SYSTEMS AND METHODS FOR MINING AGGREGATED CLINICAL DOCUMENTATION USING CONCEPT ASSOCIATIONS

• US 20130262142A1
• Filed: 02/15/2013
• Published: 10/03/2013
• Est. Priority Date: 09/01/2010
• Status: Active Grant

• Current Assignee: Apixio LLC
• Original Assignee: Jehoshua Bruck
• Inventors: Sethumadhavan, Vishnuvyas, Campana, Mary Ellen, Rogers, Robert Derward, Dastmalchi, Shahram Shawn
• Granted Patent: US 10,453,574 B2
• Time in Patent Office: Days
• US Class Current: 705/3
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A medical processor computes concept associations by mining aggregated data from patient documents thereby reducing risk of PHI exposure. The processor identifies clinically relevant terms in patient documents, compute associations between pairs of clinically relevant terms using co-occurrences, and filter out random associations. A knowledge provider receives user query concepts, retrieves patient concepts, and extracts relevant apixions from an association matrix. The knowledge provider intersects relevant apixions with patient concepts, ranks and provides the results to the user.

Claims

1. In a Medical Information Navigation Engine (“

MINE”

), a computerized method for computing concept associations by mining patient documentation in the aggregate, the method comprising;

assembling a plurality of patient documents;

identifying clinically relevant terms in the plurality of patient documents;

computing associations between at least one pair of clinically relevant terms using co-occurrences within at least one of the plurality of patient documents;

filtering out random associations from the computed associations; and

storing the filtered computed associations.

2. The method of claim 1 wherein storing the filtered computed associations are stored in an association matrix.

3. The method of claim 1 wherein the assembling of the plurality of patient documents include aggregating patient data from the plurality of patient documents.

4. The method of claim 1 wherein the associations are computed using aggregate patient data from the at least one of the plurality of patient document thereby reducing risk of exposing protected health information (PHI).

5. The method of claim 1 wherein the filtering out of the random associations include comparing a probability of concurrent occurrences.

6. The method of claim 2 further comprising:

receiving a plurality of query concepts from a user query;

retrieving a plurality of patient concepts from at least one of the plurality of patient documents;

extracting at least one relevant apixion from the association matrix, wherein the extracted at least one relevant apixion is related to at least one of the plurality of query concepts of the user query;

intersecting the at least one relevant apixion extracted from the association matrix with the plurality of patient concepts;

ranking results of the intersecting;

processing the ranked results; and

providing the processed ranked results to a user.

7. The method of claim 6 wherein the plurality of query concepts is identified in accordance with a standard coding system.

8. The method of claim 6 wherein the processing of the ranked results is processed in a knowledge exchange and the processing includes recording whether the processed ranked results were selected by the user.

9. The method of claim 6 wherein the ranking of the results is based on at least one of recency, risk, strength of association, source of data and user context.

10. The method of claim 6 wherein the ranked results are provided to the user within a workflow of the user, and wherein the workflow includes one of at least a billing system and a case management system.

11. A computerized Medical Information Navigation Engine (“

MINE”

) configured to compute concept associations by mining patient documentation in the aggregate, the MINE comprising;

a medical processor configured to;

assemble a plurality of patient documents;

identify clinically relevant terms in the plurality of patient documents;

compute associations between at least one pair of clinically relevant terms using co-occurrences within at least one of the plurality of patient documents; and

filter out random associations from the computed associations; and

an association matrix configured to store the filtered computed associations.

12. The MINE of claim 11 wherein the assembly of the plurality of patient documents include aggregating patient data from the plurality of patient documents.

13. The MINE of claim 11 wherein the associations are computed using aggregate patient data from the at least one of the plurality of patient document thereby reducing risk of exposing protected health information (PHI).

14. The MINE of claim 11 wherein the medical processor is further configured to filter out the random associations by comparing a probability of concurrent occurrences.

15. The MINE of claim 11 further comprising a related results knowledge provider configured to:

receive a plurality of query concepts from a user query;

retrieve a plurality of patient concepts from at least one of the plurality of patient documents;

extract at least one relevant apixion from the association matrix, wherein the extracted at least one relevant apixion is related to at least one of the plurality of query concepts of the user query;

intersect the at least one relevant apixion extracted from the association matrix with the plurality of patient concepts;

rank results of the intersecting;

process the ranked results; and

provide the processed ranked results to a user.

16. The MINE of claim 15 wherein the plurality of query concepts is identified in accordance with a standard coding system.

17. The MINE of claim 15 wherein the ranked results is processed in a knowledge exchange of the knowledge provider and wherein the knowledge exchange is configured to record whether the processed ranked results were selected by the user.

18. The MINE of claim 15 wherein the ranking of the results is based on at least one of recency, risk, strength of association, source of data and user context.

19. The MINE of claim 15 wherein the ranked results are provided to the user within a workflow of the user, and wherein the workflow includes one of at least a billing system and a case management system.