System and method for performing fuzzy cluster classification of stress tests

• US 5,788,640 A
• Filed: 10/26/1995
• Issued: 08/04/1998
• Est. Priority Date: 10/26/1995
• Status: Expired due to Fees

• Current Assignee: Robert Mitchell Peters
• Original Assignee: Robert Mitchell Peters
• Inventors: Peters, Robert Mitchell
• Primary Examiner(s): Bahr, Jennifer
• Assistant Examiner(s): YARNELL, BRYAN KEITH
• Application Number:
  US08/548,734
• Time in Patent Office: 1,013 Days
• Field of Search: 128/630, 128/668, 128/670, 128/700, 128/920, 128/923, 128/924, 364/413.01, 364/413.02, 395/22, 395/51, 395/61, 395/900, 395/924, 600/300, 600/481, 600/483, 600/513
• US Class Current: 600/483
• CPC Class Codes: G06N 7/02 using fuzzy logic computing arrangements based on biological models G06N3/00; computing arrangements using knowledge-based models G06N5/00
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics
• CPC Class Codes: Y10S 706/90 Fuzzy logic
• CPC Class Codes: Y10S 706/924 Medical

Abstract

A stress test classifying system and method classify current stress test data using a processor for comparing the current stress test with previous stress test data grouped in fuzzy sets and for generating a classification of the current stress test data with respect to the fuzzy sets.

Claims

1. A stress test classifying system for classifying current stress test data comprising:

a processor for comparing the current stress test with previous stress test data grouped in fuzzy sets and for generating a classification of the current stress test data with respect to the fuzzy sets, the processor including;

a fuzzy classifier for generating a respective similarity measure for each fuzzy set using predetermined fuzzy set definitions and the current stress test data, and for generating a crisp value from the similarity measures to determine the classification.

2. The stress test classifying system of claim 1 further comprising:

a memory for storing the previous stress test data and the predetermined fuzzy set definitions.

3. The stress test classifying system of claim 1 further comprising:

a fuzzy classification program executable by the processor for implementing the fuzzy classifier.

4. The stress test classifying system of claim 3 wherein the fuzzy classifier includes:

means for generating the similarity measures from a predetermined metric applied to the current stress test data and the predetermined fuzzy set definitions.

5. The stress test classifying system of claim 3 wherein the fuzzy classifier includes:

means for generating a crisp value from a minimum value of the similarity measures.

6. The stress test classifying system of claim 4 wherein the means for generating the similarity measures applies a weighted Euclidean metric to the current stress test data, the predetermined fuzzy set definitions, and a plurality of weights associated with each fuzzy set definition.

7. The stress test classifying system of claim 1 wherein the processor generates the classification of the current stress test with respect to predetermined classes defined by fuzzy sets and with respect to a plurality of stress test variables.

8. The stress test classifying system of claim 7 wherein the plurality of stress test variables include six predetermined stress test variables.

9. The stress test classifying system of claim 1 wherein the processor includes:

a statistics generator for generating statistics of the previous stress test data grouped in the fuzzy sets to facilitate refinement of the fuzzy sets.

10. A stress test classifying system for classifying current stress test data comprising:

a memory for storing previous stress test data in a database, fuzzy set definitions grouping the previous stress test data in a plurality of fuzzy sets, and a plurality of weights associated with each fuzzy set and with a respective stress test variable;

a processor for comparing the current stress test with the previous stress test data, the processor including a fuzzy classifier for generating a crisp classification of the current stress test data with respect to the fuzzy sets using similarity measures for each respective fuzzy set with respect to each stress test variable using the predetermined fuzzy set definitions, the current stress test data, and the weights, wherein each similarity measure indicates a degree of similarity of the current stress test data to the respective fuzzy set.

11. The stress test classifying system of claim 10 wherein the fuzzy classifier includes:

means for generating the similarity measures from a multi-variable weighted Euclidean metric applied to the current stress test data and the predetermined fuzzy set definitions, wherein the multi-variable weighted Euclidean metric is weighted using the weights.

12. The stress test classifying system of claim 10 wherein the fuzzy classifier includes:

means for determining a minimum value of the similarity measures; and

wherein the minimum value, as a crisp value, determines the classification of the current stress test with respect to the fuzzy sets and with respect to the plurality of stress test variables.

13. The stress test classifying system of claim 12 wherein the memory stores a plurality of labels, each associated with a respective fuzzy set;

and

wherein the processor, responsive to the minimum value, outputs a first label of the fuzzy set corresponding to the minimum value, wherein the first label indicates the classification of the current stress test data.

14. The stress test classifying system of claim 10 wherein the plurality of stress test variables include the group consisting of ST segment change, a difference between resting systolic and peak exercise systolic blood pressure, total treadmill time, peak exercise heart rate as a percentage of 100% predicted maximum for age, time to onset of angina, and duration of repolarization abnormalities.

15. The stress test classifying system of claim 10 wherein the processor includes:

a statistics generator for generating statistics of the previous stress test data grouped in the fuzzy sets to facilitate refinement of the fuzzy sets by the user using an input device.

16. A method for classifying current stress test data comprising the steps of:

(a) storing, in a memory, fuzzy set definitions grouping the previous stress test data in a plurality of fuzzy sets;

(b) storing, in the memory, a plurality of weights associated with each fuzzy set and with a respective stress test variable;

(c) operating a processor to generate a plurality of similarity measures for each respective fuzzy set with respect to each stress test variable using the predetermined fuzzy set definitions; and

(d) operating a processor to generate a crisp value as the classification of the current stress test data with respect to the fuzzy sets using the plurality of similarity measures.

17. The method of claim 16 wherein the step (c) of operating the processor includes the steps of:

(c1) processing the current stress test data and the predetermined fuzzy set definitions to generate processed values;

(c2) weighting the processed values using a plurality of weights; and

(c3) generating the similarity measures from a multi-variable Euclidean metric applied to the processed values.

18. The method of claim 16 wherein the step (c) of operating the processor includes the steps of:

(c1) determining a minimum value of the similarity measures; and

wherein the step (d) of operating the processor including the steps of;

(d1) determining the classification of the current stress test with respect to the fuzzy sets and with respect to the plurality of stress test variables using the minimum value.

19. The method of claim 16 wherein step (d) further includes the steps of:

(d1) retrieving from the memory a first label associated with a first fuzzy set corresponding to the minimum value; and

(d2) outputting the first label to indicate the classification of the current stress test data.

20. The method of claim 16 further including, before the step (c), the steps of:

(e) executing a fuzzy classification program using the processor to operate the processor to perform steps (c) and (d).