Extracting causal information from a chaotic time series

• US 7,092,849 B2
• Filed: 10/07/2004
• Issued: 08/15/2006
• Est. Priority Date: 06/27/2003
• Status: Active Grant

• Current Assignee: DyAnsys, Inc.
• Original Assignee: DyAnsys, Inc.
• Inventors: Nageshwar, Srini, Lafitte, Melvyn Jérémie, Fèvre-Genoulaz, Marion, Sauvageot, Orin
• Primary Examiner(s): NGHIEM, MICHAEL P
• Application Number:
  US10/961,710
• Publication Number:
  US 20050075811A1
• Time in Patent Office: 677 Days
• Field of Search: 702/66, 702/69, 702/70, 702/79, 702/183, 702189-191, 702/195
• US Class Current: 702/183
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
• CPC Class Codes: A61B 5/4035 Evaluating the autonomic nervous system
• CPC Class Codes: A61B 5/413 Monitoring transplanted tissue or organ, e.g. for possible rejection reactions after a transplant
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 5/4821 Determining level or depth of anaesthesia A61B5/1106 takes precedence
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing

Abstract

Embodiments of the present invention provide a method, a system, and a computer code for analyzing the state of a first system (e.g., the autonomic system) from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system (e.g., the cardiac system) governed by the first system. In one embodiment, the method includes extracting envelope information from the time-varying signal, constructing a phase space for the time-varying signal, extracting information on the relative positions of points corresponding to the time-varying signal in the phase space, combining the envelope and the position information and, based on this combination, providing information on the state of the first system.

Claims

1. A method for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the method comprising:

extracting envelope information from the time-varying signal;

constructing a phase space for the time-varying signal;

extracting information on the relative positions of points corresponding to the time-varying signal in the phase space;

combining the envelope and the position information;

based on the combining, providing information relating to the state of the first system; and

displaying, via a display unit, information relating to the state of the first system.

2. The method according to claim 1, wherein the extracting envelope information, constructing, extracting information on the relative positions of points, combining, and providing information are repeated each time a new time interval appears in the time-varying signal.

3. The method according to claim 1, wherein extracting the envelope information comprises calculating a first upper envelope of the time-varying signal in the direction of the chronological order and calculating a second upper envelope of the time-varying signal in the direction opposite to the chronological order.

4. The method according to claim 1, wherein constructing the phase space comprises constructing vectors on the basis of values taken by the time-varying signal using a determined dimension for the phase space and a determined time lag.

5. The method according to claim 1, wherein extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a lower-dimension space on which an order relation can be established, and calculating distances between the projected points.

6. The method according to claim 5, wherein extracting information on the relative positions of points further comprises discriminating positive and negative distances in said calculated distances.

7. The method according to claim 6, wherein an index representing a probability that a change of the state of the first system occurs at the next event is calculated based on said positive distances or said negative distances.

8. The method according to claim 1, wherein extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a straight line that minimizes the average distance between said points and the straight line, and calculating distances between the projected points.

9. The method according to claim 8, wherein extracting information on the relative positions of points further comprises discriminating positive and negative distances in said calculated distances.

10. The method according to claim 9, wherein an index representing a probability that a change of the state of the first system occurs at the next event is calculated based on said positive distances or said negative distances.

11. The method according to claim 1, wherein the time-varying signal is a raw signal.

12. The method according to claim 1, wherein the first system is the autonomic nervous system.

13. The method according to claim 12, wherein the second system is the cardiac system, the quasi-periodical events are R waves of an electrocardiogram and the chaotic series of time intervals are RR intervals derived from said electrocardiogram.

14. The method according to claim 12, wherein combining comprises performing a first combination calculation providing first data representative of the parasympathetic component of the autonomic nervous system and performing a second combination calculation providing second data representative of the sympathetic component of the autonomic nervous system.

15. The method according to claim 14, wherein combining further comprises (1) calculating a first index representative of a complexity exponent of a first curve defined by said first data, or (2) calculating a second index representative of a complexity exponent of a second curve defined by said second data, or both (1) and (2).

16. The method according to claim 15, wherein combining further comprises (1) calculating a first index representative of a complexity exponent of a first curve defined by said first data, or (2) calculating a second index representative of a complexity exponent of a second curve defined by said second data, or both (1) and (2).

17. The method according to claim 12, wherein combining comprises performing a first combination calculation providing first data and performing a second combination calculation providing second data, the point-by-point subtraction of either of these first and second data from the other representing a sympathovagal balance between the parasympathetic and the sympathetic component of the first system which is the autonomic nervous system.

18. The method according to claim 1, wherein the first system is the autonomic nervous system;

extracting envelope information comprises calculating a first upper envelope ForwHull of the time-varying signal in the direction of the chronological order and calculating a second upper envelope BackwHull of the time-varying signal in the direction opposite to the chronological order;

extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a lower-dimension space on which an order relation can be established, calculating distances between the projected points and discriminating positive and negative distances in said calculated distances, and combining comprises performing the following two combination calculations;

19. The method according to claim 18, wherein combining further comprises calculating the following two indices:

20. The method according to claim 18, wherein combining further comprises:

calculating (ANSigram₂−

ANSigram₁)_previous by performing a point-by-point subtraction of points in the vectors using values ending with a previous time interval that immediately precedes the current time interval;

calculating (ANSigram₂−

ANSigram₁)_current by performing a point-by-point subtraction using values ending with the current time interval;

generating a plot of (ANSigram₂−

ANSigram₁)_previous in an x-axis and (ANSigram₂−

ANSigram₁)_current in a y-axis; and

performing a coordinate transformation of the plot to generate a rotated and normalized plot having (ANSigram₂−

ANSigram₁)_current+(ANSigram₂−

ANSigram₁)_previous in a new x-axis and (ANSigram₂−

ANSigram₁)_current−

(ANSigram₂−

ANSigram₁)_previous in a new y-axis.

21. The method according to claim 20, wherein combining further comprising:

generating the rotated and normalized plot for a plurality of consecutive time intervals;

for each time interval, selecting a point from the plurality of points based on a preset criterion; and

connecting the selected point between the time intervals in chronological order to form a subset balance curve.

22. The method according to claim 20, wherein combining further comprises generating a balance trajectory for a current time interval which includes:

connecting points of the rotated and normalized plot in chronological order to produce a balance curve for the current time interval.

23. The method according to claim 22, wherein combining further comprises:

generating a plurality of balance trajectories for a plurality of consecutive time intervals; and

connecting the plurality of balance trajectories from end point to end point in chronological order of the consecutive time intervals with a time reversal in alternative order to form an accumulated balance trajectory.

24. The method according to claim 22, wherein combining further comprises:

generating a pair of balance trajectories for two consecutive time intervals; and

performing a point-to-point comparison between the two balance trajectories for points with matching time chronology from the two balance trajectories to determine a measure of response of the second system with respect to the first system,

wherein the point-to-point comparison includes computing a distance between corresponding points from the two balance trajectories.

25. The method according to claim 22, further comprising displaying the balance curve.

26. The method according to claim 18, wherein combining further comprises:

generating a rotated and normalized accumulated plot by calculating (ANSigram₂−

ANSigram₁)_previous, calculating ANSigram₂−

ANSigram₁)_current, generating a plot of (ANSigram₂−

ANSigram₁)_previous, in an x-axis and (ANSigram₂−

ANSigram₁)_current in a y-axis, and performing a coordinate transformation of the plot to generate a rotated and normalized plot, for a plurality of time intervals; and

calculating a Balance Index=Norm * sign of (X_CG+Y_CG)

where

$x_{\mathrm{CG}}=\underset{i=1}{\overset{l}{\sum <br><br>}}<br><br>\frac{x_i}{l}$$y_{\mathrm{CG}}=\underset{i=1}{\overset{l}{\sum <br><br>}}<br><br>\frac{y_i}{l}$
Norm=√

{square root over (x²_CG+y²_CG)}

where “

1”

is the number of points in the vector ANSigram₂ or ANSigram₁.

27. The method according to claim 26, further comprising displaying the Balance Index.

28. The method according to claim 18, wherein combining further comprises:

generating a rotated and normalized static balance plot by calculating (ANSigram₂−

ANSigram₁)_previous, calculating (ANSigram₂−

ANSigram₁)_current, generating a plot of (ANSigram₂−

ANSigram₁)_previous in an x-axis and (ANSigram₂−

ANSigram₁)_current in a y-axis, and performing a coordinate transformation of the plot to generate a rotated and normalized plot for a number of time intervals, the number being at least about 100.

29. The method according to claim 28, further comprising displaying the rotated and normalized static balance plot.

30. The method according to claim 18, wherein combining further comprises calculating the following two indices:

ANSindex₁=c₁+c₂*a₁+c₃*b₁

ANSindex₂=c₄+c₅*a₂+c₆*b₂

where

31. The method according to claim 30, wherein n=13.

32. The method according to claim 30, wherein c₁=−

190.857, c₂=3.54654, c₃=−

0.6894, c₄=−

87.7243, c₅=1.8054, and c₆=−

0.350946.

33. The method according to claim 30, wherein combining further comprises determining a coupling between the two indices, ANSindex₁ and ANSindex₂.

34. The method according to claim 33, wherein determining the coupling comprises:

creating a List of distances between the two indices for a number of time intervals, List=|ANSindex₁−

ANSindex₂|;

generating a New List based on summing the List over a preset time window of w time intervals to provide a more global measure of coupling,

35. The method according to claim 33, further comprising displaying the coupling between the two indices, ANSindex₁ and ANSindex₂.

36. The method according to claim 1, further comprising filtering out an anomalous signal from the time-varying signal prior to extracting envelope information from the time-varying signal and constructing the phase space for the time-varying signal.

37. An apparatus for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the apparatus comprising:

a processor configured to;

extract envelope information from the time-varying signal;

construct a phase space for the time-varying signal;

extract information on the relative positions of points corresponding to the time-varying signal in the phase space;

combine the envelope and the position information; and

, based on combining the envelope and the position information, provide information relating to the state of the first system; and

a display to display information relating to the state of the first system.

38. The apparatus according to claim 37, wherein the processor is configured to repeat extracting envelope information, constructing, extracting information on the relative positions of points, combining, and providing information, each time a new time interval appears in the time-varying signal.

39. The apparatus according to claim 37, wherein extracting the envelope information comprises calculating a first upper envelope of the time-varying signal in the direction of the chronological order and calculating a second upper envelope of the time-varying signal in the direction opposite to the chronological order.

40. The apparatus according to claim 37, wherein constructing the phase space comprises constructing vectors on the basis of values taken by the time-varying signal using a determined dimension for the phase space and a determined time lag.

41. The apparatus according to claim 37, wherein extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a lower-dimension space on which an order relation can be established, and calculating distances between the projected points.

42. The apparatus according to claim 41, wherein extracting information on the relative positions of points further comprises discriminating positive and negative distances in said calculated distances.

43. The apparatus according to claim 42, wherein the processor is configured to calculate an index representing a probability that a change of the state of the first system occurs at the next event, based on said positive distances or said negative distances.

44. The apparatus according to claim 37, wherein extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a straight line that minimizes the average distance between said points and the straight line, and calculating distances between the projected points.

45. The apparatus according to claim 44, wherein extracting information on the relative positions of points further comprises discriminating positive and negative distances in said calculated distances.

46. The apparatus according to claim 45, wherein the processor is configured to calculate an index representing a probability that a change of the state of the first system occurs at the next event, based on said positive distances or said negative distances.

47. The apparatus according to claim 37, wherein the first system is the autonomic nervous system.

48. The apparatus according to claim 47, wherein the second system is the cardiac system, the quasi-periodical events are R waves of an electrocardiogram and the chaotic series of time intervals are RR intervals derived from said electrocardiogram.

49. The apparatus according to claim 47, wherein combining the envelope and the position information comprises performing a first combination calculation providing first data representative of the parasympathetic component of the autonomic nervous system and performing a second combination calculation providing second data representative of the sympathetic component of the autonomic nervous system.

50. The apparatus according to claim 49, wherein combining the envelope and the position information further comprises (1) calculating a first index representative of a complexity exponent of a first curve defined by said first data, or (2) calculating a second index representative of a complexity exponent of a second curve defined by said second data, or both (1) and (2).

51. The apparatus according to claim 47, wherein combining the envelope and the position information comprises performing a first combination calculation providing first data and performing a second combination calculation providing second data, the point-by-point subtraction of either of these first and second data from the other representing a sympathovagal balance between the parasympathetic and the sympathetic component of the first system which is the autonomic nervous system.

52. The apparatus according to claim 51, wherein combining the envelope and the position information further comprises (1) calculating a first index representative of a complexity exponent of a first curve defined by said first data, or (2) calculating a second index representative of a complexity exponent of a second curve defined by said second data, or both (1) and (2).

53. The apparatus according to claim 37, wherein the first system is the autonomic nervous system;

extracting envelope information comprises calculating a first upper envelope ForwHull of the time-varying signal in the direction of the chronological order and calculating a second upper envelope BackwHull of the time-varying signal in the direction opposite to the chronological order;

extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a lower-dimension space on which an order relation can be established, calculating distances between the projected points and discriminating positive and negative distances in said calculated distances, and combining comprises performing the following two combination calculations;

Coeffinc₁=B+(4−

4A−

5B+4AB)·

Cinc−

B·

Cdec

Coeffdec₁=B−

B·

Cinc+(4A−

4AB−

B)·

Cdec

54. The apparatus according to claim 37, further comprising a filter to filter out an anomalous signal from the time-varying signal prior to extracting envelope information from the time-varying signal and constructing the phase space for the time-varying signal.

55. A computer readable storage medium having a computer program for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the computer program comprising:

code for extracting envelope information from the time-varying signal;

code for constructing a phase space for the time-varying signal;

code for extracting information on the relative positions of points corresponding to the time-varying signal in the phase space;

code for combining the envelope and the position information;

code for, based on the combining, providing information relating to the state of the first system; and

code for displaying information relating to the state of the first system.

56. The computer readable storage medium according to claim 55, wherein the computer program further comprises code for repeating extracting envelope information, constructing, extracting information on the relative positions of points, combining, and providing information, each time a new time interval appears in the time-varying signal.

57. The computer readable storage medium according to claim 55, wherein the computer program further comprises code for filtering out an anomalous signal from the time-varying signal prior to extracting envelope information from the time-varying signal and constructing the phase space for the time-varying signal.

58. A method for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the method comprising:

obtaining the time-varying signal from a patient via an acquisition unit;

extracting envelope information from the time-varying signal;

constructing a phase space for the time-varying signal;

extracting information on the relative positions of points corresponding to the time-varying signal in the phase space;

combining the envelope and the position information; and

based on the combining, providing information relating to the state of the first system;

wherein the first system is the autonomic nervous system of the patient; and

wherein the second system is the cardiac system of the patient, the quasi-periodical events are R waves of an electrocardiogram of the patient, and the chaotic series of time intervals are RR intervals derived from said electrocardiogram.

59. A method for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the method comprising:

obtaining the time-varying signal from a patient via an acquisition unit;

extracting envelope information from the time-varying signal;

constructing a phase space for the time-varying signal;

extracting information on the relative positions of points corresponding to the time-varying signal in the phase space;

combining the envelope and the position information; and

based on the combining, providing information relating to the state of the first system;

wherein the first system is the autonomic nervous system of the patient; and

wherein combining comprises performing a first combination calculation providing first data representative of the parasympathetic component of the autonomic nervous system of the patient and performing a second combination calculation providing second data representative of the sympathetic component of the autonomic nervous system of the patient.

60. The method according to claim 59, wherein the first system is the autonomic nervous system;

extracting envelope information comprises calculating a first upper envelope ForwHull of the time-varying signal in the direction of the chronological order and calculating a second upper envelope BackwHull of the time-varying signal in the direction opposite to the chronological order;

extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a lower-dimension space on which an order relation can be established, calculating distances between the projected points and discriminating positive and negative distances in said calculated distances, and combining comprises performing the following two combination calculations;

Coeffinc₁=B+(4−

4A−

5B+4AB)·

Cinc−

B·

Cdec

Coeffdec₁=B−

B·

Cinc+(4A−

4AB−

B)·

Cdec

61. The method according to claim 60, wherein combining further comprises calculating the following two indices:

ANSindex₁=c₁+c₂*a₁+c₃*b₁

ANSindex₂=c₄+c₅*a₂+c₆*b₂

where

62. A method for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the method comprising:

obtaining the time-varying signal from a patient via an acquisition unit;

extracting envelope information from the time-varying signal;

constructing a phase space for the time-varying signal;

extracting information on the relative positions of points corresponding to the time-varying signal in the phase space;

combining the envelope and the position information; and

based on the combining, providing information relating to the state of the first system;

wherein the first system is the autonomic nervous system of the patient; and

wherein combining comprises performing a first combination calculation providing first data and performing a second combination calculation providing second data, the point-by-point subtraction of either of these first and second data from the other representing a sympathovagal balance between the parasympathetic and the sympathetic component of the autonomic nervous system of the patient.

63. The method according to claim 62, wherein the first system is the autonomic nervous system;

extracting envelope information comprises calculating a first upper envelope ForwHull of the time-varying signal in the direction of the chronological order and calculating a second upper envelope BackwHull of the time-varying signal in the direction opposite to the chronological order;

extracting information on the relative positions of points comprises projecting said points corresponding to the time-varying signal in the phase space onto a lower-dimension space on which an order relation can be established, calculating distances between the projected points and discriminating positive and negative distances in said calculated distances, and combining comprises performing the following two combination calculations;

Coeffinc₁=B+(4−

4A−

5B+4AB)·

Cinc−

B·

Cdec

Coeffdec₁=B−

B·

Cinc+(4A−

4AB−

B)·

Cdec

64. The method according to claim 63, wherein combining further comprises calculating the following two indices:

ANSindex₁=c₁+c₂*a₁+c₃*b₁

ANSindex₂=c₄+c₅*a₂+c₆*b₂

where

65. An apparatus for analyzing the state of a first system from a time-varying signal representing a chaotic series of time intervals between quasi-periodical events produced by a second system governed by the first system, the apparatus comprising:

means for extracting envelope information from the time-varying signal, constructing a phase space for the time-varying signal, extracting information on the relative positions of points corresponding to the time-varying signal in the phase space, combining the envelope and the position information, and, based on the combining, providing information relating to the state of the first system; and

a display for displaying information relating to the state of the first system.