System for explanation-based auditing of medical records data
View Patent ↗A system and method is provided for automatically generating explanations for individual records in an access log.
1. A method for identifying an explanation for a user access to records in a records database, the method comprising:
accessing, using a computer system, a template database containing a plurality of explanation templates, each explanation template corresponding to a potential reason for user access to the records database;
accessing, using the computer system, log data stored in an access log database and identifying the user access to the records in the records database;
automatically mining, using the computer system, the plurality of explanation templates to identify candidate explanation templates based on the access log data;
automatically extracting, using the computer system, an explanation corresponding to the user access to the records in the records database, based on the candidate explanation templates; and
storing the extracted explanation.
2. The method of claim 1 , further comprising a user identifying the log data.
3. The method of claim 1 , further comprising mining the plurality of explanation templates to identify a set of frequently used explanation templates as the candidate explanation templates.
4. The method of claim 1 , further comprising mining the plurality of explanation templates based on template type.
5. The method of claim 1 , further comprising mining the plurality of explanation templates based on the lengths of explanation templates in the template database.
6. The method of claim 1 , further comprising mining the plurality of explanation templates based on the number of tables used to construct the log data stored in the access log database.
7. The method of claim 1 , further comprising mining the plurality of explanation templates based on the amount of data in the access log database.
8. The method of claim 1 , further comprising mining the plurality of explanation templates using a one-way algorithm.
9. The method of claim 1 , further comprising mining the plurality of explanation templates using a two-way algorithm.
10. The method of claim 1 , further comprising:
inferring data from missing links between data from the log data; and
automatically mining the plurality of explanation templates to identify the candidate explanation templates based on the log data and the inferred missing data.
11. The method of claim 1 , further comprising:
inferring data from missing links between data from the log data; and
adding the inferred data to the log data.
12. The method of claim 1 , further comprising determining the accuracy extracted explanation corresponding to the user access.
13. The method of claim 1 , wherein accessing log data stored in the access log database comprising collecting and accessing data real time as the user accesses the records.
14. The method of claim 1 , wherein accessing log data stored in the access log database comprising collecting and accessing data previously stored in the access log database.
15. The method of claim 1 , wherein the extracted explanation represents at least one connection that begins at the access log database, continues through the access log database and/or the records database, and ends by returning to the access log database.
16. The method of claim 15 , wherein the at least one connection begins at data in the access log database describing information that was accessed by the user and wherein returning to the access log database comprises returning to data describing the user who accessed the information.
17. The method of claim 15 , wherein the extracted explanation represents a plurality of connection paths through the access log database.
18. The method of claim 1 , wherein the log data comprises at least two of patient data, event data, doctor data, and user data.
19. The method of claim 18 , wherein at least two of the patient data, event data, doctor data, and user data are stored in different accessible data tables in the access log database.
20. The method of claim 1 , wherein the access log database is separate from the computer system and coupled thereto through a network communication link.
21. The method of claim 20 , wherein the network communication link is a wired communication link.
22. The method of claim 20 , wherein the network communication link is a wireless communication link.
23. The method of claim 1 , wherein the template database is separate from the computer system and coupled thereto through a network communication link.
24. The method of claim 23 , wherein the network communication link is a wired communication link.
25. The method of claim 23 , wherein the network communication link is a wireless communication link.
26. The method of claim 1 , wherein the template database is stored on a remote server.
27. The method of claim 26 , wherein the access log database is stored on the remote server.
28. The method of claim 1 , wherein the records are medical records.
29. The method of claim 1 , wherein the records are government records.
30. The method of claim 1 , wherein the records are credit card records.
31. The method of claim 1 , further comprising automatically detecting misuse of the records by the user, based on the extracted explanation.
32. A user-centric auditing system configured to perform the method of claim 1 , wherein the extracted explanation is presented by the auditing system in a natural language format.
33. A computer system configured to perform the method of claim 1 .
34. A computer system comprising:
a memory;
a network interface for connecting the computer system to one or more remote network devices; and
a processor having
an explanation template collector module configured to access a template database having multiple explanation templates,
an explanation template mining module configured to automatically mine the multiple explanation templates to identify a candidate subset explanation templates,
an access log module configured to access an access log database containing log data of user access to records in a records database,
a log data inference module configured to infer data not available in the access log database containing the log data of user access to the records in the records database, and
an explanation identifier module configured to automatically identify an explanation for a particular user access to the database.
35. The computer system of claim 34 , wherein at least one of the records database, the access log database, and the template database are remote to the computer system and connected to the computer system through the network interface.
36. A user-centric auditing system configured to identify one or more explanations for user access to records in a records database, the auditing system comprising:
a processor; and
a memory storing computer-readable instructions that when executed by the processor cause the auditing system to,
access a template database containing a plurality of explanation templates, each explanation template corresponding to a potential reason for user access to the records database,
access log data stored in an access log database and identify the user access to the records in the records database,
automatically mine the plurality of explanation templates to identify candidate explanation templates based on the access log data,
automatically extract one or more explanations corresponding to the user access to the records in the records database, based on the candidate explanation templates, wherein the one or more explanations are in a natural language format, and
present the one or more explanations in the natural language format to a reviewing user.
37. The user-centric auditing system of claim 36 , wherein at least one extracted explanation explains numerous user accesses to the records in the records database.
38. The user-centric auditing system of claim 36 , wherein the extracted explanation is determined from a parameterized description string.
39. The user-centric auditing system of claim 36 , wherein a plurality of extracted explanations are presented to the reviewing user, wherein the memory further stores computer-readable instructions that when executed by the processor cause the auditing system to present the plurality of extracted explanations in a ranked order.
40. The user-centric auditing system of claim 36 , wherein the one or more explanations are suspicious explanations of access to be reviewed by the reviewing user.
41. The user-centric auditing system of claim 36 , wherein the one or more explanations are presented to the reviewing user for detecting misuse of the records by the user.