Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to endSame or similar label represents same or similar element or the element with same or like function.Below with reference to attachedThe embodiment of figure description is exemplary, is only used for explaining the present invention, and is not considered as limiting the invention.
In the description of the invention, it is to be understood that term " multiple " refers to two or more；Term " first "," second " is only used for describing purpose, and it is not intended that instruction or hint relative importance.
Because the symptom of type-II diabetes may be similar to a patients with type Ⅰ DM, and symptom is often unreal apparent.SoIt may be just diagnosed to be with diabetes after morbidity for many years, complication now occurred.Because type-II diabetes are by heredityCaused by factor and environmental factor collective effect, belong to multigenic disease, therefore, the embodiment combination inherent cause and environmentA kind of factor, it is proposed that method and device that based on Rating Model type-II diabetes are carried out with risk score.
Below with reference to the accompanying drawings description is according to embodiments of the present invention is commented type-II diabetes progress risk based on Rating ModelThe method and device divided.
Fig. 1 is the method that based on Rating Model type-II diabetes are carried out with risk score according to one embodiment of the inventionFlow chart.
As shown in figure 1, the side that based on Rating Model type-II diabetes are carried out with risk score according to embodiments of the present inventionMethod, comprise the following steps.
S11, obtain the genetic test result of user and outside pathogenetic feature data.
The genetic test result and outside uploaded as a kind of exemplary embodiment, reception user by terminal is caused a diseaseCharacteristic.
Wherein, terminal can be the hardware device that computer, tablet personal computer, smart mobile phone etc. have various operating systems.
For example, during smart mobile phone, user can by smart mobile phone by the genetic test result of itself andOutside pathogenetic feature data upload onto the server.
Wherein, comprising the gene data for having substantial connection with type-II diabetes in genetic test result, for example, baseBecause TCF7L2 genes can be included in testing result.
, wherein it is desired to explanation, most of all relevant with β cell functions with the related gene of diabetes.
Wherein, outside pathogenetic feature data refer to outside other relevant with causing type-II diabetes in addition to genetic factorsPortion's factor.
Wherein, outside pathogenetic feature data can include but is not limited to lifestyle data, family's medical history data, pharmacohistoryData and physical trait data.
Wherein, the data in terms of lifestyle data can include but is not limited to mean motion amount, diet are such as whether preferenceThe data such as rich food dish, coffee.
Data can include the disease condition of lineal relative's type-II diabetes within three generations during h disease.
Pharmacohistory data can include but is not limited to glucocorticoid, thiazide diuretic (Thiazide), beta receptor retardanceThe use feelings of agent (β-blockers), atypical antipsychotic (Atypical antipsychotic) and statinsCondition., wherein it is desired to explanation, the use of said medicine can improve the risk of diabetes.
Physical trait data can include but is not limited to personal body-mass index BMI (Body Mass Index), yearThe data such as discipline, sex, the length of one's sleep.
As a kind of exemplary embodiment, can provide the user a kind of including life style, family's medical history, pharmacohistoryThe survey of option, and the data acquisition life style number filled according to user in survey must be filled out with physical trait etc.According to, outside pathogenetic feature data such as family's medical history data, pharmacohistory data and physical trait data.
For example, when user opens the product for assessing type-II diabetes risk by terminal, type-II diabetes are assessedThe product of risk can prompt the genetic test result that user uploads itself, and provide comprising life style, family's medical history, pharmacohistoryThe user interface of option must be filled out with physical trait etc., and receives user and is directed to the related data that respective selection is filled in, and will be usedThe related data that family is filled in uploads onto the server so that server obtain the lifestyle data of user, family's medical history data,The outside pathogenetic feature data such as pharmacohistory data and physical trait data.
As another exemplary embodiment, the genetic test result that user is uploaded by terminal, Yi Jicong are receivedPre-save the outside pathogenetic feature data that corresponding user is obtained in the database of the outside pathogenetic feature data of user.
, wherein it is desired to illustrate, the outside pathogenetic feature data of the user preserved in above-mentioned database can be by moreKind of mode obtains, for example, in user's registration, could fill out said external pathogenetic feature data, or, server is from other medical caresThe outside pathogenetic feature data of user are obtained in system.
S12, according to the Rating Model of genetic test result, outside pathogenetic feature data and training in advance, determine user'sThe risk score of type-II diabetes.
In one embodiment of the invention, after the genetic test result of user and outside pathogenetic feature data are obtained,Can be by genetic test result and outside pathogenetic feature data input to the Rating Model of training in advance, Rating Model is by analyzing baseBecause testing result and outside pathogenetic feature data determine the risk score of the type-II diabetes of user.
, wherein it is desired to understand, above-mentioned Rating Model is obtained by training in advance.
In one embodiment of the invention, the detailed process of Rating Model is trained, as shown in Fig. 2 can include：
S21, obtain the sample genetic test result and sample foreign pathogenetic feature data of sample of users.
As a kind of exemplary embodiment, caused in the sample foreign that sample of users is obtained by way of surveyDuring sick characteristic, after the Questionnaire results of sample of users are obtained, it can determine whether that the sample foreign in Questionnaire results causesWhether sick characteristic is complete, when the sample foreign pathogenetic feature data in determining Questionnaire results are imperfect, can incite somebody to actionCorresponding Questionnaire results are deleted, to reduce incomplete sample foreign pathogenetic feature data to follow-up training Rating ModelInfluence.
As another exemplary embodiment, the sample foreign of sample of users is being obtained by way of surveyDuring pathogenetic feature data, after the Questionnaire results of sample of users are obtained, the sample foreign in Questionnaire results can determine whetherWhether pathogenetic feature data are complete, can when the sample foreign pathogenetic feature data in determining Questionnaire results are imperfectDetermine whether incomplete ratio exceedes predetermined threshold value in sample foreign pathogenetic feature data, if it exceeds default thresholdValue, corresponding Questionnaire results are deleted, to reduce incomplete sample foreign pathogenetic feature data to follow-up training scoringThe influence of model.
If it is determined that incomplete ratio is not less than predetermined threshold value in sample foreign pathogenetic feature data, then according to other samplesPathogenetic feature data complete questionnaire adjustment result in this outside adjusts incomplete part in result to the questionnaire and handled, withMake the sample foreign pathogenetic feature data in questionnaire adjustment result complete.
Wherein, predetermined threshold value is pre-set.
For example, it is assumed that not comprising age information in current questionnaire adjustment result, and determine current questionnaire adjustmentAs a result the incomplete ratio of sample foreign pathogenetic feature data is not less than predetermined threshold value, now, if according to other sample foreignsThe complete questionnaire adjustment result of pathogenetic feature data determines that the average value at age is 30 years old, then is adjusted 30 years old as current questionnaireAs a result the age information in.
S22, obtain scoring labeled data corresponding with sample genetic test result and sample foreign pathogenetic feature data.
S23, sample genetic test result, sample foreign pathogenetic feature data and scoring labeled data are trained, withObtain Rating Model.
As a kind of exemplary embodiment, server can utilize sample of the machine learning method to great amount of samples userGenetic test result, sample foreign pathogenetic feature data and scoring labeled data are trained, to determine sample genetic testAs a result, the corresponding relation between sample foreign pathogenetic feature data and scoring labeled data, and according to the corresponding relation obtainedEstablish Rating Model.
That is, the embodiment is entered by machine learning method to a large amount of medical datas related on type-II diabetesRow analysis, by the method for machine learning and data mining, have found the potential rule of type-II diabetes, has trained high accuracyThe high performance model for assessing risk.
Wherein, scoring labeled data is the sample genetic test result and sample foreign pathogenetic feature number according to sample of usersAccording to the risk score of the type-II diabetes marked in advance.
, wherein it is desired to understand, the height of risk score is relevant with following factor：The sample gene inspection of sample of usersSurvey result and determine whether some crucial protein related to type-II diabetes undergo mutation, and determine sample genetic testAs a result whether the gene related to type-II diabetes undergos mutation in, and is undergone mutation in the gene related to type-II diabetesWhen, the type (such as missing, termination in advance) of gene mutation is determined, and determine that the gene related to type-II diabetes occurs prominentWhether change influences β cell functions, and the source of mutation.
Wherein, the source of mutation may be from paternal or maternal.
Wherein, the source of mutation can be determined by the paternal or maternal genetic test result of sample of users.
In one embodiment of the invention, in order to accurately establish Rating Model, random forests algorithm can be based on, to sampleGenetic test result, sample foreign pathogenetic feature data and scoring labeled data are trained, to obtain Rating Model.
Specifically, after sample genetic test result, sample foreign pathogenetic feature data and scoring labeled data is obtained,Random forests algorithm can be first passed through to carry out sample genetic test result, sample foreign pathogenetic feature data and scoring labeled dataTraining, to obtain the Rating Model trained.
After Rating Model is trained, test data set can be obtained, wherein, test data set includes test cdnaTesting result, the outside pathogenetic feature data of test and corresponding scoring labeled data.Then, by test cdna testing result, surveyThe outside pathogenetic feature data input of examination is into Rating Model, to obtain the risk score result exported in Rating Model, afterwards, leads toCross corresponding scoring labeled data in the risk score result exported in Rating Model and test data set and determine Rating ModelThe degree of accuracy whether exceed the degree of accuracy threshold value that pre-sets, if not less than the degree of accuracy threshold value pre-set, to scoringThe model parameter of model carries out tuning processing, to improve the degree of accuracy of Rating Model by adjusting model parameter.
, wherein it is desired to explanation, test data set obtains in advance, for example, can press the data collected in advanceRatio is randomly divided into training dataset and test data set, to obtain test data set by this way.
, wherein it is desired to understand, during using Rating Model, in order to improve constantly the accurate of Rating ModelProperty, Rating Model is updated based on the training data set after renewal after preset time can be spaced.
S13, the advisory information according to corresponding to obtaining risk score, and risk score and corresponding advisory information are providedTo user.
As a kind of exemplary embodiment, determining that the risk that user suffers from type-II diabetes is commented by Rating ModelAfter point, it can be thought according to the risk score for suffering from type-II diabetes pre-saved with the corresponding pass of advisory information, obtained and corresponding windAdvisory information corresponding to the scoring of danger, and risk score and corresponding advisory information are supplied to user.
, wherein it is desired to understand, the risk that risk score is smaller to show that user suffers from type-II diabetes is lower, risk scoreIt is higher show that user suffers from type-II diabetes risk it is higher.
The method that based on Rating Model type-II diabetes are carried out with risk score of the embodiment of the present invention, by by acquired inUser genetic test result and outside pathogenetic feature data input into the Rating Model of training in advance, pass through Rating ModelThe risk score of the type-II diabetes of user, and the advisory information according to corresponding to obtaining risk score are determined, and user is suffered fromThe risk score of type-II diabetes and corresponding advisory information are supplied to user, thus, there is provided one kind passes through Rating ModelThe method of the risk score of the type-II diabetes of accurate evaluation user, the risk score that user obtains type-II diabetes is facilitated,And facilitate user to be known from suffering from the risk of disease according to risk score, relative to conventional method, reduce user and obtain two typesThe time of the risk score of diabetes and cost, improve the user experience of user.
Fig. 3 is the side that based on Rating Model type-II diabetes are carried out with risk score according to another embodiment of the present inventionThe flow chart of method.
As shown in figure 3, the side that based on Rating Model type-II diabetes are carried out with risk score according to embodiments of the present inventionMethod, comprise the following steps.
S31, obtain the genetic test result of user and outside pathogenetic feature data.
S32, according to the Rating Model of genetic test result, outside pathogenetic feature data and training in advance, determine user'sThe risk score of type-II diabetes.
, wherein it is desired to explanation, the explanation to step S11-S12 are also applied for step S31-S32, not gone to live in the household of one's in-laws on getting married hereinState.
S33, judges whether risk score exceedes predetermined threshold value, if so, then performing step S34, otherwise performs step S35.
Wherein, predetermined threshold value is the threshold value of the risk score pre-set.
S34, the advisory information for providing a user risk score and further checking.
That is, when judging that the risk score of type-II diabetes of user exceedes predetermined threshold value, it may be determined that userType-II diabetes are very likely suffered from, in order to find and treat in time, now, can suggest that user carries out a step inspection to hospital.
S35, provide a user risk score and prevent the advisory information of type-II diabetes.
Wherein, dietary recommendation and exercise suggestion can be included but is not limited to by preventing the advisory information of type-II diabetes.
The method that based on Rating Model type-II diabetes are carried out with risk score of the embodiment of the present invention, by by acquired inUser genetic test result and outside pathogenetic feature data input into the Rating Model of training in advance, pass through Rating ModelThe risk score of the type-II diabetes of user is determined, and gives user rational advisory information according to risk score, facilitates useFamily further checks or understood the information relevant with prevention type-II diabetes according to advisory information.
In order to realize above-described embodiment, the invention also provides one kind based on Rating Model to type-II diabetes progress riskThe device of scoring.
Fig. 4 is the device that based on Rating Model type-II diabetes are carried out with risk score according to one embodiment of the inventionStructural representation.
As shown in figure 4, based on Rating Model type-II diabetes should be carried out with the device of risk score includes acquisition module110th, determining module 120 and processing module 130, wherein：
Acquisition module 110 is used to obtain the genetic test result of user and outside pathogenetic feature data.
Wherein, outside pathogenetic feature data can include but is not limited to lifestyle data, family's medical history data, pharmacohistoryData and physical trait data.
Wherein, outside pathogenetic feature data refer to outside other relevant with causing type-II diabetes in addition to genetic factorsPortion's factor.
Wherein, outside pathogenetic feature data can include but is not limited to lifestyle data, family's medical history data, pharmacohistoryData and physical trait data.
Wherein, the data in terms of lifestyle data can include but is not limited to mean motion amount, diet are such as whether preferenceThe data such as rich food dish, coffee.
Data can include the disease condition of lineal relative's type-II diabetes within three generations during h disease.
Pharmacohistory data can include but is not limited to glucocorticoid, thiazide diuretic (Thiazide), beta receptor retardanceThe use feelings of agent (β-blockers), atypical antipsychotic (Atypical antipsychotic) and statinsCondition., wherein it is desired to explanation, the use of said medicine can improve the risk of diabetes.
Physical trait data can include but is not limited to personal body-mass index BMI (Body Mass Index), yearThe data such as discipline, sex, the length of one's sleep.
Determining module 120 is used for the Rating Model according to genetic test result, outside pathogenetic feature data and training in advance,Determine the risk score of the type-II diabetes of user.
Processing module 130 is used for the advisory information according to corresponding to obtaining risk score, and by risk score and correspondingAdvisory information is supplied to user.
In one embodiment of the invention, in order to pass through the risk of the type-II diabetes of Rating Model accurate evaluation userScoring, on the basis of the embodiment shown in Fig. 4, as shown in figure 5, the device can also include training module 140, wherein, instructionPractice sample genetic test result and sample foreign pathogenetic feature data that module 140 is used to obtain sample of users, and acquisition and sampleScoring labeled data corresponding to this genetic test result and sample foreign pathogenetic feature data, and according to sample genetic test knotFruit, sample foreign pathogenetic feature data and scoring labeled data are trained, to obtain Rating Model.
In one embodiment of the invention, in order to train to obtain accurate Rating Model, training module 140 is specifically usedIn：Based on random forests algorithm, entered according to sample genetic test result, sample foreign pathogenetic feature data and scoring labeled dataRow training, to obtain Rating Model.
, wherein it is desired to understand, after Rating Model is trained by training module 140, the device can also includeTest module (not shown), test module are used to obtain test data set, wherein, test data set includes testGenetic test result, the outside pathogenetic feature data of test and corresponding scoring labeled data.Then, test cdna is detected and tiedFruit, the outside pathogenetic feature data input of test are into Rating Model, to obtain the risk score result exported in Rating Model, itAfterwards, scoring is determined by corresponding scoring labeled data in the risk score result exported in Rating Model and test data setWhether the degree of accuracy of model exceedes the degree of accuracy threshold value pre-set, if not less than the degree of accuracy threshold value pre-set, it is rightThe model parameter of Rating Model carries out tuning processing, to improve the degree of accuracy of Rating Model by adjusting model parameter.
, wherein it is desired to understand, during using Rating Model, in order to improve constantly the accurate of Rating ModelProperty, Rating Model is updated based on the training data set after renewal after preset time can be spaced.
In one embodiment of the invention, in order to reasonably suggest to user, on the basis of the embodiment shown in Fig. 5On, as shown in fig. 6, the processing module 130, which can include judging unit 131, first, provides the offer unit of unit 132 and second133, wherein：
Judging unit 131 is used to judge whether risk score exceedes predetermined threshold value.
First provide unit 132 be used for when judging that risk score exceedes predetermined threshold value, provide a user risk score withThe advisory information further checked.
Second, which provides unit 133, is used for when judging risk score not less than predetermined threshold value, provides a user risk scoreWith the advisory information of prevention type-II diabetes.
, wherein it is desired to explanation, the foregoing method to based on Rating Model type-II diabetes are carried out with risk score are realThe explanation for applying example is also applied for the device that based on Rating Model type-II diabetes are carried out with risk score of the embodiment, thisPlace repeats no more.
The device that based on Rating Model type-II diabetes are carried out with risk score of the embodiment of the present invention, by by acquired inUser genetic test result and outside pathogenetic feature data input into the Rating Model of training in advance, pass through Rating ModelThe risk score of the type-II diabetes of user, and the advisory information according to corresponding to obtaining risk score are determined, and user is suffered fromThe risk score of type-II diabetes and corresponding advisory information are supplied to user, thus, facilitate user and obtain two type glycosuriasThe risk score of disease, and facilitate user to be known from suffering from the risk of disease according to risk score, relative to conventional method, reduceUser obtains time and the cost of the risk score of type-II diabetes, improves the user experience of user.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically showThe description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example descriptionPoint is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term notIdentical embodiment or example must be directed to.Moreover, specific features, structure, material or the feature of description can be with officeCombined in an appropriate manner in one or more embodiments or example.In addition, in the case of not conflicting, the skill of this areaArt personnel can be tied the different embodiments or example and the feature of different embodiments or example described in this specificationClose and combine.
In addition, term " first ", " second " are only used for describing purpose, and it is not intended that instruction or hint relative importanceOr the implicit quantity for indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can be expressed orImplicitly include at least one this feature.In the description of the invention, " multiple " are meant that two or more, unless separatelyThere is clearly specific limit.
Any process or method described otherwise above description in flow chart or herein is construed as, and represents to includeModule, fragment or the portion of the code of the executable instruction of one or more the step of being used to realize specific logical function or processPoint, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discuss suitableSequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the inventionEmbodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered useIn the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, forInstruction execution system, device or equipment (such as computer based system including the system of processor or other can be held from instructionThe system of row system, device or equipment instruction fetch and execute instruction) use, or combine these instruction execution systems, device or setIt is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or passDefeated program is for instruction execution system, device or equipment or the dress used with reference to these instruction execution systems, device or equipmentPut.The more specifically example (non-exhaustive list) of computer-readable medium includes following：Electricity with one or more wiringConnecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read-only storage(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only depositsReservoir (CDROM).In addition, computer-readable medium, which can even is that, to print the paper of described program thereon or other are suitableMedium, because can then enter edlin, interpretation or if necessary with it for example by carrying out optical scanner to paper or other mediaHis suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentionedIn embodiment, software that multiple steps or method can be performed in memory and by suitable instruction execution system with storageOr firmware is realized.If, and in another embodiment, can be with well known in the art for example, realized with hardwareAny one of row technology or their combination are realized：With the logic gates for realizing logic function to data-signalDiscrete logic, have suitable combinational logic gate circuit application specific integrated circuit, programmable gate array (PGA), sceneProgrammable gate array (FPGA) etc..
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method carriesSuddenly it is that by program the hardware of correlation can be instructed to complete, described program can be stored in a kind of computer-readable storage mediumIn matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, can alsoThat unit is individually physically present, can also two or more units be integrated in a module.Above-mentioned integrated mouldBlock can both be realized in the form of hardware, can also be realized in the form of software function module.The integrated module is such asFruit is realized in the form of software function module and as independent production marketing or in use, can also be stored in a computerIn read/write memory medium.
Storage medium mentioned above can be read-only storage, disk or CD etc..Although have been shown and retouch aboveEmbodiments of the invention are stated, it is to be understood that above-described embodiment is exemplary, it is impossible to be interpreted as the limit to the present inventionSystem, one of ordinary skill in the art can be changed to above-described embodiment, change, replace and become within the scope of the inventionType.