CN110176283A - The associated chronic diseases management method of space-time big data, storage medium and terminal - Google Patents
The associated chronic diseases management method of space-time big data, storage medium and terminal Download PDFInfo
- Publication number
- CN110176283A CN110176283A CN201910438740.6A CN201910438740A CN110176283A CN 110176283 A CN110176283 A CN 110176283A CN 201910438740 A CN201910438740 A CN 201910438740A CN 110176283 A CN110176283 A CN 110176283A
- Authority
- CN
- China
- Prior art keywords
- disease
- individual activity
- item set
- frequent item
- information
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Epidemiology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Pathology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
A kind of associated chronic diseases management method of space-time big data, storage medium and terminal, which comprises obtain the slow sick relevant information and personal lifestyle habits information of patient；The slow sick relevant information and personal lifestyle habits information of the patient of acquisition are pre-processed；Slow sick relevant information and personal lifestyle habits information based on pretreated patient, generate corresponding training data；Clustering is carried out to the training data of generation according to time and space, obtains corresponding multiple clusters；Analysis is associated to obtained multiple clusters, obtains the frequent item set between individual activity and disease；Based on the frequent item set between individual activity and disease, the strong rules results collection between individual activity and disease is calculated；Strong rules results collection between the individual activity and disease is visualized and exports display.The accuracy and efficiency of chronic diseases management can be improved in above-mentioned scheme.
Description
Technical field
The invention belongs to field of medical technology, more particularly to a kind of associated chronic diseases management method of space-time big data, deposit
Storage media and terminal.
Background technique
The data of magnanimity are generated daily in each industry of big data era, and medical industry is no exception.With powerful
Data storage, the development of computing platform and mobile Internet, present trend are a large amount of outbursts of medical data and quickly electric
Sub-figure drives medical big data and is widely applied in core realms such as clinical research, healthy chronic diseases management, public health,
So that medical field becomes the important research part of big data field.Go out data using excavating resource abundant in big data information
The meaning of behind has immeasurable effect for medical industry.For example, there are between information and information in diagnostic message
Incidence relation, by the association analysis mining data in data mining hide a large amount of correlation rules, to auxiliary doctor do
Chronic diseases management scheme is of great significance.Correlation rule --- refer to frequent in record according to data item appearance in database
Degree, the derivation rule about data item obtained.The motivation that correlation rule initially proposes is for market basket analysis (Market
Basket Analysis) problem proposition, by finding that customer is put into the association between the different commodity in " shopping basket ", point
Analyse the purchasing habits of customer, this associated discovery can help retailer which commodity understood frequently by customer while to purchase
It buys, so that them be helped to develop better marketing strategy.Correlation rule can also apply to chronic diseases management simultaneously, and doctor is helped to do
Better chronic diseases management scheme out.General correlation rule has two steps, and the first step is looked in total data according to minimum support
Frequent item set out, second step find Strong association rule according to min confidence in frequent item set.
Data show that for the chronic in China more than 300,000,000 people, the lethal number of chronic disease has accounted for China because of disease at present
The 85% of death toll, caused Disease Spectrum have accounted for the 70% of total Disease Spectrum.It was predicted that the year two thousand thirty, the slow disease in the whole world
Relevant total death toll will rise to the 70% of the total death toll in the world.China's 65 one full year of life above old man's number is about at present
1.4 hundred million people account for total population 10.47%, and aging ratio improves year by year, and potential slow patient group radix certainly will continue to expand, society
The slow disease challenge got worse will be faced.Meanwhile rejuvenation development trend, serious shadow has been presented in chronic disease taking diabetes as an example
The quality of life and health for arriving resident are rung, chronic diseases management is very urgent.
Currently, domestic and international big data is concentrated mainly on essential information and trouble in clinical diagnosis information in the research of chronic diseases management
The excavation of sick information assists chronic diseases management using essential information and illness information, this just affects the accuracy of chronic diseases management
And efficiency.
Summary of the invention
Present invention solves the technical problem that being how to improve the accuracy and efficiency of chronic diseases management.
In order to achieve the above object, the present invention provides a kind of associated chronic diseases management method of space-time big data, the method
Include:
Obtain the slow sick relevant information and personal lifestyle habits information of patient；
The slow sick relevant information and personal lifestyle habits information of the patient of acquisition are pre-processed；
Slow sick relevant information and personal lifestyle habits information based on pretreated patient, generate corresponding trained number
According to；
Clustering is carried out to the training data of generation according to time and space, obtains corresponding multiple clusters；
Analysis is associated to obtained multiple clusters, obtains the frequent item set between individual activity and disease；
Based on the frequent item set between individual activity and disease, the strong rule knot between individual activity and disease is calculated
Fruit collection；
Strong rules results collection between the individual activity and disease is visualized and exports display.
Optionally, the method also includes:
It is exported to user and shows corresponding chronic diseases management scheme.
Optionally, the slow sick relevant information and personal lifestyle habits information of the patient of described pair of acquisition pre-processes, and wraps
It includes at least one of following:
Accuracy audit is carried out to the slow sick relevant information and personal lifestyle information of acquired patient；
The slow sick relevant information and personal lifestyle information of the patient of non-legible form are converted to the patient's of written form
Slow disease relevant information and personal lifestyle information；
Numerical value in the slow sick relevant information and personal lifestyle information of acquired patient is carried out to the discretization of successive value
Processing and the processing of missing values polishing.
Optionally, the slow sick relevant information and personal lifestyle habits information based on pretreated patient, generation pair
The training data answered, comprising:
Preset slow disease is chosen from the slow sick relevant information and personal lifestyle habits information of the pretreated patient
Correlated characteristic data；
Selected slow sick correlated characteristic data are subjected to Feature Conversion, obtain corresponding training data.
Optionally, described that analysis is associated to obtained multiple clusters, obtain the frequency between individual activity and disease
Numerous item collection, comprising:
Analysis is associated to the training data under each cluster, obtains the frequent item set of disease 1- disease 2；
Based on the frequent item set of obtained disease 1- disease 2, the frequent episode between individual activity and disease 1- disease 2 is obtained
Collection；
Frequent item set between the individual activity and disease 1- disease 2 is screened, corresponding individual activity is obtained
With the final frequent item set between disease 1- disease 2.
Optionally, the training data under each cluster is associated analysis, obtains the frequent episode of disease 1- disease 2
Collection, comprising:
Calculate the probability of occurrence of every kind of disease in the data under each cluster；
When the probability of occurrence of disease is greater than preset first support threshold, corresponding training data is added to correspondence
Common disease Candidate Set；
Calculate the second support of every two kinds of diseases in the common disease Candidate Set；
When determining that the second support being calculated is greater than preset second support threshold, by corresponding training data
It is added in the frequent item set between corresponding disease 1- disease 2.
Optionally, the frequent item set based on obtained disease 1- disease 2, obtains individual activity and disease 1- disease 2
Between frequent item set, comprising:
The every two kinds thirds between disease and individual activity in the frequent item set between the disease 1- disease 2 are calculated to support
Degree；
When determining that the third support being calculated is greater than preset third support threshold, by corresponding training data
It is added in the frequent item set between corresponding individual activity and disease 1- disease 2；Between individual activity and disease 1- disease 2
Frequent item set includes the frequent item set that individual activity arrives individual activity to the frequent item set and disease 1- disease 2 of disease 1- disease 2.
Optionally, the frequent item set based between individual activity and disease, be calculated individual activity and disease it
Between strong rules results collection, comprising:
Calculate the confidence that the strong rules results between the individual activity and disease concentrate disease 1- disease 2- individual activity
Degree；
When the confidence level being calculated is greater than preset confidence threshold value, corresponding training data is added to described
Strong rules results between people's activity and disease are concentrated.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer instruction, described
Computer instruction executes the step of space-time big data described in any of the above embodiments associated chronic diseases management method when running.
The embodiment of the invention also provides a kind of terminal, including memory and processor, energy is stored on the memory
Enough computer instructions run on the processor, the processor execute any of the above-described when running the computer instruction
The step of space-time big data associated chronic diseases management method.
Compared with prior art, the invention has the benefit that
Above-mentioned scheme is located in advance by the slow sick relevant information and personal lifestyle habits information of the patient to acquisition
Reason；Slow sick relevant information and personal lifestyle habits information based on pretreated patient, generate corresponding training data；According to
Time and space carry out clustering to the training data of generation, obtain corresponding multiple clusters；To obtained multiple clusters
It is associated analysis, obtains the frequent item set between individual activity and disease；Based on the frequent episode between individual activity and disease
Collection, is calculated the strong rules results collection between individual activity and disease；Due to the training number according to time and space to generation
According to clustering is carried out, data volume can be substantially reduced, the efficiency of chronic diseases management is improved；Meanwhile obtained individual activity and disease
Strong rules results collection between disease embodies the relationship between individual activity and disease, can to give doctor and patient provide it is quasi-
True chronic diseases management suggestion.
Detailed description of the invention
In order to more clearly explain the technical solutions in the embodiments of the present application, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, the drawings in the following description are only some examples of the present application, for
For those of ordinary skill in the art, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is the flow diagram of a kind of associated chronic diseases management method of space-time big data of the embodiment of the present invention；
Fig. 2 is a kind of structural schematic diagram of the associated chronic diseases management device of space-time big data of the embodiment of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.Related directionality instruction in the embodiment of the present invention (such as upper and lower, left and right,
It is forward and backward etc.) it is only used for the relative positional relationship explained under a certain particular pose (as shown in the picture) between each component, movement feelings
Condition etc., if the particular pose changes, directionality instruction is also correspondingly changed correspondingly.
As stated in the background art, it is concentrated mainly in clinical diagnosis information substantially in the research of chronic diseases management in the prior art
The excavation of information and illness information assists chronic diseases management using essential information and illness information, this just affects chronic diseases management
Accuracy and efficiency.
Technical solution of the present invention is carried out by the slow sick relevant information and personal lifestyle habits information of the patient to acquisition
Pretreatment；Slow sick relevant information and personal lifestyle habits information based on pretreated patient, generate corresponding training data；
Clustering is carried out to the training data of generation according to time and space, obtains corresponding multiple clusters；To obtained multiple
Cluster is associated analysis, obtains the frequent item set between individual activity and disease；Based on the frequency between individual activity and disease
The strong rules results collection between individual activity and disease is calculated in numerous item collection；Due to the instruction according to time and space to generation
Practice data and carry out clustering, data volume can be substantially reduced, improve the efficiency of chronic diseases management；Meanwhile obtained individual activity
Strong rules results collection between disease embodies the relationship between individual activity and disease, can be to giving doctor and patient mentions
For accurate chronic diseases management suggestion.
It is understandable to enable above-mentioned purpose of the invention, feature and beneficial effect to become apparent, with reference to the accompanying drawing to this
The specific embodiment of invention is described in detail.
Fig. 1 is the flow diagram of a kind of associated chronic diseases management method of space-time big data of the embodiment of the present invention.Referring to
A kind of Fig. 1, associated chronic diseases management method of space-time big data, can specifically include following step:
Step S101: the slow sick relevant information and personal lifestyle habits information of patient are obtained.
In specific implementation, the slow sick relevant information and personal lifestyle habit letter of patient can be obtained according to the actual needs
Breath.Wherein, patient is obtained by wearable device, electronic health record data, hand written case histories data, individual activity information questionnaire form
Slow sick relevant information and personal lifestyle habits information, provide data source for prediction.
For example, obtaining data by wearable device, the terminal device data of patient worn are mostly come from, such as intelligence
Blood glucose, blood pressure, blood lipid instrument, wearable dynamic electrocardiogram recording instrument, smartwatch cardiotachometer, main includes the relevant information of slow disease,
Including blood glucose, blood lipid, blood pressure and heart rate etc.；Slow sick class data are obtained by electronic health record, it is main all including paper case history
Information is with electronic equipment, such as computer, health card, preservation, management, the medical treatment for the digitized patient transmitted and reappeared
Record, can suitably replace hand-written paper case history；Obtain slow sick class data by hand written case histories, be due to present hospital data not
Interconnection, there are information barrier, hand written case histories are still the important form of diagnosis information record when participating in the cintest, and believe in area medical
Breath not yet realize it is shared under the premise of, the Medical record information content is to understand patient to go to a doctor the important evidence of situation in Different hospital；
The acquisition of individual activity information can obtain personal essential information, row by questionnaire survey form, Effect of follow-up visit by telephone, video return visit etc.
For information, action message.Including daily life system, living habit, eating habit etc..
Step S102: the slow sick relevant information and personal lifestyle habits information of the patient of acquisition are pre-processed.
In specific implementation, the slow sick relevant information and personal lifestyle habits information for obtaining patient may be it is incomplete,
Noise, inconsistent, high latitude, duplicate, these data will affect prediction result, it is not possible to directly use.Therefore, exist
The slow sick relevant information and personal lifestyle habits information of the patient of acquisition are pre-processed, can be accorded with pretreated data
The requirement of molding type obtains the data of high quality, improves the predictablity rate of data.
In specific implementation, it is pre-processed in the slow sick relevant information and personal lifestyle habits information of the patient to acquisition
When, may include at least one of following three:
(1) accuracy audit is carried out to the slow sick relevant information and personal lifestyle information of acquired patient.Slow disease data
From a wealth of sources, wherein further including the processing of the different modes of data, this will lead to data correctness decline, while be likely to occur weight
Multiple extra data, need to screen clear, and the data correction of mistake is deleted.For example, some obvious incorrect data, such as year
In age 200 etc., this kind of data can be deleted directly or telephone verification is corrected.
(2) the slow sick relevant information and personal lifestyle information of the patient of non-legible form are converted to the patient of written form
Slow sick relevant information and personal lifestyle information.The slow sick relevant information and personal lifestyle information of the patient of non-legible form include
The slow sick relevant information and personal lifestyle information of the patient of the forms such as picture, video, voice is converted into text.For example, cardiopulmonary disease
Some images, hand written case histories in disease are converted to picture and extract information etc. in picture again.For another example, for adaptive model, guarantee
Picture is changed into text or number by the uniformity of model, convenient for extracting feature and related information.In addition, remembering for some follow-ups
Record, such as including video, Effect of follow-up visit by telephone information, need to change into the information of alphanumeric readability, in order to model training with
The extraction of information.
(3) numerical value in the slow sick relevant information and personal lifestyle information of acquired patient is subjected to the discrete of successive value
Change processing and the processing of missing values polishing.It mainly include the numerical value class data for picture, video, voice or electronic health record itself
Data processing is carried out, discretization, the processing of numeric data missing values including successive value etc.
Step S103: slow sick relevant information and personal lifestyle habits information based on pretreated patient are generated and are corresponded to
Training data.
In specific implementation, slow sick relevant information and personal lifestyle habits information based on pretreated patient, generate
Corresponding training data is the process that initial data is changed into the training data of model, it is therefore an objective to obtain preferably training
Data characteristics, so that the upper limit of Model approximation data.This process can specifically include:
Firstly, being chosen from the slow sick relevant information and personal lifestyle habits information of the pretreated patient preset
Slow disease correlated characteristic data, i.e. progress feature selecting, it is uncorrelated or redundancy feature including rejecting, reduce of validity feature
Number, to reduce the time of model training.Simultaneously, it is thus necessary to determine that name, gender, age, occupation, daily life system, the diet of patient
The correlated characteristics such as habit, disease name, illness description, consultation time, site of disease.
Then, selected slow sick correlated characteristic data are subjected to Feature Conversion, obtain corresponding training data.First carry out
In other words feature combination that is to say that the part by original feature is converted into having relevant physical significance and reality with slow disease
The feature of meaning.For example, the system is to need for be converted in the place of illness longitude and dimension, and will under space-time big data
Longitude and dimension combine to form a new feature.Then, characteristic value is reprocessed: the characteristic value of the Combination nova of extraction is carried out
Data processing again makes it meet model needs.For example, the characteristic value normalization of the Combination nova of extraction is handled etc..
Step S104: carrying out clustering to the training data of generation according to time and space, obtains corresponding multiple poly-
Class.
In specific implementation, clustering is carried out to the training data of generation according to time and space, that is to say will be regular
Data according to time and space clustering.Wherein, time and space respectively indicate age and the site of disease of patient.
Step S105: analysis is associated to obtained multiple clusters, is obtained frequent between individual activity and disease
Item collection.
In specific implementation, analysis is associated to obtained multiple clusters, under each cluster, to regular number
According to analysis is associated, the frequent item set between individual activity and disease is excavated.This process includes the following steps:
Firstly, being associated analysis to the training data under each cluster, the frequent item set of disease 1- disease 2 is obtained.Tool
For body, for each cluster in multiple clusters for more obtaining, (i.e. herein is slow for all diseases for traversing in each cluster
Disease) type, count the frequency M of every kind of diseasexWith data sum M, the probability P that every kind of disease occurs is calculated, it may be assumed that
When be calculated every kind of disease appearance probability be, by by under each cluster every kind of disease occur and probability
It is compared with preset first support threshold, to filter out common disease Candidate Set.In other words, when determining certain disease
Probability of occurrence be greater than first support threshold when, corresponding training data is retained；Otherwise, by corresponding training data
It deletes, to filter out common disease Candidate Set.
Then, it for the common disease Candidate Set filtered out, is calculated using FP-Growth algorithm comprising two kinds of diseases
Frequent item set.Specifically, counting the frequency M of former piece disease x1 to consequent disease x2 firstx1→x2, and count in Candidate Set H1
In all disease x1 to disease x2 sum Mx1x2, and calculate the second support:
It wherein, will be right when the second support of two kinds of diseases being calculated is greater than preset second support threshold
The training data answered is retained, to filter out the frequent item set between qualified two kinds of diseases to get disease 1- is arrived
The frequent item set of disease 2；
Later, the frequent item set based on obtained disease 1- disease 2, obtains between individual activity and disease 1- disease 2
Frequent item set.Specifically, for filtering out the frequent item set between qualified two kinds of diseases, FG-Growth is recycled to calculate
Method is calculated including former piece disease x1 to consequent disease x2 to individual activity or individual activity to former piece disease x1 to consequent disease
X2 frequent item set, that is, the result set of sequencing is not present between individual activity and disease.In an embodiment of the present invention,
Former piece disease x1 to consequent disease x2 is counted to individual activity or individual activity to former piece disease x1 to the frequency of consequent disease x2
M(x1→x2)∩y, count all sum Ms comprising from disease x1 and disease x2 and individual activity yx1x2y, calculate third support:
When the third support being calculated, pass through the third support and preset third support that will be calculated
Threshold value is compared, and the corresponding trained number of the disease for remaining larger than third support, so that last frequent item set is filtered out,
The result set of sequencing is not present i.e. between individual activity and disease, namely obtains corresponding individual activity and disease 1- disease
Final frequent item set between 2.
Step S106: it based on the frequent item set between individual activity and disease, is calculated between individual activity and disease
Strong rules results collection.
In specific implementation, it when executing above-mentioned association analysis operation completion to each cluster, obtains under each cluster
Multiple individual activities and disease between be not present sequencing result set.Finally, further according to multiple under each cluster
There is no the result sets of sequencing to obtain the correlation rule between feature between people's activity and disease, and is built according to correlation rule
Vertical " disease 1- disease 2- individual activity " results set H, it may be assumed that
First, the confidence level of disease 1- disease 2- individual activity y is calculated:
Wherein, C (x1 → x2 → y) indicates the confidence level of disease 1- disease 2- individual activity y, and N indicates the frequency occurred, then
N ((x1 → x2) ∩ y) is indicated while disease 1 is occurred to disease 2, the frequency of individual activity y, and disease 1 occurs in N (x1 → x2) expression
To the frequency of disease 2
Later, the confidence level being calculated is compared with preset confidence threshold value, when what determination was calculated sets
Reliability be greater than preset confidence threshold value when, then for strong correlation rule and retain this rule into strong rules results collection H, thus
To " disease 1- disease 2- individual activity " strong rules results set.
Step S107: the strong rules results collection between the individual activity and disease is visualized and exports display.
In specific implementation, when obtaining the strong rules results set between the individual activity and disease, by described
Strong rules results collection between people's activity and disease visualizes and exports display, the clinical information of itself and patient are mutually tied
It closes, the activity of clinician can be alleviated in a manner of really effective, suggestion is provided for patient and give patient reminders.For example,
Doctor's knowledge heuristic is given by the strong rules results set between the individual activity and disease, with the troubling possibility of determination
The case where.
In an embodiment of the present invention, the method can also include:
Step S108: it is exported to user and shows corresponding chronic diseases management scheme.
In specific implementation, when obtaining the strong rules results set between the individual activity and disease, disease of waiting a moment patient
When Associated Acute event occurs, patient is reminded to see a doctor in time and provide timely processing scheme and measure, the latter is according to result set
And the information of patient is combined, suggestion etc. is provided for the possibility variation for the treatment of plan.
The above-mentioned associated chronic diseases management method of space-time big data in the embodiment of the present invention is described in detail, below
The above-mentioned corresponding device of method will be introduced.
Fig. 2 shows the structural representations of the associated chronic diseases management device of one of embodiment of the present invention space-time big data
Figure.Referring to fig. 2, the associated chronic diseases management device 20 of the space-time big data may include:
Acquiring unit 201, suitable for obtaining the slow sick relevant information and personal lifestyle habits information of patient；
Pretreatment unit 202, slow sick relevant information and personal lifestyle habits information suitable for the patient to acquisition carry out pre-
Processing；In specific implementation, the pretreatment unit 202, slow sick relevant information and personal lifestyle suitable for the patient to acquisition
Habits information is pre-processed, and includes at least one of the following: slow sick relevant information and personal lifestyle letter to acquired patient
Breath carries out accuracy audit；The slow sick relevant information and personal lifestyle information of the patient of non-legible form are converted into written form
Patient slow sick relevant information and personal lifestyle information；By the slow sick relevant information and personal lifestyle information of acquired patient
In numerical value carry out successive value sliding-model control and missing values polishing processing.
Training data generation unit 203, suitable for slow sick relevant information and personal lifestyle habit based on pretreated patient
Used information, generates corresponding training data；In an embodiment of the present invention, the training data generation unit 203 is suitable for from institute
It states and chooses preset slow sick correlated characteristic data in the slow sick relevant information and personal lifestyle habits information of pretreated patient；
Selected slow sick correlated characteristic data are subjected to Feature Conversion, obtain corresponding training data.
Cluster cell 204 obtains corresponding suitable for carrying out clustering to the training data of generation according to time and space
Multiple clusters；
Association analysis unit 205 obtains individual activity and disease suitable for being associated analysis to obtained multiple clusters
Between frequent item set；In an embodiment of the present invention, association analysis unit 205, suitable for the training data under each cluster
It is associated analysis, obtains the frequent item set of disease 1- disease 2；Based on the frequent item set of obtained disease 1- disease 2, obtain a
Frequent item set between people's activity and disease 1- disease 2；To the frequent item set between the individual activity and disease 1- disease 2 into
Row screening, obtains the final frequent item set between corresponding individual activity and disease 1- disease 2.In another embodiment of the present invention
In, the association analysis unit 205, suitable for calculating the probability of occurrence of every kind of disease in the data under each cluster；When disease
When probability of occurrence is greater than preset first support threshold, it is candidate that corresponding training data is added to corresponding common disease
Collection；Calculate the second support of every two kinds of diseases in the common disease Candidate Set；When determining the second support being calculated
When greater than preset second support threshold, corresponding training data is added to frequent between corresponding disease 1- disease 2
In item collection.In still another embodiment of the process, the association analysis unit 205, suitable for calculating the frequency between described two diseases
The every two kinds third supports between disease and individual activity in numerous item collection；When the determining third support being calculated is greater than in advance
If third support threshold when, corresponding training data is added between corresponding individual activity and disease 1- disease 2
In frequent item set；Frequent item set between individual activity and disease 1- disease 2 includes individual activity to the frequent of disease 1- disease 2
Item collection and disease 1- disease 2 arrive the frequent item set of individual activity.
Computing unit 206, suitable for individual activity and disease is calculated based on the frequent item set between individual activity and disease
Strong rules results collection between disease；In an embodiment of the present invention, the computing unit 206 is suitable for calculating the individual activity
Strong rules results between disease concentrate the confidence level of disease 1- disease 2- individual activity；When the confidence level being calculated is greater than
When preset confidence threshold value, the strong rules results collection that corresponding training data is added between the individual activity and disease
In.
Display unit 207 is exported, suitable for visualizing the strong rules results collection between the individual activity and disease
And export display.In an embodiment of the present invention, the output display unit 207 is further adapted for corresponding to user's output display
Chronic diseases management scheme.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer instruction, described
The step of space-time big data associated chronic diseases management method is executed when computer instruction is run.Wherein, the space-time
The associated chronic diseases management method of big data refers to the introduction of preceding sections, repeats no more.
The embodiment of the invention also provides a kind of terminal, including memory and processor, energy is stored on the memory
Enough computer instructions run on the processor, the processor execute the space-time when running the computer instruction
The step of big data associated chronic diseases management method.Wherein, the associated chronic diseases management method of the space-time big data refers to
The introduction of preceding sections, repeats no more.
Using the above scheme in the embodiment of the present invention, pass through the slow sick relevant information and personal lifestyle of the patient to acquisition
Habits information is pre-processed；Slow sick relevant information and personal lifestyle habits information based on pretreated patient, generation pair
The training data answered；Clustering is carried out to the training data of generation according to time and space, obtains corresponding multiple clusters；It is right
Obtained multiple clusters are associated analysis, obtain the frequent item set between individual activity and disease；Based on individual activity with
The strong rules results collection between individual activity and disease is calculated in frequent item set between disease；Due to according to time and sky
Between clustering is carried out to the training data of generation, data volume can be substantially reduced, improve the speed and efficiency of data processing；Together
When, the strong rules results collection between obtained individual activity and disease embodies the relationship between individual activity and disease, can be with
Accurate chronic diseases management suggestion is provided to doctor and patient is given.
The basic principles, main features and advantages of the present invention have been shown and described above.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, the present invention
Claimed range is delineated by the appended claims, the specification and equivalents thereof from the appended claims.
Claims (10)
1. a kind of associated chronic diseases management method of space-time big data characterized by comprising
Obtain the slow sick relevant information and personal lifestyle habits information of patient；
The slow sick relevant information and personal lifestyle habits information of the patient of acquisition are pre-processed；
Slow sick relevant information and personal lifestyle habits information based on pretreated patient, generate corresponding training data；
Clustering is carried out to the training data of generation according to time and space, obtains corresponding multiple clusters；
Analysis is associated to obtained multiple clusters, obtains the frequent item set between individual activity and disease；
Based on the frequent item set between individual activity and disease, the strong rules results between individual activity and disease are calculated
Collection；
Strong rules results collection between the individual activity and disease is visualized and exports display.
2. the associated chronic diseases management method of space-time big data according to claim 1, which is characterized in that further include:
It is exported to user and shows corresponding chronic diseases management scheme.
3. the associated chronic diseases management method of space-time big data according to claim 1, which is characterized in that described pair acquisition
The slow sick relevant information and personal lifestyle habits information of patient pre-processes, and includes at least one of the following:
Accuracy audit is carried out to the slow sick relevant information and personal lifestyle information of acquired patient；
The slow sick relevant information and personal lifestyle information of the patient of non-legible form are converted to the slow disease of the patient of written form
Relevant information and personal lifestyle information；
Numerical value in the slow sick relevant information and personal lifestyle information of acquired patient is carried out to the sliding-model control of successive value
With the processing of missing values polishing.
4. the associated chronic diseases management method of space-time big data according to claim 1, which is characterized in that described based on pre- place
The slow sick relevant information and personal lifestyle habits information of patient after reason, generates corresponding training data, comprising:
It is related that preset slow disease is chosen from the slow sick relevant information and personal lifestyle habits information of the pretreated patient
Characteristic；
Selected slow sick correlated characteristic data are subjected to Feature Conversion, obtain corresponding training data.
5. the associated chronic diseases management method of space-time big data according to claim 1, which is characterized in that described to acquired
Multiple clusters be associated analysis, obtain the frequent item set between individual activity and disease, comprising:
Analysis is associated to the training data under each cluster, obtains the frequent item set of disease 1- disease 2；
Based on the frequent item set of obtained disease 1- disease 2, the frequent item set between individual activity and disease 1- disease 2 is obtained；
Frequent item set between the individual activity and disease 1- disease 2 is screened, corresponding individual activity and disease are obtained
Final frequent item set between sick 1- disease 2.
6. the associated chronic diseases management method of space-time big data according to claim 5, which is characterized in that described to each poly-
Training data under class is associated analysis, obtains the frequent item set of disease 1- disease 2, comprising:
Calculate the probability of occurrence of every kind of disease in the data under each cluster；
When the probability of occurrence of disease is greater than preset first support threshold, corresponding training data is added to corresponding normal
See disease Candidate Set；
Calculate the second support of every two kinds of diseases in the common disease Candidate Set；
When determining that the second support being calculated is greater than preset second support threshold, corresponding training data is added
Into the frequent item set between corresponding disease 1- disease 2.
7. the associated chronic diseases management method of space-time big data according to claim 5 or 6, which is characterized in that described to be based on
The frequent item set of obtained disease 1- disease 2 obtains the frequent item set between individual activity and disease 1- disease 2, comprising:
Calculate the every two kinds third supports between disease and individual activity in the frequent item set between the disease 1- disease 2；
When determining that the third support being calculated is greater than preset third support threshold, corresponding training data is added
Into the frequent item set between corresponding individual activity and disease 1- disease 2；It is frequent between individual activity and disease 1- disease 2
Item collection includes the frequent item set that individual activity arrives individual activity to the frequent item set and disease 1- disease 2 of disease 1- disease 2.
8. the associated chronic diseases management method of space-time big data according to claim 1, which is characterized in that described based on individual
Frequent item set between activity and disease, is calculated the strong rules results collection between individual activity and disease, comprising:
Calculate the confidence level that the strong rules results between the individual activity and disease concentrate disease 1- disease 2- individual activity；
When the confidence level being calculated is greater than preset confidence threshold value, corresponding training data is added to described personal living
The dynamic strong rules results between disease are concentrated.
9. a kind of computer readable storage medium, is stored thereon with computer instruction, which is characterized in that the computer instruction fortune
Perform claim requires the step of 1 to 8 described in any item space-time big datas associated chronic diseases management method when row.
10. a kind of terminal, which is characterized in that including memory and processor, storing on the memory can be at the place
The computer instruction run on reason device, perform claim requires any one of 1 to 8 institute when the processor runs the computer instruction
The step of space-time big data stated associated chronic diseases management method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910438740.6A CN110176283A (en) | 2019-05-24 | 2019-05-24 | The associated chronic diseases management method of space-time big data, storage medium and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910438740.6A CN110176283A (en) | 2019-05-24 | 2019-05-24 | The associated chronic diseases management method of space-time big data, storage medium and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110176283A true CN110176283A (en) | 2019-08-27 |
Family
ID=67692093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910438740.6A Withdrawn CN110176283A (en) | 2019-05-24 | 2019-05-24 | The associated chronic diseases management method of space-time big data, storage medium and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110176283A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580940A (en) * | 2019-08-28 | 2019-12-17 | 北京好医生云医院管理技术有限公司 | Chronic disease management method and device based on big data |
| CN112037876A (en) * | 2020-09-16 | 2020-12-04 | 陈俊霖 | System, device and storage medium for chronic disease course stage analysis |
| CN114694852A (en) * | 2022-04-13 | 2022-07-01 | 武汉科瓴智能科技有限公司 | Chronic disease analysis method and system |
-
2019
- 2019-05-24 CN CN201910438740.6A patent/CN110176283A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580940A (en) * | 2019-08-28 | 2019-12-17 | 北京好医生云医院管理技术有限公司 | Chronic disease management method and device based on big data |
| CN112037876A (en) * | 2020-09-16 | 2020-12-04 | 陈俊霖 | System, device and storage medium for chronic disease course stage analysis |
| CN114694852A (en) * | 2022-04-13 | 2022-07-01 | 武汉科瓴智能科技有限公司 | Chronic disease analysis method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Johnson et al. | Machine learning and decision support in critical care | |
| Ben-Bassat et al. | Pattern-based interactive diagnosis of multiple disorders: The MEDAS system | |
| CN103493054A (en) | Healthcare information technology system for predicting development of cardiovascular conditions | |
| CN110176283A (en) | The associated chronic diseases management method of space-time big data, storage medium and terminal | |
| Kour et al. | Usage and implementation of neuro-fuzzy systems for classification and prediction in the diagnosis of different types of medical disorders: a decade review | |
| CN108140175A (en) | Labor management system, Method of labor management in road and labor management program | |
| CN111145909B (en) | Diagnosis and treatment data processing method and device, storage medium and electronic equipment | |
| CN112786131B (en) | Method and device for identifying treatment information, electronic equipment and storage medium | |
| WO2021158796A1 (en) | Forecasting and explaining user health metrics | |
| CN102405473A (en) | A point-of-care enactive medical system and method | |
| Hsu | A decision-making mechanism for assessing risk factor significance in cardiovascular diseases | |
| CN110164520A (en) | The associated chronic diseases management device of space-time big data | |
| LK et al. | COVID-19 outbreak based coronary heart diseases (CHD) prediction using SVM and risk factor validation | |
| KR20210112041A (en) | Smart Healthcare Monitoring System and Method for Heart Disease Prediction Based On Ensemble Deep Learning and Feature Fusion | |
| CN114420279A (en) | Medical resource recommendation method, device, equipment and storage medium | |
| Adebayo | Predictive model for the classification of hypertension risk using decision trees algorithm | |
| CN116453694A (en) | Disease risk prediction method and system based on under-sampling integrated framework with replacement | |
| Umapathy et al. | Artificial intelligence for Healthcare 4.0 | |
| Chowdary et al. | Multiple Disease Prediction by Applying Machine Learning and Deep Learning Algorithms | |
| Tolentino et al. | CAREdio: Health screening and heart disease prediction system for rural communities in the Philippines | |
| JP2021189093A (en) | Diagnostic device for autoimmune bullosis | |
| Uddin et al. | A Voice assistive mobile application tool to detect cardiovascular disease using machine learning approach | |
| Chai et al. | Edge Computing with Fog-cloud for Heart Data Processing using Particle Swarm Optimized Deep Learning Technique | |
| Sharma et al. | A Novel Heart Disease Prediction System Using XGBoost Classifier Coupled With ADASYN SMOTE | |
| Yousuff et al. | The Role of Artificial Intelligence for Intelligent Mobile Apps |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190827 |
|
| WW01 | Invention patent application withdrawn after publication |