Ontologically driven procedure coding

1. A computer-implemented method for identifying characters for insertion into positions of a medical code, the computer-implemented method comprising:

segmenting, using one or more processors and a natural language processing engine, unstructured narrative text of clinical documentation having boundaries between sections of narrative text within the clinical documentation, based at least in part on the boundaries defined in the clinical documentation to generate segmented narrative text;

sequencing, using the one or more processors and the natural language processing engine, words of the narrative text selected at least in part by a document filter filtering the unstructured narrative text based at least in part on the segmented narrative text to generate sequenced words;

mapping, using the one or more processors, one or more of the sequenced words to one or more candidate semantic objects of a plurality of semantic objects of an ontology, wherein (a) the ontology defines classes of the plurality of semantic objects corresponding to a multi-axial coding scheme, (b) a multi-axial coding scheme defines each position of a character within a medical code, and (c) each position of a character corresponds to a semi-independent axis of classification of a procedure, a diagnosis, or both;

filtering, using the one or more processors executing a Procedure Coding System (PCS) filter module, the one or more candidate semantic objects to select a first semantic object based at least in part on medical code filtering criteria relevant to generate a medical code;

determining, using the one or more processors, a first character to insert into a first position of the medical code based at least in part on the first semantic object and the multi-axial coding scheme;

inferring, using the one or more processors, a second semantic object of the plurality of semantic objects based at least in part on the first semantic object selected via the PCS filter module, wherein the second semantic object is not mapped to the one or more of the sequenced words based at least in part on the segmented narrative text;

determining, using the one or more processors, a second character to insert into a second position of the medical code based at least in part on the second semantic object and the multi- axial coding scheme;

inserting, using the one or more processors, the first character into the first position of the medical code and the second character into the second position of the medical code; and

outputting, using the one or more processors, the medical code comprising at least the first character and the second character.

2. The method of claim 1 , wherein outputting the medical code comprising at least the first character and the second character comprises outputting the medical code to a user interface.

3. The method of claim 1 , wherein (a) the first position comprising the first character corresponds to a first semi-independent axis of classification of a medical procedure or diagnosis, and (b) the second position comprising the second character corresponds to a second semi-independent axis of classification of the medical procedure or diagnosis.

4. The method of claim 1 , further comprising generating first trace data that comprises linking information to a location of the narrative text of the clinical documentation from which the first character was derived, wherein outputting the medical code further comprises providing the first trace data.

5. The method of claim 1 , further comprising receiving the clinical documentation from a memory.

6. The method of claim 1 , further comprising inserting a wildcard character in an additional slot position of the medical code in response to a determination that a medical concept necessary to derive a character to be positioned in the additional slot is missing from the narrative text.

7. The method of claim 1 , wherein the classes of the plurality of semantic objects define a hierarchical structure of the ontology.

8. The method of claim 7 , wherein the hierarchical structure of the ontology defines conditions on and relationships between the plurality of semantic objects.

9. The method of claim 8 , wherein the hierarchical structure further defines relationships between the first character and the second character based at least in part on the multi-axial coding scheme.

10. A computer system comprising one or more processor, a natural language processing engine, and one or more memory storage areas, the system configured to:

segment unstructured narrative text of clinical documentation having boundaries between sections of narrative text within the clinical documentation, based at least in part on the boundaries defined in the clinical documentation to generate segmented narrative text;

sequence words of the narrative text selected at least in part by a document filter filtering the unstructured narrative text based at least in part on the segmented narrative text to generate sequenced words;

map one or more of the sequenced words to one or more candidate semantic objects of a plurality of semantic objects of an ontology, wherein (a) the ontology defines classes of the plurality of semantic objects corresponding to a multi-axial coding scheme, (b) a multi-axial coding scheme defines each position of a character within a medical code, and (c) each position of a character corresponds to a semi-independent axis of classification of a procedure, a diagnosis, or both;

filter, via a Procedure Coding System (PCS) filter module, the one or more candidate semantic objects to select a first semantic object based at least in part on medical code filtering criteria relevant to generate a medical code;

determine a first character to insert into a first position of the medical code based at least in part on the first semantic object and the multi-axial coding scheme;

infer a second semantic object of the plurality of semantic objects based at least in part on the first semantic object selected via the PCS filter module, wherein the second semantic object is not mapped to the one or more of the sequenced words based at least in part on the segmented narrative text;

determine a second character to insert into a second position of the medical code based at least in part on the second semantic object and the multi-axial coding scheme;

insert the first character into the first position of the medical code and the second character into the second position of the medical code; and

output the medical code comprising at least the first character and the second character.

11. The computer system of claim 10 , wherein to output the medical code comprising at least the first character and the second character comprises outputting the medical code to a user interface.

12. The computer system of claim 10 , wherein (a) the first position comprising the first character corresponds to a first semi-independent axis of classification of a medical procedure or diagnosis, and (b) the second position comprising the second character corresponds to a second semi-independent axis of classification of the medical procedure or diagnosis.

13. The computer system of claim 10 , wherein the system is further configured to generate first trace data that comprises linking information to a location of the narrative text of the clinical documentation from which the first character was derived, wherein outputting the medical code further comprises providing the first trace data.

14. The computer system of claim 13 , wherein the system is further configured to receive the clinical documentation from a memory.

15. The computer system of claim 10 , wherein the system is further configured to insert a wildcard character in an additional slot position of the medical code in response to a determination that a medical concept necessary to derive a character to be positioned in the additional slot is missing from the narrative text.

16. The computer system of claim 10 , wherein the classes of the plurality of semantic objects define a hierarchical structure of the ontology.

17. A non-transitory computer-readable storage medium having program code embedded thereon, the program code executable on a processor of a computer system to:

segment, via a natural language processing engine, unstructured narrative text of clinical documentation having boundaries between sections of narrative text within the clinical documentation, based at least in part on the boundaries defined in the clinical documentation to generate segmented narrative text;

sequence, via the natural language processing engine, words of the narrative text selected at least in part by a document filter filtering the unstructured narrative text based at least in part on the segmented narrative text to generate sequenced words;

map one or more of the sequenced words to one or more candidate semantic objects of a plurality of semantic objects of an ontology, wherein (a) the ontology defines classes of the plurality of semantic objects corresponding to a multi-axial coding scheme, (b) a multi-axial coding scheme defines each position of a character within a medical code, and (c) each position of a character corresponds to a semi-independent axis of classification of a procedure, a diagnosis, or both;

filter, via a Procedure Coding System (PCS) filter module, the one or more candidate semantic objects to select a first semantic object based at least in part on medical code filtering criteria relevant to generate a medical code;

determine a first character to insert into a first position of the medical code based at least in part on the first semantic object and the multi-axial coding scheme;

infer a second semantic object of the plurality of semantic objects based at least in part on the first semantic object selected via the PCS filter module, wherein the second semantic object is not mapped to the one or more of the sequenced words based at least in part on the segmented narrative text;

determine a second character to insert into a second position of the medical code based at least in part on the second semantic object and the multi-axial coding scheme;

insert the first character into the first position of the medical code and the second character into the second position of the medical code; and

output the medical code comprising at least the first character and the second character.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the program code is further executable on the processor of the computer system to position insert a wildcard character in an additional slot position of the medical code in response to a determination that a medical concept necessary to derive a character to be positioned in the additional slot is missing from the narrative text.

19. The non-transitory computer-readable storage medium of claim 18 , wherein the program code is further executable on the processor of the computer system to generate first trace data that comprises linking information to a location of the narrative text of the clinical documentation from which the first character was derived, wherein outputting the medical code further comprises providing the first trace data.

20. The non-transitory computer-readable storage medium of claim 17 , wherein (a) the first position comprising the first character corresponds to a first semi-independent axis of classification of a medical procedure or diagnosis, and (b) the second position comprising the second character corresponds to a second semi-independent axis of classification of the medical procedure or diagnosis.