Patient data mining improvements

• US 20080275731A1
• Filed: 05/17/2006
• Published: 11/06/2008
• Est. Priority Date: 05/18/2005
• Status: Abandoned Application

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Krishnan, Sriram, Rao, R. Bharat, Landi, William A.
• Application Number:
  US11/435,657
• Publication Number:
  US 20080275731A1
• Time in Patent Office: Days
• US Class Current: 705/3
• 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: G16H 30/20 for handling medical images, e.g. DICOM, HL7 or PACS
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Improvements in mining information from patient records and/or use of such mined information are provided. The identity of the patient is used to link to patient records at different institutions for mining. The user controls one or more thresholds for mining and/or inferring. By providing a user interface that allows selection of a portion of the statistical summary, data supporting the statistics may be output. To assist in understanding the knowledge base used for mining or inferring, a visual representation is output. The mining may be used for diagnosis related groupings.

Claims

1. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for adherence to a clinical guideline, assessment for clinical trial and/or assessment for treatment, the storage medium comprising instructions for:

receiving input identifying a patient;

linking the patient to a first patient record at a first institution and a second patient record as a second institution different than the first institution, the linking being as a function of the input;

mining the first and second patient records; and

determining a probability of a particular disease as a function of results of the mining of the first and second patient records.

2. The instructions of claim 1 wherein mining comprises mining structured and unstructured data in at least the first patient record.

3. The instructions of claim 1 wherein the first institution is unrelated by ownership to the second institution.

4. The instructions of claim 1 wherein mining comprises extracting data from the first and second patient records as a function of domain knowledge;

and

wherein determining the probability comprises assigning probabilistic assertions to the extracted data, and combining the probabilistic assertions.

5. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for adherence to a clinical guideline, assessment for clinical trial and/or assessment for treatment, the storage medium comprising instructions for:

mining a patient record as a function of domain knowledge;

inferring a patient state from outputs of the mining; and

receiving user input of at least a first threshold;

wherein mining, inferring or mining and inferring are performed as a function of the first threshold.

6. The instructions of claim 5 wherein mining comprises extracting data from the patient record as a function the domain knowledge;

and

wherein inferring comprises assigning probabilistic assertions to the extracted data, and combining the probabilistic assertions.

7. The instructions of claim 5 wherein mining comprises mining from both structured and unstructured information.

8. The instructions of claim 5 wherein mining comprises mining for elements of the patient record as a function of domain knowledge, the domain knowledge indicating elements with a probability greater than the first threshold of indicating the patient state.

9. The instructions of claim 5 wherein inferring comprises inferring from elements with a probability greater than the first threshold of indicating the patient state.

10. The instructions of claim 5 wherein the first threshold corresponds to an upper limit of normal for the patient state.

11. The instructions of claim 5 wherein the first threshold corresponds to a definition of information used in the domain knowledge.

12. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for adherence to a clinical guideline, assessment for clinical trial and/or assessment for treatment, the storage medium comprising instructions for:

outputting a statistical summary of clinical information for a plurality of patients;

receiving a selection of a portion of the statistical summary; and

outputting data supporting the portion of the statistical summary.

13. The instructions of claim 12 wherein outputting the statistical summary comprises outputting a pie chart, graph, or combinations thereof, and wherein receiving the selection comprises receiving an indication of a pie chart wedge, a location along an axis or combinations thereof.

14. The instructions of claim 12 wherein outputting the data comprises outputting a table listing the patients of the plurality associated with the selected portion.

15. The instructions of claim 14 further comprising:

receiving a patient selection from the table; and

outputting patient record information for the patient.

16. The instructions of claim 14 further comprising:

mining patient records including unstructured information; and

inferring patient states as a function of the mining;

wherein the statistical summary is responsive to the inferring.

17. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for adherence to a clinical guideline, assessment for clinical trial and/or assessment for treatment, the storage medium comprising instructions for:

mining a patient record as a function of first domain knowledge;

inferring, as a function of second domain knowledge, a patient state from outputs of the mining; and

outputting a visual representation of a relationship of the patient state to the patient record.

18. The instructions of claim 17 wherein outputting comprises outputting elements of the patient record used to infer the patient state as linked to the patient state.

19. The instructions of claim 17 wherein outputting comprises outputting a flow chart graph representing the relationship.

20. The instructions of claim 17 wherein the visual representation is different for two different users.

21. The instructions of claim 17 wherein mining comprises mining, at least in part, from unstructured data of the patient record.

22. The instructions of claim 17 wherein inferring comprises assigning probabilistic assertions to mined elements of the patient record, and combining the probabilistic assertions.

23. In a computer readable storage medium having stored therein data representing instructions executable by a programmed processor for billing for medical treatment, the storage medium comprising instructions for:

mining at least unstructured data of a patient record, the mining being a function of domain knowledge; and

determining a diagnosis related grouping as a function of results of the mining.

24. The instructions of claim 23 wherein mining as a function of the first domain knowledge comprises mining for one or more elements comprising diagnosis codes, co-morbidities, surgical procedures, age, sex, discharge status, or combinations thereof;

and

wherein determining comprises determining as a function of the diagnosis codes, co-morbidities, surgical procedures, age, sex, discharge status, or combinations thereof.

25. The instructions of claim 24 wherein mining comprises mining for at least three of the elements from the list of:

diagnosis codes, co-morbidities, surgical procedures, age, sex, and discharge status.

26. The instructions of claim 24 wherein mining comprises inferring at least one of the elements from other data;

and

wherein determining the diagnosis related grouping comprises determining a probability of the diagnosis related grouping.

27. The instructions of claim 23 further comprising:

comparing the diagnosis related grouping to a previously assigned diagnosis related grouping.

28. The instructions of claim 23 wherein mining comprises identifying a secondary diagnosis, and determining a co-morbidity as a function of the secondary diagnosis;

and

wherein determining the diagnosis related grouping comprises determining as a function of the co-morbidity.