CN110544538A - Five-organ attribution judging system based on concept of traditional Chinese medicine - Google Patents
Five-organ attribution judging system based on concept of traditional Chinese medicine Download PDFInfo
- Publication number
- CN110544538A CN110544538A CN201910786244.XA CN201910786244A CN110544538A CN 110544538 A CN110544538 A CN 110544538A CN 201910786244 A CN201910786244 A CN 201910786244A CN 110544538 A CN110544538 A CN 110544538A
- Authority
- CN
- China
- Prior art keywords
- module
- organ
- learning
- formula
- prescription
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/90—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to alternative medicines, e.g. homeopathy or oriental medicines
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Pharmacology & Pharmacy (AREA)
- Medicinal Chemistry (AREA)
- Toxicology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Alternative & Traditional Medicine (AREA)
- Chemical & Material Sciences (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a system for judging five-organ attribution of traditional Chinese medicine concepts, which is characterized in that various judgment rule tree sets are constructed based on a data source matched by a prescription and a traditional Chinese medicine pathogenesis, and then the judgment rule tree sets are combined to form a traditional Chinese medicine five-organ judgment model, and the five-organ attribution of different traditional Chinese medicine concepts is quantitatively judged through the model. The system comprises a keyword screening module, a prescription data processing module, a five-organ discrimination model generation module, a five-organ model application and result output module and an intelligent regulation and control module. The system adopts a data mining technology and a computer artificial intelligence technology, applies VS.net and R language toolkits (random forest), simulates the whole process of establishing corresponding thinking of Chinese medical evidence to judge the Chinese medical five-organ model and applying the model to quantify the five-organ attributes of different Chinese medical concepts. The invention provides a new research method tool for the research of Chinese medicine literature, in particular to the modern scientific explanation of the basic theory of Chinese medicine.
Description
Technical Field
the invention relates to the technical field of data mining and computer artificial intelligence, in particular to a five-organ attribution judging system of a traditional Chinese medicine concept.
Background
with the development of data technology and artificial intelligence, the prescription information which originally exists in a large amount is digitalized and standardized, and a solid foundation is laid for further intellectualization. Because the corresponding thinking mode of the traditional Chinese medicine dialectics is a unique thinking mode established on the basis of understanding and associating prescription data information, the data mining and artificial intelligence technology, particularly the application of the latest machine learning method, is adopted, so that the process of corresponding thinking of the traditional Chinese medicine dialectics in computer simulation is realized.
data mining generally refers to a process of searching information hidden in a large amount of data through an algorithm, and is generally related to computer science, and the above-mentioned objects are achieved through various methods such as statistics, online analysis and processing, information retrieval, machine learning, expert systems (depending on past rules of thumb), pattern recognition and the like. Artificial intelligence is the subject of research on making computer to simulate some human thinking process and intelligent behavior (such as learning, reasoning, thinking, planning, etc.), and mainly includes the principle of computer to implement intelligence and the manufacture of computer similar to human brain intelligence to make computer implement higher-level application. The corresponding indications refer to the presumption of the pathogenesis or symptoms of the treated subjects according to the herb and flavor composition and the efficacy of the formula.
Patent application No. 201910142196.0 discloses a quantitative determination system for similarity of traditional Chinese medicine names, which is based on a data source formed by matching prescription compositions with traditional Chinese medicine names, constructs various determination rule tree sets, and further combines the rule tree sets to form a traditional Chinese medicine name determination model, so that the similarity of different disease names is quantified through the model, and the correlation between the traditional Chinese medicine names can be quantified based on the system. The defects are as follows:
1) the quantitative determination system is required to be based on the existing disease names and keywords, and the professional knowledge of the traditional Chinese medicine theory, the traditional Chinese medicine treatment method, the traditional Chinese medicine formula and the traditional Chinese medicine can not be comprehensively applied, so that the application range of the quantitative determination system is limited.
2) The learning model construction of the quantitative determination system needs to establish a learning set and a corresponding comparison set. Because many concepts of traditional Chinese medicine are mutually contained, only a learning set can be established but a control set cannot be established, which further limits the application range of the quantitative determination system.
Disclosure of Invention
the invention aims to provide a system for judging the five-organ affiliation of the traditional Chinese medicine concept, which combines data mining, artificial intelligence and massive traditional Chinese medicine prescription data, thereby simulating the corresponding thinking process of the traditional Chinese medicine prescription and quantifying the five-organ affiliation degree of different traditional Chinese medicine concepts.
The technical problem solved by the invention can be realized by adopting the following technical scheme:
A system for determining the affiliation of the five zang organs in the concept of traditional Chinese medicine comprises
The keyword screening module A is used for inputting the five internal organs and the Chinese medicine concepts needing to be judged and outputting keywords and weights; the specific process comprises the following steps: constructing keywords by combining the default concept of the five zang organs with the basic theory of the traditional Chinese medicine and the basic concept of a dialectical system, wherein the weight value of the keywords is 1; the weight value can be automatically adjusted through parameter feedback of the prescription data processing module B and the five-organ discrimination model generation module C;
The prescription data processing module B comprises a prescription information collection module B1 and a plurality of learning modules;
The prescription information collection module B1 is used for screening out all data containing keywords from the basic data base, importing the data into the basic data base, and obtaining a basic prescription information set after standardization;
each learning module adopts different rules to divide the standardized basic prescription information set to obtain a corresponding learning set;
after repeated for many times, a judged concept prescription set module B9 is established;
the five-organ discrimination model generation module C comprises a plurality of rule generation modules, and each rule generation module generates a rule tree through regulation and control on the basis of a random forest algorithm on the basis of a corresponding learning set;
All rule trees form a proportional combination five-organ judgment model C8;
and the five-organ model application and result output module D is used for applying the data of the judged concept and prescription collection module B9 to the five-organ combination judgment model C8 according to the proportion, obtaining the similarity degree of the traditional Chinese medicine concept and the five organs to be judged, and outputting a result.
Further, the learning module comprises a five-organ prescription composition set learning module B2, a five-organ prescription dose set learning module B3, a five-organ prescription meridian tropism set learning module B4, a five-organ prescription sex and taste set learning module B5, a five-organ prescription processing set learning module B6, a five-organ prescription decoction method set learning module B7 and a five-organ prescription dynasty set learning module B8.
Further, the learning modules are based on the prescription information collection module B1, and the standardized basic prescription information set is divided by using the composition data amount, the dose data amount, the meridian tropism data amount, the sex-flavor data amount, the processing data amount, the decoction method data amount, and the dynasty data amount as rules through artificial intelligence, so as to obtain a corresponding composition learning set, a dose learning set, a meridian tropism learning set, a sex-flavor learning set, a processing learning set, a decoction method learning set, and a dynasty learning set.
Further, the rule generation module includes a five-organ prescription composition rule generation module C1, a five-organ prescription dose rule generation module C2, a five-organ prescription meridian rule generation module C3, a five-organ prescription taste rule generation module C4, a five-organ prescription processing rule generation module C5, a five-organ prescription decoction method rule generation module C6, and a five-organ prescription meridian rule generation module C7.
Further, the rule generation module generates a corresponding rule tree through artificial intelligence regulation and control on the basis of a random algorithm on the basis of a five-organ prescription composition set learning module B2, a five-organ prescription dose set learning module B3, a five-organ prescription meridian tropism set learning module B4, a five-organ prescription sex set learning module B5, a five-organ prescription processing set learning module B6, a five-organ prescription decoction method set learning module B7 and a five-organ prescription dynasty set learning module B8.
Furthermore, an intelligent regulation module E is established on the proportional combination five-organ judgment model C8 and is used for intelligently regulating the keyword screening module A and the prescription data processing module B according to the entropy of the generated rule tree and the out-of-bag error of the rule set.
Further, the method for adjusting the weight value of the keyword comprises the following steps: the rationality of the authority change is verified by directionally increasing or decreasing the weight of a certain keyword and comparing the verification parameters of the overall model after the weight is changed; if the weight is reasonable, the weight is kept to be changed; if not, restoring the initial weight; and repeating the process to complete the weight adjustment of all the keywords and obtain the most reasonable corresponding weight of the model.
compared with the prior art, the invention has the beneficial effects that:
1) The method establishes a five-organ training set by comprehensively applying the traditional Chinese medicine theory, the traditional Chinese medicine treatment method, the traditional Chinese medicine formula and the professional knowledge of traditional Chinese medicine and establishing a keyword word bank of related concepts of the five organs. Compared with a judgment system established based on the existing disease names and keywords, the method has wider universality.
2) in the model training of machine learning, the similarity can be judged by establishing a learning set with the attribute of 'yes' without establishing a corresponding attribute of 'no-me' contrast set. The advantages of this method are: because many concepts of traditional Chinese medicine are mutually involved, it is often difficult to find out a "reference set" rather than a "learning set". Compared with a judgment system established based on the existing disease names and keywords, the invention has wider application range for the research of the concept of the traditional Chinese medicine.
Drawings
Fig. 1 is a schematic diagram of a system for determining the affiliation of the five zang organs according to the concept of the present invention.
Fig. 2 is a schematic flow chart of keywords and weights according to the present invention.
Fig. 3 is a schematic diagram illustrating a method for adjusting a weight value of a keyword according to the present invention.
Fig. 4 is a schematic diagram illustrating a method for adjusting a weight value of a keyword according to the present invention.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.
Referring to fig. 1 and 2, the system for determining the affiliation of five zang-organs in the concept of traditional Chinese medicine according to the present invention includes
The module A, namely a keyword screening module, inputs the five internal organs and the Chinese medicine concepts needing to be judged, and automatically screens related keywords through artificial intelligence.
the module B, namely a prescription data processing module, comprises a B1 prescription information collection module, a B2 five-organ prescription composition collection learning module, a B3 five-organ prescription dose collection learning module, a B4 five-organ prescription meridian collection learning module, a B5 five-organ prescription and taste collection learning module, a B6 five-organ prescription processing collection learning module, a B7 five-organ prescription decocting method collection learning module, a B8 five-organ prescription generation collection learning module and a B9 judged concept prescription collection module.
on the basis of the module A, a module B1, namely a prescription information collection module, is established, a five-organ model prescription information collection and a prescription data collection corresponding to the Chinese medicine concept to be judged are intelligently screened on the basis of a basic data total database, and the five-organ model prescription information collection and the prescription data collection are standardized through an artificial intelligence technology.
On the basis of the module B1, a module B2, namely a five-organ prescription composition set learning module, is established, and a proper composition data volume is selected through artificial intelligence to segment the standardized basic prescription information set so as to segment a proper composition learning set.
on the basis of the module B1, a module B3, namely a five-organ prescription dose set learning module, is established, and a proper dose data volume is selected through artificial intelligence to segment the standardized basic prescription information set and segment a proper dose learning set.
On the basis of the module B1, a module B4 is established, namely a five-organ meridian tropism set learning module, a proper meridian tropism data volume is selected through artificial intelligence, a standardized basic prescription information set is segmented, and a proper meridian tropism learning set is segmented.
on the basis of the module B1, a module B5, namely a five-organ prescription nature and taste set learning module, is established, and a proper nature and taste data volume is selected through artificial intelligence to segment the standardized basic prescription information set and segment a proper nature and taste learning set.
on the basis of the module B1, a module B6, namely a learning module of the five-organ prescription processing set, is established, and through artificial intelligence and selection of appropriate processing data volume, the standardized basic prescription information set is divided into appropriate processing learning sets.
On the basis of the module B1, a module B7, namely a five-organ prescription dose set decocting method module, is established, a proper decocting method data volume is selected through artificial intelligence, a standardized basic prescription information set is divided, and a proper decocting method learning set is divided.
On the basis of the module B1, a module B8, namely a five-organ prescription dynasty set learning module, is established, a proper dynasty data volume is selected through artificial intelligence, the standardized basic prescription information set is divided, and a proper dynasty learning set is divided.
by repeating the modules B2, B3, B4, B5, B6, B7, and B8, a module B9 is established, i.e., a concept prescription set module is determined, i.e., an application set consisting of corresponding prescriptions, dosage, meridian tropism, flavor, processing, decoction method, and generation data.
On the basis of the module B, a module C, namely a five-zang-organ judgment model generation module is established and comprises a C1 five-zang-organ prescription composition rule generation module, a C2 five-zang-organ prescription dose rule generation module, a C3 five-zang-organ prescription meridian-passing rule generation module, a C4 five-zang-organ prescription sex and taste rule generation module, a C5 five-zang-organ prescription processing rule generation module, a C6 five-zang-organ prescription decoction method rule generation module, a C7 five-zang-organ prescription generation rule generation module and a C8 quantitative judgment model which are combined in proportion.
And on the basis of the module B2, a module C1 is established, namely a rule generation module for the composition of the five-organ formulas, and a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm on the basis of the composition of different formulas of the five-organ formulas.
and on the basis of the module B3, a module C2, namely a five-organ prescription dose rule generation module, is established, and a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm on the basis of different prescription doses of the five organs.
And on the basis of the module B4, a module C3 is established, namely a five-organ prescription meridian tropism rule generation module, and a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm on the basis of meridian tropism of different prescriptions of the five organs.
on the basis of the module B5, a module C4, namely a five-organ prescription and taste rule generation module, is established, and a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm based on meridian tropism of different prescriptions of each organ of the five organs.
And on the basis of the module B6, a module C5, namely a five-organ prescription processing rule generation module, is established, processing is carried out on the basis of different prescriptions of each organ of the five organs, and a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm.
And on the basis of the module B7, a module C6 is established, namely a five-organ prescription decocting method rule generating module, and on the basis of the different prescription decocting methods of the five organs, a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm.
And on the basis of the module B8, a module C7 is established, namely a five-organ generation rule generation module, and a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm on the basis of different five-organ generation formulas.
On the basis of the modules C1, C2, C3, C4, C5, C6 and C7, a module C8 is established, namely a quantitative judgment model is combined according to the proportion, and the proportion of rule trees of different categories is adjusted and matched by artificial intelligence to form the quantitative judgment model.
On the basis of the modules B9 and C8, a module D, namely a module for applying a five-organ model and outputting a result is established, a module C8 model is applied to the data of the module B9, the similarity between the compared and judged concept and the five-organ is obtained, and the similarity is displayed and output.
At a module C8, a module E, namely an intelligent regulation and control module is established, and according to the total entropy of the spanning rule tree and the total out-of-bag error of the rule set, the screening weight of the keywords of the intelligent regulation module A and the information of the modules B2, B3, B4, B5, B6, B7 and B8 are split.
Examples
Taking the classification of five zang organs in the concept of "three jiao" in traditional Chinese medicine as an example, a classification system of five zang organs is applied.
Module a, i.e., a keyword screening module, inputs the determined concept of the chinese medicine "triple energizer", defaults to the five internal organs "heart", "lung", "spleen", "liver" and "kidney" of the chinese medicine, constructs basic keywords such as "kidney deficiency", "spleen deficiency", "kidney qi", "lung qi", and the like, with an initial weight value of 1, in combination with the basic concepts of "qi", "blood", "phlegm", "deficiency", "excess", and the like of the basic theory and dialectical system of the chinese medicine; and the weight of the basic keyword is automatically adjusted through parameter feedback of the model constructed by the module B and the module C.
The method for adjusting the weight value of the keyword comprises the following steps: the rationality of the authority change is verified by directionally increasing or decreasing the weight of a certain keyword and comparing the verification parameters of the overall model after the weight is changed; if the weight is reasonable, the weight is kept to be changed; if not, restoring the initial weight; and repeating the process to complete the weight adjustment of all the keywords and obtain the most reasonable corresponding weight of the model.
In a weight change mode: increasing; weight increase amplitude: 1; weight increase limit: 5; weight adjustment function: out-of-bag error rate; the specific keywords "heart qi" and "heart deficiency" are examples.
Referring to fig. 3, the initial weight "heart qi" in the module a is 1, and the error rate outside the bag obtained by the module C after the first operation through the module B is 25.87%; and the weight is adjusted to be 2, the second time of operation is carried out, the error rate outside the bag obtained by the module B and the module C is 25.89 percent and is higher than the error rate outside the bag for the first time, the weight change is stopped, and the 'mood' is restored to be 1.
Referring to fig. 4, the initial weight "heart deficiency" in the module a is 1, and the error rate outside the bag obtained by the module C after the first run through module B is 25.87%; the weight is adjusted to be 2, the second time of operation is carried out through the module B and the module C, the out-of-bag error rate is 25.61 percent and is lower than the first out-of-bag error rate; continuing weight change, adjusting the weight to be 3, and running for the third time through the module B and the module C to obtain the out-of-bag error rate of 25.57 percent which is lower than the out-of-bag error rate of the second time; continuing weight change, adjusting the weight to be 4, running for the fourth time through the module B, and obtaining the out-of-bag error rate of 25.54 percent by the module C, wherein the out-of-bag error rate is lower than that of the third time; and (4) continuing weight change, adjusting the weight to be 5, running for the fifth time, passing through the B, obtaining 25.51% of out-of-bag error rate by the C module, wherein the out-of-bag error rate is lower than that of the fourth time, reaching the authority increase limit, stopping continuing to increase, and setting the weight of heart deficiency to be 5.
| Keyword | Weight of |
| kidney deficiency | 1 |
| spleen deficiency | 1 |
| Kidney qi | 1 |
| Lung qi | 1 |
| heart-qi | 1 |
| Lung heat | 1 |
| heart deficiency | 1 |
| Spleen qi | 1 |
| Injuring spleen | 1 |
| …… | …… |
Module B1, a prescription information aggregation Module, in the Foundation databank
And searching keywords screened from the A, such as 'spleen deficiency', 'kidney deficiency' and the like. The data is imported into a basic database and is standardized into the following format.
On the basis of the module B1, a module B2, namely a five-organ prescription composition set learning module, is established, a proper composition data volume is selected through artificial intelligence, the standardized basic prescription information set is divided, a proper composition learning set is obtained through division, and data are stored in the following format:
On the basis of the module B1, a module B3, namely a five-organ prescription dose set learning module, is established, a proper dose data volume is selected through artificial intelligence, a standardized basic prescription information set is segmented, a proper dose learning set is segmented, and data are stored in the following format:
on the basis of the module B1, a module B4 is established, namely a five-organ prescription meridian collection learning module, a proper meridian collection data volume is selected through artificial intelligence, the standardized basic prescription information set is segmented, a proper meridian collection learning set is segmented, and data are stored in the following format:
On the basis of the module B1, a module B5, namely a five-organ prescription characteristic and taste set learning module, is established, a proper characteristic and taste data volume is selected through artificial intelligence, a standardized basic prescription information set is divided, a proper characteristic and taste learning set is divided, and data are stored in the following format:
on the basis of the module B1, a module B6, namely a five-organ prescription processing set learning module, is established, a proper processing data volume is selected through artificial intelligence, the standardized basic prescription information set is divided, a proper processing learning set is obtained through division, and the data is stored in the following format:
on the basis of the module B1, a module B7, namely a learning module of the five-organ prescription decocting method set, is established, a proper decocting method data volume is selected through artificial intelligence, the standardized basic prescription information set is divided, a proper decocting method learning set is divided, and the data are stored in the following formats:
On the basis of the module B1, a module B8 is established, namely a five-organ prescription dynasty set learning module, a proper dynasty data volume is selected through artificial intelligence, the standardized basic prescription information set is divided into proper dynasty learning sets, and the data are stored in the following format:
By repeating the modules B2, B3, B4, B5, B6, B7, and B8, a module B9 is established, that is, a module of information on concept prescriptions of the determined chinese medical science, that is, data is stored in the following format:
on the basis of the module B2, a module C1 is established, namely a five-organ prescription composition rule generation module, on the basis of different prescription compositions of the five organs, a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm, and data are stored in the following format:
On the basis of the module B3, a module C2, namely a five-organ prescription dose rule generation module, is established, on the basis of different prescription doses of the five organs and a random forest algorithm, a certain number of rule trees are generated through artificial intelligence regulation, and data are stored in the following format:
On the basis of the module B4, a module C3 is established, namely a five-organ prescription meridian-returning rule generation module, a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of the meridian returning of different five-organ prescriptions and a random forest algorithm, and data are stored in the following format:
On the basis of the module B5, a module C4 is established, namely a five-organ prescription flavor rule generation module, based on different prescription flavors of the five organs and on the basis of a random forest algorithm, a certain number of rule trees are generated through artificial intelligence regulation, and data are stored in the following format:
On the basis of the module B6, a module C5 is established, namely a five-organ prescription processing rule generation module, processing is carried out on the basis of different prescriptions of the five organs, a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm, and data are stored in the following format:
On the basis of the module B7, a module C6 is established, namely a five-organ prescription decocting method rule generating module, on the basis of different five-organ prescription decocting methods, a certain number of rule trees are generated through artificial intelligence regulation and control on the basis of a random forest algorithm, and data are stored in the following format:
On the basis of the module B8, a module C7 is established, namely a five-organ prescription generation rule generation module, based on the different prescription generations of the five organs and on the basis of a random forest algorithm, a certain number of rule trees are generated through artificial intelligence regulation, and data are stored in the following format:
On the basis of the modules C1, C2, C3, C4, C5, C6 and C7, a module C8 is established, namely, a five-organ judgment model is combined in proportion, the proportions of a rule tree, a dose rule tree, a channel-returning rule tree, a sex rule tree, a processing rule tree, a decoction method rule tree and a generation rule tree are formed by artificial intelligence adjustment and collocation medicaments, a quantitative judgment model is formed, and the total bag-out error format is as follows:
On the basis of the modules B9 and C8, a module D, namely a five-zang model application and result output module is established, the model generated by the module C8 is applied to the data of the module B9, the similarity degree of the three-jiao and the five-zang is obtained, and the output display is carried out.
On the basis of the module C8, a module E, namely an intelligent regulation and control module is established, and relevant basic settings such as the entropy value of the spanning rule tree, the out-of-bag error of the rule set, the segmentation proportion of the intelligent regulation module B2 and the like are set.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. a system for determining the affiliation of the five zang organs in the concept of traditional Chinese medicine is characterized by comprising
the keyword screening module A is used for inputting the five internal organs and the Chinese medicine concepts needing to be judged and outputting keywords and weights; the specific process comprises the following steps: constructing keywords by combining the default concept of the five zang organs with the basic theory of the traditional Chinese medicine and the basic concept of a dialectical system, wherein the weight value of the keywords is 1; the weight value can be automatically adjusted through parameter feedback of the prescription data processing module B and the five-organ discrimination model generation module C;
The prescription data processing module B comprises a prescription information collection module B1 and a plurality of learning modules;
The prescription information collection module B1 is used for screening out all data containing keywords from the basic data base, importing the data into the basic data base, and obtaining a basic prescription information set after standardization;
each learning module adopts different rules to divide the standardized basic prescription information set to obtain a corresponding learning set;
after repeated for many times, a judged concept prescription set module B9 is established;
The five-organ discrimination model generation module C comprises a plurality of rule generation modules, and each rule generation module generates a rule tree through regulation and control on the basis of a random forest algorithm on the basis of a corresponding learning set;
All rule trees form a proportional combination five-organ judgment model C8;
And the five-organ model application and result output module D is used for applying the data of the judged concept and prescription collection module B9 to the five-organ combination judgment model C8 according to the proportion, obtaining the similarity degree of the traditional Chinese medicine concept and the five organs to be judged, and outputting a result.
2. The system for determining the affiliation of five zang organs of a chinese medicine concept according to claim 1, wherein the learning modules include a five zang organs formula composition set learning module B2, a five zang organs formula dose set learning module B3, a five zang organs formula meridian tropism set learning module B4, a five zang organs formula sex and taste set learning module B5, a five zang organs formula processing set learning module B6, a five zang organs formula decoction method set learning module B7, and a five zang organs formula ancestry set learning module B8.
3. The system for determining the affiliation of the five zang organs of the concept of traditional Chinese medicine according to claim 1 or 2, wherein the learning modules are based on a prescription information set module B1, and the learning modules divide the standardized basic prescription information set by using a composition data amount, a dose data amount, a meridian tropism data amount, a sex-flavor data amount, a processing data amount, a decoction method data amount, and a generation data amount as rules through artificial intelligence, so as to obtain a corresponding composition learning set, a dose learning set, a meridian tropism learning set, a sex-flavor learning set, a processing learning set, a decoction method learning set, and a generation learning set.
4. The system for determining visceral manifestation of a chinese medicine concept according to claim 1, wherein the rule generation module includes a five-organ formula composition rule generation module C1, a five-organ formula dose rule generation module C2, a five-organ formula meridian rule generation module C3, a five-organ formula sex and taste rule generation module C4, a five-organ formula processing rule generation module C5, a five-organ formula decoction method rule generation module C6, and a five-organ formula ancestry rule generation module C7.
5. the system for determining the affiliation of the five zang organs of the chinese medical concept according to claim 1 or 4, wherein the rule generation module generates the corresponding rule tree by artificial intelligence control based on the random forest algorithm based on the different composition of the five zang organs, the different formula doses, the different formula channels, the different formula natures, the different formula processing, the different formula decocting method and the different formula generations respectively on the basis of the five zang organ formula composition set learning module B2, the five zang organ formula dose set learning module B3, the five zang organ formula channel-setting set learning module B4, the five zang organ formula sex set learning module B5, the five zang organ formula processing set learning module B6, the five zang organ formula decocting method set learning module B7 and the five zang organ formula generation set learning module B8.
6. The system for determining the five zang organs attribution of the concept of traditional Chinese medicine according to claim 1, wherein an intelligent regulation module E for intelligently regulating the keyword screening module a and the prescription data processing module B according to the entropy of the generated rule tree and the out-of-bag error of the rule set is established on the proportional combination five zang organs determination model C8.
7. The system for determining five organ affiliation of chinese medicine concept according to claim 1, wherein the method of adjusting the keyword weight value is: the rationality of the authority change is verified by directionally increasing or decreasing the weight of a certain keyword and comparing the verification parameters of the overall model after the weight is changed; if the weight is reasonable, the weight is kept to be changed; if not, restoring the initial weight; and repeating the process to complete the weight adjustment of all the keywords and obtain the most reasonable corresponding weight of the model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910786244.XA CN110544538B (en) | 2019-08-23 | 2019-08-23 | Five-organ attribution judging system based on concept of traditional Chinese medicine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910786244.XA CN110544538B (en) | 2019-08-23 | 2019-08-23 | Five-organ attribution judging system based on concept of traditional Chinese medicine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110544538A true CN110544538A (en) | 2019-12-06 |
| CN110544538B CN110544538B (en) | 2022-02-18 |
Family
ID=68712162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910786244.XA Active CN110544538B (en) | 2019-08-23 | 2019-08-23 | Five-organ attribution judging system based on concept of traditional Chinese medicine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110544538B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150213207A1 (en) * | 2012-09-13 | 2015-07-30 | Parkland Center For Clinical Innovation | Holistic hospital patient care and management system and method for automated facial biological recognition |
| CN105528526A (en) * | 2016-01-15 | 2016-04-27 | 胡广芹 | Traditional Chinese medicine inheritance and big data mining-based dynamic health management system for life cycle |
| CN105574340A (en) * | 2015-12-17 | 2016-05-11 | 北京科技大学 | Method and system for quantifying property, flavor and channel tropism of traditional Chinese medicinal prescription |
| CN106778080A (en) * | 2016-11-23 | 2017-05-31 | 小红象医疗科技有限公司 | A kind of Chinese medicinal formulae analysis method and the system for realizing the method |
| CN109215777A (en) * | 2018-08-03 | 2019-01-15 | 电子科技大学 | TCM Document intelligent excavating and prescription aid decision-making method and system |
| CN109887604A (en) * | 2019-02-26 | 2019-06-14 | 上海中医药大学 | A kind of quantization decision-making system of names of disease of tcm similarity |
| CN109903854A (en) * | 2019-01-25 | 2019-06-18 | 电子科技大学 | A kind of core drug recognition methods based on TCM Literature |
-
2019
- 2019-08-23 CN CN201910786244.XA patent/CN110544538B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150213207A1 (en) * | 2012-09-13 | 2015-07-30 | Parkland Center For Clinical Innovation | Holistic hospital patient care and management system and method for automated facial biological recognition |
| CN105574340A (en) * | 2015-12-17 | 2016-05-11 | 北京科技大学 | Method and system for quantifying property, flavor and channel tropism of traditional Chinese medicinal prescription |
| CN105528526A (en) * | 2016-01-15 | 2016-04-27 | 胡广芹 | Traditional Chinese medicine inheritance and big data mining-based dynamic health management system for life cycle |
| CN106778080A (en) * | 2016-11-23 | 2017-05-31 | 小红象医疗科技有限公司 | A kind of Chinese medicinal formulae analysis method and the system for realizing the method |
| CN109215777A (en) * | 2018-08-03 | 2019-01-15 | 电子科技大学 | TCM Document intelligent excavating and prescription aid decision-making method and system |
| CN109903854A (en) * | 2019-01-25 | 2019-06-18 | 电子科技大学 | A kind of core drug recognition methods based on TCM Literature |
| CN109887604A (en) * | 2019-02-26 | 2019-06-14 | 上海中医药大学 | A kind of quantization decision-making system of names of disease of tcm similarity |
Non-Patent Citations (1)
| Title |
|---|
| 都广礼等: "方证与方剂运用", 《陕西中医学院学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110544538B (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Filar et al. | Competitive Markov decision processes | |
| US8793144B2 (en) | Treatment effect prediction system, a treatment effect prediction method, and a computer program product thereof | |
| CN109636061A (en) | Training method, device, equipment and the storage medium of medical insurance Fraud Prediction network | |
| Stevens et al. | A cluster analysis of challenging behaviors in autism spectrum disorder | |
| CN110504031B (en) | Cloud management database establishment method and system for health behavior intervention | |
| CN111008333B (en) | Shopping system and method based on user characteristic information and product component information | |
| CN106529110A (en) | Classification method and equipment of user data | |
| Bekiroglu et al. | Control engineering methods for the design of robust behavioral treatments | |
| KR20110098286A (en) | Self health diagnosis system of oriental medicine using fuzzy inference method | |
| CN110544538B (en) | Five-organ attribution judging system based on concept of traditional Chinese medicine | |
| Smits | On the effect of adding clinical samples to validation studies of patient-reported outcome item banks: a simulation study | |
| CN112037888A (en) | Physiological health characteristic data monitoring method, device, equipment and storage medium | |
| CN109887604B (en) | Quantitative determination system for similarity of traditional Chinese medicine disease names | |
| KR102437661B1 (en) | A System and Method for Classifying diseases using Electronic Medical Record for Animal Hospitals | |
| Trisuciana et al. | Clustering of COVID-19 vaccination recipients in DKI Jakarta using the K-medoids algorithm | |
| Dipro et al. | A federated learning approach for detecting Parkinson’s disease through privacy preserving by blockchain | |
| Farahani et al. | LASSO Based Causal Variable Selection Addressing Time-Varying Confounding in Interdisciplinary Pain Management | |
| CN116795896B (en) | Big data-based rehabilitation exercise strategy generation method and device | |
| Tong | Breast Cancer Prediction from Genome Segments with Machine Learning | |
| CN117688226B (en) | Intelligent pre-diagnosis self-service bill making method and system based on similar child patient matching | |
| Sadhasivam et al. | Diet and workout recommendation using ML | |
| Karim | Causal inference approaches for dealing with time-dependent confounding in longitudinal studies, with applications to multiple sclerosis research | |
| Sun | Exploring the Educational Potential of Online Platforms in Adolescent Physical and Mental Health Promotion | |
| CN116401399A (en) | Music library generation system and method for music therapy | |
| Tyagi et al. | Deep Learning Techniques for using Intelligent Internet of Things Appliances to Protect Health Department Information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |