Method for constructing database to deduce disease and providing U-health service using a subset of ontologies
View Patent ↗A method for constructing a database to deduce a disease and constructing a Bayesian network for U-Health application, the method including the steps of analyzing U-Health information, which includes a disease, a symptom, and a treatment, and constructing a plurality of U-Health ontologies required for service provision, setting a meta-model defining cause-and-effect relationships between the constructed U-Health ontologies and selecting at least two specific U-Health ontologies from among the plurality of U-Health ontologies, setting the selected U-Health ontologies as nodes, and generating a Bayesian network by applying the meta-model to the set nodes.
1. A method, implemented in at least one processor, for constructing a database to deduce a disease, the method comprising:
the at least one processor:
analyzing U-Health information, which includes a disease, a symptom, and a treatment, and constructing a plurality of U-Health ontologies required for service provision;
setting a meta-model defining cause-and-effect relationships between the constructed U-Health ontologies; and
selecting a subset of the constructed U-Health ontologies comprising at least two specific U-Health ontologies from among the plurality of U-Health ontologies,
setting the selected U-Health ontologies as nodes,
generating a Bayesian network by applying the meta-model to the set nodes to thereby construct a customized database for inferring a disease of a user;
whereby upon receiving personal information of the user,
a disease of a user is inferred using a probabilistic analysis of said Bayesian network.
2. The method as claimed in claim 1 , wherein said analyzing comprises:
setting common ontologies which comprises various kinds of ontologies required for disease deduction; and
setting an intermediate ontology, which comprises at least one specific ontology relating to a specific disease, from among the common ontologies.
3. The method as claimed in claim 1 , wherein the U-Health ontologies comprise:
a disease ontology in which information on kinds of disease is defined,
a symptom ontology in which information on symptoms caused by diseases is defined,
an environment ontology in which information on environments of a person targeted for disease deduction is defined, and
a treatment ontology in which treatment methods for diseases are defined.
4. The method as claimed in claim 3 , wherein the U-Health ontologies further comprises:
user information ontology, including age, sex, family disease history, and biological signals.
5. The method as claimed in claim 3 , wherein the U-Health ontologies further comprise:
a sensor ontology including sensor information including kinds of sensors used for measuring biological signals of a user, and information on values measured by the sensors.
6. The method as claimed in claim 1 , wherein the U-Health ontology corresponds to an ontology input from a user through an interface which enables the user to edit the ontology.
7. The method as claimed in claim 2 , further comprising:
constructing the intermediate ontology by:
receiving personal information, including age, sex, family disease history, and symptoms,
extracting specific classes corresponding to the personal information from among a plurality of classes included in the ontology, and
combining the extracted classes.
8. The method as claimed in claim 1 , wherein the meta-model corresponds to a model defining cause-and-effect relationships between a plurality of ontologies.
9. The method as claimed in claim 1 , wherein said selecting a subset comprising at least two specific U-Health ontologies comprises:
providing a user terminal with a hierarchical structure of the plurality of ontologies, and setting an abstraction level by receiving a selection of a level of a specific layer which relates to the U-Health service to be provided to a user;
wherein said setting comprises setting an ontology included in at least one selected layer as a node of the Bayesian network by taking the set abstraction level into consideration; and
said generating a Bayesian network comprises connecting nodes of the Bayesian network with each other by taking the meta-model into consideration.
10. The method as claimed in claim 9 , wherein, in the step of setting the abstraction level, a layer and a highest layer level, in which an ontology of a disease having no connection with a specific disease for which a U-Health service is provided to diagnose among the plurality of ontologies is included, are restricted from being selected.
11. The method as claimed in claim 9 , wherein the nodes of the Bayesian network comprise a main Bayesian network node constituted by an ontology which is included in a selected abstraction level's layer, and a sub-Bayesian network node constituted by an ontology which is included in a lower layer of the main Bayesian network node.
12. A method, implemented in at least one processor, for providing a U-Health service by using a Bayesian network, the method comprising the steps of:
the at least one processor:
analyzing U-Health information, which includes a disease, a symptom, and a treatment, and constructing a plurality of U-Health ontologies required for service provision;
setting a meta-model defining cause-and-effect relationships between the constructed U-Health ontologies;
selecting a subset of the constructed U-Health ontologies comprising at least two specific U-Health ontologies from among the plurality of U-Health ontologies, setting the selected U-Health ontologies as nodes, and generating a Bayesian network by applying the meta-model to the set nodes; and
receiving personal information, which includes age, sex, family disease history, and symptoms, from a user, substituting the personal information into a customized database constructed based on the Bayesian network, and probabilistically inferring a disease of a user.
13. A system for providing a U-Health service, the system comprising:
a user terminal; and
a server connected with the user terminal through a communication network,
wherein the server comprises:
an ontology management unit for providing an interface to receive an ontology from the user terminal, and storing a plurality of ontologies received from a plurality of users in a database;
a meta-model management unit for providing an interface to receive a meta-model defining cause-and-effect relationships between ontologies from the user terminal, and storing the received meta-model in a database;
an abstraction level setting unit for providing the user terminal with a hierarchical structure of a plurality of ontologies, receiving a user selection of a level of a specific layer relating to a U-Health service to be provided to a user, and setting an abstraction level;
a Bayesian-network node generation unit for setting an ontology included in at least one selected layer as a node of a Bayesian network by taking the set abstraction level into consideration;
a Bayesian-network link generation unit for connecting nodes of the Bayesian network with each other by taking the meta-model into consideration whereby the Bayesian network is generated as a customized database on the basis of the user selection;
a U-Healthcare management unit for receiving personal information, which includes a user's age, sex, family disease history, and symptoms, from the user terminal, substituting the personal information into a database constructed based on the Bayesian network, and probabilistically inferring a disease of a user and a result of a diagnosis; and
a web page management unit for providing an input/output interface with the user terminal through a web page, and providing the user terminal with a result of the inference by the U-Healthcare management unit.
14. Apparatus for constructing a database to deduce a disease and constructing a Bayesian network for U-Health application, comprising:
a memory including code which when accessed by a processor causes the processor to execute the steps of:
analyzing U-Health information, which includes a disease, a symptom, and a treatment, and constructing a plurality of U-Health ontologies required for service provision;
setting a meta-model defining cause-and-effect relationships between the constructed U-Health ontologies; and
selecting a subset of the constructed U-Health ontologies comprising at least two specific U-Health ontologies from among the plurality of U-Health ontologies,
setting the selected U-Health ontologies as nodes,
generating a Bayesian network by applying the meta-model to the set nodes to thereby construct a customized database for inferring a disease of a user;
whereby upon receiving personal information of the user,
a disease of the user is inferred using a probabilistic analysis of said Bayesian network.
15. The apparatus as claimed in claim 14 , wherein step of analyzing comprises the steps of:
setting common ontologies which comprises various kinds of ontologies required for disease deduction; and
setting an intermediate ontology, which comprises at least one specific ontology relating to a specific disease, from among the common ontologies.
16. The apparatus as claimed in claim 14 , wherein the U-Health ontologies comprise:
a disease ontology in which information on kinds of disease is defined,
a symptom ontology in which information on symptoms caused by diseases is defined,
an environment ontology in which information on environments of a person targeted for disease deduction is defined, and
a treatment ontology in which treatment methods for diseases are defined.
17. The apparatus as claimed in claim 16 , wherein the U-Health ontologies further comprises:
user information ontology, including age, sex, family disease history, and biological signals.
18. The apparatus as claimed in claim 16 , wherein the U-Health ontologies further comprise:
a sensor ontology including sensor information including kinds of sensors used for measuring biological signals of a user, and information on values measured by the sensors.
19. The apparatus as claimed in claim 14 , wherein the U-Health ontology corresponds to an ontology input from a user through an interface which enables the user to edit the ontology.
20. The apparatus as claimed in claim 15 , wherein said processor constructs the intermediate ontology by executing the steps of:
receiving personal information, including age, sex, family disease history, and symptoms,
extracting specific classes corresponding to the personal information from among a plurality of classes included in the ontology, and
combining the extracted classes.
21. The apparatus as claimed in claim 14 , wherein the meta-model corresponds to a model defining cause-and-effect relationships between a plurality of ontologies.
22. The apparatus as claimed in claim 14 , wherein the step of selecting at least two specific U-Health ontologies comprises the steps of:
providing a user terminal with a hierarchical structure of the plurality of ontologies, and setting an abstraction level by receiving a selection of a level of a specific layer which relates to the U-Health service to be provided to a user;
setting an ontology included in at least one selected layer as a node of the Bayesian network by taking the set abstraction level into consideration; and
connecting nodes of the Bayesian network with each other by taking the meta-model into consideration.
23. The apparatus as claimed in claim 22 , wherein, in the step of setting the abstraction level, a layer and a highest layer level, in which an ontology of a disease having no connection with a specific disease for which a U-Health service is provided to diagnose, among the plurality of ontologies included, are restricted from being selected.
24. The apparatus as claimed in claim 22 , wherein the nodes of the Bayesian network comprise a main Bayesian network node constituted by an ontology which is included in a selected abstraction level's layer, and a sub-Bayesian network node constituted by an ontology which is included in a lower layer of the main Bayesian network node.