Methods and systems for utilizing prediction models in healthcare

• US 10,515,188 B2
• Filed: 06/07/2017
• Issued: 12/24/2019
• Est. Priority Date: 03/04/2005
• Status: Active Grant

• Current Assignee: Terumo Medical Corporation (Terumo Corporation)
• Original Assignee: Health Outcomes Sciences, Inc. (Terumo Corporation)
• Inventors: Soto, Gabriel Enrique, Spertus, John Albert
• Primary Examiner(s): Craig, Dwin M
• Application Number:
  US15/616,706
• Publication Number:
  US 20170357761A1
• Time in Patent Office: 930 Days
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A method for providing decision support includes using a programmed computer to input a regression model specification, and to repeat the input a plurality of times to obtain and store a plurality of regression model specifications. The method further includes using the programmed computer to analyze selected regression model specifications to determine at least one of common variables and functions of common variables, to thereby determine a reduced-redundancy request for input of variables, when a plurality of the stored regression model specifications are selected for use.

Claims

1. A system, comprising:

a server computer comprising hardware including a processing device;

non-transitory computer readable memory that stores instructions, that when executed by the server computer causes the system to perform operations comprising;

analyzing sets of predictors associated with one or more health outcomes for a patient,

wherein the predictors in the sets of predictors are collected from one or more networked devices;

generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors, based at least in part on a removal of one or more duplicate predictors or a removal of one or more predictors that can be mathematically or logically derived from other predictors within the collected sets of predictors; and

providing a user interface comprising the optimal set of predictors for display on a computer display, or a programmatic interface configured to electronically provide the optimal set of predictors, or both a user interface comprising the optimal set of predictors for display on a computer display and a programmatic interface configured to electronically provide the optimal set of predictors.

2. The system as defined in claim 1, the operations further comprising dynamically generating a patient information request interface based at least in part on the sets of predictors.

3. The system as defined in claim 1, the operations further comprising dynamically generating a patient information request interface based at least in part on the sets of predictors, wherein the dynamically generated interface obtains patient information from a patient information system using an XML web service without displaying the patient information request interface.

4. The system as defined in claim 1, the operations further comprising dynamically generating a patient information request interface based at least in part on the sets of predictors, wherein the system comprises a network interface configured to transmit the dynamically generated patient information request interface over a network to a client computer.

5. The system as defined in claim 1, wherein generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors further comprises determining whether there are variables that may be shared among the sets of predictors.

6. The system as defined in claim 1, wherein generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors further comprises utilizing regression model specifications for a plurality of regression models to determine common variables and/or functions of common variables, to thereby determine a reduced-redundancy request for input of at least one variable.

7. The system as defined in claim 1, wherein providing the optimal set of predictors to project a health outcome of the patient further comprises providing visual representations of predicted health outcomes.

8. The system as defined in claim 1, the operations further comprising generating a graphical user interfaces that enables a user to graphically build, edit, and visualize statistical health outcome predictive models.

9. The system as defined in claim 1, the system further comprising a parameter library manager, and outcome library manger, and a model library manager.

10. The system as defined in claim 1, the operations further comprising:

providing for display, using a visual model editor, a selection of parametric regression forms;

receive a selection of a first of the parametric regression forms;

provide for display requests for an outcome type and one or more regression model parameter names and corresponding parameter types;

display a visual selection of at least one parameter name and mathematical transform;

accept at least one parameter name and at least one mathematical transform;

obtain values of coefficients for selected regression model type and outcome;

store, in association, the outcome, and associated coefficients.

11. A non-transitory computer readable medium that stores instructions, that when executed by a computing device cause the computing device to perform operations comprising:

analyzing sets of predictors associated with one or more health outcomes for a patient,

wherein the predictors in the sets of predictors are collected from one or more networked devices;

generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors, based at least in part on a removal of one or more duplicate predictors or a removal of one or more predictors that can be mathematically or logically derived from other predictors within the collected sets of predictors; and

providing a user interface comprising the optimal set of predictors for display on a computer display, or a programmatic interface configured to electronically provide the optimal set of predictors, or both a user interface comprising the optimal set of predictors for display on a computer display and a programmatic interface configured to electronically provide the optimal set of predictors.

12. A computer-implemented method, the method comprising:

analyzing, by a computer system comprising processing device, sets of predictors associated with one or more health outcomes for a patient,

wherein the predictors in the sets of predictors are collected from one or more networked devices;

generating, by the computer system, an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors,

providing by the computer system, over a network, a user interface comprising the optimal set of predictors for display on a computer display, or a programmatic interface configured to electronically provide the optimal set of predictors, or both a user interface comprising the optimal set of predictors for display on a computer display and a programmatic interface configured to electronically provide the optimal set of predictors.

13. The method as defined in claim 12, the method further comprising dynamically generating a patient information request interface based at least in part on the sets of predictors.

14. The method as defined in claim 12, the method further comprising dynamically generating a patient information request interface based at least in part on the sets of predictors, wherein the dynamically generated interface obtains patient information from a patient information system using an XML web service without displaying the patient information request interface.

15. The method as defined in claim 12, the method further comprising dynamically generating a patient information request interface based at least in part on the sets of predictors, wherein the system comprises a network interface configured to transmit the dynamically generated patient information request interface over a network to a client computer.

16. The method as defined in claim 12, wherein generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors further comprises determining whether there are variables that may be shared among the sets of predictors.

17. The method as defined in claim 12, wherein generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors further comprises utilizing regression model specifications for a plurality of regression models to determine common variables and/or functions of common variables, to thereby determine a reduced-redundancy request for input of at least one variable.

18. The method as defined in claim 12, the method further comprising generating a graphical user interfaces that enables a user to graphically build, edit, and visualize statistical health outcome predictive models.

19. The method as defined in claim 12, the method further comprising:

providing for display, using a visual model editor, a selection of parametric regression forms;

receive a selection of a first of the parametric regression forms;

provide for display requests for an outcome type and one or more regression model parameter names and corresponding parameter types;

display a visual selection of at least one parameter name and mathematical transform;

accept at least one parameter name and at least one mathematical transform;

obtain values of coefficients for selected regression model type and outcome;

store, in association, the outcome, and associated coefficients.

20. The method as defined in claim 12, wherein generating an optimal set of predictors containing reduced informational redundancy using the collected sets of predictors further comprises removing of one or more duplicate predictors or removing one or more predictors that can be mathematically or logically derived from other predictors within the collected sets of predictors.