Methods and apparatus for identifying unspecified diagnoses in clinical documentation

• US 10,460,288 B2
• Filed: 02/18/2011
• Issued: 10/29/2019
• Est. Priority Date: 02/18/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Flanagan, James R., Montyne, Frank, Decraene, David, Casella dos Santos, Mariana, Holvoet, Marnix, Doyle, Karen Anne, Manoukian, Roubik, Elcocks, Anthony J., Shelton, Matthew R., Abrutyn, Scott, Habboush, Isam
• Primary Examiner(s): Nguyen, Tran N
• Application Number:
  US13/030,969
• Publication Number:
  US 20120215551A1
• Time in Patent Office: 3,175 Days
• Field of Search: 705 2
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A set of one or more clinical facts may be collected from a clinician'"'"'s encounter with a patient. It may be determined that an unspecified diagnosis not included in the set of facts may possibly be ascertained from the patient encounter. A user may be alerted that the unspecified diagnosis may possibly be ascertained from the patient encounter.

Claims

1. A method comprising:

processing documentation of a clinician'"'"'s encounter with a patient and/or the patient'"'"'s history to produce a set of one or more clinical facts, a first fact of the set of one or more clinical facts identifying a first diagnosis or a procedure applying to the patient, wherein the processing of the documentation comprises extracting at least one fact of the set of one or more clinical facts from a free-form narration documenting the clinician'"'"'s encounter with the patient, by processing at least a portion of the free-form narration using natural language understanding;

determining, using at least one processor, whether the set of one or more clinical facts that was produced by the processing of the documentation and that identifies the first diagnosis or procedure indicates one or more additional clinical facts for a second diagnosis that may apply to the patient, is different from the first diagnosis, and is not specified in the documentation, wherein the determining comprises;

processing at least a portion of the set of one or more clinical facts, the portion including the first fact, using the at least one processor and a statistical model trained on past clinical reports and/or medical literature and that associates co-occurrence of clinical facts with complications or comorbidities to identify whether the first fact co-occurs in the set with at least one other fact; and

in response to identifying that the first fact co-occurs in the set with the at least one other fact and that the statistical model associates co-occurrence of the first fact and the at least one other fact with at least the second diagnosis, determining at least one value indicative of a likelihood that the second diagnosis applies to the patient as a complication or comorbidity of the first diagnosis or the procedure; and

in response to determining that the likelihood exceeds a threshold, alerting a user that the second diagnosis may possibly apply to the patient as a complication or comorbidity of the first diagnosis or the procedure.

2. The method of claim 1, wherein the processing of the documentation to produce the set of one or more clinical facts comprises determining at least one of the one or more clinical facts from a discrete structured data item input by the user that includes the at least one of the one or more clinical facts.

3. The method of claim 1, wherein the second diagnosis is not included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly apply to the patient as a comorbidity of the first diagnosis.

4. The method of claim 1, wherein the second diagnosis is not included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly apply to the patient as a complication of the first diagnosis or the procedure.

5. The method of claim 1, wherein the processing of the at least the portion of the set of one or more clinical facts comprises applying the statistical model to associate the co-occurrence of the first fact and the at least one other fact with the second diagnosis.

6. The method of claim 1, wherein the determining whether the set of one or more clinical facts indicates one or more additional clinical facts comprises querying the patient'"'"'s history previous to the patient encounter for information that increases or decreases the likelihood of the second diagnosis applying to the patient.

7. The method of claim 1, wherein the alerting comprises displaying a visual alert to the user.

8. The method of claim 1, wherein the alerting comprises providing an audio alert to the user.

9. The method of claim 1, wherein the alerting comprises presenting one or more options corresponding to one or more diagnoses not included in the set of one or more clinical facts, and detecting a selection by the user of an option from among the one or more options.

10. The method of claim 9, further comprising storing the diagnosis corresponding to the option chosen by the user as a discrete structured data item.

11. The method of claim 9, further comprising editing a data representation of the free-form narration of the patient encounter in accordance with the option chosen by the user.

12. The method of claim 1, wherein the user is the clinician.

13. The method of claim 1, wherein the user is a person other than the clinician, wherein the alerting comprises prompting the user to provide information corresponding to the second diagnosis, and wherein the method further comprises prompting the clinician to approve the information provided by the user.

14. The method of claim 1, further comprising simultaneously displaying to the user the at least one fact and a textual representation of the portion of the free-form narration from which the at least one fact was extracted.

15. The method of claim 1, wherein the alerting is performed before the clinician finally approves the set of one or more clinical facts.

16. The method of claim 1, wherein the determining and the alerting are performed automatically.

17. An apparatus comprising:

at least one processor; and

a memory storing processor-executable instructions that, when executed by the at least one processor, perform a method comprising;

processing documentation of a clinician'"'"'s encounter with a patient and/or the patient'"'"'s history to produce a set of one or more clinical facts, wherein the processing of the documentation comprises extracting at least one fact of the set of one or more clinical facts from a free-form narration documenting the clinician'"'"'s encounter with the patient, by processing at least a portion of the free-form narration using natural language understanding, wherein a first fact of the set of one or more clinical facts identifies a first diagnosis or a procedure applying to the patient;

determining whether the set of one or more clinical facts produced by the processing of the documentation indicates one or more additional clinical facts that may apply to the patient, wherein the determining comprises processing, using a statistical model trained on past clinical reports and/or medical literature and that associates co-occurrence of clinical facts with complications or comorbidities, at least a portion of the set of one or more clinical facts produced from the processing of the documentation to identify whether the first fact co-occurs in the set with at least one other fact, wherein the statistical model associates co-occurrence of the first fact and the at least one other fact with at least a second diagnosis different from the first diagnosis, and wherein processing the at least the portion of the set of one or more clinical facts using the statistical model to identify whether the first fact co-occurs in the set with the at least one other fact comprises determining, based on whether the first fact co-occurs with one or more of the at least one other fact, at least one value indicative of a likelihood that the second diagnosis applies to the patient as a complication or comorbidity of the first diagnosis or the procedure; and

in response to the likelihood exceeding a threshold, alerting a user that the second diagnosis may possibly apply to the patient as a complication or comorbidity of the first diagnosis or the procedure.

18. The apparatus of claim 17, wherein the processing of the documentation to produce the set of one or more clinical facts comprises determining at least one of the one or more clinical facts from a discrete structured data item input by the user that includes the at least one of the one or more clinical facts.

19. The apparatus of claim 17, wherein the second diagnosis is not included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly apply to the patient as a comorbidity of the first diagnosis.

20. The apparatus of claim 17, wherein the second diagnosis is not included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly apply to the patient as a complication of the first diagnosis or the procedure.

21. The apparatus of claim 17, wherein the second diagnosis is included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly be a complication of the first diagnosis or the procedure.

22. The apparatus of claim 17, wherein the processing of the at least the portion of the set of one or more clinical facts comprises applying the statistical model to associate the co-occurrence of the first fact and the at least one other fact with the second diagnosis.

23. The apparatus of claim 17, wherein the determining whether the set of one or more clinical facts indicates one or more additional clinical facts comprises querying the patient'"'"'s history previous to the patient encounter for information that increases or decreases the likelihood of the second diagnosis applying to the patient.

24. The apparatus of claim 17, wherein the alerting comprises displaying a visual alert to the user.

25. The apparatus of claim 17, wherein the alerting comprises providing an audio alert to the user.

26. The apparatus of claim 17, wherein the alerting comprises presenting one or more options corresponding to one or more diagnoses not included in the set of one or more clinical facts, and detecting a selection by the user of an option from among the one or more options.

27. The apparatus of claim 26, wherein the method further comprises storing the diagnosis corresponding to the option chosen by the user as a discrete structured data item.

28. The apparatus of claim 26, wherein the method further comprises editing a data representation of the free-form narration of the patient encounter in accordance with the option chosen by the user.

29. The apparatus of claim 17, wherein the user is the clinician.

30. The apparatus of claim 17, wherein the user is a person other than the clinician, wherein the alerting comprises prompting the user to provide information corresponding to the second diagnosis, and wherein the method further comprises prompting the clinician to approve the information provided by the user.

31. The apparatus of claim 17, wherein the method further comprises simultaneously displaying to the user the at least one fact and a textual representation of the portion of the free-form narration from which the at least one fact was extracted.

32. The apparatus of claim 17, wherein the alerting is performed before the clinician finally approves the set of one or more clinical facts.

33. The apparatus of claim 17, wherein the determining and the alerting are performed automatically.

34. At least one non-transitory computer-readable storage medium encoded with a plurality of computer-executable instructions that, when executed, perform a method comprising:

processing documentation of a clinician'"'"'s encounter with a patient and/or the patient'"'"'s history to produce a set of one or more clinical facts, wherein the processing of the documentation comprises extracting at least one fact of the set of one or more clinical facts from a free-form narration documenting the clinician'"'"'s encounter with the patient, by processing at least a portion of the free-form narration using natural language understanding, wherein a first fact of the set of one or more clinical facts identifies a first diagnosis or a procedure applying to the patient;

determining whether the set of one or more clinical facts produced by the processing of the documentation indicates one or more additional clinical facts that may apply to the patient, wherein the determining comprises processing, using a statistical model trained on past clinical reports and/or medical literature and that associates co-occurrence of clinical facts with complications or comorbidities, at least a portion of the set of one or more clinical facts produced from the processing of the documentation to identify whether the first fact co-occurs in the set with at least one other fact, wherein the statistical model associates co-occurrence of the first fact and the at least one other fact with at least a second diagnosis different from the first diagnosis, and wherein processing the at least the portion of the set of one or more clinical facts using the statistical model to identify whether the first fact co-occurs in the set with the at least one other fact comprises determining, based on whether the first fact co-occurs with one or more of the at least one other fact, at least one value indicative of a likelihood that the second diagnosis applies to the patient as a complication or comorbidity of the first diagnosis or the procedure; and

in response to the likelihood exceeding a threshold, alerting a user that the second diagnosis may possibly apply to the patient as a complication or comorbidity of the first diagnosis or the procedure.

35. The at least one non-transitory computer-readable storage medium of claim 34, wherein the processing of the documentation to produce the set of one or more clinical facts comprises determining at least one of the one or more clinical facts from a discrete structured data item input by the user that includes the at least one of the one or more clinical facts.

36. The at least one non-transitory computer-readable storage medium of claim 34, wherein the second diagnosis is not included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly apply to the patient as a comorbidity of the first diagnosis.

37. The at least one non-transitory computer-readable storage medium of claim 34, wherein the second diagnosis is not included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly apply to the patient as a complication of the first diagnosis or the procedure.

38. The at least one non-transitory computer-readable storage medium of claim 34, wherein the second diagnosis is included in the set of one or more clinical facts, and wherein the alerting comprises alerting the user that the second diagnosis may possibly be a complication of the first diagnosis or the procedure.

39. The at least one non-transitory computer-readable storage medium of claim 34, wherein the processing of the at least the portion of the set of one or more clinical facts comprises applying the statistical model to associate the co-occurrence of the first fact and the at least one other fact with the second diagnosis.

40. The at least one non-transitory computer-readable storage medium of claim 34, wherein the determining whether the set of one or more clinical facts indicates one or more additional clinical facts comprises querying the patient'"'"'s history previous to the patient encounter for information that increases or decreases the likelihood of the second diagnosis applying to the patient.

41. The at least one non-transitory computer-readable storage medium of claim 34, wherein the alerting comprises displaying a visual alert to the user.

42. The at least one non-transitory computer-readable storage medium of claim 34, wherein the alerting comprises providing an audio alert to the user.

43. The at least one non-transitory computer-readable storage medium of claim 34, wherein the alerting comprises presenting one or more options corresponding to one or more diagnoses not included in the set of one or more clinical facts, and detecting a selection by the user of an option from among the one or more options.

44. The at least one non-transitory computer-readable storage medium of claim 43, wherein the method further comprises storing the diagnosis corresponding to the option chosen by the user as a discrete structured data item.

45. The at least one non-transitory computer-readable storage medium of claim 43, wherein the method further comprises editing a data representation of the free-form narration of the patient encounter in accordance with the option chosen by the user.

46. The at least one non-transitory computer-readable storage medium of claim 34, wherein the user is the clinician.

47. The at least one non-transitory computer-readable storage medium of claim 34, wherein the user is a person other than the clinician, wherein the alerting comprises prompting the user to provide information corresponding to the second diagnosis, and wherein the method further comprises prompting the clinician to approve the information provided by the user.

48. The at least one non-transitory computer-readable storage medium of claim 34, wherein the method further comprises simultaneously displaying to the user the at least one fact and a textual representation of the portion of the free-form narration from which the at least one fact was extracted.

49. The at least one non-transitory computer-readable storage medium of claim 34, wherein the alerting is performed before the clinician finally approves the set of one or more clinical facts.

50. The at least one non-transitory computer-readable storage medium of claim 34, wherein the determining and the alerting are performed automatically.