SYSTEM AND METHOD FOR ANALYZING COMPLEX CURVATURE OF ECG CURVES

• US 20100069767A1
• Filed: 11/28/2007
• Published: 03/18/2010
• Est. Priority Date: 11/30/2006
• Status: Active Grant

• Current Assignee: Aalborg Universitet
• Original Assignee: Aalborg Universitet
• Inventors: Xue, Joel Q., Toft, Egon, Struijk, Johannes Jan, Andersen, Mads Peter, Graff, Claus, Hardahl, Thomas Bork
• Granted Patent: US 8,260,406 B2
• Time in Patent Office: Days
• US Class Current: 600/509
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/355 Detecting T-waves
• CPC Class Codes: A61B 5/7239 using differentiation including higher order derivatives
• CPC Class Codes: G06F 2218/10 by analysing the shape of a waveform, e.g. extracting parameters relating to peaks

Abstract

The present invention relates to a system and a method for analysis of ECG curvature, which system involves a mathematical analysis, which analysis comprises the following features where a number of different parameters are isolated and stored in computer means, where a first number of parameters are selected from at least one main group, where the selected parameters are combined in at least a first mathematical analysis. It is the object of the invention to improve mathematical analysis of ECG curvature particular for complex parts of the curvature such as notches or concavities. It is a further object of the invention to improve automatic quantification of symmetry, flatness and durations of ECG curvature. This can be achieved if the parameters are selected from groups of symmetry, flatness, duration and/or complexity, which parameters are used as input to an algorithm, where concavity intervals are evaluated on subsegments of the ECG segment and form the basis of up- and downwards concavity quantification, where the system based on an algorithm detects and quantifies concavity on ECG curvatures. Herby can be achieved that a complex curvature can be analysed and the complex curvature might indicate symptoms of diseases which are very difficult to read on ECG curvature by existing technology. This can lead to a more precise analysis where existing systems might by analysing the curvature give a first indication of a symptom. By then performing an analysis of the complex part of the curvature, further symptoms might be indicated.

Claims

1. System for analysis of ECG curvature comprising a computer system, which computer system performs a mathematical analysis, which mathematical analysis comprises at least the following steps:

a;

the system comprises input means connected to at least one ECG source,

b;

a number of different parameters are isolated and stored in the computer system,

c;

the computer system select a first number of parameters from at least one main group,

d;

the computer system combine the selected parameters in at least a first mathematical analysis,

whereby the parameters are selected from groups of symmetry (S1-S72), flatness (F1-F91), duration (D1-D47) and/or complexity (C1-C468),

e;

extract a segment in of the ECG curvature in the mathematical analysis,

f;

which segment is corresponding to a interval between defined electrocardiographic points,

g;

which segment is used as input to a first algorithm,

h;

where a second fiducial point algorithm automatically calculates fiducial points at the segment,

i;

where concavity intervals are evaluated on subsegments of the ECG segment and form the basis of up- and downwards concavity quantification,

j;

where the system based on at least the first and the second algorithm detects and quantifies concavity on ECG curvatures.

2. A system according to claim 1, wherein at least one segment of the ECG repolarization signal is selected, corresponding to the interval between the Twave onset and Twave offset, which segment is normalized so that the amplitude range (y) is equal to a first predefined value in the selected interval.

3. A system according to claim 2, wherein at least one segment of the ECG repolarization signal is selected, corresponding to the interval between the Twave onset and Twave offset, which segment is normalized so that the duration (x) is equal to a second predefined value in the selected interval.

4. A system according to one of the claim 3, wherein the first and second derivatives (with respect to time) of the ECG repolarization segment are determined on the interval between the Twave onset and Twave offset, where the two derivative signals are subsequently used to calculate a radius of curvature (Rcurv) signal using the equation 1:

5. A system according to claim 4, wherein local maxima (LmRcurv) on the Rcurv signal between any two repolarization fiducial points is determined to identify places of local downwards concavity on a repolarization segment of an ECG.

6. A system according to claim 5, wherein local minima on the Rcurv signal between any two repolarization fiducial points are determined to identify places of local upwards concavity on a repolarization segment of an ECG.

7. A system according to claim 6, wherein the Rcurv signal is investigated further to identify any contiguous configuration of local maximum neighbored by local minimum on the Rcurv signal, where the values of the identified maximum and minimum are used to evaluate the size of any concavity on a repolarization segment of an ECG.

8. A system according to claim 4, wherein a window is defined in a curve segment stating at a first defined period before Twave peak and ending at a second defined period after Twave peak, in which window the curvature is excluded from further analysis of the Rcurv signal, and where any number of LmRcurv signal values with amplitudes below zero is excluded from further analysis, where any number of LmRcurv signal values, which correspond in time to ECG repolarization amplitudes below any given threshold are excluded from further analysis.

9. A system according to claim 4, wherein positive segment durations on the Rcurv on either side of all reference points are multiplied by the value of the corresponding LmRcurv values to obtain certainty measures (Cm) of the presence of local downward concavities.

10. A system according to claim 9, wherein certainty measures exceeding a threshold (Thr) are defined as places around which concavity intervals are defined and local downward concavities are quantified.

11. A system according to claim 10, wherein no certainty measures exceeding a threshold (Thr) are present, where a first subsegment is generated by two piecewise tilted segments, where a first tilted subsegment (T_1tilt) is given by the repolarization interval between the Twave onset and Twave peak from which the linear function between the two endpoints is subtracted, where a second tilted subsegment (T_2tilt) is given by the repolarization interval between the Twave peak and Twave offset from which the linear function between the two endpoints is subtracted.

12. A system according to claim 11, wherein an upwards concavity score is calculated, where the least negative minimum on either piecewise repolarization subsegment is used as an initial upwards concavity score (IuS), which score is corrected to obtain a final upwards concavity score (CuS), where the relationship is given by equation 3.

13. A system according to claim 10, wherein certainty measures exceeding a threshold (Thr) are present, where at least one third subsegment is generated by piecewise tilted segments in concavity intervals, from which the linear function between the two endpoints in concavity intervals is subtracted.

14. A system according to claim 13, wherein all of the tilted subsegments (N) are evaluated and maximum values (T_3tiltmax) to (T_Ntiltmax) on the tilted subsegments in the concavity intervals are determined, where the overall maximum value is the final concave down score (CdS) of the ECG repolarization segment given by equation 2:

CdS=max([T_{3tilt max} . . . T_{Ntilt max}]). 

 

Equation 2

15. A system according to claim 1, wherein the peakedness of the T wave peak (Peak_Curvature) is evaluated by calculating the radius of curvature value (Rcurv) in Tpeak, which radius of curvature is given by:

16. A system according to claim 1, wherein the symmetric properties of the T wave are evaluated by comparing the T wave curve on the right side of Tpeak with the T wave curve on the left side of Tpeak, which comparison of left and right side is performed at any representation of the T wave, such as the time domain ECG signal, the first derivative of the signal, the second derivative of the signal or any combination of these such as the radius of curvature of the signal Rcurv.

17. A system according to claim 16, wherein the Twave representation of the right side of Tpeak is flipped around a vertical line through Tpeak to cover the left side of Tpeak.

18. A system according to claim 17, wherein a first or second derivative is used for the mathematical analysis, where the Twave representation of the right side of Tpeak is flipped around the x axis to compensate for the sign difference between the two sides.

19. A system according to claim 1, wherein the system analyzes ECG curvature for drug influence.

20. A system for analysing ECG curvature according to claim 1, wherein the system analyzes ECG curvature for Long QT Syndrome.

21. A method for analysis of ECG curvature, which method determines a number of parameters in a computer system, where the method for analysing the ECG curvature at least incorporates the steps of:

a. receiving at least one ECG curvature from at least one source,

b;

indicating a number of different parameters determined from the received ECG curvature,

c;

storing the parameters in storage means,

d;

selecting disease specific parameters in the storage means,

e;

selecting a first number of parameters from at least one main group, which main groups comprise parameters of symmetry, flatness, duration and/or complexity,

f;

combining the selected parameters in at least a first mathematical analysis,

whereby the groups of parameters comprises parameters of symmetry (S1-S72), flatness (F1F91), duration (D1-D47) and/or complexity (C1-C468),

g;

the mathematical analysis is extracting a segment of the ECG curvature, which segment is corresponding to the interval between defined electrocardiographic points,

h;

which mathematical analysis uses the segment as input to a first algorithm,

i;

which mathematical analysis calculates fiducial points at the segment, by a second fiducial point algorithm,

j;

which mathematical analysis evaluates concavity intervals on subsegments of the ECG segment and form the basis of upwards and downwards concavity quantification,

k;

which mathematical analysis indicates symptoms of electrocardiographic abnormalities by detection and quantification of at least one concavity of an ECG curvature.