Algorithmic method for generating a medical utilization profile for a patient and to be used for medical risk analysis decisioning

• US 8,015,136 B1
• Filed: 04/03/2008
• Issued: 09/06/2011
• Est. Priority Date: 04/03/2008
• Status: Active Grant

• Current Assignee: Dynamic Healthcare Systems, Inc.
• Original Assignee: Dynamic Healthcare Systems, Inc.
• Inventors: B'Far, Reza M, Baker, Jay
• Primary Examiner(s): VINCENT, DAVID ROBERT
• Application Number:
  US12/080,497
• Time in Patent Office: 1,251 Days
• Field of Search: 706/12, 706 45- 48, 706/52
• US Class Current: 706/45
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

An algorithmic method for generating a medical utilization profile for a patient to facilitate decision making in the medical risk analyses, wherein the algorithmic method is comprised of four key components: (1) key independent variables recognized including medical data sources; (2) an ontology based process to organize these variables; (3) rules (probability distributions etc.) that can be used to draw correlative information from the given data and its ontological organization; and (4) usage of reasoning engine to process the gathered information and draw inferences.

Claims

1. A computer-implemented algorithmic method for generating a medical utilization profile for a patient, comprising steps of:

a. determining medical data sources which contain medical data of said patient;

b. applying reasoning engines to define an initial ontology model which matches said data sources;

c. creating an ontology implementation that includes characterization, and

classification of said medical data in a given analysis sequence;

d. enabling said initial ontology model to thereby modify said initial model; and

e. comparing the initial ontology model with the modified ontology model through a Set Theory including comparing with intersections and unions which yields results that drive business rules; and

f. generating a patient profile.

2. The algorithmic method in accordance with claim 1, wherein said medical data sources are provided including initially and sub-sequentially.

3. The algorithmic method in accordance with claim 1, wherein said medical data are provided including initially and sub-sequentially.

4. The algorithmic method in accordance with claim 1, said reasoning engine including a device of computation.

5. The algorithmic method in accordance with claim 1, wherein said base ontology model includes a hierarchical structure which is of a tree structure comprising various branches having the respective nodes and predicates, said nodes are represented by the respective medical data sources, and said predicates represent the respective relationships between the respective medical data sources.

6. The algorithmic method in accordance with claim 5, wherein said nodes represent including subjects and objects.

7. The algorithmic method in accordance with claim 5, wherein said predicates represent including mathematical, statistical and stochastic relationships.

8. The algorithmic method in accordance with claim 1, wherein said base ontology model includes structure of a canonical model.

9. The algorithmic method in accordance with claim 1, further comprising that various risk models apply to said base ontology model, which results in association of risk attributes with said subjects, objects or predicates of said base ontology model.

10. The algorithmic method in accordance with claim 1, wherein said medical data sources are gathered through methods including prevalence, expert knowledge, crosswalks, and coding intensity.

11. The algorithmic method in accordance with claim 1, wherein said ontology implementation includes applying said reasoning engines to fill the initial medical data into the respective nodes of the initial ontology model.

12. The algorithmic method in accordance with claim 1, wherein said enabling includes quantifying the mathematical, statistical and stochastic relationships between the respective nodes of the initial ontology model, so that their respective relationships and subject-object positions can be confirmed or changed which leads to modify the initial ontology model.

13. The algorithmic method in accordance with claim 1, wherein said comparison includes a hierarchical comparison.

14. The algorithmic method in accordance with claim 13, wherein said hierarchical comparison starts comparing from central nodes to edge nodes of the respective entire initial and modified ontology model.

15. The algorithmic method in accordance with claim 1, wherein said comparison includes a side-by-side comparison.

16. The algorithmic method in accordance with claim 15, wherein said side-by-side comparison compares with two nodes connected by a same predicate of said initial ontology model with the respective two nodes connected by the respective same predicate of a modified ontology model, and further comparing a subsequent respective two nodes till all of nodes of the respective ontology models are compared.

17. The algorithmic method in accordance with claim 1, further comprising a feed-back loop mechanism for refining said initial base ontology model.

18. The algorithmic method in accordance with claim 1, wherein said business rule includes a decision making in medical risk analyses.

19. The algorithmic method in accordance with claim 18, wherein said risk analyses include an analysis of risk which may be generated by ICD-9 coded diseases.

20. The algorithmic method in accordance with claim 1, further comprising a mathematical convolution operation while the comparison window of the convolution, itself, could be over any area of the ontology and the convolution could be of any degree of complexity mathematically with potential stochastic and statistical properties.