Hybrid Query System for Electronic Medical Records

• US 20120066197A1
• Filed: 09/13/2011
• Published: 03/15/2012
• Est. Priority Date: 09/15/2010
• Status: Active Grant

• Current Assignee: Epic Systems Corporation
• Original Assignee: Epic Systems Corporation
• Inventors: Rana, Sumit, Bhattacharya, Sagnik, Kappes, Steven Ross
• Granted Patent: US 9,147,039 B2
• Time in Patent Office: Days
• US Class Current: 707/706
• CPC Class Codes: G06F 16/2272 Management thereof
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records

Abstract

An electronic medical record employs a hybrid search engine which can perform structured queries for comprehensive search results and text searches for rapid identification of relevant information. A standard structured database of medical data is rendered accessible to text searches by generating pseudo-documents from the database that can then be subjected to standard information retrieval pre-processing.

Claims

1. A hybrid electronic medical record system comprising:

a structured database providing a plurality of records associated with particular patients and providing clinically derived medical data separated in different identified data fields;

a database management engine communicating with the structured database and responding to structured queries to provide predefined reports collecting and organizing information from different fields of selected records according to the query;

a document generator generating documents based on records associated with particular patients;

an inverted index indexing the documents according to text terms in the documents; and

a text search engine communicating with the inverted index and responding to text queries to identify multiple ranked documents based on the indexing of the inverted index.

2. The hybrid electronic medical record system of claim 1 wherein the document generator further responds to a user selection of a ranked document to provide an associated predefined report via the database management engine.

3. The hybrid electronic medical record system of claim 2 wherein the associated predefined report includes text entry fields and the document generator response to data entry in the text entry fields by changing the medical data in the structured database.

4. The hybrid electronic medical record system of claim 2 wherein the associated predefined report includes user-activatable links to structured queries to provide different predefined reports collecting and organizing information from different fields of selected records.

5. The hybrid electronic medical record system of claim 1 wherein the database management engine provides a user interface receiving user commands to execute predefined structured queries with the database management engine to generate predefined reports and wherein the document generator generates documents that duplicate the predefined reports.

6. The hybrid electronic medical record system of claim 5 wherein the document generator generates documents based both on dynamic data from the structured database contained in the predefined reports and on static text data contained in the predefined reports.

7. The hybrid electronic medical record system of claim 6 wherein the static text data is associated with text descriptions of the fields of the structured database from which the dynamic data of the predefined reports is obtained.

8. The hybrid electronic medical record system of claim 5 wherein the document generator generates documents including chronological information describing a date of clinical observations of data of the predefined report in the structured database at the time of generation of the documents.

9. The hybrid electronic medical record system of claim 5 wherein the document generator generates documents including data from the structured database not in the predefined report but related to data used in the predefined report by structure of the database.

10. The hybrid electronic medical record system of claim 1 wherein further including a data entry interface allowing highlighting of data in the structured database as particularly significant for an information retrieval search and wherein the document generator ranks documents at least in part according to whether the data of a documents is highlighted.

11. The hybrid electronic medical record system of claim 1 wherein the text search engine provides a ranking based on term frequency of the query text.

12. The hybrid electronic medical record system of claim 1 wherein the search engine provides a ranking based at least in part on chronology information associated with the text terms.

13. The hybrid electronic medical record system of claim 1 wherein the search engine further includes a query expander processing medical words from the text query to add additional medical words corresponding to the medical words of the query.

14. The hybrid electronic medical record system of claim 13 wherein the additional medical words are synonyms.

15. The hybrid electronic medical record system of claim 13 wherein the additional medical words are words related by co-morbidity

16. The hybrid electronic medical record system of claim 13 wherein the additional medical words are words related as co-symptoms.

17. The hybrid electronic medical record system of claim 1 wherein further including a heuristic monitor monitoring selection of ranked documents to provide a ranking based at least in part on historical selection of particular report types.

18. The hybrid electronic medical record system of claim 1 wherein the ranked documents are represented by abstracts of the ranked documents generated at a time of a text query using a predefined abstract report collecting predefined information from the structured database