Multi-dimensional dynamical analysis

• US 20060287607A1
• Filed: 01/26/2006
• Published: 12/21/2006
• Est. Priority Date: 08/27/2002
• Status: Abandoned Application

• Current Assignee: University of Florida Research Foundation, Incorporated (State University System of Florida), Arizona Board of Regents (University of Arizona)
• Original Assignee: University of Florida Research Foundation, Incorporated (State University System of Florida), Arizona Board of Regents (University of Arizona)
• Inventors: Dance, Linda K., Iasemidis, Leonidas D., Shiau, Deng-Shan, Sackellares, James Chris, Chaovalitwongse, Wanpracha A., Pardalos, Panos M.
• Application Number:
  US11/339,606
• Publication Number:
  US 20060287607A1
• Time in Patent Office: Days
• US Class Current: 600/544
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G06F 2218/22 Source localisation; Inverse modelling

Abstract

Techniques for investigating dynamical behavior of complex systems for monitoring, diagnosing, and predicting future behavior and trends are presented. A method includes generating a multi-dimensional representation for each signal acquired from corresponding channels of a multi-dimensional system, and generating dynamical profiles for each channel in accordance with one of multiple dynamical metrics. The method includes calculating multiple first statistical measures for a group of the channels that reflect a level of interaction among the channel group associated with at least one of the metrics. The method includes calculating in an initialization period a second statistical measure for each of the first statistical measures that reflect the association of the first statistical measures with related occurrences under investigation. The method includes selecting in the initialization period at least one of the metrics based on the second statistical measures, and identifying the first statistical measures corresponding to the selected metrics to characterize dynamical behavior.

Claims

1. A method of monitoring dynamical behavior of a multi-dimensional system comprising:

generating a multi-dimensional representation for each of a plurality of signals acquired from corresponding channels of the multi-dimensional system;

generating a plurality of dynamical profiles for each channel based on the corresponding multi-dimensional representations, wherein each dynamical profile reflects dynamical characteristics of the associated channel in accordance with one of a plurality of dynamical metrics;

calculating a plurality of first statistical measures for a group of a plurality of the channels, wherein each first statistical measure reflects a level of interaction among the channel group associated with at least one of the plurality of dynamical metrics;

calculating in an initialization period a second statistical measure for each of the first statistical measures, wherein the second statistical measure reflects a level of association of the first statistical measure with related occurrences being monitored;

selecting in the initialization period at least one of the plurality of dynamical metrics based on the second statistical measures calculated for each of the plurality of first statistical measures; and

identifying in the initialization period the first statistical measures corresponding to the selected dynamical metrics to characterize the dynamical behavior of the multi-dimensional system.

2. The method of claim 1, further comprising analyzing a set of the dynamical metrics each with a set of parameters identified by user input.

3. The method of claim 2, wherein the set of the dynamical metrics includes at least one of a maximum short-term Lyapunov exponent (STL_MAX), a rate of change in angular frequency (Ω

_MAX), an approximate entropy (ApEn), a pattern-match approximate entropy (PM-ApEn), a Pesin'"'"'s identity (h_μ

), and a Lyapunov dimension (d_L).

4. The method of claim 1, further comprising implementing a statistical test identified by user input for calculating the first statistical measures.

5. The method of claim 4, wherein the statistical test includes a nonparametric Analysis of Variance (ANOVA) test.

6. The method of claim 1, wherein the selecting step comprises selecting at least one of the plurality of dynamical metrics corresponding to the most robust first statistical measure during the initialization period based on the second statistical measures.

7. The method of claim 6, wherein the selecting step comprises selecting at least one of the plurality of dynamical metrics corresponding to the first statistical measure having a desired level of sensitivity and specificity during the initialization period based on the second statistical measures.

8. The method of claim 1, wherein the identifying step comprises identifying in the initialization period the first statistical measures corresponding to the selected dynamical metrics to predict at least one occurrence.

9. The method of claim 8, further comprising monitoring the identified first statistical measures for a condition indicative of the at least one occurrence.

10. The method of claim 9, further comprising comparing the identified first statistical measures to at least one threshold value.

11. The method of claim 9, further comprising implementing at least one of a warning and an intervention in response to detecting the condition indicative of the at least one occurrence.

12. The method of claim 1, further comprising:

repeating the calculating, selecting, and identifying steps in the initialization period for additional channel groups; and

selecting a particular channel group based on a comparison of the second statistical measures for the respective channel groups.

13. A computer readable medium having stored therein a program, which when executed causes a processor to perform the following functions for monitoring dynamical behavior of a multi-dimensional system:

generate a multi-dimensional representation for each of a plurality of signals acquired from corresponding channels of the multi-dimensional system;

generate a plurality of dynamical profiles for each channel based on the corresponding multi-dimensional representations, wherein each dynamical profile reflects dynamical characteristics of the associated channel in accordance with one of a plurality of dynamical metrics;

calculate a plurality of first statistical measures for a group of a plurality of the channels, wherein each first statistical measure reflects a level of interaction among the channel group associated with at least one of the plurality of dynamical metrics;

calculate in an initialization period a second statistical measure for each of the first statistical measures, wherein the second statistical measure reflects a level of association of the first statistical measure with related occurrences being monitored;

select in the initialization period at least one of the plurality of dynamical metrics based on the second statistical measures calculated for each of the plurality of first statistical measures; and

identify in the initialization period the first statistical measures corresponding to the selected dynamical metrics to characterize the dynamical behavior of the multi-dimensional system.

14. The computer readable medium of claim 13, wherein the program further causes the processor to analyze a set of the dynamical metrics each with a set of parameters identified by user input.

15. The computer readable medium of claim 14, wherein the set of the dynamical metrics includes at least one of a maximum short-term Lyapunov exponent (STL_MAX), a rate of change in angular frequency (Ω

_MAX), an approximate entropy (ApEn), a pattern-match approximate entropy (PM-ApEn), a Pesin'"'"'s identity (h_μ

), and a Lyapunov dimension (d_L).

16. The computer readable medium of claim 13, wherein the program further causes the processor to implement a statistical test identified by user input for calculating the first statistical measures.

17. The computer readable medium of claim 16, wherein the statistical test includes a nonparametric Analysis of Variance (ANOVA) test.

18. The computer readable medium of claim 13, wherein the program further causes the processor to select at least one of the plurality of dynamical metrics corresponding to the most robust first statistical measure during the initialization period based on the second statistical measures.

19. The computer readable medium of claim 18, wherein the program further causes the processor to select at least one of the plurality of dynamical metrics corresponding to the first statistical measure having a desired level of sensitivity and specificity during the initialization period based on the second statistical measures.

20. The computer readable medium of claim 13, wherein the program further causes the processor to identify in the initialization period the first statistical measures corresponding to the selected dynamical metrics to predict at least one occurrence.

21. The computer readable medium of claim 20, wherein the program further causes the processor to monitor the identified first statistical measures for a condition indicative of the at least one occurrence.

22. The computer readable medium of claim 21, wherein the program further causes the processor to compare the identified first statistical measures to at least one threshold value.

23. The computer readable medium of claim 21, wherein the program further causes the processor to implement at least one of a warning and an intervention in response to detecting the condition indicative of the at least one occurrence.

24. The computer readable medium of claim 13, wherein the program further causes the processor to perform the following functions:

repeat the calculating, selecting, and identifying steps in the initialization period for additional channel groups; and

select a particular channel group based on a comparison of the second statistical measures for the respective channel groups.

25. A system for monitoring dynamical behavior of a multi-dimensional system, comprising:

a processing device that executes the following steps;

generating a multi-dimensional representation for each of a plurality of signals acquired from corresponding channels of the multi-dimensional system;

generating a plurality of dynamical profiles for each channel based on the corresponding multi-dimensional representations, wherein each dynamical profile reflects dynamical characteristics of the associated channel in accordance with one of a plurality of dynamical metrics;

calculating a plurality of first statistical measures for a group of a plurality of the channels, wherein each first statistical measure reflects a level of interaction among the channel group associated with at least one of the plurality of dynamical metrics;

calculating in an initialization period a second statistical measure for each of the first statistical measures, wherein the second statistical measure reflects a level of association of the first statistical measure with related occurrences being monitored;

selecting in the initialization period at least one of the plurality of dynamical metrics based on the second statistical measures calculated for each of the first statistical measures; and

identifying in the initialization period the first statistical measures corresponding to the selected dynamical parameters to characterize the dynamical behavior of the multi-dimensional system.

26. The system of claim 25, further comprising a display device coupled to the processing device for visualization of the acquired signals, the dynamical profiles, and the first and second statistical measures.

27. The system of claim 25, further comprising a data acquisition device coupled to the processing device, wherein the data acquisition device acquires the plurality of signals from at least one sensor of the multi-dimensional system.

28. The system of claim 25, further comprising an end user device coupled to the processing device, wherein the processing device transmits occurrence information to the end user device based on the characterization of the dynamical behavior of the multi-dimensional system.

29. The system of claim 28, wherein the end user device initiates intervention in response to the transmitted occurrence information.