CN111815487A - Health education assessment method, device and medium based on deep learning - Google Patents
Health education assessment method, device and medium based on deep learning Download PDFInfo
- Publication number
- CN111815487A CN111815487A CN202010596162.1A CN202010596162A CN111815487A CN 111815487 A CN111815487 A CN 111815487A CN 202010596162 A CN202010596162 A CN 202010596162A CN 111815487 A CN111815487 A CN 111815487A
- Authority
- CN
- China
- Prior art keywords
- health
- data
- value
- health education
- deep learning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Educational Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Security & Cryptography (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention relates to a health education assessment method, a device and a medium based on deep learning, which comprises the following steps: acquiring multiple items of dimensional data influencing the health education project through distributed acquisition equipment and distributed storage equipment; calculating a plurality of statistical indexes related to health problems according to the plurality of dimensional data; carrying out quantization processing on the multiple items of dimensional data and constructing panel data; establishing one or more deep learning sequence prediction algorithm models according to the panel; inputting the statistical indexes into a deep learning sequence prediction algorithm model, and setting a possibility value of influencing the health education project in a time period in the future; and setting a threshold value based on the possible value, outputting a time point at which the possible value is greater than the threshold value within a set time period in the future, and taking the time point as an optimal time point for developing health education. The invention has the beneficial effects that: the related data are acquired through the distributed big data, quantitative processing is carried out, learning calculation is carried out, and the health education project is effectively evaluated.
Description
Technical Field
The invention relates to the field of computers, in particular to a health education assessment method, device and medium based on deep learning.
Background
The concept and practical application of health education and health promotion are in a great significance in the whole process of health management. The design, implementation and evaluation of the health education program or the health management program is the best mapping of the main implementation steps or measures of health education or health management, such as data collection, demand evaluation, intervention implementation and effect evaluation. Among them, the design of the health education program, the health education diagnosis, also called health education need assessment, is the first and most important step in the design of the health education program.
The prior art lacks effective technical means to carry out data quantification and evaluation on health education.
Disclosure of Invention
The invention aims to solve at least one technical problem in the prior art, and provides a health education assessment method, device and medium based on deep learning.
The technical scheme of the invention comprises a health education assessment method based on deep learning, which is characterized by comprising the following steps: s100, acquiring multiple items of dimensional data influencing the health education project through distributed acquisition equipment and distributed storage equipment; s200, calculating a plurality of statistical indexes related to health problems according to the plurality of items of dimensional data; s300, carrying out quantization processing on a plurality of items of dimension data and constructing panel data; s400, establishing one or more deep learning sequence prediction algorithm models according to the panel; s500, inputting the statistical indexes into the deep learning sequence prediction algorithm model, and setting a possibility value of influencing healthy education projects in a time period in the future; s600, setting a threshold value based on the possible value, outputting a time point in a set time period in the future when the possible value is larger than the threshold value, and taking the time point as an optimal time point for developing health education.
According to the health education assessment method based on deep learning, the distributed acquisition equipment and the distributed storage equipment are based on the combination of Hadoop, Spark and Pysspark distributed frames, and the distributed storage equipment adopts Hive distributed storage to realize the building of a big data environment.
The method for evaluating health education based on deep learning, wherein S100 comprises: acquiring data of one or more parameters corresponding to quality of life data, social environment data, economic environment data, cultural environment data, policy data and social resource data by the distributed acquisition equipment; and calculating one or more indexes corresponding to the quality of life data, the social environment data, the economic environment data, the cultural environment data, the policy data and the social resource data according to the parameters.
The method for evaluating health education based on deep learning, wherein S200 comprises: according to the dimension data of the S100, the severity and the harmfulness of the health problems, the main health problems and the main risk factors of the health problems are counted; wherein, the calculation of the severity and the harmfulness medical statistical index of the health problem mainly comprises the following steps: if a certain health problem has short course and acute characteristics or is epidemic disease, calculating the morbidity, mortality and fatality rate of the health problem or the disease in a target area or a certain unit time in the population; if a certain health problem has long course of disease, is difficult to cure or is a chronic disease, calculating the morbidity, mortality and fatality rate of a certain unit time in a target area or a crowd; the calculation of medical statistical indexes of the main health problems and the main risk factors mainly comprises the steps of calculating the relative risk RR and ratio OR between the influence factors of certain health problems OR diseases and the diseases, and the attributive risk AR, the attributive risk percentage AR% and the population attributive risk percentage PAR between the main health problems OR the diseases;
the calculation method comprises the following steps:
AR-incidence in the influencer exposed group-incidence in the influencer unexposed group,
the method for evaluating health education based on deep learning, wherein S200 comprises: carrying out digital processing on the collected non-digital dimensional data to obtain quantized data, and calculating a prior intervention health problem value, a health problem influence factor value, a health education content value and a health education project possibility value by taking the dimensional data and the quantized data as original data; and constructing panel data by using any one of time, the quantized data and the prior intervention health problem value, the health problem influence factor value, the health education content value and the health education project possibility value corresponding to the quantized data.
The method for evaluating health education based on deep learning, wherein S400 comprises: establishing a corresponding prediction model through an OpenAR and ARIMA deep learning time series prediction algorithm, wherein the prediction model training and verification data are respectively based on the panel data.
The method for evaluating health education based on deep learning, wherein S500 comprises: and inputting the panel data into a time series prediction model, and operating and predicting a prior intervention health problem value, a health problem influence factor value, a health education content value and a health education possibility value in a specific time period in the future, wherein the prior intervention health problem value, the health problem influence factor value and the health education possibility value are predicted by adopting but not limited to an OpenAR multi-dimensional time series prediction model, and the health education content value is predicted by adopting but not limited to an ARIMA single-dimensional time series prediction model.
The method for evaluating health education based on deep learning, wherein the method further comprises the following steps: recording a group of values with highest priority intervention health problem value, health problem influence factor value, health education content value and health education possibility value and corresponding selected health education development time, health education problems, health problem influence factors and health education evaluation values, and marking by a one-hot coding principle; and when the evaluation information is accumulated to a set number, inputting the data into a time series prediction model for early simulation, and evaluating the success degree of the health education project plan.
The technical scheme of the invention also comprises a health education assessment device based on deep learning, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor is characterized by implementing any one of the method steps when executing the computer program.
The technical solution of the present invention further includes a computer-readable storage medium, in which a computer program is stored, and when the processor executes the computer program, any of the method steps is implemented.
The invention has the beneficial effects that: based on theoretical knowledge of epidemiology and medical statistics, the accurate assessment of health education projects is realized by combining big data and deep learning technology.
Drawings
The invention is further described below with reference to the accompanying drawings and examples;
FIG. 1 illustrates an overall flow diagram according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a distributed apparatus according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a detailed method for evaluating health education according to an embodiment of the present invention;
FIG. 4 shows a diagram of an apparatus and media according to an embodiment of the invention.
Detailed Description
Reference will now be made in detail to the present preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like elements throughout.
In the description of the present invention, the meaning of a plurality of means is one or more, the meaning of a plurality of means is two or more, and larger, smaller, larger, etc. are understood as excluding the number, and larger, smaller, inner, etc. are understood as including the number.
In the description of the present invention, the consecutive reference numbers of the method steps are for convenience of examination and understanding, and the implementation sequence between the steps is adjusted without affecting the technical effect achieved by the technical solution of the present invention by combining the whole technical solution of the present invention and the logical relationship between the steps.
In the description of the present invention, unless otherwise explicitly defined, terms such as set, etc. should be broadly construed, and those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention in combination with the detailed contents of the technical solutions.
Fig. 1 shows a general flow diagram according to an embodiment of the invention, the flow comprising: s100, collecting multiple items of dimensional data influencing the health education project through distributed collection equipment and distributed storage equipment; s200, calculating a plurality of statistical indexes related to health problems according to the plurality of dimensional data; s300, carrying out quantization processing on the multiple items of dimensional data and constructing panel data; s400, establishing one or more depth learning sequence prediction algorithm models according to a panel; s500, inputting the statistical indexes into a deep learning sequence prediction algorithm model, and setting a possibility value of influencing the health education project in a future set time period; s600, setting a threshold value based on the possible value, outputting a time point with the possibility value larger than the threshold value in a set time period in the future, and taking the time point as an optimal time point for developing health education; s700, inputting any health education item meeting the data requirements of the deep learning sequence prediction algorithm model to obtain a prediction or evaluation result.
Fig. 2 is a schematic diagram of a distributed apparatus according to an embodiment of the present invention, which includes a plurality of distributed acquisition devices, where the distributed acquisition devices may be various mobile devices, such as mobile phones, notebooks, mobile medical devices, or various computer devices with acquisition functions, and the distributed storage device may be various cloud servers. The big data storage and computing environment of Hive + Hadoop + Spark + Pysspark is adopted, Hive distributed storage and distributed computing based on Hadoop + Spark + Pysspark are adopted, and the big data environment is built.
Fig. 3 is a flowchart of a detailed method for evaluating health education according to an embodiment of the present invention, in which only step two, step three, step seven and step eight are indicated, and step one, step four, step five and step six mainly implement data acquisition, analysis and prediction model construction, as follows:
the method is divided into 7 modules: social diagnostics (specified community/population), epidemiological diagnostics, environmental diagnostics, behavioral diagnostics, educational and organizational diagnostics, administrative and policy diagnostics, and priority diagnostics. The 7 modules are independent and related to each other.
The first module, social diagnosis (designated community/crowd), aims to realize the social situation and demand assessment of the research target community or crowd. This module contains 7 submodules, is respectively: quality of life diagnosis, social environment diagnosis, economic environment diagnosis, cultural environment diagnosis, relevant policy diagnosis, and social resource diagnosis.
And the second module is epidemiological diagnosis, and aims to research the main health problems affecting the quality of life and the influencing factors thereof and determine the prior intervention health problems. This module contains 3 submodules, is respectively: health problem severity and hazard diagnosis, main health problem and main risk factor diagnosis and preferential intervention are established.
And a third module, environmental diagnosis (related to the specified health problem), aiming at determining environmental factors influencing the health condition and the quality of life and determining environmental factors for prior intervention. This module contains 2 submodules, is respectively: social factor diagnosis and substance condition factor diagnosis.
And a fourth module, behavior diagnosis, which aims to determine the behavior factors influencing the health condition and the quality of life so as to determine the behavior life style of the prior intervention. This module contains 3 submodules, is respectively: behavioral factor and non-behavioral factor diagnosis, important behavior and unimportant behavior diagnosis, high variable behavior and low variable behavior diagnosis.
And a fifth module, education and organization diagnosis, which aims to analyze relevant behaviors and environmental factors influencing health and causing specific health problems and provide effective basis for the specification of health education intervention strategies. This module contains 3 submodules, is respectively: predisposing factor diagnosis, contributing factor diagnosis and augmenting factor diagnosis.
And a sixth module, management and policy diagnosis, which aims to evaluate the resources and environment of the health education and evaluate the possibility and depth level of implementing the intervention measures of the health education. This module contains 3 submodules, is respectively: tissue resource diagnostics, external force diagnostics, and policy context diagnostics.
And a seventh module, priority project diagnosis, which aims to determine the health problems and the intervention behaviors which are preferentially solved by combining project feasibility and expected effects, so as to determine the health education projects which are preferentially realized. The module has 1 submodule in total and is used for evaluating and screening the priority project.
The invention is totally divided into 8 steps, which are respectively as follows: the method comprises the steps of big data environment building, data acquisition, epidemiology and medical statistical index calculation, data acquisition quantification and panel data building, time series prediction algorithm model building based on deep learning, model operation and health education project content building, priority health education project building and priority project evaluation.
Step one, building a big data environment. The method has the advantages that the specific requirements are stored for hardware according to the experimental data volume, and in addition, GPU hardware calculation is required for calculation. In the step, a Hive + Hadoop + Spark + Pysspark big data storage and calculation environment is adopted, Hive distributed storage and Hadoop + Spark + Pysspark are adopted as basic distributed calculation, and the big data environment is built.
Step two, data acquisition, wherein the data content acquired in the step includes but is not limited to:
data acquisition for module one:
quality of life related data (e.g., average consumption level of the city/region of the community or population, average income level of residents, residential living condition, average number of residents per square meter, Enger coefficient);
social environment-related data (e.g., religious popularity of the community or group, preference of residents for something due to the temperament of the local community or group, acceptance of new things or lifestyle by residents, understanding of health education by residents, understanding of health management by residents, popularity of human health concepts, etc.);
the economic environment related data (such as GDP, average consumption level, ratio of non-agricultural economic income to total income, dominant income of urban residents, total regional production value, third industry added value, fixed asset investment of the whole society, total retail of social consumer goods, total regional production rate increase, specific gravity of three-yield value and two-yield value, financial institution deposit, and total headquarter number of 500-strength enterprises in China);
cultural environment-related data (e.g., the number of universities in the target area 985/211, the number of patents applied/granted, the number of national key labs, the amount of R & D invested per year, the number of non-material cultural heritages, the level of preference and preference of residents for healthy food due to religious beliefs, the degree of acceptance of new things or lifestyle by residents due to religious beliefs);
policy-related data (e.g., population proportion of resident health record establishment, number of annual executions and proportion of residents in a health education lecture, vaccination popularity, number and proportion of annual family visits for neonatal health management, number and proportion of annual postnatal visits for puerperae, free blood pressure measurement and number and proportion of annual visits for 35 year old hypertensive patients, free fasting blood glucose measurement and number and proportion of annual visits for type II diabetic patients, free annual health examination and proportion of annual visits for severe psychotic patients, and proportion of healthy management for tuberculosis patients, etc.);
social resource-related data (e.g., target area human resource level assessment (including percentage of employees in the country, percentage of employees in the country who have a higher degree of school assignment than a scholarer, etc.), total amount of government assistance for health).
And (3) data acquisition of a module II:
calculating the data related to the severity and harmfulness of the health problem (for example, counting the health problems or diseases recorded by the community health service station, and counting the data for the health problems or diseases, wherein the frequency and the number of residents in the target area feed back the health problems or diseases each year, the number of confirmed health problems or diseases, the awareness rate of the health problems or diseases, the treatment rate and the control rate, the disability rate of the health problems or diseases, the fatality rate of the health problems or diseases, and the average influence degree of the health problems or diseases on the daily life of the confirmed patients);
data relating the health issue to its major risk factors is calculated (e.g., based on the health issues or diseases recorded by the community health service stations described above, data is counted for which the specific gravity of the health issue or disease due to genetic factors, the specific gravity of the health issue or disease due to smoking, the specific gravity of the health issue or disease due to insufficient physical activity, the specific gravity of the health issue or disease due to viral infection, the specific gravity of the health issue or disease due to dietary factors, the specific gravity of the health issue or disease due to obesity or overweight, the specific gravity of the health issue or disease due to hypertension, the specific gravity of the health issue or disease due to household factors).
For data acquisition in module three:
social factor-related data (the section including the social factor-related data collection for achieving health issues and diseases based on the above-described collected health issues and disease categories; e.g., the number of regulatory provisions by which the target community contributes to preventing the health issues or diseases, the number of measure items by which the target community contributes to preventing the health issues or diseases, the effectiveness of measures by which the target community contributes to preventing the health issues or diseases);
material condition factor-related data (which includes, in part, the collection of material condition factor-related data that enables the health issue and disease based on the types of health issues and diseases collected as described above; e.g., the number of facilities in the target community to diagnose the health issue or disease, the average life environment index of the target community to help prevent the health issue or disease, the average work environment index of the target community to help prevent the health issue or disease);
for data acquisition for module four:
this section includes enabling individualized health issue or disease initiation or causative factor-related data collection based on the health issues and disease categories collected as described above. These factors were then scored for behavioral factor health management based on the following criteria:
diagnosing the behavioral factors and the non-behavioral factors (judging whether the initiation or pathogenic factors of the specified health problems or diseases belong to the behavioral factors (such as improper diet) or the non-behavioral factors (such as genetic tendency), and if the diagnosis is the behavioral factors, recording the behavioral factor health management score as 5 points, and if the diagnosis is the non-behavioral factors, recording the behavioral factor health management score as 1 point);
diagnosis of important behavior versus unimportant behavior (judging whether a behavior factor that specifies the initiation or pathogenesis of a health issue or disease is an important behavior factor (e.g., a frequently occurring behavior or a cause and effect relationship of the behavior directly to the specified health issue or disease) or a non-important behavior factor (such that the behavior is not closely or frequently associated with the specified health issue or disease), if the diagnosis is important, the behavior factor health management score is multiplied by 10 times, and if the diagnosis is unimportant, the behavior is not changed);
high variable behavior and low variable behavior diagnosis (judging whether the cause of a given health problem or disease or the causative or pathogenic behavior factor is a variable behavior factor (e.g., behavior in the just formed or developing stage, evidence of successful change in other plans, social disapproval behavior) or a non-variable behavior factor (e.g., long formed time, rooted in cultural tradition or traditional lifestyle, no successful change example in the past), if high variable behavior is diagnosed, the behavior factor health management score is multiplied by 10, and if low variable behavior is diagnosed, the score is divided by 10).
For data acquisition for module five:
the part comprises the statistics of different health behaviors and influence factors thereof on members with health management in a target community or a group of people. Such factors can be used as educational directions for a given health education program, and these factors are classified into the following three categories:
diagnosing the tendency factor (for example, the individual has strong change will), and if the diagnosis is the tendency factor, recording the importance score of the health education of the factor as 2 points;
if the diagnosis is a tendency factor, the importance score of the factor health education is marked as 1 point;
if the diagnosis is a tendency factor, the importance score of the factor health education is 3 points;
for data acquisition for module six:
organizing resource diagnosis, including organizing internal strength of resources for developing the health education, such as the number of organizing people, capital, whether a site is determined, the experience level of personnel, the number of health managers in the organization, and the like;
external force diagnosis, including external support for the current health education, such as sponsorship amount, government fund withdrawal amount, and the like;
and the policy environment diagnosis comprises support and affirmation degree of local government and health departments on the health education work.
And thirdly, calculating epidemiology and medical statistical indexes, wherein index calculation is realized by combining the theory of medical statistics of the epidemiology based on the data collected in the second step, and related modules comprise a module II for calculating the medical statistical indexes of the severity and the harmfulness of the health problems, the main health problems and the main risk factors of the health problems.
The epidemiological and medical statistical indicators for module two were calculated as follows:
the step is based on the collected data, and the severity and the harmfulness of the health problems, the main health problems and the main risk factors of the health problems are calculated.
The calculation of the severity and the harmfulness medical statistical index of the health problem mainly comprises the following steps: if a certain health problem has short course and acute characteristics or is epidemic disease, calculating the morbidity, mortality and fatality rate of the health problem or disease in a target area or a certain unit time in the population; if a health problem has long course, is difficult to cure or is a chronic disease, calculating the morbidity, mortality and fatality rate of a certain unit time in the target area or the crowd.
The calculation of medical statistical indexes of the main health problems and the main risk factors mainly comprises the steps of calculating the Relative Risk (RR) and the ratio (OR) between the influencing factors of a certain health problem OR disease and the disease, and the Attribution Risk (AR), the attribution risk percentage (AR%) and the population attribution risk percentage (PAR%) between the main health problem OR disease and the disease. The calculation formula is as follows:
incidence rate of RR-certain influencing factor exposure group/incidence rate of certain influencing factor non-exposure group
OR-the number of exposed groups in case group/non-exposed groups in case group-the number of exposed groups in control group/non-exposed groups in control group
Incidence in the exposure group for certain influencing factor-incidence in the non-exposure group for certain influencing factor
Incidence rate of AR%
PAR％＝(P(RR-1))/(P(RR-1)+1)×100％
Step four, quantizing the collected data and establishing the panel data
The step two comprises the quantification of the collected data and the establishment of panel data.
The quantization process involves digitizing the acquired non-digital data, for example: the organization resource diagnoses whether the collected data is "site-determined", replaces the data of "yes" with "1", and replaces the data of "no" with "0".
The invention takes the collected data and the quantized data as the original data, and calculates the prior intervention health problem value, the health problem influence factor value, the health education content value and the health education project possibility value based on the original data.
A priority intervention health issue value is an assessment of how well a health issue is prioritized for intervention, the higher the value, the more the health issue should be prioritized for intervention. The index is calculated by performing regularization processing on the dimension index corresponding to the X axis of the following table related to the specific health problem, calculating the information entropy of regularized data of each index and calculating the average value as the value of the health problem to be intervened preferentially. Table 1 of the panel data is established based on the data, i.e., X-axis is time; the Y axis is information entropy data of the following indexes: quality of life related data, social environment related data, economic environment related data, cultural environment related data, related policy related data and social resource related data, social factor related data and material condition factor related data, health problem severity and harmfulness related data, morbidity/morbidity, mortality; the Z-axis is the priority intervention health issue value.
Table 1, the health issue influence factor value is the evaluation of the influence degree of the specified health influence factor on the specific health issue, and the larger the health issue influence factor value is, the higher the influence degree of the factor on the specified health issue is. The calculation method of the index comprises the steps of conducting regularization processing on the behavior factor health management score, the relative risk and ratio between the specified health problem or disease influence factor and the disease, the attribution risk percentage and the crowd attribution risk percentage, solving the information entropy of regularized results, and solving the mean value of the information entropy results to obtain the health problem influence factor value of the health influence factor on the specified health problem. Table 2 of the panel data was created based on the data that is time on the X-axis; the Y-axis is the behavioral factor health management score, the relative risk and ratio between the impact factors of a given health problem or disease and the disease, the attributable risk, the percentage of population attributable risk; the Z-axis is the health issue influencing factor value.
Table 2 health problem influence factor value calculation rule and corresponding panel data establishment method the health education content value is to evaluate the receptivity of the educated person to a certain health education content, and the higher the health education content value is, the more acceptable the educated person is to the health education content. The calculation method of the index is as follows: and regularizing the health education importance score data of the specified factors, and solving the information entropy of the regularization result, wherein the information entropy is the health education content value of the health influence factors to the specified health problems. A panel data table 3 is established based on the data that the X axis is time; the Y axis is health education importance evaluation data of specified factors; the Z axis is the health education content value.
| X axis | Time of day |
| Y-axis | Health education importance score data for specified factors |
| Z axis | Health education content value |
Table 3 health education contents value calculation rule and corresponding panel data establishment method the health education possibility value is a value for evaluating the degree of possibility of carrying out the health education, and the greater the index is, the more health education can be realized under limited resources and conditions. The calculation rule of the index is as follows: the method comprises the steps of regularizing data related to quality of life, data related to social environment, data related to economic environment, data related to cultural environment, data related to related policy and social resource, data related to social factor and material condition factor, data related to organizational resource, data related to external force and data related to policy environment, carrying out information entropy calculation on normalization results, and obtaining an index information entropy mean value as a healthy education possibility value. Establishing a panel data table 4 based on the data that the X axis is time; the Y axis is life quality related data, social environment related data, economic environment related data, cultural environment related data, related policy related data and social resource related data, social factor related data and material condition factor related data, organization resource related data, external force related data and policy environment related data; the Z-axis is the health education likelihood value.
TABLE 4 calculation method of health education probability values and corresponding panel data establishment method
Step five, establishing a time series prediction algorithm model based on deep learning
The method adopts a prediction algorithm which is not limited to OpenAR and ARIMA deep learning time series to establish four prediction models. And (4) respectively establishing four prediction model training and verification data based on the four panel data obtained in the step (4), wherein input nodes are Y-axis data, and output nodes are Z-axis data. The time sequence prediction of OpenAR in multiple dimensions is widely applied, and the time sequence prediction of ARIMA in a single dimension is applied.
Step six, operating the model and establishing the content of the health education project,
the four-panel data are input into the time series prediction model established in the fifth step, and the prior intervention health problem value, the health problem influence factor value, the health education content value and the health education possibility value in a specific time period in the future are operated and predicted. The prediction of the prior intervention health problem value, the health problem influence factor value and the health education possibility value is not limited to a time series prediction model which is more unique to OpenAR, and the health education content value is not limited to a time series prediction model with ARIMA single dimension.
And step seven, establishing a priority health education project.
Setting a threshold value aiming at the health education possibility value, outputting a time point (the time can be months and days) when the health education possibility value is larger than the threshold value in a specific time period (the time period can be one year, half a year and the like) in the future, and selecting the time point with the maximum health education possibility value as an optimal time point for developing the health education.
And selecting a series of health problems with the highest prior intervention health problem value, a group of health problem influence factors with the highest corresponding health problem influence factor values, and a series of health education contents with the highest health education content values as the key points of the health education.
Step eight, priority item evaluation
When the method is used for the first time, a group of values with the most appropriate (highest) priority intervention health question value, health question influence factor value, health education content value and health education possibility value selected in the step seven and corresponding selected health education development time, health education problems and health question influence factors, namely health education content are recorded and marked by adopting a one-hot coding principle (0/1 principle). And evaluating the health education plan developed this time after the health education is started.
And when the evaluation information is accumulated to a certain amount, inputting the data into the multi-unique time sequence model established in the step five, realizing advanced simulation, and evaluating the success degree of the current preferred project plan.
FIG. 4 shows a diagram of an apparatus and media according to an embodiment of the invention. Fig. 4 shows a schematic view of an apparatus according to an embodiment of the invention. The apparatus comprises a memory 100 and a processor 200, wherein the processor 200 stores a computer program for performing: acquiring multiple items of dimensional data influencing the health education project through distributed acquisition equipment and distributed storage equipment; calculating a plurality of statistical indexes related to health problems according to the plurality of dimensional data; carrying out quantization processing on the multiple items of dimensional data and constructing panel data; establishing one or more deep learning sequence prediction algorithm models according to the panel; inputting the statistical indexes into a deep learning sequence prediction algorithm model, and setting a possibility value of influencing the health education project in a time period in the future; and setting a threshold value based on the possible value, outputting a time point with the possibility value larger than the threshold value in a set time period in the future, and taking the time point as an optimal time point for developing the health education. Wherein the memory 100 is used for storing data.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.
Claims (10)
1. A health education assessment method based on deep learning is characterized by comprising the following steps:
s100, acquiring multiple items of dimensional data influencing the health education project through distributed acquisition equipment and distributed storage equipment;
s200, calculating a plurality of statistical indexes related to health problems according to the plurality of items of dimensional data;
s300, carrying out quantization processing on a plurality of items of dimension data and constructing panel data;
s400, establishing one or more deep learning sequence prediction algorithm models according to the panel;
s500, inputting the statistical indexes into the deep learning sequence prediction algorithm model, and setting a possibility value of influencing health education projects in a future set time period;
s600, setting a threshold value based on the possible value, outputting a time point in a set time period in the future when the possible value is larger than the threshold value, and taking the time point as an optimal time point for developing health education.
2. The deep learning-based health education assessment method according to claim 1, wherein the distributed acquisition devices and the distributed storage devices are based on a combination of Hadoop, Spark and Pysspark distributed frameworks, and the distributed storage devices adopt Hive distributed storage to realize big data environment construction.
3. The deep learning-based health education assessment method according to claim 1, wherein the S100 includes:
one or more parameters corresponding to the quality of life data, social environment data, economic environment data, cultural environment data, policy data and social resource data are acquired through the distributed acquisition equipment;
and calculating one or more indexes corresponding to the quality of life data, the social environment data, the economic environment data, the cultural environment data, the policy data and the social resource data according to the parameters.
4. The deep learning-based health education assessment method according to claim 1, wherein the S200 includes:
according to the dimension data of the S100, the severity and the harmfulness of the health problems, the main health problems and the main risk factors of the health problems are counted;
wherein, the calculation of the severity and the harmfulness medical statistical index of the health problem mainly comprises the following steps: if a certain health problem has short course and acute characteristics or is epidemic disease, calculating the morbidity, mortality and fatality rate of the health problem or the disease in a target area or a certain unit time in the population; if a certain health problem has long course of disease, is difficult to cure or is chronic disease, calculating the morbidity, mortality and fatality rate of a certain unit time in a target area or a crowd;
the calculation of medical statistical indexes of the main health problems and the main risk factors mainly comprises the steps of calculating the relative risk RR and ratio OR between the influence factors of certain health problems OR diseases and the diseases, and the attributive risk AR, the attributive risk percentage AR% and the crowd attributive risk percentage PAR between the main health problems OR the diseases;
the calculation method comprises the following steps:
AR-incidence in the influencer exposed group-incidence in the influencer unexposed group,
5. the deep learning-based health education assessment method according to claim 1, wherein the S200 includes:
carrying out digital processing on the collected non-digital dimensional data to obtain quantized data, and calculating a prior intervention health problem value, a health problem influence factor value, a health education content value and a health education project possibility value by taking the dimensional data and the quantized data as original data;
and constructing panel data by using any one value of time, the quantized data and the value of the priority intervention health problem value, the value of the health problem influence factor, the value of the health education content and the value of the health education item possibility corresponding to the quantized data.
6. The deep learning-based health education assessment method of claim 5, wherein the S400 includes:
establishing a corresponding prediction model through an OpenAR and ARIMA deep learning time series prediction algorithm, wherein the prediction model training and verification data are respectively based on the panel data.
7. The deep learning-based health education assessment method according to claim 6, wherein the S500 includes:
and inputting the panel data into a time series prediction model, and operating and predicting a prior intervention health problem value, a health problem influence factor value, a health education content value and a health education possibility value in a specific time period in the future, wherein the prior intervention health problem value, the health problem influence factor value and the health education possibility value are predicted by adopting but not limited to an OpenAR multi-dimensional time series prediction model, and the health education content value is predicted by adopting but not limited to an ARIMA single-dimensional time series prediction model.
8. The deep learning-based health education assessment method of claim 7, further comprising:
recording a group of values with highest priority intervention health problem value, health problem influence factor value, health education content value and health education possibility value and corresponding selected health education development time, health education problems, health problem influence factors and health education evaluation values, and marking by a one-hot coding principle;
and when the evaluation information is accumulated to a set number, inputting the data into a time series prediction model for early simulation, and evaluating the success degree of the health education project plan.
9. A deep learning based health education assessment apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor when executing the computer program implements the method steps of any of claims 1-8.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method steps of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010596162.1A CN111815487B (en) | 2020-06-28 | 2020-06-28 | Deep learning-based health education assessment method, device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010596162.1A CN111815487B (en) | 2020-06-28 | 2020-06-28 | Deep learning-based health education assessment method, device and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111815487A true CN111815487A (en) | 2020-10-23 |
| CN111815487B CN111815487B (en) | 2024-02-27 |
Family
ID=72855511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010596162.1A Active CN111815487B (en) | 2020-06-28 | 2020-06-28 | Deep learning-based health education assessment method, device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111815487B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112890816A (en) * | 2020-12-11 | 2021-06-04 | 万达信息股份有限公司 | Health index scoring method and device for individual user |
| CN114139883A (en) * | 2021-11-10 | 2022-03-04 | 云南电网有限责任公司信息中心 | Calculation method for material domain evaluation indexes of power enterprise |
Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060190303A1 (en) * | 2005-02-23 | 2006-08-24 | The Trinity Management Group, Llc | Coordinated health and human services delivery system and process |
| US20100082362A1 (en) * | 2008-09-17 | 2010-04-01 | Baker Salsbury | Method and Apparatus for Assessing Salient Characteristics of a Community |
| US20110196212A1 (en) * | 2010-02-09 | 2011-08-11 | John Peters | Methods and Systems for Health Wellness Management |
| WO2012089539A1 (en) * | 2010-12-16 | 2012-07-05 | Csi Ricerca & Ambiente Srl | A system for evaluating the current evolutionary status of events or processes in a geographical region and for generating and visualizing predictions of the evolutionary dynamics in time and space |
| JP2013088981A (en) * | 2011-10-17 | 2013-05-13 | Chugoku Electric Power Co Inc:The | Time prediction device, time prediction method and program |
| CN107316501A (en) * | 2017-06-28 | 2017-11-03 | 北京航空航天大学 | A kind of SVMs Travel Time Estimation Method based on grid search |
| CN108875842A (en) * | 2018-06-29 | 2018-11-23 | 山东师范大学 | A kind of Financial Time Series Forecasting method, server and device |
| WO2018236242A1 (en) * | 2017-06-21 | 2018-12-27 | Общество С Ограниченной Ответственностью "Оптимальное Управление" | METHOD AND DEVICE FOR PLANNING OF OPERATIONS WITH ENTERPRISE ASSETS |
| CN109284861A (en) * | 2018-08-31 | 2019-01-29 | 杭州网易再顾科技有限公司 | Predictive processing method, medium, apparatus and computing device |
| CN109376969A (en) * | 2018-12-14 | 2019-02-22 | 中南大学 | Method and device for dynamic prediction of urban refined population distribution based on deep learning |
| WO2019053234A1 (en) * | 2017-09-15 | 2019-03-21 | Spherical Defence Labs Limited | Detecting anomalous application messages in telecommunication networks |
| CN109637669A (en) * | 2018-11-22 | 2019-04-16 | 中山大学 | Generation method, device and the storage medium of therapeutic scheme based on deep learning |
| CN109801094A (en) * | 2018-12-07 | 2019-05-24 | 珠海中科先进技术研究院有限公司 | The method and system of prediction model are recommended in a kind of business analysis management |
| CN109875565A (en) * | 2019-01-25 | 2019-06-14 | 电子科技大学 | An automatic assessment method for upper limb motor function in stroke based on deep learning |
| CN110598299A (en) * | 2019-09-04 | 2019-12-20 | 北京理工大学 | A Surface Integrity Evaluation Method Based on Sensor Fusion and Deep Learning |
| CN110728425A (en) * | 2019-09-09 | 2020-01-24 | 南京起跑线数据科技有限公司 | Teaching quality analysis system for physical education and health education of primary and secondary schools |
| US20200074269A1 (en) * | 2018-09-05 | 2020-03-05 | Sartorius Stedim Data Analytics Ab | Computer-implemented method, computer program product and system for data analysis |
| CN111047094A (en) * | 2019-12-12 | 2020-04-21 | 国网浙江省电力有限公司 | Meter reading data anomaly analysis method based on deep learning algorithm |
| CN111191962A (en) * | 2020-01-10 | 2020-05-22 | 国网江苏省电力有限公司 | Analysis system and analysis method for omnibearing power economy prosperity index |
| CA3025302A1 (en) * | 2018-11-26 | 2020-05-26 | Loreen M. Sherman | Process to integrate quantified qualitative data into analytics |
-
2020
- 2020-06-28 CN CN202010596162.1A patent/CN111815487B/en active Active
Patent Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060190303A1 (en) * | 2005-02-23 | 2006-08-24 | The Trinity Management Group, Llc | Coordinated health and human services delivery system and process |
| US20100082362A1 (en) * | 2008-09-17 | 2010-04-01 | Baker Salsbury | Method and Apparatus for Assessing Salient Characteristics of a Community |
| US20110196212A1 (en) * | 2010-02-09 | 2011-08-11 | John Peters | Methods and Systems for Health Wellness Management |
| WO2012089539A1 (en) * | 2010-12-16 | 2012-07-05 | Csi Ricerca & Ambiente Srl | A system for evaluating the current evolutionary status of events or processes in a geographical region and for generating and visualizing predictions of the evolutionary dynamics in time and space |
| JP2013088981A (en) * | 2011-10-17 | 2013-05-13 | Chugoku Electric Power Co Inc:The | Time prediction device, time prediction method and program |
| WO2018236242A1 (en) * | 2017-06-21 | 2018-12-27 | Общество С Ограниченной Ответственностью "Оптимальное Управление" | METHOD AND DEVICE FOR PLANNING OF OPERATIONS WITH ENTERPRISE ASSETS |
| CN107316501A (en) * | 2017-06-28 | 2017-11-03 | 北京航空航天大学 | A kind of SVMs Travel Time Estimation Method based on grid search |
| WO2019053234A1 (en) * | 2017-09-15 | 2019-03-21 | Spherical Defence Labs Limited | Detecting anomalous application messages in telecommunication networks |
| CN108875842A (en) * | 2018-06-29 | 2018-11-23 | 山东师范大学 | A kind of Financial Time Series Forecasting method, server and device |
| CN109284861A (en) * | 2018-08-31 | 2019-01-29 | 杭州网易再顾科技有限公司 | Predictive processing method, medium, apparatus and computing device |
| US20200074269A1 (en) * | 2018-09-05 | 2020-03-05 | Sartorius Stedim Data Analytics Ab | Computer-implemented method, computer program product and system for data analysis |
| CN109637669A (en) * | 2018-11-22 | 2019-04-16 | 中山大学 | Generation method, device and the storage medium of therapeutic scheme based on deep learning |
| CA3025302A1 (en) * | 2018-11-26 | 2020-05-26 | Loreen M. Sherman | Process to integrate quantified qualitative data into analytics |
| CN109801094A (en) * | 2018-12-07 | 2019-05-24 | 珠海中科先进技术研究院有限公司 | The method and system of prediction model are recommended in a kind of business analysis management |
| CN109376969A (en) * | 2018-12-14 | 2019-02-22 | 中南大学 | Method and device for dynamic prediction of urban refined population distribution based on deep learning |
| CN109875565A (en) * | 2019-01-25 | 2019-06-14 | 电子科技大学 | An automatic assessment method for upper limb motor function in stroke based on deep learning |
| CN110598299A (en) * | 2019-09-04 | 2019-12-20 | 北京理工大学 | A Surface Integrity Evaluation Method Based on Sensor Fusion and Deep Learning |
| CN110728425A (en) * | 2019-09-09 | 2020-01-24 | 南京起跑线数据科技有限公司 | Teaching quality analysis system for physical education and health education of primary and secondary schools |
| CN111047094A (en) * | 2019-12-12 | 2020-04-21 | 国网浙江省电力有限公司 | Meter reading data anomaly analysis method based on deep learning algorithm |
| CN111191962A (en) * | 2020-01-10 | 2020-05-22 | 国网江苏省电力有限公司 | Analysis system and analysis method for omnibearing power economy prosperity index |
Non-Patent Citations (1)
| Title |
|---|
| 孙世岩;张钢;田福庆;梁伟阁;: "多输入混合深度学习网络的健康因子构建方法", 系统工程与电子技术, no. 10 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112890816A (en) * | 2020-12-11 | 2021-06-04 | 万达信息股份有限公司 | Health index scoring method and device for individual user |
| CN114139883A (en) * | 2021-11-10 | 2022-03-04 | 云南电网有限责任公司信息中心 | Calculation method for material domain evaluation indexes of power enterprise |
| CN114139883B (en) * | 2021-11-10 | 2024-03-29 | 云南电网有限责任公司信息中心 | Calculation method of material domain evaluation index of power enterprise |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111815487B (en) | 2024-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7430295B2 (en) | Individual chronic disease progression risk visualization evaluation method and system | |
| US20180211727A1 (en) | Automated Evidence Based Identification of Medical Conditions and Evaluation of Health and Financial Benefits Of Health Management Intervention Programs | |
| CN117592822A (en) | Data quality comprehensive evaluation method based on dynamic weight | |
| WO2003096237A2 (en) | Electronic data processing system and method of using an electronic data processing system for automatically determining a risk indicator value | |
| CN110136836A (en) | A Disease Prediction Method Based on Cluster Analysis of Physical Examination Reports | |
| CN110619084B (en) | Method for recommending books according to borrowing behaviors of library readers | |
| CN118173253A (en) | System and method for analyzing and managing based on patient data | |
| CN113342939A (en) | Data quality monitoring method and device and related equipment | |
| CN112418699A (en) | Resource allocation method, device, equipment and storage medium | |
| CN119622555A (en) | A knowledge and data-integrated chronic disease collaborative management method and system | |
| CN114141379A (en) | Sleep disorder attribution analysis method, device and system based on knowledge graph | |
| CN116759081A (en) | A machine learning-based monitoring and early warning method for adolescents’ psychological and behavioral abnormalities | |
| CN111815487A (en) | Health education assessment method, device and medium based on deep learning | |
| CN113793693A (en) | Method and device for predicting epidemic trend of infectious diseases | |
| CN113313470B (en) | Employment type assessment method and system based on big data | |
| CN119273218A (en) | Core competence evaluation method, device and storage medium for full-time hospital infection management personnel | |
| CN114662963A (en) | Expert intelligent review management method | |
| CN119990341A (en) | An intelligent question-answering data processing system based on artificial intelligence | |
| CN112465277A (en) | Scientific score forecasting method, system, device and storage medium for students at school | |
| CN119993429A (en) | Medical queuing calling method, system, device and storage medium | |
| CN113435748A (en) | Dot state determination method and apparatus, electronic device and storage medium | |
| CN118885832A (en) | A user portrait construction and clustering method based on multi-dimensional attributes | |
| US20230409926A1 (en) | Index for risk of non-adherence in geographic region with patient-level projection | |
| CN114429820A (en) | Intelligent rehabilitation evaluation system and method for hospital rehabilitation department | |
| CN115205072A (en) | Cognitive diagnosis method for long-period evaluation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |