Measurement and analysis of trends in physiological and/or health data

• US 20060122525A1
• Filed: 01/06/2006
• Published: 06/08/2006
• Est. Priority Date: 05/30/2000
• Status: Active Grant

• Current Assignee: Vladimir Shusterman
• Original Assignee: Vladimir Shusterman
• Inventors: Shusterman, Vladimir
• Granted Patent: US 7,485,095 B2
• Time in Patent Office: Days
• US Class Current: 600/513
• CPC Class Codes: A61B 5/0022 Monitoring a patient using a global network, e.g. telephone networks, internet
• CPC Class Codes: A61B 5/243 specially adapted for magnetocardiographic [MCG] signals
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/411 Detecting or monitoring allergy or intolerance reactions to an allergenic agent or substance
• CPC Class Codes: A61B 5/6822 Neck
• CPC Class Codes: A61B 5/7232 involving compression of the physiological signal, e.g. to extend the signal recording period
• CPC Class Codes: A61B 5/726 using Wavelet transforms

Abstract

System comprised of a portable medical device and method for non-contact registering at least one of electrocardiographic (ECG), magnetocardiographic (MCG), physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, blood flow, neural activity, and other physiological, and health data, extracting and representing the most significant parameters from time series (trends) of said data. The system achieves the necessary sensitivity (signal-to-noise ratio) in order to miniaturize the device by collecting data of at least one fiducial point in a cardiac complex over a period of at least one, and preferably, several seconds, and extracting the underlying typical patterns from these data. Due to the miniaturization (pocket-size), the system can be implemented in a shape of a pen (or another miniature shape) that can be worn in a pocket.

Claims

1. A method for non-contact registering at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic data, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, extracting and representing the most significant parameters from series of said data, said method comprising:

collecting said data from at least one fiducial point in a physiological cycle over a period of at least several seconds so that the minimum data collection time depends on the sensitivity (signal-to-noise ratio) of the data collection device; and

extracting typical patterns from said data using at least one of mathematical transformation and mathematical modeling.

2. A method as set forth in claim 1 in which said extracting of typical patterns includes:

collecting at least one of electrocardiographic (ECG), magnetocardiographic (MCG), ballistocardiographic, mechanocardiographic, physical activity, body position, respiration, blood pressure, vasomotor activity, temperature, physiological, and health data over a period of at least several seconds so that the minimum data collection time depends on the sensitivity (signal-to-noise ratio) of the data collection device;

finding fiducial points indicative of a specific phase of physiological cycle;

aligning said data using said fiducial points;

applying a mathematical transformation to the said aligned time series of said data to obtain a typical pattern or waveform or a set of coefficients or primary elements that represent the most typical features of the said data in the corresponding time interval; and

forming time series from the said typical patterns or waveforms or coefficients to analyze trends.

3. A method as set forth in claim 1 in which said mathematical transformation and mathematical modeling includes at least one of the following methods:

averaging the signal over a period of several cardiac beats (seconds), orthogonal and non-orthogonal decompositions, principal component analysis, mappings and projections, wavelet transform, Fourier transform, Laplace transform, and Hilbert transform, linear and non-linear correlations, linear and non-linear regression models, methods of artificial intelligence, neural networks, expert systems, fuzzy logic, statistical estimators, and filtering techniques.

4. A method said forth in claim 1, said method comprising:

detecting fiducial points indicative of a specific phase of physiological cycle;

time-aligning the cardiac beats using said fiducial points; and

applying said mathematical transformation to extract typical patterns from said time-aligned series of said data.

5. A method said forth in claim 4, said detecting of fiducial points is performed by measuring at least one of the following signals:

mechanical movements of the chest and other parts of the human body measured by a 1, 2, or 3-axial accelerometer, giant magnetic resistance (GMR), transmitter and receiver of radio waves, or another sensor determining position and movements of the chest and body, mechanical sounds of the heart, electrical signals generated by the heart, light, infrared, and ultrasound signals measuring cardiac mechanical activity.

6. A method for analysis of serial changes in at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic data, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, said method comprising:

registering at least one type of said data using at least one non-contact sensor over a period of at least several seconds so that the minimum registration time is determined by the sensor sensitivity (signal-to-noise ratio);

determining a typical pattern or waveform or a set of coefficients or wavelets by applying a mathematical transformation to at least one of said registered data over a period of at least several seconds;

forming a time series from said typical patterns or waveforms or coefficients or wavelets corresponding to the consecutive time intervals; and

characterizing trends in said time series using a mathematical transformation.

7. A method as set forth in claim 6 in which said mathematical transformation and mathematical modeling includes at least one of the following methods:

averaging said serial changes in the cardiac beats using at least one of orthogonal and non-orthogonal decompositions, principal component analysis, mappings and projections, wavelet transform, Fourier transform, Laplace transform, and Hilbert transform, linear and non-linear correlations, linear and non-linear regression models, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, and probabilistic methods, Bayesian probability and Mahalanobis distance, pattern recognition, methods of artificial intelligence, fuzzy logic, neural networks, expert systems, statistical estimators, filtering techniques, and hybrid artificial intelligence systems.

8. A method as said forth in claim 6, said method comprising:

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.

9. A system for non-contact registering at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, blood pressure, vasomotor activity, respiration, temperature, physiological, and health data, extracting and representing the most significant parameters from time series of said data, said system comprising:

at least one of miniaturized, non-contact physiological sensors to collect said data over a period of at least several seconds with the minimum time for the data collection being determined using the sensitivity (signal-to-noise ratio) of the said sensors; and

software or hardware for extracting typical patterns from these data using at least one method selected from mathematical transformation, modeling, statistics and artificial intelligence.

10. A system as set forth in claim 9 in which said extracting of typical patterns includes:

software or hardware for finding fiducial points indicative of specific phase of physiological cycle;

software or hardware for aligning said data using said fiducial points; and

software or hardware for applying a mathematical transformation to the said aligned time series of said data.

11. A system as set forth in claim 9 in which said mathematical transformation and mathematical modeling includes at least one of the following methods:

averaging the signal over a period of at least several cardiac beats, orthogonal and non-orthogonal decompositions, principal component analysis, mappings and projections, wavelet transform, Fourier transform, Laplace transform, and Hilbert transform, linear and non-linear correlations, linear and non-linear regression models, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, and probabilistic methods, Bayesian probability and Mahalanobis distance, pattern recognition, methods of artificial intelligence, fuzzy logic, neural networks, expert systems, statistical estimators, filtering techniques, and hybrid artificial intelligence systems.

12. A system as said forth in claim 9, said system comprising software or hardware for:

detecting fiducial points indicative of a specific phase of physiological cycle;

time-aligning the cardiac beats using said fiducial points; and

applying said mathematical transformation to extract typical patterns from said time-aligned series of said data.

13. A system as said forth in claim 9, said system used as a simple diagnostic and screening tool during regular daily activities and during night-time, for analysis of slow, long-term trends in cardiac electromagnetic signals that develop gradually over periods of days-weeks-months, for detection of slow changes in the pattern of at least one of electrocardiographic (ECG), magnetocardiographic (MCG), physical activity, body position, blood pressure, vasomotor activity, respiration, temperature, physiological, and health data, including changes in the heart rate, amplitude, polarity, and duration of P, Q, R, S, T and U waves, ST-segment, and the intervals between the peaks (including PR, QRS, and QT intervals) for periodic personal self-examination for subjects at risk or with a history of cardiovascular disease, for monitoring slow changes in the said data due to development of cardiovascular disease, treatment results, physical exercises, behavioral, environmental, psychophysiological, pharmacological, and health factors.

14. A system as said forth in claim 9, said system applied for analysis of slow changes in the pattern of the signal and representation of the results at different levels of detail using the color-coded scales which are understandable to a lay person and a medical professional.

15. A system as said forth in claim 13, in which said different levels of detail include:

a low-resolution Scale I that represents a small number of primary elements;

an intermediate-resolution Scale II that represents serial changes in each of the said elements using a mathematical decomposition into series of representative indicators; and

a high-resolution Scale III that combines serial changes in all elements of at least one of electrocardiographic (ECG), magnetocardiographic (MCG), physical activity, body position, respiration, blood pressure, vasomotor activity, temperature, physiological, and health data to provide complete information about the dynamics of said data.

16. A system as set forth in claim 9, in which:

the measurement of biomagnetic signals of the heart is performed using at least one of the following magnetometers;

fluxgate, optical, laser-optical, and superconducting quantum interference device (SQUID);

or

the measurement of electrocardiographic signals is performed by at least one non-contact sensor.

17. A system as set forth in claim 9 in which said at least one non-contact physiological sensor is capable of exchanging information with an implantable device.

18. A system as set forth in claim 9 in which said at least one non-contact physiological sensor is adapted to be implanted under the skin.

19. A system as set forth in claim 9 in which said at least one non-contact physiological sensor records at least one signal related to at least one of peripheral, spinal, and cranial neural activity.

20. A system as set forth in claim 9 in which said at least one non-contact physiological sensor records at least one signal related to the central (brain) neural activity.

21. A system for analysis of serial changes or trends in at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, said system comprising:

at least one miniaturized, non-contact sensor for measuring said data from at least one fiducial point;

software or hardware for forming a time series from said data obtained from a period of at least several seconds, with the minimum time interval depending on the sensitivity (signal-to-noise ratio) of the said non-contact sensor;

software or hardware for extracting a typical pattern or waveform or a set of coefficients or wavelets representing characteristic features of the said data during the corresponding time interval using an averaging or another mathematical transformation;

software or hardware for forming time series from said typical patterns or waveforms or coefficients or wavelets representing consecutive time intervals; and

software or hardware for analyzing trends or changes in said time series using a mathematical transformation.

22. A system set forth in claim 21 said system comprised of software and hardware for filtering at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic data, blood pressure, vasomotor activity, physical activity, body position, respiration, temperature, physiological, and health data, rejecting noisy or corrupted data prior to said extraction of typical patterns.

23. A system as set forth in claim 21 in which said mathematical transformation and mathematical modeling includes at least one of the following methods:

averaging the signal over a period of at least several cardiac beats (seconds), orthogonal and non-orthogonal decompositions, principal component analysis, mappings and projections, wavelet transform, Fourier transform, Laplace transform, and Hilbert transform, linear and non-linear correlations, linear and non-linear regression models, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, and probabilistic methods, Bayesian probability and Mahalanobis distance, pattern recognition, methods of artificial intelligence, fuzzy logic, neural networks, expert systems, statistical estimators, filtering techniques, and hybrid artificial intelligence systems.

24. A system as said forth in claim 21, said system comprising software or hardware for:

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.

25. A system as said forth in claim 24, said detecting of fiducial points is performed by measuring at least one of the following signals:

mechanical movements of the chest and other parts of the human body measured by a 1, 2, or 3-axial accelerometer, giant magnetic resistance (GMR), or another sensor determining position and movements of the chest and body, mechanical sounds of the heart, electrical signals generated by the heart, light, infrared, and ultrasound signals measuring cardiac mechanical activity.

26. A portable system for registering at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, respiration, temperature, blood pressure, central neural activity, spinal cord activity, peripheral neural activity, cranial nerve activity, neural ganglia activity, vasomotor activity, physiological, and health data, extracting and representing the most significant parameters from series of cardiac beats, said system comprising software or hardware for:

at least one sensor for collecting the data for at least one fiducial point over a period of at least several seconds with the minimum collection time being dependent on the system sensitivity (signal-to-noise ratio); and

a processor for extracting typical patterns from these data using a mathematical transformation.

27. A portable system as set forth in claim 26 in which said extracting of typical patterns includes:

at least one sensor for finding fiducial points indicative of specific phase of the physiological cycle;

a processor for aligning the said data using said fiducial points; and

applying a mathematical transformation to the said aligned time series of said data.

28. A portable system as set forth in claim 26 in which said mathematical transformation and mathematical modeling includes at least one of the following methods:

averaging the signal over a period of at least several cardiac beats, orthogonal and non-orthogonal decompositions, principal component analysis, mappings and projections, wavelet transform, Fourier transform, Laplace transform, and Hilbert transform, linear and non-linear correlations, linear and non-linear regression models, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, and probabilistic methods, Bayesian probability and Mahalanobis distance, pattern recognition, methods of artificial intelligence, fuzzy logic, neural networks, expert systems, statistical estimators, filtering techniques, and hybrid artificial intelligence systems.

29. A portable system as set forth in claim 28 in which said mathematical transformation and mathematical modeling includes at least one of the following methods:

averaging the signal over a period of at least several cardiac beats, orthogonal and non-orthogonal decompositions, principal component analysis, mappings and projections, wavelet transform, Fourier transform, Laplace transform, and Hilbert transform, linear and non-linear correlations, linear and non-linear regression models, analysis of variance, cluster analysis, factor analysis, canonical analysis, regression and discriminant function analyses, and probabilistic methods, Bayesian probability and Mahalanobis distance, pattern recognition, methods of artificial intelligence, fuzzy logic, neural networks and expert systems, statistical estimators, filtering techniques, and hybrid artificial intelligence systems.

30. A portable system set forth in claim 28, said system comprising a processor for:

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.

31. A portable system set forth in claim 28, said system comprising software or hardware for:

detecting fiducial points indicative of a specific phase of the physiological cycle;

time-aligning the said data using said fiducial points; and

applying said mathematical transformation to extract typical patterns from said time-aligned series of said data.

32. A portable system set forth in claim 29, in which said detecting of fiducial points is performed by measuring at least one of the following signals:

mechanical movements of the chest and other parts of the human body measured by a 1, 2, or 3-axial accelerometer, giant magnetic resistance (GMR), optical sensor, transmitter and receiver of radio waves, or another sensor determining position and movements of the chest and body, mechanical sounds of the heart, electrical signals generated by the heart, light, infrared, and ultrasound signals measuring cardiac mechanical activity.

33. A portable system set forth in claim 26, said system used as a simple diagnostic and screening tool during regular daily activities and during night-time, for analysis of slow, long-term trends in cardiac electromagnetic signals that develop gradually over periods of days-weeks-months, for detection of slow changes and abnormalities in the pattern of at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, blood pressure, vasomotor activity, respiration, temperature, physiological, and health data, including changes in the heart rate, amplitude, polarity, and duration of P, Q, R, S, T and U waves, ST-segment, and the intervals between the peaks (including PR, QRS, and QT intervals), for periodic personal self-examination for subjects at risk or with a history of cardiovascular disease, for monitoring slow changes in the said data due to development of cardiovascular disease, treatment results, physical exercises, behavioral, environmental, psychophysiological, pharmacological, and health factors.

34. A portable system set forth in claim 26, said system applied for analysis of slow changes in the pattern of the signal and representation of the results at different levels of detail using the color-coded scales which are understandable to a lay person and a medical professional.

35. A portable system set forth in claim 32, in which said different levels of detail include:

a low-resolution Scale I that represents a small number of primary elements;

an intermediate-resolution Scale II that represents serial changes in each of the said elements using a mathematical decomposition into series of representative indicators; and

a high-resolution Scale III that combines serial changes in all elements of the said data to provide complete information about the dynamics of the signal

36. A portable system as set forth in claim 26 that is wearable, being worn in a pocket or attached to the clothes or a belt or attached using a stripe.

37. A portable system as set forth in claim 26 that includes:

data acquisition unit measuring at least one of electrocardiographic (ECG), magnetocardiographic (MCG), blood pressure, vasomotor activity, physical activity, body position, respiration, temperature, physiological, and health data;

processing unit collecting the data and extracting the typical patterns;

communication unit sending the data and receiving information from other devices.

38. A portable system as set forth in claim 36, in which:

measurement of biomagnetic signals is performed using at least one of the following sensors;

fluxgate, optical, laser-optical, and superconducting quantum interference devices (SQUIDs);

or

measurement of electrocardiographic signals is performed by a non-contact sensor.

39. A portable system as set forth in claim 36, in which communication with other devices is wireless.

40. A portable system as set forth in claim 36, which also has sensors for measuring at least one of the following signals:

a 1, 2, or 3-axial accelerometer, giant magnetic resistance (GMR), or another sensor determining position and movements of the chest and body, and physical activity of a subject, optical light or infra-red sensor for measuring cardiac mechanical activity, sonic or ultra-sonic sensors for measuring heart and lung sounds, blood flow, and blood pressure, non-invasive, non-contact sensors for measuring glucose, hemoglobin, and other biochemical, hormonal, biophysical, respiratory, genetic, proteomic, environmental, and health data.

41. A portable system for analysis of serial changes or trends in at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, said system comprising:

at least one miniaturized, non-contact sensor for measuring said data from at least one fiducial point in a physiological cycle;

software or hardware for forming a time series from said data obtained from a period of at least several physiological cycles;

software or hardware for extracting a typical pattern or waveform or a set of coefficients or wavelets representing characteristic features of the said data during the corresponding time interval using an averaging or another mathematical transformation;

software or hardware for forming time series from said typical patterns or waveforms or coefficients or wavelets representing consecutive time intervals; and

software or hardware for analyzing trends or changes in said time series using a mathematical transformation.

42. A system set forth in claim 41 said system comprised of software and hardware for filtering at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, rejecting noisy or corrupted data prior to said extraction of typical patterns.

43. A portable system for dynamic long-term analysis of serial changes or trends in at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, said system comprising:

at least one miniaturized, non-contact sensor for measuring said data from at least one fiducial point in a physiological cycle;

software or hardware for forming a time series from said data obtained from a period of at least several days;

software or hardware for extracting a typical pattern of at least one waveform, coefficient, and wavelet representing characteristic features of the said data during the corresponding time interval using at least one method selected from averaging, mathematical transformation, modeling, statistics, and artificial intelligence;

software or hardware for forming time series from said at least one typical pattern, waveform, coefficient, and wavelet representing consecutive time intervals;

software or hardware for analyzing trends or changes in said time series using at least one method selected from mathematical transformation, modeling, statistics and artificial intelligence; and

a communications unit for sending at least one parameter of said data to a computer device for processing and detailed analysis of serial changes in said data, said computer device using at least one of the methods selected from mathematical decomposition, modeling, statistics, and methods of artificial intelligence.

44. A system as set forth in claim 43 in which said computer device is selected from a specialized computer, a personal computer, a computer organizer (PDA), a cell phone, a smart phone, a group of computers connected via at least one of a local network, wireless network, and the Internet.

45. A method for dynamic analysis of serial changes in at least one of electrocardiographic (ECG), magnetocardiographic (MCG), mechanocardiographic, ballistocardiographic data, physical activity, body position, respiration, temperature, blood pressure, vasomotor activity, physiological, and health data, said method comprising:

registering at least one type of said data using at least one non-contact sensor over a period of at least several days so that the minimum registration time is determined by the sensor sensitivity (signal-to-noise ratio);

determining at least one typical pattern, waveform, coefficient, and wavelet by applying at least one method selected from mathematical transformation, modeling, statistics, and artificial intelligence to at least one of said registered data over a period of at least several days;

forming a time series from said at least one typical pattern, waveform, coefficient, and wavelet corresponding to the consecutive time intervals; and

characterizing trends in said time series using at least one method selected from mathematical transformation, modeling, statistics, and artificial intelligence.

46. A method as set forth in claim 45 in which said serial changes in data are communicated to a network of computers for processing and detailed analysis of the serial changes in a higher level of resolution analysis.

47. A method as set forth in claim 45 in which said registering of data occurs over a period of many days.