A device for collecting wearing feedback data of smart underwear

Abstract

The present invention discloses a method for collecting wearing feedback data of smart underwear, which relates to the technical field of wearing data collection, including: after analyzing and obtaining the state index of pre-processed data, switching the data sending state of the smart underwear; after receiving the health monitoring data of patients with chronic diseases, identifying the abnormal data, and issuing an abnormal reminder if there is abnormal data; building the aggregation degree of abnormal reminders according to the time nodes when the abnormal reminders are received within the content recommendation cycle, if the aggregation degree exceeds expectations, identifying the health trend of the chronic disease patients, and if it does not meet the expectations, using a collaborative filtering algorithm to output recommended content for the chronic disease patients; analyzing and obtaining the improvement degree, if the improvement degree exceeds expectations, switching the working state of the smart underwear; switching the working mode of the smart underwear according to the sleep state of the chronic disease patients, so that the smart underwear can realize multi-state operation and ensure the physical state of the patients.

Description

Technical Field

[0001] The present invention relates to the technical field of wearing data collection, and in particular to a device for collecting wearing feedback data of smart underwear. Background Art

[0002] Chronic diseases are a general term for diseases that have a slow onset, a long course, and often no clear infectious biological cause. These diseases are usually hidden and the initial symptoms are not obvious, but as time goes by, the condition gradually worsens, causing long-term and far-reaching effects on the patient's physical health. Chronic diseases include but are not limited to cardiovascular and cerebrovascular diseases (such as hypertension, coronary heart disease), diabetes, malignant tumors, chronic respiratory diseases, mental disorders and mental illnesses. They not only seriously threaten the patient's quality of life and life expectancy, but also bring heavy economic and psychological burdens to the patient's family.

[0003] Smart underwear is a new type of product that combines the comfort of traditional underwear with modern technology elements. It can be simply understood as ordinary underwear + smart chips and sensors. Compared with traditional underwear, smart underwear has the following characteristics: Real-time monitoring of health data: Through built-in sensors, smart underwear can monitor and evaluate the wearer's health status in real time, such as heart rate, body temperature, respiratory rate, etc. Data transmission and analysis: Some smart underwear supports wireless transmission function, which can transmit the monitored data to mobile phones or computers for users to view and analyze at any time.

[0004] The Chinese invention patent with the authorization announcement number CN105512476B involves a data collection method and a smart wearable device. The smart wearable device obtains the user's electronic medical record; the smart wearable device determines the parameters that need to be monitored based on the electronic medical record; when the parameters do not fall within the normal parameter range, the smart wearable device sends a reminder to the user. The smart wearable device can ensure that the patient understands his or her physical condition in real time and ensures that he or she takes timely measures.

[0005] Combined with the above application and the contents of the prior art:

[0006] Since patients with chronic diseases are in a sub-healthy state for a long time, their health indicators may be abnormal or fluctuate greatly. In this case, it is necessary to collect health data of chronic diseases regularly and deal with them in time when the health data of chronic disease patients is abnormal, so as to protect the health of chronic disease patients.

[0007] When patients with chronic diseases wear clothes, especially smart underwear, since smart underwear usually has data collection function, the health status of chronic disease patients can be grasped in real time by collecting various health data of chronic disease patients in real time; but when the patient is in a sleeping state, the various health indicators of chronic disease patients may be in a relatively stable state, therefore, some sensors of smart underwear may also enter a dormant or low-power state; however, considering that the patient's physical health may also be poor when the sleep quality is poor, therefore, if the smart underwear chooses its own working mode according to whether the patient is asleep, it may be difficult to provide effective protection for the health of chronic disease patients in the long run.

[0008] To this end, the present invention provides a device for collecting wearing feedback data of smart underwear. Summary of the invention

[0009] 1. Technical issues to be resolved

[0010] In view of the deficiencies in the prior art, the present invention provides a device for collecting wearing feedback data of smart underwear, which identifies abnormal data and issues an abnormal reminder if any; constructs the aggregation degree of abnormal reminders based on the time nodes when the abnormal reminders are received within the content recommendation cycle; if the aggregation degree exceeds expectations, identifies the health trends of chronic disease patients; if the health trends do not meet expectations, uses a collaborative filtering algorithm to output recommended content for chronic disease patients; analyzes and obtains the improvement degree; if the improvement degree exceeds expectations, switches the working state of the smart underwear; switches the working mode of the smart underwear based on the sleep state of the chronic disease patient, so that the smart underwear can achieve multi-state operation and ensure the patient's physical condition, thereby solving the technical problems raised in the background technology.

[0011] (II) Technical solution

[0012] To achieve the above objectives, the present invention is implemented through the following technical solutions: a device for collecting wearing feedback data of smart underwear, including collecting sleep data of patients with chronic diseases, using a trained sleep recognition model to identify the current sleep pattern of the patients with chronic diseases, performing similarity analysis between the current sleep pattern and the reference sleep pattern, generating a difference Cy from the acquired analysis data, and issuing a deep work instruction if the difference Cy exceeds expectations;

[0013] Start the sensor network on the smart underwear, set the data collection frequency for the sensor, collect and pre-process the data of various physical indicators of patients with chronic diseases, analyze and obtain the status indicators of the pre-processed data, and issue status analysis instructions to the outside;

[0014] The sending state coefficient Zts is generated by the data state indicator set, and the data sending state of the smart underwear is switched according to the sending state coefficient Zts; after receiving the health monitoring data of patients with chronic diseases, the abnormal data is identified, and if there is abnormal data, an abnormal reminder is issued to the outside;

[0015] The aggregation degree Jos of abnormal reminders is constructed based on the time nodes of abnormal reminders received during the content recommendation cycle. If the aggregation degree exceeds expectations, the health trend of chronic disease patients is identified. If the health trend does not meet expectations, the collaborative filtering algorithm is used to output recommended content for chronic disease patients.

[0016] The improvement degree Gts is obtained by analyzing the sleep data within the feedback cycle. If the improvement degree Gts exceeds expectations, the working state of the smart underwear is switched. If not, a multi-objective optimization algorithm is used to optimize the adjustment recommendation content, and the optimized recommendation content is output to chronic disease patients in the next feedback cycle.

[0017] Furthermore, after the chronic disease patients enter the sleep state, the sleep quality and physiological parameters of the chronic disease patients are collected and monitored, and the collected data are summarized as sleep data to obtain a sleep data set of the chronic disease patients; the sleep data of the chronic disease patients is used as input, and the identified sleep patterns are summarized to generate a sleep pattern set.

[0018] Furthermore, description information is added to the sleep mode, and a similarity analysis is performed between the corresponding description information and the description information of the reference sleep mode. Under dimensionless conditions, the difference Cy is generated by obtaining the corresponding similarity data. If the difference Cy exceeds the difference threshold, the smart underwear is switched from the shallow working state to the deep working state; wherein,

[0019]

[0020] Among them, n is the number of description information, Xs i is the similarity between the description information of the ith item and the description information of the reference sleep mode, Xs a is the average similarity; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1.

[0021] Furthermore, a sensor network is installed on the smart underwear. After receiving a deep work instruction, the dormant sensor is started, and a data collection frequency is set for the sensor. Data of various physical indicators of chronic disease patients are collected at the collection nodes that meet the constraint conditions, and the collected data are summarized to generate a health monitoring data set.

[0022] Preprocess the monitoring data in the health monitoring data set, analyze and obtain the status indicators of the preprocessed data, and generate a data status indicator set after aggregation.

[0023] Furthermore, the constraint conditions for the sensor data acquisition frequency are as follows:

[0024]

[0025] where n is the number of data acquisitions within the data acquisition period, T ij is the time interval from the i-th data acquisition to the j-th data acquisition, T a is the average value of the time intervals, and Nts is the energy consumption within the sensor data acquisition period.

[0026] Furthermore, after receiving the status analysis instruction, a transmission status coefficient Zts of the preprocessed data is constructed from the data status index set, where linear normalization processing is performed on the transmission duration Ct and the response duration Xt in the following manner:

[0027]

[0028] Weight coefficients, 0 ≤ α ≤ 1, 0 ≤ β ≤ 1.

[0029] Furthermore, if the transmission status coefficient Zts exceeds the status threshold, the preprocessed data is sent to the data receiving end by the Bluetooth module; otherwise, it is sent by the Wi-Fi module;

[0030] Using the health monitoring data received by the data receiving end as input, the trained abnormal data recognition model is used to recognize abnormal data.

[0031] Furthermore, a content recommendation period is preset, the time nodes when abnormal reminders are received are recorded within the content recommendation period, and an aggregation degree Jos of the abnormal reminders is constructed. If the aggregation degree exceeds the aggregation degree threshold, a trend recognition instruction is sent to the outside, where:

[0032]

[0033] In the formula: d ij <R, e takes the value 2.718, weight coefficients, 0 ≤ F1 ≤ 1, 0 ≤ F2 ≤ 1, and F1 + F2 = 1; R is the radius of the content recommendation period; p i The weighted value of the i-th abnormal reminder, d ij is the time interval from the i-th abnormal reminder to the j-th abnormal reminder, and n is the number of time nodes.

[0034] Furthermore, using the obtained health monitoring data as input, the trained health trend recognition model is used to recognize the health trend of chronic disease patients; if the recognized health trend does not match the expectation, a chronic disease health report is generated for the chronic disease patients based on the obtained health monitoring data.

[0035] Furthermore, after feature engineering is performed on the contents recorded in the chronic disease health report, the corresponding health features are obtained; a recommended content library is pre-built, and the health features of chronic disease patients are used as input, and the recommended content library outputs recommended content for chronic disease patients.

[0036] Furthermore, after obtaining the sleep data in each sub-period, a difference degree Cy is generated, and several difference degrees Cy are arranged along the time axis and analyzed to obtain the improvement degree Gts. If the improvement degree Gts exceeds the improvement threshold, the smart underwear is switched back to the shallow working state; if no improvement is achieved, a content optimization instruction is issued to the outside.

[0037] Furthermore, the method of arranging several difference degrees Cy along the time axis and analyzing and obtaining the improvement degree Gts is as follows:

[0038]

[0039] Among them, Cy i is the difference within the ith sub-period, Cy a is the corresponding mean, Cy b is its qualified standard value;

[0040] Pt i is the i-th improved intermediate value, Pt a is its mean, i=1, 2,…, k, k is the number of sub-periods, and the weight coefficients are: 0≤F1≤1, 0≤F2≤1 and F2+F1=1.

[0041] Furthermore, after receiving the content optimization instruction, feedback data from chronic disease patients after wearing smart underwear is collected, and features are extracted from the feedback data. Based on the corresponding optimization features obtained, a multi-objective optimization algorithm is used to optimize the adjusted recommended content to obtain optimized recommended content.

[0042] (III) Beneficial effects

[0043] The present invention provides a device for collecting wearing feedback data of smart underwear, which has the following beneficial effects:

[0044] 1. The health of chronic disease patients can be judged based on their sleep patterns. When the sleep effect of chronic disease patients is poor, their sleep state can be adjusted in time to avoid the physical condition of chronic disease patients being affected by poor sleep.

[0045] 2. By constructing the difference degree Cy, the sleep status of chronic disease patients in the current stage can be comprehensively evaluated. When the overall sleep status of chronic disease patients is poor, measures can be taken to protect the health of chronic disease patients.

[0046] 3. By setting two different working modes, shallow and deep, for the smart underwear and its sensor network, after entering the working state, the corresponding data collection frequency and number of collection times are set for each sensor to maintain a balance between the sensor's data collection and operating state.

[0047] 4. Select a corresponding data transmission mode for the collected data based on the transmission state coefficient Zts, and form a certain balance between data transmission and energy consumption by switching the data transmission mode.

[0048] 5. Identify abnormal data through the abnormal data recognition model, determine whether there are abnormalities in the health indicators of chronic disease patients and the frequency of abnormalities, evaluate the physical condition of chronic disease patients, adjust the health status of chronic disease patients in time, and protect the health of chronic disease patients.

[0049] 6. Based on the aggregation degree Jos, the current health status of chronic disease patients can be evaluated holistically. By identifying the collected health monitoring data, the health trends of chronic disease patients can be judged and a chronic disease health report can be generated. Using the chronic disease health report as feedback, the current health status of chronic disease patients can be described.

[0050] 7. After chronic disease patients obtain the recommended content, they can use it as a reference to adjust their sleep status or physical health status. After the smart underwear data is collected and processed, the physical condition of chronic disease patients can be adjusted and improved.

[0051] 8. The improvement degree of the sleep status of patients with chronic diseases can be evaluated based on the improvement degree Gts. If the sleep status of patients with chronic diseases has improved after data collection and processing, the working mode of the smart underwear can be switched again; the working mode of the smart underwear can be switched based on the sleep status of patients with chronic diseases, so that the smart underwear can achieve multi-state operation and effectively protect the physical condition of patients with chronic diseases.

[0052] 9. By optimizing and adjusting the recommended content, the reference effect can be improved, and the recommended content can be kept in a state of continuous dynamic adjustment, making the recommended content more compatible with chronic disease patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of the method for collecting wearable feedback data of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0055] See also Figure 1 The present invention provides a method for collecting wearing feedback data of smart underwear, comprising:

[0056] Step 1: Collect sleep data of patients with chronic diseases, use the trained sleep recognition model to identify the current sleep pattern of patients with chronic diseases, perform similarity analysis on it and the reference sleep pattern, generate a difference Cy from the acquired analysis data, and issue a deep work instruction to the outside if the difference Cy exceeds expectations;

[0057] The step 1 includes the following contents:

[0058] Step 101, setting a sleep monitoring cycle including several monitoring sub-cycles, collecting and monitoring the sleep quality and physiological parameters of the chronic disease patient after the chronic disease patient enters a sleep state, such as sleep duration, sleep efficiency, sleep stages, heart rate, breathing rate and pattern, blood oxygen saturation and blood pressure, etc. These can be monitored using portable sleep monitoring equipment, such as smart bracelets, smart sleep monitors and other equipment; the collected data is summarized as sleep data to obtain a sleep data set of the chronic disease patient;

[0059] Step 102: training a convolutional neural network with the labeled sample data to obtain a trained sleep recognition model, taking the sleep data of the chronic disease patient as input, using the trained sleep recognition model to identify the current sleep pattern of the chronic disease patient, and aggregating the identified sleep patterns to generate a sleep pattern set;

[0060] When in use, when the chronic disease patient is in a sleeping state, various sleep and related data of the chronic disease patient are collected. By identifying the sleep state of the chronic disease patient, the physical health of the chronic disease patient can be judged based on the chronic disease patient's sleep pattern. When the chronic disease patient's sleep effect is poor, the sleep state of the chronic disease patient can be adjusted in time to prevent the chronic disease patient from being affected by poor sleep and affecting his physical condition.

[0061] Step 103: add description information to each sleep mode, perform similarity analysis between the corresponding description information and the description information of the reference sleep mode, obtain corresponding similarity data, generate difference Cy from the similarity data under dimensionless conditions, and evaluate the sleep quality in the current stage based on the obtained difference Cy, in the following manner:

[0062]

[0063] Among them, n is the number of description information, Xs i is the similarity between the description information of the ith item and the description information of the reference sleep mode, Xs a is the average value of similarity; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; the weight coefficient can be obtained by referring to the hierarchical analysis method;

[0064] Pre-set difference thresholds based on historical data and expectations for managing sleep patterns in patients with chronic diseases;

[0065] If the difference Cy exceeds the difference threshold, it means that the chronic disease patient's current sleep state is poor. When the chronic disease patient has a chronic disease, the chronic disease symptoms of the chronic disease patient may worsen. At this time, the smart underwear needs to be switched from a shallow working state to a deep working state, and a deep working instruction is issued to the outside.

[0066] When using, combine the contents in steps 101 to 103:

[0067] After continuously identifying and obtaining the sleep patterns of chronic disease patients for several times, by comparing the current sleep pattern with the target sleep pattern and constructing the difference degree Cy through the comparison results, it is possible to make a holistic assessment of the sleep state of chronic disease patients in the current stage and verify the current sleep effect of chronic disease patients. When the overall sleep state of chronic disease patients is poor, it is helpful to protect the physical health of chronic disease patients from the perspective of sleep or similar.

[0068] When patients with chronic diseases wear clothes, especially smart underwear, since smart underwear usually has data collection function, the health status of chronic disease patients can be grasped in real time by collecting various health data of chronic disease patients in real time; but when the patient is in a sleeping state, the various health indicators of chronic disease patients may be in a relatively stable state, therefore, some sensors of smart underwear may also enter a dormant or low-power state; however, considering that the patient's physical health may also be poor when the sleep quality is poor, therefore, if the smart underwear chooses its own working mode according to whether the patient is asleep, it may be difficult to provide effective protection for the health of chronic disease patients in the long run.

[0069] Step 2: Start the sensor network on the smart underwear, set the data collection frequency for the sensor, collect and preprocess the data of various physical indicators of patients with chronic diseases, analyze and obtain the status indicators of the preprocessed data, and then send a status analysis instruction to the outside;

[0070] The step 2 includes the following contents:

[0071] Step 201: Install a sensor network on the smart underwear, including heart rate, temperature, humidity, motion and pressure sensors, etc.; after receiving a deep work instruction, start the sensor in the dormant state and set the data collection frequency for the sensor. The collection frequency constraint conditions are as follows:

[0072]

[0073] Where n is the number of data collection times in the data collection cycle, T ij is the time interval from the i-th data collection to the j-th data collection, T a is the average value of the time interval, Nts is the energy consumption during the sensor data collection period;

[0074] At the collection nodes that meet the constraints, data on various physical indicators of patients with chronic diseases are collected, and the collected data are summarized to generate a health monitoring data set;

[0075] When in use, two different working modes, shallow and deep, are set for the smart underwear and its sensor network. After entering the working state, by setting the corresponding data collection frequency and number of collection times for each sensor, the balance between the data collection and the operating state of the sensor can be maintained;

[0076] Step 202: preprocess the monitoring data in the health monitoring data set, including data cleaning, data transformation, data smoothing, feature extraction, etc., to obtain preprocessed data; analyze and obtain status indicators of the preprocessed data, including data volume, estimated transmission time when data is sent, and response time of receiving data; generate a data status indicator set after aggregation, and issue a status analysis instruction to the outside;

[0077] When using, combine the contents in steps 201 to 202:

[0078] After completing data collection, the data quality of the collected data can be improved by preprocessing the collected data. The transmission efficiency of the collected data can be improved by collecting data status indicators and selecting data output methods.

[0079] Step 3: Generate a sending state coefficient Zts from the data state indicator set, and switch the data sending state of the smart underwear according to the sending state coefficient Zts; after receiving the health monitoring data of patients with chronic diseases, identify abnormal data, and if there is abnormal data, send an abnormal reminder to the outside;

[0080] The step three includes the following contents:

[0081] Step 301: After receiving the state analysis instruction, the transmission state coefficient Zts of the pre-processed data is constructed by the data state indicator set, wherein the data transmission time Ct and the response time Xt of the received data are linearly normalized in the following manner:

[0082]

[0083] Weight coefficient, 0≤α≤1, 0≤β≤1; the weight coefficient can be obtained by referring to the hierarchical analysis method;

[0084] Step 302: pre-set a state threshold according to historical data and management expectations for data transmission; switch the data transmission state according to the relationship between the transmission state coefficient and the state threshold; if the transmission state coefficient Zts exceeds the state threshold, the pre-processed data is sent from the Bluetooth module to the data receiving end; otherwise, it is sent by the Wi-Fi module;

[0085] When in use, after completing the construction of the sending state coefficient Zts, the corresponding data sending mode is selected for the collected data according to the sending state coefficient Zts, and by switching the data sending mode, a certain balance is formed between data transmission and energy consumption, and the operating state of the smart underwear is adjusted;

[0086] Step 303: train a convolutional neural network with the labeled sample data to obtain a trained abnormal data recognition model. The health monitoring data received by the data receiving end is used as input, and the trained abnormal data recognition model is used to recognize abnormal data. When abnormal data exists, an abnormal reminder is issued to the outside.

[0087] When using, combine the contents in steps 301 to 303:

[0088] After completing data collection and transmission, abnormal data is identified through the abnormal data recognition model to determine whether there are abnormalities in the health indicators of chronic disease patients and the frequency of abnormalities, so as to evaluate the physical condition of chronic disease patients. After receiving abnormal reminders, the health status of chronic disease patients can be adjusted in time, so that when the smart underwear is in operation, chronic disease patients can be monitored to protect their health.

[0089] Step 4: Construct the aggregation degree Jos of abnormal reminders according to the time nodes of abnormal reminders received during the content recommendation cycle. If the aggregation degree exceeds expectations, identify the health trend of chronic disease patients. If the health trend does not meet expectations, use collaborative filtering algorithm to output recommended content for chronic disease patients.

[0090] The step 4 includes the following contents:

[0091] Step 401: Preset the content recommendation period. Record the time nodes when abnormal reminders are received within the content recommendation period, and construct the aggregation degree Jos of the abnormal reminders in the following way:

[0092]

[0093] In the formula: d ij <R, e takes the value of 2.718, the weight coefficient, 0 ≤ F1 ≤ 1, 0 ≤ F2 ≤ 1, and F1 + F2 = 1; R is the radius of the content recommendation period; p i The weighted value of the i-th abnormal reminder, whose value falls within [0.2, 0.5], d ij is the time interval from the i-th abnormal reminder to the j-th abnormal reminder, and n is the number of time nodes;

[0094] Preset the aggregation degree threshold based on historical data and the management of the severity of abnormal reminders;

[0095] If the aggregation degree exceeds the aggregation degree threshold, it indicates that the physical state of the chronic disease patients may be abnormal during the current period. At this time, send a trend recognition instruction to the outside;

[0096] When in use, after continuously receiving several abnormal reminders within the content recommendation period, generate the aggregation degree Jos of the abnormal reminders. If the aggregation degree of the abnormal reminders is higher and the reminders are sent more densely, the physical state of the chronic disease patients may be worse. Therefore, the current health state of the chronic disease patients can be evaluated comprehensively based on the aggregation degree;

[0097] Step 402: Use the labeled sample data to train the convolutional neural network to obtain the trained health trend recognition model. Use the obtained health monitoring data as the input, and use the trained health trend recognition model to identify the health trend of the chronic disease patients; if the identified health trend does not match the expectation, generate a chronic disease health report for the chronic disease patients based on the obtained health monitoring data;

[0098] When in use, when the current physical state of the chronic disease patients is poor, identify the health trend of the chronic disease patients based on the collected health monitoring data, and generate a chronic disease health report. Using the chronic disease health report as feedback, the current health state of the chronic disease patients can be described;

[0099] Step 403: After performing feature engineering on the content recorded in the chronic disease health report, obtain the corresponding health features;

[0100] Through retrieval, collection and formulation, a number of push content items are obtained, such as exercise types, diet plans and daily routines, etc. After aggregation, a recommended content library is pre-built. Using collaborative filtering algorithms, the health characteristics of chronic disease patients are used as input, and the recommended content library outputs recommended content for chronic disease patients;

[0101] When using, combine the contents in steps 401 to 403:

[0102] Based on the chronic disease health report, a collaborative filtering algorithm is used to output recommended content for chronic disease patients from the recommended content library. After the chronic disease patients obtain the recommended content, they can use it as a reference to adjust their sleep status or physical health status. After the smart underwear data is collected and processed, the physical condition of chronic disease patients can be adjusted and improved.

[0103] Step 5: Obtain the improvement degree Gts by analyzing the sleep data in the feedback cycle. If the improvement degree Gts exceeds expectations, switch the working state of the smart underwear. If the improvement degree Gts does not exceed expectations, use a multi-objective optimization algorithm to optimize the adjustment recommendation content, and output the optimized recommendation content to the chronic disease patient in the next feedback cycle.

[0104] The step five includes the following contents:

[0105] Step 501: pre-set a feedback cycle including several sub-cycles, generate a difference degree Cy after obtaining the sleep data in each sub-cycle, arrange the several difference degrees Cy along the time axis and analyze to obtain the improvement degree Gts, and judge whether the sleep state of the chronic disease patient is improved by the improvement degree Gts, according to the following method:

[0106]

[0107] Among them, Cy i is the difference within the ith sub-period, Cy a is the corresponding mean, Cy b is its qualified standard value;

[0108] Pt i is the i-th improved intermediate value, Pt a is its mean, i=1, 2, …, k, k is the number of sub-periods, weight coefficient: 0≤F1≤1, 0≤F2≤1 and F2+F1=1; the weight coefficient is consistent with the previous value;

[0109] Pre-set improvement thresholds based on historical data and expectations for improvement in sleep status for patients with chronic diseases;

[0110] If the improvement degree Gts exceeds the improvement threshold, it means that the sleep quality of the chronic disease patient has gradually improved, and the smart underwear is switched back to the light working state; if it has not improved, a content optimization instruction is sent to the outside to optimize the recommended content received by the chronic disease patient;

[0111] When in use, after the chronic disease patient obtains the recommended content as a reference, the improvement degree Gts is constructed according to the degree of change in the chronic disease patient's sleep state. According to the improvement degree Gts, the degree of improvement in the chronic disease patient's sleep state can be evaluated. If the chronic disease patient's sleep state has improved after data collection and processing, the working mode of the smart underwear can be switched again; based on data collection and processing, the working mode of the smart underwear can be switched according to the chronic disease patient's sleep state, so that the smart underwear can achieve multi-state operation, effectively realizing protection of the physical condition of chronic disease patients.

[0112] Step 502: after receiving the content optimization instruction, the wearing experience data, sleep quality data and body monitoring data of the chronic disease patients after wearing the smart underwear are collected, and the data are summarized as the feedback data of the smart underwear. The feedback data is subjected to feature extraction, and the multi-objective optimization algorithm is used to optimize the adjustment recommendation content according to the corresponding optimization features obtained, so as to obtain the optimized recommendation content;

[0113] When using, combine the contents in steps 501 and 502:

[0114] When smart underwear fails to achieve the expected results, the reference effect can be improved by optimizing and adjusting the recommended content, and the recommended content can be kept in a state of continuous dynamic adjustment, making the recommended content more compatible with patients with chronic diseases.

[0115] The Analytic Hierarchy Process (AHP) is a decision-making method that decomposes decision-related elements into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. It is particularly suitable for dealing with target systems with hierarchical and staggered evaluation indicators, and when the target value is difficult to describe quantitatively, the AHP is an effective decision-making tool.

[0116] The core of the hierarchical analysis method is to decompose the decision problem into multiple levels to form a hierarchical structure, which usually includes the target level, the criterion level, the sub-criterion level and the solution level. By solving the eigenvector of the judgment matrix, the priority weight of each element of each level to an element of the previous level is obtained, and the final weight of each alternative solution to the total goal is recursively merged by the weighted sum method, so as to find the optimal solution.

[0117] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0118] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0123] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code. The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A device for collecting wearing feedback data of smart underwear, characterized in that: include, The data collection unit collects the sleep data of patients with chronic diseases, uses the trained sleep recognition model to identify the current sleep pattern of patients with chronic diseases, performs similarity analysis on it and the reference sleep pattern, and generates a difference degree from the acquired analysis data. , if the difference Exceed expectations and issue deep work orders; The data processing unit starts the sensor network on the smart underwear, sets the data collection frequency for the sensor, collects and preprocesses the various physical indicators of patients with chronic diseases, analyzes and obtains the status indicators of the preprocessed data, and issues a status analysis instruction; wherein, the constraints on the sensor data collection frequency are as follows: ； in, n is the number of data collections within the data collection cycle, It is i Data collection to the first j The time interval between data collections, is the average value of the time interval, is the energy consumption during the sensor data collection cycle; Abnormal data identification unit, generates the sending state coefficient from the data state indicator set , according to the sending state coefficient Switch the data transmission status of the smart underwear; after receiving the health monitoring data of patients with chronic diseases, identify the abnormal data, and issue an abnormal reminder if there is abnormal data; after receiving the status analysis instruction, the data status indicator set is used to construct the transmission status coefficient of the preprocessed data , where the transmission time and response time Perform linear normalization as follows: ； Weight coefficient, , ; Working state switching unit, if the state coefficient is sent If the state threshold is exceeded, the pre-processed data is sent from the Bluetooth module to the data receiving end. Otherwise, it is sent from the Wi-Fi module. The health monitoring data received by the data receiving end is used as input, and the trained abnormal data recognition model is used to identify abnormal data. The content recommendation matching unit builds the aggregation of abnormal alerts based on the time nodes of abnormal alerts received within the content recommendation cycle. If the degree of aggregation exceeds expectations, identify the health trends of chronic disease patients. If the health trends do not match expectations, use collaborative filtering algorithms to output recommended content for chronic disease patients. Use the acquired health monitoring data as input and use the trained health trend recognition model to identify the health trends of chronic disease patients. If the identified health trends do not match expectations, generate chronic disease health reports for chronic disease patients based on the acquired health monitoring data. After performing feature engineering on the content recorded in the chronic disease health report, the corresponding health features are obtained; a recommended content library is pre-built, and the health features of chronic disease patients are used as input, and the recommended content library outputs recommended content for chronic disease patients; Improvement analysis unit, obtains improvement degree by analyzing sleep data within the feedback cycle If the improvement If the result exceeds the expectation, the working state of the smart underwear is switched. If the result does not exceed the expectation, the multi-objective optimization algorithm is used to optimize the adjustment recommendation content, and the optimized recommendation content is output to the chronic disease patients in the next feedback cycle. If the improvement degree If the improvement threshold is exceeded, the smart underwear will be switched back to a shallow working state; if no improvement is achieved, a content optimization instruction will be issued to the outside world; After the chronic disease patient enters a sleeping state, the sleep quality and physiological parameters of the chronic disease patient are collected and monitored, and the collected data are summarized as sleep data to obtain a sleep data set of the chronic disease patient; the sleep data of the chronic disease patient is used as input, and the identified sleep patterns are summarized to generate a sleep pattern set; Add description information to the sleep mode, perform similarity analysis between the corresponding description information and the description information of the reference sleep mode, and generate the difference degree by obtaining the corresponding similarity data , if the difference When the difference threshold is exceeded, the smart underwear will be switched from the shallow working state to the deep working state; ； in, n is the number of descriptive information, It is i The similarity between the description information and the description information of the reference sleep mode, is the average similarity; weight coefficient: , ,and ; A sensor network is installed on the smart underwear. After receiving the deep work instruction, the dormant sensors are started and the data collection frequency is set for the sensors. The various physical indicators of chronic disease patients are collected at the collection nodes that meet the constraints, and the collected data are summarized to generate a health monitoring data set. Preprocess the monitoring data in the health monitoring data set, analyze and obtain the status indicators of the preprocessed data, and generate a data status indicator set after aggregation.

2. The device for collecting wearing feedback data of smart underwear according to claim 1, characterized in that: Record the time nodes when abnormal alerts are received during the content recommendation cycle, and build the aggregation of abnormal alerts , if the aggregation When the concentration threshold is exceeded, a trend identification instruction is issued to the outside, where: ； Where: , The value is 2.718, the weight coefficient, , ,and ; is the radius of the content recommendation cycle; The weighted value of the i-th abnormal reminder, is the time interval from the i-th abnormal reminder to the j-th abnormal reminder, and n is the number of time nodes.

3. The device for collecting wearing feedback data of smart underwear according to claim 2, characterized in that: After obtaining the sleep data in each sub-period, the difference is generated , several differences Arrange along the time axis and analyze to get the improvement ; Analyze and obtain improvement The way is as follows: ； in, is the difference within the i-th sub-period, is the corresponding mean, is its qualified standard value; For the i Improve the median value, is its mean value, , is the number of sub-periods, and the weight coefficient is: , ,and .

4. The device for collecting wearing feedback data of smart underwear according to claim 3, characterized in that: After receiving the content optimization instruction, the feedback data of chronic disease patients after wearing smart underwear is collected, and features are extracted from the feedback data. Based on the corresponding optimization features obtained, a multi-objective optimization algorithm is used to optimize the adjusted recommended content to obtain the optimized recommended content.

Images