Cloud platform-based chronic disease health data management system

Abstract

The application provides a chronic disease health data management system based on a cloud platform, relates to the technical field of health data management, and comprises a data collection module, a data recognition module, an incentive execution module, an association acquisition module and a storage module.The data collection module is used for collecting health data of a target user based on a cloud platform; the data recognition module is used for recognizing health data through a chronic disease type to obtain a target chronic disease type; the incentive execution module is used for performing incentive execution on the target chronic disease type according to an incentive mechanism to obtain an optimized chronic disease type; the association acquisition module is used for acquiring an associated disease type of the optimized chronic disease type; and the storage module is used for storing the optimized chronic disease type and the associated disease type and giving an early warning. The application can solve the technical problem that potential comorbidity risks are not identified in time due to the lack of dynamic association recognition capability between disease types in the prior art, achieve the technical target of association recognition and dynamic monitoring between multiple disease types, and achieve the technical effects of improving the accuracy of early screening of chronic diseases, optimizing intervention schemes and intelligent early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of health data management, and particularly relates to a chronic disease health data management system based on a cloud platform. BACKGROUND

[0002] With the aggravation of population aging and the change of lifestyle, the prevalence of chronic diseases continues to rise, which has become an important problem affecting the health level of the whole people and the allocation of medical resources. In response to this challenge, health data management technology relying on big data and the Internet of Things has gradually developed, and especially the chronic disease management system based on the cloud platform has played a preliminary role in health monitoring, data collection, remote follow-up, etc.

[0003] At present, the existing technology usually collects the basic health parameters such as blood pressure, blood sugar, heart rate and weight of the user through wearable devices, mobile terminals or medical devices, uploads them to the cloud platform for storage and analysis, and preliminarily identifies abnormal data according to the set threshold. However, these technologies still have many defects, which limit their in-depth application in early screening, intervention and personalized management of chronic diseases. The current system generally lacks the ability of dynamic correlation identification between diseases, and only monitors and processes a single chronic disease, without considering the comorbidity relationship and development path between chronic diseases. For example, diabetic patients often have high blood pressure, hyperlipidemia or diabetic nephropathy. If the system cannot timely identify these potential associated diseases, it is difficult to give an effective early warning and delay the intervention opportunity.

[0004] In summary, the existing technology has the technical problem that due to the lack of dynamic correlation identification between diseases, potential comorbidity risks cannot be timely identified, which further affects the early screening, precise intervention and personalized management of chronic diseases. SUMMARY

[0005] The purpose of the present application is to provide a chronic disease health data management system based on a cloud platform, to solve the technical problem in the prior art that due to the lack of dynamic correlation identification between diseases, potential comorbidity risks cannot be timely identified, which further affects the early screening, precise intervention and personalized management of chronic diseases.

[0006] In view of the above problems, the present application provides a chronic disease health data management system based on a cloud platform, comprising: a data collection module for collecting health data of a target user based on a cloud platform; a data identification module for identifying the health data through a chronic disease type to obtain a target chronic disease type; an incentive execution module for performing incentive execution on the target chronic disease type according to an incentive mechanism to obtain an optimized chronic disease type; an association acquisition module for acquiring associated diseases of the optimized chronic disease type; and a storage module for storing and warning according to the optimized chronic disease type and the associated diseases.

[0007] The technical solutions provided in the application have at least the following technical effects or advantages: by achieving the technical target of multi-disease association recognition and dynamic monitoring, the technical effects of improving the accuracy of early screening of chronic diseases, optimizing intervention schemes, realizing personalized health management and intelligent early warning are achieved.

[0008] The above description is only a summary of the technical solutions of the application. In order to more clearly understand the technical means of the application, the specific embodiments of the application can be implemented according to the content of the description, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the provided drawings.

[0010] Figure 1 The structural schematic diagram of the chronic disease health data management system based on the cloud platform of the application;

[0011] Figure 2 The structural schematic diagram of the data recognition module in the chronic disease health data management system based on the cloud platform of the application;

[0012] Figure 3 The schematic logic diagram of the chronic disease health data management system based on the cloud platform of the application.

[0013] Explanation of reference signs: data acquisition module 1, data recognition module 2, incentive execution module 3, association acquisition module 4, storage module 5. DETAILED DESCRIPTION

[0014] The application provides a chronic disease health data management system based on a cloud platform, which solves the technical problem in the prior art that due to the lack of dynamic association recognition ability between diseases, potential comorbidity risks cannot be identified in time, which further affects the early screening, precise intervention and personalized management of chronic diseases.

[0015] Below, the technical solutions in the present application will be described clearly and completely with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application. In addition, it should be noted that, for the convenience of description, only parts related to the present application are shown in the drawings, not all.

[0016] Please refer to the accompanying Figure 1 The present application provides a chronic disease health data management system based on a cloud platform, specifically comprising:

[0017] A data collection module 1 is configured to collect health data of a target user based on a cloud platform.

[0018] Specifically, collecting health data of a target user based on a cloud platform means that various health information of a user is collected through devices and then uploaded to a remote server for unified storage, processing and management by the system, by means of internet technology and cloud computing architecture. The cloud platform is a platform that provides services through a remote data center connected by a network without relying on local hardware resources of the user, and has high reliability, high concurrent processing capability and data security guarantee, and can receive and update health data of thousands of users in real time, support remote access and cross-institutional sharing.

[0019] The target user refers to an object that needs to be monitored and served, including chronic disease patients, high-risk groups or health management participants. For example, a hypertensive patient is a target user, whose identity is confirmed by personal information, medical records, medical behavior, etc. bound by the system. When collecting data, the system only records the long-term tracking of the identified target user for personalized analysis and intervention.

[0020] Health data refers to information closely related to the physiological state of the human body, disease risk and living habits, such as body temperature, blood pressure, heart rate, blood glucose, sleep duration, exercise steps, medication records, etc. The data sources can include smart watches, blood pressure meters, blood glucose meters, hospital information systems, community health stations, home self-testing devices, etc. For example, the fasting blood glucose value, postprandial blood glucose value and sleep quality score uploaded by a diabetic patient through a blood glucose meter every day are important health data content.

[0021] The data collection module 1 connects with the user's wearable device (such as smart bracelet, sphygmomanometer, blood glucose meter, etc.) or mobile terminal through API interface, collects the user's multi-dimensional health parameter data in real time, including blood pressure, blood glucose, heart rate, body temperature, weight, sleep duration, activity steps, etc., and uploads the data to the cloud database in a unified format (such as JSON). The data is transmitted by using HTTPS protocol to ensure data security.

[0022] The data recognition module 2 is used to identify the health data by chronic disease type, and obtain the target chronic disease type.

[0023] Specifically, after collecting the user's health data, the preset recognition model is used to analyze and match the health data, so as to determine the chronic disease type that the user may currently have or have a risk of. The recognition process depends on multiple physiological indicators, such as blood pressure, blood glucose, blood lipid, weight, heart rate, sleep quality, etc., and may also combine the user's living habits, past medical history and family genetic conditions for comprehensive judgment.

[0024] The data recognition module 2 establishes the mapping relationship between the health indicators and the disease risk by configuring the chronic disease type grid model. For example, for common chronic diseases such as diabetes and hypertension, the basic screening indicators such as fasting blood glucose greater than 7 mmol / L, systolic blood pressure greater than 140 mmHg, etc. are set to form an index threshold comparison table. Rule matching and machine learning algorithm (such as decision tree, XGBoost) are combined for chronic disease recognition, automatically determine the disease type that the target user may have, and output the disease type label and matching confidence.

[0025] The incentive execution module 3 is used to perform incentive execution on the target chronic disease type according to the incentive mechanism, and obtain the optimized chronic disease type.

[0026] Specifically, after identifying the chronic disease type that the user currently has, the preset incentive mechanism is combined to guide, intervene or adjust the health behavior, index control and lifestyle related to the disease type, observe the feedback results after intervention, and judge whether the chronic disease type changes from abnormal or critical state to stable or improved state based on the feedback effect, so as to be defined as the optimized chronic disease type. The incentive mechanism can include personalized health advice, behavior feedback, point reward, lifestyle intervention, etc. For example, it is monitored that a user has abnormal blood lipid performance, and the target chronic disease type of hyperlipidemia is preliminarily identified, and then incentive actions are performed, such as recommending light diet, increasing aerobic exercise, controlling oil intake, etc., and at the same time, the user's behavior changes are recorded through daily health punch card or smart device.

[0027] The incentive mechanism includes intervention actions of user health behaviors and evaluation mechanism of corresponding feedback effects. The intervention actions are, for example, pushing dietary suggestions, exercise plans or medication reminders; the feedback effects include user index improvement conditions (such as blood glucose decrease, weight loss, etc.). Based on the positive feedback effects, the intervention strategies are automatically adjusted by using the reinforcement learning idea, so as to improve the individualization level and effectiveness of health intervention. For example, when the blood glucose of a certain user decreases by 2 mmol / L after 3 days of persistent execution of dietary control suggestions, it is judged by the system that the intervention is a positive feedback, and the intervention action parameters (such as adjusting the suggestion frequency, content individualization) are optimized.

[0028] The association acquisition module 4 is configured to acquire the associated disease of the optimized chronic disease.

[0029] Specifically, after determining that a certain chronic disease has achieved state optimization through the incentive mechanism, other diseases highly related to the optimized disease are further screened in the chronic disease grid or database based on information such as physiological mechanisms, clinical associations and index coincidence degrees between diseases, and are identified as associated diseases, so as to be monitored and warned in advance.

[0030] The association acquisition module 4 models the potential co-occurrence relationship between different diseases based on a disease comorbidity network graph. Using historical health data of a user group, the association degree between diseases is calculated through statistical analysis (such as Pearson correlation coefficient, point two column correlation coefficient). Taking the identified target disease as a seed node, association propagation is performed to screen other diseases with an association degree exceeding a set threshold (such as 0.6), and the potential comorbidity risk of the target user is output. For example, after identifying diabetes, the system can associate two associated diseases, hypertension and diabetic retinopathy, and perform data monitoring in advance.

[0031] The storage module 5 is configured to store and warn according to the optimized chronic disease and the associated disease.

[0032] Specifically, after acquiring the chronic disease and identifying the potential associated disease, the health data and labeled information of the user are stored in the cloud platform, and a dynamic monitoring and intelligent warning mechanism is established based thereon, so as to realize early intervention and risk control of disease development.

[0033] MongoDB is used to store unstructured health data through the cloud platform, and Elasticsearch is used to realize efficient labeled indexing. Through the system, each piece of data is bound with user portrait, disease label and timestamp information when it is stored, forming a traceable and searchable health data set. The label data is used to drive the subsequent identification, warning and feedback cycle.

[0034] Further, as Figure 2As shown, the present application further comprises: a grid configuration unit 21 for configuring a chronic disease grid associated with the health data; and a grid matching unit 22 for matching the chronic disease grid based on the health data to obtain the target chronic disease of the target grid.

[0035] Specifically, configuring a chronic disease grid associated with health data means that the system establishes a regular mapping relationship between different chronic diseases and corresponding health indicators. Each grid cell represents a chronic disease and is connected to a specific combination of health data indicators. The chronic disease grid is a disease characteristic classification system for structuring and classifying complex health information. For example, the hypertension grid may be associated with indicators such as systolic pressure, diastolic pressure, and heart rate, while the diabetes grid may include fasting blood glucose, glycosylated hemoglobin, and insulin levels. Each grid sets its standard value interval according to the data parameters required by different disease types.

[0036] Based on the health data matching the chronic disease grid, the target chronic disease of the target grid is obtained by comparing the real-time health data uploaded by the user with the chronic disease grid item by item, checking which grid interval the user's health indicators fall into, and determining the specific chronic disease the user may have.

[0037] Further, the present application further comprises: a basic indicator setting unit for setting basic screening indicators based on the chronic disease grid; a risk identification unit for identifying the health data through the basic risk threshold of the basic screening indicators to determine a target grid group; an extended indicator setting unit for setting extended screening indicators based on the chronic disease group of the target grid group; a data group obtaining unit for combining the basic collection package obtained by screening the health data through the basic screening indicators and the extended collection package obtained through the extended screening indicators to obtain a target disease data group; and a disease matching unit for performing disease matching of the target disease data group according to the target user portrait determined based on the chronic disease record of the target user to obtain the target chronic disease.

[0038] Specifically, setting the basic screening indicators based on the chronic disease grid means that based on the chronic disease grid, a group of health indicators with strong representativeness, convenient collection, and high sensitivity are selected as the basis for the first-stage screening according to the core judgment elements of each disease. For example, in the hypertension disease grid, the basic screening indicators may include systolic pressure, diastolic pressure, and heart rate, while in the diabetes grid, they may be fasting blood glucose and body mass index. The role of the basic screening indicators is to quickly lock the health risk direction and facilitate preliminary grouping.

[0039] The health data is identified by a basic index risk threshold of a basic screening index, and the target grid group is determined by setting a risk limit for each basic index. When the health data of a user exceeds these threshold values, it is determined that the user is in a potential risk state, and the user is classified into a corresponding chronic disease grid set, i.e., a target grid group. For example, if the systolic pressure of a user is 165 mmHg and the fasting blood glucose is 7.5 mmol / L, the user will be classified into both the hypertension and diabetes grids, and form a target grid group.

[0040] The extended screening index is set according to the disease types that the target user may be involved in, and the health index that is more valuable for diagnosis is further configured. For example, for the diabetes grid, the extended index may include glycosylated hemoglobin and insulin resistance index; for the cardiovascular disease type, the extended index may include blood lipids, homocysteine, and the like, which assist medical personnel in disease judgment.

[0041] The target disease data group is obtained by combining the basic collection package obtained by screening the health data through the basic screening index and the extended collection package obtained by screening the health data through the extended screening index, i.e., all the health data of the user in the two screening stages are integrated to form a complete data package containing basic information and detailed indexes, which is used for disease identification and analysis. For example, the basic collection package of a user contains blood pressure and blood glucose, and the extended collection package contains blood lipids and electrocardiogram data. After integration, a target disease data group for hypertension and cardiovascular disease is obtained.

[0042] The target chronic disease type is obtained by matching the target disease data group with the target user portrait determined according to the chronic disease record of the target user, i.e., the personalized user portrait model is constructed by combining the past chronic disease history, family history, age, gender, and lifestyle of the user, and then the model is matched with the currently collected disease data group, so as to determine the chronic disease type that best matches the characteristics of the user. For example, a middle-aged woman who has a history of gestational diabetes will be identified as a diabetes patient even if her health data just exceeds the diabetes judgment threshold, because the high-risk factors in her portrait are weighted.

[0043] Further, the application also includes: an association degree acquisition unit configured to acquire a disease association degree according to the chronic disease type grid; and a disease screening unit configured to perform association disease screening based on the disease association degree in the target grid group according to the optimized chronic disease type, to obtain the association disease.

[0044] Specifically, after the grid structure of chronic disease types is constructed, the correlation between each disease type is calculated and labeled. The chronic disease type grid is a structure that divides different chronic disease types according to their characteristic indicators, occurrence mechanisms and development paths into multiple grid areas, and the grids are associated through common physiological indicators, concurrent mechanisms or clinical statistical data. The disease type correlation degree reflects the strength of the mutual relationship between two or more chronic disease types in terms of onset time, symptom intersection, physiological mechanism, etc. For example, there is a high correlation between hypertension and chronic kidney disease, partly because long-term hypertension can damage the glomerulus and thus cause chronic kidney dysfunction.

[0045] After a certain chronic disease type has been identified as optimized, in its target grid group, based on the correlation between it and other disease types, potential related disease types are screened. The target grid group refers to a group of grids that are spatially adjacent or have similar indicators to the grid where the optimized disease type is located. Risk assessment will be conducted on the disease types in the grid, especially focusing on the parts with high correlation with the current optimized disease type, to determine whether the user may have or develop the disease type in the future. For example, if a diabetic patient's health indicators improve significantly during management, it is determined that his diabetes has been optimized, and then based on the high correlation between diabetes and high blood fat, high blood pressure, etc., the changes in blood fat level and blood pressure in the adjacent grid are screened to confirm whether early intervention or monitoring of related disease types is needed.

[0046] Further, the present application also includes: a basic counting unit for counting the number of exclusive basic indicators of the grid in the target grid group according to the basic screening indicators of the target grid group; a grid pair acquisition unit for acquiring a target grid pair through the number of exclusive basic indicators, wherein the number of exclusive indicators between the target grid pair is the least; a grid pair connection unit for connecting the target grid pair based on the same disease type in the target grid pair to obtain a target grid chain; a first judgment unit for obtaining the related disease type according to the target grid chain if the optimized chronic disease type exists in the target grid chain.

[0047] Specifically, in the target grid group, the basic screening indicators associated with each grid are counted, and it is identified which indicators are unique to the grid. The basic screening indicators refer to key physiological parameters used to initially identify chronic disease types, such as blood pressure, fasting blood glucose, body mass index, heart rate, etc. The exclusive basic indicators refer to indicators that are only unique to a grid and are not shared with other grids among the screening parameters used for disease type identification. The number of these exclusive indicators for each grid is counted in order to further analyze the uniqueness and degree of overlap between grids.

[0048] From the target grid group, select any two grids to form a grid pair, compare the exclusivity of the basic indicators between them, and select the grid pair with the least number of exclusive basic indicators. The selected target grid pair has a high degree of overlap in basic indicators, indicating that they have similarities in the performance of chronic diseases, which facilitates subsequent aggregation analysis. For example, two grids represent high blood lipids and diabetes, respectively, and they both use body mass index and fasting blood glucose as basic indicators, so the number of exclusive indicators between them is small, and they can form a target grid pair.

[0049] In the grid pair with high similarity, further filter the grids related to the same chronic disease or related disease, and sequentially connect them through commonality to construct a continuous grid sequence with potential disease evolution relationship, that is, the target grid chain.

[0050] Once the optimized chronic disease has appeared in the target grid chain, take the disease as the starting point to analyze the evolution relationship and clinical co-occurrence frequency between other diseases in the entire grid chain, so as to identify its potential associated diseases.

[0051] Further, the application also includes: an extension counting unit for counting the number of exclusive extension indicators of the grids in the target grid chain according to the extension screening indicators; a deletion and supplement unit for supplementing the target grid chain with the target grid group through the number of exclusive extension indicators to obtain a target grid cluster; a second judgment unit for obtaining the associated diseases from the target grid cluster if the optimized chronic disease exists in the target grid cluster.

[0052] Specifically, on the basis of the constructed target grid chain, further statistics of the extension screening indicators involved in each grid are counted, and it is judged which of the extension indicators are unique to the grid. The extension screening indicators are a supplement to the basic screening indicators, such as ECG abnormalities, liver and kidney function indicators, microalbuminuria, and inflammatory factor levels, which are used to identify chronic diseases with high complexity and intertwined pathogenesis. The number of exclusive extension indicators reflects the uniqueness of each grid in the extension indicator dimension, which helps to identify whether there is redundancy, omission or potential contact between grids.

[0053] According to the degree of overlap of the extension indicators, the original target grid chain is reduced or supplemented. If the exclusive extension indicators of some grids are highly repeated with other grids, they may represent information redundancy and can be deleted; on the contrary, if the indicators of some potential associated diseases are not included in the grid chain, the corresponding grid can be supplemented through the extension indicators to form a new grid cluster. The target grid cluster is a more compact, richer indicator, and more comprehensive chronic disease identification unit.

[0054] In the grid cluster, if a chronic disease type that has been determined to be in an optimization state is included, the optimization disease type is taken as a core to analyze disease type distribution and index evolution path in the entire grid cluster, so as to identify associated disease types that are most likely to have a linkage or transformation relationship with the optimization disease type.

[0055] Further, the application further comprises: an incentive mechanism setting unit configured to set an incentive mechanism, wherein the incentive mechanism comprises an association relationship between an incentive action and an incentive effect; an incentive execution unit configured to execute the incentive action on the target chronic disease type according to the association relationship to obtain a target incentive effect; an incentive adjustment unit configured to adjust the incentive action to obtain a target incentive action in a manner of guiding a positive feedback effect of the target incentive effect; and an optimization result obtaining unit configured to obtain the optimized chronic disease type according to the target incentive effect and the target incentive action.

[0056] Specifically, the incentive mechanism is set, wherein the association relationship between the incentive action and the incentive effect in the incentive mechanism refers to introducing a positive feedback strategy in the chronic disease management process to guide the user to continuously improve the health behavior. The incentive action in the incentive mechanism can be an intervention means adopted by the system, such as a push exercise reminder, a diet guide, a reward point, a doctor remote follow-up, and the like, and the incentive effect refers to a change result in the user health data or behavior performance after the actions are implemented, for example, blood glucose reduction, weight loss, activity amount increase, and the like. By analyzing the statistical or logical relationship between the incentive action and the incentive effect, an effect-oriented association model is established.

[0057] The target incentive effect obtained by executing the incentive action on the target chronic disease type according to the association relationship refers to that after a target chronic disease type existing in a certain user is identified, a suitable intervention measure is selected and implemented according to a response model of the disease type in the past. For example, for a user with mild hypertension, a daily exercise prompt and a salt-restricted diet suggestion can be arranged. After a period of tracking, if the systolic pressure is reduced from 150 mmHg to 135 mmHg, a positive target incentive effect is obtained.

[0058] The target incentive action obtained by adjusting the incentive action in a manner of guiding a positive feedback effect of the target incentive effect refers to that when it is found that a certain incentive action can bring good health improvement, the original action is strengthened, personalized or optimized to make it more in line with the actual needs of the user. For example, the original daily walking target of 5000 steps is automatically adjusted to 7000 steps, and a diet control suggestion is matched to form a more reasonable target incentive action.

[0059] According to the target incentive effect and the target incentive action, the optimized chronic disease type is obtained by re-evaluating the current disease type state of the user and judging the development trend of the chronic disease type. If the user realizes index stability, symptom relief or recurrence reduction through the incentive in a period of time, the original target chronic disease type state can be updated to an optimized state, such as updating from moderate hypertension to stable period hypertension, or reducing from prediabetes to good blood sugar control state. Table 1 is the record of the last incentive execution.

[0060] Table 1: Record of the last incentive execution

[0061]

[0062] Further, the application further comprises: a third judgment unit, configured to obtain a positive feedback state if the target incentive effect is a positive feedback effect and the target incentive action is a positive feedback action; and a screening unit, configured to screen in the chronic disease type grid according to the positive feedback state to obtain the optimized chronic disease type.

[0063] Specifically, if the target incentive effect is a positive feedback effect and the target incentive action is a positive feedback action, obtaining a positive feedback state means that in the process of chronic disease management, the intervention measures and the corresponding health changes are continuously tracked and analyzed. If the user actively responds to the incentive action in a period of time, and the health data indeed shows an improvement trend, this situation is identified as a positive feedback state. For example, a user's blood pressure gradually decreases and is stable for a long time under the incentive of low-salt diet and daily walking, and the exercise frequency increases, which indicates that there is a direct positive relationship between the incentive action and health improvement, so it is judged that the intervention path is effective, forming a positive feedback state.

[0064] According to the positive feedback state, the optimized chronic disease type is obtained by screening in the chronic disease type grid, which means that when a user is identified in a positive feedback state, the distribution position of the health data of the user in the chronic disease type grid is traced back, and the disease type risk level and state trend are recalculated. The updated health indicators are remapped to the chronic disease type grid, and combined with the current incentive response of the user, it is confirmed whether the disease has been substantially improved. If the health indicators have deviated far from the danger threshold and the trend is continuously good, the original target chronic disease type will be marked as an optimized chronic disease type. For example, a user originally belongs to the high blood sugar grid, and after intervention, his blood sugar has been in the normal range for more than half a year, so he can be adjusted from a high-risk chronic disease state to an optimized state and included in preventive monitoring rather than key management.

[0065] Further, the application further comprises: a path establishment unit, configured to establish an optimized path in the chronic disease type grid according to the positive feedback state; and a step number determination unit, configured to determine a step number maximum grid of the number of positive feedback steps in the optimized path to obtain the optimized chronic disease type of the optimized disease type grid.

[0066] Specifically, establishing an optimization path in the chronic disease grid according to the positive feedback state means that after identifying that a certain user is in a positive feedback state, an optimization path for continuously improving the health status is designed based on the state, including the health behavior steps gradually taken by the user in the chronic disease management process, and dynamically adjusting according to the effects of the steps. For example, the optimization path can start with increasing daily exercise, gradually advancing to improving dietary structure, reducing stress management, regular monitoring, and other strategies.

[0067] Determining the step maximum grid of the number of positive feedback steps in the optimization path, obtaining the optimization chronic disease grid of the optimization disease grid, means that in the entire optimization path, the chronic disease grid corresponding to each incentive step is quantitatively analyzed to determine which grid area has the most significant positive feedback effect, that is, the step maximum grid. The step maximum grid refers to the chronic disease grid interval that has the greatest impact and the most obvious effect among the implemented health behaviors and intervention measures. For example, if a user's blood glucose value in the diabetes grid has significantly decreased by increasing the number of exercise steps, but the effect in the hypertension grid is relatively flat, then the diabetes grid will be considered as the step maximum grid.

[0068] According to the feedback effect of the step maximum grid, the user's chronic disease management strategy will be re-evaluated, and the optimized chronic disease will be determined. For example, according to the significant improvement of the diabetes grid, the user's disease will be optimized to stable period diabetes, so that the health status will be transferred to a more relaxed monitoring mode instead of intensive management of high-risk diseases.

[0069] Further, the present application also includes: a storage record obtaining unit for storing the health data, the optimization chronic disease, and the associated disease in the cloud platform to obtain a storage record; a tagging unit for tagging the storage record in combination with the chronic disease record, the target user portrait, and the storage record to obtain tagged health data; a coordination unit for realizing dynamic coordination of chronic disease identification, incentive feedback, associated disease acquisition, and early warning according to the tagged health data.

[0070] Specifically, a unified data storage module is established in the cloud platform to store the user's health monitoring data and the identified chronic disease in the optimization state and the possible associated disease in a certain data structure in an orderly manner. Associated storage not only includes basic data such as blood pressure, blood glucose, weight, etc., but also stores structured information such as label results generated by the analysis model, evaluation scores, disease relationship networks, etc.

[0071] The user's chronic disease diagnosis history, user profile (including gender, age, genetic history, lifestyle, etc.), and real-time storage records are used to automatically generate labels for each piece of health data. The purpose of the labeling process is to quickly search, classify, and analyze massive amounts of data. For example, a blood glucose data will be labeled as diagnosed with diabetes, over 60 years old, family history, high-sugar diet, etc. when stored. The combination of labels forms labeled health data, making subsequent screening and identification more accurate and personalized.

[0072] Through real-time monitoring and analysis of labeled health data, automatic collaboration between modules is achieved in identifying chronic diseases, judging their trends, selecting appropriate incentive mechanisms, obtaining potential associated diseases, and issuing risk warnings. The sensitivity of the identification model and the warning threshold are dynamically adjusted based on the user's current labeled data flow. For example, Figure 3 A diagrammatical illustration of the cloud-based chronic disease health data management system.

[0073] According to the actual specific example, first, the user background is acquired, including user A, male, 45 years old, weight 85 kg, with a family history of diabetes, no clear past medical history, and participates in chronic disease screening and management for the first time through the health management platform. Then, health data collection and uploading are performed. Among them, user A uses a smart watch and a Bluetooth blood glucose meter to collect the following indicators multiple times within a week and synchronizes them to the cloud platform, including blood glucose (fasting 7.8 mmol / L, postprandial 10.5 mmol / L), diastolic blood pressure (95 mmHg), systolic blood pressure (142 mmHg), heart rate (88 beats per minute), BMI (28.3 kg / m2), and through the cloud platform, the data is uploaded once a day according to the set data uploading period. Further, chronic disease type grid matching and preliminary screening identification are performed. Among them, according to the built-in chronic disease type grid, the collected data is matched and judged to obtain blood glucose level exceeding the basic threshold of diabetes, blood pressure exceeding the screening standard of hypertension, and two preliminary screening diseases (diabetes and hypertension) are matched and output. Further, the incentive mechanism is executed. Among them, an intervention action plan is automatically generated for user A, including a diet control plan (low-sugar diet guide, report diet diary once a week), exercise incentive (three times of fast walking per week, each for thirty minutes), and intervention period (two weeks), and real-time recording of exercise data and user feedback. Further, positive feedback identification and action adjustment are performed. Among them, after two weeks, the data is re-collected, and it is found that the fasting blood glucose is reduced to 6.4 mmol / L, and the diastolic blood pressure is reduced to 88 mmHg, and then it is determined that the positive feedback effect is obtained, and the intervention plan is further adjusted through the cloud platform to increase the high-intensity exercise suggestion and push the reward point mechanism (such as completing the plan to exchange health product discount). Further, optimized chronic disease type and comorbidity identification are performed. Among them, diabetes is identified as an "optimized chronic disease type", and based on the disease comorbidity atlas, possible comorbidity diseases are screened out in the diabetes related grid, including diabetic nephropathy and hyperlipidemia, and then the user is prompted to perform extended screening index detection, including creatinine, urine protein, blood lipids, etc. Further, data labeling and storage are performed. Among them, the process data of user A is structured and stored through the cloud platform, and labels are generated, including label one (middle-aged male), label two (diabetes preliminary screening to positive feedback improvement), label three (hypertension early warning), and label four (active response to incentive mechanism), and the label data is used for subsequent chronic disease early warning, recommended strategy optimization, and group model training. Finally, collaborative management and intelligent early warning are performed. Among them, if abnormal blood lipids or elevated urine protein are found in subsequent detection, the system automatically links the early warning module to send a message to remind the user to review or seek medical treatment, forming a dynamic collaborative management closed loop.

[0074] In summary, the chronic disease health data management system based on the cloud platform has the following technical effects: through the technical targets of realizing the multi-disease association recognition and dynamic monitoring, the technical effects of improving the early screening accuracy of chronic diseases, optimizing the intervention scheme, realizing the personalized health management and intelligent early warning are achieved.

[0075] The above description of disclosed embodiments enables one skilled in the art to make or use the application. Numerous modifications to these embodiments would be apparent to those of skill in the art, and the general principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0076] Obviously, various modifications and changes can be made to the present application by those skilled in the art without departing from the spirit and scope of the present application. Thus, it is intended that the present application cover the modifications and changes as long as they come within the scope of the application, and its equivalents.

Claims

1. A cloud platform based chronic health data management system, characterized in that, The method comprises the steps of: a data collection module for collecting health data of a target user based on a cloud platform; a data recognition module for identifying the health data through chronic disease types to obtain a target chronic disease type; an incentive execution module for performing incentive execution on the target chronic disease type according to an incentive mechanism to obtain an optimized chronic disease type; an association acquisition module for acquiring an associated disease type of the optimized chronic disease type; a storage module for storing and warning according to the optimized chronic disease type and the associated disease type; The data recognition module comprises: a grid configuration unit for configuring a chronic disease type grid associated with the health data; a grid matching unit for matching the chronic disease type grid based on the health data to obtain the target chronic disease type of a target grid; The grid matching unit comprises: a basic index setting unit for setting a basic screening index based on the chronic disease type grid; a risk identification unit for identifying the health data through a basic index risk threshold of the basic screening index to determine a target grid group; an extended index setting unit for setting an extended screening index based on a chronic disease type group of the target grid group; a data group obtaining unit for combining a basic collection package obtained by screening the health data through the basic screening index and an extended collection package obtained through the extended screening index to obtain a target disease type data group; a disease type matching unit for matching the target disease type data group according to a target user portrait determined based on a chronic disease record of the target user to obtain the target chronic disease type; The association acquisition module comprises: an association degree acquisition unit for acquiring a disease type association degree based on the chronic disease type grid; a disease type screening unit for screening an associated disease type based on the disease type association degree in a target grid group according to the optimized chronic disease type to obtain the associated disease type; The disease type screening unit comprises: a basic counting unit for counting a number of exclusive basic indexes of a grid in the target grid group according to a basic screening index of the target grid group; a grid pair acquisition unit for acquiring a target grid pair through the number of exclusive basic indexes, wherein the number of exclusive indexes between the target grid pair is the least; a grid pair connection unit for connecting the target grid pair based on the same disease type in the target grid pair to obtain a target grid chain; a first judgment unit for obtaining the associated disease type according to the target grid chain if the optimized chronic disease type exists in the target grid chain; The first judgment unit comprises: an extended counting unit for counting a number of exclusive extended indexes of a grid in the target grid chain according to an extended screening index; a deletion and supplement unit for deleting and supplementing the target grid chain in the target grid group through the number of exclusive extended indexes to obtain a target grid cluster; a second judgment unit for obtaining the associated disease type according to the target grid cluster if the optimized chronic disease type exists in the target grid cluster.

2. The cloud platform based chronic health data management system as claimed in claim 1, wherein, The incentive execution module comprises: an incentive mechanism setting unit for setting an incentive mechanism, wherein the incentive mechanism comprises an associated relationship between an incentive action and an incentive effect; The incentive execution unit is configured to perform an incentive action on the target chronic disease type according to the association relationship, and obtain a target incentive effect; The incentive adjustment unit is configured to adjust the incentive action to obtain a target incentive action, with a positive feedback effect of the target incentive effect as a guide; The optimization result obtaining unit is configured to obtain the optimized chronic disease type according to the target incentive effect and the target incentive action. 3.The cloud platform-based chronic disease health data management system of claim 2, wherein, The optimization result obtaining unit comprises: The third judgment unit is configured to obtain a positive feedback state if the target incentive effect is a positive feedback effect and the target incentive action is a positive feedback action; The screening unit is configured to screen in the chronic disease type grid according to the positive feedback state, and obtain the optimized chronic disease type. 4.The cloud platform-based chronic disease health data management system of claim 3, wherein, The screening unit comprises: The path establishment unit is configured to establish an optimization path in the chronic disease type grid according to the positive feedback state; The step number determination unit is configured to determine a step number maximum grid of a positive feedback step number in the optimization path, and obtain the optimized chronic disease type of the optimization disease type grid. 5.The cloud platform-based chronic disease health data management system of claim 1, wherein, The storage module comprises: The storage record obtaining unit is configured to store the health data, the optimized chronic disease type and the associated disease type in a cloud platform to obtain a storage record; The tagging unit is configured to perform a tagging process on the storage record in combination with a chronic disease record, a target user portrait and the storage record, and obtain tagged health data; The cooperation unit is configured to realize dynamic cooperation of chronic disease type identification, incentive feedback, associated disease type acquisition and early warning according to the tagged health data.

Images