Reflecting the quantitative impact of ordinal indicators

• US 8,972,333 B2
• Filed: 06/13/2014
• Issued: 03/03/2015
• Est. Priority Date: 06/22/2011
• Status: Active Grant

• Current Assignee: James R. Miller III
• Original Assignee: James R. Milller III
• Inventors: Miller, James R. III
• Primary Examiner(s): CHANG, LI WU
• Application Number:
  US14/304,776
• Publication Number:
  US 20140297581A1
• Time in Patent Office: 263 Days
• Field of Search: None
• US Class Current: 706/52
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Embodiments of the invention may help improve the accuracy of individually tailored prognoses in dealing with individual patient prognoses for cancer, diabetes, multiple sclerosis, heart disease, and/or other diseases. Alternate embodiments of the invention may be useful in improving the accuracy of predictions in a variety of other contexts and fields. A method according to various embodiments of the invention includes generating a univariate impact reflecting index (UIRI) based on an indicator, the indicator having a monotonic relationship to a focal conclusion, wherein the UIRI spaces subscales of the indicator according to a magnitude of impact of the indicator to the focal conclusion.

Claims

1. A computer-implemented method comprising:

generating, by a computer system, a univariate impact reflecting index (UIRI) based on an indicator, the indicator having a monotonic relationship to a focal conclusion, wherein the UIRI spaces subscales of the indicator according to a magnitude of impact of the indicator on the focal conclusion, and wherein generating the UIRI includes forming partitions in the scale for the indicator using one or more cut points within the scale and spacing the partitions.

2. The method of claim 1, wherein forming partitions in the scale and spacing the partitions are performed separately on one or more stratified samples.

3. The method of claim 1, determining, by the computer system, a probability that the focal conclusion will occur, wherein determining the probability is based at least partially on one of:

a value in the UIRI, or a regression analysis of the UIRI.

4. The method of claim 3, further comprising determining a level of accuracy for the determined probability.

5. The method of claim 4, wherein determining the level of accuracy for the determined probability is based on one or more of:

a receiver operating characteristic (ROC) analysis;

an area under the ROC curve (AUC) analysis;

a maximum probability of correct prediction analysis;

an absolute probabilistic error analysis; and

a probability scale characteristics analysis.

6. The method of claim 3, wherein the probability that the focal conclusion will occur is determined for a single entity.

7. The method of claim 6, wherein the indicator is based upon measured data from one or more other entities stratified into separate respective samples.

8. The method of claim 7, further comprising selecting the one or more other entities based on one or more similarities between the single entity and the one or more other entities.

9. The method of claim 3, wherein the regression analysis is selected from the group consisting of:

a logistic regression analysis;

a Cox proportional hazards analysis; and

a linear regression analysis.

10. The method of claim 3, further comprising:

generating a respective UIRI for each of a plurality of indicators, wherein determining the probability is based at least partially on a regression analysis of each of the generated UIRIs.

11. The method of claim 10, further comprising weighting values of an UIRI index of an indicator of the plurality of indicators based on the indicator'"'"'s propensity to correctly predict whether the focal conclusion will occur.

12. The method of claim 10, wherein each indicator of the plurality of indicators is based upon measured data from one or more other entities, and wherein at least one of the one or more entities includes measured data for a first indicator of the plurality of indicators and does not have measured data for a second indicator of the plurality of indicators.

13. The method of claim 1, further comprising selecting the indicator from a plurality of possible indicators.

14. The method of claim 13, wherein selecting the indicator is at least partially based on at least one of a desired sensitivity and a desired specificity.

15. The method of claim 13, wherein selecting the indicator includes verifying that the indicator has a monotonic relationship to the focal conclusion.

16. The method of claim 13, wherein selecting the indicator includes:

determining one or more cut points for the indicator;

determining whether a partition formed by one or more of the cut points has at least a minimum scale partition size; and

determining, based the one or more cut points, whether the scale partition has a monotonic relationship to the focal conclusion.

17. The method of claim 1, further comprising verifying the indicator'"'"'s admissibility, where verifying the indicator'"'"'s admissibility includes verifying an ordinal scale for the indicator.

18. The method of claim 17, wherein generating the UIRI is based on the verified ordinal scale for the indicator.

19. The method of claim 1, wherein generating the UIRI includes presenting the UIRI to a user of the computer system, wherein the UIRI is presented to the user in the form of one of:

a list of values of the UIRI, or a graph of the UIRI.

20. A tangible, non-transitory computer-readable medium storing instructions that, when executed by a computing device, cause the computing device to:

generate a univariate impact reflecting index (UIRI) based on an indicator, the indicator having a monotonic relationship to a focal conclusion, wherein the UIRI spaces subscales of the indicator according to a magnitude of impact of the indicator on the focal conclusion, and wherein generating the UIRI includes forming partitions in the scale for the indicator using one or more cut points within the scale and spacing the partitions.

21. A system comprising:

a processor;

a memory coupled to the processor and storing instructions that, when executed by the processor, cause the system to;

generate a univariate impact reflecting index (UIRI) based on an indicator, the indicator having a monotonic relationship to a focal conclusion, wherein the UIRI spaces subscales of the indicator according to a magnitude of impact of the indicator on the focal conclusion, and wherein generating the UIRI includes forming partitions in the scale for the indicator using one or more cut points within the scale and spacing the partitions.