System and device for multi-scale analysis and representation of physiological data

• US 6,925,324 B2
• Filed: 04/17/2002
• Issued: 08/02/2005
• Est. Priority Date: 05/30/2000
• Status: Expired due to Fees

• Current Assignee: Vladimir Shusterman
• Original Assignee: Vladimir Shusterman
• Inventors: Shusterman, Vladimir
• Primary Examiner(s): Getzow, Scott M.
• Application Number:
  US10/124,651
• Publication Number:
  US 20020143263A1
• Time in Patent Office: 1,203 Days
• Field of Search: 600508-521
• US Class Current: 600/509
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/7232 involving compression of the physiological signal, e.g. to extend the signal recording period

Abstract

System comprised of a medical device and method for analyzing physiological and health data and representing the most significant parameters at different levels of detail which are understandable to a lay person and a medical professional. Low, intermediate and high-resolution scales can exchange information between each other for improving the analyses; the scales can be defined according to the corresponding software and hardware resources. A low-resolution Scale I represents a small number of primary elements such as intervals between the heart beats, duration of electrocardiographic PQ, QRS, and QT-intervals, amplitudes of P-, Q-, R-, S-, and T-waves. This real-time analysis is implemented in a portable device that requires minimum computational resources. The set of primary elements and their search criteria can be adjusted using intermediate or high-resolution levels. At the intermediate-resolution Scale II, serial changes in each of the said elements can be determined using a mathematical decomposition into series of basis functions and their coefficients. This scale can be implemented using a specialized processor or a computer organizer. At the high-resolution Scale III, combined serial changes in all primary elements can be determined to provide complete information about the dynamics of the signal. This scale can be implemented using a powerful processor, a network of computers or the Internet. The system can be used for personal or group self-evaluation, emergency or routine ECG analysis, or continuous event, stress-test or bed-side monitoring.

Claims

1. A method for analysis of physiological or health data in at least two levels of detail, said method comprising:

analyzing a plurality of primary elements in said data in first scale, low level resolution to detect one-time changes in such primary elements and thereby identify abnormal or unstable primary elements;

comparing said primary elements with reference values for such primary elements, and representing deviation of primary elements from the reference values in a form understandable to a lay person;

analyzing at least some of the same said primary elements in said data in second scale, higher level resolution using mathematical decomposition to provide detailed characterization of serial changes in said abnormal or unstable primary elements; and

exchanging information between said analyzing in said first and second levels of resolution to improve at least one of said first and second analyses.

2. A method as set forth in claim 1 in which said analyzing in a second scale, higher level resolution uses orthogonal mathematical decomposition.

3. A method as set forth in claim 2 in which said analyzing in said second scale, higher resolution further includes:

combining a plurality of said serial changes in the abnormal or unstable primary elements and analyzing the combined changes using orthogonal mathematical decomposition; and

selecting a combination of parameters for tracking the changes in said abnormal or unstable elements and sending information respecting the selected combination to said low-resolution analysis for adjustment of monitoring parameters.

4. A method as set forth in claim 2 in which said analyzing in said second scale, higher resolution comprises:

forming a time series from said serial changes;

characterizing said time series using Principal Component Analysis (PCA) and generating PCA-coefficients indicative of both linear and nonlinear changes in the individual pattern; and

determining the magnitude of said linear and nonlinear changes by using time varying mean and variance of said PCA-coefficients and determining the complexity of said linear and nonlinear changes by calculating the number of PCA-coefficients that exhibit substantially simultaneous changes.

5. A method as set forth in claim 2 in which said analyzing in said second scale, higher resolution comprises:

forming time series from said serial changes;

characterizing said time series using Principal Component Analysis (PCA) and generating PCA-coefficients indicative of both linear and nonlinear changes in the individual pattern;

presenting said PCA-coefficients as a vector in n-dimensional space wherein n is the number of variables analyzed; and

determining the magnitude and direction of the combined changes in the characteristic variables by the changes in the magnitude and direction of said n-dimensional vector.

6. A method as set forth in claim 2 which includes adjusting said analyzing in low resolution based on said detailed characterization of serial changes in the primary elements.

7. A method as set forth in claim 6 that further comprises representing deviations of said primary elements from said reference values in a quantitative form for use by a medical professional.

8. A method as set forth in claim 2 which includes producing a visual or audio warning signal when said low resolution detects an abnormal or unstable primary element.

9. A method as set forth in claim 2, in which the multi-resolution analysis and representation are applied for

improved detection of changes during one-time examination,

assessment of short-term and long term dynamics, which include fitness level, disease progression, treatment, complications and side-effects control,

physical examination, early detection of subtle changes, and timely initiation or adjustment of therapy,

early prediction and prevention of physiological disorders and abnormalities,

comparison of the values of data obtained from individual patients against averages of values obtained from a group of patients or population of patients to facilitate analysis of individual data and to determine the values that characterize groups of patients with similar characteristics and/or similar disorders.

10. A method as set forth in claim 1 in which orthogonal mathematical decomposition is applied to both said resolution scales.

11. A method as set forth in claim 1 in which the changes in basis vectors (eigenvectors) are used to evaluate the changes and variables in the data at both resolution scales.

12. A method as set forth in claim 11 in which changes in said eigenvalues are used for estimation of the changes in the variables in the data at both resolution scales so that at the low resolution scale, the analysis is limited to the estimation of changes in the spectral characteristics of a few, most significant eigenvectors and the corresponding eigenvalues, and at the higher resolution scales, the analysis includes a greater number of studied eigenvectors and eigenvalues, and estimation of their combined changes.

13. A method as set forth in claim 1 in which analyzing said data in low and higher resolution to provide detailed characterization of serial changes in said abnormal or unstable primary elements is performed using at least one method selected from the following:

multidimensional scaling non-metric distances, mapping techniques, non-orthogonal linear mappings, nonlinear mappings, projection methods, statistical estimators, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, probabilistic methods, Bayesian probability and Mahalanobis distance.

14. A method as set forth in claim 1, in which the multi-resolution analysis and representation are applied to other physiological signals selected from at least one of the following signals:

blood pressure, cardiac output, vascular activity, temperature, respiration, cardiac, abdominal, or breathing sounds, blood flow, hormonal concentration, neural activity, electroencephalographic activity, and other electrical, mechanic, sonic, biochemical, biophysical processes in the human body, and/or demographic, psychological, or environmental data.

15. A system for detection of serial changes in physiological or health data and analysis in at least two levels of detail for use by both lay persons and medical professionals, said system comprising:

an acquisition unit for collecting physiological or health data from a subject over a period of at least several seconds;

a first analysis and processing unit for detecting a plurality of primary elements from said data and processing said primary elements in low level resolution to generate data respecting time intervals or amplitudes of said primary elements, storing reference values of said plurality of primary elements, and comparing said reference values with data newly received by said first analysis and processing unit and producing qualitative indicators and quantitative data of differences between said recorded data and said newly received data;

an output unit for displaying said qualitative indicators in a form understandable by lay persons and quantitative data for medical professionals; and

a communications unit for sending quantitative data of said primary elements to a second analysis unit for processing and detailed analysis of serial changes in at least some of the same said primary elements in said data, said second analysis unit using a higher scale of resolution using mathematical decomposition for assessing small changes in serial data and for exchanging information with said first analysis and processing unit to improve at least one of said low level or higher level resolution.

16. A system as set forth in claim 15 in which said second analysis uses orthogonal mathematical decomposition.

17. A system for detection and analysis as set forth in claim 16 in which said second analysis unit characterizes time series of at least one primary element or index in said set of primary elements and indexes using Principal Component Analysis (PCA) and generating PCA coefficients indicative of both linear and nonlinear changes in the individual pattern;

determines the magnitude of said changes in the characteristic variables by using time varying mean and variance of said PCA coefficients and determining the complexity of said changes by calculating the number of PCA coefficients that exhibit substantially simultaneous changes;

presents said combination of PCA coefficients as a vector in n-dimensional space wherein n is the number of variables analyzed; and

determines the magnitude and complexity of the changes in the combination of the characteristic variables by the changes in said n-dimensional vector.

18. A detection and analysis system as set forth in claim 16 in which said first analysis unit includes means for receiving commands from said second analysis unit to adjust the primary elements to be detected from said data.

19. A detection and analysis system as set forth in claim 16 in which said output unit for qualitative representation includes a static screen for one-time presentation and a dynamic screen for presenting changes in serial recordings.

20. A detection and analysis system as set forth in claim 16 in which said set of primary elements includes a complete set of characteristic fragments of ECG signals and their parameters, said set of characteristic fragments including P, Q, R, S, T and U waves, ST-segment, PR, QRS and QT-intervals, said parameters including the amplitude and duration of said waves and segments, and the time intervals between the heart beats.

21. A detection and analysis system as set forth in claim 15 in which said first and second analysis units use changes in basis vectors (eigenvectors) to evaluate the changes and variables in the data in both levels of resolution.

22. A detection and analysis system as set forth in claim 21 in which changes in said eigenvectors are used for estimation of the changes in the variables in the data at both resolution scales so that, at the low resolution scale, the analysis is limited to the estimation of the changes in the spectral characteristics of a few, most significant eigenvectors and the corresponding eigenvalues, and at the higher resolution scale, the analysis includes a greater number of studied eigenvectors and eigenvalues, and estimation of their combined changes.

23. A detection and analysis system as set forth in claim 15 in which said first and second analysis units analyze said data in low and higher resolution respectively to provide detailed characterization of serial changes in said abnormal or unstable primary elements using at least one method selected from the following:

multidimensional scaling, non-metric distances, mapping techniques, non-orthogonal linear mappings, nonlinear mappings, projection methods, statistical estimators, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, probabilistic methods, Bayesian probability and Mahalanobis distance.

24. A detection and analysis system as set forth in claim 15 in which said first and second analysis units analyze other physiological data selected from at least one of the following signals:

blood pressure, cardiac output, vascular activity, temperature, respiration, cardiac, abdominal, or breathing sounds, blood flow, hormonal concentration, neural activity, electroencephalographic activity, and other electrical, mechanic, sonic, biochemical, biophysical processes in the human body and/or demographic, psychological, or environmental data.

25. A detection and analysis system as set forth in claim 15 in which said output unit includes a screen for displaying said qualitative indicators of a minimal set of most significant indexes to indicate a substantially abnormal physiological or health data and a screen for quantitative representation that represents a complete set of characteristic fragments and their parameters.

26. A detection and analysis system as set forth in claim 15 in which said first analysis and processing unit includes means for decomposing physiological or health data into a plurality of elements, and said decomposing means performs an orthogonal decomposition to extract the most significant information about said physiological or health data.

27. A detection and analysis system as set forth in claim 15 in which said first analysis and processing unit, output unit and communications unit are combined in one pocket size unit connected to said communication unit.

28. A detection and analysis system as set forth in claim 15 in which said communication unit is wireless.

29. A detection and analysis system as set forth in claim 15 in which said first analysis and processing unit includes storage for a physiologic range of the primary elements and detector of error and noise to reject the data upon detection of the primary elements or the indexes beyond the physiologic range.

30. A detection and analysis system as set forth in claim 15 in which said output unit includes a display for viewing the physiological data.

31. A portable system for monitoring physiological or health data and analyzing the data in at least two levels of detail (or resolution) for displaying changes detected in the data, said portable system comprising:

an acquisition unit for receiving physiological or health data generated by monitoring a subject for at least several seconds;

at least one analysis unit for detecting a plurality of primary elements from said signals to detect one-time changes in such primary elements and thereby identify abnormal or unstable primary elements, storing said plurality of primary elements, comparing said plurality of primary elements which have been stored with a plurality of primary elements newly received from said analysis nodule and producing both qualitative and quantitative information representing the differences in the data in low level resolution, and using mathematical decomposition to provide detailed characterization of serial changes in said abnormal or unstable primary elements in higher level resolution;

an output unit for displaying qualitative information understandable by lay persons and quantitative information useful to medical professionals; and

a communications unit for sending data to a remote processing unit for analysis by medical professionals and exchanging information between said at least one analysis unit and said remote processing unit to improve functionality of at least one of said one analysis unit and said remote processing unit.

32. A portable system as set forth in claim 31 which forms a time series from said serial changes, characterizes said time series using Principal Component Analysis (PCA) and generates PCA-coefficients indicative of both linear and nonlinear changes in the individual pattern, determines the magnitude of said linear and nonlinear changes by using time varying mean and variance of said PCA-coefficients and determines the complexity of said linear and nonlinear changes by calculating the number of PCA-coefficients that exhibit substantially simultaneous changes.

33. A portable system as set forth in claim 31 which includes means for adjusting said characterization in low level resolution based on characterization of the serial changes in high level resolution.

34. A portable system as set forth in claim 31 that further includes an analysis unit for analyzing said primary elements in third level high resolution using orthogonal mathematical decomposition to provide more detailed characterization of serial changes in the abnormal or unstable primary elements.

35. A portable system as set forth in claim 31 in which said set of primary elements includes a complete set of characteristic fragments of ECG signals and their parameters, said set of characteristic fragments including P, Q, R, S, T and U waves, ST segment, PR, QRS and QT intervals, and said parameters including the amplitude and duration of said waves and segments, and the time intervals between heart beats.

36. A portable system as set forth in claim 31 in which said output unit includes a display for the physiological or health data.

37. A portable system as set forth in claim 31 in which said first and second analysis units use changes in basis vectors (eigenvectors) to evaluate the changes and variables in the data in both levels of resolution.

38. A portable system as set forth in claim 37 in which changes in said eigenvectors are used for estimation of the changes in the variables in the data at both resolution scales so that, at the low resolution scale, the analysis is limited to the estimation of the changes in the spectral characteristics of a few, most significant eigenvectors and the corresponding eigenvalues, and at the higher resolution scale, the analysis includes a greater number of studied eigenvectors and eigenvalues, and estimation of their combined changes.

39. A portable system as set forth in claim 31 in which said first and second analysis units analyze said data in low and higher resolution respectively to provide detailed characterization of serial changes in said abnormal or unstable primary elements using at least one method selected form the following:

multidimensional scaling, non-metric distances, mapping techniques, non-orthogonal linear mappings, nonlinear mappings projection methods, statistical estimators, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, probabilistic methods, Bayesian probability and Mahalanobis distance.

40. A portable system as set forth in claim 31 in which said first and second analysis units analyze other physiological data selected from at least one of the following signals:

blood pressure, cardiac output, vascular activity, temperature, respiration, cardiac, abdominal, or breathing sounds, blood flow, hormonal concentration, neural activity, electroencephalographic activity, and other electrical, mechanic, sonic, biochemical, biophysical processes in the human body and/or demographic, psychological, or environmental data.

41. A system for detection of serial changes in health data and analysis of the data for use by both lay persons and medical professionals, said system comprising:

an acquisition unit for collecting health data from a subject over a peirod of at least several seconds;

an analysis unit for detecting a plurality of primary elements from said data and processing said primary elements to generate references values respecting said primary elements, storing said references values, and comparing said reference values with data newly received by said analysis and processing unit usign mathematical decomposition and producing qualitative and quantitative data of differences between said reference data and said newly received data; and

an output unit for displaying said quantitative and qualitative data.

42. A system as set forth in claim 41 in which said mathematical decomposition is linear orthogonal.

43. A system as set forth in claim 41 in which said mathemetical decomposition is linear non-orthogonal.

44. A system as set forth in claim 41 in which said mathematical decomposition is nonlinear.

45. A system as set forth in claim 41 which includes a communications unit for sending data from said analysis unit to a second analysis unit for processing and detailed analysis of serial changes in at least some of the same primary elements in said data.

46. A method for detecting serial changes in health data and analysis of the data for use by both lay persons and medical professionals, said system comprising:

collecting health data from a subject over a period of at least several seconds;

analyzing said health data to detect a plurality of primary elements in said data and processing said primary elements to generate reference values respecting said primary elements, storing said reference values, and comparing said reference values with data newly received by said analysis and processing unit using mathematical decomposition and producing qualitative and quantitative data of differences between said reference data and said newly received data; and

displaying said quantitative and qualitative data.

47. A method as set forth in claim 46 in which said mathematical decomposition is linear orthogonal.

48. A method as set forth in claim 46 in which said mathematical decomposition is linear non-orthogonal.

49. A method as set forth in claim 46 in which said mathematical decomposition is nonlinear.

50. A method as set forth in claim 46 which includes transmitting said quantitative data to a remote analysis unit and processing system for processing and detailed analysis of serial changes in at least some of the same primary elements in said data.

51. A system for detection of serial changes in health data and analysis of the data, said system comprising:

an implantable acquisition unit for collecting health data from a subject; and

an external processing unit in wireless communication with said implantable acquisition unit;

wherein at least one of said implantable acquisition unit and said external device having the capability of detecting a plurality of primary elements from said data and processing said primary elements to generate reference values respecting said primary elements, storing said reference values, comparing said reference values with data newly received by said implantable acquisition unit using mathematical decomposition to generate health data of differences between said reference data and said newly received data.

52. A system as set forth in claim 51 in which said external processing unit is selected from a personal computer, a specialized processor and a computer organizer.

53. A system as set forth in claim 52 in which said external processing unit has a wireless bi-directional communication with said acquisition device and with a remote server.

54. A system as set forth in claim 52 in which said external processing unit has a display for showing health information.

55. A system as set forth in claim 52 in which said external processing unit has a function for manual input of data.

56. A system as set forth in claim 51 in which said acquisition device includes means for receiving commands from said external processing device to adjust the primary elements to be detected.

57. A system as set forth in claim 51 in which said mathematical decomposition is linear orthogonal.

58. A system as set forth in claim 51 in which said mathematical decomposition is linear non-orthogonal.

59. A system as set forth in claim 51 in which said mathematical decomposition is nonlinear.

60. A system as set forth in claim 51 in which said mathematical decomposition is performed using at least one of the following methods:

multidimensional scaling, non-meteric distances, mapping techniques, non-orthogonal linear mappings, non-linear mappings, projection methods, ststistical estimators, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, probabilistic methods, Bayesian proability and Mahalanobis distance.

61. A system as set forth in claim 51 in which said acquisition unit processes said primary elements at a first level of resolution and said external processing unit processes said health data at a second higher level of resolution.

62. A system as set forth in claim 51 in which said external processing unit processes said health data at more than one higher levels of resolution.

63. A system as set forth in claim 51 in which said external processing unit consist of at least two units in communication with each other.

64. An external processing unit as set forth in claim 63 in which said communication between said at least two units is wireless.

65. A system as set forth in claim 51 in which said health data are selected from at least one of the following signals:

blood pressure, cardiac output, vascular activity, temperature, respiration, cardiac, abdominal, or breathing sounds, blood flow, hormonal concentration, neural activity,electroencephalographic activity, and other electrical, mechanical, sonic, biochemical, biophysical processes in the human body and/or demographic, psychological, or enviromental data.