CN111899828A - Knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system - Google Patents
Knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system Download PDFInfo
- Publication number
- CN111899828A CN111899828A CN202010758816.6A CN202010758816A CN111899828A CN 111899828 A CN111899828 A CN 111899828A CN 202010758816 A CN202010758816 A CN 202010758816A CN 111899828 A CN111899828 A CN 111899828A
- Authority
- CN
- China
- Prior art keywords
- diagnosis
- breast cancer
- treatment
- precondition
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 124
- 206010006187 Breast cancer Diseases 0.000 title claims abstract description 85
- 208000026310 Breast neoplasm Diseases 0.000 title claims abstract description 85
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 32
- 238000013507 mapping Methods 0.000 claims description 12
- 238000012423 maintenance Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 230000010365 information processing Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000003058 natural language processing Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 2
- 238000012549 training Methods 0.000 claims description 2
- JMANVNJQNLATNU-UHFFFAOYSA-N oxalonitrile Chemical compound N#CC#N JMANVNJQNLATNU-UHFFFAOYSA-N 0.000 abstract 1
- 239000003814 drug Substances 0.000 description 9
- 229940079593 drug Drugs 0.000 description 8
- 201000010099 disease Diseases 0.000 description 7
- 206010028980 Neoplasm Diseases 0.000 description 6
- 238000000034 method Methods 0.000 description 6
- 238000003759 clinical diagnosis Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 208000024891 symptom Diseases 0.000 description 3
- 102000053602 DNA Human genes 0.000 description 2
- 108020004414 DNA Proteins 0.000 description 2
- 239000002246 antineoplastic agent Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 210000000481 breast Anatomy 0.000 description 2
- 201000011510 cancer Diseases 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 229920002477 rna polymer Polymers 0.000 description 2
- 101150042997 21 gene Proteins 0.000 description 1
- 206010061623 Adverse drug reaction Diseases 0.000 description 1
- 206010067484 Adverse reaction Diseases 0.000 description 1
- 206010006223 Breast discharge Diseases 0.000 description 1
- 206010006272 Breast mass Diseases 0.000 description 1
- 208000030453 Drug-Related Side Effects and Adverse reaction Diseases 0.000 description 1
- 108090000790 Enzymes Proteins 0.000 description 1
- 102000004190 Enzymes Human genes 0.000 description 1
- 208000005726 Inflammatory Breast Neoplasms Diseases 0.000 description 1
- 206010021980 Inflammatory carcinoma of the breast Diseases 0.000 description 1
- ZOKXTWBITQBERF-UHFFFAOYSA-N Molybdenum Chemical compound [Mo] ZOKXTWBITQBERF-UHFFFAOYSA-N 0.000 description 1
- 102000043276 Oncogene Human genes 0.000 description 1
- 108700020796 Oncogene Proteins 0.000 description 1
- 102000044209 Tumor Suppressor Genes Human genes 0.000 description 1
- 108700025716 Tumor Suppressor Genes Proteins 0.000 description 1
- 230000006838 adverse reaction Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 229940041181 antineoplastic drug Drugs 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 229940044683 chemotherapy drug Drugs 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000000857 drug effect Effects 0.000 description 1
- 230000036267 drug metabolism Effects 0.000 description 1
- 239000003596 drug target Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 210000004602 germ cell Anatomy 0.000 description 1
- 210000004907 gland Anatomy 0.000 description 1
- 230000035876 healing Effects 0.000 description 1
- 201000004653 inflammatory breast carcinoma Diseases 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 210000001165 lymph node Anatomy 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000036210 malignancy Effects 0.000 description 1
- 208000027202 mammary Paget disease Diseases 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000011987 methylation Effects 0.000 description 1
- 238000007069 methylation reaction Methods 0.000 description 1
- 229910052750 molybdenum Inorganic materials 0.000 description 1
- 239000011733 molybdenum Substances 0.000 description 1
- 238000013188 needle biopsy Methods 0.000 description 1
- 239000002773 nucleotide Substances 0.000 description 1
- 125000003729 nucleotide group Chemical group 0.000 description 1
- 230000000771 oncological effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
- 230000008719 thickening Effects 0.000 description 1
- 230000005186 women's health Effects 0.000 description 1
Images
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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- 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
- G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
- G16H20/17—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients delivered via infusion or injection
-
- 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
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medicinal Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention provides a knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system, which constructs a breast cancer knowledge graph model based on multiple guidelines (NCCN, CSCO), supports an index system set and a recommendation scheme set derived from multiple guidelines, and assists a doctor to provide a diagnosis and treatment scheme with multiple continuity of a whole course for a patient.
Description
Technical Field
The application relates to the field of medical information, in particular to a recommendation system for breast cancer diagnosis and treatment schemes.
Background
At present, artificial intelligence technology is in the fields of traffic, finance, security and the like, and for the medical industry, an intelligent medical age is coming from intelligent diagnosis guidance, image recognition, auxiliary diagnosis, new medicine research and development. New modes of medical services featuring informatization and intelligence are emerging. The core of the "information and intelligence" service model is a Clinical Decision Support System (CDSS). The CDSS is a medical information technology application system based on human-computer interaction, and aims to provide clinical decision support for doctors and other health practitioners and complete clinical decision with assistance of data, models and the like. The CDSS can achieve the aim of improving the medical service quality of medical health institutions by reducing the missed diagnosis rate and the misdiagnosis rate and standardizing the diagnosis and treatment behaviors and processes. At present, the leading and authoritative CDSS in foreign technology has the BMJ optimal clinical practice developed by BMJ group, UpToDate developed by Wolter Kluwer medical group and the like. The development of the recent technologies such as artificial intelligence machine learning and natural language processing brings new vitality to the CDSS, and IBM Watton for Oncology is a typical representative thereof.
Although the development of CDSS in China is starting to be late, the development is rapid in recent years. At present, CDSS widely applied in China mainly comprises a personal hygiene clinical assistant, a per-unit clinical decision auxiliary system and the like of a personal hygiene publishing company. The intelligent medical assistant robot for science and science news can comprehensively learn mass data such as related medical professional teaching materials, clinical guidelines and classical cases, grasp a large amount of medical knowledge, has an intelligent voice recognition function, and can quickly generate an electronic medical record according to the conversation between a doctor and a patient and give auxiliary diagnosis.
Breast cancer is one of the major diseases faced by women, the morbidity and mortality of breast cancer are continuously increased, and serious harm is caused to the health of women, so that a medical system capable of effectively performing auxiliary diagnosis on breast cancer diseases is one of the medical systems which are urgently needed at present.
Disclosure of Invention
The invention aims to provide a knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system, which is used for carrying out data structuring on international and domestic breast cancer diagnosis and treatment guidelines such as NCCN, CSCO and the like, comprises an index system, an entity and a relation construction to form a knowledge graph, and establishes rules of the index system and the diagnosis and treatment scheme for logic in the guidelines by using a rule engine drools.
In order to achieve the above object, the system for recommending a diagnosis and treatment plan for breast cancer driven by knowledge graph comprises:
an input module for receiving input of patient information;
the system comprises a user information management module, a diagnosis and treatment stage information processing module and a diagnosis and treatment information processing module, wherein the user information management module is used for storing patient information and diagnosis and treatment stage information, and the diagnosis and treatment stage information represents breast cancer diagnosis and treatment operations which are already performed by a patient;
the system comprises a knowledge graph maintenance module, a data processing module and a data processing module, wherein the knowledge graph maintenance module takes professional clinical documents including a breast cancer diagnosis and treatment guide as data sources to generate a knowledge graph aiming at breast cancer diseases; generating a knowledge map for a breast cancer disease further comprises: extracting concept terms, relations and attributes of professional clinical documents, and defining a reference ontology; and establishing a mapping relation between the definition reference ontology and terms in the local hospital database, and if the local database uses standard clinical terms, the standard clinical terms are merged into the reference ontology, otherwise, establishing a mapping relation between the terms in the local database and the reference ontology.
The rule base module abstracts the inference rule of the diagnosis and treatment scheme into a triple structure < Precondition, Option and Function >, and constructs an inference rule base aiming at the diagnosis and treatment scheme of the breast cancer disease by using the triple structure as a data source and using professional clinical documents including a breast cancer diagnosis and treatment guide;
wherein Precondition represents a diagnosis Precondition, that is, a Precondition required to be satisfied when the diagnosis Precondition is satisfied; the precondition is usually plural, and the individual conditions are referred to as conditional elements; the Option represents a series of diagnosis and treatment operations which need to be executed when the Precondition is satisfied, namely, a corresponding diagnosis and treatment conclusion when the Precondition is satisfied.
The decision generation module is used for generating a corresponding breast cancer diagnosis and treatment scheme according to the patient information and the diagnosis and treatment stage information; the decision generation module is connected with the knowledge graph maintenance module and the rule base module, receives patient information and diagnosis and treatment stage information and takes the patient information and the diagnosis and treatment stage information as facts to be inferred; processing the facts to be retired by using natural language processing rules, and mapping the facts to the knowledge graph aiming at the breast cancer diseases so that local terms are mapped to a reference ontology; generating the Precondition required by the diagnosis and treatment scheme inference rule according to the reference ontology, and carrying out inference matching on the diagnosis and treatment scheme corresponding to the Precondition by loading the inference rule stored by the rule base module.
And the output module is used for outputting the diagnosis and treatment scheme.
Further, the decision generation module comprises at least one rule inference network, the rule inference network comprises an alpha region and a beta region, the alpha region is used for storing conditional elements of the Precondition, namely, each irreparable Precondition is used as a unit for storage, the beta region is used for recording matching intermediate results of each fact to be inferred and the preconditions stored in the alpha region, and the intermediate results are connected to obtain corresponding inference results under a plurality of matched preconditions.
According to the scheme, an effective solution is provided for recommending uncertain diagnosis and treatment schemes of complex conditions under multiple conditions by constructing a breast cancer knowledge graph model (including a breast cancer related medical entity E and a breast cancer related medical fact relation R in a graph) based on multiple guidelines (NCCN and CSCO), training and extracting corresponding diagnosis and treatment scheme rules and introducing an improved drools rule editor.
According to the invention, the breast cancer disease knowledge map and the actual diagnosis and treatment clinical data are combined to carry out deep analysis and identification on specific case conditions, so that the accuracy of disease condition identification of clinical patients is greatly improved, the construction of the breast cancer disease knowledge map is guided by the clinical data through mapping, and the diagnosis and treatment rules have practical value for diagnosis and treatment of breast cancer diseases.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a logical architecture diagram of a breast cancer diagnosis and treatment plan recommendation system driven by knowledge-graph provided by the present invention.
FIG. 2 is a schematic diagram of rule-based inference network inference provided by the present invention.
Fig. 3 is a flow chart of reasoning and decision for a breast cancer diagnosis and treatment plan provided by the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments that can be derived by one of ordinary skill in the art from the embodiments given herein are intended to be within the scope of the present invention.
The clinical practice guidelines for various malignancies released annually by the National Comprehensive Cancer Network (NCCN) have been recognized and followed by clinicians worldwide. The aim of NCCN, a non-profit academic organization composed of 21 top centers of tumors in the united states, is to increase the level of tumor service worldwide, benefiting tumor patients. Breast cancer is one of the accepted women health killers, and more than 100 million women with breast cancer patients are newly added every year in the world, wherein at least 40 million people die from breast cancer. Like most other countries, breast cancer is also the most common cancer among women in china, with increasing incidence and mortality from year to year.
Taking breast cancer as an example, the types and stages of breast cancer have different resistance to anticancer drugs and the applicability of clinical diagnosis and treatment methods. The germ line polymorphism variation of drug metabolizing enzymes, drug transporters and drug target related genes can cause the difference of drug metabolism speed, required drug dosage, adverse reaction, drug effect and the like of different patient individuals. In omics, oncogenes and cancer suppressor genes are targets of most targeted drugs, and variations in different omics, such as single nucleotide variation, insertion deletion, copy number variation, methylation and the like in the deoxyribonucleic acid (DNA) level, expression level variation and fusion variation in the ribonucleic acid (RNA) level, and expression variation in the protein level all affect the sensitivity of a patient to the targeted drugs, the efficacy of chemotherapeutic drugs, and the effect after healing. In order to improve the pertinence of diagnosis and treatment schemes for breast cancer diseased conditions of different individuals, the clinical influence of different variations on different breast cancer types and different drugs is fully considered, information such as a breast cancer related clinical guideline is integrated, and a knowledge graph for breast cancer diseases is constructed.
Definition 1 (breast cancer-related medical entity E):
breast cancer-related medical entity E refers to various uniquely identifiable medical entities in clinical patient medical records. Patient entities, basic information entities, breast cancer condition diagnostic entities, oncological examination entities, and the like are typically included.
Definition 2 (breast cancer-related medical fact relationship R):
the breast cancer related medical fact relationship represents a medical fact relationship R { Ei, Ej } occurring between different breast cancer related medical entities, wherein Ei, Ej are the breast cancer related medical entities. R may specifically include the following:
(1) and the has _ a relationship represents the membership between the entity A and the entity B.
(2) instance _ of relationship represents an instance relationship between entity A and entity B. I.e. entity B is an instance of entity a.
(3) attribute of relationship, indicating that entity a is an attribute value of entity B.
(4) part of relationship denotes the relationship of whole and part. For example, the characterization entity a in the inspection report is part of the inspection report entity B.
(5) Diagnostis relationship indicates that there is a diagnostic relationship between diagnostic entity A and patient entity B.
On the basis of defining the medical entity and medical fact relationship of the breast tumor, the formalization of the breast tumor knowledge map is defined as follows:
defining 3 (breast cancer disease knowledge graph G), wherein the breast cancer disease knowledge graph is a directed label graph G ═ E, R and T, wherein E is a vertex set of the knowledge graph and is used for representing a breast cancer related medical entity set; r is an edge set of the knowledge graph and is used for representing the medical fact relation related to the breast cancer; t is a function of E → R.
In actual clinical practice, different people have different description habits on the same symptom, the same character or diagnosis and treatment advice, so that the disease description and the diagnosis and treatment scheme record of the patient in clinical documents have differences such as spoken language, individuation and the like. Therefore, in the process of constructing G ═ E, R, T above, the concept terms, relationships and attributes of the professional clinical documents are extracted, the reference ontology is defined, the mapping relationship between the definition reference ontology and the terms in the hospital local database is established, if the local database uses the standard clinical terms, the standard clinical terms are incorporated into the reference ontology, otherwise, the mapping relationship between the terms in the local database and the reference ontology is established.
The recommendation system provided by the embodiment recommends a diagnosis and treatment scheme in a rule-based manner. In the field, a production formula such as IF-THEN is generally used for rule representation, but particularly in decision recommendation of a breast cancer diagnosis and treatment scheme, a certain precondition of each inference rule may appear in a plurality of inference rules, so that data redundancy is high, reuse and maintenance of the rules are not facilitated, and potential relations between symptoms and diseases are not found. Therefore, the invention generates a rule inference network based on an improved drools technology:
drools is an open source project developed by Bob mccuirter, which is a deductive inference based rule engine written in the Java language. The starting point is that the matching efficiency of the system mode is improved by utilizing two characteristics of an expert system based on rule reasoning, namely time redundancy and structural similarity. The core idea is that the matching efficiency is improved by saving the matching intermediate process and reducing repeated matching, and the realization mechanism is that a rule reasoning network is formed to carry out pattern matching.
The inference flow of the rule inference network is shown in the attached figure 2: when the rules satisfy facts C0, C1, trigger activity P1 to execute, when the rules satisfy facts C1, C2, C4, trigger activity P2 to execute, and when the rules satisfy facts C1, C2, C4, C5, trigger activity P3 to execute.
In constructing the rule inference network:
the method comprises the steps of dividing the method into an alpha area and a beta area, wherein the alpha area is used for storing conditional elements of the Precondition, namely, each indivisible Precondition is used as a unit for storage, the beta area is used for recording matching intermediate results of each fact to be presumed and the preconditions stored in the alpha area, and the intermediate results are connected to obtain corresponding reasoning results under the condition that a plurality of preconditions are matched.
Generally speaking, each fact to be inferred must be processed by the rule inference network to obtain a definite result. In clinical diagnosis and treatment decision, especially for breast cancer disease diagnosis, a plurality of rules are associated in fact in many cases, and only effective guidance can be given to doctors by exploring potential risk factors as much as possible, so that the probability of missed diagnosis is reduced, and the diagnosis and treatment efficiency is improved. In addition, with the increase of clinical diagnosis and treatment rules, the rules stored in the diagnosis and treatment rule base may reach tens of thousands or even more, and if the intermediate results stored in the β region are not screened, the rule matching speed cannot meet the requirements of practical application.
Based on the above reasons, the invention optimizes the matching mode, and applies the optimized algorithm to the reasoning decision of the diagnosis and treatment scheme of the breast cancer disease:
recording the confidence coefficient of the matching corresponding result of each condition element in the alpha region during the rule inference network test, and storing the confidence coefficient and the condition element into the alpha region; recording the confidence sum of each condition element meeting the condition when performing the inter-condition variable connection test in the beta region, and storing the confidence sum into the beta region;
judging whether a fact queue to be inferred (namely patient information and diagnosis and treatment stage information) is empty or not, if so, considering that inference is finished, and packaging result confidence coefficient data matched by inference together; and if a plurality of matching results exist finally, sorting according to the confidence degree, and returning a result set which is well arranged in sequence.
Through the matching mode, the rule inference network not only supports the deterministic inference of diagnosis and treatment schemes, but also supports the uncertain inference of a plurality of diagnosis and treatment schemes, and can provide more possible diagnosis and treatment schemes for complex diseases.
FIG. 1 shows a main logic architecture diagram of a recommendation system constructed based on a knowledge graph and a rule inference network of breast cancer diseases. Based on the constructed knowledge graph and rule inference network of breast cancer diseases, the recommendation system provided by the embodiment of the invention specifically comprises the following modules:
an input module for receiving input of patient information;
the system comprises a user information management module, a diagnosis and treatment stage information processing module and a diagnosis and treatment information processing module, wherein the user information management module is used for storing patient information and diagnosis and treatment stage information, and the diagnosis and treatment stage information represents breast cancer diagnosis and treatment operations which are already performed by a patient;
the system comprises a knowledge graph maintenance module, a data processing module and a data processing module, wherein the knowledge graph maintenance module takes professional clinical documents including a breast cancer diagnosis and treatment guide as data sources to generate a knowledge graph aiming at breast cancer diseases; generating a knowledge map for a breast cancer disease further comprises: extracting concept terms, relations and attributes of professional clinical documents, and defining a reference ontology; and establishing a mapping relation between the definition reference ontology and terms in the local hospital database, and if the local database uses standard clinical terms, the standard clinical terms are merged into the reference ontology, otherwise, establishing a mapping relation between the terms in the local database and the reference ontology.
The rule base module abstracts the inference rule of the diagnosis and treatment scheme into a triple structure < Precondition, Option and Function >, and constructs an inference rule base aiming at the diagnosis and treatment scheme of the breast cancer disease by using the triple structure as a data source and using professional clinical documents including a breast cancer diagnosis and treatment guide; wherein Precondition represents a diagnosis Precondition, that is, a Precondition required to be satisfied when the diagnosis Precondition is satisfied; the precondition is usually plural, and the individual conditions are referred to as conditional elements; the Option represents a series of diagnosis and treatment operations which need to be executed when the Precondition is satisfied, namely, a corresponding diagnosis and treatment conclusion when the Precondition is satisfied.
In order to facilitate the explanation of the inference rules of the rule base module, only a few condition elements are used as representatives for explanation. For example, the condition may include various influencing factors related to breast cancer diagnosis, such as female age, lymph node type, RS content value, clinical symptoms, and the like, and the Option includes various corresponding detection operations, such as 21 gene detection, PAM 50, Endo prediction, needle biopsy, molybdenum target examination, and the like, and drug recommendation constitutes a diagnosis and treatment plan.
The decision generation module is used for generating a corresponding breast cancer diagnosis and treatment scheme according to the patient information and the diagnosis and treatment stage information; the decision generation module is connected with the knowledge graph maintenance module and the rule base module, receives patient information and diagnosis and treatment stage information and takes the patient information and the diagnosis and treatment stage information as facts to be inferred; processing the facts to be retired by using natural language processing rules, and mapping the facts to the knowledge graph aiming at the breast cancer diseases so that local terms are mapped to a reference ontology; generating the Precondition required by the diagnosis and treatment scheme inference rule according to the reference ontology, and carrying out inference matching on the diagnosis and treatment scheme corresponding to the Precondition by loading the inference rule stored by the rule base module.
For example, when preconction constitutes a precondition: when the patient is aged more than or equal to 30 years old, and has breast lumps, asymmetrical thickening of glands or nodule and breast skin change (inflammatory breast cancer and breast Paget disease), and continuous, spontaneous, unilateral, single-duct, bloody and serous bloody nipple discharge, the decision generation module analyzes to obtain a diagnosis and treatment scheme corresponding to the patient, wherein the diagnosis and treatment scheme is an inference result which takes the X-ray breast examination as a main means and the ultrasonic examination as an auxiliary means. The example is only a simple example, and when an actual clinical diagnosis and treatment scheme is recommended, the complexity of the processed disease condition is far higher than that of the example, and the diagnosis and treatment scheme suggestion can be provided with uncertainty.
And the output module is used for outputting the diagnosis and treatment scheme.
After the diagnosis and treatment scheme is output, the user information management module updates the diagnosis and treatment stage information according to the recommended diagnosis and treatment scheme so as to construct a full course record of the patient; the decision generation module is also used for generating a continuous breast cancer diagnosis and treatment scheme according to the updated diagnosis and treatment stage information, and supporting a doctor to provide a scheme suggestion with a plurality of continuous whole course for a patient.
The valley provided by the invention supports diagnosis and treatment recommendation of multiple guidelines, multiple documents and multiple expert experiences aiming at a recommendation system of a breast cancer diagnosis and treatment scheme, and simultaneously provides scheme support of multiple continuous whole course of disease for a patient; in addition, based on the expansion of data sources in the knowledge graph, the clinical diagnosis and treatment work of doctors can be supported from the early warning and prevention of adverse drug reaction risks, the use knowledge of drugs, the multi-dimensionality of clinical research information and the like.
Claims (8)
1. A knowledge-graph-driven breast cancer diagnosis and treatment scheme recommendation system is characterized by comprising:
an input module for receiving input of patient information;
the system comprises a user information management module, a diagnosis and treatment stage information processing module and a diagnosis and treatment information processing module, wherein the user information management module is used for storing patient information and diagnosis and treatment stage information, and the diagnosis and treatment stage information represents breast cancer diagnosis and treatment operations which are already performed by a patient;
the system comprises a knowledge graph maintenance module, a data processing module and a data processing module, wherein the knowledge graph maintenance module takes professional clinical documents including a breast cancer diagnosis and treatment guide as data sources to generate a knowledge graph aiming at breast cancer diseases;
the rule base module is used for extracting reasoning rules aiming at the breast cancer disease diagnosis and treatment scheme by taking professional clinical documents including a breast cancer diagnosis and treatment guide as a data source to form a diagnosis and treatment scheme rule base;
the decision generation module is used for generating a corresponding breast cancer diagnosis and treatment scheme according to the patient information and the diagnosis and treatment stage information;
and the output module is used for outputting the diagnosis and treatment scheme.
2. The breast cancer diagnosis and treatment plan recommendation system according to claim 1, wherein:
generating a knowledge map for a breast cancer disease further comprises: extracting concept terms, relations and attributes of professional clinical documents, and defining a reference ontology;
and establishing a mapping relation between the definition reference ontology and terms in the local hospital database, and if the local database uses standard clinical terms, the standard clinical terms are merged into the reference ontology, otherwise, establishing a mapping relation between the terms in the local database and the reference ontology.
3. The breast cancer diagnosis and treatment plan recommendation system according to claim 2, wherein the generation of the diagnosis and treatment plan rule base specifically includes:
abstracting an inference rule of a diagnosis and treatment scheme into a triple structure < Precondition, Option and Function >, and constructing an inference rule base aiming at the diagnosis and treatment scheme of the breast cancer diseases by using the triple structure;
wherein Precondition represents a diagnosis Precondition, that is, a Precondition required to be satisfied when the diagnosis Precondition is satisfied; the precondition is usually plural, and the individual conditions are referred to as conditional elements; the Option represents a series of diagnosis and treatment operations which need to be executed when the Precondition is satisfied, namely, a corresponding diagnosis and treatment conclusion when the Precondition is satisfied.
4. The breast cancer diagnosis and treatment scheme recommendation system according to claim 3, wherein the decision generation module is connected with the knowledge graph maintenance module and the rule base module, and specifically comprises:
receiving patient information and diagnosis and treatment stage information as facts to be inferred;
processing the fact to be reasoned by using a natural language processing rule, and mapping the fact to be reasoned into the knowledge graph aiming at the breast cancer disease so that local terms are mapped to a reference ontology;
generating the Precondition required by the diagnosis and treatment scheme inference rule according to the reference ontology, and carrying out inference matching on the diagnosis and treatment scheme corresponding to the Precondition by loading the inference rule stored by the rule base module.
5. The breast cancer diagnosis and treatment plan recommendation system according to any one of claims 1 to 4, wherein:
the decision generation module comprises at least one rule inference network;
the rule inference network comprises an alpha region and a beta region, wherein the alpha region is used for storing conditional elements of Precondition, namely each irreparable Precondition is used as a minimum unit for storage; the beta region is used for recording the matching intermediate results of each fact to be presumed and the preconditions stored in the alpha region, and performing binding operation on each intermediate result to obtain corresponding reasoning results under the condition of matching a plurality of preconditions.
6. The breast cancer diagnosis and treatment plan recommendation system according to claim 5, wherein:
when the rule inference network is constructed, training of an alpha region and a beta region is carried out by loading the breast cancer disease diagnosis and treatment rules in the rule base module, and confidence degrees of successful matching of conditional elements and successful matching of intermediate results are respectively introduced into the alpha region and the beta region and used for inference of an uncertain diagnosis and treatment scheme.
7. The breast cancer diagnosis and treatment plan recommendation system according to claim 6, wherein:
recording the confidence coefficient of the matching corresponding result of each condition element in the alpha region when the rule inference network is trained, and storing the confidence coefficient and the condition element into the alpha region; recording the confidence sum of each condition element meeting the condition when performing the inter-condition variable connection test in the beta region, and storing the confidence sum into the beta region;
judging whether the fact to be inferred is empty or not, if so, considering that inference is finished, and packaging the result confidence coefficient data matched by inference together; and if a plurality of matching results exist finally, sorting according to the confidence degree, and returning a result set which is well arranged in sequence.
8. The breast cancer diagnosis and treatment plan recommendation system according to claim 1, wherein:
the user information management module updates diagnosis and treatment stage information according to the recommended diagnosis and treatment scheme so as to construct a full course record of the patient;
and the decision generation module is used for generating a continuous breast cancer diagnosis and treatment scheme according to the updated diagnosis and treatment stage information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010758816.6A CN111899828A (en) | 2020-07-31 | 2020-07-31 | Knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010758816.6A CN111899828A (en) | 2020-07-31 | 2020-07-31 | Knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111899828A true CN111899828A (en) | 2020-11-06 |
Family
ID=73182869
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010758816.6A Pending CN111899828A (en) | 2020-07-31 | 2020-07-31 | Knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111899828A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112786198A (en) * | 2021-02-08 | 2021-05-11 | 科大讯飞股份有限公司 | Diagnosis and treatment information recommendation model construction method, diagnosis and treatment information recommendation method and device |
CN113393933A (en) * | 2021-07-12 | 2021-09-14 | 华东理工大学 | Gastric cancer decision-making auxiliary treatment system based on state transition map |
CN113506625A (en) * | 2021-08-21 | 2021-10-15 | 江泽飞 | Diagnosis and treatment suggestion scoring system based on csco guide |
CN113571179A (en) * | 2021-07-09 | 2021-10-29 | 清华大学 | Index extraction method and device based on knowledge graph |
CN113590775A (en) * | 2021-06-22 | 2021-11-02 | 北京百度网讯科技有限公司 | Diagnosis and treatment data processing method and device, electronic equipment and storage medium |
CN113838574A (en) * | 2021-09-29 | 2021-12-24 | 杭州海心智医信息科技有限公司 | Database application system of tumor disease medical record |
CN115762707A (en) * | 2023-01-09 | 2023-03-07 | 神州医疗科技股份有限公司 | Knowledge graph-based liver cancer medication recommendation method and system |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050049497A1 (en) * | 2003-06-25 | 2005-03-03 | Sriram Krishnan | Systems and methods for automated diagnosis and decision support for breast imaging |
CN103020219A (en) * | 2012-12-10 | 2013-04-03 | 广东电网公司电力科学研究院 | Network-based expert system tool for power grid fault diagnosis |
CN106295215A (en) * | 2016-08-18 | 2017-01-04 | 京东方科技集团股份有限公司 | Clinical assistant diagnosis system |
CN107578797A (en) * | 2016-07-04 | 2018-01-12 | 胡峰 | The method and its system that a kind of health diagnosis and treatment scheme based on cloud computing is recommended |
CN107808194A (en) * | 2017-10-26 | 2018-03-16 | 北京科技大学 | A kind of complex inference method based on Rete inference networks |
CN109830303A (en) * | 2019-02-01 | 2019-05-31 | 上海众恒信息产业股份有限公司 | Clinical data mining analysis and aid decision-making method based on internet integration medical platform |
CN110459320A (en) * | 2019-08-20 | 2019-11-15 | 山东众阳健康科技集团有限公司 | A kind of assisting in diagnosis and treatment system of knowledge based map |
CN110751621A (en) * | 2019-09-05 | 2020-02-04 | 五邑大学 | Breast cancer auxiliary diagnosis method and device based on deep convolutional neural network |
CN111370127A (en) * | 2020-01-14 | 2020-07-03 | 之江实验室 | Decision support system for early diagnosis of chronic nephropathy in cross-department based on knowledge graph |
CN111382275A (en) * | 2018-12-28 | 2020-07-07 | 医渡云(北京)技术有限公司 | Construction method, device and medium of medical knowledge graph and electronic equipment |
-
2020
- 2020-07-31 CN CN202010758816.6A patent/CN111899828A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050049497A1 (en) * | 2003-06-25 | 2005-03-03 | Sriram Krishnan | Systems and methods for automated diagnosis and decision support for breast imaging |
CN103020219A (en) * | 2012-12-10 | 2013-04-03 | 广东电网公司电力科学研究院 | Network-based expert system tool for power grid fault diagnosis |
CN107578797A (en) * | 2016-07-04 | 2018-01-12 | 胡峰 | The method and its system that a kind of health diagnosis and treatment scheme based on cloud computing is recommended |
CN106295215A (en) * | 2016-08-18 | 2017-01-04 | 京东方科技集团股份有限公司 | Clinical assistant diagnosis system |
CN107808194A (en) * | 2017-10-26 | 2018-03-16 | 北京科技大学 | A kind of complex inference method based on Rete inference networks |
CN111382275A (en) * | 2018-12-28 | 2020-07-07 | 医渡云(北京)技术有限公司 | Construction method, device and medium of medical knowledge graph and electronic equipment |
CN109830303A (en) * | 2019-02-01 | 2019-05-31 | 上海众恒信息产业股份有限公司 | Clinical data mining analysis and aid decision-making method based on internet integration medical platform |
CN110459320A (en) * | 2019-08-20 | 2019-11-15 | 山东众阳健康科技集团有限公司 | A kind of assisting in diagnosis and treatment system of knowledge based map |
CN110751621A (en) * | 2019-09-05 | 2020-02-04 | 五邑大学 | Breast cancer auxiliary diagnosis method and device based on deep convolutional neural network |
CN111370127A (en) * | 2020-01-14 | 2020-07-03 | 之江实验室 | Decision support system for early diagnosis of chronic nephropathy in cross-department based on knowledge graph |
Non-Patent Citations (2)
Title |
---|
汪鹏等: "大数据驱动的临床决策支持***设计与实践", 《中国数字医学》 * |
高雅: "开放式临床诊疗智能决策支持***中规则引擎的研究与实现", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112786198A (en) * | 2021-02-08 | 2021-05-11 | 科大讯飞股份有限公司 | Diagnosis and treatment information recommendation model construction method, diagnosis and treatment information recommendation method and device |
CN112786198B (en) * | 2021-02-08 | 2024-05-31 | 科大讯飞股份有限公司 | Diagnosis and treatment information recommendation model construction method, diagnosis and treatment information recommendation method and device |
CN113590775A (en) * | 2021-06-22 | 2021-11-02 | 北京百度网讯科技有限公司 | Diagnosis and treatment data processing method and device, electronic equipment and storage medium |
CN113590775B (en) * | 2021-06-22 | 2023-07-21 | 北京百度网讯科技有限公司 | Diagnosis and treatment data processing method and device, electronic equipment and storage medium |
CN113571179A (en) * | 2021-07-09 | 2021-10-29 | 清华大学 | Index extraction method and device based on knowledge graph |
CN113393933A (en) * | 2021-07-12 | 2021-09-14 | 华东理工大学 | Gastric cancer decision-making auxiliary treatment system based on state transition map |
CN113506625A (en) * | 2021-08-21 | 2021-10-15 | 江泽飞 | Diagnosis and treatment suggestion scoring system based on csco guide |
CN113838574A (en) * | 2021-09-29 | 2021-12-24 | 杭州海心智医信息科技有限公司 | Database application system of tumor disease medical record |
CN115762707A (en) * | 2023-01-09 | 2023-03-07 | 神州医疗科技股份有限公司 | Knowledge graph-based liver cancer medication recommendation method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111899828A (en) | Knowledge graph driven breast cancer diagnosis and treatment scheme recommendation system | |
Solares et al. | Deep learning for electronic health records: A comparative review of multiple deep neural architectures | |
US10957433B2 (en) | Clinical concept identification, extraction, and prediction system and related methods | |
Peek et al. | Thirty years of artificial intelligence in medicine (AIME) conferences: A review of research themes | |
US20200381087A1 (en) | Systems and methods of clinical trial evaluation | |
Mansingh et al. | Using ontologies to facilitate post-processing of association rules by domain experts | |
Guh et al. | Integrating genetic algorithm and decision tree learning for assistance in predicting in vitro fertilization outcomes | |
US11636951B2 (en) | Systems and methods for generating a genotypic causal model of a disease state | |
AU2005321925A1 (en) | Methods, systems, and computer program products for developing and using predictive models for predicting a plurality of medical outcomes, for evaluating intervention strategies, and for simultaneously validating biomarker causality | |
CN113161001B (en) | Improved LDA-based process path mining method | |
WO2022060949A1 (en) | Systems and methods for automatically identifying a candidate patient for enrollment in a clinical trial | |
CN112270988B (en) | Auxiliary diagnosis method for rare diseases | |
Benhar et al. | A systematic mapping study of data preparation in heart disease knowledge discovery | |
Gao et al. | Using case-level context to classify cancer pathology reports | |
Lai et al. | BioREx: improving biomedical relation extraction by leveraging heterogeneous datasets | |
Johari et al. | Smart self-checkup for early disease prediction | |
US20230253122A1 (en) | Systems and methods for generating a genotypic causal model of a disease state | |
Uppal et al. | Enhancing accuracy in brain stroke detection: Multi-layer perceptron with Adadelta, RMSProp and AdaMax optimizers | |
Pathak | Automatic structuring of breast cancer radiology reports for quality assurance | |
Chen et al. | A multi-channel convolutional neural network for ICD coding | |
Rao et al. | Clinical and financial outcomes analysis with existing hospital patient records | |
Baron | Artificial Intelligence in the Clinical Laboratory: An Overview with Frequently Asked Questions | |
Yang et al. | Modeling Clinical Thinking Based on Knowledge Hypergraph Attention Network and Prompt Learning for Disease Prediction | |
Abdullah et al. | Disease diagnosis using soft computing model: a digest | |
Gupta | Developing Clinical Decision Support Systems for Sepsis Prediction Using Temporal and Non-temporal Machine Learning Methods |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201106 |
|
RJ01 | Rejection of invention patent application after publication |