CN106682412A - Diabetes prediction method based on medical examination data - Google Patents
Diabetes prediction method based on medical examination data Download PDFInfo
- Publication number
- CN106682412A CN106682412A CN201611199219.4A CN201611199219A CN106682412A CN 106682412 A CN106682412 A CN 106682412A CN 201611199219 A CN201611199219 A CN 201611199219A CN 106682412 A CN106682412 A CN 106682412A
- Authority
- CN
- China
- Prior art keywords
- examination data
- physical examination
- diabetes
- data
- suffering
- 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
- 206010012601 diabetes mellitus Diseases 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012549 training Methods 0.000 claims abstract description 21
- 230000000694 effects Effects 0.000 claims abstract description 5
- 230000004927 fusion Effects 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 7
- 239000000203 mixture Substances 0.000 claims description 7
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000012952 Resampling Methods 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 3
- 238000003058 natural language processing Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 6
- 201000010099 disease Diseases 0.000 abstract description 3
- 238000003745 diagnosis Methods 0.000 description 7
- 230000036541 health Effects 0.000 description 7
- 238000007477 logistic regression Methods 0.000 description 7
- 238000003066 decision tree Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 102000017011 Glycated Hemoglobin A Human genes 0.000 description 2
- 108010014663 Glycated Hemoglobin A Proteins 0.000 description 2
- 206010018473 Glycosuria Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 102000010445 Lactoferrin Human genes 0.000 description 1
- 108010063045 Lactoferrin Proteins 0.000 description 1
- 206010034719 Personality change Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000002513 implantation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- CSSYQJWUGATIHM-IKGCZBKSSA-N l-phenylalanyl-l-lysyl-l-cysteinyl-l-arginyl-l-arginyl-l-tryptophyl-l-glutaminyl-l-tryptophyl-l-arginyl-l-methionyl-l-lysyl-l-lysyl-l-leucylglycyl-l-alanyl-l-prolyl-l-seryl-l-isoleucyl-l-threonyl-l-cysteinyl-l-valyl-l-arginyl-l-arginyl-l-alanyl-l-phenylal Chemical compound C([C@H](N)C(=O)N[C@@H](CCCCN)C(=O)N[C@@H](CS)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CC=1C2=CC=CC=C2NC=1)C(=O)N[C@@H](CCC(N)=O)C(=O)N[C@@H](CC=1C2=CC=CC=C2NC=1)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CCSC)C(=O)N[C@@H](CCCCN)C(=O)N[C@@H](CCCCN)C(=O)N[C@@H](CC(C)C)C(=O)NCC(=O)N[C@@H](C)C(=O)N1CCC[C@H]1C(=O)N[C@@H](CO)C(=O)N[C@@H]([C@@H](C)CC)C(=O)N[C@@H]([C@@H](C)O)C(=O)N[C@@H](CS)C(=O)N[C@@H](C(C)C)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](C)C(=O)N[C@@H](CC=1C=CC=CC=1)C(O)=O)C1=CC=CC=C1 CSSYQJWUGATIHM-IKGCZBKSSA-N 0.000 description 1
- 229940078795 lactoferrin Drugs 0.000 description 1
- 235000021242 lactoferrin Nutrition 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a diabetes prediction method based on medical examination data. The method includes that 1, medical examination data of each user is processed, and complete medical examination data is obtained; 2, the complete medical examination data of users with diabetes is used as a positive training sample, and the complete medical examination data of users without diabetes is used as a negative training sample; a GBDT+LR model is adopted to conduct training, model adjustment and fusion are conducted according to the effect of the model, and a final prediction model is obtained; 3, processed examination data of new users is used as a prediction sample which is input into the final prediction model, and the probability that the new users suffer from diabetes is obtained. The method can assist doctors in making better judgments, and by the method, patients know better about the risks of their own diseases.
Description
Technical field
The invention belongs to big data medical field, and in particular to a kind of diabetes Forecasting Methodology of medical examination data.
Background technology
With the improvement of people ' s living standards, the enhancing of health care consciousness, health examination is increasingly becoming a kind of social fashion, people
Changed the traditional concept of the Cai Qu hospitals only when falling ill, have regular physical checkups and received by most people.Therefore,
Hospital have accumulated the electronics physical examination data of magnanimity, big data is had ample scope for abilities.
Big data medical treatment is a current focus, refers to the data that medical field is analyzed by big data correlation technique
And knowledge therein is excavated so as to increase substantially medical services.In the past few decades, big data influences dearly
Each enterprise, including health care industry.Nowadays, substantial amounts of data can allow health care more efficient, more individual character
Change.
This year, the World Health Organization (WHO) gives a warning, and China there are about 1.1 hundred million diabetics, account for Chinese adult
The 1/10 of people's sum.If not taking action as early as possible, reduce unhealthy diet and lack the hazards in the life styles such as motion,
It is expected that the numeral will increase to 1.5 hundred million people in the year two thousand forty, brings and have a strong impact on to common people's health and social economy.Diabetes are except right
Patient and its friends and family cause the injury of body and mind, also bring huge economic loss.China puts into nearly 173,400,000,000 people every year
(25,000,000,000 dollars) of coin is used for diabetes management;Direct medical expense for diabetes accounts for the 13% of Chinese medical expenditure.These
Data do not include the economic loss that diabetes related diseases bring to patient home and company also.Big data is introduced into diabetes doctor
Treatment field, can not only reduce surgeon stress, moreover it is possible to allow patient usually to pass more comfortable.
The content of the invention
In view of above-mentioned, the invention provides a kind of diabetes Forecasting Methodology based on medical examination data, the method is logical
The diagnosis of every data target and doctor of patient in analysis physical examination data to patient's physical examination data is crossed, physical examination data and body is set up
Whether the association between inspection diagnosis, prediction patient may suffer from diabetes, so as to aid in doctor preferably to be judged, make patient more
The good risk for being known from suffering from disease.
A kind of diabetes Forecasting Methodology based on medical examination data, comprises the following steps:
(1) the physical examination data to each user are processed, and obtain complete physical examination data;
(2) using the complete physical examination data for suffering from diabetes as Positive training sample, by the complete physical examination for being not suffering from diabetes
Data are used as Negative training sample;It is trained using GBDT+LR models, and model adjustment fusion is carried out according to the effect of model,
Obtain final forecast model;
(3) using treatment after the physical examination data of new user be input to final forecast model as forecast sample, newly used
Diabetes probability is suffered from family.
Step 1 is concretely comprised the following steps:
(1-1) is pre-processed to the physical examination data of each user, obtains the physical examination data of same form;
(1-2) is equalized to the physical examination data of same form, the physical examination data of being equalized;
(1-3) carries out shortage of data value filling to the physical examination data for equalizing, and obtains complete physical examination data.
In step (1-1), the process of the data prediction that checks UP is:First, to diagnosis primary in physical examination data
As a result, physical examination project name and physical examination project result, are analyzed using natural language processing method, obtain analysis result;
Then, analysis result is further cleaned and is standardized, be converted to the physical examination data of same form, additional information is used.
In step (1-2), because in physical examination data, the user for suffering from diabetes only occupies a part therein, therefore,
By expanding the physical examination data (small sample) for suffering from diabetes, the method that diminution is not suffering from the physical examination data (large sample) of diabetes is obtained
To the equal positive negative example training sample of composition of quantity, to reach the equalization of positive and negative example sample, it is easy to follow-up model to use.
Data sample is equalized:In classification problem, the situation that positive and negative example sample data volume is not waited is frequently encountered, such as
Positive example sample is 10w datas, and negative example sample only has 1w datas, now needs to carry out the equalization of sample so that positive and negative example
Sample reaches balance.
There are three kinds to the method that physical examination data are equalized, respectively:
(a) resampling method:The physical examination data for suffering from diabetes by repeated sampling are to expand the physical examination data for suffering from diabetes
Quantity;To reach the equilibrium of positive negative example training sample of composition.
(b) lack sampling method:The physical examination data for being not suffering from diabetes by a small amount of sampling are not suffering from the physical examination number of diabetes to reduce
According to quantity, to reach the equilibrium of positive negative example training sample of composition.
(c) weighed value adjusting method:By the weights for changing the physical examination data for suffering from diabetes and the physical examination data for being not suffering from diabetes
Ratio to cause the total weight value of positive and negative example training sample consistent, to reach the equilibrium of positive negative example training sample of composition.
Preferably, by the way of resampling method is combined with lack sampling method, i.e., stochastical sampling positive example suffers from the body of diabetes
Inspection data, and the less negative example of extraction missing data that sorts is not suffering from the physical examination data of diabetes.Positive example sample was so both expanded
Data volume, has screened poor negative example sample again.
In step (1-3), data value missing refers in data acquisition because natural cause and artificial origin lead
Cause data imperfect, the situation of data value missing is equally there is also in physical examination data, accordingly, it would be desirable to be lacked to physical examination data
Value filling.The method for carrying out Missing Data Filling includes three kinds, respectively:
(a) direct elimination method:Directly delete the physical examination data for having missing data.
B () calculates sample data completion method:By calculating the median of physical examination data, mode, average and random point
Implantation etc., the missing values in filling physical examination data.
(c) comprehensive whole sample data completion method:Most like physical examination data are found, using lacking for its data that check UP
Mistake value is filled, or will missing characteristic value mapping higher dimensional space.
Preferably, the present invention carries out shortage of data value filling using comprehensive whole sample data completion method, specially:Adopt
Physical examination shortage of data value is predicted with K arest neighbors (k-Nearest Neighbor, kNN) algorithm, using user itself its
His feature finds k most like user, and check UP the missing values of data of the similitude weighted average of comprehensive k user are filled out
Fill, k is the number of user.
In step (2), using GBDT (Gradient Boosting Decision Tree) and LR (Logistic
Regression) model is trained, and carries out model adjustment fusion according to the effect of model, obtains final model.
In step (3), first, at step (1-1)~physical examination data of the step (1-3) to each new user
Reason, then, using treatment after the physical examination data of new user be input to final forecast model as forecast sample, obtain new user's
Suffer from diabetes probability.
Diabetes Forecasting Methodology of the present invention based on medical examination data is the physical examination data by analyzing user, using big
The means of data analysis, judge that user's suffers from diabetes disease risk.So as to promote the development of each paradiabetes medical applications, not only
It is the quick judgement provided auxiliary foundation of doctor, while making patient get more information about the potential risk of itself, has
Advantage is as follows:
(1) medical examination data are pre-processed, more available physical examination data is converted into normal data, not only increased
Many training samples, also can predict and service for more complicated physical examination data be provided.
(2) particularity of physical examination data is combined, while final sample quantity is expanded, low-quality sample is carried out
Screening.
(3) KNN algorithms have been used to fill missing values, and local directed complete set optimizes, and can both be pushed away using data with existing
Survey, and do not expend excessive computing resource.
(4) GBDT+LR models are employed, the link that artificial treatment analyzes feature has both been saved, is enhanced again non-linear pre-
Survey ability.
Brief description of the drawings
Fig. 1 is the structure chart of diabetes Forecasting Methodology of the present invention based on medical examination data;
Diagnostic result cleanings and standardization schematic diagram of the Fig. 2 for medical examination data;
Physical examination project name and result cleanings and standardization schematic diagram of the Fig. 3 for medical examination data;
Fig. 4 is positive and negative example diabetes examinee data balancing schematic diagram;
Fig. 5 is shortage of data value fill method analysis chart;
Fig. 6 is part physical examination data glycosuria disease forecasting result figure.
Specific embodiment
In order to more specifically describe the present invention, below in conjunction with the accompanying drawings and specific embodiment is to technical scheme
It is described in detail.
As shown in figure 1, the present invention is based on comprising the following steps that for the diabetes Forecasting Methodology of medical examination data:
Step 1, data prediction:Physical examination data to each user are pre-processed, and obtain the physical examination number of same form
According to.
As shown in Fig. 2 for primary diagnosis data, due to coming from different doctors and specific different scenes, producing
Raw diagnostic result is complicated, it is impossible to directly used.For example in the diabetes diagnosis of required judgement there are diabetes, diabetes
Property PVR, height onset diabetes risk etc., it is necessary to can be used as the mark of diagnosis by data cleansing and standardization
Label are used.By after natural language processing, obtaining all diagnosis words relevant with diabetes first, judged and related by artificial
The auxiliary of medical knowledge, is finally divided into three labels:Diabetes, doubtful diabetes, non-diabetic.Meanwhile, also related physical examination
Project name is also required to be cleaned and is standardized, as shown in Figure 3.Such as glycosylated hemoglobin project, there may be saccharification blood
Lactoferrin A1, glycosated Hb A 1 (HbA1), glycosylated hemoglobin, A1 (HbA1) etc., they are all referring to for same physical examination
Project, simply there is different titles in different combined health checkup services.Except physical examination project name, also physical examination project result is also required to
Cleaning standardization is carried out, for example:Result may be refuse to examine, refuse to survey, not examining, 21,88cm, left hand:135 right hands:129th, 76 not, ++
32 etc., these data can all have unified data form and unit after cleaning and standardization.
Step 2, data sample equalization:Physical examination data to same form are equalized, the physical examination of being equalized
Data.
Such as Fig. 4, after data normalization, a number of sample data can be got, what this thing was often present asks
Topic is, positive and negative example imbalanced training sets, because the user for suffering from diabetes simply occupies a part in all of physical examination user.
To equalize positive and negative example sample data, the present embodiment uses stochastical sampling positive example and suffers from the data of diabetes user, and sorts
Extract the less negative example sample data of missing data.Positive example sample data volume was so both expanded, poor negative example had been screened again
Sample.
Step 3, Missing Data Filling:Physical examination data to equalizing carry out shortage of data value filling, obtain complete physical examination
Data.
In the data of equalization, however it remains many shortage of data values are, it is necessary to be filled, such as Fig. 5.In the present embodiment
Selection is predicted using other features to missing values.The method that intermediate value, average for simple computation data etc. are filled,
Have that randomness is larger, can artificially increase the problem of noise, the accuracy of data can be reduced.And for missing characteristic value is mapped
To the method for higher dimensional space, then can increase amount of calculation, it is necessary to larger resource.The use of selection other features are carried out to missing values
The method of prediction, it is usually required mainly for rely on the correlation of its dependent variable, is more adapted to for physical examination data.Specifically use KNN
Algorithm calculates the most like k bars record of the data, and final Filling power is obtained according to its similitude weighted average.Namely utilize
User itself other features find k most like user, and comprehensively check UP the missing values of data of the value of k user are filled out
Fill.
Step 4, model training:Using the complete physical examination data for suffering from diabetes as Positive training sample, complete is not suffering from
The physical examination data of diabetes are used as Negative training sample;It is trained using GBDT+LR models, and mould is carried out according to the effect of model
Type adjustment fusion, obtains final forecast model.
GBDT is called MART (Multiple Additive Regression Tree), is a kind of conventional nonlinear model
Type, it is based on the boosting thoughts in integrated study, and each iteration all newly sets up one certainly in the gradient direction for reducing residual error
Plan tree, iteration how many times will generate how many decision trees.The thought of GBDT makes it have inherent advantage it can be found that various have
The feature and combinations of features of distinction, the path of decision tree can use directly as LR input feature vectors, eliminate and manually seek
The step of looking for feature, combinations of features.LR is a kind of linear fit model, it is possible to use Logistic functions (or be Sigmoid
Function) become grader.
Step 5, model prediction:The physical examination data of the new user after using treatment are input to final prediction mould as forecast sample
Type, obtain new user suffers from diabetes probability.
After getting model result, for each new user's physical examination data, it is only necessary to automate above-mentioned flow
Obtain him suffers from diabetes probability.As shown in Fig. 6 part physical examination data glycosuria disease forecasting result figure, analysis can be obtained from Fig. 6:
The diabetes predictablity rate for obtaining is predicted using the method fine.
Above-described specific embodiment has been described in detail to technical scheme and beneficial effect, Ying Li
Solution is to the foregoing is only presently most preferred embodiment of the invention, is not intended to limit the invention, all in principle model of the invention
Interior done any modification, supplement and equivalent etc. are enclosed, be should be included within the scope of the present invention.
Claims (6)
1. a kind of diabetes Forecasting Methodology based on medical examination data, comprises the following steps:
(1) the physical examination data to each user are processed, and obtain complete physical examination data;
(2) using the complete physical examination data for suffering from diabetes as Positive training sample, by the complete physical examination data for being not suffering from diabetes
As Negative training sample;It is trained using GBDT+LR models, and model adjustment fusion is carried out according to the effect of model, is obtained
Final forecast model;
(3) using treatment after the physical examination data of new user be input to final forecast model as forecast sample, obtain new user's
Suffer from diabetes probability.
2. the diabetes Forecasting Methodology of medical examination data is based on according to claim 1, it is characterised in that:The tool of step 1
Body step is:
(1-1) is pre-processed to the physical examination data of each user, obtains the physical examination data of same form;
(1-2) is equalized to the physical examination data of same form, the physical examination data of being equalized;
(1-3) carries out shortage of data value filling to the physical examination data for equalizing, and obtains complete physical examination data.
3. the diabetes Forecasting Methodology of medical examination data is based on according to claim 2, it is characterised in that:In step (1-
1) in, the process of the data prediction that checks UP is:First, to diagnostic result primary in physical examination data, physical examination project name
And physical examination project result, it is analyzed using natural language processing method, obtain analysis result;Then, analysis result is entered
The cleaning of one step ground and standardization, are converted to the physical examination data of same form.
4. the diabetes Forecasting Methodology of medical examination data is based on according to claim 2, it is characterised in that:To physical examination data
The method for being equalized has three kinds, respectively:
(a) resampling method:The physical examination data for suffering from diabetes by repeated sampling are expanding the quantity of the physical examination data for suffering from diabetes;
To reach the equilibrium of positive negative example training sample of composition;
(b) lack sampling method:The physical examination data for being not suffering from diabetes by a small amount of sampling are not suffering from the physical examination data of diabetes to reduce
Quantity, to reach the equilibrium of positive negative example training sample of composition;
(c) weighed value adjusting method:By the weighting ratio for changing the physical examination data for suffering from diabetes and the physical examination data for being not suffering from diabetes
To cause the total weight value of positive and negative example training sample consistent, to reach the equilibrium of positive negative example training sample of composition.
5. the diabetes Forecasting Methodology of medical examination data is based on according to claim 4, it is characterised in that:To physical examination data
The method for being equalized is:By the way of resampling method is combined with lack sampling method, stochastical sampling positive example suffers from the body of diabetes
Inspection data, and the less negative example of extraction missing data that sorts is not suffering from the physical examination data of diabetes.
6. the diabetes Forecasting Methodology of medical examination data is based on according to claim 2, it is characterised in that:In step (1-
3) in, physical examination shortage of data value is predicted using kNN algorithms, using user itself, other features find most like k
User, the similitude weighted average of comprehensive k user checks UP the Missing Data Filling of data, and k is the number of user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611199219.4A CN106682412A (en) | 2016-12-22 | 2016-12-22 | Diabetes prediction method based on medical examination data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611199219.4A CN106682412A (en) | 2016-12-22 | 2016-12-22 | Diabetes prediction method based on medical examination data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106682412A true CN106682412A (en) | 2017-05-17 |
Family
ID=58871277
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611199219.4A Pending CN106682412A (en) | 2016-12-22 | 2016-12-22 | Diabetes prediction method based on medical examination data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106682412A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107767960A (en) * | 2017-09-13 | 2018-03-06 | 温州悦康信息技术有限公司 | Data processing method, device and the electronic equipment of clinical detection project |
CN108717867A (en) * | 2018-05-02 | 2018-10-30 | 中国科学技术大学苏州研究院 | Disease forecasting method for establishing model and device based on Gradient Iteration tree |
CN108962386A (en) * | 2017-05-27 | 2018-12-07 | 中国移动通信有限公司研究院 | A kind of data processing method, apparatus and system |
CN109214437A (en) * | 2018-08-22 | 2019-01-15 | 湖南自兴智慧医疗科技有限公司 | A kind of IVF-ET early pregnancy embryonic development forecasting system based on machine learning |
CN109378072A (en) * | 2018-10-13 | 2019-02-22 | 中山大学 | A kind of abnormal fasting blood sugar method for early warning based on integrated study Fusion Model |
CN109448855A (en) * | 2018-09-17 | 2019-03-08 | 大连大学 | A kind of diabetes glucose prediction technique based on CNN and Model Fusion |
CN109524118A (en) * | 2018-11-01 | 2019-03-26 | 上海海事大学 | A kind of screen method for gestational diabetes based on machine learning and physical examination data |
CN111403024A (en) * | 2019-01-02 | 2020-07-10 | 天津幸福生命科技有限公司 | Method and device for obtaining disease judgment model based on medical data |
WO2020181805A1 (en) * | 2019-03-12 | 2020-09-17 | 平安科技(深圳)有限公司 | Diabetes prediction method and apparatus, storage medium, and computer device |
CN111696667A (en) * | 2020-06-11 | 2020-09-22 | 吾征智能技术(北京)有限公司 | Common gynecological disease prediction model construction method and prediction system |
WO2022063047A1 (en) | 2020-09-22 | 2022-03-31 | 博邦芳舟医疗科技(北京)有限公司 | Photoplethysmography-based non-invasive diabetes prediction system and method |
-
2016
- 2016-12-22 CN CN201611199219.4A patent/CN106682412A/en active Pending
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108962386A (en) * | 2017-05-27 | 2018-12-07 | 中国移动通信有限公司研究院 | A kind of data processing method, apparatus and system |
CN107767960A (en) * | 2017-09-13 | 2018-03-06 | 温州悦康信息技术有限公司 | Data processing method, device and the electronic equipment of clinical detection project |
CN108717867A (en) * | 2018-05-02 | 2018-10-30 | 中国科学技术大学苏州研究院 | Disease forecasting method for establishing model and device based on Gradient Iteration tree |
CN109214437A (en) * | 2018-08-22 | 2019-01-15 | 湖南自兴智慧医疗科技有限公司 | A kind of IVF-ET early pregnancy embryonic development forecasting system based on machine learning |
CN109448855A (en) * | 2018-09-17 | 2019-03-08 | 大连大学 | A kind of diabetes glucose prediction technique based on CNN and Model Fusion |
CN109378072A (en) * | 2018-10-13 | 2019-02-22 | 中山大学 | A kind of abnormal fasting blood sugar method for early warning based on integrated study Fusion Model |
CN109524118A (en) * | 2018-11-01 | 2019-03-26 | 上海海事大学 | A kind of screen method for gestational diabetes based on machine learning and physical examination data |
CN111403024A (en) * | 2019-01-02 | 2020-07-10 | 天津幸福生命科技有限公司 | Method and device for obtaining disease judgment model based on medical data |
WO2020181805A1 (en) * | 2019-03-12 | 2020-09-17 | 平安科技(深圳)有限公司 | Diabetes prediction method and apparatus, storage medium, and computer device |
CN111696667A (en) * | 2020-06-11 | 2020-09-22 | 吾征智能技术(北京)有限公司 | Common gynecological disease prediction model construction method and prediction system |
WO2022063047A1 (en) | 2020-09-22 | 2022-03-31 | 博邦芳舟医疗科技(北京)有限公司 | Photoplethysmography-based non-invasive diabetes prediction system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106682412A (en) | Diabetes prediction method based on medical examination data | |
Guo et al. | Using bayes network for prediction of type-2 diabetes | |
CN113011485B (en) | Multi-mode multi-disease long-tail distribution ophthalmic disease classification model training method and device | |
CN107193919A (en) | The search method and system of a kind of electronic health record | |
CN108305690A (en) | The nervous system disease point examines the system for distribution of out-patient department of method and the nervous system disease | |
CN108511056A (en) | Therapeutic scheme based on patients with cerebral apoplexy similarity analysis recommends method and system | |
CN109256212A (en) | Bone health assessment models construction method, device, equipment, medium and appraisal procedure | |
CN113990494B (en) | Tic disorder auxiliary screening system based on video data | |
CN110739076A (en) | medical artificial intelligence public training platform | |
CN110246577A (en) | A method of based on artificial intelligence auxiliary gestational diabetes genetic risk prediction | |
CN110706822B (en) | Health management method based on logistic regression model and decision tree model | |
KR20150078236A (en) | Method and System for personalized healthcare | |
CN113838577B (en) | Convenient layered old people MODS early death risk assessment model, device and establishment method | |
CN103678534A (en) | Physiological information and health correlation acquisition method based on rough sets and fuzzy inference | |
CN106485061A (en) | A kind of life risk assessment and the method for building up of improvement system | |
CN107145715B (en) | Clinical medicine intelligence discriminating gear based on electing algorithm | |
CN110491479A (en) | A kind of construction method of sclerotin status assessment model neural network based | |
CN104462744A (en) | Data quality control method suitable for cardiovascular remote monitoring system | |
CN111370124A (en) | Health analysis system and method based on facial recognition and big data | |
CN116189866A (en) | Remote medical care analysis system based on data analysis | |
CN115985515A (en) | Amblyopia correction effect prediction method, device and equipment based on machine learning | |
He et al. | Quantification of cognitive function in Alzheimer’s disease based on deep learning | |
CN113440101B (en) | Vertigo diagnosis device and system based on ensemble learning | |
CN109192312A (en) | A kind of heart failure patient adverse events intelligent management system and method | |
CN103735247A (en) | Palm color analysis method for health state identification |
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: 20170517 |