EKG system and method using statistical analysis of heartbeats and topographic mapping of body surface potentials

• US 4,974,598 A
• Filed: 03/28/1989
• Issued: 12/04/1990
• Est. Priority Date: 04/22/1988
• Status: Expired due to Term

• Current Assignee: Heart Map Incorporated
• Original Assignee: Heart Map Incorporated
• Inventors: John, Erwin R.
• Primary Examiner(s): Jaworski, Francis
• Assistant Examiner(s): MANUEL, GEORGE C
• Application Number:
  US07/329,583
• Time in Patent Office: 616 Days
• Field of Search: 128/695, 128/696, 128/700, 128/702, 128/703, 128/704, 128/705, 128/706, 128/709, 364/413.05, 364/413.06
• US Class Current: 600/509
• CPC Class Codes: A61B 5/282 Holders for multiple electrodes for introduction into the body A61B5/287
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: G06F 18/00 Pattern recognition

Abstract

The Heart State Analyzer (HSA) is a system and method in medical non-invasive electrocardiographic (EKG) analysis of human heart beats for the early detection of certain heart diseases in which a large number of electrodes, for example, 32 to 64, are attached on the chest, back and sides of the patient, i.e., "body surface". The electrical signals detected by the electrodes ae converted to digital data, treated to remove muscle artifact and other noise, and then analyzed mathematically to determine the presence or absence of abnormal body surface potential distributions, or of unusual beat morphologies, compared statistically to the self-norm "typical beat" of the patient and also compared to a data base compiled from comparable normal population groups. The results of the statistical analysis are displayed as topographical maps of the body surface, color coded to represent the presence of significant derivations from the norms, defined as "abnormality", i.e., abnormal spatio-temporal patterns of voltages on the body surface, or as waveshape or histogram displays of features, similarly Z-transformed and color coded. Discriminant functions, stored in the heart state analyzer, estimate the relative probability of various cardiac pathologies.

Claims

1. An electrocardiographic system for the detection of heart disease in a patient including:

at least 20 electrodes to detect heart beat activity analog electrical signals, adapted to be removably connected to the skin of the patient in separated locations around the heart on the chest, sides and back of the patient (torso);

amplification means to amplify the analog electrical signals detected at each electrode;

digital conversion means to convert the said amplified electrical signals into EKG patient digital data;

norm storage means to receive and store normative EKG digital data representing normal heart activity electrical signals and features extracted from such signals during the time of the cardiac cycle;

digital computer statistical analysis means comprising space-time domain analysis means space-time domain and space-frequency domain analysis means to analyze the said patient digital data on a statistical basis compared to said stored normative data;

said space-frequency domain analysis means including Fourier transform (FT) means for performing from 0 to 250 Hz a fourier transform of the patient digital data to form FT transformed data;

subclass means to form homogenous subclasses of heart beats using the Fourier transform transformed data;

transform means to statistically compare said subclasses with said normative data to produce transformed Fourier transform data;

inverse Fourier transform means to inverse fourier transform said transformed Fourier transform data to produce inverse Fourier transform data;

said space-time domain analysis including means to form homogenous subclasses of EKG waveshapes and transform means to statistically compare said patient digital data with said normative data in each subclass;

topographic map display means to display the results of said statistical analysis of the signals from each electrode on a topographic map representing the chest, sides and back of the patient, each map location corresponding to the location of corresponding connected electrode; and

means to classify the type of coronary artery disease by computing discriminant functions.

4. An electrocardiographic system as in claims 1, 2 or 3 and further including multivariate extraction feature means to form a heart state vector calculated from the Mahalanobis length which takes account of the covariance matrix of the features extracted by said multivariate extraction feature means.

5. An electrocardiographic system as in claims 1, 2 or 3 and further including multivariate extract feature means to form a heart state vector and Z-transformation means operatable with multi-variant feature means to Z-transform the multi-variant features and form a multivariate Z-score.

6. An electrocardiographic system as in claims 1, 2 or 3 wherein the number of electrodes is in the range of 20-64, each electrode being connected to a separate communication channel.

7. An electrocardiographic system as in claims 1, 2 or 3 wherein said normative data consists of data from a normal population without evidence of heart disease by current standards.

8. An electrocardiographic system as in claims 1, 2 or 3 wherein said display means includes a monitor and the statistical results are expressed in colors representing positive and negative abnormality and the extent thereof.

9. An electrocardiographic system as in claims 1, 2 or 3 wherein said system also includes fiducial selection means permitting the operator to select the portion of the heart wave which acts as a fiducial time for the analysis.

10. An electrocardiographic system as in claim 1 wherein said frequency domain transform means and time domain transform means perform (based on) Z-transformations.

11. An electrocardiographic system as in claim 10 wherein said heart activity is of heart beats having P, Q, R, S, T and U portions and said space-frequency domain analysis means further includes feature extraction means to extract features from the Z-transformed FT data, said features being selected from the group consisting of the average R-R interval, QRS amplitude, QRS power, QR power and RS power.

12. An electrocardiographic system as in claim 10 and further including means to calculate the principal components of frequency spectra to obtain a factor analysis of the IFT data and produce factor analysis scores therefrom and means to calculate the Z-transformations of the said factor analysis scores.

2. An electrocardiographic system to detect heart beats having P, Q, R, S, T and U portions as well as those containing some but not all P, Q, R, S and T portions including:

a plurality of electrodes to detect heart activity electrical analog signals adapted to be removably connected to the skin of the patient in separated locations around the heart on the chest, sides and back of the patient (torso);

amplification means to amplify the electrical signals detected at each electrode;

digital conversion means to convert the said amplified electrical signals into EKG patient digital data;

normative storage means to receive and store normal and abnormal digital data representing normal and abnormal heart activity electrical signals during the time of the cardiac cycle;

statistical analysis computer means comprising space-time domain analysis means to analyze the said EKG patient digital data on a statistical basis compared to said stored normal and abnormal data;

said space-time domain analysis means including means to construct homogenous subclasses;

means to extract univariate features selected from the group consisting of the P-Q, Q-R, R-S, S-T, T-U or R-R portions;

P, QRS, T or U amplitudes; and

P, QR, QRS, T or U powers;

means to obtain factor scores of said univariate features and to Z-transform said factor scores as compared to said normal data to form factor Z-scores;

said Z-transform being defined as;

##EQU10## where as Z_ij is the Z-score of factor score a_ij ;

P_ij is patient weighting coefficient denoting the amount of factor j contributing to the EKG wave EKG_i ;

M_ij is the normal data weighting coefficient denoting the amount of factor j contributing to the EKG wave EKG_i and σ

_ij is the standard deviation of the normative data for the factor j contribution to EKG wave EKG_i,

topographic map display means to display the said factor Z-scores on a topographic map representing the torso of the patient, each map location corresponding to the location of a connected electrode.

3. An electrocardiographic system to detect heart beats having P, Q, R, S, T and U portions, as well as those beats having some but not all of said portions, for the detection of heart disease in a patient including;

at least 20 electrodes to detect heart beat activity electrical analog signals adapted to be removably connected to the skin of the patient in separated locations around the heart on the chest, sides and back of the patient (torso);

amplification means to amplify the electrical signals detected at each electrode;

digital conversion means to convert the said amplified electrical signals into ECG patient digital data;

normative storage means to receive and store normal digital data representing normal heart activity electrical signals during the time of the cardiac cycle;

digital computer statistical analysis means comprising space-time domain analysis means to analyze the said patient digital data on a statistical basis compared to said stored normal data;

said space-time domain analysis means including means to construct homogenous subclasses of heart beats;

means to extract univariate features selected from the group consisting of the P-Q, Q-R, R-S, S-T, T-U, or R-R portions, P, QRS, T or U amplitude, P, QR, RS, QRS, T or U power;

means to obtain factor scores of said univariate features and to Z-transform said factor scores and compare to said normal data to form factor Z-scores;

where said means to obtain factor scores a_ij operates in accordance with the equation ##EQU11## where EKG_i is the voltage of the cardiac cycle seen at any position i;

a_ij is the factor score defining the contribution of factor j to EKG_ij ;

F_j (t) is the waveshape of factor j throughout the cardiac cycle; and

, where said Z-transform is defined as;

##EQU12## where as Z_ij is the Z-score of factor score a_ij ;

P_ij is patient factor score (weighting coefficient) denoting the amount of factor j contributing to the EKG wave EKG_i ;

M_ij is the normal data weighting coefficient denoting the average amount of factor j normally contributing to the EKG wave EKG_i and σ

_ij is the standard deviation of the normal data for the factor j contribution to EKG wave EKG_ij, and

display means to display the said factor Z-scores on a topographic map representing the torso of the patient, each map location corresponding to the location of the corresponding electrode.

13. The method of electrodcardiograph EKG examination of a patient for the detection of heart disease including but not limited to the steps of:

removably connecting at least 20 electrodes to detect heart activity electrical signals to the skin of the patient in separated locations around the heart on the chest, sides and back of the patient (torso);

amplifying the analog electrical signals detected at each electrode;

converting the said amplified analog electrical signals into patient EKG digital data;

receiving and storing normal and abnormal digital ECG data representing normal and abnormal heart activity electrical signals during the time of the cardiac cycle;

automatically statistically analyzing the patient EKG data in computer means, in both the space-time domain and space-frequency domain, compared to said stored normal data;

said space-frequency domain analysis including performing a fourier transform of the data to form FT transformed data;

forming homogenous subclasses of the FT transformed data;

Z-transforming said FT transformed data;

inverse fourier transforming said Z transformed FT data;

said space-time domain analysis including forming homogenous subclasses and comparing the patient data with the normal data in each subclass and Z transforming the comparison results; and

displaying the results of said space-frequency domain analysis and said space-time domain analysis of the signals from each electrode on a topographic map representing the torso of the patient, each map location corresponding to the location of the corresponding electrode.

14. An electrocardiographic method as in claim 13 wherein the number of electrodes is in the range of 20-64, each electrode being connected to a separate communication channel.

15. An electrocardiographic method as in claim 13 wherein the digital conversion step includes sampling at least at the millisecond rate in each channel.

16. An electrocardiographic method as in claim 13 wherein said normative data consists of data from a normal population without evidence of heart disease by current standards.

17. An electrocardiographic method as in claim 13 wherein said normal data consists of self-norm data obtained from said patient, which may include abnormal EKG patterns.

18. An electrocardiographic method as in claim 13 wherein the frequency domain analysis and time domain analysis results are expressed in color representing positive and negative abnormality and the extent thereof.

19. An electrocardiographic method as in claim 13 wherein said method also includes the step of selecting the portion of the heart wave which acts as a fiducial time for the analysis.

20. An electrocardiographic method as is claim 13 wherein said heart activity is of heart beats having P, Q, R, S, T and U portions and said space-frequency domain analysis further includes the step of extracting features from the Z-transformed FT data;

said features being selected from the group consisting of the average R-R interval, QRS amplitude, QRS power, QR power and RS power.

21. An electrocardiographic method as in claim 13 and further including inverse Fourier transforming the data and producing IFT data and analyzing the IFT data to produce factor scores and to Z transform the said factor scores.

22. An electrocardiographic method as in claim 13 wherein said space-time domain analysis incudes computing subclasses and computing the average and variances in each subclass.

23. An electrocardiographic method as in claim 13 wherein said space-time domain analysis includes the step of extracting features of said heart beat electrical activity and the variances of said features.