Method for Automatic Labeling of Unstructured Data Fragments From Electronic Medical Records

• US 20090299977A1
• Filed: 05/21/2009
• Published: 12/03/2009
• Est. Priority Date: 05/28/2008
• Status: Abandoned Application

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Rosales, Romer E.
• Application Number:
  US12/469,745
• Publication Number:
  US 20090299977A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24573 using data annotations, e.g. user-defined metadata
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A method for automatically labeling unstructured data from electronic medical records using a computer-based medical data processing system includes selecting a data pattern based on a desired medical finding. The selected data pattern is searched for within source data including patient records to find one or more matches. A context of a predetermined range around each data pattern match found is identified within the source data and the found contexts are associated with a particular medical finding. The medical finding can be at the patient level or document level, not necessarily at the context level. Associations between contexts and medical findings are identified. A classifier based on an association between the identified contexts and the desired medical finding is trained. The trained classifier is used to automatically identify likely instances of passages, documents or patients related to the desired medical finding from within subsequent data including patient records.

Claims

1. A method for automatically labeling unstructured data from electronic medical records using a computer-based medical data processing system, comprising:

selecting a data pattern based on a desired medical finding;

searching for the selected data pattern within source data including patient records;

identifying a context of a predetermined range around each data pattern match found within the source data;

training a classifier based on an association between the identified contexts and the desired medical finding; and

using the trained classifier to automatically identify likely instances of the desired medical finding from within subsequent data including patient records.

2. The method of claim 1, wherein the data pattern is selected from one or more words or regular expressions relating to the desired medical finding.

3. The method of claim 1, wherein the one or more words or regular expressions are selected from a description of the desired medical finding.

4. The method of claim 1, wherein the predetermined range is a fixed number of words or characters preceding and following the data pattern.

5. The method of claim 1, wherein the source data includes a medical image and the data pattern is a particular image filter, shape, or other image appearance.

6. The method of claim 5, wherein the predetermined range is a surrounding area or volume of a fixed perimeter about the data pattern.

7. The method of claim 1, wherein the desired medical finding is a diagnosis, condition, symptom or other medical concept of interest.

8. The method of claim 1, wherein the source data includes structured data indicating whether or not the desired medical finding is present and the subsequent data does not include structured data indicating whether or not the desired medical finding is present.

9. The method of claim 8, wherein the structured data indicates that the medical finding exists for a particular patient, for a particular document within the electronic medical records of the particular patient, within an image within the electronic medical records of the particular patient, or within a particular context of the electronic medical records of the particular patient.

10. The method of claim 1, wherein the classifier is trained using a machine learning technique.

11. The method of claim 1, additionally including adding to the subsequent data, structured data indicating whether or not the desired medical finding is present.

12. A method for automatically labeling unstructured data from electronic medical records using a computer-based medical data processing system, comprising:

receiving patient medical data that does not include structured data indicating whether or not a desired medical finding is present;

searching for a data pattern indicative of the desired medical finding from within the patient medical data;

identifying a context of a predetermined range around each data pattern match found within the patient medical data; and

using a trained classifier to automatically identify whether the patient medical data has the desired medical finding based on the identified contexts, wherein the trained classifier was generated based on an association between identified contexts and the desired medical finding within training data.

13. The method of claim 12, wherein using a trained classifier to automatically identity whether the patient medical data has the desired medical finding based on the identified contexts includes automatically identifying whether a particular document of the patient medical data has the desired medical findings.

14. The method of claim 12, wherein using a trained classifier to automatically identity whether the patient medical data has the desired medical finding based on the identified contexts includes automatically identifying whether a particular section of text within a particular document of the patient medical data has the desired medical findings.

15. The method of claim 12, wherein the data pattern is selected from one or more words or regular expressions relating to the desired medical finding.

16. The method of claim 15, wherein the one or more words or regular expressions are selected from a description of the desired medical finding.

17. The method of claim 12, wherein the predetermined range is a fixed number of words or characters preceding and following the data pattern.

18. The method of claim 12, wherein the desired medical finding is a diagnosis, condition, symptom or other medical concept of interest.

19. The method of claim 12, wherein the training data includes structured data indicating whether or not the desired medical finding is present and the patient medical data does not include structured data indicating whether or not the desired medical finding is present.

20. The method of claim 12, additionally including adding to the patient medical data, structured data indicating whether or not the desired medical finding is present.

21. A computer system comprising:

a processor; and

a program storage device readable by the computer system, embodying a program of instructions executable by the processor to perform method steps for automatically labeling unstructured data from electronic medical records, the method comprising;

selecting a data pattern based on a desired medical finding;

searching for the selected data pattern within source data including patient records;

identifying a context of a predetermined range around each data pattern match found within the source data;

training a classifier based on an association between the identified contexts and the desired medical finding using a machine learning technique; and

using the trained classifier to automatically identify likely instances of the desired medical finding from within subsequent data including patient records.

22. The computer system of claim 21, wherein the source data includes structured data indicating whether or not the desired medical finding is present and the subsequent data does not include structured data indicating whether or not the desired medical finding is present.

23. The computer system of claim 21, additionally including adding to the subsequent data, structured data indicating whether or not the desired medical finding is present.

24. A method for determining contextual phrases that are indicative of a particular medical finding using a computer-based medical data processing system, comprising:

selecting a data pattern based on a desired medical finding;

searching for the selected data pattern within source data including patient records;

identifying a context of a predetermined range around each data pattern match found within the source data; and

generating a set of associations between the contexts identified around each of the plurality of data pattern matches of the source data and the desired medical finding.