Method of classifying a bill

• US 10,140,636 B2
• Filed: 03/20/2014
• Issued: 11/27/2018
• Est. Priority Date: 04/02/2010
• Status: Active Grant

• Current Assignee: Mitchell International Incorporated (Mitchell International Holding, Inc.)
• Original Assignee: Mitchell International Incorporated (Mitchell International Holding, Inc.)
• Inventors: Pepper, Noah Mantell, Francis, Charles Cooper, Quarum, Merritt Laird
• Primary Examiner(s): Frenel, Vanel
• Application Number:
  US14/221,183
• Publication Number:
  US 20140289083A1
• Time in Patent Office: 1,713 Days
• Field of Search: 705 34, 705311, 705400, 705 2
• CPC Class Codes: G06Q 30/04 Billing or invoicing
• CPC Class Codes: G06Q 40/12 Accounting

Abstract

Briefly, embodiments of a method of classifying a bill are disclosed.

Claims

1. A method for automated electronic document auditing and analysis to facilitate electronic document classification, the method implemented by one or more computing devices and comprising:

deriving at least one machine learning bill classification scheme using one or more statistical based decision processes by encoding at least a first set of electronic medical bills to generate at least one sparse binary matrix and generating one or more models using the at least one sparse binary matrix as input to a support vector machine (SVM) process;

applying to a second set of electronic medical bills the one or more models, wherein at least some of the second set of electronic medical bills comprise different electronic billing forms and the second set of electronic medical bills is different from the first set of electronic medical bills;

automatically classifying one or more of the second set of electronic medical bills as associated with one of a plurality of adjudication types based on the application of the one or more models to the second set of electronic medical bills; and

inserting the classified one or more of the second set of electronic medical bills into an electronic database.

2. The method of claim 1, wherein the one of the adjudication types comprises a medical bill origin.

3. The method of claim 1, further comprising, prior to applying the at least one bill classification scheme, applying a rule-based classification scheme.

4. The method of claim 1, wherein the at least one bill classification scheme tolerates false negatives more than false positives.

5. The method of claim 1, wherein the at least one bill classification scheme adjusts for gradual shifts in a population of bills to be classified.

6. The method of claim 1, wherein the at least one bill classification scheme includes a classification for ambiguous bills.

7. The method of claim 1, further comprising automatically verifying, based on the classification, that one or more of the one or more of the second set of electronic medical bills is associated with one or more provided medical services that have been accurately billed and outputting a result of the verification.

8. An apparatus comprising memory comprising programmed instructions stored thereon and one or more processors configured to be capable of executing the stored programmed instructions to:

derive at least one machine learning bill classification scheme using one or more statistical based decision processes by encoding at least a first set of electronic medical bills to generate at least one sparse binary matrix and generating one or more models using the at least one sparse binary matrix as input to a support vector machine (SVM) process;

apply to a second set of electronic medical bills the one or more models, wherein at least some of the second set of electronic medical bills comprise different electronic billing forms and the second set of electronic medical bills is different from the first set of electronic medical bills;

automatically classify one or more of the second set of electronic medical bills as associated with one of a plurality of adjudication types based on the application of the one or more models to the second set of electronic medical bills; and

insert the classified one or more of the second set of electronic medical bills into an electronic database.

9. The apparatus of claim 8, wherein one of the adjudication types comprises a medical bill origin.

10. The apparatus of claim 8, wherein the one or more processors are further configured to be capable of executing the stored programmed instructions to apply a rule-based classification scheme.

11. The apparatus of claim 8, wherein the at least one bill classification scheme tolerates false negatives more than false positives.

12. The apparatus of claim 8, wherein the at least one bill classification scheme adjusts for gradual shifts in a population of bills to be classified.

13. The apparatus of claim 8, wherein the at least one bill classification scheme includes a classification for ambiguous bills.

14. The apparatus of claim 8, wherein the one or more processors are further configured to be capable of executing the stored programmed instructions to verify, based on the classification, that one or more of the one or more of the second set of electronic medical bills is associated with one or more provided medical services that have been accurately billed and output a result of the verification.

15. A non-transitory computer readable medium having stored thereon instructions for automated electronic document auditing and analysis to facilitate electronic document classification comprising machine executable code which when executed by at least one processor, causes the processor to:

derive at least one machine learning bill classification scheme using one or more statistical based decision processes by encoding at least a first set of electronic medical bills to generate at least one sparse binary matrix and generating one or more models using the at least one sparse binary matrix as input to a support vector machine (SVM) process;

apply to a second set of electronic medical bills the one or more models, wherein at least some of the second set of electronic medical bills comprise different electronic billing forms and the second set of electronic medical bills is different from the first set of electronic medical bills;

automatically classify one or more of the second set of electronic medical bills as associated with one of a plurality of adjudication types based on the application of the one or more models to the second set of electronic medical bills; and

insert the classified one or more of the second set of electronic medical bills into an electronic database.

16. The medium of claim 15, wherein the at least one bill classification scheme tolerates false negatives more than false positives.

17. The medium of claim 15, wherein the at least one bill classification scheme adjusts for gradual shifts in a population of bills to be classified.

18. The medium of claim 15, wherein the at least one bill classification scheme includes a classification for ambiguous bills.

19. The medium of claim 15, wherein the one of the adjudication types comprises a medical bill origin.

20. The medium of claim 15, wherein the machine executable code, when executed by the processor further causes the processor to apply a rule-based classification scheme.

21. The medium of claim 15, wherein the machine executable code, when executed by the processor further causes the processor to verify, based on the classification, that one or more of the one or more of the second set of electronic medical bills is associated with one or more provided medical services that have been accurately billed and output a result of the verification.