Method and system for reducing a data set

• US 7,805,421 B2
• Filed: 11/02/2007
• Issued: 09/28/2010
• Est. Priority Date: 11/02/2007
• Status: Expired due to Fees

• Current Assignee: Caterpillar Incorporated
• Original Assignee: Caterpillar Incorporated
• Inventors: Taylor, Michael Lee, Nikolopoulos, Christos, Hart, Gabriel Carl, Mason, James Robert, Cler, Meredith Jaye, Grichnik, Anthony J.
• Primary Examiner(s): Fleurantin; Jean B.
• Assistant Examiner(s): Nguyen; Phong
• Application Number:
  US11/979,409
• Publication Number:
  US 20090119323A1
• Time in Patent Office: 1,061 Days
• Field of Search: 707/101, 707/696, 707/802, 707/E17.002, 705/2, 705/3
• US Class Current: 707/696
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

System and methods are provided for applying a dimensionality reduction process to a data set. The method includes obtaining a data set including a first set of variables and identifying collections of the first set of variables for replacement by a second set of variables. The method also includes replacing the collections of the first set of variables with the second set of variables and eliminating the first set of variables from the data set.

Claims

1. A computer-implemented method for applying a dimensionality reduction process to a medical data set, comprising:

obtaining a data set stored in a medical database, the data set including a set of medical data entries relating to medical events for individuals in a population;

identifying a collection of the set of medical data entries relating to a common medical event for an individual in the population for replacement by a representative medical data entry for the common medical event, the identifying including identifying medical data entries of the set that are associated with medical events relating to the common medical event and occurring within a predetermined period of time; and

reducing the size of the medical database by;

replacing the identified collection of medical data entries with the representative medical data entry for the common medical event in the medical database, the representative medical data entry including a numerical indicator representing a number of the medical data entries in the identified collection; and

eliminating the identified collection of medical data entries from the medical database.

2. The method of claim 1, wherein:

the set of medical data entries includes medical codes and dates of occurrence of the events related to the medical data entries; and

the medical events include at least one of a diagnosis of an individual and creating a prescription for an individual.

3. The method of claim 1, wherein:

the identified collection of medical data entries represents a diagnosis of the individual for one or more diseases; and

if a value of the numerical indicator is below a threshold, the individual is diagnosed as not having the disease.

4. The method of claim 1, wherein;

the set of medical data entries includes medical codes; and

replacing includes combining the medical codes associated with the identified collection of medical data entries into a string.

5. A computer-readable medium comprising instructions which, when executed by a processor, perform a method for applying a dimensionality reduction process to a medical data set, the method comprising:

obtaining a data set stored in a medical database, the data set including a set of medical data entries relating to medical events for individuals in a population;

identifying a collection of the set of medical data entries relating to a common medical event for an individual in the population for replacement by a single medical data entry representative of at least some of the medical data entries relating to the common event, the identifying including identifying medical data entries of the set that are associated with medical events relating to the common medical event and occurring within a predetermined period of time; and

reducing the size of the medical database by;

replacing the identified collection of medical data entries with the representative medical data entry for the common medical event in the medical database, the representative medical data entry including a numerical indicator representing a number of the identified collection of medical data entries; and

eliminating the identified collection of medical data entries from the medical database.

6. The computer-readable medium of claim 5, wherein:

the set of medical data entries includes medical codes and dates of occurrence of the events related to the medical data entries; and

the medical events include at least one of a diagnosis of an individual and creating a prescription for an individual.

7. The computer-readable medium of claim 5, wherein:

the identified collection of medical data entries represents a diagnosis of the individual for one or more diseases; and

if a value of the numerical indicator is below a threshold, the individual is diagnosed as not having the disease.

8. The computer-readable medium of claim 5, wherein:

the set of medical data entries includes medical codes; and

replacing includes combining the medical codes associated with the identified collection of medical data entries into a string.

9. The computer-readable medium of claim 5, wherein the method further includes using the data set to diagnose a patient for one or more diseases.

10. A system for applying a dimensionality reduction process to a data set, the system comprising:

a medical database storing a data set including medical data entries relating to medical events for individuals in a population;

at least one input device; and

a central processing unit in communication with the medical database and the at least one input device, wherein the central processing unit;

obtains the data set from the medical database;

identifies a collection of the set of medical data entries relating to a common medical event for an individual in the population for replacement by a representative medical data entry for the common medical event, the identifying including identifying medical data entries of the set that are associated with medical events relating to the common medical event and occurring within a predetermined period of time; and

reduces the size of the medical database by;

replacing the identified collection of medical data entries with the representative medical data entry for the common medical event in the medical database, the representative medical data entry including a numerical indicator representing a number of the identified collection of medical data entries; and

deleting at least some of the identified collection of medical data entries from the medical database.

11. The system of claim 10, wherein:

the set of medical data entries includes medical codes and dates of occurrence of the medical events related to the medical data entries; and

the medical events include at least one of a diagnosis of an individual and creating a prescription for an individual.

12. The system of claim 10, wherein:

the identified collection of medical data entries represents a diagnosis of the individual for one or more diseases; and

if a value of the numerical indicator is below a threshold, the central processing unit is configured to diagnose the individual as not having the disease, and to update the medical database based on the diagnosis.

13. The system of claim 10, wherein:

the set of medical data entries includes medical codes; and

the central processing unit replaces the identified collection of medical data entries by combining the medical codes associated with the identified collection of medical data entries into a string.