Extracting clinical entities from clinical documents

1. A non-transitory machine-readable medium storing a program executable by at least one processor of a device, the program comprising sets of instructions for:

receiving a clinical document partitioned into a set of sections, each section in the set of sections comprising a plurality of terms;

identifying a section in the set of sections;

determining a classification of the section using a section classification map;

selecting a subset of a plurality of entity recognizers based on the classification of the section, each entity recognizer in the plurality of entity recognizers configured to identify terms in the section as being associated with a particular type of data that is associated with the classification;

sending the section to the subset of the entity recognizers for processing; and

generating a clinical statement based on the terms identified in the section by the subset of the plurality of entity recognizers.

2. The non-transitory machine-readable medium of claim 1 , wherein the program further comprises sets of instructions for:

sending the section to a terminology service for processing, the terminology service configured to determine medical codes based on the section and associating the medical codes with a subset of the terms identified in the section by the plurality of entity recognizers; and

receiving the section from the terminology service with the medical codes associated with the subset of the terms.

3. The non-transitory machine-readable medium of claim 1 , wherein the program further comprises sets of instructions for:

receiving the section classification map associated with the clinical document, wherein each section entry of the section classification map comprises a classification and a confidence score for the classification.

4. The non-transitory machine-readable medium of claim 1 , wherein the program further comprises a set of instructions for sending the clinical statement to an aggregation service for processing.

5. The non-transitory machine-readable medium of claim 1 , wherein the subset of the plurality of entity recognizers is a first subset of the plurality of entity recognizers, wherein the first subset of the plurality of entity recognizers is included in a first natural language processing (NLP) pipeline, wherein the program further comprises sets of instructions for:

determining a second subset of the plurality of entity recognizers based on the classification of the section; and

sending the section to the second subset of the entity recognizers for processing, wherein generating the clinical statement is further based on terms identified in the section by the second subset of the plurality of entity recognizers.

6. The non-transitory machine-readable medium of claim 1 , wherein the clinical statement is a first clinical statement, wherein the program further comprises sets of instructions for generating a second clinical statement based on terms identified in the section by the subset of the plurality of entity recognizers.

7. The non-transitory machine-readable medium of claim 1 , wherein the section is a first section in the set of sections, wherein the subset of the plurality of entity recognizers is a first subset of the plurality of entity recognizers, wherein the clinical statement is a first clinical statement, wherein the program further comprises sets of instructions for:

identifying a second section in the set of sections;

determining a second subset of the plurality of entity recognizers based on a classification of the second section;

sending the second section to the second subset of the entity recognizers for processing; and

generating a second clinical statement based on terms identified in the second section by the second subset of the plurality of entity recognizers.

8. A method comprising:

receiving a clinical document partitioned into a set of sections, each section in the set of sections comprising a plurality of terms;

identifying a section in the set of sections;

determining a classification of the section;

selecting a subset of a plurality of entity recognizers based on the classification of the section, each entity recognizer in the plurality of entity recognizers configured to identify terms in the section associated with a particular type of data that is associated with the classification;

sending the section to the subset of the entity recognizers for processing; and

generating a clinical statement based on terms identified in the section by the subset of the plurality of entity recognizers.

9. The method of claim 8 further comprising:

sending the section to a terminology service for processing, the terminology service configured to determine medical codes based on the section and associating the medical codes with a subset of the terms identified in the section by the plurality of entity recognizers; and

receiving the section from the terminology service with the medical codes associated with the subset of the terms.

10. The method of claim 8 further comprising:

receiving a section classification map associated with the clinical document;

wherein the classification of the section is determined based on the section classification map.

11. The method of claim 8 further comprising sending the clinical statement to an aggregation service for processing.

12. The method of claim 8 , wherein the subset of the plurality of entity recognizers is a first subset of the plurality of entity recognizers, wherein the first subset of the plurality of entity recognizers is included in a first natural language processing (NLP) pipeline, wherein the method further comprises:

determining a second subset of the plurality of entity recognizers based on the classification of the section; and

sending the section to the second subset of the entity recognizers for processing, wherein generating the clinical statement is further based on terms identified in the section by the second subset of the plurality of entity recognizers.

13. The method of claim 8 , wherein the clinical statement is a first clinical statement, wherein the method further comprises generating a second clinical statement based on terms identified in the section by the subset of the plurality of entity recognizers.

14. The method of claim 8 , wherein the section is a first section in the set of sections, wherein the subset of the plurality of entity recognizers is a first subset of the plurality of entity recognizers, wherein the clinical statement is a first clinical statement, wherein the program further comprises sets of instructions for:

identifying a second section in the set of sections;

determining a second subset of the plurality of entity recognizers based on a classification of the second section;

sending the second section to the second subset of the entity recognizers for processing; and

generating a second clinical statement based on terms identified in the second section by the second subset of the plurality of entity recognizers.

15. A system comprising:

at least one processor; and

a memory having stored thereon program code that, when executed by the at least one processor, causes the processor to:

receive a clinical document partitioned into a set of sections, each section in the set of sections comprising a plurality of terms;

identify a section in the set of sections;

determine a classification of the section;

select a subset of a plurality of entity recognizers based on the classification of the section, each entity recognizer in the plurality of entity recognizers configured to identify terms in the section as being associated with a particular type of data that is associated with the classification;

send the section to the subset of the entity recognizers for processing; and

generate a clinical statement based on terms identified in the section by the subset of the plurality of entity recognizers.

16. The system of claim 15 , wherein the instructions further cause the at least one processor to:

send the section to a terminology service for processing, the terminology service configured to determine medical codes based on the section and associating the medical codes with a subset of the terms identified in the section by the plurality of entity recognizers; and

receive the section from the terminology service with the medical codes associated with the subset of the terms.

17. The system of claim 15 , wherein the instructions further cause the at least one processor to:

receive a section classification map associated with the clinical document; wherein the classification of the section is determined based on the section classification map.

18. The system of claim 15 , wherein the subset of the plurality of entity recognizers is a first subset of the plurality of entity recognizers, wherein the first subset of the plurality of entity recognizers is included in a first natural language processing (NLP) pipeline, wherein the instructions further cause the at least one processor to:

determine a second subset of the plurality of entity recognizers based on the classification of the section; and

send the section to the second subset of the entity recognizers for processing, wherein generating the clinical statement is further based on terms identified in the section by the second subset of the plurality of entity recognizers.

19. The system of claim 15 , wherein the clinical statement is a first clinical statement, wherein the instructions further cause the at least one processor to generate a second clinical statement based on terms identified in the section by the subset of the plurality of entity recognizers.

20. The system of claim 15 , wherein the section is a first section in the set of sections, wherein the subset of the plurality of entity recognizers is a first subset of the plurality of entity recognizers, wherein the clinical statement is a first clinical statement, wherein the instructions further cause the at least one processor to:

identify a second section in the set of sections;

determine a second subset of the plurality of entity recognizers based on a classification of the second section;

send the second section to the second subset of the entity recognizers for processing; and

generate a second clinical statement based on terms identified in the second section by the second subset of the plurality of entity recognizers.