CN114969820A - Management platform based on occupational health medical big data - Google Patents
Management platform based on occupational health medical big data Download PDFInfo
- Publication number
- CN114969820A CN114969820A CN202210620754.1A CN202210620754A CN114969820A CN 114969820 A CN114969820 A CN 114969820A CN 202210620754 A CN202210620754 A CN 202210620754A CN 114969820 A CN114969820 A CN 114969820A
- Authority
- CN
- China
- Prior art keywords
- occupational health
- model
- cloud
- occupational
- big data
- 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
- 230000036541 health Effects 0.000 title claims abstract description 156
- 238000007726 management method Methods 0.000 claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 29
- 238000012544 monitoring process Methods 0.000 claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 230000006870 function Effects 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000005457 optimization Methods 0.000 claims abstract description 14
- 238000012800 visualization Methods 0.000 claims abstract description 12
- 238000012423 maintenance Methods 0.000 claims abstract description 10
- 230000005540 biological transmission Effects 0.000 claims abstract description 6
- 238000010276 construction Methods 0.000 claims description 31
- 238000000034 method Methods 0.000 claims description 14
- 208000028571 Occupational disease Diseases 0.000 claims description 12
- 230000010354 integration Effects 0.000 claims description 10
- 238000013473 artificial intelligence Methods 0.000 claims description 9
- 230000005802 health problem Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 201000010099 disease Diseases 0.000 claims description 6
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 6
- 238000013439 planning Methods 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000004962 physiological condition Effects 0.000 claims description 3
- 238000013468 resource allocation Methods 0.000 claims description 3
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 230000000007 visual effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 9
- 238000007405 data analysis Methods 0.000 abstract description 8
- 238000003860 storage Methods 0.000 abstract description 6
- 238000013500 data storage Methods 0.000 abstract description 5
- 238000013523 data management Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000002265 prevention Effects 0.000 description 2
- 206010008428 Chemical poisoning Diseases 0.000 description 1
- 208000035473 Communicable disease Diseases 0.000 description 1
- 241000282414 Homo sapiens Species 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 208000025157 Oral disease Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004141 dimensional analysis Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 208000030194 mouth disease Diseases 0.000 description 1
- 206010035653 pneumoconiosis Diseases 0.000 description 1
- 230000002285 radioactive effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 208000023504 respiratory system disease Diseases 0.000 description 1
- 208000017520 skin disease Diseases 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/008—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols involving homomorphic encryption
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Bioethics (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Computer Hardware Design (AREA)
- Public Health (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Educational Administration (AREA)
- Epidemiology (AREA)
- Computing Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a management platform based on occupational health medical treatment big data, and belongs to the technical field of occupational health data management. According to the invention, a hybrid cloud building mode is adopted, and the private clouds are connected to the same network through the central server, so that the problem of single and insufficient data source is solved; a federal learning mode is adopted, model training is carried out through parameter transmission, and privacy information is protected through a differential privacy, homomorphic encryption and cooperative training mode; by adopting a mode of deploying public cloud Hadoop big data clusters, functions of private cloud user non-sensitive data storage, data analysis and the like can be realized, and the pressure of user storage and calculation is relieved; intelligent analysis and establishment of a model algorithm library can be realized through the acquired multi-source data; and based on different applications and visualization requirements in multiple scenes, an application module is constructed, and an application layer function module comprising a basic function, monitoring early warning, maintenance optimization, special analysis and presentation is provided.
Description
Technical Field
The invention relates to the technical field of occupational health data management, in particular to a management platform based on occupational health medical treatment big data.
Background
The main work content of occupational health is to study the health and health problems of human beings in various occupational labories, and aims to protect the health of workers from harmful factors in the occupational activities, including the influence of the working environment on the health of workers and countermeasures for preventing occupational hazards.
More importantly, acquisition, storage, management, calculation and analysis of occupational health data are not developed along with rapid development of internet, cloud calculation, big data and artificial intelligence technologies. Some scholars count the professional health information systems of the national level, the provincial level and part of central enterprises in 2020, and 57 professional health information systems of various types comprise 4 national level systems, 44 provincial level systems and 9 industry systems. Therefore, the problems of scattered construction, respective construction and the like exist in the informatization construction of the occupational health in China, and effective data communication sharing, data analysis and mining, intelligent calculation and the like are lacked among related systems. In addition, in the aspect of privacy protection, occupational health data relate to user sensitive information, and a general occupational health management platform may cause the risk of sensitive information leakage; in the aspect of big data medical information processing, the problems of poor performance, high cost, difficult capacity expansion and the like exist; in the aspects of model algorithm and application planning, multi-party occupational health data are not fully utilized, and a complete occupational health model algorithm library is not constructed, so that the efficiency is not high, and the error rate is large. Nor is there an application layer complete functional programming design.
The traditional occupational health data management platform has the following defects: 1) the multi-party data cannot be fully utilized. Information islands are caused by scattered construction; 2) the privacy disclosure problem may occur in each link which is difficult to avoid due to potential safety hazards of user sensitive information; 3) Most of the data analysis and visualization are completed by adopting a relational database and a self-built machine room deployment application server, and under the condition of large data set analysis, the performance may not meet the calculation requirement, so that the construction cost is further increased and the capacity expansion is inconvenient. 4) And various data sources cannot be integrated to perform model modeling optimization. 5) And a uniformly optimized professional health algorithm model library cannot be built. 6) The application layer cannot be completely designed for function planning. Therefore, a management platform based on occupational health medical treatment big data is provided.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: how to solve the above-mentioned problem that traditional occupational health data management platform exists provides a management platform based on occupational health medical treatment big data.
The invention solves the technical problems through the following technical scheme, and the invention comprises the following steps: the system comprises a hybrid cloud construction module, a privacy protection module, a federal learning module, an algorithm model base module and an application module;
the hybrid cloud construction module is used for integrating the public cloud and the private cloud to construct a hybrid cloud;
the privacy protection module is used for introducing various privacy protection schemes;
the federal learning module is used for realizing the training of the occupational health model by federal learning;
the algorithm model library module is used for providing an occupational health monitoring model, an occupational health prediction model, an occupational health comprehensive scoring model, an occupational health warning line identification model and an artificial intelligence algorithm library;
and the application module is used for realizing occupational health visualization in various forms by means of multi-source data modeling analysis.
Furthermore, the hybrid cloud construction module comprises a private cloud integration unit, a public cloud construction unit and a hybrid cloud network construction unit; the system comprises a private cloud integration unit, a public cloud construction unit, a hybrid cloud network construction unit and a public cloud management unit, wherein the private cloud integration unit is used for constructing a data requirement and planning and selecting a data party private cloud according to a model, then determining data resources, transmission modes and model parameter information required by each model, the public cloud construction unit is used for constructing a public cloud platform so as to realize non-private data backup of each private cloud data party, analysis based on a Hadoop big data platform and model parameter tuning optimization based on federal learning, and the hybrid cloud network construction unit is used for making network connection between the private cloud and the public cloud and constructing connection between a plurality of private clouds and the public cloud so as to realize functions of private cloud non-private data backup, private cloud Hadoop tenants, resource allocation, private cloud customized computing and joint model federal learning; the public cloud is the central server.
Furthermore, the privacy protection module comprises a differential privacy unit, a homomorphic encryption unit and a cooperative training unit; the differential privacy unit is used for adding noise in the parameters through a differential privacy protection random mechanism before the parameters shared by the private cloud participants are sent to the central server, so that malicious participants cannot use the parameters of the shared global model to deduce the information of other private cloud participants; the collaborative training unit is used for enabling the private cloud participants not to upload the complete occupational health parameter set generated after local training to the central server, updating the whole occupational health global model to the local, but selectively uploading and downloading, and determining the number of shared occupational health parameters according to conditions.
Further, in the federal learning module, the process of federal learning includes the steps of:
s1: the central server generates a public key and a private key, and issues the public key to the private cloud participants to download the latest model from the central server respectively;
s2: each private cloud participant trains a model by using local data, encrypts gradient and loss and uploads the gradient and loss to the central server, and the central server is used for aggregating gradient update model parameters of each user;
s3: the central server returns the updated model to each private cloud participant;
s4: each private cloud participant updates its respective model.
Furthermore, the occupational health monitoring model is used for monitoring the occupational health of the user through regular working environment assessment, physiological index monitoring and health condition prediction for the crowd with occupational disease risks, and reporting the conditions of abnormal relevant indexes to supervision departments, factories and individuals.
Furthermore, the occupational health prediction model is used for collecting information such as age, gender, BMI, occupation, industry attributes, occupational health risk level, working age, occupational disease type, current illness state risk level and the like through integration of multi-source data and by utilizing historical occupational health data, establishing the occupational health prediction model and achieving prediction of occupational health problems.
Furthermore, the comprehensive occupational health scoring model is used for carrying out comprehensive scoring on the occupational health of the individual according to the possibility of occupational disease risk and the physiological condition, and scoring results are used for showing the overall occupational health risk condition of the individual, a factory and an area.
Furthermore, the occupational health warning line identification model is used for establishing an occupational health rating standard and identifying occupational health standard systems of safety, low risk, medium risk, high risk and the like.
Furthermore, the artificial intelligence algorithm library is used for customizing the analysis model according to the actual situation of the private cloud participant, and the respective analysis modeling requirements are met.
Furthermore, the application module comprises a basic function unit, a monitoring early warning unit, a maintenance optimization unit, a special analysis unit and a presentation unit; the basic function unit is used for evaluating the occupational health degree of a user and tracing and positioning the reasons of low health degree to realize visualization of the occupational health problems of the basic function, and the monitoring and early warning unit is used for early warning the crowd with risks through an occupational health monitoring model and an occupational health warning line identification model and monitoring and supervising the factory with potential risks; the system maintenance optimizing unit is used for tracking user occupational health and optimizing occupational health, the thematic analysis is used for analyzing the professional disease thematic and the administrative region health degree thematic, and the presenting unit is used for visually displaying occupational health through a visual way and a terminal display mode.
Compared with the prior art, the invention has the following advantages: the management platform based on the occupational health medical big data adopts a hybrid cloud building mode, and all private clouds are connected to the same network through the central server, so that the problem of single and insufficient data source is solved; a federal learning mode is adopted, model training is carried out through parameter transmission, and privacy information is protected through a differential privacy, homomorphic encryption and cooperative training mode; by adopting a mode of deploying public cloud Hadoop big data clusters, functions of private cloud user non-sensitive data storage, data analysis and the like can be realized, and the pressure of user storage and calculation is relieved; intelligent analysis and establishment of a model algorithm library can be realized through the acquired multi-source data; and based on different applications and visualization requirements in multiple scenes, an application module is constructed, and an application layer function module comprising a basic function, monitoring early warning, maintenance optimization, special analysis and presentation is provided.
Drawings
FIG. 1 is a schematic overall architecture diagram of an intelligent management platform based on occupational health and medical care big data according to an embodiment of the invention;
FIG. 2 is a flowchart of an implementation of an intelligent management platform based on occupational health and medical care big data according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a hybrid cloud construction process in the embodiment of the present invention.
Fig. 4 is a schematic diagram of a federal learning system architecture in an embodiment of the present invention.
Detailed Description
The following examples are given for the detailed implementation and specific operation of the present invention, but the scope of the present invention is not limited to the following examples.
As shown in fig. 1, the present embodiment provides a technical solution: an intelligent management platform based on occupational health medical big data mainly comprises a hybrid cloud construction module, a privacy protection module, a federal learning module, an algorithm model library module and an application module. Fig. 2 is a flowchart illustrating implementation of each module in the intelligent management platform. The invention aims to construct an intelligent management platform for occupational health medical treatment big data, which is safe, interconnected, efficient, low in cost, multi-model, high in precision and wide in application, by utilizing technologies such as federal learning, mixed cloud, Hadoop cluster and artificial intelligence algorithm.
The following describes the modules of the present invention:
1) hybrid cloud construction module
Most of the existing occupational health medical treatment big data platforms are constructed dispersedly, each data party maintains own private data, information interaction cannot be carried out among all the platforms, and a plurality of information isolated islands are formed. In addition, with the advent of the big data age, data has grown exponentially. On one hand, in the face of high-speed increase of data, capacity expansion needs to be carried out on hardware in time, and the data storage needs are met. On the other hand, the potential value of the big data needs to be mined by means of big data analysis, and the statistics, analysis and prediction capabilities of the medical data under multiple scenes are realized, so that the medical service level is improved. In addition, the private data source is single, and the generalization capability of the model result is insufficient, so that the optimization of the model parameters by means of external data is required. In summary, the public cloud and the private cloud can be integrated to construct a hybrid cloud, so as to solve the above problems, the hybrid cloud construction process is shown in fig. 3, and the specific process is as follows:
integrating private clouds
The construction of an algorithm model base of the occupational health medical treatment big data intelligent management platform depends on data and parameters provided by each data party to carry out model training, so that data integration needs to be carried out on data private parties participating in the model training, and a data party private cloud is constructed. The primary data parties include various government regulatory agencies, hospitals, physical examination centers, enterprises, factories, and the like. And selecting a private cloud of a data party according to the data requirements of model construction and later planning, and then determining information details such as data resources, transmission modes, model parameters and the like required by each model. In addition, the non-sensitive information needing backup and analysis is determined and periodically uploaded to a large data platform of the central server, so that the basic data storage and data analysis capacity is realized, and the local private cloud storage and calculation pressure is relieved.
② building up public cloud
The public cloud is the central server in fig. 3, and mainly plays roles of data backup, intelligent computing and federal learning. The public cloud is convenient to expand, low in storage cost and good in platform integration performance, and unified management and safety protection are convenient to carry out on the whole data, so that the public cloud platform is built to realize non-private data backup of each private cloud data side, intelligent analysis based on a Hadoop big data platform and model parameter adjustment and optimization based on federal learning.
Construction of hybrid cloud network
And (3) breaking through network connection between the private cloud and the public cloud, and constructing connection between a plurality of private clouds and the public cloud, thereby realizing the functions of private cloud non-private data backup, private cloud Hadoop tenants, resource allocation, private cloud customized intelligent computing and joint model federal learning.
2) Privacy protection module
Because occupational health data relate to user privacy information, based on the characteristics of federal learning, private cloud users only share parameter models but not share data, and therefore the data information of the private cloud users is guaranteed not to be leaked. However, it is still possible to infer information about the user's gender, occupation, geographic location, etc. from the model parameters of the private cloud user. Therefore, in order to prevent the private cloud user from maliciously backstepping other user sensitive information through shared parameters, various privacy protection technologies are introduced into the intelligent management platform based on occupational health medical big data, and occupational health data are prevented from being leaked.
Specifically, various privacy protection techniques are included:
difference privacy
Differential privacy based solutions. The method mainly aims at the fact that a malicious private cloud data party wants to deduce the privacy information of other participants, noise is added to parameters through a differential privacy protection random mechanism such as a Gaussian mechanism before the parameters shared by the private cloud participants are sent to a central server, and therefore the malicious participants cannot use the parameters of a shared global model to deduce the information of other private cloud participants.
② homomorphic encryption
A solution based on homomorphic encryption. The method encrypts the training parameters of the private cloud participants using a homomorphic encryption technique before sending them to the central server. The occupational health privacy of the user can be effectively guaranteed not to be revealed while data calculation and model parameter prediction are not influenced.
③ cooperative training
Based on a collaborative training solution. The private cloud participants selectively upload and download the shared occupational health parameter set according to conditions without uploading the occupational health complete parameter set generated after local training to a central server or updating the whole occupational health global model to the local.
3) Federal learning module
The model precision is not influenced while the occupational health multi-source data user privacy is ensured. And (4) realizing training of the occupational health model by adopting federal learning. Fig. 4 shows a federal learning system architecture diagram of an intelligent management platform based on occupational health and medical care big data, which specifically includes the following processes:
firstly, a central server generates a public key and a private key, and issues the public key to private cloud participants to download the latest model from the central server respectively;
secondly, each private cloud participant trains a model by using local data, the encrypted gradient and loss are uploaded to a central server, and the central server aggregates gradient update model parameters of each user;
thirdly, the central server returns the updated model to each private cloud participant;
and fourthly, updating the respective model by each private cloud participant.
Through the federal learning training, the occupational health model can be trained by utilizing multi-source data while privacy is protected, so that the occupational health model is output, and an algorithm model base is constructed.
4) Algorithm model library module
The algorithm model library module is used for providing the following models or algorithm libraries to complete corresponding functions;
occupational health monitoring model
The occupational health monitoring model realizes monitoring of the occupational health of users mainly through regular work environment assessment, physiological index monitoring and health condition prediction for people with occupational disease risks, reports the conditions of abnormal relevant indexes to supervision departments, factories and individuals in time, and can play roles in early supervision, treatment and prevention respectively.
Occupational health prediction model
The occupational health prediction model is mainly used for establishing the occupational health prediction model by integrating multi-source data and collecting information such as age, gender, BMI, occupational, industry attributes, occupational health risk levels, working years, occupational disease types and current disease risk levels by means of historical occupational health data, so that accurate prediction of occupational health problems is achieved, advance prevention is achieved, and occupational health risks brought in the working process of worker groups are reduced.
Comprehensive scoring model for occupational health
The comprehensive occupational health scoring model scores the individuals comprehensively according to factors such as the possibility of occupational disease risks, physiological conditions and the like, and the scoring results can clearly show the overall occupational health risk conditions of the individuals, factories and areas. The system can be used for checking reasons of individuals with low comprehensive occupational health scores, checking and correcting problems of factories with low comprehensive occupational health scores, and performing overall check on supervision departments or regional enterprises in regions with low comprehensive occupational health scores. Thereby improving the level of the user's occupational health from multiple aspects.
Occupational health warning line recognition model
The method comprises the steps of establishing a occupational health rating standard, identifying occupational health standard systems such as safety, low risk, medium risk and high risk, carrying out early warning on crowds with high and high vigilance in time, tracing reasons, and reducing risks of occupational diseases or aggravation of diseases.
Artificial intelligence algorithm library
The artificial intelligence algorithm library aims at establishing four algorithm models except the four algorithm models, and can customize analysis models according to the actual conditions of private cloud participants so as to meet the respective analysis modeling requirements. And further expanding an artificial intelligence algorithm library for occupational health by taking the model as a base point according to a valuable model.
5) Application module
The intelligent management platform based on occupational health medical big data relies on multi-source data modeling analysis to provide functions including a basic function, monitoring early warning, maintenance optimization, thematic analysis and presentation module, and the occupational health visualization function is realized in various forms.
The concrete description is as follows:
basic function
The basic function module comprises user occupational health degree evaluation, low health degree reason tracing positioning and the like, and visualization of basic function occupational health problems is achieved.
② monitoring and early warning
And the health early warning module comprises user occupational health monitoring, user occupational health early warning and the like. The method mainly comprises the steps of carrying out timely early warning on people with risks through an occupational health monitoring model and an occupational health warning line identification model, and monitoring and supervising factories with potential risks.
(iii) maintenance optimization
And the maintenance optimization module comprises user occupational health tracking, occupational health optimization and the like. The private cloud data party provides humanistic care and high-quality customer service for employees, patients and the like, pays attention to the professional health of customers, and pays attention to the employees or the customers through multiple aspects such as visiting, inquiring and the like so as to improve the overall level of the professional health and enhance the attribution of the employees and the customer stickiness of medical institutions.
Analysis of special subject
The special topic analysis module comprises professional disease special topics, administrative region health special topics and the like. The statistical analysis is carried out from the whole and parts by classifying occupational pneumoconiosis and other respiratory diseases, occupational skin diseases, occupational ophthalmopathy, occupational otorhinolaryngological oral diseases, occupational chemical poisoning, occupational diseases caused by physical factors, occupational radioactive diseases, occupational infectious diseases, occupational tumors and other occupational diseases. And by combining GIS information, multi-dimensional visualization is realized according to administrative regions, and the capacity of regional drilling and multi-dimensional analysis is met.
Is presented
And the presentation module comprises visualization ways such as tables, graphs and maps and terminal presentation modes such as PC, APP, large screen and WeChat. Therefore, a multi-scene display occupational health visualization platform is realized, and the presentation module effectively supports the validity verification and demonstration application popularization of core technology and theoretical feasibility.
To sum up, the intelligent management platform based on the occupational health medical big data of the embodiment adopts a hybrid cloud building mode aiming at the problems that multi-party data cannot be fully utilized and the construction is dispersed, and the private clouds are connected to the same network through the central server, so that the problem of single and insufficient data source is solved; aiming at the problems that potential safety hazards exist in sensitive information of users and privacy leakage is possible in each link difficult to avoid, a Federal learning mode is adopted, model training is carried out through parameter transmission, and privacy information is protected through a differential privacy, homomorphic encryption and cooperative training mode; aiming at the conditions of insufficient resources and inconvenient capacity expansion under the condition of private cloud user big data analysis, the method for deploying the public cloud Hadoop big data cluster is adopted, so that the functions of private cloud user non-sensitive data storage, data analysis and the like can be realized, and the pressure of user storage and calculation is relieved; aiming at the problem that a traditional occupational health platform cannot effectively introduce multi-party data to establish and optimize an algorithm model base, an intelligent management platform based on occupational medical treatment big data can realize intelligent analysis and establishment of the model algorithm base through acquired multi-source data; and based on different applications and visualization requirements in multiple scenes, an application module is constructed, and an application layer function module comprising a basic function, monitoring early warning, maintenance optimization, special analysis and presentation is provided.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A management platform based on occupational health medical big data is characterized by comprising: the system comprises a hybrid cloud construction module, a privacy protection module, a federal learning module, an algorithm model base module and an application module;
the hybrid cloud construction module is used for integrating the public cloud and the private cloud to construct a hybrid cloud;
the privacy protection module is used for introducing various privacy protection schemes;
the federal learning module is used for realizing the training of the occupational health model by federal learning;
the algorithm model library module is used for providing an occupational health monitoring model, an occupational health prediction model, an occupational health comprehensive scoring model, an occupational health warning line identification model and an artificial intelligence algorithm library;
and the application module is used for realizing occupational health visualization in various forms by means of multi-source data modeling analysis.
2. The occupational health medical big data based management platform according to claim 1, wherein the occupational health medical big data comprises: the hybrid cloud construction module comprises a private cloud integration unit, a public cloud construction unit and a hybrid cloud network construction unit; the system comprises a private cloud integration unit, a public cloud construction unit, a hybrid cloud network construction unit and a public cloud management unit, wherein the private cloud integration unit is used for constructing a data requirement and planning and selecting a data party private cloud according to a model, then determining data resources, transmission modes and model parameter information required by each model, the public cloud construction unit is used for constructing a public cloud platform so as to realize non-private data backup of each private cloud data party, analysis based on a Hadoop big data platform and model parameter tuning optimization based on federal learning, and the hybrid cloud network construction unit is used for making network connection between the private cloud and the public cloud and constructing connection between a plurality of private clouds and the public cloud so as to realize functions of private cloud non-private data backup, private cloud Hadoop tenants, resource allocation, private cloud customized computing and joint model federal learning; the public cloud is the central server.
3. The occupational health medical big data based management platform according to claim 2, wherein the occupational health medical big data comprises: the privacy protection module comprises a differential privacy unit, a homomorphic encryption unit and a cooperative training unit; the differential privacy unit is used for adding noise in the parameters through a differential privacy protection random mechanism before the parameters shared by the private cloud participants are sent to the central server, so that malicious participants cannot use the parameters of the shared global model to deduce the information of other private cloud participants; the collaborative training unit is used for enabling the private cloud participants not to upload the complete occupational health parameter set generated after local training to the central server, updating the whole occupational health global model to the local, but selectively uploading and downloading, and determining the number of shared occupational health parameters according to conditions.
4. The occupational health medical big data based management platform according to claim 3, wherein the occupational health medical big data comprises: in the federal learning module, the process of federal learning includes the following steps:
s1: the central server generates a public key and a private key, and issues the public key to the private cloud participants to download the latest model from the central server respectively;
s2: each private cloud participant trains a model by using local data, encrypts gradient and loss and uploads the gradient and loss to the central server, and the central server is used for aggregating gradient update model parameters of each user;
s3: the central server returns the updated model to each private cloud participant;
s4: each private cloud participant updates its respective model.
5. The occupational health medical big data based management platform according to claim 4, wherein the occupational health medical big data comprises: the occupational health monitoring model is used for monitoring the occupational health of users through regular work environment assessment, physiological index monitoring and health condition prediction for people with occupational disease risks, and reporting the conditions of abnormal relevant indexes to supervision departments, factories and individuals.
6. The occupational health medical big data based management platform according to claim 5, wherein the occupational health medical big data comprises: the occupational health prediction model is used for collecting information such as age, gender, BMI, occupation, industry attributes, occupational health risk level, working age, occupational disease type, current illness state risk level and the like through integration of multi-source data and by utilizing historical occupational health data, establishing the occupational health prediction model and achieving prediction of occupational health problems.
7. The occupational health medical big data based management platform according to claim 6, wherein the occupational health medical big data comprises: the comprehensive occupational health scoring model is used for carrying out comprehensive scoring on the occupational health of the individual according to the possibility that occupational disease risks exist and the physiological conditions, and scoring results are used for showing the overall occupational health risk conditions of the individual, a factory and a region.
8. The occupational health medical big data based management platform according to claim 7, wherein: the occupational health warning line identification model is used for establishing an occupational health rating standard and identifying occupational health standard systems such as safety, low risk, medium risk and high risk.
9. The occupational health medical big data based management platform according to claim 8, wherein: the artificial intelligence algorithm library is used for customizing an analysis model according to the actual situation of the private cloud participant, and the respective analysis modeling requirements are met.
10. The occupational health medical big data based management platform according to claim 9, wherein: the application module comprises a basic function unit, a monitoring early warning unit, a maintenance optimization unit and a thematic analysis and presentation unit; the basic function unit is used for evaluating the occupational health degree of a user and tracing and positioning the reasons of low health degree to realize visualization of the occupational health problems of the basic function, and the monitoring and early warning unit is used for early warning the crowd with risks through an occupational health monitoring model and an occupational health warning line identification model and monitoring and supervising the factory with potential risks; the system maintenance optimizing unit is used for tracking user occupational health and optimizing occupational health, the thematic analysis is used for analyzing the professional disease thematic and the administrative region health degree thematic, and the presenting unit is used for visually displaying occupational health through a visual way and a terminal display mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210620754.1A CN114969820A (en) | 2022-06-01 | 2022-06-01 | Management platform based on occupational health medical big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210620754.1A CN114969820A (en) | 2022-06-01 | 2022-06-01 | Management platform based on occupational health medical big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114969820A true CN114969820A (en) | 2022-08-30 |
Family
ID=82958740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210620754.1A Pending CN114969820A (en) | 2022-06-01 | 2022-06-01 | Management platform based on occupational health medical big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114969820A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116362627A (en) * | 2023-06-01 | 2023-06-30 | 工福(北京)科技发展有限公司 | Staff integrated chemotherapy rest information analysis method of improved KCF algorithm model |
-
2022
- 2022-06-01 CN CN202210620754.1A patent/CN114969820A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116362627A (en) * | 2023-06-01 | 2023-06-30 | 工福(北京)科技发展有限公司 | Staff integrated chemotherapy rest information analysis method of improved KCF algorithm model |
| CN116362627B (en) * | 2023-06-01 | 2023-08-04 | 工福(北京)科技发展有限公司 | Staff integrated chemotherapy rest information analysis method of improved KCF algorithm model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104809933B (en) | A kind of power grid is without script emergency drilling system, drilling method and equipment | |
| CN105069025A (en) | Intelligent aggregation visualization and management and control system for big data | |
| CN114598539B (en) | Root cause positioning method and device, storage medium and electronic equipment | |
| CN112712903A (en) | Infectious disease monitoring method based on human-computer three-dimensional cooperative sensing | |
| CN107016457A (en) | One kind realizes community's hazardous act pre-warning system and method | |
| CN111370133A (en) | Medical data management platform based on block chain | |
| CN113488155A (en) | Refined management system based on hospital operation model | |
| CN114969820A (en) | Management platform based on occupational health medical big data | |
| CN116050836A (en) | Accident prevention service datamation construction method and system based on double prevention mechanism | |
| Lixin et al. | Analysis of social vulnerability of residential community to hazards in Tianjin, China | |
| CN107577769A (en) | A kind of method for digging and system for measuring expert data | |
| Higgins et al. | The evolution of geographical information systems for fire prevention support | |
| CN114650179A (en) | Risk data monitoring method, device and system, electronic equipment and storage medium | |
| Taub et al. | The synthetic data challenge | |
| CN107403257A (en) | One kind production basic data index analysing system | |
| Bashir et al. | MedCloud: cloud-based disease surveillance and information management system | |
| CN115147086A (en) | Monitoring and early warning platform system and method for salary payment of agricultural workers | |
| CN114118696A (en) | Transportation index management system based on space dimensionality | |
| Jing et al. | [Retracted] The Construction and Development of App Application Platform for Public Information Products of Urban Grand Media in the Context of Artificial Intelligence | |
| Yin et al. | Application of key technologies of distributed storage based on the internet of things in urban fire protection | |
| Arai et al. | Building a data ecosystem for using telecom data to inform the COVID-19 response efforts | |
| Garg et al. | Smart Infrastructure: Role of Information Technology in Smart Cities | |
| Chen et al. | Visual method of analyzing COVID-19 case information using spatio-temporal objects with multi-granularity | |
| Lei | Emergency information communication structure by using multimodel fusion and artificial intelligence algorithm | |
| Shoghri et al. | How disease spread dynamics evolve over time |
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 |