CN108717875A - A kind of chronic disease intelligent management system based on big data - Google Patents
A kind of chronic disease intelligent management system based on big data Download PDFInfo
- Publication number
- CN108717875A CN108717875A CN201810481954.7A CN201810481954A CN108717875A CN 108717875 A CN108717875 A CN 108717875A CN 201810481954 A CN201810481954 A CN 201810481954A CN 108717875 A CN108717875 A CN 108717875A
- Authority
- CN
- China
- Prior art keywords
- data
- chronic disease
- user
- module
- analysis result
- 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
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
- G16H50/70—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
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a kind of chronic disease intelligent management system based on big data.Including intelligent wearable device：The analysis result of vital sign data and reception network transmission module passback for acquiring user；Network transmission module：Data analysis module for the collected vital sign data of intelligent wearable device to be sent to high in the clouds, and the analysis result of data analysis module is passed back into the intelligent wearable device that user dresses；Data analysis module：It is matched to obtain analysis result with the vital sign data that network transmission module is sent for establishing pathological model, and by pathological model；Big data memory module：Vital sign data for storing pathological model, acquisition and analysis result.The present invention have the characteristics that can to alleviate disease on outpatients mental state pressure, provide a good decision-making foundation to the diagnosis of patient for doctor and diagnostic result is accurate and reliable.
Description
Technical field
The present invention relates to medical information field, especially a kind of chronic disease intelligent management system based on big data.
Background technology
It is quick with internet, mobile Internet, cloud computing, the rapid rising of Internet of Things and mobile intelligent terminal
Development, explosive growth is presented in chronic disease data, and the content of data is more and more abundant, and relationship becomes increasingly complex, update
Speed is getting faster.The digital research report that International Data Corporation (IDC) provides shows that only 2011 whole world creates and the data of duplication
Amount is more than 1.8ZB, if data growth trend follows new Moore's Law, i.e., double about every two years for global metadata, then the year two thousand twenty
Data volume will be up to 30ZB, and single computer performance can not support the storage and analysis to such huge data.
According to statistics, Chinese chronic disease number of patients has surpassed mostly 300,000,000, and the trend to double with every ten years at present
Developing.In addition, rejuvenation is increasingly presented in chronic disease now, patient groups are distributed in the above crowd of threescore since most
Start invasion 40 till now and arrives quinquagenary crowd.WHO data show that chronic disease makes 38,000,000 people lose life every year, complete
It is dead caused by chronic disease to account for the 60% of all death within the scope of ball, wherein 80% chronic disease death by accident is happened at low receipts
Enter with middle income country (including China), these national chronic disease death tolls account for global chronic disease death toll four/
Three (about 28,000,000 people).In China, chronic disease death toll accounts for about total death toll 85%, and chronic disease has become Chinese residence
The main reason for people are dead.Most of chronic disease can all have certain signs, such as heart murmur or blood before morbidity
The variation of the data such as pressure is analyzed according to data variation rule and just can if these data can be monitored in real time
Then the disease for having found that it is likely that appearance in time takes corresponding measure that can effectively reduce and certain chronic disease probability occurs, or
Some chronic diseases are enable effectively to be cured.
Definition according to MBA think tanks to chronic disease, chronic disease refer to long-term, be unable to self-healing, can hardly also be controlled
More disease.So the treatment of chronic disease is a long-term process, and chronic disease patient is gone due to disease before meeting periodically
Hospital carries out the detection of various indexs.And China or even the medical resource in the whole world all compare shortage relatively, a doctor is every
Balance is required to 10 multidigit patient's interrogations, this allows for many chronics and goes to also need to while enduring disease and tormenting
The problem of in face of the difficulty of getting medical service.Meanwhile hospital admission often being gone to will also result in prodigious burden to the psychology of patient.
Invention content
The object of the present invention is to provide a kind of chronic disease intelligent management system based on big data.The present invention has
Can alleviate disease on outpatients mental state pressure, provide a good decision-making foundation to the diagnosis of patient for doctor and examine
The accurate and reliable feature of disconnected result.
Technical scheme of the present invention：A kind of chronic disease intelligent management system based on big data, including
Intelligent wearable device：The analysis knot of vital sign data and reception network transmission module passback for acquiring user
Fruit；
Network transmission module：Data point for the collected vital sign data of intelligent wearable device to be sent to high in the clouds
Module is analysed, and the analysis result of data analysis module is passed back to the intelligent wearable device of user's wearing；
Data analysis module：For establishing pathological model, and the life entity that pathological model and network transmission module are sent
Sign data are matched to obtain analysis result；
Big data memory module：Vital sign data for storing pathological model, acquisition and analysis result.
Chronic disease intelligent management system above-mentioned based on big data further includes：
Mobile client：Data are independently uploaded for binding multidigit user, user, receive analysis result；
Private clound client：For establishing private clound account for user, the data independently uploaded from user are received, according to
The data of the big data memory module storage of data and high in the clouds that user uploads are that user establishes personalized chronic disease management
Center；The data that chronic disease management can in real time upload user are matched with the pathological model of data analysis module, are obtained
To analysis result, when had found that it is likely that in analysis result there is certain chronic disease when, analysis result is passed through into network transmission in time
Module passes back to intelligent wearable device and/or mobile client；
Hospital's client：For the data of user's private clound client upload to be filed, received for user, go user every time
The later data of hospital diagnosis upload to the private clound client of user and upload newest chronic disease information.
Chronic disease intelligent management system above-mentioned based on big data further includes voice broadcast module：For according to analysis
As a result periodically current physical condition is reported for user.
In chronic disease intelligent management system above-mentioned based on big data, the pathological model is to utilize Bayes
Sorting algorithm is trained data set.
In chronic disease intelligent management system above-mentioned based on big data, the big data memory module is
The distributed parallel file system of Hadoop frames；The data analysis module is based on Spark distributed computing frameworks
Data analysis module；The pathological model is established as follows：By hospital, medical web site, increase income on network
Data set is constituted with chronic disease index, the relevant data of prophylactic treatment needed for database acquisition, is then stored in data set
In corresponding database or data warehouse, the data set in database or data warehouse is imported into structure using Sqoop tools
To be stored in the big data memory module of distributed file system, later based on the data analysis mould under Spark frames
Using Bayesian Classification Arithmetic data set is trained to obtain pathological model in block.
In chronic disease intelligent management system above-mentioned based on big data, the analysis that is stored in big data memory module
As a result, being exported from big data memory module by Sqoop tools, then the intelligence that user dresses is passed back to by network transmission module and is worn
Wear equipment and/or mobile client.
It is described based on the number under Spark frames in chronic disease intelligent management system above-mentioned based on big data
According to being trained to obtain pathological model to data set using Bayesian Classification Arithmetic in analysis module, carry out as follows：It is right
Data set carries out converging operation, and using chronic disease type as label, statistical data concentrates the number that all labels occur；It will be each
The corresponding set of disorders of kind chronic disease, obtains the corresponding illness of each chronic disease；By each chronic disease and its corresponding
Illness is as format sample, this uses aggregate function to plaid matching style, carries out aggregate statistics operation to the data of same label, later
Prior probability, the conditional probability that each label is calculated by polymerization result, obtain pathological model.
It is described by pathological model and network transmission mould in chronic disease intelligent management system above-mentioned based on big data
The vital sign data that block is sent is matched to obtain analysis result, is to calculate sample using the vital sign data as sample
Originally the prior probability and conditional probability for belonging to the chronic disease in each pathological model, pass through each prior probability and conditional probability meter
Calculation obtains the probability that sample belongs to each chronic disease, takes the chronic disease classification of maximum probability as sample class, will be described
The probability of sample class is compared with the probability for the chronic disease classification for corresponding to classification in pathological model, obtains analysis result.
Advantageous effect
Compared with prior art, the chronic disease intelligent management system of the invention based on big data, utilizes big data platform
In conjunction with machine learning related algorithm to the symptom of chronic disease different phase and it is possible that chronic disease some symptoms into
Pathological model is established in row training.The user data for acquiring, uploading to high in the clouds in real time by embedded platform compares and analyzes, and is
User provides rational suggestion, realizes the functions such as prevention, prediction, the control that chronic disease is carried out to user.In addition, the present invention trains
Modeling diagnostic message of the data from the corresponding section office authoritative expert of each large hospital used so that patient does not spend hospital and just can
Enough to oneself current health status, there are one rough assessments, alleviate disease to the pressure on outpatients mental state.Simultaneously as
The analysis platform of big data of the present invention is a continuous process to the analysis of user, it is possible to from this continuous process-based view
It examines to obtain the progressive formation of user's body health status, a good decision-making foundation is provided to the diagnosis of patient for doctor.
The present invention is a chronic disease intelligent management system efficiently, practical, utilization is embedded, big data, machine learning,
The technologies such as Android exploitation realize the functions such as prevention, prediction, diagnosis, treatment, the real-time analysis of health data of chronic disease.This hair
The bright application for having gathered big data technology and machine learning the relevant technologies, solve single computer cannot to large-scale data into
The problem of row processing, improve the efficiency of data analysis and the real-time of processing.By chronic to magnanimity in big data platform
The analysis of disease related data so that analysis result has higher reliability, accuracy.
Description of the drawings
Fig. 1 is the principle of the present invention frame diagram；
Fig. 2 is Sqoop tool work block diagrams；
Fig. 3 is HDFS storage organization figures；
Fig. 4 is Spark Organization Charts.
Specific implementation mode
The present invention is further illustrated with reference to the accompanying drawings and examples, but be not intended as to the present invention limit according to
According to.
Embodiment 1.A kind of chronic disease intelligent management system based on big data, principle framework is as shown in Figure 1, system structure
At including
Intelligent wearable device：The analysis knot of vital sign data and reception network transmission module passback for acquiring user
Fruit；The vital sign data includes the data such as heart rate, blood pressure, blood oxygen, body temperature, body fat, internal organ grade；
Network transmission module：Data point for the collected vital sign data of intelligent wearable device to be sent to high in the clouds
Module is analysed, and the analysis result of data analysis module is passed back to the intelligent wearable device of user's wearing, to feed back to user；
Data analysis module：For establishing pathological model, and the life entity that pathological model and network transmission module are sent
Sign data are matched to obtain analysis result；
Big data memory module：Vital sign data for storing pathological model, acquisition and analysis result.
Chronic disease intelligent management system above-mentioned based on big data further includes：
Mobile client：Data are independently uploaded for binding multidigit user, user, receive analysis result；It can bind more
Position user so that mutual physical condition can be understood between household at any time, and timely processing is carried out according to analysis result；
Private clound client：For establishing private clound account for user, the data independently uploaded from user are received, according to
The data of the big data memory module storage of data and high in the clouds that user uploads are that user establishes personalized chronic disease management
Center；The data that chronic disease management can in real time upload user are matched with the pathological model of data analysis module, are obtained
To analysis result, this method is realized to the function of Users'Data Analysis, certain chronic disease occurs when being had found that it is likely that in analysis result
When sick, analysis result is passed back into intelligent wearable device and/or mobile client by network transmission module in time, with to user
Carry out healthy early warning；
Hospital's client：For the data of user's private clound client upload to be filed, received for user, go user every time
The later data of hospital diagnosis upload to the private clound client of user and upload newest chronic disease information to facilitate to building
The pathological model for the chronic disease stood carries out real-time update.
Chronic disease intelligent management system above-mentioned based on big data further includes voice broadcast module：For according to analysis
As a result periodically current physical condition is reported for user.
Pathological model above-mentioned is trained to obtain using Bayesian Classification Arithmetic to data set.
Big data memory module above-mentioned is the distributed parallel file system of Hadoop frames；The data analysis
Module is the data analysis module based on Spark distributed computing frameworks；The pathological model is to establish as follows
's：By hospital, medical web site, the database increased income on network obtain needed for it is related to chronic disease index, prophylactic treatment
Data constitute data set, then data set is stored in corresponding database or data warehouse, using Sqoop tools by data
Data set in library or data warehouse imported into the big data memory module that structure is distributed file system (HDFS) and carries out
Storage, is later trained data set using Bayesian Classification Arithmetic in based on the data analysis module under Spark frames
Obtain pathological model.Since when being trained, required data volume is especially huge, so in structure big data memory module
When need to build the data processing platform (DPP) of High Availabitity, in entire data handling procedure, pass through the Zookeeper tools of Hadoop
The case where entire cluster, is managed.The core of Spark is RDD, and RDD belongs to a kind of data set of distributed memory system
Using the data set stored in HDFS can be imported so that Spark platforms can be to the data in distributed file system HDFS
It is operated.
The data such as the real time data or the real-time blood pressure of user, body fat of usual hospital are stored in relevant database
In, with the increase of data volume, single computer can not store the data information of chronic disease, or even cannot be to it
The disease data data of middle a certain kind chronic disease is stored；At this time, it may be necessary to which a kind of tool can be by a huge data
Collection is cut into multiple small data sets, and each small data are then stored in different servers respectively again.The big number of the present invention
It is HDFS structures according to memory module, HDFS is the distributed parallel file system of Hadoop frames, it has the fault-tolerant of certain altitude
Property, and the data access of high-throughput is provided, the application being highly suitable on large-scale dataset.But by relationship type
Data in database imported into HDFS, it is necessary to complete to operate accordingly using Sqoop tools.Sqoop tools are mainly
For the tool for mutually being shifted the data in relevant database and Hadoop, Sqoop tools are the equal of ditch clearance
It is type database and the bridge of Hadoop, the data in relevant database can be both imported into the distributed document of Hadoop
In system (HDFS), the data in Hadoop distributed file systems can also be imported into relevant database.Sqoop is main
Used the two tools of import and export complete the migrations of data with it is synchronous, detailed process is as shown in Figure 2.Work as data
Amount after reaching certain scale needs that the data set is analyzed or counted, at this point, if simple uses relationship type number
It according to library, can not accurately be analyzed, moreover, being directed to medical data, the ability that relevant database can store
It is limited, it at this moment needs that data are imported into HDFS distributed file systems from service database using the import tools of Sqoop,
Then it is analyzed using data analysis module.
The storage organization of HDFS is as shown in figure 3, the System Framework of HDFS is Master/Slave structures.One typical
HDFS is usually made of a NameNode, a SecondaryNameNode and multiple DataNode, wherein Namenode
It is a central server, is responsible for data block mapping, handles the read-write requests of client, configure copy, manage HDFS's
Name space, that store in NameNode memories is fsimage and fsedits, and wherein fsimage is the mirror image text of metadata
Part, edits are the operation logs of metadata, i.e., the record of modification that file system is made etc. operation.
The general individually placed server of SecondaryNameNode, it is every to obtain fsimage from NameNode at regular intervals
It is merged with edits, is then then forwarded to NameNode.In emergency circumstances SeondaryNameNode can assist restoring
NameNode.DataNode is mainly responsible for provides storage soon for HDFS, stores actual data, and stored information and be sent to
NameNode.File storage units of the HDFS using " block " as its management.Block number evidence in HDFS is located on a node,
And a mass file can be stored in multiple pieces of unit, block where a file can be located at different nodes.In addition,
HDFS also has the characteristics that high fault tolerance, and when user upload the data to HDFS, this document can be cut into multiple areas by system
Block, a file tile system default copy is at 3 parts (copy number can be configured according to user demand), when file block
When damage, NameNode understands copy of the Automatic-searching on other DataNode to restore data.
The analysis result being stored in big data memory module need to be exported by Sqoop tools from big data memory module,
The intelligent wearable device and/or mobile client for passing back to user's wearing by network transmission module again, to feed back to user.Due to
The data that mobile client (such as mobile phone and tablet computer) is read come from relevant database, so needing again will be by big data
Memory module (HDFS) analyzes the analysis result come and imported into relevant database, needs to utilize Sqoop's at this time
Export tools, as shown in Figure 2.The operation principle of export tools is to be read according to the separator that user specifies and parse number
According to, be then converted into insert/update sentences and import data to relational database, when data imported into relevant database with
Afterwards, mobile client can read the data in relevant database, finally be presented to the user analysis result, complete from
Data upload the health datas such as blood pressure, body fat and obtain the process of oneself present health condition analysis result to internet to user.
It is above-mentioned that data set is carried out using Bayesian Classification Arithmetic in based on the data analysis module under Spark frames
Training obtains pathological model, carries out as follows：Converging operation is carried out to data set, using chronic disease type as label, system
Count the number for concentrating all labels to occur；By the corresponding set of disorders of each chronic disease, each chronic disease is obtained
Corresponding illness；Using each chronic disease and its corresponding illness as format sample, this uses aggregate function to plaid matching style, right
The data of same label carry out aggregate statistics operation, calculate prior probability, the condition of each label by polymerization result later
Probability obtains pathological model.
After the big data memory module of HDFS structures is stored data, in the data analysis module of Spark frames
Data are handled using Bayesian Classification Arithmetic.Spark is the big data parallel computation frame calculated based on memory, right
The real-time and high efficiency of data processing are improved when data are handled.Spark frameworks are as shown in figure 4, to data
When being analyzed, client (client) submits task, host node (Master) to find working node idle in current cluster
(Worker) start the driving (Driver) of operation task, Driver applies for resource to Master, later by Task Switching at bullet
Property distributed data collection (RDD) directed acyclic graph, then convert RDD directed acyclic graphs to having for Stage by DAGScheduler
TaskSheduler is submitted to acyclic figure, submits task to be executed to Executor by TaskSheduler.Spark's
RDD can obtain new RDD by " conversion " operation, but can't execute at once, be divided into RDD operators in Spark
Two classes, one kind are transformation operator (Transformation), and one kind is action operator (Action), and wherein transformation operator does not touch
Hair submits operation, it only completes the processing of operation pilot process, and action operator can just trigger submission operation.To being utilized when data analysis
Bayesian Classification Arithmetic.Bayesian algorithm provides a kind of natural method for indicating cause and effect, potential between data for finding
Relationship.By Bayesian Classification Arithmetic it can be found that between the variation and certain chronic diseases of the health datas such as heart rate, body fat
Relevance.
Bayes' theorem is the first theorem of the conditional probability about chance event A, B.Under normal conditions, each is chronic
Symptom corresponding to disease can be by obtaining, for a user more concerned with how being according to intelligence in doctor and medicine books
Can the data that detected of equipment judge that certain chronic disease oneself may be suffered from, Bayesian use is i.e. by according to trouble
The probability of certain symptoms occurs in certain upper chronic disease, and to be inferred to occur certain symptom for user, may to change certain chronic
The probability of disease.Assuming that it is event A to suffer from certain chronic disease, it is event B certain symptom occur, then Bayes' theorem is mainly
It is for stating between the probability of probability of event A under conditions of event B occurs and event B under conditions of event A occurs
Relationship.
In formula (1), and P (A | B) it indicates under the premise of event B has occurred and that, the probability that event A occurs is called event B
The conditional probability of lower event A occurs.
In formula (2), and P (B | A) it indicates under the premise of the time, A had occurred and that, the probability that event B occurs is called event A
The conditional probability of lower event B occurs.Formula (2) is the definition of Bayesian formula, is illustrated between P (A | B) and P (B | A)
Relationship, calculate when can one of probability of happening obtain the probability for being inferred to another event.
Bayesian Classification Arithmetic is the general name of oneclass classification algorithm, Naive Bayes Classification Algorithm category used in the present invention
In the one of which of Bayesian Classification Arithmetic, idea basis is：For the item to be sorted provided, solves and to occur in this project
Under the conditions of the probability that occurs of each classification, find out maximum probability, it is believed that the classification residing for maximum probability is belonging to item to be sorted
Classification.Assuming that x={ a1, a2... amIt is an item to be sorted, it is the symptom occurred, C={ y in conjunction with the content of present invention1,
y2..., ynIt is a set for having classification, the i.e. set of disease data, it can be derived as follows by Bayes' theorem：
Wherein yiIndicate that user has suffered from i-th kind of disease, x then indicates the symptom occurred.P(yi| x) indicate certain diseases occur
The probability of certain chronic disease is suffered from when shape, P (x) indicates the probability for symptom occur.Because denominator P (x) is for all categories
Constant (numerical value is calculated by the data of upload), so only needing molecule maximization can be obtained P (yi| x) most
Big value.Because various characteristic attributes are independent, formula (4) is obtained：
Logarithm is taken to obtain formula (5) above formula：
The reluctant operation for seeking product of formula (4) Computer is converted by formula (5) in order to be easy to solve
Summation and ask the operation of logarithm.The mode of the above process is realized under Spark clusters：Converging operation, system are carried out to sample
The number that all labels occur is counted, the sum of character pair, this uses aggregate function to plaid matching style, gathers to same label data
Close statistical operation.After converging operation, prior probability, conditional probability can be calculated by polymerization result, obtain Bayes point
Class model, i.e. pathological model.
It is above-mentioned to be matched to obtain analysis result with the vital sign data that network transmission module is sent by pathological model,
Be using the vital sign data as sample, calculate sample belong to the chronic disease in each pathological model prior probability and
Conditional probability, is calculated sample by each prior probability and conditional probability and belongs to the probability of each chronic disease and (pass through priori
Probability calculation obtains conditional probability, and the probability that sample belongs to some classification chronic disease is calculated according to conditional probability), it takes general
The probability of the sample class is corresponded to the slow of classification with pathological model by the maximum chronic disease classification of rate as sample class
The probability of property disease category is compared, and obtains analysis result.
Claims (8)
1. a kind of chronic disease intelligent management system based on big data, which is characterized in that including
Intelligent wearable device：The analysis result of vital sign data and reception network transmission module passback for acquiring user；
Network transmission module：Data analysis mould for the collected vital sign data of intelligent wearable device to be sent to high in the clouds
Block, and the analysis result of data analysis module is passed back into the intelligent wearable device that user dresses；
Data analysis module：For establishing pathological model, and the vital sign number that pathological model and network transmission module are sent
According to being matched to obtain analysis result；
Big data memory module：Vital sign data for storing pathological model, acquisition and analysis result.
2. the chronic disease intelligent management system according to claim 1 based on big data, it is characterised in that：The system is also
Including
Mobile client：Data are independently uploaded for binding multidigit user, user, receive analysis result；
Private clound client：For establishing private clound account for user, the data independently uploaded from user are received, according to user
The data of upload and the data of the big data memory module storage in high in the clouds are that user establishes personalized chronic disease administrative center；
The data that chronic disease management can in real time upload user are matched with the pathological model of data analysis module, are analyzed
As a result, when had found that it is likely that in analysis result there is certain chronic disease when, in time by analysis result pass through network transmission module return
Pass to intelligent wearable device and/or mobile client；
Hospital's client：For the data of user's private clound client upload to be filed, received for user, user is gone to hospital every time
Later data are diagnosed to upload to the private clound client of user and upload newest chronic disease information.
3. the chronic disease intelligent management system according to claim 1 based on big data, it is characterised in that：The system is also
Including voice broadcast module：For being periodically that user reports current physical condition according to analysis result.
4. the chronic disease intelligent management system according to claim 1 or 2 based on big data, it is characterised in that：It is described
Pathological model, data set is trained to obtain using Bayesian Classification Arithmetic.
5. the chronic disease intelligent management system according to claim 4 based on big data, it is characterised in that：Described is big
Data memory module is the distributed parallel file system of Hadoop frames；The data analysis module is to be based on Spark
The data analysis module of distributed computing framework；The pathological model is established as follows：Pass through hospital, medical and health network
It stands, constitutes data set with chronic disease index, the relevant data of prophylactic treatment needed for the database acquisition increased income on network, so
Data set is stored in corresponding database or data warehouse afterwards, using Sqoop tools by the number in database or data warehouse
According to integrate imported into structure as the big data memory module of distributed file system in stored, later based on Spark frames
Under data analysis module in data set is trained using Bayesian Classification Arithmetic to obtain pathological model.
6. the chronic disease intelligent management system according to claim 5 based on big data, it is characterised in that：It is stored in big
Analysis result in data memory module is exported from big data memory module by Sqoop tools, then is returned by network transmission module
Pass to the intelligent wearable device and/or mobile client of user's wearing.
7. the chronic disease intelligent management system according to claim 5 based on big data, it is characterised in that：It is described
Based on being trained to obtain pathology mould to data set using Bayesian Classification Arithmetic in the data analysis module under Spark frames
Type carries out as follows：Converging operation is carried out to data set, using chronic disease type as label, statistical data is concentrated all
The number that label occurs；By the corresponding set of disorders of each chronic disease, the corresponding illness of each chronic disease is obtained；It will be every
Kind chronic disease and its corresponding illness are as format sample, this uses aggregate function to plaid matching style, to the data of same label
Aggregate statistics operation is carried out, prior probability, the conditional probability of each label is calculated by polymerization result later, obtains pathology mould
Type.
8. the chronic disease intelligent management system according to claim 7 based on big data, it is characterised in that：The general
Pathological model is matched to obtain analysis result with the vital sign data that network transmission module is sent, and is with the life entity
Sign data are sample, prior probability and conditional probability that sample belongs to the chronic disease in each pathological model are calculated, by each
The probability that sample belongs to each chronic disease is calculated in prior probability and conditional probability, takes the chronic disease classification of maximum probability
As sample class, the probability that the probability of the sample class is corresponded to the chronic disease classification of classification with pathological model carries out
It compares, obtains analysis result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810481954.7A CN108717875A (en) | 2018-05-18 | 2018-05-18 | A kind of chronic disease intelligent management system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810481954.7A CN108717875A (en) | 2018-05-18 | 2018-05-18 | A kind of chronic disease intelligent management system based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108717875A true CN108717875A (en) | 2018-10-30 |
Family
ID=63899953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810481954.7A Pending CN108717875A (en) | 2018-05-18 | 2018-05-18 | A kind of chronic disease intelligent management system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108717875A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109273085A (en) * | 2018-11-23 | 2019-01-25 | 南京清科信息科技有限公司 | The method for building up in pathology breath sound library, the detection system of respiratory disorder and the method for handling breath sound |
| CN109935319A (en) * | 2019-03-15 | 2019-06-25 | 南京邮电大学 | Chronic disease systematic management system |
| CN110211699A (en) * | 2019-06-12 | 2019-09-06 | 湖南智超医疗科技有限公司 | A kind of osteoporosis intelligence screening system |
| CN110299207A (en) * | 2019-05-08 | 2019-10-01 | 天津市第四中心医院 | For chronic disease detection in based on computer prognosis model data processing method |
| CN111506659A (en) * | 2020-04-20 | 2020-08-07 | 杭州数澜科技有限公司 | Data synchronization method, system and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105054909A (en) * | 2015-07-15 | 2015-11-18 | 江西师范大学 | Method for remotely monitoring old people based on intelligent watch and Bluetooth blood oxygen probe |
| CN106066934A (en) * | 2016-05-27 | 2016-11-02 | 山东大学苏州研究院 | A kind of Alzheimer based on Spark platform assistant diagnosis system in early days |
-
2018
- 2018-05-18 CN CN201810481954.7A patent/CN108717875A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105054909A (en) * | 2015-07-15 | 2015-11-18 | 江西师范大学 | Method for remotely monitoring old people based on intelligent watch and Bluetooth blood oxygen probe |
| CN106066934A (en) * | 2016-05-27 | 2016-11-02 | 山东大学苏州研究院 | A kind of Alzheimer based on Spark platform assistant diagnosis system in early days |
Non-Patent Citations (2)
| Title |
|---|
| 叶天瑜: "医疗健康大数据在慢性病管理中的应用研究", 《卫生经济研究》 * |
| 闫茜: "海量医疗数据挖掘平台的研究与设计", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109273085A (en) * | 2018-11-23 | 2019-01-25 | 南京清科信息科技有限公司 | The method for building up in pathology breath sound library, the detection system of respiratory disorder and the method for handling breath sound |
| CN109273085B (en) * | 2018-11-23 | 2021-11-02 | 南京清科信息科技有限公司 | Pathological respiratory sound library establishing method, respiratory disease detection system and respiratory sound processing method |
| CN109935319A (en) * | 2019-03-15 | 2019-06-25 | 南京邮电大学 | Chronic disease systematic management system |
| CN110299207A (en) * | 2019-05-08 | 2019-10-01 | 天津市第四中心医院 | For chronic disease detection in based on computer prognosis model data processing method |
| CN110211699A (en) * | 2019-06-12 | 2019-09-06 | 湖南智超医疗科技有限公司 | A kind of osteoporosis intelligence screening system |
| CN111506659A (en) * | 2020-04-20 | 2020-08-07 | 杭州数澜科技有限公司 | Data synchronization method, system and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717875A (en) | A kind of chronic disease intelligent management system based on big data | |
| Souri et al. | A new machine learning-based healthcare monitoring model for student’s condition diagnosis in Internet of Things environment | |
| El Aboudi et al. | Big data management for healthcare systems: architecture, requirements, and implementation | |
| Kalyankar et al. | Predictive analysis of diabetic patient data using machine learning and Hadoop | |
| Eswari et al. | Predictive methodology for diabetic data analysis in big data | |
| CN109830303A (en) | Clinical data mining analysis and aid decision-making method based on internet integration medical platform | |
| Asri et al. | Reality mining and predictive analytics for building smart applications | |
| CN108197737A (en) | A kind of method and system for establishing medical insurance hospitalization cost prediction model | |
| CN105095653A (en) | Basic service system for medical large data application | |
| CN109785927A (en) | Clinical document structuring processing method based on internet integration medical platform | |
| US20170060622A1 (en) | System and method for a data processing architecture | |
| CN107301118A (en) | A kind of fault indices automatic marking method and system based on daily record | |
| Eiras-Franco et al. | A scalable decision-tree-based method to explain interactions in dyadic data | |
| Bellini et al. | Data flow management and visual analytic for big data smart city/IOT | |
| CN107046557A (en) | The intelligent medical calling inquiry system that dynamic Skyline is inquired about under mobile cloud computing environment | |
| CN112331349B (en) | Cerebral apoplexy relapse monitoring system | |
| CN113130085A (en) | 5G intelligent sensing control prediction system based on big data | |
| Demirbaga et al. | Social media data analysis using MapReduce programming model and training a Tweet Classifier using Apache Mahout | |
| CN109192312B (en) | Intelligent management system and method for adverse events of heart failure patients | |
| Afyouni et al. | E-ware: a big data system for the incremental discovery of spatio-temporal events from microblogs | |
| Zhao et al. | Analyzing COVID-19 epidemiological data | |
| CN110837970A (en) | Regional health platform quality control method and system | |
| Leung et al. | Big Data Mining on Health Informatics Data for Cities | |
| Niu et al. | CNN autoencoders and LSTM-based reduced order model for student dropout prediction | |
| CN112562862A (en) | Epidemic situation information identification method, device and equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181030 |