Dynamic topological system and method for efficient claims processing

• US 10,474,792 B2
• Filed: 11/12/2015
• Issued: 11/12/2019
• Est. Priority Date: 05/18/2015
• Status: Active Grant

• Current Assignee: Change Healthcare Holdings, LLC (UnitedHealth Group Incorporated)
• Original Assignee: Change Healthcare Holdings, LLC (UnitedHealth Group Incorporated)
• Inventors: Alstad, Colin Erik, Tanner, Theodore, Gosnell, Denise Koessler
• Primary Examiner(s): Pauls, John A
• Application Number:
  US14/939,973
• Publication Number:
  US 20160342751A1
• Time in Patent Office: 1,461 Days
• Field of Search: 705 2, 705 4
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms

Abstract

A dynamic topological system and method for efficient claims processing are provided. The dynamic topological system and method for efficient claims processing may be used in a healthcare system. The dynamic topological system and method for efficient claims processing is easily extensible, maintainable and extendable.

Claims

1. A healthcare claims processing apparatus, comprising:

a computer system having a processor and a memory;

a database associated with the computer system that stores one or more claims records wherein each claim record contains data about a claim and the database stores data about a plurality of claims;

a claims processing component that generates a compressed representation of the plurality of claims including a particular claim in the one or more claims records, the compressed representation having a plurality of nodes with which the particular claim is associated, partitions the compressed representation into one or more neighborhoods of nodes using a weighted undirected graph having a plurality of vertexes and a plurality of edges that connect two of the plurality of vertexes and each vertex represents a neighborhood of nodes and has a weight and each edge joins two vertexes only if at least one of the claims in the two vertexes exists in each vertex, identifies the neighborhood of nodes that includes the particular claim and determines, using a classifier for the identified neighborhood of nodes for the particular claim, a status of the particular claim, wherein the status is one of denied, overpaid or underpaid.

2. The apparatus of claim 1, wherein the compressed representation is a simplicial complex.

3. The apparatus of claim 2, wherein the claims processing component converts the one or more claims records into a feature matrix before generating the compressed representation of the one or more claims records.

4. The apparatus of claim 1, wherein the claims processing component trains one or more classifiers for each neighborhood of nodes.

5. The apparatus of claim 4, wherein the claims processing component selects a best classifier from the trained classifiers for a particular neighborhood of nodes wherein the particular claim is associated with the particular neighborhood of nodes.

6. The apparatus of claim 1, wherein the claims processing component gathers the status of the claim and feeds back the status of the claim.

7. A method for healthcare claims processing, comprising:

obtaining one or more claims records wherein each claim record contains data about a claim;

generating a compressed representation of a plurality of claims including a particular claim in the one or more claims records, the compressed representation having a plurality of nodes with which the particular claim is associated;

partitioning the compressed representation into one or more neighborhoods of nodes using a weighted undirected graph having a plurality of vertexes and a plurality of edges that connect two of the plurality of vertexes and each vertex represents a neighborhood of nodes from the plurality of claims and each edge joins two vertexes only if at least one of the claims in the two vertexes exists in each vertex;

identifying the neighborhood of nodes that includes the particular claim; and

determining, using a classifier for the identified neighborhood of nodes for the particular claim, a status of the particular claim, wherein the status is one of denied, overpaid or underpaid.

8. The method of claim 7, wherein the compressed representation is a simplicial complex.

9. The method of claim 8 further comprising converting the one or more claims records into a feature matrix before generating the compressed representation of the one or more claims records.

10. The method of claim 7, wherein determining using the classifier further comprises training one or more classifiers for each neighborhood of nodes.

11. The method of claim 10, wherein determining using the classifier further comprises selecting a best classifier from the trained classifiers for a particular neighborhood of nodes wherein the particular claim is associated with the particular neighborhood of nodes.

12. The method of claim 7 further comprising gathering the status of the claim and feeding back the status of the claim.

13. A healthcare system, comprising:

a computer system having a processor and a memory;

a health marketplace system hosted by the computer system;

a database associated with the computer system that stores one or more claims records wherein each claim record contains data about a claim and the database stores data about a plurality of claims;

a claims processing component that generates a compressed representation of the plurality of claims including a particular claim in the one or more claims records, the compressed representation having a plurality of nodes with which the particular claim is associated, partitions the compressed representation into one or more neighborhoods of nodes using a weighted undirected graph having a plurality of vertexes and a plurality of edges that connect two of the plurality of vertexes and each vertex represents a neighborhood of nodes from the plurality of claims and each edge joins two vertexes only if at least one of the claims in the two vertexes exists in each vertex, identifies the neighborhood of nodes that includes the particular claim, and determines, using a classifier for the identified neighborhood of nodes for the particular claim, a status of the particular claim, wherein the status is one of denied, overpaid or underpaid.

14. The system of claim 13, wherein the compressed representation is a simplicial complex.

15. The system of claim 14, wherein the claims processing component converts the one or more claims records into a feature matrix before generating the compressed representation of the one or more claims records.

16. The system of claim 13, wherein the claims processing component trains one or more classifiers for each neighborhood of nodes.

17. The system of claim 16, wherein the claims processing component selects a best classifier from the trained classifiers for a particular neighborhood of nodes wherein the particular claim is associated with the particular neighborhood of nodes.

18. The system of claim 13, wherein the claims processing component gathers the status of the claim and feeds back the status of the claim.