Data processing method and device for patient intelligent sleep-aiding glasses

Through intelligent sleep aid glasses, predict the physiological data and sleep status of the patient at the next moment, and adjust the sleep aid plan in real time, solving the problem of sleep aid music failure caused by timely adjustment in the existing technology, and improving the quality of sleep aid and user experience.

CN118557867BActive Publication Date: 2024-11-15SUZHOU MUNICIPAL HOSPITAL
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202410629842.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-21
Publication Date
2024-11-15
Estimated Expiration
2044-05-21

AI Technical Summary

Technical Problem

Existing sleep aid glasses adjust the sleep aid music after determining the sleep state, resulting in failure to adjust in time, causing the patient to wake up and fail to fall asleep, making it difficult to improve the quality of sleep aid and user experience.

Method used

Through intelligent sleep aid glasses, the physiological data of the patient is collected at the moment, the physiological data of the next moment is predicted, the sleep aid plan is generated based on changes in sleep state, and the music and light are adjusted in real time to assist sleep.

Benefits of technology

It has achieved high-quality sleep aid for users' sleep, ensured the quality of sleep, improved the user experience, and avoided the problem of waking up caused by the failure to adjust music in time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118557867B_ABST
    Figure CN118557867B_ABST
Patent Text Reader

Abstract

The present application provides a data processing method and device for smart sleep-aid glasses for patients. The method includes: a first server generates physiological data of the target patient at the next moment based on a pre-acquired physiological data sequence of the target patient at the current moment through a pre-trained sleep-aid prediction model, and sends it to the smart sleep-aid glasses, and the smart sleep-aid glasses generate a target sleep-aid solution based on the physiological data at the current moment and the physiological data at the next moment. Through the above method, the user's sleep is helped to achieve high-quality sleep, the user's sleep quality is guaranteed, and the user experience is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of sleep assistance technology, and in particular to a data processing method and device for a patient's smart sleep-aiding glasses. Background Art

[0002] As people's quality of life improves, they will choose to go to the hospital for treatment when they have health problems. Hospitals have a large number of staff and a lot of large medical equipment. Only by ensuring adequate rest can hospitalized patients improve their recovery ability and recover soon. However, in the complex scenes of hospitals, the noise in the wards is diverse, which makes it difficult for hospitalized patients to fall asleep. Therefore, ensuring the sleep quality of hospitalized patients becomes the key.

[0003] At present, most sleep-aiding programs for patients are based on EEG data to determine the patient's current sleep state, and then adjust and release sleep-aiding music according to the current sleep state. However, the above method is to adjust the sleep-aiding music after determining the sleep state. At this time, the patient has entered the current sleep state. If the sleep-aiding music is not adjusted in time, it will wake up the patient and cause the patient to fail to fall asleep.

[0004] In summary, how to improve the sleep-aiding quality of sleep-aid glasses and improve user experience is a technical problem that urgently needs to be solved in this field. Summary of the invention

[0005] The present application provides a data processing method and device for a patient's smart sleep-aid glasses, which are used to solve the problem of how to improve the sleep-aiding quality of the sleep-aid glasses and improve the user experience.

[0006] In a first aspect, the present application provides a data processing method for smart sleep-aid glasses for patients, which is applied to a sleep-aid system, wherein the sleep-aid system comprises smart sleep-aid glasses, a first server for sleep-aid prediction, and a second server for model training, wherein the smart sleep-aid glasses, the first server, and the second server are in communication connection; the method comprises:

[0007] The first server generates physiological data of the target patient at the next moment based on the pre-acquired physiological data sequence of the target patient at the current moment by calculating the pre-trained sleep-aid prediction model, and sends the data to the smart sleep-aid glasses, wherein the physiological data sequence includes the physiological data at the current moment and the physiological data at n moments before the current moment, and the sleep-aid prediction model is obtained by the second server training based on the meta-learning model;

[0008] The smart sleep-aiding glasses generate a target sleep-aiding program according to the physiological data at the current moment and the physiological data at the next moment, and the target sleep-aiding program is used to assist the sleep of the target patient.

[0009] In combination with the first aspect, in some embodiments, the first server generates physiological data of the target patient at the next moment according to a pre-acquired physiological data sequence of the target patient at the current moment by calculating using a pre-trained sleep-aid prediction model, including:

[0010] Dividing the physiological data sequence into a first support set and a first query set;

[0011] Inputting the first support set into the sleep-aid prediction model, optimizing the prediction module in the sleep-aid prediction model through the parameter update learning module of the sleep-aid prediction model, and obtaining the prediction module corresponding to the target patient;

[0012] The first query set is input into the prediction module corresponding to the target patient for calculation to generate the physiological data at the next moment.

[0013] In combination with the first aspect, in some embodiments, the smart sleep-aiding glasses generate a target sleep-aiding solution according to the physiological data at the current moment and the physiological data at the next moment, including:

[0014] Determining the sleep state of the target patient at the current moment and the sleep state at the next moment according to the physiological data at the current moment and the physiological data at the next moment;

[0015] Determining whether the sleep state of the target patient changes according to the sleep state at the current moment and the sleep state at the next moment;

[0016] If it is determined that the sleep state of the target patient has changed, then according to the sleep state at the next moment, the music corresponding to the sleep state at the next moment in the pre-set music set is determined as the target sleep-aiding program.

[0017] In combination with the first aspect, in some embodiments, the method further includes:

[0018] The second server divides the pre-acquired task set into a second support set and a second query set, wherein the task set includes a physiological data sequence of at least one patient;

[0019] The second server performs multi-task training on the meta-learning model according to the second support set and the second query set to obtain the sleep-aiding prediction model, and sends the model to the first server.

[0020] In combination with the first aspect, in some embodiments, the second server performs multi-task training on the meta-learning model according to the second support set and the second query set to obtain the sleep-aiding prediction model, including:

[0021] Inputting the second support set into the prediction module of the meta-learning model to obtain an output result of the prediction module;

[0022] The output result of the prediction module is input into the parameter update learning module of the meta-learning model, and after multiple iterations of calculation training, a trained prediction module is obtained;

[0023] Inputting the second query set into the trained prediction module, and obtaining target parameters of the parameter update learning module through back propagation calculation;

[0024] The target parameters are configured in the parameter update learning module to obtain the sleep-aiding prediction model.

[0025] In combination with the first aspect, in some embodiments, the method further includes:

[0026] If it is determined that the sleep state of the target patient has not changed, the sleep-aiding scheme corresponding to the physiological data at the current moment is determined as the target sleep-aiding scheme.

[0027] In combination with the first aspect, in some embodiments, the method further includes:

[0028] The smart sleep-aid glasses acquire sensor data, where the sensor data is body monitoring data of the target patient collected in real time based on sensors configured by the smart sleep-aid glasses;

[0029] Preprocessing and feature extraction are performed on the sensor data to obtain an initial physiological data sequence;

[0030] The initial physiological data sequence is normalized according to a preset normalization formula to obtain the physiological data sequence.

[0031] In combination with the first aspect, in some embodiments, the physiological data sequence includes electroencephalogram data, heart rate data, respiratory rate data and muscle activity data.

[0032] In a second aspect, the present application provides a data processing device for a patient's smart sleep-aiding glasses, comprising:

[0033] A prediction module, which is used for the first server to calculate the physiological data sequence of the target patient at the current moment based on the pre-acquired physiological data sequence of the target patient through the pre-trained sleep-aid prediction model, to generate the physiological data of the target patient at the next moment, and send it to the smart sleep-aid glasses, wherein the physiological data sequence includes the physiological data at the current moment and the physiological data at n moments before the current moment, and the sleep-aid prediction model is obtained by the second server through training based on the meta-learning model;

[0034] A generation module is used for the smart sleep-aid glasses to generate a target sleep-aiding program based on the physiological data at the current moment and the physiological data at the next moment, wherein the target sleep-aiding program is used to assist the sleep of the target patient.

[0035] In conjunction with the second aspect, in some embodiments, the prediction module includes:

[0036] a division unit, configured to divide the physiological data sequence into a first support set and a first query set;

[0037] an optimization unit, configured to input the first support set into the sleep-aid prediction model, optimize the prediction module in the sleep-aid prediction model through a parameter update learning module of the sleep-aid prediction model, and obtain a prediction module corresponding to the target patient;

[0038] A generating unit is used to input the first query set into a prediction module corresponding to the target patient for calculation, so as to generate the physiological data at the next moment.

[0039] In conjunction with the second aspect, in some embodiments, the generating module includes:

[0040] A first determining unit, configured to determine the sleep state of the target patient at a current moment and the sleep state of the target patient at a next moment according to the physiological data at the current moment and the physiological data at the next moment;

[0041] A second determining unit, configured to determine whether the sleep state of the target patient changes according to the sleep state at the current moment and the sleep state at the next moment;

[0042] The third determining unit is configured to determine, if it is determined that the sleep state of the target patient has changed, music corresponding to the sleep state at the next moment in a pre-set music set as the target sleep-aiding program according to the sleep state at the next moment.

[0043] In conjunction with the second aspect, in some embodiments, the device further includes:

[0044] A division module, configured for the second server to divide the pre-acquired task set into a second support set and a second query set, wherein the task set includes a physiological data sequence of at least one patient;

[0045] A training module is used for the second server to perform multi-task training on the meta-learning model according to the second support set and the second query set, obtain the sleep-aid prediction model, and send it to the first server.

[0046] In conjunction with the second aspect, in some embodiments, the training module includes:

[0047] An input unit, used for inputting the second support set into the prediction module of the meta-learning model to obtain an output result of the prediction module;

[0048] An iterative unit, used for inputting the output result of the prediction module into the parameter update learning module of the meta-learning model, and obtaining a trained prediction module after multiple iterative calculation training;

[0049] A reverse calculation unit, used for inputting the second query set into the trained prediction module, and obtaining target parameters of the parameter update learning module through reverse propagation calculation;

[0050] A configuration unit is used to configure the target parameter in the parameter update learning module to obtain the sleep-aid prediction model.

[0051] In conjunction with the second aspect, in some embodiments, the generating module includes:

[0052] The fourth determining unit is configured to determine the sleep-aiding scheme corresponding to the physiological data at the current moment as the target sleep-aiding scheme if it is determined that the sleep state of the target patient has not changed.

[0053] In conjunction with the second aspect, in some embodiments, the device further includes:

[0054] An acquisition module, used for the smart sleep-aid glasses to acquire sensor data, where the sensor data is body monitoring data of the target patient collected in real time based on sensors configured by the smart sleep-aid glasses;

[0055] A processing and extraction module, used for preprocessing and feature extraction of the sensor data to obtain an initial physiological data sequence;

[0056] The normalization module is used to perform normalization processing on the initial physiological data sequence according to a preset normalization formula to obtain the physiological data sequence.

[0057] In combination with the second aspect, in some embodiments, the physiological data sequence includes electroencephalogram data, heart rate data, respiratory rate data, and muscle activity data.

[0058] In a third aspect, the present application provides smart sleep-aid glasses, comprising: a processor, and a memory, a sensor, a communication interface, and a speaker that are communicatively connected to the processor;

[0059] The memory stores computer-executable instructions;

[0060] The processor executes the computer-executable instructions stored in the memory to implement the data processing method of the patient's smart sleep-aiding glasses described in any one of the first aspects.

[0061] In a fourth aspect, the present application provides a sleep aid system, comprising: smart sleep aid glasses, a first server for sleep aid prediction and a second server for model training, the smart sleep aid glasses, the first server and the second server being communicatively connected; and used to execute the data processing method for the patient's smart sleep aid glasses described in any one of the first aspects.

[0062] In a fifth aspect, the present application provides a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed by a processor, they are used to implement the data processing method for the patient's smart sleep-aid glasses described in any one of the first aspects.

[0063] In a sixth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the data processing method for the patient's smart sleep-aid glasses described in any one of the first aspects.

[0064] The data processing method and device of the patient smart sleep-aid glasses provided by the present application, the first server generates the physiological data of the target patient at the next moment according to the physiological data sequence of the target patient at the current moment acquired in advance, through the calculation of the pre-trained sleep-aid prediction model, and sends it to the smart sleep-aid glasses, and the smart sleep-aid glasses generate the target sleep-aid scheme according to the physiological data at the current moment and the physiological data at the next moment. Through the above method, the user's sleep is helped to achieve high-quality sleep-aiding effect, the user's sleep quality is guaranteed, and the user experience is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0066] Figure 1 This is an application scenario diagram of the data processing method of the patient's smart sleep-aiding glasses provided in an embodiment of the present application;

[0067] Figure 2 A flowchart of a first embodiment of a method for processing data of the patient's smart sleep-aiding glasses provided in an embodiment of the present application;

[0068] Figure 3 A flowchart of Embodiment 2 of the data processing method of the patient intelligent sleep-aiding glasses provided in the embodiment of the present application;

[0069] Figure 4 A flowchart of a third embodiment of a method for processing data of the smart sleep-aiding glasses for patients provided in an embodiment of the present application;

[0070] Figure 5A flowchart of a fourth embodiment of a method for processing data of the smart sleep-aiding glasses for patients provided in an embodiment of the present application;

[0071] Figure 6 A flowchart of a fifth embodiment of a method for processing data of the smart sleep-aiding glasses for patients provided in an embodiment of the present application;

[0072] Figure 7 A flowchart of a sixth embodiment of a method for processing data of the smart sleep-aiding glasses for patients provided in an embodiment of the present application;

[0073] Figure 8 A structural diagram of a first embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application;

[0074] Fig. 9 A structural diagram of Embodiment 2 of a data processing device for the patient's intelligent sleep-aiding glasses provided in an embodiment of the present application;

[0075] Fig.10 A structural diagram of a third embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application;

[0076] Fig.11 A structural schematic diagram of a fourth embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application;

[0077] Fig.12 A structural diagram of a fifth embodiment of a data processing device for the smart sleep-aiding glasses for patients provided in an embodiment of the present application;

[0078] Fig.13 A structural diagram of a sixth embodiment of a data processing device for the smart sleep-aiding glasses for patients provided in an embodiment of the present application;

[0079] Fig.14 A schematic diagram of the structure of the smart sleep-aiding glasses provided in an embodiment of the present application;

[0080] Fig.15 This is a schematic diagram of the architecture of the sleep aid system provided in an embodiment of the present application.

[0081] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0082] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0083] As people's quality of life improves, people will choose to go to the hospital for treatment when they have health problems. The hospital has a complex staff and a lot of large medical equipment. Only by ensuring adequate rest can hospitalized patients improve their recovery ability and recover soon. However, in the complex scenes of the hospital, the noise in the ward is diverse, which will make it difficult for hospitalized patients to fall asleep. Therefore, ensuring the sleep quality of hospitalized patients becomes the key. At present, most of the patient-based sleep aid programs are based on EEG data to determine the patient's current sleep state, and then adjust and release sleep-aid music according to the current sleep state. However, the above method is to adjust the sleep-aid music after determining the sleep state. At this time, the patient has entered the current sleep state. If the sleep-aid music is not adjusted in time, it will wake up the patient and cause the patient to fail to fall asleep.

[0084] In response to the above problems, the present application provides a data processing method and device for smart sleep-aid glasses for patients, which improves the sleep-aiding quality of the sleep-aid glasses and enhances the user experience. Specifically, in the prior art, the sleep-aiding schemes based on patients are mostly based on EEG data to determine the patient's current sleep state, and then adjust and release the sleep-aiding music according to the current sleep state. However, the above method is to adjust the sleep-aiding music after determining the sleep state. At this time, the patient has entered the current sleep state. If the sleep-aiding music is not adjusted in time, it will cause the patient to wake up and cause the patient to fail to fall asleep. Taking these problems into consideration, the inventor studied whether it is possible to collect the patient's physiological data at the current moment through smart sleep-aid glasses, and then predict the patient's physiological data at the next moment. Through the changes in the patient's sleep state at two moments, the noodle cooking plan is determined, and then high-quality sleep assistance is achieved for the patient. Based on this, the technical solution of the present application is proposed.

[0085] Figure 1The application scenario diagram of the data processing method of the patient smart sleep-aid glasses provided in the embodiment of the present application, the scenario includes at least one user and a sleep-aid system, wherein the sleep-aid system includes smart sleep-aid glasses, a first server for sleep-aid prediction and a second server for model training, and the smart sleep-aid glasses, the first server and the second server are connected in communication, which can be connected via Bluetooth communication or wireless communication, which is not specifically limited in the present application. The user can be a patient user in the hospital, or a user with insomnia, or a user without any disease or other troubles. The present application does not specifically limit the specific type of user. The smart sleep-aid glasses are equipped with sensors, which can collect the physiological data of the user wearing them in real time, and then send the real-time collected physiological data to the first server. The first server can predict the physiological data of the user at the next moment based on the physiological data at the current moment, and then determine the change in the user's sleep state, and generate a sleep-aid plan based on the change in the user's sleep state, and send the sleep-aid plan to the smart sleep-aid glasses. The smart sleep-aid glasses implement the plan according to the sleep-aid plan, thereby achieving the effect of helping the user sleep.

[0086] Optionally, the second server can train a sleep aid prediction model based on the physiological data of different users collected by the smart sleep aid glasses, and then send the trained sleep aid prediction model to the first server. The first server predicts the user's physiological data at the next moment through the sleep aid prediction model.

[0087] It should be noted that the specific types and forms of the physical devices mentioned above are not specifically limited in this application.

[0088] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0089] Figure 2 This is a flow chart of a first embodiment of a method for processing data of the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Figure 2 As shown, the method is applied to a sleep aid system, wherein the sleep aid system includes smart sleep aid glasses, a first server and a second server, and the smart sleep aid glasses, the first server and the second server are communicatively connected; then the method specifically includes:

[0090] S201: The first server generates the physiological data of the target patient at the next moment based on the pre-acquired physiological data sequence of the target patient at the current moment through calculation using a pre-trained sleep-aid prediction model, and sends the generated physiological data to the smart sleep-aid glasses.

[0091] In this step, in order to achieve high-quality sleep-aiding effects for users, it is necessary to determine a sleep-aiding plan based on the user's personalized characteristics. The first server pre-acquires the physiological data sequence of the target patient at the current moment, and then inputs the physiological data sequence at the current moment into a pre-trained sleep prediction model to predict the physiological data of the target patient at the next moment, wherein the physiological data sequence includes the physiological data at the current moment and the physiological data at n moments before the current moment, and the sleep-aiding prediction model is trained based on a meta-learning model.

[0092] Specifically, the physiological data sequence is divided into a first support set and a first query set, and then the first support set is input into the sleep-aid prediction model, and the prediction module in the sleep-aid prediction model is optimized through the parameter update learning module of the sleep-aid prediction model to obtain the prediction module corresponding to the target patient. The first query set is input into the prediction module corresponding to the target patient for calculation to generate the physiological data at the next moment.

[0093] By adopting the above-mentioned sleep-aid prediction model, individual differentiation can be achieved, so that the smart sleep-aid glasses can achieve high-quality sleep-aid for different users.

[0094] For example, the physiological data sequence can be expressed as {X t-H+1 ,X 1-H+2 ,...,X t}, where X t represents the physiological data at the current moment, and H represents the preset time length. Assuming n=1 and the preset time length is H, the form of the physiological data sequence at the above n+1 moments is described.

[0095] In a possible implementation, the time corresponding to the data in the physiological data sequence at time n+1 overlaps. For example, the physiological data sequence at time t1 {X t1 -H+1,…,X t1-1 , X t1}, physiological data sequence at time t2 {X t1 -H+2,…,X t1 , X t2}.

[0096] In another possible implementation, the time corresponding to the data in the physiological data sequence at the n time points does not overlap. For example, the physiological data sequence at time t1 {X t1 -H+1,…,X t1-1 , X t1}, physiological data sequence at time t2 {X t2 -H+1,…,X t2-1 , X t2}.

[0097] S202: The smart sleep-aiding glasses generate a target sleep-aiding plan based on the physiological data at the current moment and the physiological data at the next moment.

[0098] In this step, after receiving the physiological data of the target patient at the next moment predicted by the first server, the smart sleep-aid glasses determine the changes in the sleep state of the target patient based on the physiological data at the current moment and the physiological data at the next moment, thereby generating a target sleep-aid plan.

[0099] Specifically, the target patient's sleep state at the current moment and the sleep state at the next moment are determined based on the physiological data at the current moment and the physiological data at the next moment, and then whether the target patient's sleep state has changed is determined based on the sleep state at the current moment and the sleep state at the next moment. If it is determined that the target patient's sleep state has changed, then based on the sleep state at the next moment, the music in the pre-set music set corresponding to the sleep state at the next moment is determined as the target sleep-aid plan; if it is determined that the target patient's sleep state has not changed, then the sleep-aid plan corresponding to the physiological data at the current moment is determined as the target sleep-aid plan.

[0100] In one possible implementation, the targeted sleep-aid plan can also include a light adjustment plan. After determining the changes in the target patient's sleep state, the sleep-aid glasses can also collect real-time light information of the target patient's external environment, and then determine the light adjustment plan based on the target patient's sleep state at the next moment. By adjusting the light on the lenses of the sleep-aid glasses, the user is provided with a visual environment that is more conducive to falling asleep, thereby achieving high-quality sleep assistance.

[0101] In another possible implementation method, the target sleep-aiding plan can also include a massage plan. The sleep-aiding glasses generate a massage plan according to the changes in the target patient's sleep state. A small mechanical hammer is configured on the sleep-aiding glasses to relax the target patient's nerves and achieve the effect of helping the target patient sleep.

[0102] The data processing method of the patient smart sleep-aid glasses provided in this embodiment is that the first server generates the physiological data of the target patient at the next moment based on the physiological data sequence of the target patient at the current moment acquired in advance through the pre-trained sleep-aid prediction model, and sends it to the smart sleep-aid glasses. The smart sleep-aid glasses generate a target sleep-aid solution based on the physiological data at the current moment and the physiological data at the next moment. Through the above method, the user's sleep is helped to achieve high-quality sleep, the user's sleep quality is guaranteed, and the user experience is improved.

[0103] Figure 3 This is a flow chart of Embodiment 2 of the data processing method of the patient intelligent sleep aid glasses provided in the present application embodiment. Figure 3As shown, based on the above embodiment, step S201 specifically includes:

[0104] S301: Divide the physiological data sequence into a first support set and a first query set.

[0105] In this step, after the first server obtains the physiological data sequence, in order to more accurately predict the physiological data of the target patient at the next moment, it is necessary to divide the physiological data sequence into a first support set and a first query set, and then perform prediction through the first support set and the first query set.

[0106] Specifically, the first support set is used to adapt to individual differences of users, and the first query set is used for prediction of physiological data.

[0107] Regarding how to obtain the first support set and the first query set, optionally, the physiological data sequence is divided into the first support set and the first query set according to a preset ratio in chronological order.

[0108] Exemplarily, the ratio of the number of data in the first support set to the first query set is 8: 2. It should be noted that the above is only an illustration of a ratio provided in the embodiment of the present application, and is not specifically limited.

[0109] Take the example that the physiological data sequence includes 10 physiological data. Specifically, the first 4 data are used as the first support subset {X1, X2, X3, X4}, and the difference between data X4 and data X5 is used as the label of the first support subset; the second support subset {X2, X3, X4, X5}, and the difference between data X5 and data X6 is used as the label of the second support subset, and similarly, until the fifth support subset {X5, X6, X7, X8} and the label of the fifth support subset are obtained. The first to fifth support subsets together constitute the above-mentioned first support set.

[0110] The first query set includes the first query subset {X6, X7, X8, X9}, and the labels of the first query subset are data X9 and data X 10 The difference between the two, and the second query subset {X7, X8, X9, X 10}, the label of the second query subset is data X 10 With data X 11 The difference.

[0111] S302: Inputting the first support set into the sleep-aid prediction model, optimizing the prediction module in the sleep-aid prediction model through the parameter update learning module of the sleep-aid prediction model, and obtaining the prediction module corresponding to the target patient.

[0112] In this step, in order to adapt to individual differences, the first support set is input into the sleep aid prediction model. Through the calculation of the prediction module, the difference between the prediction result and the actual value is input into the parameter update learning module for parameter optimization calculation to obtain the parameters of the prediction module. The optimization is terminated through continuous iterative calculation until the difference between the prediction result of the prediction module and the actual data reaches a preset threshold. The result output by the last iteration of the parameter update learning module is the parameter of the prediction module that is more suitable for the target patient. The parameter is configured in the prediction module to obtain the prediction module corresponding to the target patient.

[0113] It should be noted that the above-mentioned parameter update learning module is used to update the parameters of the above-mentioned prediction module, and can be a deep learning network model, such as a recurrent neural network (RNN) model, a long short-term memory (LSTM) neural network model, or other models that can realize updating the parameters of the prediction module, which is not limited in this application.

[0114] The above prediction module is used to obtain the prediction result according to the physiological data sequence of the target patient at the current moment, which can be a deep learning network model, such as Convolutional Neural Networks (CNN), or other models that can realize the prediction data, which is not limited in this application. The prediction result is used to characterize the difference between the predicted physiological data at the next moment and the physiological data at the current moment.

[0115] S303: Input the first query set into the prediction module corresponding to the target patient for calculation to generate physiological data at the next moment.

[0116] In this step, after the prediction module corresponding to the target patient is obtained, the first query set is input into the prediction module corresponding to the target patient for calculation, and the obtained prediction result is the physiological data of the target patient at the next moment.

[0117] The data processing method of the patient intelligent sleep-aid glasses provided in this embodiment divides the physiological data sequence into a first support set and a first query set, and then inputs the first support set into the sleep-aid prediction model, and optimizes the prediction module in the sleep-aid prediction model through the parameter update learning module of the sleep-aid prediction model to obtain the prediction module corresponding to the target patient. The first query set is input into the prediction module corresponding to the target patient for calculation to generate the physiological data at the next moment. Through the above method, the applicability of individual differences is achieved, thereby achieving high-quality sleep aid for different users.

[0118] Figure 4This is a flow chart of Embodiment 3 of the data processing method of the patient intelligent sleep-aiding glasses provided in the present application. Figure 4 As shown, based on the above embodiments, step S202 specifically includes:

[0119] S401: Determine the target patient's sleep state at the current moment and the sleep state at the next moment according to the physiological data at the current moment and the physiological data at the next moment.

[0120] In this step, after receiving the physiological data of the target patient at the next moment predicted by the first server, the smart sleep-aid glasses determine the sleep state of the target patient at the current moment and the next moment based on the physiological data of the target patient at the current moment and the physiological data of the next moment, in order to assist the target patient's sleep with high quality, avoid the target patient from being disturbed by the outside world, and ensure the sleep quality of the target patient. The sleep state includes the awake state and the non-rapid eye movement sleep state.

[0121] Specifically, non-rapid eye movement sleep states include: N1, N2 and N3. The awake state is manifested by high-frequency and low-amplitude EEG signals. Stage N1 is the transition stage from the awake state to sleep, which usually lasts for a short time, accounting for about 5% of the total sleep time. Stage N2 is a light sleep stage, accounting for about 50% of the total sleep time. In this stage, brain activity gradually becomes slower, but there is still intermittent rapid activity. Stage N3 is a deep sleep stage, also known as slow-wave sleep, accounting for about 20-25% of the total sleep time. This stage of sleep is the deepest, difficult to wake up, and is essential for physical recovery and rest. At this stage, the EEG shows a large amount of slow wave (low-frequency and high-amplitude) activity.

[0122] Exemplarily, when the user is in different sleep states, the phase and frequency information of the slow wave frequency in the EEG signal are also different, so the phase and frequency information of the slow wave frequency in the EEG data can be obtained. The sleep state is determined by presetting the mapping relationship between the phase, frequency information and the sleep state. For example, when the user is awake, the characteristic wave is mainly composed of α waves and β waves, when the user is in the N1 stage of non-rapid eye movement sleep, the characteristic wave is mainly composed of θ waves and α waves, when the user is in the N3 stage of non-rapid eye movement sleep, the characteristic wave is mainly composed of δ waves, and when the user is in the N2 stage of non-rapid eye movement sleep, the characteristic wave is mainly composed of θ waves, α waves and Sawtooth waves.

[0123] S402: Determine whether the sleep state of the target patient changes according to the sleep state at the current moment and the sleep state at the next moment.

[0124] S403: If it is determined that the sleep state of the target patient has changed, then according to the sleep state at the next moment, music corresponding to the sleep state at the next moment in the pre-set music set is determined as the target sleep-aiding solution.

[0125] By comparing the sleep status at the current moment and the next moment, if the sleep status is different, it means that the sleep status of the target patient has changed, and the sleep aid plan can be adjusted. According to the sleep status at the next moment, the music that best matches the sleep status at the next moment is matched in the pre-set music collection and determined as the target sleep aid plan.

[0126] Optionally, the target sleep-aiding program also includes the volume of playing the above-mentioned music.

[0127] S404: If it is determined that the sleep state of the target patient has not changed, the sleep-aiding program corresponding to the physiological data at the current moment is determined as the target sleep-aiding program.

[0128] By comparing the sleep status at the current moment and the next moment, if the sleep status is the same, it means that the sleep status of the target patient has not changed. At this time, there is no need to modify the sleep aid plan, and the current sleep aid plan can continue to be used as the target sleep aid plan.

[0129] In one possible implementation, if the target patient's sleep state changes, the smart sleep-aid glasses can also adjust the volume of music playing in the current sleep-aid program based on the noise collected from the external environment.

[0130] The data processing method of the patient intelligent sleep-aid glasses provided in this embodiment determines the target patient's sleep state at the current moment and the sleep state at the next moment based on the physiological data at the current moment and the physiological data at the next moment, and determines whether the target patient's sleep state has changed based on the sleep state at the current moment and the sleep state at the next moment. If it is determined that the target patient's sleep state has changed, then based on the sleep state at the next moment, the music in the pre-set music set corresponding to the sleep state at the next moment is determined as the target sleep-aid solution; if it is determined that the target patient's sleep state has not changed, then the sleep-aid solution corresponding to the physiological data at the current moment is determined as the target sleep-aid solution. By analyzing the changes in the patient user's sleep state at adjacent moments, the sleep-aid solution for the target patient is determined, thereby ensuring high-quality sleep assistance and improving the user experience.

[0131] Figure 5 This is a flow chart of a fourth embodiment of a method for processing data of the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Figure 5 As shown, based on the above embodiments, the method further includes:

[0132] S501: The second server divides the pre-acquired task set into a second support set and a second query set.

[0133] In this step, the second server may further divide the pre-acquired task set into a second support set and a second query set for training the sleep aid prediction model, wherein the task set includes a physiological data sequence of at least one patient.

[0134] The specific implementation process of dividing the second support set and the second query set is the same as the specific implementation method of step S301 in the above embodiment, which will not be repeated here.

[0135] S502: The second server performs multi-task training on the meta-learning model according to the second support set and the second query set to obtain a sleep-aiding prediction model, and sends the model to the first server.

[0136] After obtaining the second support set and the second query set, the second server performs multi-task training on the meta-learning model according to the second support set and the second query set, and sends the trained sleep-aid prediction model to the first server.

[0137] Specifically, the second support set is input into the prediction module of the meta-learning model to obtain the output result of the prediction module, and then the output result of the prediction module is input into the parameter update learning module of the meta-learning model. After multiple iterative calculations and training, the trained prediction module is obtained, and then the second query set is input into the trained prediction module. After back propagation calculation, the target parameters of the parameter update learning module are obtained. Finally, the target parameters are configured in the parameter update learning module to obtain the sleep-aid prediction model.

[0138] In the data processing method of the patient intelligent sleep-aiding glasses provided in this embodiment, the second server divides the pre-acquired task set into a second support set and a second query set, and then the second server performs multi-task training on the meta-learning model according to the second support set and the second query set to obtain a sleep-aiding prediction model, and sends it to the first server. By pre-training the meta-learning model, a sleep-aiding prediction model with strong adaptability and capable of providing high-quality personalized sleep-aiding is obtained.

[0139] Figure 6 This is a flow chart of a fifth embodiment of a method for processing data of the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Figure 6 As shown, based on the above embodiments, step S502 specifically includes:

[0140] S601: Input the second support set into the prediction module of the meta-learning model to obtain an output result of the prediction module.

[0141] In this step, the second support set is calculated by the prediction module of the meta-learning model to obtain the output result, wherein the output result is the loss function value and the gradient value, and the loss function value can be the mean absolute error (MAE), the root mean square error (RMSE), etc. The gradient value is used to indicate the changing trend of the loss function.

[0142] S602: Input the output result of the prediction module into the parameter update learning module of the meta-learning model, and obtain a trained prediction module after multiple iterations of calculation training.

[0143] The output results of each prediction module are input into the parameter update learning module of the meta-learning model for calculation, so as to obtain the parameters of the updated prediction module. After multiple iterative calculations, the update is terminated until the output results of the prediction module meet the preset threshold, and the trained prediction module is obtained.

[0144] For example, taking the initial parameter of the prediction module as θ1, the subsets in the second support set are calculated in chronological order. In the first iteration calculation, the output results of the prediction module are L1 and After the parameter update learning module calculates, the parameter θ2 of the prediction module is output. By configuring θ2 in the prediction module, the updated prediction module is used to calculate the subset in the second support set, and the output results L2 and L3 are obtained. Then continue to input the obtained output result into the parameter update learning module for calculation, and output the parameter θ3 of the prediction module. Repeat the iterative calculation to obtain the optimal parameters of the prediction module and update them in the prediction module.

[0145] S603: Input the second query set into the trained prediction module, and obtain the target parameters of the parameter update learning module through back propagation calculation.

[0146] In this step, after obtaining the trained prediction module, the parameters of the prediction module are fixed, and the parameters of the parameter update learning module are updated, thereby obtaining the target parameters of the parameter update learning module.

[0147] Specifically, according to the second query set, for the above-mentioned last training iteration, the parameter update learning module can use the back propagation mechanism to obtain the parameter ψ′ of the parameter update learning module based on the output result of the penultimate training iteration of the trained prediction module and the parameters adopted in the penultimate training iteration of the prediction module. The parameter ψ′ is the parameter adopted by the parameter update learning module in the last training iteration.

[0148] Exemplarily, the data acquired for the first time is the first subset {X4, X5, X6, X7} in the second query set. Referring to the example in the previous step, the parameter used by the trained prediction module is θ4. For the first training iteration:

[0149] (1) The trained prediction module adopts the parameter θ4 and uses the data obtained for the first time to obtain the output results L1 and L2 of the first training iteration of the prediction module.

[0150] (2) The parameter update learning module is based on the parameter θ4 used by the prediction module, as well as the output results L1 and L2 of the first training iteration of the prediction module. The back propagation mechanism is used to obtain the parameter ψ′1 of the parameter update learning module in the second training iteration, and the parameter update learning module is updated using the parameter used in the second training iteration of the parameter update learning module.

[0151] The data obtained for the second time is the second subset {X5, X6, X7, X8} in the second query set, for the second training iteration:

[0152] (1) The trained prediction module adopts the parameter θ4 and uses the data obtained for the second time to obtain the output results L2 and L3 of the second training iteration of the prediction module.

[0153] (2) The parameter update learning module is based on the parameter θ4 used by the prediction module, as well as the output results L2 and L4 of the second training iteration of the prediction module. The back-propagation mechanism is used to obtain the parameter ψ′2 of the parameter update learning module, and the parameter ψ′2 adopted by the parameter update learning module is used to update the above parameter update learning module to obtain a trained parameter update learning module.

[0154] S604: configuring the target parameters in the parameter update learning module to obtain a sleep-aiding prediction model.

[0155] For example, using the method described in the above steps, a sleep-aid prediction model is finally obtained, wherein the parameter used by the parameter update learning module of the sleep-aid prediction model is ψ′2; the parameters of the prediction module of the sleep-aid prediction model need to be determined by the parameter update learning module based on the physiological data sequence input into the sleep-aid prediction model. The parameters used by the prediction module can initially be the initialization parameters, or the parameters θ4 determined by the above training.

[0156] The data processing method of the patient intelligent sleep-aid glasses provided in this embodiment inputs the second support set into the prediction module of the meta-learning model to obtain the output result of the prediction module, and then inputs the output result of the prediction module into the parameter update learning module of the meta-learning model. After multiple iterations of calculation and training, the trained prediction module is obtained, and then the second query set is input into the trained prediction module. After back-propagation calculation, the target parameters of the parameter update learning module are obtained, and finally the target parameters are configured in the parameter update learning module to obtain the sleep-aid prediction model. By training the meta-learning model, the optimal parameters of the parameter update learning module are determined, and a sleep-aid prediction model that can adapt to individual differences is obtained, providing high-quality personalized sleep assistance.

[0157] Figure 7 This is a flow chart of a sixth embodiment of the data processing method of the patient intelligent sleep-aiding glasses provided in the present application. Figure 7 As shown, based on the above embodiments, the method further includes:

[0158] S701: The smart sleep-aid glasses obtain sensor data.

[0159] In this step, in order to achieve real-time monitoring of the target patient and thus ensure the quality of sleep aid, sensors are configured in the smart sleep aid glasses to collect the user's sensor data in real time. The sensor data is body monitoring data of the target patient collected in real time based on the sensors configured in the smart sleep aid glasses, including EEG data, heart rate data, respiratory rate data, muscle activity data, body temperature data, ambient light data, head posture and movement data, ambient sound data, etc.

[0160] S702: Preprocess and extract features of sensor data to obtain an initial physiological data sequence.

[0161] S703: Performing normalization processing on the initial physiological data sequence according to a preset normalization formula to obtain a physiological data sequence.

[0162] In order to extract data that has an impact on the target patient's sleep from the collected sensor data, the sensor data obtained in the above steps is preprocessed and feature extracted to obtain an initial physiological data sequence. In order to ensure the accuracy of the initial physiological data sequence, the initial physiological data sequence needs to be normalized to obtain a physiological data sequence.

[0163] Optionally, preprocessing of sensor data includes data cleaning, detecting and processing outliers and missing values ​​in the data to ensure the integrity and consistency of the data. Data standardization, standardizing the data collected by different sensors to have similar scales and ranges to avoid the model being affected by different data ranges. Data dimensionality reduction, standardizing the data collected by different sensors to have similar scales and ranges to avoid the model being affected by different data ranges. Data smoothing, for data that may be noisy, filtering techniques such as moving average or Kalman filtering can be used to smooth the data to improve the stability and reliability of the data. Time alignment, if the collected data has different sampling rates or timestamps, it needs to be time aligned to ensure the consistency of the data at the same time point. Data segmentation, segmenting the data into appropriate time windows or paragraphs so that the model can process continuous data streams and improve the accuracy of predictions.

[0164] The above are feasible processing methods for preprocessing. The specific processing method is not specifically limited in this embodiment and can be selected according to the type of data collected.

[0165] The initial physiological data sequence can be normalized using the Min-Max normalization formula.

[0166] Optionally, standardization can be used.

[0167] The present embodiment provides a data processing method for smart sleep-aid glasses for patients. The smart sleep-aid glasses obtain sensor data, then pre-process and extract features of the sensor data to obtain an initial physiological data sequence. Finally, the initial physiological data sequence is normalized according to a preset normalization formula to obtain a physiological data sequence. Through the above method, the accuracy of the data is guaranteed, providing a data basis for subsequent predictions.

[0168] Figure 8 This is a structural diagram of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application. Figure 8 As shown, the data processing device 800 of the patient's smart sleep-aiding glasses includes:

[0169] Prediction module 801 is used for the first server to generate the physiological data of the target patient at the next moment based on the pre-acquired physiological data sequence of the target patient at the current moment through the pre-trained sleep-aid prediction model, and send it to the smart sleep-aid glasses. The physiological data sequence includes the physiological data at the current moment and the physiological data at n moments before the current moment. The sleep-aid prediction model is trained by the second server based on the meta-learning model.

[0170] The generation module 802 is used for the smart sleep-aiding glasses to generate a target sleep-aiding solution according to the physiological data at the current moment and the physiological data at the next moment, and the target sleep-aiding solution is used to assist the sleep of the target patient.

[0171] Fig. 9 This is a structural diagram of a second embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Fig. 9 As shown, the prediction module 801 includes:

[0172] The division unit 901 is configured to divide the physiological data sequence into a first support set and a first query set.

[0173] The optimization unit 902 is used to input the first support set into the sleep-aiding prediction model, optimize the prediction module in the sleep-aiding prediction model through the parameter update learning module of the sleep-aiding prediction model, and obtain the prediction module corresponding to the target patient.

[0174] The generating unit 903 is used to input the first query set into the prediction module corresponding to the target patient for calculation, so as to generate physiological data at the next moment.

[0175] Fig.10 This is a structural diagram of a third embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Fig.10 As shown, the generating module 802 includes:

[0176] The first determining unit 1001 is used to determine the sleep state of the target patient at the current moment and the sleep state of the target patient at the next moment according to the physiological data at the current moment and the physiological data at the next moment.

[0177] The second determining unit 1002 is used to determine whether the sleep state of the target patient changes according to the sleep state at the current moment and the sleep state at the next moment.

[0178] The third determining unit 1003 is used to determine the music corresponding to the sleep state at the next moment in the pre-set music set as the target sleep-aiding solution according to the sleep state at the next moment if it is determined that the sleep state of the target patient has changed.

[0179] The fourth determining unit 1004 is configured to determine the sleep-aiding scheme corresponding to the physiological data at the current moment as the target sleep-aiding scheme if it is determined that the sleep state of the target patient has not changed.

[0180] Fig.11 This is a structural diagram of a fourth embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Fig.11 As shown, the data processing device 800 of the patient's smart sleep-aiding glasses also includes:

[0181] The division module 1101 is used for the second server to divide the pre-acquired task set into a second support set and a second query set, where the task set includes a physiological data sequence of at least one patient.

[0182] The training module 1102 is used for the second server to perform multi-task training on the meta-learning model according to the second support set and the second query set to obtain a sleep-aiding prediction model, and send it to the first server.

[0183] Fig.12 This is a structural diagram of a fifth embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Fig.12 As shown, the training module 1102 specifically includes:

[0184] The input unit 1201 is used to input the second support set into the prediction module of the meta-learning model to obtain the output result of the prediction module.

[0185] The iteration unit 1202 is used to input the output result of the prediction module into the parameter update learning module of the meta-learning model, and obtain the trained prediction module after multiple iterations of calculation and training.

[0186] The reverse calculation unit 1203 is used to input the second query set into the trained prediction module, and obtain the target parameters of the parameter update learning module through reverse propagation calculation.

[0187] The configuration unit 1204 is used to configure the target parameters in the parameter update learning module to obtain the sleep-aid prediction model.

[0188] Fig.13 This is a structural diagram of a sixth embodiment of a data processing device for the patient's smart sleep-aiding glasses provided in an embodiment of the present application, as shown in FIG. Fig.13 As shown, the data processing device 800 of the patient's smart sleep-aiding glasses also includes:

[0189] The acquisition module 1301 is used for the smart sleep-aid glasses to acquire sensor data, where the sensor data is body monitoring data of the target patient collected in real time based on the sensors configured by the smart sleep-aid glasses.

[0190] The processing and extraction module 1302 is used to pre-process and extract features from the sensor data to obtain an initial physiological data sequence.

[0191] The normalization module 1303 is used to perform normalization processing on the initial physiological data sequence according to a preset normalization formula to obtain a physiological data sequence.

[0192] Optionally, based on the above-mentioned various device embodiments, the physiological data sequence includes EEG data, heart rate data, respiratory rate data and muscle activity data.

[0193] The data processing device of the patient's smart sleep-aid glasses provided in the above-mentioned embodiments is used to execute the data processing method of the patient's smart sleep-aid glasses in any of the above-mentioned method embodiments. Its implementation principle and technical effect are similar and will not be repeated here.

[0194] Fig.14 The structure diagram of the smart sleep-aid glasses provided in the embodiment of the present application is as follows: the smart sleep-aid glasses 1400 include: a processor 1402, and a memory 1401 connected to the processor 1402 for communication, a sensor 1403, a communication interface 1404, and a speaker 1405;

[0195] Memory 1401 stores computer-executable instructions.

[0196] The processor 1402 executes the computer-executable instructions stored in the memory 1401 to implement the data processing method of the patient's smart sleep-aiding glasses in any of the aforementioned method embodiments.

[0197] Sensor 1403 is used to collect the patient's physiological data in real time.

[0198] The communication interface 1404 is used to implement data communication with other external devices.

[0199] Speaker 1405 is used to achieve in-ear sound playback for the patient.

[0200] Fig.15 This is a schematic diagram of the architecture of a sleep aid system provided in an embodiment of the present application. The sleep aid system 1500 includes:

[0201] Smart sleep-aid glasses 1400, a first server 1501 for sleep-aid prediction, and a second server 1502 for model training, wherein the smart sleep-aid glasses 1400, the first server 1501, and the second server 1502 are communicatively connected. The system is used to execute the data processing method of the patient's smart sleep-aid glasses in any of the aforementioned method embodiments. The technical principles and technical effects implemented are similar and will not be repeated here.

[0202] An embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the data processing method of the patient's smart sleep-aid glasses in any embodiment.

[0203] The computer-readable storage medium mentioned above can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0204] Optionally, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0205] An embodiment of the present application also provides a computer program product, which includes a computer program. The computer program is stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, the technical solution provided by any of the above method embodiments can be implemented.

[0206] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0207] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A sleep aid system, comprising smart sleep aid glasses, a first server for sleep aid prediction, and a second server for model training, wherein the smart sleep aid glasses, the first server, and the second server are in communication connection; The first server generates physiological data of the target patient at the next moment based on the pre-acquired physiological data sequence of the target patient at the current moment by calculating the pre-trained sleep-aid prediction model, and sends the data to the smart sleep-aid glasses, wherein the physiological data sequence includes the physiological data at the current moment and the physiological data at n moments before the current moment, and the sleep-aid prediction model is obtained by the second server training based on the meta-learning model; The smart sleep-aiding glasses generate a target sleep-aiding solution according to the physiological data at the current moment and the physiological data at the next moment, wherein the target sleep-aiding solution is used to assist the sleep of the target patient; The first server generates physiological data of the target patient at the next moment according to the pre-acquired physiological data sequence of the target patient at the current moment by calculating the pre-trained sleep-aid prediction model, including: The first server divides the physiological data sequence into a first support set and a first query set; The first server inputs the first support set into the sleep-aid prediction model, optimizes the prediction module in the sleep-aid prediction model through the parameter update learning module of the sleep-aid prediction model, and obtains the prediction module corresponding to the target patient; The first server inputs the first query set into a prediction module corresponding to the target patient for calculation, so as to generate the physiological data at the next moment; The smart sleep-aiding glasses generate a target sleep-aiding solution according to the physiological data at the current moment and the physiological data at the next moment, including: The smart sleep-aiding glasses determine the target patient's sleep state at the current moment and the sleep state at the next moment according to the physiological data at the current moment and the physiological data at the next moment; The smart sleep-aiding glasses determine whether the sleep state of the target patient changes according to the sleep state at the current moment and the sleep state at the next moment; If the smart sleep-aid glasses determine that the sleep state of the target patient has changed, then, according to the sleep state at the next moment, music in a pre-set music set corresponding to the sleep state at the next moment is determined as the target sleep-aid solution; Among them, it also includes: The second server divides the pre-acquired task set into a second support set and a second query set, wherein the task set includes a physiological data sequence of at least one patient; The second server performs multi-task training on the meta-learning model according to the second support set and the second query set to obtain the sleep-aiding prediction model, and sends the model to the first server; The second server performs multi-task training on the meta-learning model according to the second support set and the second query set to obtain the sleep-aiding prediction model, including: The second server inputs the second support set into the prediction module of the meta-learning model to obtain an output result of the prediction module; The second server inputs the output result of the prediction module into the parameter update learning module of the meta-learning model, and obtains a trained prediction module after multiple iterative calculation training; The second server inputs the second query set into the trained prediction module, and obtains the target parameters of the parameter update learning module through back propagation calculation; The second server configures the target parameter in the parameter update learning module to obtain the sleep-aid prediction model; Among them, it also includes: If the smart sleep-aid glasses determine that the sleep state of the target patient has not changed, the sleep-aid program corresponding to the physiological data at the current moment is determined as the target sleep-aid program.

2. The system according to claim 1, characterized in that Also includes: The smart sleep-aid glasses acquire sensor data, where the sensor data is body monitoring data of the target patient collected in real time based on sensors configured by the smart sleep-aid glasses; The smart sleep-aiding glasses preprocess and extract features from the sensor data to obtain an initial physiological data sequence; The smart sleep-aiding glasses perform normalization processing on the initial physiological data sequence according to a preset normalization formula to obtain the physiological data sequence.

3. The system according to any one of claims 1 to 2, characterized in that: The physiological data sequence includes electroencephalogram data, heart rate data, respiratory rate data and muscle activity data.

Citation Information

Patent Citations

  • Sleep state detection method, device and equipment and computer readable medium

    CN112690761A

  • Model acquisition system, gesture recognition method and device, equipment and storage medium

    CN114781439A

  • Sleep intervention equipment control method and device, storage medium and computer equipment

    CN115206491A

  • Intelligent sleep aiding method and device and wearable equipment

    CN116726344A

  • Music sleep aiding method, device and equipment based on sleep monitoring

    CN116803337A