Method and system for automatically evaluating the quality of medical records

Patent metadata

Field	Value
Publication number	US-9798858-B2
Application number	US-43034809-A
Country	US
Kind code	B2
Filing date	Apr 27, 2009
Priority date	Apr 28, 2008
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

A method, a system, and a computer readable article of manufacture tangibly embodying computer readable instructions for executing a computer implemented method for automatically evaluating the quality of a medical record. The method includes: generating temporary facts from a medical record to be evaluated; constructing a query to a knowledge base of standard facts of standard medical records from the temporary facts; executing the query to the knowledge base by using a reasoning engine; and determining the quality of a medical record to be evaluated based on the result of the query.

First claim

What is claimed is:

1. A method for automatically evaluating the quality of a medical record, the method comprising:
providing a knowledge base of standard facts of a standard medical record pattern;
generating temporary facts from a medical record to be evaluated;
constructing a query to said knowledge base from said temporary facts, comprising:
forming a list of the temporary facts and a list of the standard facts;
removing all facts that are present in both lists; and
forming a query using remaining standard facts and negations of remaining temporary facts;

executing said query by a reasoning engine; and
determining the quality of said medical record to be evaluated using a processor to produce an evaluation result based on said query.

2. The method according to claim 1 , wherein generating temporary facts from the medical record to be evaluated comprises:
providing semantic annotations to the medical record to be evaluated; and
generating temporary facts from the medical record to be evaluated having the semantic annotations.

3. The method according to claim 1 , wherein generating temporary facts from the medical record to be evaluated comprises:
transforming the medical record to be evaluated to an instance of the unified medical record model; and
generating the temporary facts from the instance of the unified medical record model.

4. The method according to claim 3 , wherein the instance of the unified medical record model comprises one or more sections, each section containing one or more entries, and each entry containing one or a set of semantically independent statements.

5. The method according to claim 1 , wherein determining the quality of the medical record comprises:
determining whether the medical record to be evaluated is incomplete.

6. The method according to claim 1 , wherein constructing the query to a knowledge base of standard facts of a standard medical record pattern from the temporary facts further comprises:
adding a corresponding target predicate to the query, and
wherein determining the quality of a medical record to be evaluated based on the result of the query comprises:
determining the medical record to be evaluated not specific when the result of the query is not in the temporary facts.

7. The method according to claim 1 , wherein constructing the query to a knowledge base of standard facts of a standard medical record pattern from the temporary facts further comprises:
adding a target predicate indicating the sequence of statements to the query, and
wherein determining the quality of a medical record to be evaluated based on the result of the query comprises:
determining the sequence of statements in the medical record to be evaluated to be incorrect when the result of the query is not empty.

8. The method according to claim 1 , wherein constructing the query to a knowledge base of standard facts of a standard medical record pattern from the temporary facts further comprises:
adding a corresponding target predicate to the query, and
wherein determining the quality of a medical record to be evaluated based on the result of the query comprises:
determining the medical record to be evaluated contains incorrect terms when the result of the query is not empty.

9. The method according to claim 1 , further comprising:
forming the knowledge base of the standard facts.

10. The method according to claim 9 , wherein forming the knowledge base comprises:
obtaining a standard medical record pattern;
generating standard facts from the standard medical record pattern;
storing the standard facts into the knowledge base;
obtaining a universal terminology base regarding medical records;
obtaining standard facts from the universal terminology base regarding the terms in medical records and the relationships thereof; and
storing the standard facts into the knowledge base.

11. The method according to claim 9 , wherein forming the knowledge base of standard facts of standard medical records comprises:
defining medical record quality evaluation rules; and
storing the medical record quality evaluation rules as standard facts into the knowledge base.

12. A system for automatically evaluating the quality of a medical records, comprising:
a fact generator for generating temporary facts from a medical record to be evaluated;
a query constructor and executor for constructing a query to a knowledge base of standard facts of a standard medical record pattern from the temporary facts, configured to form a list of the temporary facts and a list of the standard facts, to remove all facts that are present in both lists, and to form a query using remaining standard facts and negations of remaining temporary facts;
a reasoning engine for executing the query to the knowledge base; and
a processor configured to determine the quality of the medical record to be evaluated based on the result of the query.

13. The system according to claim 12 , further comprising:
a universal terminology base regarding medical records; and
wherein the knowledge base is further for receiving and storing standard facts regarding terms in medical records and the relationships thereof from the universal terminology base.

14. The system according to claim 12 , further comprising:
a medical record semantic annotator for providing semantic annotations to the medical record to be evaluated; and
wherein the fact generator is for generating temporary facts from the medical record to be evaluated having the semantic annotations.

15. The system according to claim 12 , further comprising:
a medical record transformer for transforming the medical record to be evaluated into an instance of the unified medical record model; and
wherein the fact generator is for generating the temporary facts from the instance of the unified medical record model.

16. The system according to claim 15 , wherein the instance of the unified medical record model comprises one or more sections, each section containing one or more entries, and each entry containing one or a set of semantically independent statements.

17. A computer readable article of manufacture tangibly embodying computer readable instructions for executing a computer implemented method for automatically evaluating the quality of a medical record by a computer, the method comprising:
providing a knowledge base of standard facts of a standard medical record pattern;
generating temporary facts from a medical record to be evaluated;
constructing a query to said knowledge base from said temporary facts, comprising:
forming a list of the temporary facts and a list of the standard facts;
removing all facts that are present in both lists; and
forming a query using remaining standard facts and negations of remaining temporary facts;

executing said query by a reasoning engine; and
determining the quality of said medical record to be evaluated to produce an evaluation result based on said query.

Assignees

• Li Bo
• Hou Xue Qiao
• Hsiao Hui-I
• Hu Gang
• Pan Yue
• IBM

Inventors

• Li Bo
• Hou Xue Qiao
• Hsiao Hui-I
• Hu Gang
• Pan Yue

Classifications

• G06F19/324 Primary

  Physics · mapped topic

• G06Q50/22

  Social work or social welfare, e.g. community support activities or counselling services · CPC title

• G06Q50/24

  Physics · mapped topic

• G16H10/60 Primary

  for patient-specific data, e.g. for electronic patient records · CPC title

• G16H70/60

  relating to pathologies · CPC title