System and method for predicting disease onset

• US 6,110,109 A
• Filed: 03/26/1999
• Issued: 08/29/2000
• Est. Priority Date: 03/26/1999
• Status: Expired due to Term

• Current Assignee: Orthus Health, LLC (TMA @ Your Service LLC)
• Original Assignee: Biosignia, Inc.
• Inventors: Hu, Guizhou, Root, Martin M.
• Primary Examiner(s): O'Connor, Cary
• Assistant Examiner(s): CARTER, RYAN CHRISTOPHER
• Application Number:
  US09/277,257
• Time in Patent Office: 522 Days
• Field of Search: 600/300, 128/920, 128/923-924
• US Class Current: 600/300
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics
• CPC Class Codes: Y10S 128/923 by comparison of patient data to other data

Abstract

A method and apparatus for assessing a person'"'"'s disease status is provided using a plurality of disease prediction factors for that person and applying a multivariate disease prediction equation to the data to assess the disease status of that person. The multivariate disease prediction equation includes the contribution for each disease prediction factor in a comprehensive set of disease prediction factors. Each of the plurality of disease prediction factors for which data are available from the test person is included in the comprehensive set. The multivariate disease prediction equation is obtained by

(a) obtaining disease association data for each disease prediction factor in a first subset of the comprehensive set of disease prediction factors that are used in the multivariate disease prediction equation;

(b) obtaining additional disease association data for each disease prediction factor in at least one additional subset of the comprehensive set of disease prediction factors;

(c) obtaining cross-correlation data between each of the disease prediction factors in the comprehensive set, the cross-correlation data being obtained from a database in which all of the disease prediction factors in the comprehensive set are included; and

(d) using the disease association data together with the cross-correlation data to develop the multivariate disease prediction equation.

Claims

1. A method for assessing disease status comprising:

obtaining data on a plurality of disease prediction factors for a person;

selecting a multivariate disease prediction equation for assessing disease status for a specified disease; and

applying said multivariate disease prediction equation to said data to determine the disease status of said person;

said multivariate disease prediction equation including a term for quantifying the contribution to a specified disease status for each disease prediction factor in a comprehensive set of disease prediction factors;

each of said plurality of disease prediction factors for said person being included in said comprehensive set; and

said multivariate disease prediction equation being obtained by;

(a) obtaining disease association data for each disease prediction factor in a first subset of said comprehensive set of disease prediction factors;

(b) obtaining additional disease association data for each disease prediction factor in at least one additional subset of said comprehensive set of disease prediction factors;

the total number of disease prediction factors in each subset being less than the total number of disease prediction factors in said comprehensive set and each of said disease prediction factors in said comprehensive set being included in at least one subset;

(c) obtaining cross-correlation data between each of said disease prediction factors in said comprehensive set, said cross-correlation data being obtained from a database in which all of said disease prediction factors in said comprehensive set are included; and

(d) using the disease association data together with the cross-correlation data to develop said multivariate disease prediction equation.

2. The method of claim 1, wherein said multivariate disease prediction equation is a linear regression equation and said contribution to said specified disease status comprises a linear regression coefficient for each disease prediction factor.

3. The method of claim 2, wherein said linear regression equation is of the form:

space="preserve" listing-type="equation">Y=a+Σ

b.sub.i X.sub.i (I)

where Y represents said specified disease status;

a is a constant;

X_i represents said disease prediction factors in said comprehensive set of disease prediction factors; and

b_i represents the partial regression coefficients for each of said disease prediction factors.

4. The method of claim 3, wherein said partial regression coefficients are obtained using an equation of the form

5. The method of claim 1, wherein said multivariate disease prediction equation is a logistic regression equation and said contribution to said specified disease status comprises a logistic regression coefficient for each disease prediction factor.

6. The method of claim 1, wherein said multivariate disease prediction equation is of the form:

space="preserve" listing-type="equation">ln(P/(1-P))=a+Σ

b.sub.i X.sub.i

where P is the probability of having a specified disease status;

a is a constant;

X_i represents said disease prediction factors in said comprehensive set of disease prediction factors; and

b_i represents the partial logistic regression coefficients for each of said disease prediction factors.

7. The method of claim 1, wherein said specified disease status comprises said person currently having a specified disease.

8. The method of claim 1, wherein said specified disease status comprises said person currently having a specified probability of a specified future disease onset.

9. The method of claim 1, wherein said specified disease status comprises a continuous variable.

10. The method in claim 1, wherein said specified disease status is a dichotomous variable.

11. An apparatus for assessing disease status comprising:

a processor;

a port coupled to said processor; and

a memory coupled to said processor, said memory storing instructions adapted to be executed by said processor, said instructions including instructions for

receiving data on a plurality of disease prediction factors for a person; and

applying a multivariate disease prediction equation to said data to determine the disease status of said person;

said multivariate disease prediction equation including a term for quantifying the contribution to a specified disease status for each disease prediction factor in a comprehensive set of disease prediction factors;

each of said plurality of disease prediction factors for said person being included in said comprehensive set; and

said multivariate disease prediction equation being obtained by;

(a) obtaining disease association data for each disease prediction factor in a first subset of said comprehensive set of disease prediction factors;

(b) obtaining additional disease association data for each disease prediction factor in at least one additional subset of said comprehensive set of disease prediction factors;

the total number of disease prediction factors in each subset being less than the total number of disease prediction factors in said comprehensive set and each of said disease prediction factors in said comprehensive set being included in at least one subset;

(c) obtaining cross-correlation data between each of said disease prediction factors in said comprehensive set, said cross-correlation data being obtained from a database in which all of said disease prediction factors in said comprehensive set are included; and

(d) using the disease association data together with the cross-correlation data to develop said multivariate disease prediction equation.

12. The apparatus of claim 11, wherein said multivariate disease prediction equation is a linear regression equation and said contribution to said specified disease status comprises a linear regression coefficient for each disease prediction factor.

13. The apparatus of claim 12, wherein said linear regression equation is of the form:

space="preserve" listing-type="equation">Y=a+Σ

b.sub.i X.sub.i (I)

where Y represents said specified disease status;

a is a constant;

X_i represents said disease prediction factors in said comprehensive set of disease prediction factors; and

b_i represents the partial regression coefficients for each of said disease prediction factors.

14. The apparatus of claim 13, wherein said partial regression coefficients are obtained using an equation of the form

15. The apparatus of claim 11, wherein said multivariate disease prediction equation is a logistic regression equation and said contribution to said specified disease status comprises a logistic regression coefficient for each disease prediction factor.

16. The apparatus of claim 11, wherein said multivariate disease prediction equation is of the form

17. The apparatus of claim 11, wherein said specified disease status comprises said person currently having a specified disease.

18. The apparatus of claim 11, wherein said specified disease status comprises said person currently having a specified probability of a specified future disease onset.

19. The apparatus of claim 11, wherein said specified disease status comprises a continuous variable.

20. The apparatus of claim 11, wherein said specified disease status is a dichotomous variable.

21. A medium storing instructions for assessing a person'"'"'s disease status comprising a medium containing a memory storing instructions adapted to be executed by a processor, the instructions including:

receiving data on a plurality of disease prediction factors for a person; and

applying a multivariate disease prediction equation to said data to determine the disease status of said person;

said multivariate disease prediction equation including a term for quantifying the contribution to a specified disease status for each disease prediction factor in a comprehensive set of disease prediction factors;

each of said plurality of disease prediction factors for said person being included in said comprehensive set; and

said multivariate disease prediction equation being obtained by;

(a) obtaining disease association data for each disease prediction factor in a first subset of said comprehensive set of disease prediction factors;

(b) obtaining additional disease association data for each disease prediction factor in at least one additional subset of said comprehensive set of disease prediction factors;

the total number of disease prediction factors in each subset being less than the total number of disease prediction factors in said comprehensive set and each of said disease prediction factors in said comprehensive set being included in at least one subset;

(c) obtaining cross-correlation data between each of said disease prediction factors in said comprehensive set, said cross-correlation data being obtained from a database in which all of said disease prediction factors in said comprehensive set are included; and

(d) using the disease association data together with the cross-correlation data to develop said multivariate disease prediction equation.

22. The medium of claim 21, wherein said multivariate disease prediction equation is a linear regression equation and said contribution to said specified disease status comprises a linear regression coefficient for each disease prediction factor.

23. The medium of claim 22, wherein said linear regression equation is of the form:

space="preserve" listing-type="equation">Y=a+Σ

b.sub.i X.sub.i (I)

where Y represents said specified disease status;

a is a constant;

X_i represents said disease prediction factors in said comprehensive set of disease prediction factors; and

b_i represents the partial regression coefficients for each of said disease prediction factors.

24. The medium of claim 23, wherein said partial regression coefficients are obtained using an equation of the form

25. The medium of claim 21, wherein said multivariate disease prediction equation is a logistic regression equation and said contribution to said specified disease status comprises a logistic regression coefficient for each disease prediction factor.

26. The medium of claim 21, wherein said multivariate disease prediction equation is of the form

27. The medium of claim 21, wherein said specified disease status comprises said person currently having a specified disease.

28. The medium of claim 21, wherein said specified disease status comprises said person currently having a specified probability of a specified future disease onset.

29. The medium of claim 21, wherein said specified disease status comprises a continuous variable.

30. The medium of claim 21, wherein said specified disease status is a dichotomous variable.