CN113539492A - Urban health index prediction system, prediction analysis method and storage medium thereof - Google Patents
Urban health index prediction system, prediction analysis method and storage medium thereof Download PDFInfo
- Publication number
- CN113539492A CN113539492A CN202110667979.8A CN202110667979A CN113539492A CN 113539492 A CN113539492 A CN 113539492A CN 202110667979 A CN202110667979 A CN 202110667979A CN 113539492 A CN113539492 A CN 113539492A
- Authority
- CN
- China
- Prior art keywords
- data
- health
- index
- health index
- prediction
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention belongs to the technical field of city monitoring, and discloses a city health index prediction system, a prediction division method and a storage medium thereof, wherein health data are collected, processed and analyzed through a health index cloud platform; the urban health index system adopts a visualization technology to carry out comprehensive health condition monitoring and risk prediction in different areas; the personal health analysis system excavates disease risk factors, adopts artificial intelligence and machine learning technology to establish an accurate prediction model, carries out risk factor analysis based on big data, fuses an expert knowledge base, and provides individualized prevention and intervention suggestions for accurate people. The invention carries out overall evaluation of indexes and data with different properties, different dimensions and magnitude in all directions, multiple fields and multiple aspects, and finally realizes the visual display of the prediction result according to the requirements and scenes of individuals and decision makers.
Description
Technical Field
The invention belongs to the technical field of urban monitoring, discloses an urban health index prediction system, a prediction division method and a storage medium thereof, and particularly relates to urban health index model establishment, function definition, index design, score proportion and prediction system establishment, and health data acquisition, processing and analysis are carried out through a health index cloud platform.
Background
The health level of people is continuously improved, and simultaneously, China also faces new challenges brought by industrialization, urbanization, population aging, disease spectrum, ecological environment, life style continuous change and the like, and needs to comprehensively solve major and long-term problems related to the health of people. In order to adapt to the new development trend of the economic society, the deep integration of high technology and modern medicine becomes a necessary development trend.
At present, the research on urban health indexes in China is less, and the construction of an information platform and the data release based on the urban health indexes are also rarely reported. At the present stage, the definition of the healthy city is single, and the relevance of the content containing the uncapping is poor.
However, the existing health analysis technologies are mainly directed to individuals and not directed to regions, meanwhile, a traditional health prediction model is usually established based on a small number of known risk factors, and a modeling method generally adopts a traditional statistical analysis method, so that the prediction accuracy is limited. Traditional models lack accurate disease prevention decision support. At present, decision suggestions for disease prevention are generally extensive, and individual prevention intervention suggestions cannot be given for accurate people generally.
Through the above analysis, the problems and defects of the prior art are as follows:
(1) the prior art does not have a method for forecasting the health index of a city or region.
(2) Traditional disease prediction models are usually built based on a small number of known risk factors, and the modeling method generally adopts a traditional statistical analysis method, so that the prediction accuracy is limited.
(3) Lack of accurate disease prevention decision support by traditional models
(4) At present, decision suggestions for disease prevention are generally extensive, and individual prevention intervention suggestions cannot be given for accurate people generally.
The difficulty in solving the above problems and defects is:
firstly, all medical and health institution information systems in a specific city realize interconnection and intercommunication, and services are shared cooperatively; secondly, information such as various medical treatment records, chronic disease records, physical examination results, planned immunization records, past medical history, various inspection and examination results, treatment records, medicine allergy history, behavior risk factors, records participating in health education activities and the like of all masses in the urban district are collected to an urban health index cloud platform, and then management and innovative application of health files are realized by means of cloud computing, big data, artificial intelligence and the like; and thirdly, data of all departments such as natural environment, economic indexes, urban construction level, education resources, life expectancy of everyone, traffic and the like in the city need to be interconnected and intercommunicated.
The significance of solving the problems and the defects is as follows:
health portrayal and scoring are carried out on each city according to multiple dimensions such as a health resource index, a natural environment index, a life expectancy index, a life style index and a chronic disease index, the health level of the city is comprehensively displayed, the urban health index serves as the basic public service field of people, and the method becomes possible to continuously adapt to the development of the economic society and the health requirements of people.
Firstly, the sanitary resource index provides the feeling most suitable for the living standard of people for the index of the healthy city. The health service system capable of sustainable development is intuitively shown in the aspects of the proportion of technicians in every thousand population, the proportion of technicians in every thousand population hospitals, the proportion of technicians in every thousand population health institutions, the proportion of beds in every thousand population medical health institutions, the proportion of comprehensive hospitals in every thousand population, doctors in every thousand population executives (assistants), registered nurses in every thousand population, the number of staff in professional public health institutions in every thousand population, the number of staff in basic health institutions in every thousand population, the proportion of beds in every thousand population hospitals in every thousand population, the number of third-level hospitals in every thousand population, the number of hospitals above second-level in every thousand population, the number of medical health institutions in every thousand population and the like. And evaluating the fairness of the medical resource allocation and the utilization level of the medical service of each region by using the density index of the sanitary resources and the coefficient of the kini and according to the sequential optimization technology of the similarity of the ideal scheme.
And secondly, the natural environment index provides a green development visual feeling for the healthy city index. This part is mainly analyzed by collecting and analyzing inhalable particles (PM2.5) sulfur dioxide (SO) in unit hour2) Nitrogen dioxide (NO)2) And (4) integrating the values to obtain the air quality index (AQI index).
And thirdly, the life expectancy index provides a reference of life expectancy for the healthy city index. The system management rate of pregnant and lying-in women, the system management rate of children under 3 years old, the system health care rate of children under 7 years old, the low weight morbidity of children under 5 years old, the mortality rate of pregnant and lying-in women, the mortality rate of infants, the postpartum visit rate, the prenatal examination rate of the pregnant and lying-in women and the expected life of the pregnant and lying-in women in urban jurisdictions are comprehensively analyzed, and urban vitality is comprehensively displayed.
And fourthly, the life style index provides a habitual activity reference of the persistent behavior mode for the healthy city index. The research mainly analyzes and researches from the aspects of daily diet, physical activity, substance abuse, social and interpersonal interaction and the like.
And fifthly, the chronic disease index provides a basis for healthy city index chronic disease prevention and treatment, health literacy knowledge and skill propaganda activity, so that the knowledge level of the chronic diseases of people is improved, and the initiative of the people participating in the chronic disease prevention and treatment is increased. Meanwhile, through the study of chronic disease indexes, the utilization of personal health files and physical examination information can be greatly improved, the management of important chronic diseases such as diabetes, hypertension and the like is developed, and the management is possible for early discovery and early treatment and the enhancement of the daily management of patients with chronic diseases.
According to the invention, through the construction of an urban health index system, the health portrait and the score of each city are comprehensively displayed by applying the emerging technologies such as big data, cloud computing, artificial intelligence, Internet + and the like from multiple dimensions such as a health resource index, a natural environment index, a life expectancy index, a life style index, a chronic disease index and the like, and the urban health index is taken as the basic public service field of people, so that the urban health index system continuously adapts to the development of economic society and the health requirements of people.
Disclosure of Invention
Aiming at the problems in the prior art, the invention establishes an urban health index prediction system, a prediction division method and a storage medium thereof, particularly relates to an urban health index model, creatively realizes the definition of the functions of a healthy city, the design of indexes, the ratio of scores and the establishment of a prediction system, and simultaneously acquires, processes and analyzes health data through a health index cloud platform.
The invention is realized in this way, a city health index prediction system, comprising:
the health index data cloud platform comprises a data docking module and a data analysis module; the health data acquisition, processing and analysis are carried out;
the urban health index system is used for carrying out comprehensive health condition monitoring and risk prediction in different areas by adopting radar maps, color block tables, trend maps, geographic information systems and other visualization technologies;
the personal health analysis system is used for mining disease risk factors from a large amount of medical data and medical documents based on big data, establishing an accurate prediction model by adopting artificial intelligence and machine learning technology, analyzing the risk factors based on the big data, fusing expert knowledge, and providing individualized prevention and intervention suggestions for accurate people.
Further, the health index data cloud platform comprises:
the data docking module is used for acquiring corresponding health data of each health data platform, the diagnosis and treatment system and other prevention and control public health data platforms;
the data analysis module comprises a data integration unit, a data conversion unit, a data cleaning unit and a missing value processing unit; the method is used for integrating, converting, cleaning and processing missing values of the acquired data.
Further, the city health index system includes:
the medical big data deep mining module comprises a health index modeling unit and a health index prediction health index visualization unit; the method is used for constructing and visualizing the urban health index model;
the urban health index building module comprises a health index system building unit and a health index forecasting unit; the method is used for constructing an urban health index system and carrying out health forecast.
Further, the medical big data deep mining module comprises:
the health index modeling unit is used for constructing a health panorama reflecting the region, sensing the current situation in real time based on big data and expressing a health index model of the whole health portrait of the dynamically monitored city;
a health index prediction health index visualization unit; the index model display method is used for displaying the results of the index model in a layering mode by using radar maps, color block tables, trend maps, geographic information systems and other visualization technologies and monitoring the index dynamics of all levels.
Further, the city health index construction module comprises:
the health index system building unit is used for building a health index system comprising a chronic non-infectious disease index, a sanitary resource index, life expectancy, a natural environment index and a regional health index;
and the health index forecasting unit is used for forecasting the urban health based on the health index indexes.
Further, the integrated health monitoring and risk prediction includes monitoring and prediction of chronic non-infectious diseases, health services index, natural environment index.
Another object of the present invention is to provide a method for predicting an urban health index applied to the urban health index prediction system, the method comprising:
step one, collecting, processing and analyzing urban and individual health data;
step two, comprehensive health condition monitoring and risk prediction of different areas are carried out by adopting radar maps, color block tables, trend maps, geographic information systems and other visualization technologies;
and step three, excavating disease risk factors from a large amount of medical data and medical documents based on big data, establishing an accurate prediction model by adopting artificial intelligence and machine learning technology, analyzing the risk factors based on the big data, fusing expert knowledge, and providing individualized prevention and intervention suggestions for accurate people.
Further, in the first step, the processing the acquired data includes:
and (3) carrying out data integration: extracting various types of acquired original data and integrating the data of different data sources according to a certain contact rule to obtain a unified data set comprising basic information, baseline medical history, inspection, diagnosis/prescription, follow-up visit and other various types of data;
and (3) carrying out data conversion: performing data type conversion, data semantic conversion, data value domain conversion, data granularity conversion, table/data splitting, row-column conversion, data discretization, data standardization, new field refinement and other processing on the obtained unified data set;
and (3) data cleaning: determining different cleaning rules according to different data types and error types, and regulating the data format and processing or modifying unreasonable data based on the determined data cleaning rules;
and (3) carrying out missing value processing: filling missing feature values based on rules, filling missing patient data with statistics of other patients, or supplementing missing data using linear regression/machine learning methods.
Further, the establishing of the accurate prediction model by adopting artificial intelligence and a machine learning technology comprises:
and establishing an accurate prediction model by adopting a method based on traditional regression analysis, traditional machine learning and ensemble learning.
By combining all the technical schemes, the invention has the advantages and positive effects that: the invention utilizes the existing medical health big data, carries out comprehensive, multi-field and multi-aspect overall evaluation containing indexes and data with different properties, different dimensions and magnitude orders by the technical means of description analysis, intelligent prediction modeling research, artificial intelligence and the like of the medical health big data, and finally realizes the visual display of the prediction result according to the requirements and scenes of individuals and decision makers.
The invention can also carry out risk prediction by butting resident electronic health files and health physical examination data and adopting a high-precision chronic disease prediction model, gives the risk of each resident about hypertension, diabetes, coronary heart disease and cerebral apoplexy in the next five years, gives high, medium and low risk levels and gives risk details and personalized health management suggestions.
Drawings
FIG. 1 is a schematic structural diagram of a city health index prediction system provided by an embodiment of the present invention;
in the figure: 1. a health index data cloud platform; 2. an urban health index system; 3. a personal health analysis system.
FIG. 2 is a flow chart of a city health index prediction method provided by an embodiment of the invention;
FIG. 3 is a multi-dimensional map of an urban health index provided by an embodiment of the invention;
FIG. 4 is an index of health resources for an index of urban health provided by an embodiment of the present invention;
FIG. 5 is a natural environment index of the urban health index provided by an embodiment of the present invention;
FIG. 6 is a life expectancy index for an urban health index provided by an embodiment of the present invention;
FIG. 7 is a city health index lifestyle index provided by an embodiment of the present invention;
fig. 8 is a city health index lentigo index provided by an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In view of the problems in the prior art, the present invention provides a city health index prediction system, which is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the urban health index prediction system provided by the embodiment of the present invention includes:
the health index data cloud platform 1 comprises a data docking module and a data analysis module; the health data acquisition, processing and analysis are carried out;
the urban health index system 2 is used for carrying out comprehensive health condition monitoring and risk prediction in different areas by adopting radar maps, color block tables, trend maps, geographic information systems and other visualization technologies;
the personal health analysis system 3 is used for mining disease risk factors from a large amount of medical data and medical documents based on big data, establishing an accurate prediction model by adopting artificial intelligence and machine learning technology, analyzing the risk factors based on the big data, fusing expert knowledge, and providing individualized prevention and intervention suggestions for accurate people.
The health index data cloud platform provided by the embodiment of the invention comprises:
the data docking module is used for acquiring corresponding health data of each health data platform, the diagnosis and treatment system and other prevention and control public health data platforms;
the data analysis module comprises a data integration unit, a data conversion unit, a data cleaning unit and a missing value processing unit; the method is used for integrating, converting, cleaning and processing missing values of the acquired data.
The urban health index system provided by the embodiment of the invention comprises:
the medical big data deep mining module comprises a health index modeling unit and a health index prediction health index visualization unit; the method is used for constructing and visualizing the urban health index model;
the urban health index building module comprises a health index system building unit and a health index forecasting unit; the method is used for constructing an urban health index system and carrying out health forecast.
The medical big data depth mining module provided by the embodiment of the invention comprises:
the health index modeling unit is used for constructing a health panorama reflecting the region, sensing the current situation in real time based on big data and expressing a health index model of the whole health portrait of the dynamically monitored city;
a health index prediction health index visualization unit; the index model display method is used for displaying the results of the index model in a layering mode by using radar maps, color block tables, trend maps, geographic information systems and other visualization technologies and monitoring the index dynamics of all levels.
The urban health index construction module provided by the embodiment of the invention comprises:
the health index system building unit is used for building a health index system comprising a chronic non-infectious disease index, a sanitary resource index, life expectancy, a natural environment index and a regional health index;
and the health index forecasting unit is used for forecasting the urban health based on the health index indexes.
The comprehensive health condition monitoring and risk prediction provided by the embodiment of the invention comprises monitoring and prediction of chronic non-infectious diseases, health service indexes and natural environment indexes.
As shown in fig. 2, the urban health index prediction method provided by the embodiment of the present invention includes:
s101, collecting, processing and analyzing urban and individual health data;
s102, comprehensive health condition monitoring and risk prediction in different areas are carried out by adopting radar maps, color block tables, trend maps, geographic information systems and other visualization technologies;
s103, disease risk factors are mined from a large amount of medical data and medical documents based on big data, an accurate prediction model is established by adopting artificial intelligence and machine learning technology, risk factor analysis is carried out based on the big data, expert knowledge is fused, and individualized prevention and intervention suggestions are provided for accurate people.
The processing of the acquired data provided by the embodiment of the invention comprises the following steps:
and (3) carrying out data integration: extracting various types of acquired original data and integrating the data of different data sources according to a certain contact rule to obtain a unified data set comprising basic information, baseline medical history, inspection, diagnosis/prescription, follow-up visit and other various types of data;
and (3) carrying out data conversion: performing data type conversion, data semantic conversion, data value domain conversion, data granularity conversion, table/data splitting, row-column conversion, data discretization, data standardization, new field refinement and other processing on the obtained unified data set;
and (3) data cleaning: determining different cleaning rules according to different data types and error types, and regulating the data format and processing or modifying unreasonable data based on the determined data cleaning rules;
and (3) carrying out missing value processing: filling missing feature values based on rules, filling missing patient data with statistics of other patients, or supplementing missing data using linear regression/machine learning methods.
The method for establishing the accurate prediction model by adopting artificial intelligence and machine learning technology provided by the embodiment of the invention comprises the following steps:
and establishing an accurate prediction model by adopting a method based on traditional regression analysis, traditional machine learning and ensemble learning.
The technical solution of the present invention is further illustrated by the following specific examples.
Example 1:
the construction of the system takes urban and rural residents as a center, integrates resources, provides personal disease prediction service and urban health supervision, focuses on service and supervision effects, can conveniently improve the statistical analysis and decision support capability of managers after the system is built, and can better serve the urban and rural residents and health managers.
(1) Integrated design concept
The integrated design concept is adopted, the urban health supervision service is covered in all directions, and the relevant information system is integrated. The management, prevention and control of chronic diseases are more favorably enhanced, and the intelligent and precise health management is promoted.
(2) Intensive data center deployment
The urban health index and personal disease prediction project adopts a cloud computing intensive data center deployment mode, provides a complete cloud computing mode infrastructure construction scheme or a public cloud base implementation renting scheme, and realizes one-center and multi-role access use.
Compared with a traditional single-institution construction mode, the urban health index and personal disease prediction platform can realize maintenance-free operation, the cost of informatization infrastructure implementation construction and operation and maintenance is greatly reduced, and the construction efficiency is improved.
(3) SaaS application mode
In terms of software application, the urban health index and disease prediction platform adopts a SaaS software application service mode, provides a software application service multi-tenant mechanism, and realizes one-center deployment and multi-mechanism use.
Architectural design
Information interconnection and intercommunication are already realized on the national health informatization platform in Gansu province, the existing medical health big data are utilized, the overall evaluation of indexes and data which are all-around, multi-field and multi-aspect and comprise different properties, different dimensions and magnitude levels is carried out through the technical means of description analysis, intelligent prediction modeling research, artificial intelligence and the like of the medical health big data, and finally the visual display of prediction results is realized according to the demands and scenes of individuals and decision makers.
The Gansu health index analysis system comprises index research and development, visual display, systematic service management and the like.
System deployment
A Gansu health index and disease prediction analysis system is deployed in a data center of a national health information platform of Gansu province, data exchange with other systems is realized based on a platform data exchange sharing system, meanwhile, a test point mechanism is selected to run on line, and after experience is obtained, popularization of the Gansu province is realized.
2. System function implementation
2.1 building health index data cloud platform
2.1.1 data docking
The method is characterized by comprising the steps of connecting a health data platform of the whole people of Gansu province to a public health data cloud platform of public health care and precision prevention and control based on a health file and disease diagnosis and treatment 'one card through' information network, and establishing a working tamping data source foundation for a Gansu health index and disease prediction system. Standardizing data push standards, realizing integrated management of public health data resources and integration, mining and publishing of data, and gradually constructing an informatization application mode of three-in-one service cooperation of medical data push, disease control service management and community follow-up intervention. Through the platform, public information resources of a social network are integrated, a disease sensitive information early warning mechanism is perfected, the occurrence trend of diseases of a whole crowd is mastered and dynamically analyzed in time, and the early warning and emergency response capabilities of emergent public health events are improved. And multi-party monitoring data such as environmental sanitation, drinking water, health hazard factors, medical media biology and the like are integrated, and social factors influencing health are effectively evaluated.
Wherein, cloud computing center is project construction's infrastructure support, mainly includes: machine rooms, general and special infrastructure, network equipment, power distribution, lighting, integrated wiring. The cloud computing center mainly realizes unified storage and sharing of data, unified integration and mutual integration of services, and support of subsequent extended application; monitoring and management of the operation of the basic operation supporting platform are realized; realizing the timed backup management of data; and monitoring management of network communication is realized.
2.1.2 data analysis
The data normalization processing is to perform a series of steps of integration, conversion, cleaning and missing value processing on a data set, so that the data quality is improved, and the data normalization processing is a core support for constructing a data cloud platform. Medical data content covers a wide variety of data types, including: options, date and time, numeric data, character data, and non-numeric data. Aiming at different data types, different processing methods are required to carry out targeted data preprocessing.
2.1.2.1 data integration
Data integration is the integration of data from different sources, formats, and characteristics to provide comprehensive data for later analysis. Data provided by different systems are different in structure, and data sources are independent and closed, so that data fusion of different data sources is completed through a data integration step.
The first step of data integration is to export various types of raw data from the hospital information system. For example, basic information about a patient can be generally obtained from the HIS system in a hospital; the baseline medical history is typically generated by an EMR system; the inspection data is mostly recorded in the LIS system; diagnosis/prescription can also be derived from the EMR system in most cases; the follow-up information can typically be derived from the EMR system, or a corresponding follow-up system.
The extraction of the original data is completed, the second step of data integration is to integrate the data of different data sources according to a certain contact rule, generally speaking, the contact rule is mainly formulated according to the patient main index (such as ID number) or group serial number, the different data sources are connected together with the information of the same ID number or serial number, and thus all the data from the different data sources are integrated together. After the data integration is completed, the data from the different data sources is integrated into a unified data set.
2.1.2.2 data conversion
The unified data set obtained by data integration comprises various data such as basic information, baseline medical history, examination, diagnosis/prescription, follow-up visit and the like, and can be related by a patient main index or a group serial number and the like. Data transformation is an important step of the data preprocessing process, which is a standard process on one of the data, and is involved in almost all data preprocessing processes. Common content for data conversion includes: data type conversion, data semantic conversion, data value domain conversion, data granularity conversion, table/data splitting, line and column conversion, data discretization, data normalization, refining new fields (such as summation, average, mode, etc. of fields), and the like.
2.1.2.3 data cleansing
The data cleaning task is to clean data from the aspects of accuracy, consistency, non-redundancy and the like through various measures, improve the data quality, facilitate the operation of next missing data processing and prepare for subsequent data verification. The reasons for the data quality problem are manifold and therefore represent different forms of problems. For example, there may be problems with datetime-type data with an unqualified date format, such as 10/1/2015 and 2005-5-10, and with date values outside of a reasonable time range, such as 2115 for a certain diagnostic time. Common problems with numeric data include uniform units of values, and out of reasonable ranges of values. Other types of data may also relate to content that is not structurally uniform or multiple selection results for a single selection option, etc.
And for different data types, summarizing common error types by observing common errors, and designing different cleaning rules. These cleansing rules are used to satisfy the above-mentioned normalization of data formats and the processing or modification of unreasonable data. After the data cleaning is finished, the accuracy, consistency and non-redundancy of the original data are improved, and the overall data quality is improved.
2.1.2.4 missing value handling
Missing data processing refers to performing a series of processes on data missing from a data set in preparation for later data verification. Some missing data inevitably exists in the data set to be processed. Meanwhile, the data set of the medical field characteristic variables is generally sparse in time sequence, and the data verification after the data is further hindered by the absence of the data. The core goal of missing data processing is to fill in missing data as scientifically and reasonably as possible. The missing data processing methods can be classified into the following categories, as shown in the following table:
2.2 urban health index System construction
Gansu health indexes are issued periodically (quarterly and yearly), and adopt visualization technologies such as radar maps, color block tables, trend maps and geographic information systems to provide comprehensive health condition monitoring and risk prediction information (including monitoring and prediction of chronic non-infectious diseases, health service indexes, natural environment indexes and the like) of different administrative regions, and are assisted with intervention measures.
2.2.1 deep mining of data based on medical big data platform
2.2.1.1 health index modeling
The health index model aims to reflect a health panorama of a region, senses the current situation in real time based on big data, dynamically monitors the whole health portrait of a city, and has the five characteristics of comprehensiveness, rigidness, objectivity, authority and flexibility. Not only timely and effectively collects resources for government health departments and plays a role in assisting decision making in pertinently developing prevention work, but also can effectively develop disease propaganda and education, and fundamentally improves and promotes the health behaviors of the whole society.
2.2.1.2 health index prediction health index visualization
The urban health index is in principle composed of a primary index (urban health index), a secondary index (e.g. disease index) and a tertiary index (e.g. sanitary resource, chronic non-infectious disease index, etc.). In order to visually display each index, the module is used for hierarchically displaying the result of the index model by using the visualization technologies such as a radar chart, a color block table, a trend chart, a geographic information system and the like, and the system monitors the dynamic indexes at all levels.
Big data visualization techniques include traditional scientific visualization and information visualization. From the big data analysis perspective, information mining and insight knowledge are used as targets. Although the medical health big data information is various in types, the readability of the medical big data can be enhanced through the visual presentation mode, and the medical big data can be further understood and applied conveniently.
And aiming at the result of big data analysis, realizing the visual display of the data analysis result according to the requirements and scenes of individuals and decision makers. The method provides high-dimensional display, ad hoc query, statistical report forms and other display modes, the graphics comprise dozens of graphs, and besides common bar graphs, line graphs, scatter diagrams and radar graphs, various special graphs for big data analysis are provided, such as path analysis graphs and character cloud graphs. The method supports a dragging type definition chart, performs combined calculation on dimensions and measurement, and supports complex calculation of customer retention rate, percentage and the like; and functions of in-table brushing selection, cross-table linkage and the like are supported.
Based on the basis of medical big data of the Gansu province and the construction work of each health index, a visual analysis platform of 'Gansu health index' of the Gansu province is comprehensively constructed. A real LBS Gansu province city map scene is built, three aspects of data dimension analysis including internal factors (demographic characteristics, health history, physical examination data and health risk factors), external factors (transportation, environment, public safety and city planning) and medical data (mechanism setting, mechanism quality, outpatient service information and hospitalization information) are displayed in a big data visualization analysis platform, and disease health risk assessment and health risk model prediction are achieved. In conclusion, the big data visualization platform can cross-compare and visualize relevant factors of various diseases, show the distribution overview of high risk groups of the diseases, simultaneously show the progressive display overview of various risk levels and various time periods, assist in realizing the prediction and early screening of the high risk groups of the diseases and provide reasonable opinions and suggestions.
2.2.2 application construction of urban health index
2.2.2.1 Gansu health index System construction
The Gansu health index model is intended to combine the expert opinions in multiple fields such as epidemiology, clinical medicine, environmental science, health economics, data science and the like, and a big data analysis method is applied to realize the health condition and development trend of the area from multiple aspects and multiple levels.
Specifically, the city health index system encompasses several modules.
2.2.2.1.1 construction of Chronic non-infectious disease index
Chronic non-infectious diseases refer to a class of diseases with long disease course, no infectivity, no self-healing and almost no cure at present. Because the etiology of the diseases is complex, the diseases are related to various factors such as bad behaviors, life styles and the like; the diseases have long incubation period, no definite disease acquisition time and long course of disease, and are manifested as functional impairment or disability along with the disease development. The diseases comprise cardiovascular and cerebrovascular diseases, malignant tumors, diabetes, chronic obstructive pulmonary diseases and the like.
The method is used for screening, evaluating and predicting key diseases based on multi-source heterogeneous data formed by traditional data of disease control, physical examination organizations, multi-level medical organizations and the like. By means of artificial intelligence and big data technology, fusion of multi-source heterogeneous data is effectively achieved, relevant influence factors are deeply mined, and all-around and multi-angle automatic comprehensive assessment and prediction are conducted on relevant disease problems (such as hypertension, diabetes, cerebral apoplexy, coronary heart disease and the like). Visual data value presentation is realized through a visualization technology, the readability of data is improved, and early disease screening of high-risk groups with chronic diseases and auxiliary health management of patients are finally realized.
2.2.2.1.2 sanitary resource index construction
Health services are a large important component for measuring public health of people, and comprise public health services and medical health services, and the health service quality and coverage are improved, so that the health level of people can be directly improved.
The health resource index construction relates to various aspects of people, property and material related to health services, a perfect evaluation system is established by monitoring indexes of health personnel conditions (such as the number of average people of health personnel, the number of doctors, the academic proportion and the title proportion), medical guarantee conditions (such as the health financial expenditure, the medical insurance coverage condition and the like), health institution/facility conditions (such as the number of hospitals, the number of disease control centers, the number of basic health service centers, the number of beds and the like), a quantitative evaluation basis of health services is provided for the Gansu health index, and a comprehensive health service capability index comprising the health personnel index, the medical guarantee index and the health service index is established. By the machine learning technology, factors, incidence relations and influence ratios which influence health services can be deeply mined. Thereby providing quantitative basis for the establishment of urban health index and improvement measures.
2.2.2.1.3 Life expectancy index construction
The life expectancy index is compiled based on population information indexes which can directly reflect the health level of residents, such as the average life expectancy, the death rate, the birth rate, the natural population growth rate and the like, and is used for evaluating the health quality and the life quality of the residents. The continuous monitoring of the index variation trend and the distribution of each index in different ages and gender groups can provide quantitative basis for the health policy and population policy of relevant government departments.
2.2.2.1.4 Natural Environment index construction
With the global climate change and the human living activities causing the pollution of air and water resources to be aggravated, the influence of environmental factors such as weather and environmental quality on human health is increasingly emphasized. The research and development content comprehensively analyzes medical big data and environmental protection data of Gansu regions through a big data machine learning method, and researches the correlation between air pollutants (such as particles, ozone, nitrogen dioxide, sulfur dioxide and the like) and typical weather factors (such as temperature, air pressure, wind speed, temperature and the like) and diseases in the environment where people are located, and the urban health index.
Through symptom and environment correlation analysis, the significance of the influence of various environmental factors on the health of the disease is analyzed, through establishing an environmental factor such as PM2.5, PM10, CO, NO2, SO2, air pressure and temperature and a correlation analysis model of the disease (such as respiratory system disease) and exploring population health data and disease control related data in Gansu regions, a prediction early warning index of partial diseases is formed, and the scientific basis of the health early warning index is perfected. And subsequently, according to the change trends of weather, air quality and the like, the prediction early warning model of the respiratory system diseases is used as an incision, the environmental factors of other diseases are gradually increased, the algorithm and the model are further optimized, and the disease prediction early warning model of other related environmental influences is gradually perfected. With the model, residents are reminded to take protective measures in time, and the healthy life of the residents is served.
Through the visual presentation of environmental data dimension, for example, the data visual presentation of PM 2.5's numerical value height under the long-term exposure condition to the danger and the injury degree that chronic disease caused, according to the data trend, improve the environmental quality decision-making, reduce sick probability, propose reasonable suggestion to chronic disease crowd early.
2.2.2.1.5 Gansu health index construction
The urban health index is composed of, in principle, a primary index (urban health index), a secondary index (e.g., disease index), and a tertiary index (e.g., various types of chronic non-infectious diseases). Parameters such as structure (logical, comprehensive, feasibility, validity, algorithm rationality, etc.), execution (data accessibility, whether cooperation/linkage with other units is required, sustainability, etc.), performance (predictable time span, timeliness, accuracy, qualitative/quantitative, etc.), etc. are evaluated and defined as quantitative values.
2.2.2.2 prediction of Gansu health index
The health index and the intelligent forecasting platform aim to be established, an intelligent service system from perception, forecasting and prevention is created, single forecasting indexes aiming at influenza indexes, hand-foot-and-mouth diseases and the like are provided for individual cities in China at present, complex monitoring data are converted into shallow and understandable risk grade modes by public health products, and targeted prevention and control suggestions are attached, so that health knowledge is effectively publicized, and health behaviors of citizens are promoted to be cultured.
The forecasting center of the Gansu health index (PHI) is used for carrying out comprehensive, multi-field and multi-aspect overall evaluation on cities, and comprises indexes and data with different properties, different dimensions and magnitude orders. The method carries out multi-angle analysis and mining on data such as urban disease monitoring, health service monitoring and the like, establishes a Gansu health index reflecting the local overall health condition by utilizing big data analysis and artificial intelligence means, carries out regional disease prediction based on a powerful intelligent engine, and dynamically monitors the urban health appearance and the change trend thereof; and a group prevention and control suggestion is provided to guide residents to prevent diseases, so that government departments are assisted to improve the efficiency in the prevention and control work of related diseases and reduce the prevention and control cost of the diseases.
On a data platform after realizing the data docking of the province, a corresponding group monitoring map is established for each disease. The map is provided with a draggable time axis so as to realize the time-crossing crowd distribution.
The development characteristics of various diseases in multiple dimensions of time and space are displayed on a platform through a data visualization method. The propagation paths of various diseases and the pathogenic accumulation areas on the population level can be observed more intuitively.
2.2.2.3 characterization of Gansu health index
The Gansu health index has the following characteristics:
(1) the urban health portrait is comprehensively drawn by covering all-round data such as chronic diseases, natural environment, life style, health level, health service resources and the like;
(2) the system can be directly butted and applied with a regional health big data platform, and has real-time performance and practicability;
(3) the authority of the index is ensured by integrating the expert opinions in multiple fields of public health, clinical medicine, environmental science and data science;
(4) objective statistical data and expert opinions are fused to determine index weight, and a big data analysis method is adopted to correct the multivariate redundancy among multiple indexes, so that the method has the characteristics of rigor and objectivity;
(5) the urban health index is a developmental index, and indexes can be increased or deleted according to needs, so that the urban health index is advanced with time;
(6) the urban health indexes can be flexibly split and combined, and sub-indexes (second-level indexes and third-level indexes) of each level can also be displayed as independent indexes. Therefore, high risk indexes affecting regional health are monitored in a focused manner while the health overall appearance is displayed.
2.3 prediction of personal diseases
The individual disease prediction system mines disease risk factors from a large amount of medical data and medical documents based on big data, and adopts artificial intelligence and machine learning technology to establish an accurate prediction model. Risk factor analysis is carried out based on big data, expert knowledge is fused, and individualized prevention intervention suggestions are provided for accurate crowds.
The purpose of disease prediction is to predict an individual's future risk of developing certain diseases and to manage the diseases hierarchically according to the risk level. According to the scheme, a prediction target is determined firstly, and then risk prediction is carried out on the basis of population descriptive analysis by adopting different prediction methods, wherein the method comprises the following steps:
(1) method based on traditional regression analysis
Statistical methods such as logistic regression, COX regression, etc. are suitable for finding linear or near linear relationships between variables and targets. A linear prediction model can be constructed. Such models are well interpretable, but the predictive performance is generally modest.
(2) Method based on traditional machine learning
A disease prediction model is established by adopting a single traditional machine learning model such as a decision tree, a support vector machine, naive Bayes and the like, needs prior knowledge and is often good in small sample data performance.
(3) Ensemble learning based method
A plurality of weak classifiers are fused, and a result which is more robust and better than a single classification is obtained through strategies such as voting or averaging, and the like, wherein the results generally comprise two major methods of serial integrated processing Boosting and parallel integrated processing Bagging, and the commonly used methods comprise Random Forest, XGBOST, LightGBM and the like.
The methods can further determine core risk factors in the feature set of the data, screen out core features relevant to the target, eliminate irrelevant or if relevant features, improve the accuracy of risk prediction, ensure the interpretability of a prediction model and provide strong support for healthy hierarchical management.
2.3.1 difficulties in predicting personal diseases
The main difficulties in disease prediction are as follows:
(1) the accuracy of the traditional disease prediction model is to be improved
Traditional disease prediction models are usually built based on a small number of known risk factors, and the modeling method generally adopts a traditional statistical analysis method, so that the prediction accuracy is limited.
(2) Lack of accurate disease prevention decision support by traditional models
At present, decision suggestions for disease prevention are generally extensive, and individual prevention intervention suggestions cannot be given for accurate people generally.
2.3.2 models and assessments of personal disease prediction
The unified data set after data preprocessing can further generate a specific data set through steps of constructing crowds, screening characteristics and the like according to different task requirements, and the specific data set is transmitted to the model through a specified API. Next, the present invention requires evaluation of the results returned by the model on a given data set. The evaluation methods used are also different, since the tasks to which different models are directed are different. The invention mainly introduces a model evaluation method for disease prediction.
2.3.3 disease prognosis/disease diagnosis
For disease prediction/disease diagnosis models, model outcomes can generally be evaluated from three aspects: discrimination, consistency, comparison with existing models, overall performance.
(1) Distinction (distinction)
Discrimination refers to the ability of the model to correctly classify the study object, i.e., the model can correctly distinguish individuals with high probability and low probability of outcome. Common indicators include Sensitivity (Sensitivity), Specificity (Specificity), ROC curve, AUC, etc., and specific descriptions of these indicators are given in the following table.
(2) Consistency (Calibration).
The consistency refers to the consistency degree of the model output probability and the actual event occurrence probability, and the accuracy of evaluating and measuring the model output absolute probability is evaluated. Commonly used indicators include the Home-Lemeshow test, Calibration plot, etc.
(3) Comparison with existing models.
Model comparison is the evaluation of differences in efficacy between the new and old models, and commonly used indices include AUC (area Under cut)
2.3.4 support disease species
The supporting disease species mainly comprise:
(1) cardiovascular and cerebrovascular diseases
Cardiovascular and cerebrovascular diseases are the general term for cardiovascular and cerebrovascular diseases, and refer broadly to ischemic or hemorrhagic diseases of the heart, brain and systemic tissues caused by blood viscosity, atherosclerosis, hypertension, etc. Hypertension, hyperglycemia and hyperlipidemia are typical symptoms, and are developed in middle-aged and elderly people.
(2) Chronic disease
Cardiovascular and cerebrovascular diseases, diabetes and chronic diseases are the most major public health problems in the world to date. The information released by the Wei-Jian-Wei-Jian-Wei disease control department shows that death caused by chronic diseases such as cardiovascular and cerebrovascular diseases and diabetes mellitus are the first four death causes in urban and rural areas of China, and the risk factors related to the chronic diseases commonly exist in people.
The current version shows models of hypertension, diabetes, coronary heart disease, apoplexy and heart age
2.3.5 health intervention
According to the disease risk level given by the prediction model and the corresponding risk factor, the health intervention measures which are evaluated and formulated by a medical expert group are given, the risk of the patient suffering from the disease is reduced through scientific health intervention measures, and the purposes of preventing the disease before and reducing the treatment cost of the subsequent disease are achieved.
2.3.6 individual disease prediction and group risk display
And (3) butting the resident electronic health files and health physical examination data, performing risk prediction by adopting a high-precision chronic disease prediction model, giving the risk of each resident about hypertension, diabetes, coronary heart disease and cerebral apoplexy in five years in the future, giving high, medium and low risk levels, and giving risk details and personalized health management suggestions.
In the above embodiment, fig. 3 is a multidimensional image of the urban health index provided by the present invention;
FIG. 4 is an index of health resources for an index of urban health provided by the present invention;
FIG. 5 is a city health index natural environment index provided by the present invention;
FIG. 6 is a city health index life expectancy index provided by the present invention;
FIG. 7 is a city health index lifestyle index provided by the present invention;
FIG. 8 is a city health index lentigo index provided by the present invention.
It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus and its modules of the present invention may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of hardware circuits and software, e.g., firmware. The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A city health index prediction system, the city health index prediction system comprising:
the health index data cloud platform comprises a data docking module and a data analysis module; the health data acquisition, processing and analysis are carried out;
the urban health index system is used for carrying out comprehensive health condition monitoring and risk prediction in different areas by adopting radar maps, color block tables, trend maps, geographic information systems and other visualization technologies;
the personal health analysis system is used for mining disease risk factors from a large amount of medical data and medical documents based on big data, establishing an accurate prediction model by adopting artificial intelligence and machine learning technology, analyzing the risk factors based on the big data, fusing expert knowledge, and providing individualized prevention and intervention suggestions for accurate people.
2. The urban health index prediction system of claim 1 wherein the health index data cloud platform comprises:
the data docking module is used for acquiring corresponding health data of each health data platform, the diagnosis and treatment system and other prevention and control public health data platforms;
the data analysis module comprises a data integration unit, a data conversion unit, a data cleaning unit and a missing value processing unit; the method is used for integrating, converting, cleaning and processing missing values of the acquired data.
3. The urban health index prediction system of claim 1 wherein the urban health index system comprises:
the medical big data deep mining module comprises a health index modeling unit and a health index prediction health index visualization unit; the method is used for constructing and visualizing the urban health index model;
the urban health index building module comprises a health index system building unit and a health index forecasting unit; the method is used for constructing an urban health index system and carrying out health forecast.
4. The urban health index prediction system of claim 3 wherein the medical big data deep mining module comprises:
the health index modeling unit is used for constructing a health panorama reflecting the region, sensing the current situation in real time based on big data and expressing a health index model of the whole health portrait of the dynamically monitored city;
a health index prediction health index visualization unit; the index model display method is used for displaying the results of the index model in a layering mode by using radar maps, color block tables, trend maps, geographic information systems and other visualization technologies and monitoring the index dynamics of all levels.
5. The urban health index prediction system of claim 3 wherein the urban health index construction module comprises:
the health index system building unit is used for building a health index system comprising a chronic non-infectious disease index, a sanitary resource index, life expectancy, a natural environment index and a regional health index;
and the health index forecasting unit is used for forecasting the urban health based on the health index indexes.
6. The city health index prediction system of claim 1 wherein the integrated health monitoring and risk prediction includes monitoring and prediction of chronic non-infectious diseases, health services indices, natural environment indices.
7. A city health index prediction method applied to the city health index prediction system according to any one of claims 1 to 6, wherein the city health index prediction method comprises:
step one, collecting, processing and analyzing urban and individual health data;
step two, comprehensive health condition monitoring and risk prediction of different areas are carried out by adopting radar maps, color block tables, trend maps, geographic information systems and other visualization technologies;
and step three, excavating disease risk factors from a large amount of medical data and medical documents based on big data, establishing an accurate prediction model by adopting artificial intelligence and machine learning technology, analyzing the risk factors based on the big data, fusing expert knowledge, and providing individualized prevention and intervention suggestions for accurate people.
8. The method for predicting the urban health index as recited in claim 7, wherein in the first step, the processing the collected data comprises:
and (3) carrying out data integration: extracting various types of acquired original data and integrating the data of different data sources according to a certain contact rule to obtain a unified data set comprising basic information, baseline medical history, inspection, diagnosis/prescription, follow-up visit and other various types of data;
and (3) carrying out data conversion: performing data type conversion, data semantic conversion, data value domain conversion, data granularity conversion, table/data splitting, row-column conversion, data discretization, data standardization, new field refinement and other processing on the obtained unified data set;
and (3) data cleaning: determining different cleaning rules according to different data types and error types, and regulating the data format and processing or modifying unreasonable data based on the determined data cleaning rules;
and (3) carrying out missing value processing: filling missing feature values based on rules, filling missing patient data with statistics of other patients, or supplementing missing data using linear regression/machine learning methods.
9. The method of predicting the urban health index as recited in claim 7, wherein the step of building the accurate prediction model by using artificial intelligence and machine learning techniques comprises:
and establishing an accurate prediction model by adopting a method based on traditional regression analysis, traditional machine learning and ensemble learning.
10. A computer-readable storage medium, storing a computer program which, when executed by a processor, causes the processor to carry out the city health index prediction method of claims 7-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110667979.8A CN113539492A (en) | 2021-06-16 | 2021-06-16 | Urban health index prediction system, prediction analysis method and storage medium thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110667979.8A CN113539492A (en) | 2021-06-16 | 2021-06-16 | Urban health index prediction system, prediction analysis method and storage medium thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113539492A true CN113539492A (en) | 2021-10-22 |
Family
ID=78125039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110667979.8A Pending CN113539492A (en) | 2021-06-16 | 2021-06-16 | Urban health index prediction system, prediction analysis method and storage medium thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113539492A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860537A (en) * | 2022-11-30 | 2023-03-28 | 杭州光海科技有限公司 | Campus card based physique test system and method |
| CN116663893A (en) * | 2023-05-16 | 2023-08-29 | 中国疾病预防控制中心环境与健康相关产品安全所 | Comprehensive data processing method and system for environmental health risk assessment |
| CN117976148A (en) * | 2024-04-01 | 2024-05-03 | 成都市双流区妇幼保健院 | Machine learning-based prediction method and system for mental health problems of children |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530175A (en) * | 2016-11-01 | 2017-03-22 | 北京师范大学 | Urban ecosystem health diagnosis system |
| CN107705019A (en) * | 2017-09-30 | 2018-02-16 | 安徽四创电子股份有限公司 | A kind of smart city Huimin service level evaluation method |
| CN109119130A (en) * | 2018-07-11 | 2019-01-01 | 上海夏先机电科技发展有限公司 | A kind of big data based on cloud computing is health management system arranged and method |
| CN111612294A (en) * | 2020-04-08 | 2020-09-01 | 广州市城市规划勘测设计研究院 | City physical examination system based on space-time big data |
| CN112951430A (en) * | 2021-03-29 | 2021-06-11 | 阿斯利康投资(中国)有限公司 | Multiple chronic disease joint management apparatus and computer-readable storage medium |
-
2021
- 2021-06-16 CN CN202110667979.8A patent/CN113539492A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530175A (en) * | 2016-11-01 | 2017-03-22 | 北京师范大学 | Urban ecosystem health diagnosis system |
| CN107705019A (en) * | 2017-09-30 | 2018-02-16 | 安徽四创电子股份有限公司 | A kind of smart city Huimin service level evaluation method |
| CN109119130A (en) * | 2018-07-11 | 2019-01-01 | 上海夏先机电科技发展有限公司 | A kind of big data based on cloud computing is health management system arranged and method |
| CN111612294A (en) * | 2020-04-08 | 2020-09-01 | 广州市城市规划勘测设计研究院 | City physical examination system based on space-time big data |
| CN112951430A (en) * | 2021-03-29 | 2021-06-11 | 阿斯利康投资(中国)有限公司 | Multiple chronic disease joint management apparatus and computer-readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 殷利霞 等: "智慧居民健康档案研究", 《中国卫生信息管理杂志》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860537A (en) * | 2022-11-30 | 2023-03-28 | 杭州光海科技有限公司 | Campus card based physique test system and method |
| CN115860537B (en) * | 2022-11-30 | 2023-08-22 | 杭州光海科技有限公司 | Campus card-based physique testing system and method |
| CN116663893A (en) * | 2023-05-16 | 2023-08-29 | 中国疾病预防控制中心环境与健康相关产品安全所 | Comprehensive data processing method and system for environmental health risk assessment |
| CN117976148A (en) * | 2024-04-01 | 2024-05-03 | 成都市双流区妇幼保健院 | Machine learning-based prediction method and system for mental health problems of children |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Galetsi et al. | Values, challenges and future directions of big data analytics in healthcare: A systematic review | |
| CN113539492A (en) | Urban health index prediction system, prediction analysis method and storage medium thereof | |
| CN112133445A (en) | Cardiovascular disease management service method and system | |
| JP7430295B2 (en) | Individual chronic disease progression risk visualization evaluation method and system | |
| Lin et al. | Abnormal diagnosis of Emergency Department triage explored with data mining technology: An Emergency Department at a Medical Center in Taiwan taken as an example | |
| US20210280320A1 (en) | Community Vulnerability Index Dashboard | |
| CN113345583A (en) | Method and system for constructing full life cycle resident intelligent health file | |
| Sun et al. | Applying machine learning algorithms to electronic health records to predict pneumonia after respiratory tract infection | |
| Jiang et al. | A systematic review of the modelling of patient arrivals in emergency departments | |
| TW201837745A (en) | Cross-platform analyzing and display system of clinical data | |
| Bunaciu et al. | Measuring recovery capital for people recovering from alcohol and drug addiction: a systematic review | |
| Campbell et al. | Do scores on the HCR-20 and FAM predict frequency of self-harm in females within a secure psychiatric hospital? | |
| Posey et al. | A fatal review: Exploring how children’s deaths are reported in the United States | |
| Demir et al. | Enabling better management of patients: discrete event simulation combined with the STAR approach | |
| Chen et al. | Evaluating the risk of hypertension in residents in primary care in Shanghai, China with machine learning algorithms | |
| Nates et al. | Cross-validation of a modified score to predict mortality in cancer patients admitted to the intensive care unit | |
| Treasury | The Magenta Book: guidance notes for policy evaluation and analysis | |
| CN114664447A (en) | Osteoporosis health risk assessment method and equipment | |
| Schuurman | Metadata as a site for imbuing GIS with qualitative information | |
| Chatterjee et al. | Kalyani cohort–the first platform in Eastern India for longitudinal studies on health and disease parameters in peri-urban setting | |
| Hu et al. | A retrospective study of suicide attempts in patients with pulmonary hypertension | |
| Jiang | RETRACTED: Design of enterprise employee pension platform based on complex embedded system | |
| Alam | Identification of malignant mesothelioma risk factors through association rule mining | |
| Simon | Covid-19 data in historical perspective: Lessons from Chicago’s 1995 heat wave | |
| Camacho et al. | Low-cost predictive models of dementia risk using machine learning and exposome predictors |
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 |