System and method for multidimensional extension of database information using inferred groupings

• US 7,865,375 B2
• Filed: 01/03/2006
• Issued: 01/04/2011
• Est. Priority Date: 08/28/2003
• Status: Active Grant

• Current Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Original Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Inventors: McNair, Douglas S., Gragg, John H., Lancaster, Brian J., Parkins, Kent D., Yarbrough, Michael E.
• Primary Examiner(s): O'Connor; Gerald J.
• Assistant Examiner(s): Altschul; Amber
• Application Number:
  US11/325,004
• Publication Number:
  US 20070005154A1
• Time in Patent Office: 1,827 Days
• Field of Search: 705/2, 705/3, 705/4
• US Class Current: 705/3
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A system and method for receiving medical or other database information and pregrouping and extending that data include a data enhancement layer configured to generate additional stored dimensions capturing the data and relevant attributes. Data sources such as hospitals, laboratories and others may therefore communicate their clinical data to a central warehousing facility which may assemble and extend the resulting aggregated data for data mining purposes. Varying source format and content may be conditioned and conformed to a consistent physical or logical structure. The source data may be extended and recombined into additional related dimensions, pre-associating meaningful attributes for faster querying and storage. The attributes, data and other pieces of information may likewise in embodiments be subjected to an inference analysis to determine whether previously unidentified or unexploited relationships may exist within the universe of source data, for instance using correlation, inference or other analytic techniques. Newly detected, identified or inferred data groupings, which may for instance reveal hidden trends or patterns residing in the data, may then be added back to the enhanced data groupings. Users running analytics against the resulting medical or other datamarts may therefore access a richer set of related information, more powerful sets of predictive models as well as have their queries and other operations run more efficiently.

Claims

1. A system for generating an inferred data grouping, comprising:

an input interface receiving clinically relevant source data from a first data source and clinically relevant source data from a second data source, wherein each of the first data source and the second data source comprises a healthcare provider, a hospital, an insurer, a laboratory, or a medical facility;

an inference engine, the inference engine

communicating with the input interface to access the clinically relevant source data,

automatically detecting an inferred clinical relationship between at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data in the clinically relevant source data from the second data source being separate from the first data source comprising the healthcare provider, the hospital, the insurer, the laboratory, or the medical facility, wherein the inferred clinical relationship is automatically detected using an inference measure to detect and substantiate the inferred clinical relationship,

generating an inferred data grouping in accordance with the detected inferred clinical relationship, and

combining the inferred data grouping with an existing clinical data grouping to generate an inferentially enhanced data grouping that is used to analyze past healthcare-related trends and model future healthcare-related events to project a healthcare-related cost, outcome, trend, or interest; and

a query engine, the query engine

receiving a user query,

in response to receiving the user query, utilizing the inferentially enhanced data grouping to analyze one or more healthcare-related trends, and

providing a user with output based on the one or more healthcare-related trends.

2. A system according to claim 1, wherein the clinically relevant source data from the first data source comprises data generated by at least one of a health care provider, a hospital, an insurer and a laboratory.

3. A system according to claim 1, wherein the inference measure comprises a correlation measure between the at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data in the clinically relevant source data from the second data source.

4. A system according to claim 1, wherein the inferred data grouping comprises a dimensional or multidimensional data grouping.

5. A system according to claim 4, wherein the dimensional or multidimensional data grouping is combined with the existing clinical data grouping to generate an inferentially enhanced dimensional or multidimensional data grouping.

6. A system according to claim 1, wherein the inference engine stores the inferred data grouping to a datamart.

7. A system according to claim 6, wherein the datamart is configured to receive queries.

8. A system according to claim 7, wherein the queries comprise standard query language queries.

9. A system according to claim 7, wherein the queries comprise at least one of analytic and predictive processing applied to the inferred data groupings.

10. A system according to claim 9, wherein the inferred data groupings are associated with clinical trial activity.

11. A computer-implemented method for generating an inferred data grouping, comprising:

receiving clinically relevant source data from a first data source and clinically relevant source data from a second data source, wherein each of the first data source and the second data source comprises a health care provider, a hospital, an insurer, a laboratory, or a medical facility;

automatically detecting an inferred clinical relationship between at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data in the clinically relevant source data from the second data source being separate from the first data source comprising the healthcare provider, the hospital, the insurer, the laboratory, or the medical facility, wherein the inferred clinical relationship is automatically detected using an inference measure to detect and substantiate the inferred clinical relationship;

generating, via a data enhancement layer, an inferred data grouping in accordance with the detected inferred clinical relationship;

combining the inferred data grouping with an existing clinical data grouping to generate an inferentially enhanced data grouping that is used to analyze past healthcare-related trends and model future healthcare-related events to project a healthcare-related cost, outcome, trend, or interest;

in response to receiving a user query, utilizing the inferentially enhanced data grouping to analyze one or more healthcare-related trends; and

providing a user with output based on the one or more healthcare-related trends.

12. A method according to claim 11, wherein the clinically relevant source data from the first data source comprises data generated by at least one of a health care provider, a hospital, an insurer and a laboratory.

13. A method according to claim 11, wherein the inference measure comprises a correlation measure between the at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data in the clinically relevant source data from the second data source.

14. A method according to claim 11, wherein the inferred data grouping comprises a dimensional or multidimensional data grouping.

15. A method according to claim 14, wherein the dimensional or multidimensional data grouping is combined with the existing clinical data grouping to generate an inferentially enhanced dimensional or multidimensional data grouping.

16. A method according to claim 11, further comprising storing the inferred data grouping to a datamart.

17. A method according to claim 16, wherein the datamart is configured to receive queries.

18. A method according to claim 17, wherein the queries comprise standard query language queries.

19. A method according to claim 17, further comprising applying at least one of analytic and predictive processing to the inferred data groupings based on the queries.

20. A method according to claim 19, wherein the inferred data groupings are associated with clinical trial activity.

21. A method according to claim 17, wherein the queries comprise standard query language queries.

22. A method according to claim 17, wherein the method further comprises applying at least one of analytic and predictive processing to the inferred data groupings based on the queries.

23. A method according to claim 22, wherein the inferred data groupings are associated with clinical trial activity.

24. A method for generating an inferred data grouping, the method comprising:

receiving clinically relevant source data from a first data source and clinically relevant source data from a second data source, wherein each of the first data source and the second data source comprises a health care provider, a hospital, an insurer, a laboratory, or a medical facility;

automatically detecting an inferred clinical relationship between at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data in the clinically relevant source data from the second data source being separate from the first data source comprising the healthcare provider, the hospital, the insurer, the laboratory, or the medical facility, wherein the inferred clinical relationship is automatically detected using an inference measure to detect and substantiate the inferred clinical relationship;

generating, via a data enhancement layer, an inferred data grouping in accordance with the detected inferred clinical relationship;

combining the inferred data grouping with an existing clinical data grouping to generate an inferentially enhanced data grouping that is used to analyze past healthcare-related trends and model future healthcare-related events to project a healthcare-related cost, outcome, trend, or interest;

in response to receiving a user query, utilizing the inferentially enhanced data grouping to analyze one or more healthcare-related trends; and

providing a user with output based on the one or more healthcare-related trends.

25. A method according to claim 24, wherein the clinically relevant source data from the first data source comprises data generated by at least one of a health care provider, a hospital, an insurer and a laboratory.

26. A method according to claim 24, wherein the inference measure comprises a correlation measure between the at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data from the second data source.

27. A method according to claim 24, wherein the inferred data grouping comprises a dimensional or multidimensional data grouping.

28. A method according to claim 27, wherein the dimensional or multidimensional data grouping is combined with the existing clinical data grouping to generate an inferentially enhanced dimensional or multidimensional data grouping.

29. A method according to claim 24, wherein the method further comprises storing the inferred data grouping to a datamart.

30. A method according to claim 29, wherein the datamart is configured to receive queries.

31. A method for generating an output based on an inferred data grouping, the method comprising:

receiving clinically relevant source data from a first data source and clinically relevant source data from a second data source, wherein each of the first data source and the second data source comprises a health care provider, a hospital, an insurer, a laboratory, or a medical facility;

automatically detecting an inferred clinical relationship between at least one attribute or data in the clinically relevant source data from the first data source and at least one attribute or data in the clinically relevant source data from the second data source being separate from the first data source comprising the healthcare provider, the hospital, the insurer, the laboratory, or the medical facility, wherein the inferred clinical relationship is automatically detected using an inference measure to detect and substantiate the inferred clinical relationship, the inference measure comprising a statistical correlation measure between the at least one attribute or data in the clinically relevant source data from the first data source and the at least one attribute or data in the clinically relevant source data from the second data source generating, via a data enhancement layer, an inferred data grouping in accordance with the detected inferred clinical relationship, the inferred data grouping comprising a hierarchical or multidimensional data grouping;

combining the inferred data grouping with an existing clinical data grouping to generate an inferentially enhanced hierarchical or multidimensional data grouping, wherein the inferentially enhanced data grouping is used to analyze past healthcare-related trends and model future healthcare-related events, wherein one or more users can project a healthcare-related cost, outcome, trend, or interest;

in response to receiving a user query, utilizing the inferentially enhanced data grouping to model future healthcare-related events pertaining to disease, drug efficacy, therapeutic, or demographic trends; and

providing a user with output based on the modeling of future healthcare-related events.

32. A method according to claim 31, wherein the clinically relevant source data from the first data source comprises data generated by at least one of a health care provider, a hospital, an insurer and a laboratory.