Computerized systems and methods for facilitating clinical decision making

• US 10,431,336 B1
• Filed: 09/30/2011
• Issued: 10/01/2019
• Est. Priority Date: 10/01/2010
• Status: Active Grant

• Current Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Original Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Inventors: Murrish, John Christopher, McNair, Douglas S., Kailasam, Kanakasabha
• Primary Examiner(s): Burgess, Joseph D
• Application Number:
  US13/250,072
• Time in Patent Office: 2,923 Days
• Field of Search: 705 2, 705 3
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 20/10 relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A system, method, and computer-readable media are provided for facilitating clinical decision making, and in particular, facilitating treatment of a person having congestive heart failure. The method includes the step of receiving patient information for a patient. The method also includes the steps of determining whether the patient information suggests a trigger event, and upon determining a trigger event, determining at least one goal associated with the trigger event. The method also includes the steps of selecting a first plan, from a library of plans, corresponding to the goal and executing the plan, which further includes determining a solver to determine patient conditions or recommended treatments, receiving parameters for the determined solver, and preparing patient information for the determined solver. The method further includes instantiating the solver based on the received parameters and the prepared patient information, and applying the solver to determine a patient condition or recommended treatment.

Claims

1. A computerized method for determining treatment for a patient having congestive heart failure and at least one other medical condition, the method comprising:

receiving, by one or more computer processing components, patient-results information for a patient from a patient information database, wherein the patient information database is remote from at least one of the one or more computer processing components;

extract, by the one or more computer processing components, discrete patient data from the patients-results information;

receiving, by the one or more computer processing components, one or more rules from a parameters database, wherein the parameters database is remote from the one or more computer processing components, and wherein the one or more rules comprise a rule relevant to the discrete patient data;

based on said discrete patient data and the one or more rules, determining, by a first processing component of the one or more computer processing components, whether said patient-results information suggests a trigger event;

upon a determination that said trigger event has occurred, determining by a second processing component of the one or more computer processing components, at least one goal based on said trigger event, wherein the at least one goal is communicated to the one or more computer processing components by the parameters database;

based on said goal, selecting, by the one or more computer processing components, a first plan, from a library of plans, corresponding to said determined goal, wherein the library of plans is communicated to the one or more computer processing components by the parameters database;

in response to selecting the first plan, receiving additional patient-results information specified by said plan, wherein the additional patient-results information is communicated to the one or more computer processing components by the patient information database; and

executing said first plan, wherein said execution comprises;

(1) from among a library of solvers, determining a first solver to determine patient conditions and recommended treatments, said first solver comprising a finite state machine;

(2) receiving first-solver parameters for said first solver, wherein said first solver is running on the one or more computer processing components;

(3) preparing patient-results information for said first solver;

(4) instantiating said first solver based on said prepared patient-results information and said first-solver parameters;

(5) applying said first solver to determine said patient conditions and recommended treatments, each patient condition comprising an evaluated state for the patient condition;

(6) determining that a second solver is needed to assist said first solver to determine patient conditions and recommended treatments, wherein the second solver is running on the one or more computer processing components;

(7) invoking the second solver to assist said first solver to determine patient conditions and recommended treatments, said second solver comprising a mixed-integer linear solver, said second solver being invoked by the first solver;

(8) communicating the evaluated states for the determined said patient conditions from the first solver to the second solver;

(9) preparing patient-results information for said second solver;

(10) instantiating said second solver based on said prepared patient-results information and said evaluated states;

(11) based on said determined patient conditions and recommended treatments, communicating actions and dispositions specific to said patient from the one or more computer processing components to a second computer processing component, wherein the second computer processing component is remote from the one or more computer processing components; and

(12) displaying the communicated actions and dispositions specific to said patient on a user interface of the second computer processing component.

2. The computerized method of claim 1, wherein said actions and dispositions specific to said patient are determined by invoking an expert rules engine, and wherein said determined actions and dispositions include at least one of an order, patient condition, recommended treatment, recommended additional testing, or recommended execution of a second plan.

3. The computerized method of claim 1, wherein said determined actions and dispositions specific to said patient are processed for presentation.

4. The computerized method of claim 3, wherein said determined actions and dispositions processed for presentation further includes functionality for presenting information enabling a health-care provider to see which plan, of said library of plans, was executed and information associated with said plan including said trigger event, patient-results information, and goals.

5. The computerized method of claim 1, wherein the determination of at least one of said trigger event, goal, or plan selection are performed by a software agent.

6. The computerized method of claim 1, wherein the determination of at least one of said trigger event, goal, or plan selection are performed using logic comprising at least one of a rules engine, a Boolean evaluation, or a rules-based determination.

7. A decision support system determining treatment for a patient having congestive heart failure and at least one other medical condition, comprising:

a library of healthcare agent solvers configured for evaluating patient information to facilitate clinical decision support for the patient;

one or more computer processors; and

one or more computer storage media storing computer-useable instructions that, when executed by the one or more processors, implement a method comprising;

receiving, by one or more computer processing components, patient information for the patient from a patient information database, wherein the patient information database is remote from at least one of the one or more computer processing components;

extract, by the one or more computer processing components, discrete patient data from the patients-results information;

receiving, by the one or more computer processing components, one or more rules from a parameters database, wherein the parameters database is remote from the one or more computer processing components, and wherein the one or more rules comprise a rule relevant to the discrete patient data;

based on the discrete patient data, determining, by a first processing component of the one or more computer processing components, two or more clinical conditions of the patient, at least one clinical condition including heart failure;

based on the two or more clinical conditions, determining, by the first processing component of the one or more computer processing components a set of solver-content parameters;

from a library of solvers, using the set of solver-content parameters, generating a finite state machine running on the one or more computer components having states and transition-rules corresponding to the patient information and solver-content parameters;

determining that a second solver is needed to assist said finite state machine to determine the conditions and a recommended treatment for the patient wherein the second solver is running on the one or more computer processing components;

invoking, by said finite state machine, a linear solver to assist said finite state machine to determine the conditions and the recommended treatment for the patient;

evaluating the finite state machine to determine the conditions and a recommended treatment for the patient, each state being evaluated using the linear solver, the finite state machine passing states to the linear solver;

based on the determined patient conditions and recommended treatment, communicating actions and dispositions specific to said patient from the one or more computer processing components to a second computer processing component, wherein the second computer processing component is remote from the one or more computer processing components; and

displaying the communicated actions and dispositions specific to said patient on a user interface of the second computer processing component.

8. The decision support system of claim 7, wherein said action or disposition is determined by invoking an expert rules engine, and wherein said determined action or disposition includes at least one of an order, patient condition, recommended treatment, recommended additional testing.

9. The decision support system of claim 8, wherein said determined action or disposition specific to said patient are processed for presentation.

10. The decision support system of claim 7, wherein said determination of an indication of heart failure is performed by a software agent.

11. The decision support system of claim 7, wherein said determination of an indication of heart failure is performed using logic comprising at least one of a rules engine, a Boolean evaluation, or a rules-based determination.

12. A method of determining treatment for a patient having congestive heart failure, the method comprising:

receiving, by one or more computer processing components, patient-results data, associated with a patient from a patient information database, wherein the patient information database is remote from at least one of the one or more computer processing components;

based on said discrete patient data, determining discrete patient data operable for use by a finite state machine solver and a mixed-integer linear solver, wherein the finite state machine solver and the mixed-integer linear solver are running on the one or more computer processing components;

determining whether said discrete patient data is indicative of heart failure;

based on said determination of whether said patient information is indicative of heart failure, accessing heart-failure content parameters operable for use by a finite state machine solver from a memory associated with a parameters database, the parameters database being remote from the one or more computer processing components;

based on said received content parameters and said discrete patient data, instantiating the finite state machine solver;

determining that the mixed-integer linear solver is needed to assist said finite state machine solver to determine patient condition or recommended treatment;

invoking, by said finite state machine solver, the mixed-integer linear solver to assist said finite state machine solver to determine patient condition or recommended treatment;

using the finite state machine solver and the mixed-integer linear solver, determining a patient condition or recommended treatment, based on said discrete patient data and said solver-content parameters, said finite state machine solver passing states to said mixed integer linear solver;

based on said determined patient condition or recommended treatment, applying a rules engine to determine an action; and

initiating the action.

13. The method of claim 12, wherein said finite state machine is instantiated such that states and transition-rules for said finite state machine are determined based on said discrete patient data and said solver-content parameters, and wherein said each state of said finite state machine is evaluated using said mixed-integer linear solver.

14. The method of claim 13, wherein said determined action or disposition is processed for presentation.

15. The method of claim 12, wherein said determination of an indication of heart failure is performed by a software agent.

16. The method of claim of 12, wherein said determination of an indication of heart failure is performed using logic comprising at least one of a rules engine, a Boolean evaluation, or a rules-based determination.

17. A system for facilitating clinical decision making, comprising:

patient information related to a patient;

solver-content parameters operable for use to instantiate one or more solvers, from a library of solvers, for use to determine at least one condition or recommended treatment associated with said patient;

a data-extraction solver operable to determine discrete patient data from said patient information;

a condition and recommendation resolver agent (“

resolver agent”

) operable to receive said discrete patient data and said solver-content parameters, and based on said discrete data and solver-content parameters, determine at least one condition or recommended treatment associated with said patient; and

an expert rules engine operable for receiving at least one condition or recommended treatment and determining an action or disposition based on said at least one condition or recommended treatment;

(1) said resolver agent further comprising a generic finite state machine and a mixed-integer linear solver;

(2) said resolver agent being further operable to instantiate and execute a patient-specific finite state machine, based on said generic finite state machine, having states and transition-rules corresponding to said discrete patient data and said solver-content parameters, the patient-specific finite state machine determining that a second solver is needed to assist said first patient-specific finite state machine and invoking the mixed-integer linear solver to evaluate each state of the patient-specific finite state machine to determine the at least one condition or recommended treatment; and

(3) said resolver agent being further operable to use said mixed integer linear solver to evaluate each state of the finite state machine.

18. The system of claim 17, wherein the generic finite state machine returns an actual state for each clinical condition of the patient, the actual states being passed to the mixed integer linear solver as parameters, to apply the mixed integer solver based on a clinical state, and content tables.