Method and apparatus for physiologic system identification

• US 4,930,517 A
• Filed: 04/25/1989
• Issued: 06/05/1990
• Est. Priority Date: 04/25/1989
• Status: Expired due to Term

• Current Assignee: Massachusetts Institute of Technology
• Original Assignee: Massachusetts Institute of Technology
• Inventors: Berger, Ronald D., Appel, Marvin L., Cohen, Richard J.
• Primary Examiner(s): Jaworski, Francis
• Assistant Examiner(s): MANUEL, GEORGE C
• Application Number:
  US07/343,404
• Time in Patent Office: 406 Days
• Field of Search: 128/671, 128/695, 128/696, 364/413.01, 364/413.02, 364/413.03, 364/413.04, 364/413.05, 364/413.06
• US Class Current: 600/484
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/7257 using Fourier transforms

Abstract

A method and apparatus for estimating transfer functions among multiple physiologic or biologic signals in the presence of feedback. The invention comprises the injection of broad band purturbation into one or more of the subsystems under study, and measuring signals from the subsystems. These signals are transformed to generate a new set of n signals. Casual transfer functions between the signals and additive noise sources are used to represent the relationships between the n signals. Parametric system identification techniques are then used to characterize quantitatively at least two casual transfer functions and noise sources. This method and apparatus provides a powerful tool with which to characterize the interactions of subsystems in the presence of feedback.

Claims

1. A method for analyzing a biological or physiological system comprising the steps of:

A. Applying at least one exogenous broad-band perturbation to said physiological or biological system;

B. Measuring m signals from the biological or physiological system;

C. Transforming the m signals to generate a new set of n signals where n may be greater than, equal to or less than m;

D. Representing the relationships between the new signals by causal transfer functions and additive noise sources; and

E. Using a parametric system identification technique to characterize quantitatively at least two causal transfer functions and the noise sources.

2. The method of claim 1 wherein the quantitative characterization includes the selection of an optimal model order for the causal transfer function according to determined criteria.

3. The method of claim 1 where the parametric identification technique is an autoregressive-moving average, parametrization with a plurality of known exogenous inputs.

4. The method of claim 1 wherein said new signals are heart rate, arterial blood pressure, and instantaneous lung volume.

5. The method of claim 4 wherein said at least one broad-band perturbation is applied to the respiratory system.

6. The method of claim 5 wherein said at least one broad-band perturbation is applied to said respiratory system by random interval breathing.

7. The method of claim 4 wherein said at least one broad-band perturbation is applied to the cardiac system.

8. The method of claim 7 wherein said at least one broad-band perturbation is applied to said cardiac system by direct random ventricular stimulation.

9. The method of claim 1 wherein said new signals are heart rate, arterial blood pressure, arterial resistance and instantaneous lung volume.

10. The method of claim 1 wherein one measured signal is arterial blood pressure measured by intra-arterial catheter.

11. The method of claim 1 wherein one measured signal is arterial blood pressure measured by continuous noninvasive arterial blood pressure transducer.

12. The method of claim 1 wherein one measured signal is cardiac output.

13. The method of claim 1 wherein one measured signal is central venous pressure.

14. The method of claim 1 wherein said at least one broad-band perturbation is introduced into a physiological system of an organism by conscious acts by said organism.

15. The method of claim 14 wherein said conscious acts are induced by sensory-cues.

16. The method of claim 15 wherein said sensory-cues are aural.

17. The method of claim 15 wherein said sensory-cues are visual.

18. The method of claim 15 wherein said sensory-cues are tactile.

19. The method of claim 1 where said at least one broad-band perturbation is introduced by changes in posture.

20. The method of claim 1 where said at least one broad-band perturbation is introduced by changes in lower body negative pressure.

21. The method of claim 1 where said at least one broad-band perturbation is introduced by changes in body tilt.

22. The method of claim 1 where said at least one broad-band perturbation is introduced by exercise.

23. The method of claim 1 wherein said at least one broad-band perturbation has a Poisson distribution.

24. The method of claim 23 wherein said Poisson distribution is modified to have a maximum value.

25. The method of claim 24 wherein said maximum value is approximately 15 sec.

26. The method of claim 23 wherein said Poisson distribution is modified to have a minimum value.

27. The method of claim 26 wherein said minimum value is approximately 1 sec.

28. The method of claim 1 wherein said at least one broad-band perturbation has a Gaussian distribution.

29. The method of claim 1 wherein the representation of the interaction between the n signals results in n(n-1) transfer functions and n additive noise sources.

30. The method of claim 1 wherein the representing of the interaction between the n signals includes the removal of transfer functions and noise sources known to be of negligible significance.

31. The method of claim 1 applied to several physiological states.

32. The method of claim 31 wherein changes of physiological state are induced by changing posture.

33. The method of claim 31 wherein changes of physiological state are induced by changes in lower body pressure.

34. The method of claim 31 wherein changes of physiological state are induced by changes in body tilt.

35. The method of claim 31 wherein changes of physiological state are induced by exercise.

36. The method of claim 1 wherein at least one signal transformation is nonlinear.

37. The method of claim 1 wherein at least one signal transformation is linear.

38. A system for analyzing a biological or physiological system comprising:

A. An apparatus adapted to introduce at least one exogenous broad-band pertubation to said physiological or biological system;

B. An apparatus adapted to measure m signals from said physiological or biological system; and

C. A computing apparatus programmed to

i. Transform said m signals to generate a new set of n signals;

ii. Represent the relationships between the new n signals by causal transfer functions and additive noise sources; and

iii. Characterize quantitatively at least two causal transfer functions and the noise sources.

39. The system of claim 38 wherein said signals are measured by electrocardiograph apparatus, radial artery catheter apparatus and plethysmograph apparatus.

40. The system of claim 39 wherein said plethysmograph is a two-belt inductance plethysmograph apparatus.

41. A method for analyzing transfer functions and noise sources for heart rate, arterial blood pressure and respiration in an organism, comprising the steps of:

A. Applying broad-band perturbation by random interval breathing and/or random interval neck suction;

B. Measuring the electrocardiogram, arterial blood pressure and respiration signals;

C. Deriving a heart rate signal by nonlinear transformation of the electrocardiogram or arterial blood pressure signal;

D. Representing the relationship between the heart rate, arterial blood pressure and respiration signals by causal transfer functions and additive noise sources; and

E. Using parametric system identification to characterize quantitatively at least two causal transfer functions and the noise sources.

42. A method for analyzing transfer functions and noise sources for heart rate, arterial blood pressure, vascular resistance and respiration in an organism, comprising the steps of:

A. Applying broad-band perturbation by random interval breathing and/or random interval neck suction;

B. Measuring the electrocardiogram, arterial blood pressure and respiration signals;

C. Deriving a heart rate signal by nonlinear transformation of the electrocardiogram or arterial blood pressure signal, and deriving a vascular resistance signal by nonlinear transformation of the arterial blood pressure signal;

D. Representing the relationship between the heart rate, arterial blood pressure, vascular resistance, and respiration signals by causal transfer functions and noise sources; and

E. Using parametric system identification to characterize quantitatively at least two causal transfer functions and the noise sources.