Computer architecture and process of patient generation, evolution, and simulation for computer based testing system

• US 7,653,556 B2
• Filed: 09/08/2005
• Issued: 01/26/2010
• Est. Priority Date: 10/30/1996
• Status: Expired due to Fees

• Current Assignee: American Board of Family Medicine, Inc.
• Original Assignee: American Board of Family Medicine, Inc.
• Inventors: Truszczynski, Miroslav, Marek, Victor W., Sumner, Walton II, Rovinelli, Richard J.
• Primary Examiner(s): O'Connor; Gerald J.
• Assistant Examiner(s): Najarian; Lena
• Application Number:
  US11/220,638
• Publication Number:
  US 20060004608A1
• Time in Patent Office: 1,601 Days
• Field of Search: 705/2, 705/3, 600/300
• US Class Current: 705/2
• CPC Class Codes: G09B 23/28 for medicine
• CPC Class Codes: G09B 7/02 of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student
• 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

Abstract

A computer implemented simulation and evaluation method simulates interventions to a patient by a user, and evaluates the interventions responsive to predetermined criteria and the interventions. The method includes defining a test area to evaluate the user to at least one of predetermined criteria and a user profile, selecting genetic information of the patient responsive to the test area, and generating a patient history responsive to the test area and the genetic information. The method also includes receiving at least one intervention input by the user, and evaluating the user responsive to the intervention and predetermined criteria.

Claims

1. A computer system for training a user to treat a simulated patient having at least one present health state, comprising:

(a) a computer generating the simulated patient and a history for the simulated patient, wherein said generating comprises evolving the simulated patient backwards in time to generate a progression of health states and risk factors traversed by the simulated patient from a specified health condition to a normal health condition; and

(b) the computer iterating through a prescribed treatment protocol prescribed by said user, wherein said iterating comprises evolving the simulated patient forward from at least one present health state to at least one subsequent health state responsive to at least one of time and at least one intervention input by the user, and using parallel networks of health states to model transitions that occur among a set of mutually exclusive health conditions in at least one body part comprising at least one of the following interactions;

(1) independent interaction between parallel networks so that patient evolution between first and second parallel networks are unrelated to each other;

(2) unilateral interaction between the parallel networks so that patient evolution on a first parallel network is unrelated to patient evolution on a second parallel network, and patient evolution on the second parallel network is related to the patient evolution on the first parallel network; and

(3) mutually dependent interaction between the parallel networks so that patient evolution between the first and second parallel networks are related to each other.

2. A computer system according to claim 1, further comprising:

(c) accessing a profile for said user;

(d) selecting a training area to evaluate said user based on at least the user'"'"'s profile;

(e) dynamically generating multiple instances of patients responsive to said training area.

3. A computer system according to claim 1, further comprising:

(c) evolving the patient to the at least one subsequent health state responsive to the at least one intervention; and

(d) evaluating the user responsive to the at least one intervention input by the user.

4. A computer system according to claim 1, further comprising:

(c) evolving the patient to the at least one subsequent health state responsive to the at least one intervention and the patient history; and

(d) evaluating the user responsive to at least one of the at least one intervention input by the user, and the at least one subsequent health state.

5. A computer system according to claim 1, further comprising:

(c) evolving the patient to the at least one subsequent health state responsive to the at least one intervention and the patient history;

(d) receiving at least one other intervention input by the user;

(e) evolving the patient responsive to the at least one other intervention from the at least one subsequent health state to at least one other subsequent health state; and

(I) evaluating the user responsive to at least one of the at least one other intervention, the at least one subsequent health state, and the at least one other subsequent health state.

6. A computer system according to claim 1, wherein the method uses an entity relationship model.

7. A computer system according to claim 6, wherein the entity relationship model comprises population, record, agents of change, health states, findings and courses of action.

8. A computer system according to claim 7, wherein the findings include specific findings, patterns and sub-patterns describing patient behaviors and characteristics.

9. A computer system according to claim 8, wherein the patterns describe one or more features over time.

10. A computer system according to claim 8, wherein the sub-patterns describe consequences of patient related events.

11. A computer system according to claim 8, wherein the patterns model time and characterize interrelated medical observations.

12. A computer system according to claim 8, further comprising performing a differential diagnosis responsive to the findings, the patterns and the sub-patterns.

13. A computer system according to claim 8, wherein confidence in a presence of the patterns increases with passage of time.

14. A computer system according to claim 7, wherein the courses of action describe tasks and methods used to apply, modify, and evaluate health state information and characteristics described in the entity relationship model.

15. A computer system according to claim 7, wherein the courses of action describe patient activities, including at least one of medical and non-medical activities.

16. A computer system according to claim 7, wherein the courses of action describe potential interventions input by the user including at least one of diagnostic and management strategies.

17. A computer system according to claim 7, wherein the courses of action comprise one or more elementary courses of action used to construct at least one course of action, one or more types of elementary courses of action corresponding to the one or more elementary courses of action, and weighting factors corresponding to the one or more elementary courses of action.

18. A computer system according to claim 6, further comprising evolving the patient responsive to the at least one intervention, entity relations and the patient history to at least one subsequent health state.

19. A computer system according to claim 6, wherein the at least one intervention by the user is considered by the entity relationship model in evolving the patient from a first health state to a subsequent health state.

20. A computer system according to claim 6, wherein the entity relationship model includes one or more of the following relations between entities:

Population Contacts Population

Population Related to Population

Population Interacts with Courses of Action

Population Exposed to Agents of Change

Population Has Health States

Population Exhibits Findings

Agents of Change Cause Health States

Health States Lead to Health States

Findings Associated with Health States

Findings Link to Findings

Course of Action use Agents of Change

Courses of Action Identify Agents of Change

Courses of Action Treat Health States

Courses of Action Alter Findings

Courses of Action Reveal Findings

Courses of Action Evaluate Findings.

21. A computer system according to claim 8, wherein the entity relationship model links the findings with the patterns to a health state, rather than linking a range of finding values to the health state.

22. A computer system according to claim 8, wherein the patterns include sensitivity and specificity represented as age dependent, rather than as constants.

23. A computer system according to claim 1, wherein said generating the patient history is executed once for each simulation to generate the patient history used in said computer implemented simulation and evaluation method.

24. A computer system according to claim 3, further compromising repeating said evolving, and said evaluating a plurality of times.

25. A computer system according to claim 1, further comprising generating the patient history using a Monte Carlo process to generate a plurality of potential patient histories.

26. A computer system according to claim 6, wherein the entity relationship model utilizes tree structures to describe a probability density function conditioned on comorbidities, treatments, risk factors, and the interventions.

27. A computer system according to claim 6, wherein the entity relationship model includes diagnostic complexities and disease interactions.

28. A computer system according to claim 1, wherein at least one of the parallel networks comprises a single health state.

29. A computer system according to claim 1, wherein the parallel networks of health states describe at least one of a chronic condition and non-chronic condition.

30. A computer system according to claim 29, wherein the non-chronic condition includes acute exacerbations describing acute flares of illness that occur during a more chronic health condition.

31. A computer system according to claim 1, wherein at least one of the parallel networks comprises at least one health state.

32. A computer system according to claim 3, further comprising repeating said evolving responsive to:

(1) parallel health states of the patient; and

(2) a target health state and health state combinations that lead to different parallel states.

33. A computer system according to claim 1, wherein the parallel networks of health states comprise:

(1) a primary network including primary health conditions defining a health domain;

(2) a risk factor network including risk factors for progression through the primary network; and

(3) complications attributed to treating the primary health conditions in the primary network.

34. A computer system according to claim 33, wherein the parallel networks of health states are generated using the following information:

(1) how long at least one of the risk factors exists before influencing a transition between primary health conditions in the primary network;

(2) time required for transitions in the primary network, considering different combinations of the risk factors; and

(3) number of transitions the patient is allowed to make between a specified health state and a normal health state.

35. A computer system according to claim 6, wherein the entity relationship model is based on knowledge data, further comprising acquiring the knowledge data by at least one of questioning domain experts, access to scientific literature, and protocol analysis.

36. A computer system according to claim 6, wherein patterns, having a shape and reflecting probability distributions, describe at least one entity or relationship, and wherein said evolving is responsive to at least one of the patterns.

37. A computer system according to claim 36, wherein the patterns have at least one of cycles, sub-patterns and sub-sub-patterns, and wherein said evolving is responsive to at least one of the cycles, sub-patterns and sub-sub-patterns.

38. A computer system according to claim 7, wherein each health state is part of a network of related and mutually exclusive health states, and a plurality of health state networks exist in parallel, further comprising defining the parallel networks as active or inactive, and defining the active networks as primary or comorbid.

39. A computer system according to claim 38, further comprising adding one or more new states to the network using one or more irreversible interventions.

40. A computer system according to claim 38, further comprising defining an overall health of a patient by a vector indicating the patient'"'"'s current health state in each network.

41. A computer system according to claim 1, further comprising:

accessing a profile of the user;

defining the at least one present health state and the history for the patient responsive to the user profile including prior performance of the user;

generating the patient and the history for the patient responsive to the at least one present health state and the user profile; and

iterating through the prescribed treatment protocol with said user for the simulated patient, and evolving the patient from the at least one present health state to at least one subsequent health state responsive to at least one of time and at least one intervention input by the user.

42. A computer system according to claim 41, further comprising generating an initial patient history state in response to the profile of the user who is being evaluated on the basis of the at least one intervention input by the user.

43. A computer system according to claim 1, wherein said generating the history further comprises generating the history by determining at least one precursor health state for the at least one present health state and one or more risk factors.

44. A computer system according to claim 43, wherein said determining the at least one precursor health state further comprises determining one or more risk factors.

45. A computer system according to claim 43, further comprising repeating said determining iteratively until a substantially normal health state is reached.

46. A computer system according to claim 1, wherein said generating the history for the patient further comprises establishing for the patient at least one of an age, sex, racial/ethnic origin, and age of onset of the present health state, and wherein said evolving is responsive to the patient history.

47. A computer system according to claim 1, wherein said evolving further comprises creating a list of possible subsequent health states;

identifying all courses of action relevant to each of the possible subsequent health states; and

selecting an identified course of action closest to the at least one intervention input by the user.

48. A computer system for training a user to treat a simulated patient having at least one present health state, comprising:

a computer generating the simulated patient and a history for the simulated patient, wherein said generating comprises evolving the simulated patient backwards in time to generate health states traversed by the simulated patient from a specified health condition to a normal health condition; and

the computer iterating through a prescribed treatment protocol prescribed by said user, wherein said iterating comprises evolving the simulated patient from at least one present health state to at least one subsequent health state using at least one of the responsive to at least one of time and at least one intervention input by the user, and using at least one of the first and second parallel networks to model transitions that occur among a set of mutually exclusive health conditions in at least one body part comprising at least one of the following interactions;

(1) independent interaction between parallel networks so that patient evolution between first and second parallel networks are unrelated to each other;

(2) unilateral interaction between the parallel networks so that patient evolution on a first parallel network is unrelated to patient evolution on a second parallel network, and patient evolution on the second parallel network is related to the patient evolution on the first parallel network; and

(3) mutually dependent interaction between the parallel networks so that patient evolution between the first and second parallel networks are related to each other.

51. A computer system according to claim 48, 49, or 50, further comprising:

accessing at least one profile for said user;

selecting a training area to evaluate said user based on at least the user'"'"'s profile;

dynamically generating multiple instances of patients responsive to said training area.

52. A computer system according to claim 48, 49, or 50, wherein the method uses an entity relationship model.

53. A computer system according to claim 52, wherein patterns model time and characterize interrelated medical observations.

54. A computer system according to claim 52, further comprising evolving the patient responsive to the at least one intervention, entity relations defined by the entity relationship model and the patient history to at least one subsequent health state.

55. A computer system according to claim 53, wherein the entity relationship model links the findings with the patterns to a health state, rather than linking a range of finding values to the health state.

56. A computer system according to claim 48, 49, or 50, wherein said generating the patient history is executed for each simulation to generate the patient history used in said computer implemented simulation and evaluation method.

57. A computer system according to claim 48, 49, or 50, further comprising generating the patient history comprising a progression of health states and risk factors traversed by the patient from a normal health condition to a specified health condition.

58. A computer system according to claim 48, 49, or 50, further comprising generating the patient history backwards in time from a specified health condition to a normal health condition including successive precursor health states and onset times therebetween.

59. A computer system according to claim 48, 49, or 50, wherein the networks of health states are parallel networks used to model transitions that occur among a set of health conditions responsive to the at least one intervention by the user.

60. A computer system according to claim 59, further comprising repeating said evolving responsive to:

(1) parallel health states of the patient; and

(2) a target health state and health state combinations that lead to different parallel states.

61. A computer system according to claim 59, wherein the parallel networks of health states comprise:

(1) a primary network including primary health conditions defining a health domain;

(2) a risk factor network including risk factors for progression through the primary network; and

(3) complications attributed to treating the primary health conditions in the primary network.

62. A computer system according to claim 52, wherein patterns, having a shape and reflecting probability distributions, describe at least one entity or relationship, and wherein said evolving is responsive to at least one of the patterns.

63. A computer system according to claim 48, 49, or 50, further comprising generating an initial patient history state in response to the profile of the user who is being evaluated on the basis of the at least one intervention input by the user.

64. A computer system according to claim 48, 49, or 50, wherein said generating the history further comprises generating the history by determining at least one precursor health state for the at least one present health state and one or more risk factors.

65. A computer system according to claim 48, 49, or 50, wherein said evolving further comprises creating a list of possible subsequent health states;

identifying courses of action relevant to each of the possible subsequent health states; and

selecting an identified course of action closest to the at least one intervention input by the user.

49. A computer system for training a user to treat a simulated patient having at least one present health state, comprising:

a computer generating the simulated patient and a history for the simulated patient and evolving the simulated patient backwards in time to generate health states traversed by the simulated patient from a specified health condition to a normal health condition using a parallel network model comprising at least one health state, wherein the at least one health state of the simulated patient at a given time is described as part of a different network comprising a plurality of related and mutually exclusive health states and transitions therebetween, and wherein a plurality of networks exist in parallel; and

the computer iterating through a prescribed treatment protocol prescribed by said user, wherein said iterating comprises evolving the simulated patient from at least one present health state to at least one subsequent health state responsive to at least one of time and at least one intervention input by the user, and using at least one of first and second parallel networks to model transitions that occur among a set of mutually exclusive health conditions in at least one body part comprising at least one of the following interactions;

(1) independent interaction between parallel networks so that patient evolution between first and second parallel networks are unrelated to each other;

(2) unilateral interaction between the parallel networks so that patient evolution on a first parallel network is unrelated to patient evolution on a second parallel network, and patient evolution on the second parallel network is related to the patient evolution on the first parallel network; and

(3) mutually dependent interaction between the parallel networks so that patient evolution between the first and second parallel networks are related to each other.

50. A computer system for training a user to treat a simulated patient having at least one present health state, comprising:

a computer generating the simulated patient and a history for the simulated patient responsive to evolving the simulated patient backwards in time to generate health states traversed by the simulated patient from a specified health condition to a normal health condition using a parallel network model comprising at least one health state; and

the computer iterating through a prescribed treatment protocol prescribed by said user, wherein said iterating comprises evolving the simulated patient from at least one present health state to at least one subsequent health state using at least one of first and second parallel networks responsive to at least one of time and at least one intervention input by the user, and using at least one of the first and second parallel networks to model transitions that occur among a set of mutually exclusive health conditions in at least one body part comprising at least one of the following interactions;

(1) independent interaction between parallel networks so that patient evolution between first and second parallel networks are unrelated to each other;

(2) unilateral interaction between the parallel networks so that patient evolution on a first parallel network is unrelated to patient evolution on a second parallel network, and patient evolution on the second parallel network is related to the patient evolution on the first parallel network; and

(3) mutually dependent interaction between the parallel networks so that patient evolution between the first and second parallel networks are related to each other.