CN113723922A

CN113723922A - Method and system for realizing prescription auditing based on rule engine

Info

Publication number: CN113723922A
Application number: CN202111006484.7A
Authority: CN
Inventors: 于国超
Original assignee: Beijing Sagittarius Technology Co ltd
Current assignee: Beijing Sagittarius Technology Co ltd
Priority date: 2021-08-30
Filing date: 2021-08-30
Publication date: 2021-11-30
Anticipated expiration: 2041-08-30
Also published as: CN113723922B

Abstract

The invention provides a method and a system for realizing prescription auditing based on a rule engine, wherein the method comprises the following steps: step S1: constructing a rule engine based on the requirements of the user; step S2: acquiring a prescription to be audited; step S3: and auditing the prescription based on the rule engine. According to the method for realizing prescription auditing based on the rule engine, when the method is used, the prescription can be reasonably used for verifying the prescription by inputting the prescription into the rule engine, so that intelligent auditing is realized, the problems of slow manual prescription auditing, poor accuracy and difficulty in supervision are solved, meanwhile, a medical professional user can pertinently establish auditing rules, and the logic of each rule can be traced and verified, so that the specialty, reliability and expandability of the prescription auditing engine are improved.

Description

Method and system for realizing prescription auditing based on rule engine

Technical Field

The invention relates to the technical field of rule engine application, in particular to a method and a system for realizing prescription auditing based on a rule engine.

Background

At present, when a prescription is checked, the prescription is checked based on a preset reasonable medication knowledge database [ for example: the MySQL database stores medicine names, applicable diseases, age contraindications, dosage contraindications, applicable population contraindications, and the like ] for checking the prescription and determining whether the prescription is inappropriate [ for example: multiple drugs in conflict with each other ], however, this method has the following disadvantages:

1. the auditing efficiency is low and the auditing is slow; most databases are designed by professional personnel, and users need to use and learn for a period of time;

2. the supervision is not easy; different users utilize the database to check different prescriptions, and the use subjectivity is strong;

3. a medical professional user cannot design an auditor rule by himself.

Disclosure of Invention

One of the purposes of the invention is to provide a method and a system for realizing prescription auditing based on a rule engine, when the method and the system are used, the prescription can be reasonably used for verifying the prescription by inputting the prescription into the rule engine, intelligent auditing is realized, the problems of slow manual prescription auditing, poor accuracy and difficult supervision are solved, meanwhile, a medical professional user can pertinently establish an auditing rule, the logic of each rule can be traced and verified, and the specialty, the reliability and the expandability of the prescription auditing engine are improved.

The method for realizing prescription auditing based on the rule engine provided by the embodiment of the invention comprises the following steps:

step S1: constructing a rule engine based on the requirements of the user;

step S2: acquiring a prescription to be audited;

step S3: and auditing the prescription based on the rule engine.

Preferably, step S1: building a rule engine based on the requirements of the user, comprising:

extracting a plurality of requirement items in the requirement, wherein the requirement items comprise: a first drug;

establishing a drug taboo rule base;

determining at least one taboo rule corresponding to the first medicine based on a medicine taboo rule library;

acquiring a preset Drools engine, and inputting a tabu rule into the Drools engine;

when the taboo rules are all input into the Drools engine, the Drools engine is used as a rule engine to complete the construction.

Preferably, step S3: based on the rule engine, auditing the prescription, including:

extracting prescription content of a prescription;

inputting the prescription content into a rule engine to obtain an auditing result;

and determining whether the prescription is qualified or not based on the auditing result.

Preferably, establishing a drug contraindication rule base comprises:

acquiring a preset physician set, wherein the physician set comprises: a plurality of physicians;

obtaining warranty information of a doctor, wherein the warranty information comprises: at least one guarantor and at least one guaranty institution;

acquiring a first credit value of a guarantor, summarizing the first credit value and acquiring a first credit value sum;

acquiring a second credit value of the guarantee institution, and summarizing to obtain a second credit value sum;

determining a first credit value and a radius length which corresponds to the second credit value and the sum together based on a preset radius length library;

acquiring a preset virtual space, randomly representing a first position of a doctor in the virtual space, and meanwhile, constructing a virtual sphere by taking the first position as a sphere center and the radius length as a radius;

acquiring a medicine contraindication rule set input by a doctor, wherein the medicine contraindication rule set comprises the following components: a plurality of first drug contraindication rules, the first drug contraindication rules comprising: a second drug, a contra-indicated item corresponding to the second drug, and adverse consequences corresponding to the contra-indicated item;

carrying out semantic segmentation on the bad results to obtain a plurality of bad result items;

performing feature extraction on the adverse consequence item to obtain a plurality of first features;

acquiring a preset bad feature library, and matching the first feature with a second feature in the bad feature library;

if the matching is in accordance with the preset matching, acquiring a severity value corresponding to the second characteristic in accordance with the matching, and summarizing to obtain a severity value sum;

determining a severity value and a corresponding initial length based on a preset initial length library;

a second position in the virtual space, which is away from the center of the sphere in any direction and has an initial length, represents a corresponding first medicine taboo rule, and meanwhile, a connecting line between the center of the sphere and the second position is used as a moving route;

acquiring a preset node set, wherein the node set comprises: a plurality of nodes;

the method comprises the following steps of issuing a first medicine taboo rule through a node, anonymously disclosing, and obtaining a public display result at preset time intervals, wherein the public display result comprises the following steps: the first medicine contraindication rule is newly approved by at least one approver;

acquiring a first credibility of the approver, and simultaneously acquiring a second credibility of the node;

acquiring an access mode of a praise for accessing the first medicine taboo rule, wherein the access mode comprises the following steps: active access and passive access;

when the access mode of the approver is active access, determining a first moving distance corresponding to the first credibility and the second credibility together based on a preset first moving distance library, and simultaneously moving a first medicine taboo rule to the center of the sphere by the first moving distance along a moving route in a virtual space;

when the access mode of the approver is passive access, determining a second moving distance corresponding to the first credibility and the second credibility together based on a preset second moving distance library, and simultaneously moving the first medicine taboo rule to the center of the sphere by the second moving distance along the moving route in the virtual space;

when the first medicine taboo rule moves into the virtual sphere, the first medicine taboo rule is effective, anonymous disclosure of the first medicine taboo rule is stopped, and meanwhile, the first medicine taboo rule is used as a second medicine taboo rule;

acquiring a preset blank database, and storing a second medicine taboo rule into the blank database;

and when the second medicine taboo rules which need to be stored in the blank database are all stored in the blank database, taking the blank database as a medicine taboo rule base to finish the establishment.

Preferably, the method for implementing prescription auditing based on the rule engine further includes:

acquiring a preset ratio set, wherein the ratio set comprises: a plurality of first proportion of first medicine contraindication rules which are historically input by doctors and are effective in the medicine contraindication rule set;

establishing a time axis, and representing the first ratio on the time axis based on the generation time of the first ratio;

randomly selecting a first ratio on a time axis and taking the first ratio as a second ratio;

if the second proportion is less than or equal to a preset first proportion threshold, determining a first malicious value corresponding to the second proportion based on a preset first malicious value library;

if a third ratio smaller than or equal to a preset second ratio threshold exists in the first ratios of the preset first number before and/or after the second ratio, determining a second malicious value corresponding to the third ratio based on a preset second malicious value library;

if a fourth proportion smaller than or equal to a preset third proportion threshold exists in a preset second number of second proportions in the direction of deviating from the second proportion by a third proportion farthest from the second proportion before and/or after the second proportion, determining a third malicious value corresponding to the fourth proportion based on a preset third malicious value library;

summarizing the first malicious value, the second malicious value and the third malicious value to obtain a malicious value sum;

determining a malicious value and a corresponding first down-regulation amplitude based on a preset first down-regulation amplitude library, and down-regulating a first guarantee value of a guarantee person based on the first down-regulation amplitude;

and determining a malicious value and a corresponding second down-regulation amplitude based on a preset second down-regulation amplitude library, and performing down-regulation on a second guarantee value of the guarantee mechanism based on the second down-regulation amplitude.

The system for realizing prescription auditing based on the rule engine provided by the embodiment of the invention comprises:

the building module is used for building a rule engine based on the requirements of the user;

the acquisition module is used for acquiring the prescription to be audited;

and the auditing module is used for auditing the prescription based on the rule engine.

Preferably, the building block performs the following operations:

establishing a drug taboo rule base;

Preferably, the auditing module performs the following operations:

extracting prescription content of a prescription;

Preferably, the building block performs the following operations:

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

FIG. 1 is a flow chart of a method for implementing prescription audit based on a rules engine according to an embodiment of the present invention;

FIG. 2 is a flow chart of another method for implementing prescription review based on a rules engine according to an embodiment of the present invention;

FIG. 3 is a diagram of a system for implementing prescription auditing based on a rules engine according to another embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, which comprises the following steps of:

step S1: constructing a rule engine based on the requirements of the user;

step S2: acquiring a prescription to be audited;

step S3: and auditing the prescription based on the rule engine.

The working principle and the beneficial effects of the technical scheme are as follows:

obtain the prescription to be reviewed [ for example: prescription filled by doctor); based on user [ e.g.: needs of a certain medical institution [ for example: which drugs the organization relates to, build the rules engine on a targeted basis; checking the prescription based on the rule engine, if the prescription passes the checking, distributing the prescription to a corresponding patient, and otherwise returning;

when the prescription checking engine is used, the prescription can be reasonably used for checking the prescription by inputting the prescription into the rule engine, intelligent checking is realized, the problems of slow speed, poor accuracy and difficulty in supervision of manual checking of the prescription are solved, meanwhile, a medical professional user can construct checking rules in a targeted mode, the logic of each rule can be traced and checked, and the specialty, reliability and expandability of the prescription checking engine are improved.

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, as shown in fig. 2, step S1: building a rule engine based on the requirements of the user, comprising:

s101: extracting a plurality of requirement items in the requirement, wherein the requirement items comprise: a first drug;

s102: establishing a drug taboo rule base;

s103: determining at least one taboo rule corresponding to the first medicine based on a medicine taboo rule library;

s104: acquiring a preset Drools engine, and inputting a tabu rule into the Drools engine;

s105: when the taboo rules are all input into the Drools engine, the Drools engine is used as a rule engine to complete the construction.

the preset Drools engine is specifically: an open source code rule engine written in Java language;

for example: if a certain medical institution only designs n medicines, the n medicines are taken as each demand item, a taboo rule corresponding to each demand item is determined from a medicine taboo rule library, and is input into a Drools engine to configure a rule engine;

the embodiment of the invention constructs the rule engine based on the user requirements, and has pertinence; meanwhile, a medicine taboo rule base is arranged, so that taboo rules corresponding to medicines can be conveniently acquired, and the working efficiency of the system is improved;

the rule engine originates from a rule-based expert system, which is one of the branches of the expert system. An expert system falls into the category of artificial intelligence that mimics human reasoning, uses heuristic methods to reason, and uses human-understandable terms to interpret and document its reasoning conclusions. The prescription auditing rule engine employs medical experts to preset and store the calculation logic (rule) of the auditor in the rule engine, receives fact data (prescription) in the service, takes the fact as a parameter into the rule, and obtains the result of the auditor through logic calculation.

Rete matching algorithm is adopted among a plurality of rules. The Rete algorithm is currently the most widely used rule matching algorithm, invented by Charles l. The Rete algorithm is a fast Forward-changing inference algorithm, and the matching speed is independent of the number of rules. The high efficiency of Rete comes mainly from two important assumptions:

temporal redundancy: the changes of the artifacts in the inference process are slow, i.e., only a few artifacts change in each execution cycle, so that the affected rules are only a small percentage. Only already matched artifacts per execution cycle may be considered. Many rules often contain similar patterns and pattern groups.

The basic idea of Rete algorithm is to save all the information left in the past matching process in exchange for execution efficiency at a spatial cost. For each pattern, a matching element table is added to record all elements in the WorkingMemory which can be matched with the matching element table. When a new element is added into the Workingmemory, finding out all patterns which can be matched with the new element, and adding the element into a matching element table; when a null is deleted from the WorkingMemory, all patterns matching the element are also found and the element is deleted from the matching elements table. The Rete algorithm accepts a description of the modify operation on the working memory, resulting in an action to modify the conflict set.

The Rete algorithm steps are as follows:

initial data (fact) is input into the Working Memory.

Pattern Matcher comparison rules (rule) and data (fact) were used.

If there is a conflict (conflict) in executing a rule, i.e., multiple rules are activated simultaneously, the conflicting rules are placed in a conflict set.

And resolving the conflict, and putting the activated rules into Agenda in sequence.

The rules in Agenda are executed using a rules engine. Repeating the steps 2 to 5 until all the rules in Agenda are executed;

drools is an open source business rule engine based on the Charles Forgy's RETE algorithm, easy to access enterprise policies, easy to adjust and easy to manage, meets the standards in the industry, and is fast and efficient. The business analyst or auditor can use it to easily view the business rules to verify whether the encoded rules implement the desired business rules.

The categories of contraindication rules for drugs include, as shown in the following table:

the Drools engine sets specific rule algorithms for different medication contraindications.

The rule algorithm comprises rule selectable condition parameters and values, rule optional condition parameters and values, and logical relations between the condition parameters and values and the result of the auditor.

Optional conditions include:

gender, available value: male, female;

age, available value: numerical values, age units: year, month, day, minimum age, maximum age;

dose, available value: numerical values, dosage units: setting a dose unit and a conversion dictionary; single minimum dose, single maximum dose, single daily minimum dose, single daily maximum dose;

frequency, available value: setting a frequency list;

route of administration, available values: setting a route list;

the simultaneous administration of the medicines has the available value: a drug name or keyword;

simultaneous diagnosis, available values: disease name, disease code or keyword;

recipe origin, available value: setting a source list;

drug specification, available value: a list of available specifications for current drugs;

manufacturer, availability value: a list of current drug manufacturers;

the necessary conditions comprise:

the indication rule is as follows: the necessary conditions are as follows: diagnosis, logical relationship: non-inclusive, multiple diagnostic intervals are

Contraindication rules: the necessary conditions are as follows: diagnosis, logical relationship: comprising, between multiple diagnoses is or

The administration frequency rule is as follows: the necessary conditions are as follows: route of administration, logical relationship: is not composed of

The administration dosage rule is as follows: the necessary conditions are as follows: dose, including single, single day, maximum, minimum, single day minimum, logical relationship: less than, maximum for single day, logical relationship: greater than, single minimum, logical relationship: less than, single maximum, logical relationship: greater than;

among dosage parameters, the logic relation is OR;

the administration route rules are as follows: the necessary conditions are as follows: route of administration, logical relationship: non-inclusive;

interaction rules: a step of (1) a step of (b); selecting conditions as follows: simultaneous medication, logical relationship: comprises that the logic relationship among multiple medicines is OR

The repeated medication rule is as follows: the necessary conditions are as follows: simultaneous medication, logical relationship: comprises that the logic relationship among multiple medicines is OR;

the incompatibility rule of the injection is as follows: the necessary conditions are as follows: simultaneous medication, logical relationship: comprises that the logic relationship among multiple medicines is OR;

age regulation: the necessary conditions are as follows: age;

minimum age, logical relationship: is less than;

maximum age, logical relationship: greater than;

between the maximum and minimum ages, the logical relationship is OR;

sex rules: the necessary conditions are as follows: gender, logical relationship: not equal to;

each rule is combined by a plurality of optional conditions and a necessary condition, and the combination logic operates according to the logical relation of 'AND' of all the conditions. The multiple logic relation calculation between the interior and the condition of the condition is carried out, the calculation principle is that the relation between multiple parameters is firstly calculated in the condition, for example, the relation between multiple parameters is firstly calculated when multiple parameters exist in the condition, then the integral condition logic is calculated, and finally the logic operation between the conditions is carried out. Firstly, two-by-two operation is carried out on the condition or parameter to obtain a result, and then two-by-two operation is carried out on the result and the subsequent logic until the final logic is reached

The logical and (and), or, non-operation refers to the following operational relationship:

for example: the indication rules, when drug, for non-diagnostic use are: alarm should be given when the heart peduncle and the cerebral peduncle are diseased.

The logic diagnosis logic in the condition is not, the logic between diagnoses is or, the overall logic relationship is not (A or B), namely not (myocardial infarction or cerebral infarction) is hit, and an alarm is needed.

If optional conditions are added, for example: the prescription source is an outpatient prescription, and the non-diagnosis is: alarm should be given when the heart peduncle and the cerebral peduncle are diseased.

The logic relationship is as follows: is the hit between the clinic prescription and non (myocardial infarction or cerebral infarction)

The concatenation Drools grammar is

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, which comprises the following steps of S3: based on the rule engine, auditing the prescription, including:

extracting prescription content of a prescription;

extract prescription content [ for example: gender, age, medical history, diagnosed diseases, prescription drugs, usage amount, etc.) by directly inputting the prescription content into a rule engine, calling different rules by the rule engine, and executing corresponding rules each time to determine whether unqualified places exist in the prescription content;

the embodiment of the invention audits the prescription based on the rule engine, and has higher efficiency.

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, which establishes a drug taboo rule base and comprises the following steps:

the preset physician set specifically comprises: a collection containing a plurality of physicians who contract with the system [ hire their rules of providing medication contraindications ]; the preset radius length library specifically comprises: the database stores radius lengths which correspond to different types of credit values together; the preset virtual space is specifically as follows: for example, a three-dimensional space; the preset bad feature library specifically comprises: a database storing a plurality of adverse outcome characteristics and severity values corresponding to the adverse outcome characteristics; the preset initial length library specifically comprises: a database, which stores the initial lengths corresponding to different severity values; the preset node set specifically includes: a set, which contains a plurality of nodes (capable of acquiring data); the preset time interval is specifically as follows: for example, 20 minutes; the preset first moving distance library specifically comprises: the database stores moving distances corresponding to different types of credibility; the preset second moving distance library is similar to the preset second moving distance library; the preset blank database specifically comprises the following steps: a database having no content therein;

a warranty system is established, and when a doctor signs up with the system, at least one warrantee [ for example: other contracted physicians ] and at least one warranty institution [ for example: in the medical institution); when the object of the guarantor or the guarantor institution has bad records, the corresponding credit value will be adjusted downwards; determining the radius length based on the credit value, wherein the smaller the credit value is, the better the physician is, and the larger the radius length is, the easier the tabu rule is to enter the sphere; the greater the serious value of adverse consequences of the taboo rule is, the more easily the taboo rule cannot be put into the ball body, and the longer moving distance (initial length) is set; the node is used for disclosing the taboo rule (a large number of doctors are hired with higher cost, a part of doctors can be hired to only provide the taboo rule, and the rest doctors can only audit the public taboo rule), the more the praise persons are, the greater the credibility of the praise persons and the credibility of the node (the credibility of the node is determined based on the qualification of the auditor), the greater the moving distance is in each time; after the taboo rule continuously moves, entering a sphere to indicate that the taboo rule is qualified, and inputting the taboo rule into a blank database; active access [ approver self-browse access ] and passive access [ system-accurate push, for example: a certain medicine belongs to a certain field, and has different values to be pushed to corresponding praise in the certain field, so that the moving distance should be determined separately;

the embodiment of the invention establishes a guarantee system, and when a doctor produces bad behaviors, the doctor affects both a guarantee person and a guarantee mechanism for guaranteeing the doctor, so that the cost of producing the bad behaviors is increased, and the error rate is reduced; meanwhile, the radius is determined based on the guarantee information, the initial length is determined based on the adverse consequence of the tabu item, the moving distance of each time is determined based on the approval condition, the setting is reasonable, manual verification is not needed, and the working efficiency of the system is improved to a great extent; in addition, reasonable division of labor for the employed physicians can reduce costs.

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, which obtains a first credibility of a approver and comprises the following steps:

acquiring verification information of the approver;

acquiring a preset forward verification mode set, wherein the forward verification mode set comprises: the method comprises the steps that a plurality of forward verification modes are adopted, meanwhile, forward verification is conducted on verification information based on the forward verification modes, and a first verification result is obtained;

acquiring a preset reverse verification mode set, wherein the reverse verification mode set comprises: a plurality of groups of at least one reverse verification mode corresponding to the forward verification mode, and meanwhile, reverse verification is carried out on the verification information based on the reverse verification mode to obtain a second verification result;

calculating a first credibility of the approver based on the first verification result and the second verification result, wherein a calculation formula is as follows:

wherein D is₁For the first degree of confidence, γ_t,iThe method comprises the steps of carrying out forward verification on verification information based on a tth forward verification mode to obtain an ith first result value n in a first verification result_tThe total number of first result values in a first verification result obtained after forward verification is performed on the verification information based on the tth forward verification mode, O is the total number of the forward verification modes, and sigma_t,α,lIs the ith second result value J in the second verification result obtained after the verification information is reversely verified based on the alpha reverse verification mode corresponding to the t forward verification mode_αA total number of second result values, mu, in a second verification result obtained after reverse verification of the verification information based on an alpha-th reverse verification mode corresponding to the t-th forward verification mode_tIs the total number of reverse verification patterns corresponding to the tth forward verification pattern, θ₁And theta₂Is a preset weight value, beta₁And beta₂Is a predetermined comparison threshold value, A_tB is the total number of first result values which are less than or equal to a preset first result value threshold value in first verification results obtained after forward verification is carried out on the verification information based on the tth forward verification mode_αF is an intermediate variable, and else is else, in order to obtain a total number of second result values which are less than or equal to a preset second result value threshold in a second verification result obtained after the verification information is reversely verified based on an alpha-th reverse verification mode corresponding to a t-th forward verification mode;

wherein obtaining a second confidence level of the node comprises:

calculating a second reliability based on a plurality of the approved first reliabilities corresponding to the nodes, wherein the calculation formula is as follows:

wherein D is₂To said second degree of confidence, D_1,mFor the mth said approver's first confidence level corresponding to said node, Q is the total number of said approvers corresponding to said node.

the preset forward verification mode set specifically comprises the following steps: a set including a plurality of forward authentication modes; the preset reverse verification mode set specifically comprises the following steps: a set including a plurality of reverse authentication modes; the preset comparison threshold specifically comprises: for example, 85 and 90, respectively; the preset first result threshold specifically includes: for example, 78; the preset second result threshold is specifically: for example, 82;

obtain approval person's verification information [ for example: certificate of authority, work history, winning records, etc.; first, a forward authentication approach [ e.g.: verifying whether the certificate of authority contains a pharmacist certificate ] verifying the verification information, and then adopting at least one reverse verification mode corresponding to the forward verification mode [ for example: if the praise has the pharmacist certificate, the post experience which must have the pharmacist certificate is deduced reversely, whether bad behaviors exist in the post experience or not and which excellent behaviors exist in the post experience are verified, and the verification information is verified; calculating a first credibility of the approver based on each verification result; calculating a second credibility based on the first credibility of the approver corresponding to the node;

in order to avoid accidental errors of the result, the result comprises a plurality of result values, and the larger the result value is, the more the verification result in the verification mode is consistent;

in the formula, if the result value is larger, the first reliability is also larger;

the overall situation of the first result value is good,

the same way is adopted; if both are satisfied, the first confidence level of the approver should be adjusted to be greater;

the embodiment of the invention carries out forward verification and reverse verification on the verification information of the approver, calculates the first credibility based on the verification result, is reasonable in setting, and quickly judges the credibility of the approver; calculating a second credibility based on the first credibility of each approver, and comprehensively judging the credibility of the nodes; the working efficiency of the system is improved to a great extent; meanwhile, the reverse verification mode corresponds to the forward verification mode, and the credibility is determined more finely.

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, which further comprises the following steps:

the preset first ratio threshold specifically is as follows: for example, 0.95; the preset second ratio threshold specifically is: for example, 0.9; the preset third ratio threshold specifically is: for example, 0.85; the preset first malicious value library specifically comprises: a database, storing malicious values corresponding to different proportions; the preset second malicious value library and the preset third malicious value library have the same principle, but the corresponding malicious values in the first malicious value library, the second malicious value library and the third malicious value library are sequentially increased in the same proportion, and the severity is different; the preset first number specifically comprises: for example, 5; the preset second number is specifically: for example, 4; the preset first down-regulation amplitude library specifically comprises: the database stores first down-regulation amplitudes corresponding to different malicious values; the preset second down-regulation amplitude value is the same as the preset second down-regulation amplitude value, but the corresponding down-regulation amplitude of the mechanism is larger due to the same malicious value;

the credit values of the guarantor and the guarantor are adjusted downwards based on the occupation ratio generated by the doctor in history, meanwhile, a time axis is established, the adjusted amplitude values with different sizes are determined based on different severity degrees of different events, the setting is reasonable, and the punishment of the guarantor is realized.

An embodiment of the present invention provides a system for implementing prescription audit based on a rule engine, as shown in fig. 3, including:

the building module 1 is used for building a rule engine based on the requirements of users;

the acquisition module 2 is used for acquiring a prescription to be audited;

and the auditing module 3 is used for auditing the prescription based on the rule engine.

The embodiment of the invention provides a system for realizing prescription auditing based on a rule engine, wherein a construction module 1 executes the following operations:

establishing a drug taboo rule base;

the embodiment of the invention constructs the rule engine based on the user requirements, and has pertinence; meanwhile, the medicine taboo rule base is arranged, so that taboo rules corresponding to medicines can be conveniently acquired, and the working efficiency of the system is improved.

The embodiment of the invention provides a system for realizing prescription auditing based on a rule engine, wherein an auditing module 3 executes the following operations:

extracting prescription content of a prescription;

The embodiment of the invention provides a method for realizing prescription auditing based on a rule engine, wherein a construction module 1 executes the following operations:

acquiring verification information of the approver;

wherein D is₁For the first degree of confidence, γ_t,iThe method comprises the steps of carrying out forward verification on verification information based on a tth forward verification mode to obtain an ith first result value n in a first verification result_tThe total number of first result values in a first verification result obtained after forward verification is performed on the verification information based on the tth forward verification mode, O is the total number of the forward verification modes, and sigma_t,α,lIs the ith second result value J in the second verification result obtained after the verification information is reversely verified based on the alpha reverse verification mode corresponding to the t forward verification mode_αA total number of second result values, mu, in a second verification result obtained after reverse verification of the verification information based on an alpha-th reverse verification mode corresponding to the t-th forward verification mode_tIs the total number of reverse verification patterns corresponding to the tth forward verification pattern, θ₁And theta₂Is a preset weight value, beta₁And beta₂Is a predetermined comparison threshold value, A_tThe total number of first result values which are less than or equal to a preset first result value threshold value in first verification results obtained after forward verification is carried out on the verification information based on the tth forward verification mode，B_αF is an intermediate variable, and else is else, in order to obtain a total number of second result values which are less than or equal to a preset second result value threshold in a second verification result obtained after the verification information is reversely verified based on an alpha-th reverse verification mode corresponding to a t-th forward verification mode;

wherein obtaining a second confidence level of the node comprises:

the overall situation of the first result value is good,

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims

1. A method for realizing prescription auditing based on a rule engine is characterized by comprising the following steps:

step S1: constructing a rule engine based on the requirements of the user;

step S2: acquiring a prescription to be audited;

step S3: and auditing the prescription based on the rule engine.

2. The method for implementing prescription auditing based on rule engine according to claim 1, wherein the step S1: building a rule engine based on the requirements of the user, comprising:

establishing a drug taboo rule base;

determining at least one taboo rule corresponding to the first medicine based on the medicine taboo rule library;

acquiring a preset Drools engine, and inputting the tabu rule into the Drools engine;

and when the taboo rules are all input into the Drools engine, the Drools engine is used as a rule engine to complete construction.

3. The method for implementing prescription auditing based on rule engine according to claim 1, wherein the step S3: based on the rule engine, auditing the prescription, including:

extracting prescription content of the prescription;

inputting the prescription content into the rule engine to obtain an auditing result;

determining whether the prescription is qualified based on the audit result.

4. The method for auditing prescriptions based on a rules engine of claim 2 wherein establishing a drug contraindication rule base comprises:

obtaining a preset physician set, the physician set comprising: a plurality of physicians;

obtaining warranty information of the physician, the warranty information comprising: at least one guarantor and at least one guaranty institution;

acquiring a first credit value of the guarantor, summarizing the first credit value and acquiring a first credit value sum;

determining the first credit value and the radius length which corresponds to the second credit value and the sum together based on a preset radius length library;

acquiring a preset virtual space, randomly representing a first position of the doctor in the virtual space, and meanwhile, constructing a virtual sphere by taking the first position as a sphere center and the radius length as a radius;

acquiring a drug contraindication rule set input by the doctor, wherein the drug contraindication rule set comprises: a plurality of first drug contraindication rules, the first drug contraindication rules comprising: a second drug, a contra-indicated item corresponding to the second drug, and an adverse consequence corresponding to the contra-indicated item;

performing semantic segmentation on the bad results to obtain a plurality of bad result items;

determining the severity value and the corresponding initial length based on a preset initial length library;

a second position in the virtual space, which is away from the center of the sphere in any direction and has the initial length, represents a rule corresponding to the first medicine taboo, and a connecting line between the center of the sphere and the second position is used as a moving route;

the node issues the first medicine taboo rule, anonymous disclosure is carried out, and a disclosure result is obtained at preset time intervals, wherein the disclosure result comprises the following steps: the first medicine contraindication rule is newly approved by at least one approver;

acquiring a first credibility of the approver, and acquiring a second credibility of the node;

acquiring an access mode of the approver for accessing the first medicine taboo rule, wherein the access mode comprises the following steps: active access and passive access;

when the access mode of the approver is active access, determining a first moving distance corresponding to the first credibility and the second credibility together based on a preset first moving distance library, and simultaneously moving the first medicine taboo rule to the center of the sphere by the first moving distance along the moving path in the virtual space;

when the access mode of the approver is passive access, determining a second moving distance corresponding to the first credibility and the second credibility together based on a preset second moving distance library, and simultaneously moving the first medicine taboo rule to the center of the sphere along the moving path within the virtual space by the second moving distance;

acquiring a preset blank database, and storing the second medicine taboo rule into the blank database;

5. The method of claim 4, further comprising:

acquiring a preset proportion set, wherein the proportion set comprises: a plurality of first odds of the first drug contraindication rule that is historically valid in the set of drug contraindication rules entered by the physician;

randomly selecting the first ratio on the time axis and taking the first ratio as a second ratio;

if the second proportion is smaller than or equal to a preset first proportion threshold, determining a first malicious value corresponding to the second proportion based on a preset first malicious value library;

if a fourth proportion smaller than or equal to a preset third proportion threshold exists in a preset second number of second proportions in the direction of deviating from the second proportion by the third proportion farthest from the second proportion before and/or after the second proportion, determining a third malicious value corresponding to the fourth proportion based on a preset third malicious value library;

determining the malicious value and a corresponding first descending amplitude based on a preset first descending amplitude library, and descending a first guarantee value of the guarantor based on the first descending amplitude;

and determining the malicious value and a corresponding second downward adjustment amplitude based on a preset second downward adjustment amplitude library, and downward adjusting a second guarantee value of the guarantee mechanism based on the second downward adjustment amplitude.

6. A system for implementing prescription audit based on a rules engine, comprising:

the acquisition module is used for acquiring the prescription to be audited;

7. The system for implementing prescription audit based on rules engine of claim 6 wherein the construction module performs the following operations:

establishing a drug taboo rule base;

8. The system for implementing prescription audit based on rules engine of claim 6 wherein the audit module performs the following operations:

extracting prescription content of the prescription;

determining whether the prescription is qualified based on the audit result.

9. The system for implementing prescription audit based on rules engine of claim 7 wherein the build module performs the following operations:

10. The system for implementing prescription audit based on rules engine of claim 9 wherein the build module performs the following operations: