CN105956384A - Method for realizing assessment engine in health assessment system - Google Patents

Abstract

The invention relates to a method, in particular to a method for realizing an assessment engine in a health assessment system, and belongs to the technical field of health assessment. According to the technical scheme provided by the method for realizing the assessment engine in the health assessment system, the assessment engine adopts a rule engine, and a rule set is defined by utilizing rule flow to realize a required assessment process. According to the method, a decision-making step is described by adopting the rule flow, and a rule structure is layered for optimizing the rule engine, so that the problem of complex chronic disease rules is solved; and the method is well suitable for the health assessment system.

Description

The implementation method of evaluation engine in health evaluation system

Technical field

The present invention relates to a kind of method, the implementation method of evaluation engine in a kind of health evaluation system, Belong to the technical field of health evaluating.

Background technology

Chronic disease (Chronic) full name is Chronic Non-Communicable Diseases, the work and rest of this kind of disease and people and Diet has the biggest relation, therefore in time chronic disease risk is carried out health evaluating, is tied by assessment Fruit adjusts human diet's work and rest, and the risk that reduction individual is suffered from chronic disease has important meaning.

Health evaluating is exactly the health and fitness information according to user, uses specific rule to calculate for identifying user The score value of health degree.This health evaluation system utilizes regulation engine to realize various rule.

Traditional mode realizing business rule is that the form using if...then is described.And it is various chronic The health evaluating rule of disease wants complicated many, and its complexity is difficult to directly by based on rete algorithm Regulation engine utilizes its inferential capability to perform multiple if...then statements and solves.

Summary of the invention

It is an object of the invention to overcome the deficiencies in the prior art, it is provided that a kind of health evaluation system is commented Estimating the implementation method of engine, it describes steps in decision-making, by regular texture stratification pair by using regular flow Regulation engine is optimized, and solves the problem that chronic disease rule is complicated, is well applicable to health evaluating System.

The technical scheme provided according to the present invention, the implementation method of evaluation engine in described health evaluation system, Described evaluation engine uses regulation engine, and utilizes regular flow definition rule set, the assessment needed for realizing Journey.

When carrying out health risk assessment, it was predicted that model is:

$LogitV={\mathrm{\β}}_0+\underset{i=1}{\overset{n}{\Σ}}{\mathrm{\β}}_i{x}_i$

Wherein, x_iFor the risk factor of corresponding disease, β_iRepresent the weight of corresponding risk factor, β₀It is constant term, LogitV is the risk fraction of prediction, and n is the quantity of disease risk factor, then P P is:

Advantages of the present invention: by using regular flow to describe steps in decision-making, by regular texture stratification to rule Then engine is optimized, and solves the problem that chronic disease rule is complicated, is well applicable to health evaluating system System.

Accompanying drawing explanation

Fig. 1 is the flow chart of the present invention.

Fig. 2 is the regular flow schematic diagram of type 2 diabetes mellitus of the present invention.

Detailed description of the invention

Below in conjunction with concrete drawings and Examples, the invention will be further described.

As shown in Figure 1: the problem complicated in order to solve chronic disease rule, well it is applicable to health and comments Estimating system, evaluation engine of the present invention uses regulation engine, and utilizes regular flow definition rule set, with reality Existing required evaluation process.

Specifically, a basic module during regular flow is regulation engine, it is allowed to user sets up a flow chart Carry out the order that definition rule set is estimated.When the assessment rule in system is a lot, management assessment rule is held The order of row can become the most complicated.Regular flow allows to specify the flow process of a rule evaluation, is used for specifying correspondence Rule should be according to sequencing or concurrent assessment, or the prerequisite etc. of specified rule assessment.Make Used time, application program calls the corresponding regular flow of execution by regular flow engine interface, and obtains result, tool Body implementation process is that known to those skilled in the art, here is omitted.

When being embodied as, first passing through health detector and obtain health data, such as sex, body weight etc., as disease In sick risk factor incoming assessment system；Then carry out data prediction, the unit of health data is carried out Unified；Using individual health data as true (facts) incoming regulation engine, and it is the most legal to verify data, bag Including the age the most in the reasonable scope, risk factor data are the most complete；

When carrying out health risk assessment, it was predicted that model is:

$LogitV={\mathrm{\β}}_0+\underset{i=1}{\overset{n}{\Σ}}{\mathrm{\β}}_i{x}_i$

Wherein, x_iFor the risk factor of corresponding disease, β_iRepresent the weight of corresponding risk factor, β₀It is constant term, LogitV is the risk fraction of prediction, and n is the quantity of disease risk factor, then P P is:

The present invention can carry out the disease of health evaluating and mainly include that type 2 diabetes mellitus is predicted, crowd's obesity is predicted, The disease forecastings such as hypertension is predicted, coronary heart disease prediction.Concrete to the present invention as a example by type 2 diabetes mellitus below Process illustrates.

1, code of points design

Risk factor and weight thereof: risk factor is used to calculate the health data of user's disease risk degree, will It uses forecast model to calculate healthy score value as evaluation criteria.

Type 2 diabetes mellitus risk factor and the weight of correspondence thereof

2, modelling

Regulation engine can accept the outside business datum inserted, and (i.e. facts object, this refers to user health Data), the POJO of a Java can serve as a facts object, wherein can include several attributes, Each attribute has a getter of oneself, setter method, is used for revising or obtain this facts object certain Property value.In Drools regulation engine, the health data that rule is used be by facts object from application Pass over.The present invention mainly includes two kinds of facts object: evaluation object and appraisal result object.

1), evaluation object is used to preserve the basic health condition of user, life style, personal history, h disease The information such as history.It is the evaluation object in health evaluation system shown in figure below:

Below by table 4-1, the object properties in evaluation target model are illustrated:

AssessMent (evaluation object) attribute list

AssessInfo (risk assessment information object) attribute list

2), evaluation engine run time, the factor of each disease needs to be added dynamically in engine, therefore definition one Individual RiskFactor class, as appraisal result object, comprises the attributes such as title, weight, default value, appraisal result The risk elements created when using list storage running in object model AssessResult.

The results such as appraisal result object model is used to preserve the health evaluating score value that user is last, risk 's.Below the object properties in appraisal result object model are illustrated:

AssessResult (assessment result object) attribute list

RiskFactor (risk factor object) attribute list

3), rule design and realization

In the embodiment of the present invention, regular flow (Drools flow) implementation rule is mainly used to perform control.By decision-making Flow process is graphically changed intelligible mode and is shown, and helps regulation engine more reasonably to control rule in the execution phase Then perform.

As in figure 2 it is shown, the regular flow of type 2 diabetes mellitus designs, specifically, 1), health data enters health and comments After estimating module, the legal verification of data, mainly check user data, the verification age whether legal, dangerous because of Prime number is according to the most complete, such as, during pretreatment diabetes, the serum fasting glucose data in verification diabetes are No is sky etc.；As long as data violate any admittable regulation, regular flow directly exits, and regulation engine is tied Bundle assessment.Whereas if do not have any admittable regulation to be triggered, regular flow enters the definition of following risks and assumptions； 2), risk factor definition, effect is risk factor corresponding to instantiation before scoring, and sets relevant parameter, Finally risk factor are added in assessment result.Risk factor definition is to be given with the form of decision table.

Drools decision table make use of Excel Spreadsheet as rule editor, defines multiple parametrization Rule, condition and action be defined in the cell of correspondence by script, and incorporating parametric is assembled into health and comments Estimate rule.

35-74 year people taking physical examination 2 type Tang urine onset risk forecast model, risk factor (evaluation factor) Definition

After risk factor adds assessment result, enter scoring pretreatment, it is therefore an objective to wind previous step defined Strategically located and difficult of access element adds in the working memory of regulation engine；Then individual scoring is carried out, by risk factor and number of users According to carrying out rule match, Regression and Major Risk Factors is utilized to calculate dangerous score:

$\begin{array}{c}LogitV={\mathrm{\β}}_0+\underset{i=1}{\overset{n}{\Σ}}{\mathrm{\β}}_i{x}_i\\ =-15.1619+0.0203*X_1+0.3860*X_2-0.0432*X_3+\\ 0.0429*X_4+0.1436*X_5+0.0024*X_6-0.0167*X_7\end{array}$

Wherein, X₁～X₇It is the age respectively, with or without type 2 diabetes mellitus family history (0=without, 1=has), height, waistline, Serum fasting glucose, triglyceride, the value of HDL-C.Finally according to formula: Calculate this user and the probability of type 2 diabetes mellitus occurred within 5 years.

Claims (2)

1. an implementation method for evaluation engine in health evaluation system, is characterized in that: described evaluation engine is adopted With regulation engine, and utilize regular flow definition rule set, the evaluation process needed for realizing.

The implementation method of evaluation engine in health evaluation system the most according to claim 1, is characterized in that: When carrying out health risk assessment, it was predicted that model is:

$LogitV={\mathrm{\β}}_0+\underset{i=1}{\overset{n}{\Σ}}{\mathrm{\β}}_i{x}_i$

Wherein, x_iFor the risk factor of corresponding disease, β_iRepresent the weight of corresponding risk factor, β₀It is constant term, LogitV is the risk fraction of prediction, and n is the quantity of disease risk factor, then P P is: $P=\frac{e^{LogitV}}{1+e^{LogitV}}.$