Ontologically driven procedure coding

• US 10,133,727 B2
• Filed: 10/01/2013
• Issued: 11/20/2018
• Est. Priority Date: 10/01/2013
• Status: Active Grant

• Current Assignee: Optum360, LLC (UnitedHealth Group Incorporated)
• Original Assignee: A-Life Medical, LLC
• Inventors: Karres, George, Tormey, Destinee, Miller, Christopher, Potter, Brian, Morsch, Mark L.
• Primary Examiner(s): Mishra, Richa
• Application Number:
  US14/043,344
• Publication Number:
  US 20150095016A1
• Time in Patent Office: 1,876 Days
• Field of Search: None
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Computer implemented systems and methods of processing clinical documentation for a multi-axial coding scheme include inputting clinical documentation from memory operatively coupled with a computer system, and executing a natural language processor configured to process narrative text in the clinical documentation. The processor segments the narrative text based on boundaries defined in the clinical documentation, sequences words in the narrative text based on the segmentation, and maps the sequenced words to semantic objects in an ontology database. The ontology defines classes of semantic objects and relationships between them, corresponding to the multi-axial coding scheme. The semantic objects are converted into characters and output into slots in a medical code, with the characters positioned in the slots based on the multi-axial coding scheme.

Claims

1. A computer implemented method of processing clinical documentation for a multi-axial coding scheme, the method comprising:

inputting clinical documentation from memory operatively coupled with a computer system, the clinical documentation comprising narrative text;

executing a natural language processor on the computer system, the natural language processor configured to;

segment the narrative text based on boundaries defined in the clinical documentation;

sequence words in the narrative text, based on the segmented text;

map the sequenced words to semantic objects in an ontology, the ontology defining classes of the semantic objects corresponding to axes of the multi-axial coding scheme and concepts of a conceptual data resource, wherein the classes of semantic objects define a hierarchical structure in the ontology, the hierarchical structure defining conditions on and relationships between the semantic objects;

convert the semantic objects into characters;

output, to a user interface, the characters into slots in a procedural or a diagnostic medical code, wherein the characters are positioned in the slots based on the multi-axial coding scheme and wherein in the multi-axial coding system each position of a character within the procedural or diagnostic medical code corresponds to a semi-independent axis of classification of a procedure or diagnosis;

output, to the user interface, trace data that includes linking information to map characters in a selected slot in the procedural or the diagnostic medical code to particular locations in the narrative text of the clinical documentation from which the characters in the selected slot in the medical code are derived; and

infer at least one of the characters in at least one of the slots in the procedural or the diagnostic medical code, wherein the inferred character is associated with a particular semantic object that is not mapped to sequenced words in the segmented text, the particular sematic object being inferred from one or more other semantic objects that are mapped to sequenced words in the segmented text.

2. The method of claim 1, further comprising marking one or more of the slots in the medical code with a wildcard, the wildcard indicating that one or more particular medical concept necessary to specify one or more of the characters was not identified in the narrative text.

3. The method of claim 1, wherein the hierarchical structure defines relationships among the characters in the slots based on the multi-axial coding scheme.

4. The method of claim 1, wherein the conditions requiring a laterality for a character in a slot associated with particular body parts.

5. The method of claim 3, wherein the hierarchical structure defines a relationship between a character in a slot associated with a body part and appropriate characters in another slot associated with a core procedure or root operation.

6. The method of claim 3, wherein the hierarchical structure defines a semantic relationship between two or more classes in the ontology.

7. The method of claim 1, further comprising classifying a context for one or more of the semantic objects based on the narrative text, wherein the one or more semantic objects are not converted into characters for outputting into slots in the medical code, based on the context.

8. The method of claim 7, wherein the context is classified as past, negated or other context pertinent to medical concept recognition based on machine learning using sample clinical documentation.

9. The method of claim 1, further comprising filtering the semantic objects based on rules reflecting medical coding guidelines, wherein at least one of the semantic objects is not converted into a character based on an inherent procedure defined in the multi-axial coding scheme.

10. A computer system comprising:

a computer processor operatively coupled to an interactive display;

a database in communication with the computer processor, the database comprising memory for storing clinical documentation;

a natural language engine executing on the computer processor, the natural language engine configured to;

input narrative text from the clinical documentation, the narrative text comprising sections of words;

segment the sections of words based on boundaries defined between the sections of words in the documentation;

map the segmented sections of words to semantic objects in an ontology, the ontology defining classes of the semantic objects based on axes of a multi-axial coding scheme for coding the clinical documentation and based on concepts of a conceptual data resource, wherein the classes of semantic objects define a hierarchical structure in the ontology, the hierarchical structure defining conditions on and relationships between the semantic objects;

convert the semantic objects into characters;

position the characters into slots in a procedural or a diagnostic medical code, wherein positions of the slots are defined by the multi-axial coding scheme and wherein in the multi-axial coding system each position of a character within the procedural or diagnostic medical code corresponds to a semi-independent axis of classification of a procedure or diagnosis;

output the characters to the user interface;

output, to the user interface, trace data that includes linking information to map characters in a selected slot in the procedural or the diagnostic medical code to particular locations in the narrative text of the clinical documentation from which the characters in the selected slot in the medical code are derived; and

infer a character in at least one of the slots in the procedural or the diagnostic medical code, wherein the inferred character is associated with a semantic object that is not mapped to sequenced words in the segmented text, the sematic object being inferred from one or more other semantic objects that are mapped to sequenced words in the segmented text.

11. The computer system of claim 10, wherein the user interface is configured for a user to accept or reject the code based on the trace data.

12. The computer system of claim 10, further comprising marking one or more of the slots in the medical code with a wildcard, the wildcard indicating that one or more particular medical concept necessary to specify one or more of the characters was not identified in the narrative text.

13. The computer system of claim 10, wherein the user interface is further configured for the user to fill in the wildcard based on the trace data.

14. A non-transitory computer-readable storage medium having program code embedded thereon, the program code executable on a natural language processor of a computer system to perform a method comprising:

reading clinical documentation from memory operatively coupled with the computer system;

segmenting the narrative text into sections of words based on boundaries defined in the clinical documentation;

sequencing items representing the words and boundaries based on the segmented text;

mapping the sequenced items representing the words and boundaries to semantic objects in an ontology, the ontology defining classes of the semantic objects corresponding to axes of a multi-axial medical coding scheme and concepts of a conceptual data resource, wherein the classes of semantic objects define a hierarchical structure in the ontology, the hierarchical structure defining conditions on and relationships between the semantic objects;

converting the semantic objects into characters based on the multi-axial medical coding scheme;

outputting, to a user interface, the characters into slots in a procedural or a diagnostic medical code, wherein the slots correspond to the axes and the characters are positioned in the slots based on the multi-axial medical coding scheme and wherein in the multi-axial coding system each position of a character within the procedural or diagnostic medical code corresponds to a semi-independent axis of classification of a procedure or diagnosis;

outputting, to the user interface, trace data that includes linking information to map characters in a selected slot in the procedural or the diagnostic medical code to particular locations in the narrative text of the clinical documentation from which the characters in the selected slot in the medical code are derived; and

inferring a character in at least one of the slots in the procedural or the diagnostic medical code, wherein the inferred character is associated with a semantic object that is not mapped to sequenced words in the segmented text, the sematic object being inferred from one or more other semantic objects that are mapped to sequenced words in the segmented text.

15. The storage medium of claim 14, wherein the method further comprises marking at least one of the slots in the medical code with a wildcard, the wildcard indicating that one or more medical concepts required to specify one or more of the characters was not found in the narrative text.

16. The storage medium of claim 14, wherein the method further comprises classifying a context one or more of the semantic objects based on the narrative text, wherein the one or more semantic object are not converted to characters for outputting into the slots in the medical code, based on the context.

17. A method comprising:

inputting narrative text from clinical documentation stored in memory, the memory operatively coupled with a computer system having a user interface;

executing a natural language processor on the computer system, wherein the natural language processor is configured to;

segment the narrative text based on boundaries defined in the clinical documentation;

map the segmented text to semantic objects in an ontology defining classes of the semantic objects corresponding to axes of a multi-axial medical coding scheme and concepts of a conceptual data resource, wherein the classes of semantic objects define a hierarchical structure in the ontology, the hierarchical structure defining conditions on and relationships between the semantic objects;

convert the semantic objects into characters positioned into slots in a procedural or diagnostic medical code, displayed on the user interface, based on the multi-axial medical coding scheme and wherein in the multi-axial coding system each position of a character within the procedural or diagnostic medical code corresponds to a semi-independent axis of classification of a procedure or diagnosis,

mark one or more of the slots in the procedural or diagnostic medical code with a wildcard, the wildcard indicating that a particular medical concept necessary to specify one or more of the characters is not identified in the narrative text; and

output, to a user interface, trace data that includes linking information to map characters in a selected slot in the procedural or the diagnostic medical code to particular locations in the narrative text of the clinical documentation from which the characters in the selected slot in the medical code are derived; and

infer a character in at least one of the slots in the procedural or the diagnostic medical code, wherein the inferred character is associated with a semantic object that is not mapped to sequenced words in the segmented text, the sematic object being inferred from one or more other semantic objects that are mapped to sequenced words in the segmented text.

18. A method comprising:

inputting narrative text from clinical documentation stored in memory, the memory operatively coupled with a computer system;

executing a natural language processor on the computer system, wherein the natural language processor is configured to;

segment the narrative text based on boundaries defined in the clinical documentation, sequences words in the narrative text based on the segmented text;

map the sequenced words to semantic objects in an ontology defining classes of the semantic objects corresponding to axes of a multi-axial medical coding scheme;

convert the semantic objects into characters, based on the ontology;

position the characters into slots in a procedural or a diagnostic medical code based on the multi-axial medical coding scheme wherein in the multi-axial coding system each position of a character within the procedural or diagnostic medical code corresponds to a semi-independent axis of classification of a procedure or diagnosis; and

output, to a user interface, trace data that includes linking information to map characters in a selected slot in the procedural or the diagnostic medical code to particular locations in the narrative text of the clinical documentation from which the characters in the selected slot in the medical code are derived;

infer a character in at least one of the slots in the procedural or the diagnostic medical code, wherein the inferred character is associated with a semantic object that is not mapped to sequenced words in the segmented text, the sematic object being inferred from one or more other semantic objects that are mapped to sequenced words in the segmented text;

wherein the ontology comprises classes of the semantic objects defining a hierarchical structure in the ontology, the hierarchical structure defining relationships among the characters positioned in the slots based on axes of a multi-axial coding scheme for generating the medical code, wherein the hierarchical structure defines conditions on and relationships between the semantic objects; and

wherein the hierarchical structure of the ontology reflects core medical concepts used in the medical coding scheme.

19. The method of claim 18, wherein the characters define primitive medical concepts in the natural language processor.

20. The method of claim 19, wherein the medical code does not define a primitive medical concept in the natural language processor.