Service architecture for ontology linking of unstructured text
Techniques for ontology linking of unstructured text as a service are described. A service may receive a request to link unstructured text to a standardized ontology, and the service may segment and tokenize the unstructured text and send the result to multiple services implementing multiple deep machine learning models trained to identify particular entities and one or more relationships between entities. The service may perform a search of the standardized ontology to identify a set of similar candidates from the standardized ontology for the detected entities and the one or more relationships, and then rank the set of similar candidates from the standardized ontology according to their similarity to the detected entities within the unstructured text. The output from the service may include a result identifying a highest ranked candidate of the set of similar candidates from the standardized ontology for the detected entities within the unstructured text.
1. A computer-implemented method comprising:
receiving unstructured text;
executing at least one machine learning model on the unstructured text to detect named entities of a medical condition type and an anatomy type;
executing the at least one machine learning model to detect named entities of a test type, a test result type, a medication type, and a treatment type;
executing the at least one machine learning model on the named entities to identify one or more relationships between the named entities;
generating a result identifying the named entities and the one or more relationships between the named entities within the unstructured text; and
transmitting the result.
2. The computer-implemented method of claim 1 , further comprising executing the at least one machine learning model to identify a similar concept from a medical ontology based on the named entities.
3. The computer-implemented method of claim 1 , wherein the result further comprises an entity category for the named entities.
4. The computer-implemented method of claim 1 , wherein the medical condition type comprises a symptom type.
5. The computer-implemented method of claim 4 , wherein the medical condition type comprises a diagnosis type.
6. The computer-implemented method of claim 1 , wherein the medical condition type comprises a diagnosis type.
7. The computer-implemented method of claim 1 , wherein the named entities of the anatomy type comprise a body system type, an anatomic location or region, and a body site.
8. The computer-implemented method of claim 1 , wherein the named entities of the medication type comprise a dosage, a frequency, and a method of administration.
9. The computer-implemented method of claim 8 , wherein the result comprises a branded medication name for medication text in the unstructured text.
10. The computer-implemented method of claim 1 , wherein:
the executing the at least one machine learning model on the unstructured text to detect named entities of the medical condition type and the anatomy type comprises executing a first machine learning model; and
the executing the at least one machine learning model on the named entities to identify one or more relationships between the named entities comprises executing a second machine learning model.
11. A non-transitory machine-readable medium that stores code that when executed by a machine causes the machine to perform a method comprising:
receiving unstructured text;
executing at least one machine learning model on the unstructured text to detect named entities of a medical condition type and an anatomy type;
executing the at least one machine learning model to detect named entities of a test type, a test result type, a medication type, and a treatment type;
executing the at least one machine learning model on the named entities to identify one or more relationships between the named entities;
generating a result identifying the named entities and the one or more relationships between the named entities within the unstructured text; and
transmitting the result.
12. The non-transitory machine-readable medium of claim 11 , wherein the method further comprises executing the at least one machine learning model to identify a similar concept from a medical ontology based on the named entities.
13. The non-transitory machine-readable medium of claim 11 , wherein the result further comprises an entity category for the named entities.
14. The non-transitory machine-readable medium of claim 11 , wherein the medical condition type comprises a symptom type.
15. The non-transitory machine-readable medium of claim 14 , wherein the medical condition type comprises a diagnosis type.
16. The non-transitory machine-readable medium of claim 11 , wherein the medical condition type comprises a diagnosis type.
17. The non-transitory machine-readable medium of claim 11 , wherein the named entities of the anatomy type comprise a body system type, an anatomic location or region, and a body site.
18. The non-transitory machine-readable medium of claim 11 , wherein the named entities of the medication type comprise a dosage, a frequency, and a method of administration.
19. The non-transitory machine-readable medium of claim 18 , wherein the result comprises a branded medication name for medication text in the unstructured text.
20. The non-transitory machine-readable medium of claim 11 ,
wherein:
the executing the at least one machine learning model on the unstructured text to detect named entities of the medical condition type and the anatomy type comprises executing a first machine learning model; and
the executing the at least one machine learning model on the named entities to identify one or more relationships between the named entities comprises executing a second machine learning model.