REAL-TIME PREDICTIVE COMPUTER PROGRAM, MODEL, AND METHOD

• US 20080097943A1
• Filed: 02/26/2007
• Published: 04/24/2008
• Est. Priority Date: 10/23/2006
• Status: Active Grant

• Current Assignee: Health Care Information Services, LLC
• Original Assignee: Health Care Information Services, LLC
• Inventors: Kelly, Bruce, McDannald, E.R.
• Granted Patent: US 7,707,130 B2
• Time in Patent Office: Days
• US Class Current: 706/21
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A method for predicting a future occurrence of an event involves obtaining a history of prior occurrences of the event. A plurality of variables is created that are associated with the event. Weights are assigned to each variable. An artificial neural network is accessed and trained with the history of past occurrences of the event by comparing an output of the artificial neural network to the past occurrence of the event. The weights are adjusted until the output corresponds to the past occurrence of the event.

Claims

1. A method of predicting a future occurrence of an event, the method comprising the steps of:

obtaining a history of past occurrences of the event;

accessing an artificial neural network; and

training the artificial neural network with the history of past occurrences of the event to predict the future occurrence of the event.

2. The method of claim 1, wherein the method further includes creating a plurality of variables associated with the event.

3. The method of claim 2, wherein the method further includes assigning a weight to each variable.

4. The method of claim 3, wherein the method further includes adjusting the weights of the variables.

5. The method of claim 2, wherein the method further includes applying the variables to the artificial neural network.

6. The method of claim 1, wherein the method further includes comparing an output of the network with the past occurrence of the event.

7. The method of claim 1, wherein the number of occurrences of the event is a minimum of twenty-five.

8. A method of predicting a worker'"'"'s compensation injury, the method comprising the steps of:

obtaining a history of injuries that required worker'"'"'s compensation;

accessing an artificial neural network; and

training the artificial neural network with the history of worker'"'"'s compensation injuries to predict the occurrence of the next worker'"'"'s compensation injury.

9. The method of claim 8, wherein the method further includes creating a plurality of variables associated with worker'"'"'s compensation injuries.

10. The method of claim 9, wherein the variables are selected from the group consisting of pre-morbidity condition, post-event condition, post-medical condition, hospitalization, support at home, availability of home care, availability of outpatient services, suitability of individual for his job, availability of light-duty work, availability of job retraining, potential for litigation, the state in which the injury occurred, and miscellaneous factors.

11. The method of claim 9, wherein the method further includes assigning a weight to each variable.

12. The method of claim 11, wherein the method further includes adjusting the weights of the variables.

13. The method of claim 8, wherein the method further includes comparing an output of the network with a past occurrence of the worker'"'"'s compensation injury.

14. A method of predicting an acute medical situation, the method comprising the steps of:

obtaining a history of acute medical situations;

accessing an artificial neural network; and

training the artificial neural network with the history of acute medical situations to predict the occurrence of the next acute medical situation.

15. The method of claim 14, wherein the method further includes creating a plurality of variables associated with acute medical situations.

16. The method of claim 15, wherein the variables are selected from the group consisting of pre-morbidity condition, post-event status for the second day, post-event status for the third day, post-event status for the fourth day, post-event status for the fifth day, co-morbidities, medications, X-rays and diagnostic tests, laboratory tests, physical therapy, occupational therapy, activities of daily living, surgeries, support at home, availability of home care, and miscellaneous factors.

17. The method of claim 16, wherein the method further includes assigning a weight to each variable.

18. The method of claim 17, wherein the method further includes adjusting the weights of the variables.

19. The method of claim 14, wherein the method further includes comparing an output of the network with a past occurrence of the acute medical situation.

20. A method of predicting a future occurrence of an event, the method comprising the steps of:

obtaining a history of past occurrences of the event;

creating a plurality of variables associated with the event;

assigning a weight to each variable;

creating an artificial neural network; and

training the artificial neural network with the history of past occurrences of the event to predict the future occurrence of the event, including the steps—

applying the variables to the artificial neural network for each past occurrence of the event,

comparing an output of the network with each past occurrence of the event, and

adjusting the weights of the variables such that the output of the network corresponds to each past occurrence of the event.

21. The method of claim 20, wherein the event is selected from the group consisting of worker'"'"'s compensation injuries and acute medical situations.