CN104504297A - Method for using neural network to forecast hypertension - Google Patents
Method for using neural network to forecast hypertension Download PDFInfo
- Publication number
- CN104504297A CN104504297A CN201510028877.6A CN201510028877A CN104504297A CN 104504297 A CN104504297 A CN 104504297A CN 201510028877 A CN201510028877 A CN 201510028877A CN 104504297 A CN104504297 A CN 104504297A
- Authority
- CN
- China
- Prior art keywords
- hypertension
- data
- neural network
- layer
- value
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of hypertension prevalence forecasting preformed according to personal basic information and healthy information during the medical process, in particular to a method for using a neural network to forecast hypertension. The method for using the neural network to forecast the hypertension includes following steps: (1) finding out dangerous factors which influence the hypertension; (2) extracting health survey data which influences the hypertension; (3) confirming dangerous factors which really influences the hypertension; (4) collecting health information survey data of the dangerous factors which really influences the hypertension; (5) screening the data; (6) performing standardization processing on valid data; (7) building an MLP (multilayer perceptron) model of a BP (back propagation) neural network, and forecasting whether a person is suffered from the hypertension or not through the MLP model of the BP neural network; (8) comparing a forecasted result with an actual situation whether the person is suffered from the hypertension so as to obtain probability details in forecasting of the situation that the person is suffered from the hypertension and the other situation that the person is not suffered from the hypertension, performed through the method for using the neural network to forecast the hypertension. The method for using the neural network to forecast the hypertension provides a scientific basis for old people to prevent hypertension disease in advance, and enables the old people to early discover, early interfere and early cure the hypertension.
Description
Technical field
The present invention relates in medical procedure and carry out Hypertension electric powder prediction according to the essential information of individual, health and fitness information, especially relate to a kind of hypertension Forecasting Methodology utilizing neural network and genetic algorithm.
Background technology
Hypertension is global great controllability chronic disease, has become the main cause of cerebral apoplexy, coronary heart disease and cardiorenal function exhaustion etc.At present, China's Prevalence of Hypertension is greater than 20%, and patient is more than 200,000,000, and the incidence of disease is in ascendant trend year by year, and therefore, hypertensive control is the significant problem that China's population and health field face.
Hypertension drug curative effect is just from the research of group feature study direction individuation, and therefore an effective forecast model is the foundation that people's hypertension therapy provides science, also perfects rational hypertension prediction scheme provide research direction for setting up one.To controlling hypertensive morning anti-morning, the curative effect of Individual drug treatment has very important scientific meaning.At present, the Mode Road for Hypertension prediction is still not clear, and Part Methods is also immature.
Summary of the invention
The object of the invention is to avoid the deficiencies in the prior art to provide a kind of hypertension Forecasting Methodology utilizing neural network, thus effectively solve the problem of prior art.
For achieving the above object, the technical scheme that the present invention takes is: described a kind of hypertension Forecasting Methodology utilizing neural network, is characterized in comprising the steps:
(1) according to factors such as demography, anthropometry, behavioural informations, first the factor that some may affect Hypertension is determined, these factors are defined herein as independent variable, comprise 13 kinds: age, sex, height, body weight, marital status, education, annual family income, exercise situation, whether diabetes, whether high fat of blood, smoke whether more than 100, smoke, drink; Establish a kind of dependent variable factor is hypertension simultaneously;
(2) according to the health and fitness information data that continual investigating system BRFSS maximum in the world provides, the data of above correlative factor are taken out;
(3) by the above data importing SPSS software collected, binary Logistic, is chosen as hypertension by dependent variable to selection analysis---recurrence---, and Variable selection is other 13 kinds of factors;
(4) carry out correlation analysis according to binary Logistic homing method, obtain P value correlation analysis; Its result is as follows:
| variable | B | Wald | P‐value | Odd Ratio(95％CI) |
| Exercise situation | ‐0.130 | 177.438 | 0.000 | 0.878(0.861～0.895) |
| Diabetes | 0.350 | 2616.923 | 0.000 | 1.420(1.401～1.439) |
| High fat of blood | 0.748 | 7650.724 | 0.000 | 2.112(2.077～2.148) |
| Age | ‐0.046 | 18613.797 | 0.000 | 0.955(0.955～0.956) |
| Marital status | ‐0.013 | 15.459 | 0.000 | 0.987(0.981～0.993) |
| Education | 0.046 | 103.814 | 0.000 | 1.047(1.038～1.056) |
| Annual family income | 0.076 | 925.898 | 0.000 | 1.079(1.073～1.084) |
| Body weight | ‐0.011 | 8795.350 | 0.000 | 0.989(0.988～0.989) |
| Height | 0.003 | 453.368 | 0.000 | 1.003(1.003～1.003) |
| Sex | ‐0.058 | 32.369 | 0.000 | 0.944(0.925～0.963) |
| Smoke | 0.006 | 1.219 | 0.270 | 1.006(0.996～1.016) |
| Drink | ‐0.034 | 13.974 | 0.000 | 0.967(0.950～0.984) |
Be less than 0.05 according to P value and there is statistical significance, so eliminate " smoking " to hypertensive impact; Wherein whether independent variable " smokes more than 100 " due to disallowable close to constant value;
(5) obtaining according to above-mentioned analysis the factor affecting Hypertension is 11 kinds: age, sex, height, body weight, marital status, education, annual family income, tempers situation, diabetes, high fat of blood, drinks, and has filtered out the hypertensive 11 kinds of factors of impact;
(6) gather the data affecting hypertension factor, its method is: collected the elderly's essential information by telephone interview mode; Allow the elderly fill in questionnaire form by scene collect; The elderly logs in oneself lily wisdom endowment cloud platform, improves the basic health and fitness information of individual;
(7) be all be stored in background data base by concrete digital form by the data collected in the 6th step; If data are filled in not complete, be considered as invalid data;
(8) after deleting invalid data, standardization is carried out to remaining data, when level when between each index differs greatly, if directly use original index value to analyze, the effect of the higher index of numerical value in comprehensive analysis will be given prominence to, relatively weaken the effect of the lower index of numerical value level, in order to ensure result reliability, standardization is carried out to raw data, herein by data normalization process between [0,1];
(9) neural network model of the Multi-layered Feedforward Networks by the training of error backpropagation algorithm is set up, hypertension prediction is carried out according to above-mentioned data, this model is divided into input layer, hidden layer, output layer, this mode input layer number is InCode=11, comprise the hypertensive 11 kinds of factors of the impact filtered out, hidden layer number is HideCode=2*InCode+1=23, output layer number is OutCode=1, learning rate is Study_Efficient=0.01, number of run Echo_Num=200, judges precision Precision=0.5; Variable parameter in forecasting process is defined as, and input layer carries out being defined between [-1,1] to the weights W_I_H of hidden layer, hidden layer to weights W_H_O, the threshold value Bias_H of hidden layer of output layer, the threshold value Bias_O of output layer;
(10) by calculating pure input, the output of hidden layer, the pure input and output of output layer, the error of output layer, the error of hidden layer, by these parameter feedbacks to neural network, be W_Refresh by right value update, threshold value is updated to Bias_Refresh, is finally saveWV by input layer to the weights of hidden layer and hidden layer to the right value update of output layer; According to output layer final output valve be R as judgement, according to R<0.5 for suffering from hypertension, R>0.5 is not for suffer from hypertension.Statistical forecast goes out the correct number Num1 suffering from hypertension number, and dope the number Num2 correctly not suffering from hypertension number, respectively with the trouble hypertension number of data centralization with do not suffer from hypertension number and carry out ratio, obtain the method correct Prediction and suffer from hypertensive accuracy rate and correct Prediction does not suffer from hypertensive accuracy rate.
The invention has the beneficial effects as follows: described a kind of hypertension prognoses system and method, it determines by adopting Logistic linear regression model (LRM) correlation rule the factor of influence gathered, guarantee the reasonable science of Information Monitoring, adopt data analysis module storage and standardization image data, utilize optimization connection weights and threshold error to train corresponding predicted value simultaneously, improve science and the accuracy of hypertension forecast model.
Accompanying drawing explanation
Fig. 1 is data-mining module prediction algorithm process flow diagram of the present invention;
Fig. 2 is BP neural network hypertension forecast model schematic diagram of the present invention.
Embodiment
Be described principle of the present invention and feature below in conjunction with accompanying drawing, example, only for explaining the present invention, is not intended to limit scope of the present invention.
As illustrated in fig. 1 and 2, described a kind of hypertension Forecasting Methodology utilizing neural network, is characterized in comprising the steps:
(1) according to factors such as demography, anthropometry, behavioural informations, first the factor that some may affect Hypertension is determined, these factors are defined herein as independent variable, comprise 13 kinds: age, sex, height, body weight, marital status, education, annual family income, exercise situation, whether diabetes, whether high fat of blood, smoke whether more than 100, smoke, drink; Establish a kind of dependent variable factor is hypertension simultaneously;
(2) according to the health and fitness information data that continual investigating system BRFSS maximum in the world provides, the data of above correlative factor are taken out;
(3) by the above data importing SPSS software collected, binary Logistic, is chosen as hypertension by dependent variable to selection analysis---recurrence---, and Variable selection is other 13 kinds of factors;
(4) carry out correlation analysis according to binary Logistic homing method, obtain P value correlation analysis; Its result is as follows:
| variable | B | Wald | P‐value | Odd Ratio(95％CI) |
| Exercise situation | ‐0.130 | 177.438 | 0.000 | 0.878(0.861～0.895) |
| Diabetes | 0.350 | 2616.923 | 0.000 | 1.420(1.401～1.439) |
| High fat of blood | 0.748 | 7650.724 | 0.000 | 2.112(2.077～2.148) |
| Age | ‐0.046 | 18613.797 | 0.000 | 0.955(0.955～0.956) |
| Marital status | ‐0.013 | 15.459 | 0.000 | 0.987(0.981～0.993) |
| Education | 0.046 | 103.814 | 0.000 | 1.047(1.038～1.056) |
| Annual family income | 0.076 | 925.898 | 0.000 | 1.079(1.073～1.084) |
| Body weight | ‐0.011 | 8795.350 | 0.000 | 0.989(0.988～0.989) |
| Height | 0.003 | 453.368 | 0.000 | 1.003(1.003～1.003) |
| Sex | ‐0.058 | 32.369 | 0.000 | 0.944(0.925～0.963) |
| Smoke | 0.006 | 1.219 | 0.270 | 1.006(0.996～1.016) |
| Drink | ‐0.034 | 13.974 | 0.000 | 0.967(0.950～0.984) |
Be less than 0.05 according to P value and there is statistical significance, so eliminate " smoking " to hypertensive impact; Wherein whether independent variable " smokes more than 100 " due to disallowable close to constant value;
(5) obtaining according to above-mentioned analysis the factor affecting Hypertension is 11 kinds: age, sex, height, body weight, marital status, education, annual family income, tempers situation, diabetes, high fat of blood, drinks, and has filtered out the hypertensive 11 kinds of factors of impact;
(6) gather the data affecting hypertension factor, its method is: collected the elderly's essential information by telephone interview mode; Allow the elderly fill in questionnaire form by scene collect; The elderly logs in oneself lily wisdom endowment cloud platform, improves the basic health and fitness information of individual;
(7) be all be stored in background data base by concrete digital form by the data collected in the 6th step; If data are filled in not complete, be considered as invalid data;
(8) after deleting invalid data, standardization is carried out to remaining data, when level when between each index differs greatly, if directly use original index value to analyze, the effect of the higher index of numerical value in comprehensive analysis will be given prominence to, relatively weaken the effect of the lower index of numerical value level, in order to ensure result reliability, standardization is carried out to raw data, herein by data normalization process between [0,1];
(9) neural network model of the Multi-layered Feedforward Networks by the training of error backpropagation algorithm is set up, hypertension prediction is carried out according to above-mentioned data, this model is divided into input layer, hidden layer, output layer, this mode input layer number is InCode=11, comprise the hypertensive 11 kinds of factors of the impact filtered out, hidden layer number is HideCode=2*InCode+1=23, output layer number is OutCode=1, learning rate is Study_Efficient=0.01, number of run Echo_Num=200, judges precision Precision=0.5; Variable parameter in forecasting process is defined as, and input layer carries out being defined between [-1,1] to the weights W_I_H of hidden layer, hidden layer to weights W_H_O, the threshold value Bias_H of hidden layer of output layer, the threshold value Bias_O of output layer;
(10) by calculating pure input, the output of hidden layer, the pure input and output of output layer, the error of output layer, the error of hidden layer, by these parameter feedbacks to neural network, be W_Refresh by right value update, threshold value is updated to Bias_Refresh, is finally saveWV by input layer to the weights of hidden layer and hidden layer to the right value update of output layer; According to output layer final output valve be R as judgement, according to R<0.5 for suffering from hypertension, R>0.5 is not for suffer from hypertension.Statistical forecast goes out the correct number Num1 suffering from hypertension number, and dope the number Num2 correctly not suffering from hypertension number, respectively with the trouble hypertension number of data centralization with do not suffer from hypertension number and carry out ratio, obtain the method correct Prediction and suffer from hypertensive accuracy rate and correct Prediction does not suffer from hypertensive accuracy rate.
Described a kind of hypertension Forecasting Methodology utilizing neural network, during its screening, the BRFSS data set at foundation U.S.'s control and prevention of disease center is as training dataset, this system is maximum in the world, ongoing phone health survey system, I has therefrom screened 308771 data of 1996-2005 years, does the information comprised have 1: suffer from hypertension? 2: past one month took exercises? 3: suffer from diabetes? 4: suffer from hypertension? 5: your age? 6: marital status? 7: educational situation? 8: interviewee's annual family income? 9: body weight? 10: height? 11: sex? 12: whether smoking is more than 100? 13: smoking frequency? 14: drink? I utilizes logistic linear regression analysis this information, " whether will suffer from hypertension " as dependent variable, other 13 factors are as independent variable, carry out regretional analysis, result is as follows:
According to get p ?value<0.05 statistically meaningful, so (variable smoke100 is that the value collected due to it is nearly all for being pumped through 100 cigarettes to eliminate smoke and smoke100, the result of investigation is all consistent, a definite value), so I has only used all the other 11 hazards to predict whether suffer from hypertension as dependent variable in the present invention.
And be not stored in background data base according to this numeral as data in platform, thus data are screened herein, if user's registration information is imperfect, the information would not collecting this user carries out forecast analysis to test data is concentrated, and user profile stores and is that (following data wherein can not contain null data when writing test data set in a database in platform, otherwise the test data that can not write this user is concentrated): 1: sex: man: 300002, female: 300001, unknown: 300003 (unknown data is directly deleted); 2: the age: according to date of birth 1950 ?09 ?08 stored in database, after function process, obtain the age information write test data set of individual; 3: height: cm; 4: body weight: kg:5: marital status: 3401 is married, 3402 divorce, 3403 the death of one's spouse, 3404 separation, 3405 is unmarried, 3406 other; 6: education: 1 primary school and following, 2 junior middle schools, 3 special secondary schools and senior middle school, 4 junior colleges, 5 undergraduate courses, 6 masters and more than; Below 7: annual family income: 1:5000 unit, 2:5000-10000,3:10000-15000,4:15000-2500,5:25000-35000,6:35000-50000,7:50000-70000, more than 8:70000 unit; 9: whether took exercises in past one month: 3301 are, 3302 is no; 10: whether suffer from diabetes: 1 is, 2 is no; 11: whether suffer from high fat of blood: 1 is, 2 is no; 12: history of drinking history: 2901 never, and 2902 is a small amount of, 2903 moderates, and 2904 often, and 2905 is unknown, and 2906 guard against
Note: be corresponding with 1 during write test data set, 2,3,4,5,6 replace.Delete illegal data and empty data.
(this data centralization is suffered from hypertension number and is not suffered from hypertension number unequal in the impact that training dataset brings due to data set imbalance, so I has got wherein 216107 data, wherein suffer from hypertension number to equal not suffer from hypertension number, effectively eliminate the impact that data set imbalance is brought model)
Remove invalid data, again standardization is carried out to remaining data, the dimension relation between variable is eliminated during data normalization process, if when the level when between each index differs greatly, if directly use original index value to analyze, the effect of the higher index of numerical value in comprehensive analysis will be given prominence to, the effect of the lower index of relative weakening numerical value level, therefore, in order to ensure the reliability of result, need to carry out standardization to original index data.Herein I with Min ?Max standardization, the method carries out linear transformation to raw data.If minA and maxA is respectively minimum value and the maximal value of attribute A, by A original value x by min ?max standardization be mapped to interval [0,1] value in, its formula is: new data=(Yuan Shuo Ju ?minimal value)/(Ji great Zhi ?minimal value)), be converted to required data layout to store, for data-mining module provides effective data supporting.
The basic personal information of described data collecting module collected and health and fitness information comprise individual and whether suffer from hypertension (1: be, 2: no), health (1: be whether was tempered in one month, 2 is no), whether there are diabetes (1: be, 2: no), whether there is high fat of blood (1: be, 2: no), age (unit is year), marital status (1: married, 2: divorce, 3: the death of one's spouse, 4: separation, 5: unmarried, 6: other), educational situation (1 primary school and following, 2 junior middle schools, 3 special secondary schools and senior middle school, 4 junior colleges, 5 undergraduate courses, 6 masters and more than), annual family income is (below 1:5000 unit, 2:5000-10000, 3:10000-15000, 4:15000-25000, 5:25000-35000, 6:35000-50000, 7:50000-70000, more than 8:70000 unit), body weight (unit is Kg), height (unit is cm), sex (1: man, 2: female), whether past one month drank wine (1: be, 2: female).
Finally carry out correlation predictive using dynamic BP algorithm as prediction algorithm benchmark for data with data-mining module.It is characterized in that defining input layer number InCode, hidden layer number HideCode=2*InCode+1, output layer number OutCode, learning rate Study_Efficient, number of run Echo_Num, judges precision Precision.Again the input layer of neural network is defined between [-1,1] to the weights W_I_H of hidden layer, hidden layer to weights W_H_O, the threshold value Bias_H of hidden layer of output layer, the threshold value Bias_O of output layer.Then calculate again the pure input of hidden layer, export cal_I, the pure input and output of output layer, the error of output layer, the error of hidden layer, feed back to the right value update W_Refresh of neural network again, the renewal Bias_Refresh of neural network threshold value, finally by input layer to the weights of hidden layer and hidden layer to the right value update saveWV of output layer.Then therefrom draw predictablity rate prerate [q]=(double) classNum/SampleNum, wherein classNum have recorded by the number of samples of correctly classifying, and SampleNum is sample data number.
As above-mentioned embodiment, final according to the individual output valve R doped as judgement, according to R<0.5 for suffering from hypertension, R>0.5 does not suffer from hypertension, the hypertensive accuracy rate A that suffers from obtained in training pattern is prediction individual and suffers from hypertensive probability, does not suffer from hypertensive accuracy rate B and be and dope individual and do not suffer from hypertensive probability in training pattern.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (1)
1. utilize a hypertension Forecasting Methodology for neural network, it is characterized in that comprising the steps:
(1) according to factors such as demography, anthropometry, behavioural informations, first the factor that some may affect Hypertension is determined, these factors are defined herein as independent variable, comprise 13 kinds: age, sex, height, body weight, marital status, education, annual family income, exercise situation, whether diabetes, whether high fat of blood, smoke whether more than 100, smoke, drink; Establish a kind of dependent variable factor is hypertension simultaneously;
(2) according to the health and fitness information data that continual investigating system BRFSS maximum in the world provides, the data of above correlative factor are taken out;
(3) by the above data importing SPSS software collected, binary Logistic, is chosen as hypertension by dependent variable to selection analysis---recurrence---, and Variable selection is other 13 kinds of factors;
(4) carry out correlation analysis according to binary Logistic homing method, obtaining the P value that P value correlation analysis obtains smoking is 0.27, and the P value of other factors is 0; Be less than 0.05 according to P value and there is statistical significance, so eliminate " smoking " to hypertensive impact; Wherein whether independent variable " smokes more than 100 " due to disallowable close to constant value;
(5) obtaining according to above-mentioned analysis the factor affecting Hypertension is 11 kinds: age, sex, height, body weight, marital status, education, annual family income, tempers situation, diabetes, high fat of blood, drinks, and has filtered out the hypertensive 11 kinds of factors of impact;
(6) gather the data affecting hypertension factor, its method is: collected the elderly's essential information by telephone interview mode; Allow the elderly fill in questionnaire form by scene collect; The elderly logs in oneself lily wisdom endowment cloud platform, improves the basic health and fitness information of individual;
(7) be all be stored in background data base by concrete digital form by the data collected in the 6th step; If data are filled in not complete, be considered as invalid data;
(8) after deleting invalid data, standardization is carried out to remaining data, when level when between each index differs greatly, if directly use original index value to analyze, the effect of the higher index of numerical value in comprehensive analysis will be given prominence to, relatively weaken the effect of the lower index of numerical value level, in order to ensure result reliability, standardization is carried out to raw data, herein by data normalization process between [0,1];
(9) neural network model of the Multi-layered Feedforward Networks by the training of error backpropagation algorithm is set up, hypertension prediction is carried out according to above-mentioned data, this model is divided into input layer, hidden layer, output layer, this mode input layer number is InCode=11, comprise the hypertensive 11 kinds of factors of the impact filtered out, hidden layer number is HideCode=2*InCode+1=23, output layer number is OutCode=1, learning rate is Study_Efficient=0.01, number of run Echo_Num=200, judges precision Precision=0.5; Variable parameter in forecasting process is defined as, and input layer carries out being defined between [-1,1] to the weights W_I_H of hidden layer, hidden layer to weights W_H_O, the threshold value Bias_H of hidden layer of output layer, the threshold value Bias_O of output layer;
(10) by calculating pure input, the output of hidden layer, the pure input and output of output layer, the error of output layer, the error of hidden layer, by these parameter feedbacks to neural network, be W_Refresh by right value update, threshold value is updated to Bias_Refresh, is finally saveWV by input layer to the weights of hidden layer and hidden layer to the right value update of output layer; According to output layer final output valve be R as judgement, according to R<0.5 for suffering from hypertension, R>0.5 is not for suffer from hypertension.Statistical forecast goes out the correct number Num1 suffering from hypertension number, and dope the number Num2 correctly not suffering from hypertension number, respectively with the trouble hypertension number of data centralization with do not suffer from hypertension number and carry out ratio, obtain the method correct Prediction and suffer from hypertensive accuracy rate and correct Prediction does not suffer from hypertensive accuracy rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510028877.6A CN104504297A (en) | 2015-01-21 | 2015-01-21 | Method for using neural network to forecast hypertension |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510028877.6A CN104504297A (en) | 2015-01-21 | 2015-01-21 | Method for using neural network to forecast hypertension |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104504297A true CN104504297A (en) | 2015-04-08 |
Family
ID=52945694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510028877.6A Pending CN104504297A (en) | 2015-01-21 | 2015-01-21 | Method for using neural network to forecast hypertension |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104504297A (en) |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104965953A (en) * | 2015-07-13 | 2015-10-07 | 熊妹珍 | Establishing method for juvenile height predication model |
| CN106096286A (en) * | 2016-06-15 | 2016-11-09 | 北京千安哲信息技术有限公司 | Clinical path formulating method and device |
| CN106295179A (en) * | 2016-08-10 | 2017-01-04 | 广东工业大学 | A kind of influenza Forecasting Methodology and device |
| CN106384005A (en) * | 2016-09-28 | 2017-02-08 | 湖南老码信息科技有限责任公司 | Incremental neural network model-based depression prediction method and prediction system |
| CN106407694A (en) * | 2016-09-28 | 2017-02-15 | 湖南老码信息科技有限责任公司 | Neurasthenia prediction method and prediction system based on incremental neural network model |
| CN106446551A (en) * | 2016-09-28 | 2017-02-22 | 湖南老码信息科技有限责任公司 | Incremental neural network model based chronic gastroenteritis prediction method and system |
| CN106682434A (en) * | 2016-12-30 | 2017-05-17 | 深圳中科金证科技有限公司 | Method and device for assessing disease risk |
| CN106777874A (en) * | 2016-11-18 | 2017-05-31 | 中国科学院自动化研究所 | The method that forecast model is built based on Recognition with Recurrent Neural Network |
| CN106821355A (en) * | 2017-04-01 | 2017-06-13 | 泰康保险集团股份有限公司 | Method and device for predicting blood pressure |
| CN107742533A (en) * | 2017-09-13 | 2018-02-27 | 天津大学 | A method for predicting gout based on food nutrients |
| CN108351862A (en) * | 2015-08-11 | 2018-07-31 | 科格诺亚公司 | The method and apparatus for determining development progress using artificial intelligence and user's input |
| CN108523867A (en) * | 2018-03-28 | 2018-09-14 | 武汉麦咚健康科技有限公司 | A kind of self calibration PPG non-invasive blood pressure measuring methods and system |
| CN108538393A (en) * | 2018-04-02 | 2018-09-14 | 重庆邮电大学 | Sclerotin quality evaluation expert system based on big data and prediction model method for building up |
| CN109599177A (en) * | 2018-11-27 | 2019-04-09 | 华侨大学 | A method of the deep learning based on case history predicts medical track |
| CN110364258A (en) * | 2019-08-19 | 2019-10-22 | 南京鼓楼医院 | A risk prediction device for Xp11.2 translocation renal carcinoma |
| CN111553478A (en) * | 2020-05-06 | 2020-08-18 | 西安电子科技大学 | Cardiovascular disease prediction system and method for the elderly in community based on big data |
| CN111627558A (en) * | 2020-05-28 | 2020-09-04 | 南京鼓楼医院 | Waistline prediction method based on physical examination data modeling |
| CN111684536A (en) * | 2018-03-05 | 2020-09-18 | 欧姆龙株式会社 | Apparatus for predicting weight of person, and apparatus and method for health management |
| CN112041855A (en) * | 2018-04-27 | 2020-12-04 | 美敦力阿迪安卢森堡有限公司 | Identifying patients suitable for renal denervation therapy |
| CN112420195A (en) * | 2020-11-06 | 2021-02-26 | 清华大学 | Hypertension risk prediction method and device |
| CN112951421A (en) * | 2021-01-26 | 2021-06-11 | 吾征智能技术(北京)有限公司 | System, equipment and storage medium for deducing hypertension based on dietary habits |
| CN113157640A (en) * | 2020-12-31 | 2021-07-23 | 上海明品医学数据科技有限公司 | Family doctor auxiliary inquiry device, terminal and inquiry system |
| CN113257421A (en) * | 2021-05-31 | 2021-08-13 | 吾征智能技术(北京)有限公司 | Method and system for constructing hypertension prediction model |
| CN114638412A (en) * | 2022-03-11 | 2022-06-17 | 平安国际智慧城市科技股份有限公司 | Hazardous waste disposal capacity early warning method and device, electronic equipment and storage medium |
| WO2022141927A1 (en) * | 2020-12-31 | 2022-07-07 | 上海明品医学数据科技有限公司 | Health intervention system, server, and health management system |
| CN114743619A (en) * | 2022-04-20 | 2022-07-12 | 山东大学齐鲁医院 | Questionnaire quality evaluation method and system for disease risk prediction |
| CN115294671A (en) * | 2022-08-08 | 2022-11-04 | 杭州哲达科技股份有限公司 | Air compressor outlet pressure prediction method and prediction system |
| CN117198513A (en) * | 2023-11-07 | 2023-12-08 | 北京烔凡科技有限公司 | Health guidance method and system for hypertension patient |
| CN117253612A (en) * | 2023-10-09 | 2023-12-19 | 吉林大学 | Resident diet quality and chronic disease risk condition evaluation method based on scoring model |
| US11862339B2 (en) | 2019-03-22 | 2024-01-02 | Cognoa, Inc. | Model optimization and data analysis using machine learning techniques |
| US11972336B2 (en) | 2015-12-18 | 2024-04-30 | Cognoa, Inc. | Machine learning platform and system for data analysis |
| US12205725B2 (en) | 2016-11-14 | 2025-01-21 | Cognoa, Inc. | Methods and apparatus for evaluating developmental conditions and providing control over coverage and reliability |
-
2015
- 2015-01-21 CN CN201510028877.6A patent/CN104504297A/en active Pending
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104965953A (en) * | 2015-07-13 | 2015-10-07 | 熊妹珍 | Establishing method for juvenile height predication model |
| CN108351862A (en) * | 2015-08-11 | 2018-07-31 | 科格诺亚公司 | The method and apparatus for determining development progress using artificial intelligence and user's input |
| CN108351862B (en) * | 2015-08-11 | 2023-08-22 | 科格诺亚公司 | Method and apparatus for determining developmental progress using artificial intelligence and user input |
| US11972336B2 (en) | 2015-12-18 | 2024-04-30 | Cognoa, Inc. | Machine learning platform and system for data analysis |
| CN106096286A (en) * | 2016-06-15 | 2016-11-09 | 北京千安哲信息技术有限公司 | Clinical path formulating method and device |
| CN106295179A (en) * | 2016-08-10 | 2017-01-04 | 广东工业大学 | A kind of influenza Forecasting Methodology and device |
| CN106446551A (en) * | 2016-09-28 | 2017-02-22 | 湖南老码信息科技有限责任公司 | Incremental neural network model based chronic gastroenteritis prediction method and system |
| CN106384005A (en) * | 2016-09-28 | 2017-02-08 | 湖南老码信息科技有限责任公司 | Incremental neural network model-based depression prediction method and prediction system |
| CN106407694A (en) * | 2016-09-28 | 2017-02-15 | 湖南老码信息科技有限责任公司 | Neurasthenia prediction method and prediction system based on incremental neural network model |
| US12205725B2 (en) | 2016-11-14 | 2025-01-21 | Cognoa, Inc. | Methods and apparatus for evaluating developmental conditions and providing control over coverage and reliability |
| CN106777874A (en) * | 2016-11-18 | 2017-05-31 | 中国科学院自动化研究所 | The method that forecast model is built based on Recognition with Recurrent Neural Network |
| CN106682434A (en) * | 2016-12-30 | 2017-05-17 | 深圳中科金证科技有限公司 | Method and device for assessing disease risk |
| CN106821355A (en) * | 2017-04-01 | 2017-06-13 | 泰康保险集团股份有限公司 | Method and device for predicting blood pressure |
| CN107742533A (en) * | 2017-09-13 | 2018-02-27 | 天津大学 | A method for predicting gout based on food nutrients |
| CN111684536A (en) * | 2018-03-05 | 2020-09-18 | 欧姆龙株式会社 | Apparatus for predicting weight of person, and apparatus and method for health management |
| CN108523867A (en) * | 2018-03-28 | 2018-09-14 | 武汉麦咚健康科技有限公司 | A kind of self calibration PPG non-invasive blood pressure measuring methods and system |
| CN108538393A (en) * | 2018-04-02 | 2018-09-14 | 重庆邮电大学 | Sclerotin quality evaluation expert system based on big data and prediction model method for building up |
| CN108538393B (en) * | 2018-04-02 | 2021-12-10 | 重庆邮电大学 | Bone quality assessment expert system based on big data and prediction model establishing method |
| CN112041855A (en) * | 2018-04-27 | 2020-12-04 | 美敦力阿迪安卢森堡有限公司 | Identifying patients suitable for renal denervation therapy |
| CN112041855B (en) * | 2018-04-27 | 2024-06-04 | 美敦力阿迪安卢森堡有限公司 | Identifying patients suitable for renal denervation therapy |
| CN109599177A (en) * | 2018-11-27 | 2019-04-09 | 华侨大学 | A method of the deep learning based on case history predicts medical track |
| US11862339B2 (en) | 2019-03-22 | 2024-01-02 | Cognoa, Inc. | Model optimization and data analysis using machine learning techniques |
| CN110364258A (en) * | 2019-08-19 | 2019-10-22 | 南京鼓楼医院 | A risk prediction device for Xp11.2 translocation renal carcinoma |
| CN111553478A (en) * | 2020-05-06 | 2020-08-18 | 西安电子科技大学 | Cardiovascular disease prediction system and method for the elderly in community based on big data |
| CN111627558A (en) * | 2020-05-28 | 2020-09-04 | 南京鼓楼医院 | Waistline prediction method based on physical examination data modeling |
| CN112420195A (en) * | 2020-11-06 | 2021-02-26 | 清华大学 | Hypertension risk prediction method and device |
| CN113157640A (en) * | 2020-12-31 | 2021-07-23 | 上海明品医学数据科技有限公司 | Family doctor auxiliary inquiry device, terminal and inquiry system |
| WO2022141927A1 (en) * | 2020-12-31 | 2022-07-07 | 上海明品医学数据科技有限公司 | Health intervention system, server, and health management system |
| CN113157640B (en) * | 2020-12-31 | 2023-05-23 | 上海明品医学数据科技有限公司 | Auxiliary inquiry device, terminal and inquiry system for family doctor |
| CN112951421A (en) * | 2021-01-26 | 2021-06-11 | 吾征智能技术(北京)有限公司 | System, equipment and storage medium for deducing hypertension based on dietary habits |
| CN113257421B (en) * | 2021-05-31 | 2023-09-15 | 吾征智能技术(北京)有限公司 | Construction method and system of hypertension prediction model |
| CN113257421A (en) * | 2021-05-31 | 2021-08-13 | 吾征智能技术(北京)有限公司 | Method and system for constructing hypertension prediction model |
| CN114638412A (en) * | 2022-03-11 | 2022-06-17 | 平安国际智慧城市科技股份有限公司 | Hazardous waste disposal capacity early warning method and device, electronic equipment and storage medium |
| CN114638412B (en) * | 2022-03-11 | 2024-07-02 | 平安国际智慧城市科技股份有限公司 | Dangerous waste disposal capacity early warning method and device, electronic equipment and storage medium |
| CN114743619B (en) * | 2022-04-20 | 2024-03-29 | 山东大学齐鲁医院 | A questionnaire quality evaluation method and system for disease risk prediction |
| CN114743619A (en) * | 2022-04-20 | 2022-07-12 | 山东大学齐鲁医院 | Questionnaire quality evaluation method and system for disease risk prediction |
| CN115294671A (en) * | 2022-08-08 | 2022-11-04 | 杭州哲达科技股份有限公司 | Air compressor outlet pressure prediction method and prediction system |
| CN117253612A (en) * | 2023-10-09 | 2023-12-19 | 吉林大学 | Resident diet quality and chronic disease risk condition evaluation method based on scoring model |
| CN117198513A (en) * | 2023-11-07 | 2023-12-08 | 北京烔凡科技有限公司 | Health guidance method and system for hypertension patient |
| CN117198513B (en) * | 2023-11-07 | 2024-03-29 | 北京烔凡科技有限公司 | Health guidance method and system for hypertension patient |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104504297A (en) | Method for using neural network to forecast hypertension | |
| Westrupp et al. | Subjective wellbeing in parents during the COVID-19 pandemic in Australia | |
| Wolfe et al. | The income and health effects of tribal casino gaming on American Indians | |
| CN118711822B (en) | Cardiovascular and cerebrovascular disease risk prediction method and system | |
| CN111709575A (en) | Academic achievement prediction method based on C-LSTM | |
| CN106897566A (en) | A kind of construction method and device of risk prediction model | |
| Karatzas et al. | The patient’s hay-fever diary: three years of results from Germany | |
| KR20150078236A (en) | Method and System for personalized healthcare | |
| Guilbert et al. | The theory and application of UW ehealth-PHINEX, a clinical electronic health record-public health information exchange | |
| Chowdry et al. | The cost of late intervention: EIF analysis 2016 | |
| CN110729052B (en) | Elderly health data knowledge analysis method and system | |
| Zhang et al. | Analysis of spatial-temporal distribution and influencing factors of pulmonary tuberculosis in China, during 2008–2015 | |
| Hindmarsh et al. | Effect of comorbidity on relative survival following hospitalisation for fall‐related hip fracture in older people | |
| van Ooijen et al. | Health status over the life cycle | |
| Close et al. | Socioeconomic status and heart failure in Sydney | |
| Zahra et al. | Marginality and multidimensional poverty: A case study of Christian community of Lahore, Pakistan | |
| KR101274431B1 (en) | Apparatus and method for determining health using survey information, apparatus and method for generating health sort function | |
| CN117893027A (en) | Insurance business risk assessment method and apparatus | |
| Keramat et al. | Obesity, disability and self-perceived health outcomes in Australian adults: a longitudinal analysis using 14 annual waves of the HILDA cohort | |
| Johnson et al. | Predictive risk modelling under different data access scenarios: who is identified as high risk and for how long? | |
| Buyukkececi | Neighbourhood effects on early adulthood family life courses: A trajectory‐based approach | |
| CN111986801B (en) | Rehabilitation assessment method, device and medium based on deep learning | |
| Beales et al. | The predictive ability of the full and short versions of the Orebro Questionnaire for Absenteeism and Presenteeism Over the Subsequent 12 Months, in a cohort of young community-based adult workers | |
| CN115458181A (en) | Vaccine hesitation prediction method | |
| Rivera et al. | The Economic Crisis and the Death by Suicide in Spain: Empirical Evidence Based on a Data Panel and the Quantification of Losses in Labour Productivity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150408 |
|
| WD01 | Invention patent application deemed withdrawn after publication |