analysis of the impact effects of the sleep positive impact module, sleep negative impact module, and sleep fusion impact module in the sleep health monitoring model. Based on the evaluation results, the impact treatment solutions implemented by different impact modules in the sleep health monitoring model are adaptively managed dynamically. Among them, according to the sleep impact verification data corresponding to several sleep health monitorings of the user, the total number of completely effective processing, the total number of partially effective processing and the total number of completely invalid processing associated with the sleep positive impact module, the sleep negative impact module and the sleep fusion impact module in the sleep health monitoring model are counted respectively, and the total number of completely effective processing and the total number of completely invalid processing are calculated by the formula Calculate and obtain the impact reliability value Kd corresponding to different impact modules; where d is 1, 2, and 3, representing the sleep positive impact module, sleep negative impact module, and sleep fusion impact module, respectively; n1d, n2d, and n3d are the total number of completely effective treatments, partially effective treatments, and completely ineffective treatments corresponding to different impact modules, respectively; α is the error impact factor, which is a real number greater than one; Ad is the impact reliability threshold corresponding to different impact modules; Conduct data analysis on the reliability value of the impact effect to determine the sleep impact prediction processing effect of the corresponding impact module; If the impact effect reliability value is less than or equal to 1, it is determined that the sleep impact prediction processing effect of the corresponding impact module is normal, and the subsequent implementation of the existing impact processing plan of the corresponding impact module is maintained; Otherwise, the sleep impact prediction processing effect of the corresponding impact module is determined to be abnormal, and the existing impact processing plan of the impact module is dynamically managed; When only exercise data exists, the user's exercise data for the day is processed and analyzed. If the exercise time period in the exercise data is not within the user's pre-sleep influence period, a positive influence processing instruction is generated. According to the positive influence processing instruction, the exercise intensity value and exercise duration in the exercise data are analyzed through the sleep positive influence processing solution associated with the sleep positive influence module, and the corresponding sleep positive influence processing data is output; If the exercise time period in the exercise data is within the user's sleep impact period, a negative impact processing instruction is generated, and according to the negative impact processing instruction, the exercise intensity value and exercise duration in the exercise data are analyzed by the first sleep negative impact processing scheme associated with the sleep negative impact module, and the corresponding first sleep negative impact processing data is output. When both exercise data and beverage data exist, the exercise intensity value and exercise duration in the exercise data and the drinking time and drinking amount in the beverage data are analyzed through the sleep fusion impact processing solution associated with the sleep fusion impact module, and the corresponding sleep fusion impact processing data is output.

2. The deep learning model optimization method for sleep health monitoring according to claim 1, characterized in that: When only beverage data exists, the drinking time and drinking amount in the beverage data are analyzed using the second negative sleep impact processing solution associated with the negative sleep impact module, and corresponding second negative sleep impact processing data is output.

3. The deep learning model optimization method for sleep health monitoring according to claim 2, characterized in that: Obtain the user's total sleep time, total deep sleep time, and total light sleep time from the basic sleep quality data, and calculate the corresponding sleep indicator reliability value by combining the total sleep time, total deep sleep time, and total light sleep time with the sleep impact processing data.

4. The deep learning model optimization method for sleep health monitoring according to claim 3, characterized in that: Perform data analysis on the reliability values of the sleep indicators, and set the impact identification flag corresponding to the sleep indicator to 0 or 1 according to the analysis results; All sleep indicators are sorted and combined with the impact identification identifiers obtained by the corresponding processing to obtain a single impact identification processing sequence.

5. The deep learning model optimization method for sleep health monitoring according to claim 4, characterized in that: All elements of a single impact identification processing sequence are summed and analyzed; If the sum of all elements is 0, the total number of completely effective treatments of the impact treatment scheme corresponding to the sleep impact treatment data is increased by one; If the sum of all elements is greater than 0 and less than 3, the total number of effective processing of the impact processing scheme corresponding to the sleep impact processing data is increased by one; If the sum of the elements is equal to 3, the total number of completely ineffective treatments of the impact treatment scheme corresponding to the sleep impact treatment data is increased by one; The obtained single impact identification processing sequence and the total number of completely effective processing, partially effective processing or completely invalid processing updated by the corresponding impact processing scheme data are sorted and combined to obtain the sleep impact verification data corresponding to the user's single sleep health monitoring.

Images