CN112732940B - Reasoning method, device, equipment and medium of medical knowledge graph based on model - Google Patents

Reasoning method, device, equipment and medium of medical knowledge graph based on model Download PDF

Info

Publication number
CN112732940B
CN112732940B CN202110057441.5A CN202110057441A CN112732940B CN 112732940 B CN112732940 B CN 112732940B CN 202110057441 A CN202110057441 A CN 202110057441A CN 112732940 B CN112732940 B CN 112732940B
Authority
CN
China
Prior art keywords
model
function
quadruple
determining
entity
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.)
Active
Application number
CN202110057441.5A
Other languages
Chinese (zh)
Other versions
CN112732940A (en
Inventor
李林峰
闫峻
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yidu Cloud Beijing Technology Co Ltd
Original Assignee
Yidu Cloud Beijing Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yidu Cloud Beijing Technology Co Ltd filed Critical Yidu Cloud Beijing Technology Co Ltd
Priority to CN202110057441.5A priority Critical patent/CN112732940B/en
Publication of CN112732940A publication Critical patent/CN112732940A/en
Priority to PCT/CN2021/128491 priority patent/WO2022151811A1/en
Application granted granted Critical
Publication of CN112732940B publication Critical patent/CN112732940B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • 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
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • 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
    • G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • Pathology (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Animal Behavior & Ethology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The application discloses a reasoning method, a device, equipment and a storage medium of a medical knowledge graph based on a model, wherein the graph comprises a plurality of quaternions, the quaternions comprise model information of a head entity, an attribute relationship, a tail entity and a function model, the function model is a model for determining constraint conditions of the attribute relationship between the head entity and the tail entity, and the method comprises the following steps: determining a first entity and a second entity corresponding to an object to be inferred; determining at least one quadruple corresponding to the medical knowledge graph based on a first entity and a second entity corresponding to the object to be inferred; obtaining model information corresponding to at least one quadruple, and determining a corresponding function model based on the model information; and determining constraint conditions of the attribute relationship corresponding to the at least one tetrad based on the function model and the model information. According to the technical scheme, the constraint condition of the attribute relationship between the two entities can be accurately expressed, and personalized reasoning can be performed according to the individual condition of the patient.

Description

Reasoning method, device, equipment and medium of medical knowledge graph based on model
Technical Field
The present disclosure relates to the field of machine learning technologies, and in particular, to a method, an apparatus, a device, and a storage medium for reasoning a medical knowledge graph based on a model.
Background
The medical knowledge graph technology is a semantic network for revealing medical knowledge entities, and is a key technology capable of enabling a computer algorithm system to understand medical knowledge.
Medical knowledge maps are generally constructed by ternary structures, such as: < diabetic complications diabetic nephropathy >. The three elements of the triplet may be referred to as a Head Entity (Head Entity), an attribute relationship (relationship), and a Tail Entity (Tail Entity), respectively. After the knowledge graph is constructed, knowledge reasoning can be performed based on the existing medical knowledge in the medical knowledge graph of the model.
However, since the establishment or occurrence of the attribute relationship between the head entity and the tail entity of the triplet is generally constrained by a certain condition, for example, not all diabetics will develop diabetic nephropathy, and the diabetics will develop diabetic nephropathy with a certain probability, the medical knowledge graph is difficult to express the condition constraint of the attribute relationship, and personalized reasoning cannot be performed.
Disclosure of Invention
The embodiment of the application provides a method, a device, equipment and a storage medium for reasoning a medical knowledge graph based on a model, which are used for solving the problems that the medical knowledge graph is difficult to express the condition constraint of an attribute relationship and personalized reasoning cannot be performed.
In a first aspect, the present application provides a method for reasoning a model-based medical knowledge graph, the medical knowledge graph including a plurality of quaternions, the quaternions including model information of a head entity, an attribute relationship, a tail entity, and a function model, the function model being a model that determines constraints of the attribute relationship between the head entity and the tail entity, the method comprising: determining a first entity and a second entity corresponding to an object to be inferred; determining at least one quadruple corresponding to the medical knowledge graph based on the first entity and the second entity corresponding to the object to be inferred; obtaining model information corresponding to the at least one quadruple, and determining a corresponding function model based on the model information; and determining the constraint condition of the attribute relation corresponding to the at least one quadruple based on the function model and the model information.
In some example embodiments of the present application, the model information includes model types including: one or more of a constant function model, a fitting function model, a regular function model, and a complex function model.
In some example embodiments of the present application, the model information is information represented in JavaScript object notation JSON format, and the determining a corresponding function model based on the model information includes:
analyzing the model information in the JSON format, and determining the model type in the model information; and determining a corresponding function model based on the model type.
In some example embodiments of the present application, the at least one quadruple comprises a quadruple, and the determining the corresponding function model based on the model type comprises: if the model information of the four-element group comprises the rule function model and the probability model, taking the product of the rule function model and the probability model as the function model, wherein the probability model comprises the constant function model and/or the fitting function model; if the model information of the quadruple only comprises the rule function model, the rule function model is used as the function model; and if the model information of the four-element group only comprises the probability model, taking the probability model as the function model.
In some example embodiments of the present application, the at least one quadruple comprises a plurality of quadruples, the determining a corresponding function model based on the model type comprises: determining each intermediate reasoning model corresponding to the model information of each four-tuple, wherein the intermediate reasoning model comprises the rule function model and/or the probability model; and taking the product of each intermediate reasoning model as a function model corresponding to the at least one quadruple.
In some example embodiments of the present application, the model information includes model variables and model parameters, the determining the constraint condition of the attribute relationship corresponding to the at least one quadruple based on the function model and the model information includes: determining the value of a corresponding model variable from medical record information of the object to be inferred based on the model variable; substituting the values of the model variables and the values of the model parameters into the function model; and determining the constraint condition of the attribute relation corresponding to at least one quadruple based on the output result of the function model.
In some example embodiments of the present application, the constraint of the attribute relationship includes a condition that the attribute relationship holds and/or a probability that the attribute relationship occurs.
In a second aspect, there is provided an inference apparatus of a model-based medical knowledge graph including a plurality of quaternions including model information of a head entity, an attribute relationship, a tail entity, and a function model which is a model of constraint conditions determining the attribute relationship between the head entity and the tail entity, the apparatus comprising: the entity determining module is used for determining a first entity and a second entity corresponding to the object to be inferred; the quadruple determining module is used for determining at least one quadruple corresponding to the medical knowledge graph based on the first entity and the second entity corresponding to the object to be inferred; the model determining module is used for acquiring model information corresponding to the at least one quadruple and determining a corresponding function model based on the model information; and the attribute determining module is used for determining the constraint condition of the attribute relation corresponding to the at least one quaternion based on the function model and the model information.
In some example embodiments of the present application, the model information includes model types including: one or more of a constant function model, a fitting function model, a regular function model, and a complex function model.
In some example embodiments of the present application, the model information is information represented in JavaScript object notation JSON format, and the model determination module is further configured to: analyzing the model information in the JSON format, and determining the model type in the model information; and determining a corresponding function model based on the model type.
In some example embodiments of the present application, the at least one quadruple comprises one quadruple, and the model determination module is further configured to: if the model information of the four-element group comprises the rule function model and the probability model, taking the product of the rule function model and the probability model as the function model, wherein the probability model comprises the constant function model and/or the fitting function model; if the model information of the quadruple only comprises the rule function model, the rule function model is used as the function model; and if the model information of the four-element group only comprises the probability model, taking the probability model as the function model.
In some example embodiments of the present application, the at least one quadruple comprises a plurality of quadruples, the model determination module further to: determining each intermediate reasoning model corresponding to the model information of each four-tuple, wherein the intermediate reasoning model comprises the rule function model and/or the probability model; and taking the product of each intermediate reasoning model as a function model corresponding to the at least one quadruple.
In some example embodiments of the present application, the model information includes model variables and model parameters, and the attribute determination module is further configured to: determining the value of a corresponding model variable from medical record information of the object to be inferred based on the model variable; substituting the values of the model variables and the values of the model parameters into the function model; and determining the constraint condition of the attribute relation corresponding to at least one quadruple based on the output result of the function model.
In some example embodiments of the present application, the constraint of the attribute relationship includes a condition that the attribute relationship holds and/or a probability that the attribute relationship occurs.
In a third aspect, the present application provides an electronic device, comprising: at least one processor, memory, and an interface to communicate with other electronic devices; the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of reasoning about model-based medical knowledge-graph of any of the second aspects.
In a fourth aspect, the present application provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method of reasoning model-based medical knowledge-graph of any one of the first to third aspects.
One embodiment of the above application has the following advantages or benefits: on the one hand, based on a first entity and a second entity corresponding to the object to be inferred, determining at least one quadruple corresponding to the medical knowledge graph, and accurately matching the object to be inferred with the quadruple; on the other hand, the constraint condition of the attribute relationship of the two entities is expressed as a function model, the model information of the function model is used as the fourth element of the four-element group in the medical knowledge graph, the constraint condition of the attribute relationship corresponding to the four-element group is determined based on the function model and the model information, the constraint condition of the attribute relationship between the two entities of the four-element group can be accurately expressed, and personalized reasoning can be carried out according to the individual condition of the patient.
Other effects of the above alternative will be described below in connection with specific embodiments.
Drawings
The drawings are for better understanding of the present solution and do not constitute a limitation of the present application. Wherein:
FIG. 1 is a schematic diagram of a medical knowledge graph in one embodiment;
FIG. 2 is a flow diagram of a method of reasoning model-based medical knowledge-graph, provided in accordance with some embodiments of the present application;
FIG. 3 is a schematic illustration of a medical knowledge-graph provided in accordance with some embodiments of the present application;
FIG. 4 is a flow diagram of an inference method for determining a functional model provided in accordance with some embodiments of the present application;
FIG. 5 is a schematic illustration of a medical knowledge-graph provided in accordance with further embodiments of the present application;
FIG. 6 is a schematic block diagram of an inference apparatus for model-based medical knowledge-graph provided in accordance with some embodiments of the present application;
fig. 7 is a block diagram of an electronic device for implementing a method of reasoning model-based medical knowledge-graph in accordance with an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
In one technical solution, knowledge points with attributes of the attribute relationship between the head entity and the tail entity are represented by the quadruple, for example, the probability of occurrence of the attribute relationship between the head entity and the tail entity may be represented on a fourth element of the quadruple. Referring to fig. 1, the knowledge graph defines two knowledge points of "diabetes mellitus will occur in the future of 20% probability of obese patients", "diabetic nephropathy will occur in 40% probability of diabetic patients", and the corresponding quadruple is expressed as:
< probability of obesity Risk disease diabetes 0.2>
< probability of diabetic complications diabetic nephropathy >
However, in this technical solution, although the group probability value is added to the fourth element of the four-element group, it is possible to infer the group probability that the attribute relationship occurs, but it is difficult to perform personalized inference for individuals in different situations. For example, according to the four-way group described above, the probability of occurrence of diabetic nephropathy is 40% for different diabetics, which is clearly not the case. Thus, although the risk of developing diabetic nephropathy is 40% in the population with diabetes mellitus in a statistical sense, the probability of risk of developing diabetic nephropathy in each individual patient is related to the individual information of the patient, such as Body Mass Index (BMI), whether smoking, the latest glycosylated hemoglobin, urine protein/creatinine ratio, high density cholesterol, etc., and the degree of the test result. Therefore, the current medical knowledge graph cannot express the complex attribute of the attribute relationship, and cannot perform personalized reasoning according to the individual condition of the patient.
Based on the above, the application provides a reasoning method of a medical knowledge graph based on a model, which is characterized in that constraint conditions of attribute relations of triples are expressed as a function model, model information of the function model is used as a fourth element of the quadruples in the medical knowledge graph, and constraint conditions of the attribute relations corresponding to the quadruples are determined based on the function model and the model information. According to the technical scheme of the embodiment of the application, the constraint condition of the attribute relationship between the two entities of the four-element group can be accurately expressed, and personalized reasoning can be performed according to the individual condition of the patient.
Fig. 2 is a flow chart of a method for processing a medical knowledge-graph according to some embodiments of the present application. The processing method of the medical knowledge graph may be performed by a device having a calculation processing function, for example, may be performed by a server. The processing method includes steps S210 to S230. The method for processing the medical knowledge graph is described in detail below with reference to the accompanying drawings.
Referring to fig. 2, in step S210, a first entity and a second entity corresponding to an object to be inferred are determined.
In an example embodiment, the target patient is an object to be inferred, and after the inference target is determined, the first entity and the second entity corresponding to the target patient may be determined according to the inference target. For example, if the inference objective is to infer the probability that the diabetes patient a uses insulin therapy, it is determined that the first entity corresponding to the diabetes patient a to be inferred is diabetes and the second entity is insulin therapy.
In step S220, at least one quadruple corresponding to the medical knowledge-graph is determined based on the first entity and the second entity corresponding to the object to be inferred.
In an example embodiment, the medical knowledge graph includes a plurality of quaternions including model information of a head entity, an attribute relationship, a tail entity, and a functional model, the functional model being a model of constraints that determine the attribute relationship between the head entity and the tail entity. After the first entity and the second entity corresponding to the object to be inferred are determined, matching the corresponding entities in the medical knowledge graph according to the first entity and the second entity, and obtaining at least one quadruple corresponding to the medical knowledge graph.
For example, referring to fig. 3, if the first entity corresponding to the object to be inferred is diabetes, the second entity is insulin therapy, and a quadruple, that is, a quadruple < diabetes therapy regimen insulin attribute 6>, in the corresponding medical knowledge graph is determined through entity matching; if the first entity corresponding to the object to be inferred is obesity and the second entity is insulin therapy, determining two quadruples in the corresponding medical knowledge graph through entity matching, namely, the quadruples are < obesity risk disease diabetes attribute 1>, < diabetes therapy scheme insulin attribute 6>, wherein the attribute 1 and the attribute 6 can be represented by model information of a corresponding function model.
In step S230, model information corresponding to at least one quadruple is acquired, and a corresponding function model is determined based on the model information.
In an example embodiment, the model information is information expressed in JSON (JavaScript Object Notation ) format, the model information including model types including: one or more of a constant function model, a fitting function model, and a regular function model. . Analyzing the model information in the JSON format, and determining the model type in the model information; a corresponding functional model is determined based on the model type.
It should be noted that, although the model information is described as being in JSON format, it should be understood by those skilled in the art that the model information may be represented in other suitable forms, such as XML (Extensible Markup Language ), which is also within the scope of the present application.
In step S240, constraint conditions of the attribute relationship corresponding to the at least one quadruple are determined based on the function model and the model information.
In an example embodiment, the model information includes model variables and model parameters, and values of the corresponding model variables are determined from medical record information of the object to be inferred based on the model variables; substituting the values of the model variables and the values of the model parameters into the function model; and determining constraint conditions of attribute relationships corresponding to at least one four-tuple based on the output result of the function model.
According to the technical scheme in the example embodiment of fig. 2, on one hand, based on the first entity and the second entity corresponding to the object to be inferred, determining at least one quadruple corresponding to the medical knowledge graph, the object to be inferred and the quadruple can be accurately matched; on the other hand, the constraint condition of the attribute relationship of the two entities is expressed as a function model, the model information of the function model is used as the fourth element of the four-element group in the medical knowledge graph, the constraint condition of the attribute relationship corresponding to the four-element group is determined based on the function model and the model information, the constraint condition of the attribute relationship between the two entities of the four-element group can be accurately expressed, and personalized reasoning can be carried out according to the individual condition of the patient.
In an example embodiment, the model types of the function model include: one or more of a constant function model, a fitting function model, a regular function model, and a complex function model. Table 1 below shows these several functional models.
TABLE 1 functional model
Figure SMS_1
Fig. 4 is a flow diagram of an inference method for determining a functional model provided in accordance with some embodiments of the present application.
Referring to fig. 4, in step S410, if the model information of the four-tuple includes a rule function model and a probability model, the product of the rule function model and the probability model is taken as a function model, and the probability model includes a constant function model and/or a fitting function model.
In an example embodiment, the at least one quadruple comprises one quadruple, i.e. a first order reasoning. For example, in the first-order extrapolation, assume e i ,e j Are two directly related entities in the medical knowledge-graph. Then e i To e j The functional model or the inference probability model of (2) can be represented by the following formula (1):
C ij *P ij (1)
wherein C is ij For the condition that the attribute relationship of two entities is satisfied, a rule function model, i.e. the function f in the table 1, can be used 3 Expressed by P ij The probability that the attribute relationship is established can be expressed by a probability model, wherein the probability model comprises a constant function model and/or a fitting function model, and the constant function model can be a function f shown in the table 1 1 The fitting function model may be the function f of Table 1 above 2 。
Further, if a rule function model has been defined in the model information, C ij Taking a rule function model for operation; if a fitting function model, i.e. a personalized probability model, has been defined in the model information, P ij And selecting a fitting function model for operation.
In step S420, if the model information of the four-tuple includes only the rule function model, the rule function model is used as the function model.
In an example embodiment, if the probabilistic model is not defined in the model information, then P ij Taking 1, the attribute relation determination will happen, and taking a rule function model as a function model.
In step S430, if only the probability model is included in the model information of the quadruple, the probability model is used as a function model.
In an example embodiment, if the rule function model is not defined in the model information, C ij Taking 1 indicates that the attribute relationship is always true. If a fitting function model has been defined in the model information and the model-dependent variables are available in the inference context, then P ij And selecting a fitting function model for operation. If the fitted function model is not defined in the model information or the model dependent variables are not available, P ij And taking the value of a constant function model, namely a population probability model.
According to the technical solution in the example embodiment of fig. 4, by combining a rule function model and a fitting function model, i.e. a personalized reasoning model, more accurate probabilistic reasoning can be performed for different individuals.
Further, in an example embodiment, the at least one quadruple comprises a plurality of quadruples, i.e. multi-order reasoning, determining a corresponding function model based on model types, comprising: determining each intermediate reasoning model corresponding to the model information of each four-tuple, wherein the intermediate reasoning model comprises a rule function model and/or a probability model, and the probability model comprises a constant function model and/or a fitting function model; and taking the product of each intermediate reasoning model as a function model corresponding to the at least one quaternion.
For example, in multi-order extrapolation, assume e i ,e k E is the entity of two non-direct association in the medical knowledge graph i To e j The inference probability of (2) is represented by the following formula:
Figure SMS_2
wherein j is 1 ,...,j n To connect e i To e k Is a single node of the plurality of nodes. The values of C and P in the reasoning process of the multi-order reasoning are the same as those in the first-order reasoning. That is to say e i To e j Is equal to e i To e k The product of the inference probabilities in the path.
Fig. 5 is a schematic diagram of a medical knowledge-graph provided according to further embodiments of the present application.
FIG. 5 is a sub-graph of a medical knowledge graph about diabetes, with a description ID of constraints of an attribute relationship on each side of the sub-graph, represented by constraints { ID }, e.g., REL_ATT { ID }. Constraint 6 is a constraint that diabetes can treat this attribute relationship with insulin. The probability of using insulin therapy for four diabetics a, B, C, D can be inferred based on the medical knowledge graph of fig. 5. Next, referring to fig. 5, an inference process applying the inference method of the medical knowledge-graph in the exemplary embodiment of the present application will be described in detail with reference to several examples.
Example one: when only the rule function model, i.e., cond_model, is defined in constraint 6, model information corresponding to the fourth element corresponding to the quadruple is as follows:
Figure SMS_3
based on the model information, the probability of insulin therapy to be used next is inferred from the current medication conditions of four patients and the latest glycosylated hemoglobin value hba c conditions as shown in the following table (2):
table (2) probability of insulin therapy for diabetics when only the rule function model is defined in constraint 6
Patient' s Current solution Up to date hba c Probability of insulin usage
Patient A Double medicine treatment 8.0 0×1＝0
Patient B Three-medicine treatment 6.5 0×1＝0
Patient C Three-medicine treatment 8.0 1×1＝1
Patient D Three-medicine treatment 10.0 1×1＝1
In Table (2), since only the rule function model is defined, the inference model is C ij * Patient A, due to the use of dual medication, does not meet the rules, C ij 0, the probability of insulin use is 0; patient B was treated with three drugs, but the latest hba C was 6.5 to less than 7, and did not meet the rules, C ij 0, the probability of insulin use is 0; patient C is treated with three drugs, and latest hba C is 8 to 7, which meets the rules, C ij 1, the probability of insulin use is 1; patient D was treated with three drugs and was more recently hba C10 to 7, meeting the rules, C ij 1, the probability of insulin use is 1.
Example two: adding a constant function model, namely a group reasoning probability prob model, to the constraint condition 6, wherein model information corresponding to a fourth element corresponding to the four elements is as follows:
Figure SMS_4
based on the model information, the probability of the next insulin therapy is inferred from the current medication conditions of four patients and the latest glycosylated hemoglobin value hba c conditions as shown in the following table (3):
table (3) probability of diabetes patient to use insulin therapy when rule function model and population probability model are defined in constraint 6
Patient' s Current solution Up to date hba c Probability of insulin usage
Patient A Double medicine treatment 8.0 0×0.35＝0
Patient B Three-medicine treatment 6.5 0×0.35＝0
Patient C Three-medicine treatment 8.0 1×0.35＝0.35
Patient D Three-medicine treatment 10.0 1×0.35＝0.35
In Table 3, a rule function model and a constant function model are defined, and the inference model is C ij *0.35, patient A, due to the dual medication, does not meet the rules, C ij 0, the probability of insulin use is 0; patient B was treated with three drugs, but the latest hba C was 6.5 to less than 7, and did not meet the rules, C ij 0, the probability of insulin use is 0; patient C is treated with three drugs, and latest hba C is 8 to 7, which meets the rules, C ij 1, the probability of insulin use is 0.35; patient D was treated with three drugs and was more recently hba C10 to 7, meeting the rules, C ij Is a number of 1, and is not limited by the specification,the probability of insulin use was 0.35.
The reasoning probability of the example two is the same for the patient conforming to the rule function model. However, patient C and patient D have a large difference in hba C values, and patient D has significantly higher blood glucose than patient C, and the benefit obtained from treatment with insulin is more pronounced.
Example three: based on medical knowledge, whether insulin therapy is used will be affected by the patient's bmi, c_peptide, i.e., C peptide levels, current treatment regimen, hba C, etc. results. Taking a logistic regression model as an example to train a machine learning model, predicting the probability of insulin to be used by a patient for the next visit according to the current bmi, the C peptide level, the current treatment scheme and hba C, and defining the model obtained by training as an individuation reasoning model in constraint condition 6, namely model information corresponding to a fourth element of the tetrad is as follows:
Figure SMS_5
Figure SMS_6
The probability of insulin use by different patients is calculated using the LR model described above as represented by the following formula (3):
Figure SMS_7
wherein P is the output value of the LR model, bmi, c_peptide, is_current_3drugs_therapy, and last_hba1c is the model variable, -0.2, -0.3, 0.6, and 0.8 are the model parameters obtained by training respectively.
Thus, the probability of insulin administration for the four patients described above can be represented by the following table (4):
table 4 when the rule function model and the population probability model are defined in constraint 6, the probability of diabetes patient using insulin therapy
Figure SMS_8
In the above table (4), a rule function model and an individual probability function model, that is, a fitting function model are defined, the reasoning model is c×p, and the patient a, because of adopting the double-drug treatment, does not satisfy the rule, C is 0, and the probability of using insulin is 0; patient B was treated with three drugs, but the latest hba C was 6.5 less than 7, the rule was not satisfied, C was 0, and the probability of insulin use was 0; patient C is treated with three drugs, and the latest hba C is 8 or more than 7, the rule is satisfied, C is 1, P obtained by the formula (3) is 0.608, and the probability of insulin use is 0.608; patient D was treated with three doses and was more recently hba C than 7, meeting the rules, C1, P by formula (3) 0.934, and the probability of insulin administration by patient D0.35.
Fig. 6 is a schematic block diagram of an inference apparatus for model-based medical knowledge-graph provided in accordance with some embodiments of the present application.
In the exemplary embodiment of fig. 6, the medical knowledge graph includes a plurality of quaternions including model information of a head entity, an attribute relationship, a tail entity, and a function model, the function model being a model determining constraints of the attribute relationship between the head entity and the tail entity, and referring to fig. 6, the apparatus 600 includes: the entity determining module 610 is configured to determine a first entity and a second entity corresponding to an object to be inferred; a quadruple determining module 620, configured to determine at least one quadruple corresponding to the medical knowledge graph based on the first entity and the second entity corresponding to the object to be inferred; the model determining module 630 is configured to obtain model information corresponding to the at least one quadruple, and determine a corresponding function model based on the model information; an attribute determining module 640, configured to determine the constraint condition of the attribute relationship corresponding to the at least one quadruple based on the function model and the model information.
According to the technical solution in the example embodiment of fig. 6, on one hand, based on the first entity and the second entity corresponding to the object to be inferred, determining at least one quadruple corresponding to the medical knowledge graph, the object to be inferred and the quadruple can be accurately matched; on the other hand, the constraint condition of the attribute relationship of the two entities is expressed as a function model, the model information of the function model is used as the fourth element of the four-element group in the medical knowledge graph, the constraint condition of the attribute relationship corresponding to the four-element group is determined based on the function model and the model information, the constraint condition of the attribute relationship between the two entities of the four-element group can be accurately expressed, and personalized reasoning can be carried out according to the individual condition of the patient.
In some example embodiments of the present application, the model information includes model types including: one or more of a constant function model, a fitting function model, a regular function model, and a complex function model.
In some example embodiments of the present application, the model information is information represented in JavaScript object notation JSON format, and the model determination module 630 is further configured to: analyzing the model information in the JSON format, and determining the model type in the model information; and determining a corresponding function model based on the model type.
In some example embodiments of the present application, the at least one quadruple comprises one quadruple, and the model determination module 630 is further configured to: if the model information of the four-element group comprises the rule function model and the probability model, taking the product of the rule function model and the probability model as the function model, wherein the probability model comprises the constant function model and/or the fitting function model; if the model information of the quadruple only comprises the rule function model, the rule function model is used as the function model; and if the model information of the four-element group only comprises the probability model, taking the probability model as the function model.
In some example embodiments of the present application, the at least one quadruple comprises a plurality of quadruples, and the model determination module 630 is further configured to: determining each intermediate reasoning model corresponding to the model information of each four-tuple, wherein the intermediate reasoning model comprises the rule function model and/or the probability model; and taking the product of each intermediate reasoning model as a function model corresponding to the at least one quadruple.
In some example embodiments of the present application, the model information includes model variables and model parameters, and the attribute determination module 640 is further configured to: determining the value of a corresponding model variable from medical record information of the object to be inferred based on the model variable; substituting the values of the model variables and the values of the model parameters into the function model; and determining the constraint condition of the attribute relation corresponding to at least one quadruple based on the output result of the function model.
In some example embodiments of the present application, the constraint of the attribute relationship includes a condition that the attribute relationship holds and/or a probability that the attribute relationship occurs.
The reasoning device for the model-based medical knowledge graph provided in the above embodiments is used for implementing the technical scheme of the reasoning method for the model-based medical knowledge graph in any of the foregoing method embodiments, and the implementation principle and the technical effect are similar and are not repeated here.
It should be noted that, the division of the respective modules of the apparatus provided in the above embodiments is merely a division of logic functions, and may be integrated in whole or in part into one physical entity or may be physically separated. And these modules may all be implemented in software in the form of calls by the processing element; or can be realized in hardware; the method can also be realized in a form of calling software by a processing element, and the method can be realized in a form of hardware by a part of modules. For example, the benefit index determination module may be a processing element that is set up separately, may be implemented in a chip of the above-described apparatus, or may be stored in a memory of the above-described apparatus in the form of program codes, and may be called by a processing element of the above-described apparatus to execute the functions of the above-described processing module. The implementation of the other modules is similar. In addition, all or part of the modules can be integrated together or can be independently implemented. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in a software form.
Fig. 7 is a block diagram of an electronic device for implementing a method of reasoning model-based medical knowledge-graph in accordance with an embodiment of the present application. As shown in fig. 7, is a block diagram of an electronic device according to an embodiment of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the application described and/or claimed herein.
As shown in fig. 7, the electronic device includes: one or more processors 710, memory 720, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces, and interfaces for communicating with other electronic devices. The various components are interconnected using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions executing within the electronic device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In other embodiments, multiple processors and/or multiple buses may be used, if desired, along with multiple memories and multiple memories. Also, multiple electronic devices may be connected, each providing a portion of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). One processor 710 is illustrated in fig. 7.
Memory 720 is a non-transitory computer-readable storage medium provided herein. The memory stores instructions executable by the at least one processor to cause the at least one processor to perform the reasoning method of the model-based medical knowledge graph corresponding to any execution subject provided by the present application. The non-transitory computer readable storage medium of the present application stores computer instructions for causing a computer to perform the methods provided herein.
The memory 720 is used as a non-transitory computer readable storage medium for storing non-transitory software programs, non-transitory computer executable programs, and modules, such as corresponding program instructions/modules in the reasoning method of the model-based medical knowledge-graph in the embodiments of the present application. The processor 710 executes various functional applications and data processing of the server by running non-transitory software programs, instructions and modules stored in the memory 720, i.e., implementing the reasoning method of the model-based medical knowledge graph corresponding to any of the execution subjects in the above-described method embodiments.
Memory 720 may include a storage program area that may store an operating system, at least one application program required for functionality, and a storage data area; the storage data area may then store data, such as data provided by parties stored in a data processing platform, or data in a secure isolation area, etc. In addition, memory 720 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 720 may optionally include memory located remotely from processor 710, which may be connected to the data processing electronics via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Furthermore, the electronic device may further include: an input device 730 and an output device 740. Processor 710, memory 720, input device 730, and output device 740 may be connected by a bus 750 or otherwise, as exemplified in fig. 7 by a bus connection.
The input device 730 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the data processing electronic device, such as a touch screen, keypad, mouse, trackpad, touchpad, pointer stick, one or more mouse buttons, trackball, joystick, and like input devices. The output device 740 may include a display apparatus, auxiliary lighting devices (e.g., LEDs), haptic feedback devices (e.g., vibration motors), and the like. The display device may include, but is not limited to, a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display, and a plasma display. In some implementations, the display device may be a touch screen.
Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, application specific ASIC (application specific integrated circuit), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
These computing programs (also referred to as programs, software applications, or code) include machine instructions for a programmable processor, and may be implemented in a high-level procedural and/or object-oriented programming language, and/or in assembly/machine language. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
Further, the application further provides a non-transitory computer readable storage medium storing computer instructions, where the computer instructions are configured to implement the technical solution provided by any of the foregoing method embodiments after being executed by a processor.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application can be achieved, and are not limited herein.
The above embodiments do not limit the scope of the application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are intended to be included within the scope of the present application.

Claims (5)

1. A method of reasoning a model-based medical knowledge graph, the medical knowledge graph comprising a plurality of quaternions, the quaternions comprising model information of a head entity, an attribute relationship, a tail entity, and a functional model, the functional model being a model that determines constraints of the attribute relationship between the head entity and the tail entity, the method comprising:
determining a first entity and a second entity corresponding to an object to be inferred;
Determining at least one quadruple corresponding to the medical knowledge graph based on the first entity and the second entity corresponding to the object to be inferred;
obtaining model information corresponding to the at least one tetrad, and determining a corresponding function model based on the model type of the model information; wherein the model information includes model types, model variables, and model parameters, the model types including: fitting one of a function model, a rule function model and a composite function model, or a plurality of constant function models, fitting function models, rule function models and composite function models;
determining the constraint condition of the attribute relation corresponding to the at least one quadruple based on the function model and the model information comprises the following steps: determining the value of a corresponding model variable from medical record information of the object to be inferred based on the model variable; substituting the values of the model variables and the values of the model parameters into the function model; determining the constraint condition of the attribute relation corresponding to at least one quadruple based on the output result of the function model, wherein the constraint condition of the attribute relation comprises a condition that the attribute relation is established and/or probability that the attribute relation occurs;
Wherein when the at least one quadruple comprises a quadruple, the determining a corresponding function model based on the model type of the model information comprises:
if the model information of the four-element group comprises the rule function model and the probability model, taking the product of the rule function model and the probability model as the function model, wherein the probability model comprises the constant function model and the fitting function model;
if the model information of the quadruple only comprises the rule function model, the rule function model is used as the function model;
if the model information of the quadruple only comprises the probability model, the probability model is used as the function model;
when the at least one quadruple includes a plurality of quadruples, the determining a corresponding function model based on the model type includes:
determining each intermediate reasoning model corresponding to the model information of each four-tuple, wherein the intermediate reasoning model comprises the rule function model and/or the probability model;
and taking the product of each intermediate reasoning model as a function model corresponding to the at least one quadruple.
2. The method according to claim 1, wherein the model information is information expressed in JavaScript object notation JSON format, and the determining a corresponding function model based on the model type of the model information includes:
analyzing the model information in the JSON format, and determining the model type in the model information;
and determining a corresponding function model based on the model type.
3. An inference apparatus of a model-based medical knowledge graph, wherein the medical knowledge graph includes a plurality of quaternions including model information of a head entity, an attribute relationship, a tail entity, and a function model, the function model being a model that determines constraint conditions of the attribute relationship between the head entity and the tail entity, the apparatus comprising:
the entity determining module is used for determining a first entity and a second entity corresponding to the object to be inferred;
the quadruple determining module is used for determining at least one quadruple corresponding to the medical knowledge graph based on the first entity and the second entity corresponding to the object to be inferred;
the model determining module is used for acquiring model information corresponding to the at least one quadruple and determining a corresponding function model based on the model type of the model information; wherein the model information includes model types, model variables, and model parameters, the model types including: fitting one of a function model, a rule function model and a composite function model, or a plurality of constant function models, fitting function models, rule function models and composite function models;
An attribute determining module, configured to determine, based on the function model and the model information, the constraint condition of the attribute relationship corresponding to the at least one quadruple, including: determining the value of a corresponding model variable from medical record information of the object to be inferred based on the model variable; substituting the values of the model variables and the values of the model parameters into the function model; determining the constraint condition of the attribute relation corresponding to at least one quadruple based on the output result of the function model, wherein the constraint condition of the attribute relation comprises a condition that the attribute relation is established and/or probability that the attribute relation occurs;
the model determination module is further configured to, when the at least one quadruple comprises a quadruple,
the model determining module is configured to take a product of the rule function model and the probability model as the function model if the model information of the four-tuple includes the rule function model and the probability model, where the probability model includes the constant function model and the fitting function model;
if the model information of the quadruple only comprises the rule function model, the rule function model is used as the function model;
If the model information of the quadruple only comprises the probability model, the probability model is used as the function model;
when the at least one quadruple comprises a plurality of quadruples, the model determining module is used for determining each intermediate reasoning model corresponding to the model information of each quadruple, and the intermediate reasoning model comprises the rule function model and/or the probability model; and taking the product of each intermediate reasoning model as a function model corresponding to the at least one quadruple.
4. An electronic device, comprising:
at least one processor, memory, and an interface to communicate with other electronic devices;
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of reasoning about model-based medical knowledge-graph of any one of claims 1 to 2.
5. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method of reasoning model-based medical knowledge-graph of any one of claims 1 to 2.
CN202110057441.5A 2021-01-15 2021-01-15 Reasoning method, device, equipment and medium of medical knowledge graph based on model Active CN112732940B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202110057441.5A CN112732940B (en) 2021-01-15 2021-01-15 Reasoning method, device, equipment and medium of medical knowledge graph based on model
PCT/CN2021/128491 WO2022151811A1 (en) 2021-01-15 2021-11-03 Model-based medical knowledge graph reasoning method and apparatus, device, and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110057441.5A CN112732940B (en) 2021-01-15 2021-01-15 Reasoning method, device, equipment and medium of medical knowledge graph based on model

Publications (2)

Publication Number Publication Date
CN112732940A CN112732940A (en) 2021-04-30
CN112732940B true CN112732940B (en) 2023-07-14

Family

ID=75591763

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110057441.5A Active CN112732940B (en) 2021-01-15 2021-01-15 Reasoning method, device, equipment and medium of medical knowledge graph based on model

Country Status (2)

Country Link
CN (1) CN112732940B (en)
WO (1) WO2022151811A1 (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112732940B (en) * 2021-01-15 2023-07-14 医渡云（北京）技术有限公司 Reasoning method, device, equipment and medium of medical knowledge graph based on model
CN114297411B (en) * 2022-03-08 2023-05-02 支付宝(杭州)信息技术有限公司 Configuration method, analysis method and device of entities in knowledge graph
CN115658925B (en) * 2022-11-21 2023-04-25 阿里健康科技(杭州)有限公司 Medical knowledge group generation method, apparatus, computer device and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111260064A (en) * 2020-04-15 2020-06-09 中国人民解放军国防科技大学 Knowledge inference method, system and medium based on knowledge graph of meta knowledge
CN111680109A (en) * 2020-04-22 2020-09-18 北京三快在线科技有限公司 Knowledge graph representation learning model training method and device and electronic equipment
CN112037920A (en) * 2020-08-31 2020-12-04 康键信息技术（深圳）有限公司 Medical knowledge map construction method, device, equipment and storage medium

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170277841A1 (en) * 2016-03-23 2017-09-28 HealthPals, Inc. Self-learning clinical intelligence system based on biological information and medical data metrics
CN108446769B (en) * 2018-01-23 2020-12-08 深圳市阿西莫夫科技有限公司 Knowledge graph relation inference method, knowledge graph relation inference device, computer equipment and storage medium
CN111383773B (en) * 2018-12-28 2023-04-28 医渡云（北京）技术有限公司 Medical entity information processing method and device, storage medium and electronic equipment
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
CN109815343B (en) * 2019-01-28 2021-01-22 北京百度网讯科技有限公司 Method, apparatus, device and medium for obtaining data models in a knowledge graph
CN110113314B (en) * 2019-04-12 2021-05-14 中国人民解放军战略支援部队信息工程大学 Network security domain knowledge graph construction method and device for dynamic threat analysis
US11152125B2 (en) * 2019-06-06 2021-10-19 International Business Machines Corporation Automatic validation and enrichment of semantic relations between medical entities for drug discovery
US11227018B2 (en) * 2019-06-27 2022-01-18 International Business Machines Corporation Auto generating reasoning query on a knowledge graph
CN111242448B (en) * 2020-01-06 2021-05-14 思创数码科技股份有限公司 Fire hazard assessment method and device based on correlation analysis
CN111950285B (en) * 2020-07-31 2024-01-23 合肥工业大学 Medical knowledge graph intelligent automatic construction system and method with multi-mode data fusion
CN112084344A (en) * 2020-09-11 2020-12-15 清华大学 Knowledge graph reasoning method, device and storage medium
CN112732940B (en) * 2021-01-15 2023-07-14 医渡云（北京）技术有限公司 Reasoning method, device, equipment and medium of medical knowledge graph based on model

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111260064A (en) * 2020-04-15 2020-06-09 中国人民解放军国防科技大学 Knowledge inference method, system and medium based on knowledge graph of meta knowledge
CN111680109A (en) * 2020-04-22 2020-09-18 北京三快在线科技有限公司 Knowledge graph representation learning model training method and device and electronic equipment
CN112037920A (en) * 2020-08-31 2020-12-04 康键信息技术（深圳）有限公司 Medical knowledge map construction method, device, equipment and storage medium

Also Published As

Publication number Publication date
CN112732940A (en) 2021-04-30
WO2022151811A1 (en) 2022-07-21

Similar Documents

Publication Publication Date Title
CN112732940B (en) Reasoning method, device, equipment and medium of medical knowledge graph based on model
CN112732941B (en) Method, device, equipment and medium for constructing medical knowledge graph based on model
Schwab et al. Clinical predictive models for COVID-19: systematic study
CN111681726B (en) Processing method, device, equipment and medium of electronic medical record data
Kwon et al. Validation of deep-learning-based triage and acuity score using a large national dataset
Bali et al. Diabot: a predictive medical chatbot using ensemble learning
CN106407666A (en) Method, apparatus and system for generating electronic medical record information
CN110517785A (en) Lookup method, device and the equipment of similar case
US20220035998A1 (en) Obtaining supported decision trees from text for medical health applications
US20130282393A1 (en) Combining knowledge and data driven insights for identifying risk factors in healthcare
CN111275121B (en) Medical image processing method and device and electronic equipment
Ho et al. A call for an ethics and governance action plan to harness the power of artificial intelligence and digitalization in nephrology
CN114255837A (en) Data query method and device, computer readable storage medium and electronic equipment
WO2021179694A1 (en) Drug recommendation method, apparatus, computer device, and storage medium
Wardi et al. Bringing the promise of artificial intelligence to critical care: what the experience with sepsis analytics can teach us
US20200278860A1 (en) Cognitive service updates via container instantiation
Wang et al. Bayesian adaptive lasso for additive hazard regression with current status data
Heavner et al. A path to real-world evidence in critical care using open-source data harmonization tools
JP2017153691A (en) Diagnosis support apparatus, control method for diagnosis support apparatus, and program
Kuo et al. Mining surprising patterns and their explanations in clinical data
Dheeba A heart disease prognosis pipeline for the edge using federated learning
CN108292522A (en) The comprehensive verification of patient medical record
KR20220164985A (en) Method for vectorizing medical data for machine learning, data transforming apparatus and data transforming program
Hurley et al. Clinical Phenotyping with an Outcomes-driven Mixture of Experts for Patient Matching and Risk Estimation
US10574510B2 (en) System for decomposing events from managed infrastructures using syntax pattern

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