WO2023036447A1

WO2023036447A1 - Method and device for determining excitation potentials of the human heart

Info

Publication number: WO2023036447A1
Application number: PCT/EP2021/075107
Authority: WO
Inventors: Tino KREUTZBERGER; Jan KNOBELOCH
Original assignee: Mycor Gmbh
Priority date: 2021-09-13
Filing date: 2021-09-13
Publication date: 2023-03-16
Also published as: WO2023037006A9; WO2023037006A1

Abstract

The invention relates to a method for determining excitation potentials of the human heart. The invention also relates to a device for carrying out this method. The problem of the present invention is solved according to claim 1, 2 or 10 and by a device according to claim 17.

Description

Method and device for determining excitation potentials of the human heart

The invention relates to a method for determining excitation potentials of the human heart using at least 4 electrodes, which are arranged on the body surface of a person and from which at least three derivatives, preferably bipolar derivatives, of the electrical potential of the excitation of the human heart are determined, wherein the connecting lines between the electrodes arranged on the body surface, which form the at least three leads, span a three-dimensional space. In addition, the invention relates to a device for carrying out the method.

The human heart is made up of myocardial cells, which contract by passing electrical signals through the myocardial cells themselves. The excitation potentials of ventricular excitation and regression of ventricular excitation can be measured on the body surface of a human being. The course of cardiac muscle excitation and excitation recovery is also used to analyze abnormal changes in the heart. For example, the use of an electrocardiogram (ECG) to diagnose myocardial infarction is well known. Methods of cardiogoniometry are also used for the same purpose, with a cardiogoniogram being able to determine indications of an existing infarction or ischemia, i.e. an undersupply of the heart muscle tissue.

From the article "Non-invasive diagnosis of coronary artery disease using cardiogoniometry performed at rest", WM Michael Schühbach et. Al, Swiss Med WKLY 2008; 138 (15-16):230-238 it is known that a cardiac muscle undersupply can be detected via a cardiogoniometric examination determined by altered excitation training in a cardiogoniogram Cardiogoniometry and a cardiogoniometer are also known from European patent EP 0 086429 B1.

In addition to myocardial infarction or ischemia, other heart diseases are known. Amyloidosis, for example, refers to a group of diseases in which deposits of proteins occur in various human organs. In cardiac amyloidosis, the heart muscle thickens and stiffens, preventing the heart from contracting and expanding evenly. The abnormal proteins deposited in the intercellular spaces, which are called amyloid, lead to cardiac insufficiency as the disease progresses, which manifests itself in shortness of breath and a reduction in exercise capacity. There are also other heart diseases that lead to a structural change in the heart muscle tissue.

To date, imaging methods such as echocardiography or cardiac MRI have been used as diagnostic methods for cardiac amyloidosis. An examination under stress is usually also carried out. A heart muscle biopsy and scintigraphy are also used to diagnose cardiac amyloidosis. A further possibility of diagnosing amyloidosis is known according to European patent EP 3 417 295 B1 using a method for extracting and determining proteins of the amyloidosis disease using a biological sample. To diagnose cardiac amyloidosis, this method also requires a biopsy, which is a complex procedure on the heart. All previous methods are associated with relatively great effort and, due to the high costs, are not particularly suitable for screening for cardiac amyloidosis in order to detect the disease even in the early stages.

Proceeding from this prior art, it is the object of the present invention to propose a method for determining excitation potentials of the human heart using at least 4 electrodes, which is suitable for determining data for diagnosing structural diseases of the heart to be deployed. For example, a method for determining data for a diagnostic method for detecting cardiac amyloidosis is to be provided, so that a diagnosis no longer requires technically complex or invasive methods. Finally, a device for carrying out the method should also be specified.

According to a first aspect of the present invention, the stated object is achieved by a generic method mentioned at the outset for determining excitation potentials of the human heart using at least 4 electrodes in that

- at least the electrical potential of the propagation and/or regression of ventricular excitation of the heart is measured in each lead and

- the measured values of the leads are fed to a signal processor in order to obtain a processed excitation potential signal, preferably in each lead,

- the QRS excitation complexes for the propagation of the ventricular excitation of the heart in the at least three leads are detected in the processed excitation potential signal,

- the processed excitation potential signal of the leads is corrected by an offset value in order to obtain a corrected excitation potential signal,

- wherein in the corrected excitation potential signal of the at least three leads the main orientation of the corrected excitation potential is determined, preferably in the form of an electric potential vector RDir with maximum length during the propagation of the ventricular excitation of the QRS excitation complex of the heart in the spanned space and

- a measure for the spatial orientation of the ascertained corrected excitation potential of the ventricular excitation of the QRS excitation potential, preferably of the electric potential vector RDir is determined.

In principle, the method according to the invention can be carried out with at least four electrodes, which can in principle be arranged in any arrangement on the body surface of a person, as long as at least three Derivatives or the associated electrodes each have straight connecting lines that span a space. For example, the leads A, D, Ho, Ve and 1, as shown in FIG. 2, span such a space.

At least four electrodes are preferably used, which are arranged on the body surface of the patient in such a way that they allow the formation of three leads, preferably bipolar leads, which are preferably perpendicular to one another and span a space. However, any arrangement of at least four electrodes that span a space is also possible. For example, the bipolar leads A, D, Ho, Ve and 1 according to FIG. 2 are preferably used. Furthermore, in a preferred variant, the term derivation is always also understood to mean a bipolar derivation. The electrical potential of the ventricular excitation of the heart can be represented three-dimensionally as a loop via at least three leads. By detecting the QRS excitation complex of the propagation of the ventricular excitation of the heart in the signal of the leads, the evaluation of the signal can be limited to the measurement data assigned to this excitation complex. The planned signal processing of the measured values of the derivatives serves to eliminate all possible external interference, for example interference from high-frequency electromagnetic radiation from mobile phones, etc. in the signal as far as possible. In addition, the excitation potential signal is corrected by an offset value so that constant contributions in the excitation potential signal are eliminated and the main orientation of the corrected excitation potential signal in space can be determined correctly and substantially comparably.

The principal orientation of the corrected capture potential signal is preferably determined in the form of an electrical potential vector RDir of maximum length during ventricular capture of the heart. The orientation of the electric potential vector RDir describes the main orientation of the corrected excitation potential signal, ie the direction in which the corrected excitation potential signal has the greatest extent in space. The main direction of the corrected excitation potential signal can be determined not only by a maximum length electric potential vector RDir but also in other ways. For example, by considering the obtained measured values of the corrected excitation potential signal as a cloud of points, the geometric extent of which, starting from the center of gravity of the cloud of points, which has the value zero, is described using eigenvectors. Eigenvectors are perpendicular to each other. The eigenvector with the largest eigenvalue can also be used to determine the principal orientation of the corrected excitation potential signal. The latter eigenvector and the electric potential vector RDir are not mathematically identical. However, both essentially indicate the main orientation of the corrected excitation potential signal during ventricular excitation in the space formed by the chosen leads.

A measure of its orientation or position in space can then be determined from the main orientation of the corrected excitation potential signal. The measure of the spatial orientation can be, for example, an angle or another value, for example also a vector. The method according to the invention can then be used in a diagnostic method by comparison with a reference angle or a reference vector.

For example, cardiac amyloidosis has been shown to affect the principal orientation of the excitation potential of the spread of ventricular excitation of the heart, so a measure of the spatial orientation of the principal orientation of the corrected excitation potential signal, for example by comparison to a reference orientation, or reference angle, indicates cardiac amyloidosis declining, deviating ventricular excitation of the heart can be concluded in a diagnostic procedure.

The method according to the invention can also be used in diagnostic methods of other structural excitation disorders of the human heart. In particular, due to the low cost of screening methods with the Methods according to the invention, for example for cardiac amyloidosis, can be carried out easily, since a large number of people can be examined at low cost.

According to a further aspect of the present invention, alternatively or cumulatively to the detection of the QRS excitation complexes in the processed excitation potential signal, the T-wave of the excitation regressions of the heart chambers is detected in the at least three leads,

- the processed excitation potential signal is corrected by an offset value in order to obtain a corrected excitation potential signal,

- wherein in the corrected excitation potential signal of the excitation recovery of the heart of the at least three leads, the main orientation of the corrected excitation potential of the excitation recovery of the heart chambers, preferably in the form of an electric potential vector TDir with maximum length in the spanned space during the excitation recovery of the heart chambers is determined and

- a measure for the alignment of the main alignment of the corrected excitation potential signal of the excitation recovery of the T-wave, preferably the alignment of the electric potential vector TDir in the space spanned by the at least three leads is determined.

It has been shown that a measure of the alignment of the main alignment of the corrected excitation potential signal of the excitation recovery of the heart chambers (the T-waves), preferably the alignment of the electrical potential vector TDir with maximum length, can indicate a changed excitation recovery of the chambers of the heart. The determination of the main orientation of QRS excitation complexes via eigenvectors can be used here analogously for the T waves. The latter eigenvectors and the electric potential vector TDir are not mathematically identical. However, both essentially give the main orientation of the corrected excitation potential signal of the De-excitation of the heart in the space formed by the selected leads.

The respective electrical potential vectors RDir, TDir or both together can, for example, also be used in a comparison with one or more reference vectors. The reference vectors are preferably determined by a multivariate discriminant analysis. Other analysis options using the electrical potential vectors RDir, TDir or both vectors are also conceivable.

According to one embodiment of the method, at least one angle of the electrical potential vectors RDir and/or TDir in the spanned space is determined, for example. Determining an angle in space is a particularly simple measure for determining the main orientation of the corrected excitation potential. At least one of these angles can be used to discriminate against or belong to a group affected by amyloidosis.

A simplified way of determining the orientation of the electrical potential vector RDir or TDir is achieved by determining the determined electrical potential vector RDir and/or TDir in spherical coordinates in the spanned space and using the elevation angle 0 for the orientation of the electrical potential vector RDir or TDir and/or the azimuth angle cp can be determined. Other coordinate systems, such as cylindrical coordinates, can also be used in analyzing the principal orientation of the corrected excitation potential signals.

In addition, according to a further embodiment of the method, there is the possibility of determining at least one angle between the electrical potential vectors RDir and TDir. This angle can also be used in a diagnostic procedure for cardiac amyloidosis or another heart disease by comparison with reference values.

According to a further embodiment of the method, the time of the start of the QRS excitation complex and/or the T-wave is determined in at least one derivation of the processed excitation potential signal and the measured electrical potential value is determined in at least one derivation at the time of the start of the QRS excitation complex or the T-wave used as an offset value for the correction of the excitation potential in the respective lead. The beginning of the QRS excitation complex can be determined in the individual leads. The determination of the beginning of the QRS excitation complexes can also be improved statistically, for example by averaging, in the respective leads or independently of the leads. The determined candidates for the beginning of QRS excitation complexes can be combined statistically, for example by averaging, into a QRS excitation complex that is representative for the measurement, or this can be selected from the candidates.

Furthermore, according to a further embodiment of the method, electrical potentials are measured for a plurality of propagations and regressions of the ventricular excitation of the heart in each lead and the main orientation of the corrected excitation potential signal is determined using statistical measurement variables, for example medians or mean values, which are derived from the majority of the propagations and regressions measured the ventricular excitation of the heart were determined. This generally leads to higher measurement certainty with regard to determining the orientation of the corrected excitation potential signals.

The offset values used to determine the corrected excitation potential signals can be determined, for example, by detecting the QRS excitation complexes. They preferably correspond to the amplitude of the beginning of the respective QRS excitation complex in at least one derivation. In preparation for the detection of the QRS excitation complexes, for example, selection QRS-typical frequencies formed a QRS detection signal for at least one derivation. An amplification of the relevant QRS areas is then carried out in the QRS detection signals formed. The precise start and end points of the QRS ranges are determined using adaptive threshold values, which can be calculated from statistical measurement variables in the QRS detection signal. The starting points of the QRS excitation complexes form the offset values. The same procedure can also be followed when determining the start of the T-wave, whose offset value can be used as an alternative.

According to a further aspect of the present invention, as an alternative or in addition to the methods described so far, a space curve of the electrical potential is determined during the propagation and regression of the ventricular excitation of the heart in the spanned space, the space curve of the electrical potential is subjected to a shape analysis, with a projection being used for the shape analysis of the space curve of the electric potential is carried out in the main plane of the space curve or in at least one plane in space and the projection of the space curve in the respective plane is subjected to a shape analysis. The space curve has at least three dimensions due to the at least 3 derivatives used. When using 4 derivatives, the space curve can also have 4 dimensions, for example. The principal plane is defined by N coordinate axes, where N is at least 2. A comparison with reference values or a reference space curve is preferably carried out.

It was found by the inventors that the disturbances in the propagation and regression of ventricular excitation not only affect the main orientation of the excitation potentials RDir and/or TDir, but in particular also result in a change in the space curve of the electrical potential of ventricular excitation or regression of the heart.

By projecting the space curve of electric potential into the main plane of the space curve or into at least one plane in space, the problem of Analysis of the space curve can be reduced to a form analysis in two-dimensional space.

A projection into the main plane is preferably carried out using principal component analysis. Here, the main plane is the plane which, after a main component analysis, preferably results from the first two main components of the three-dimensional space curve of the electric potential. In the principal component analysis, a three-dimensional statistical variable, such as the space curve of the electric potential, is approximated and the principal components are calculated in this way. In the present case, for example, the axes of the main plane can consist of a linear combination of the spatial axes defined by the at least three derivations. If, for example, the space curve of the electrical potential of ventricular excitation essentially consists of an excitation loop lying obliquely in space, the projection of the space curve of the electrical potential into the main plane leads to a plane in which the space curve of ventricular excitation essentially lies. In addition, a projection can also be used, for example, in any plane in space, in order to only carry out an analysis in at least two dimensions in the shape analysis.

According to a further embodiment of the method according to the third teaching, the shape analysis is carried out on the basis of the aforementioned projection of the space curve. This analysis can be performed using shape-defining parameters. These are, for example, roundness, compactness or the expansion of the space curve in at least one space dimension. These parameters can be determined, for example, directly from the entire space curve or from the determined QRS and T complexes.

Furthermore, according to one embodiment, the shape analysis of the projection of the space curve can be carried out using a simplified space curve generated using Fourier descriptors, the projection of the space curve of the electrical potential being represented by the complex value c(m] in the selected plane, so that applies : c(m) = x(m) + iy(m) , [1] where x(m], y(m) represent the mth measured value of the electric potential in the respective projection and for the Fourier descriptors in the plane for N measured values applies:

When using Fourier descriptors to prepare the measurement data for the shape analysis, the measured space curve of the electrical potential of the ventricular excitation and regression of the heart is transformed into the frequency space by a discrete Fourier transformation and the associated Fourier descriptors are thus determined. An exact image of the measured space curve of the electrical potential can then be obtained by using all Fourier descriptors in the inverse transformation. However, if the number of Fourier descriptors for the inverse transformation is reduced, reduced basic forms of the measured space curves of the electrical potentials are obtained, which allow a simplified evaluation of the shape analysis of the space curves.

Similar to a complex filter, a reduced number of Fourier descriptors can smooth the measured spatial curves in the respective plane and reduce them to pathological or healthy basic forms.

For example, a particularly simple shape analysis of the two-dimensional projections of the space curve can be carried out by a further embodiment in that the roundness of the projection of the space curve is determined as a shape analysis in the respective plane, with the roundness being the ratio of the greatest width and the greatest height of the projection of the space curve is equivalent to. Roundness thus provides a particularly simple shape analysis parameter. Alternatively or in addition to this, the compactness of the projection of the space curve in the respective plane from the ratio of the square of the circumference of the projection of the space curve and the area enclosed by the projection of the space curve can also be used as a shape analysis. The compactness is at a minimum in the case of an approximately circular course of the projection of the space curve in the respective plane, since the enclosed area of the space curve is then at a maximum. The compactness can also be used to infer a changed electrical excitation of the heart, for example caused by cardiac amyloidosis.

In addition, the shape analysis can also be carried out using determined Fourier descriptors, in that a feature vector is formed with the determined Fourier descriptors as components and a comparison is carried out with at least one reference feature vector. Since the determined Fourier descriptors are used here directly for the shape analysis, the step of generating a space curve using the Fourier descriptors and the subsequent shape analysis of the space curve generated from the Fourier descriptors is omitted.

According to a further embodiment of the method, the signal processing includes at least smoothing and/or filtering of the measured electrical potentials of the at least three leads, so that the main disturbances, such as movement artifacts, mains frequency disturbances or high-frequency disturbances, such as high-frequency interference from the patient, are present in the processed excitation potential signal, for example Muscle tremors should be reduced as much as possible in all three leads.

In addition, ventricular extrasystoles of the human heart can interfere with the analysis for a structural heart disease such as amyloidosis. In a further embodiment of the method, the signal processing preferably includes a classification of the measured electrical potentials of the at least three Derivations, so that ventricular excitations of the heart, in particular ventricular extrasystoles, which deviate from the normal excitation complex, in particular the QRS complex, are not used for further excitation complex analysis. A corresponding classification can be carried out, for example, by analyzing the electrical potentials in relation to the characteristic morphology of the excitation complex, in particular the QRS complex, and in the event of a deviation this can result in this measurement of electrical potentials not being permitted for further analysis. The classification can also exclude supraventricular extrasystoles or other forms of extrasystoles that lead to a change in the excitation complex from further analysis.

Finally, according to a fourth aspect, the above-mentioned object is also achieved by a device for carrying out a method according to the invention

- Means for measuring electrical potentials of at least 4 electrodes arranged on the body surface of a person for the determination of at least three leads, preferably at least three bipolar leads, wherein the straight lines connecting the electrodes arranged on the body surface, which form the at least three leads, one span three-dimensional space,

- Means for signal processing of the measured electrical potentials of the at least three leads to generate a processed excitation potential signal, preferably in each lead,

-means for correcting the processed excitation potential signal of the leads by an offset value in order to obtain a corrected excitation potential signal,

- Means for determining an electrical potential vector RDir of the QRS excitation complex of the heart and/or an electrical potential vector TDir of the T-wave of the excitation recovery of the heart and/or means for determining the space curve of the electrical potential from the corrected excitation potential signal in the space spanned by the electrodes , - means for determining in space an electric potential vector RDir for the main orientation of the corrected excitation potential of ventricular excitation and/or an electric potential vector for the main orientation of de-excitation of the heart TDir and

- Means for determining a measure for the spatial orientation of the determined electrical potential vector RDir and/or the determined electrical potential vector TDir and/or

- Means for analyzing the shape of the projection of the space curve of the electrical potential in a main plane or in at least one plane in space, the means for analyzing the shape preferably carrying out a comparison with a reference space curve and means for outputting the determined extent of the alignment of the electrical potential vectors or the results of the shape analysis, solved.

The method according to the invention can be carried out with the device. The device is characterized by a low expenditure on equipment and can therefore be used particularly well in a diagnostic method. Due to the different possibilities for analyzing the measured electrical excitation potential signals by determining the alignment of the corrected excitation potentials on the one hand and the shape analysis of the space curve of the electrical potential on the other hand and their combined application, a very precise analysis with regard to the presence of pathologically altered electrical excitation potentials of the heart with the Device allows.

The device can be further developed, for example, in that it analyzes the measurement data online, ie during the measurement of the derivations, and can also carry out a more extensive offline analysis of the data determined. The same also applies to the method according to the invention, which can optionally be divided into an online measurement method and an analysis method carried out offline. It is also conceivable to carry out a pure online analysis of the measurement data of the at least three derivations. If only the recording of the measurement data, ie the measurement of the derivatives, takes place online, it is possible to carry out the further analysis completely offline according to a further embodiment.

The invention is to be explained in more detail below using exemplary embodiments in conjunction with the drawing. The drawing shows in

1 shows the spread of excitation and regression in the heart using a typical electrocardiogram,

Fig. 2 shows a preferred arrangement of at least four electrodes on the

surface of a patient's torso,

3a, 3b corrected excitation potential signals of two healthy hearts determined from three leads in a three-dimensional representation,

4a, 4b corrected excitation potential signals with altered electrical excitation due to cardiac amyloidosis in a three-dimensional representation,

5a, 5b a projection of space curves of the electrical potential in the main plane of the space curve, each without findings,

6a, 6b a representation of the projection of the space curves into the main plane after a discrete Fourier transformation and inverse transformation using the first three Fourier descriptors without findings,

7a, 7b projections of the space curve of the electric potential in the main plane with cardiac amyloidosis, 8a, 8b the back-transformed projection of the space curve of the electrical potential into the main plane determined using the first three Fourier descriptors,

9 shows a diagram with the 2nd and 3rd Fourier descriptors as the axis and entered feature vectors with findings (x) and without findings (o) and

10 to 14 schematic representation of different exemplary embodiments of the method according to the invention for determining excitation potentials of the human heart.

1 shows the typical course of an ECG complex from atrial stimulation to ventricular stimulation and regression of the stimulation of the human heart. The P wave is an expression of the spread of excitation in the atria starting from the sinus node of the heart. The first part of the P wave represents the electrical excitation of the right atrium, the second part the electrical excitation of the left atrium. The PQ interval comprises the time from the onset of the P wave to the beginning of ventricular excitation. It reflects the time it takes for electrical impulses generated by the sinus node to travel through the atrium, AV node, and bundle of His to the ventricles.

The QRS excitation complex then corresponds to the spread of excitation in the two heart chambers. The ventricles are fully excited at the beginning of the ST segment. The ST segment then transitions into the T wave of ventricular resuscitation. The maximum value of the T-wave represents the maximum repolarization of the regression of ventricular excitation. This is followed by the U-wave and then the propagation of excitation at the sinus node of the heart begins again. According to the invention, at least 4 electrodes are used to determine excitation potentials of the human heart in order to determine at least three leads, preferably bipolar leads. The electrodes are preferably arranged on the surface of the upper body of a person, as shown in FIG. The lines connecting the electrodes of leads Ho, Ve and 1 span a space. This electrode arrangement is also characterized by mutually perpendicular connection lines between the electrodes of leads Ho, Ve and I. In addition, a fifth electrode can also be used, which can serve as a neutral electrode. Finally, more electrodes can also be used, in which case several triple derivations that span a space can be formed from the electrodes and evaluated.

3a and 3b now show the measured spatial curves of the electrical potential of the ventricular excitation and regression of the excitation of the heart in a three-dimensional diagram for two non-pathological cases. In contrast, FIGS. 4a and 4b show the spatial curves of the electrical potential for two different pathological changes in the electrical excitation of the heart which can be attributed to cardiac amyloidosis. In all four three-dimensional diagrams, determined electric potential vectors RDir and TDir are plotted as the main orientation of the corrected excitation potential.

A comparison of the three-dimensional representations of the pathological space curves in FIGS. 4a and 4b with the non-pathological space curves in FIGS diagnostic methods can be used.

A measure of the alignment can be specified on the basis of the alignment of the electrical potential vectors RDir and TDir and also the alignment of the vectors RDir and TDir to one another, so that it is possible to conclude that electrical excitation of the heart has been altered by cardiac amyloidosis. In FIGS. 5a, 5b the projection of the space curve into the main plane is shown in a diagram in which the X and Y axes represent the first and second main components of the space curve. It can be seen that this projection of the space curve of the electrical potential onto the level of the main components clearly shows a unification of the two non-pathological space curves.

If a discrete Fourier transformation is applied to the projections of the space curves determined in this way, a further standardization of the two non-pathological space curves of the electrical potential can take place during the inverse transformation, for example using only the first three Fourier descriptors. The associated two-dimensional projections are shown in Figures 6a and 6b.

A shape analysis of these obtained basic shapes of the projection of the space curves into the main plane, for example with regard to roundness and/or compactness, shows further standardization of the values compared to a shape analysis of the projections in FIGS. 5a and 5b.

The same standardization takes place within the pathological group, as can be seen in the projections of the space curves of the electrical potential of pathological hearts with cardiac amyloidosis shown in FIGS. 7a, 7b and 8a, 8b with the projections shown. The standardization can be clearly seen both in the projection of the space curve into the main plane and in the inverse transform using only the first three Fourier descriptors after a discrete Fourier transformation. It can be clearly seen that the standardization only works within the respective group (not pathological against pathological). This makes it clear that pathological and non-pathological space curves of the electrical potential can be very clearly distinguished from one another. Here, the shape analysis values in terms of compactness or the roundness of the curves shown in Fig. 7a, 7b, 8a, 8b clearly different values than the non-pathological ones of Fig. 5a, 5b and 6a and 6b.

In FIG. 9, the Fourier descriptors FD2 and FD3 of the projections into the main plane are plotted against one another for a plurality of space curves. Pathological excitation potentials are marked with an x and non-pathological ones with an o. It is shown that pathological and non-pathological measurements can be separated with a high degree of certainty using the linear separation function shown. The separation function used is not limited to a linear form; higher-order functions are also possible. This separation function can also be expressed in terms of reference vectors. A comparison with these reference vectors can result in a diagnosis of cardiac amyloidosis, for example. The separation of the pathological and non-pathological space curves of the electrical potential can therefore be carried out with a very high level of quality. The pathological measurements can be separated from non-pathological measurements using a plurality of Fourier descriptors or also other features, such as angles or extensions of the space curves. Depending on the number of features (dimensions), a corresponding hyperplane is then necessary for this. The selected features are combined in feature vectors, which are also used to evaluate pathological or non-pathological changes in the excitation of the heart.

10 shows a schematic view of an embodiment according to the first aspect of the invention for the evaluation of the angle of the main orientation of the electrical excitation potential signal in the form of the electrical potential vector RDir of the QRS excitation complexes. For this purpose, the recorded measured values of the at least three derivatives (b1, b2, b3j) are fed to a signal processing unit in block 1. In the exemplary embodiment, the QRS excitation complexes for the propagation of the ventricular excitation of the heart in the at least three derivatives are detected in block 2. These are detected QRS Excitatory complexes are fed to the optional classification unit in block 3 and divided into different classes. This classification also includes the division of excitation potential signals into the class of extrasystoles. In addition, a normal class is preferably defined, which corresponds to the normal physiological course of the heartbeats, which is preferably evaluated, for example.

In block 4, the processed excitation potential signal is subjected to an offset correction and an offset-corrected angle determination is carried out for all detected QRS excitation complexes and their main alignment in the leads. The angles determined, for example the elevation angle and the azimuth angle in spherical coordinates as a measure of the main alignment of the QRS excitation complexes, are compared in block 5 with reference values or reference vectors, ie reference features. The result of this comparison is forwarded or output to block D, an optional diagnostics block. The diagnosis block D either outputs the determined degree of alignment as a value or already indicates a diagnosis, for example of a structural heart disease, preferably of cardiac amyloidosis.

FIG. 11 shows an embodiment according to the second aspect of the invention, in which the steps of the embodiment according to the first aspect are carried out analogously for the T-wave of the heart's resuscitation. For this purpose, the recorded measured values of the at least three leads (b1, b2, b3j) are fed to a signal processing unit in block 1. In the exemplary embodiment, the T waves of the propagation of the excitation regression of the heart are detected in the at least three leads in block 6. These detected T - Waves are fed to the optional classification unit in block 7 and divided into different classes analogous to block 3.

In block 8, an offset correction of the processed excitation potential signals and an alignment of the electrical potential vector TDir analogous to the ORS Excitation complex in the corrected excitation potential signal determined over all detected T-waves. The orientation of the electric potential vector TDir indicates the main orientation of the corrected capture potential signal of decapitation of the ventricles in the leads. The angles determined as a measure of the main orientation, for example the elevation angle and the azimuth angle of TDir in spherical coordinates, are compared in block 5 with reference values or reference vectors, ie reference features. The result of the comparison is forwarded or output to block D, an optional diagnostics block. The diagnosis block D either outputs the determined degree of alignment as a value or already indicates a diagnosis, for example of a structural heart disease, preferably of cardiac amyloidosis.

A combination of the alignment measures of the electrical potential vectors RDir and TDir of FIGS. 10 and 11 can preferably also be used to provide a combined evaluation measure of an alignment of the corrected excitation potential signals in block 5 for a diagnostic method. In addition, the main alignments RDir and TDir themselves can be combined in that the alignment with one another is evaluated and a comparison with reference values is also carried out in block 5, for example.

FIG. 12 shows a schematic view of an exemplary embodiment according to the third aspect of the invention, which includes a shape analysis of the space curve of the electrical potentials. For this purpose, the recorded measured values of the at least three leads (b1, b2, b3) are sent to a signal processing unit in block 1. In block 2, the QRS excitation complexes for the propagation of the ventricular excitation of the heart in the at least three leads are detected in the exemplary embodiment. These detected QRS excitation complexes are fed to the optional classification unit in block 3 and divided into different classes. This classification also includes marking as an extrasystole. It is also preferably a Normal class set, which corresponds to the normal physiological course of the heartbeats, which is preferably evaluated, for example.

Subsequently, in block 9, the space curves of the measured electrical potential are projected into the main plane of the space curve or into a plane in space, which can be formed by at least two derivatives, for example. In principle, a space curve based on a corrected excitation potential signal can also be used in the projection. The determined projection of the space curve is subjected to a shape analysis in block 10 . The shape analysis values determined as a measure of the shape are then fed to block 5 and compared by it with reference values or reference vectors, ie reference features. The result of the comparison is forwarded to block D, an optional diagnostics block, or output. The diagnosis block D either outputs the determined measure of the shape as a value or already indicates a diagnosis, for example of a structural heart disease, preferably of cardiac amyloidosis.

The shape analysis can be done directly using the projection of the space curve or using an approximated projection obtained by a discrete Fourier transformation of the projection of the space curve into Fourier space and the inverse transformation using, for example, only the first three Fourier descriptors. As already explained, these only have typical basic forms, which can be classified with greater certainty into at least two groups to be discriminated; these are preferably defined as non-pathological or pathological. The shape analysis of the projections of the space curve using Fourier descriptors can also take place in block 10 .

In principle, the shape analysis can be carried out in different ways. For example, as will be shown below, by determining the compactness of the projection of the space curve, which results from the ratio of the square of the circumference of the projection of the space curve with the area enclosed by the projection of the space curve, a differentiation between pathological and non-pathological space curves of the electrical excitation potential signal can be achieved and a cardiac amyloidosis can thus be concluded. This also applies, for example, to the roundness of the projection of the space curve.

Finally, all three exemplary embodiments of FIGS. 10, 11 and 12, which are shown here as alternatives, can be used cumulatively in order to further improve the quality of at least one evaluation measure, which in turn also further improves the optional diagnosis suggestion. This combination is shown in FIG. At the same time, FIG. 13 also shows an embodiment of the device V according to the fourth aspect of the invention, which can execute each individual method according to the first to third aspects of the invention, but also any combination thereof.

By combining the QRS excitation complexes and the T waves, the angles between RDir and TDir can also be determined in block 11 as a measure of the alignment with one another. In block 5, this measure of alignment is compared with reference values or reference vectors, ie reference features. Again, the result of the comparison is forwarded or output to block D, an optional diagnostics block. The diagnosis block D either outputs the determined degree of alignment as a value or already indicates a diagnosis, for example of a structural heart disease, preferably of cardiac amyloidosis.

In block 1, signal processing is carried out as for all exemplary embodiments, for example via smoothing and/or filtering of the measured electrical potentials of the at least three leads. The optional classification of the electrical excitation potential signals occurs in Block 3.

The purpose of block 5 in Fig. 13 is to perform a combination of the results of blocks 4, 8, 10 and 11 described above. Those in blocks 4, 8, 10 and 11 the detected features are output to block 5. Block 5 in FIG. 13 can then combine the features with one another and carry out a clear separation between at least two groups, preferably defined by non-pathological and pathological groups with a high hit probability. Furthermore, a separation function can be defined, which divides the feature vectors into at least two groups, of which at least one can be defined as pathological and at least one as non-pathological. The features can be linked in any combination to derive group information. The combinations include the calculated values or ratios between these calculated values. The group information can be forwarded to the optional block D.

This provides a simple way of screening for structural heart diseases, for example cardiac amyloidosis, in order to detect them as early as possible.

Claims

25

Method for determining excitation potentials of the human heart using at least 4 electrodes, which are arranged on the body surface of a human and from which at least three derivatives, preferably bipolar derivatives, of the electrical potential of the excitation of the human heart are determined and the straight lines connecting the electrodes arranged on the body surface, which form the at least three leads, span a three-dimensional space, characterized in that

- at least the electrical potential of the propagation and/or regression of the ventricular excitation of the heart is measured in each lead,

- the measured values of the leads are fed to a signal processing unit (block 1) in order to obtain a processed excitation potential signal, preferably in each lead,

- the QRS excitation complexes (block 2) for the propagation of the ventricular excitation of the heart in the at least three leads are detected in the processed excitation potential signal,

- the processed excitation potential signal of the leads is corrected by an offset value in order to obtain a corrected excitation potential signal (block 3),

- wherein in the corrected excitation potential signal of the at least three leads the main orientation of the corrected excitation potential signal, preferably in the form of an electric potential vector RDir with maximum length during the propagation of the ventricular excitation of the QRS excitation complex of the heart in the spanned space is determined (block 4) and - a measure for the spatial alignment of the determined corrected excitation potential signal of the ventricular excitation of the QRS excitation potential, preferably the electrical potential vector RDir (block 4) is determined.

2. The method according to claim 1, characterized in that alternatively or cumulatively to the detection of the QRS excitation complex in the processed excitation potential signal, the T-wave of the excitation recovery of the heart chambers is detected in the at least three leads (block 6),

- the processed excitation potential signal is corrected by an offset value in order to obtain a corrected excitation potential signal (block 8),

-wherein in the corrected excitation potential signal of the excitation recovery of the heart of the at least three leads, the main orientation of the corrected excitation potential of the excitation recovery of the heart chambers, preferably in the form of an electrical potential vector TDir with maximum length in the spanned space during the excitation recovery of the heart chambers is determined (block 8) and

- a measure for the alignment of the main alignment of the corrected excitation potential of the excitation recovery of the T-wave, preferably the alignment of the electric potential vector TDir in the space spanned by the at least three leads (block 8) is determined.

3. The method according to claim 1 or 2, characterized in that at least one angle of the electric potential vector RDir or TDir in the spanned space is determined as a measure for the orientation of the electric potential vector RDir and/or TDir.

4. The method according to any one of claims 1 to 3, characterized in that the determined electrical potential vector RDir and / or TDir in the spanned Space is determined in spherical coordinates or cylinder coordinates and the elevation angle 0 and/or the azimuth angle cp are determined in spherical coordinates for the alignment of the electric potential vector RDir and/or TDir. Method according to one of Claims 3 or 4, characterized in that a comparison of at least one angle with a threshold value and/or a comparison with a reference vector of RDir and/or TDir is carried out (block 5). Method according to one of Claims 3 to 5, characterized in that at least one angle between the electric potential vectors RDir and TDir is determined. Method according to one of claims 1 to 6, characterized in that in the processed excitation potential signal in at least one derivation of the time of the beginning of the QRS excitation complex and / or the T-wave is determined and the measured electrical potential value in at least one derivation at the time of the beginning of the QRS excitation complex or the T wave is used as an offset value for the correction of the excitation potential in the respective lead. Method according to one of Claims 1 to 7, characterized in that electrical potentials are measured in each lead for a plurality of propagations and regressions of the ventricular excitation of the heart and the main orientation of the corrected excitation potential signal is determined on the basis of statistical measurement variables, preferably medians or mean values, which 28 of the majority of the measured propagations and regressions of the ventricular excitation of the heart were determined. Method according to one of Claims 1 to 8, characterized in that mean and/or median values for the electric potential vectors RDir and TDir are used to determine the measure for the spatial alignment of the corrected excitation potential signal. Method according to claim 1 or 2, characterized in that alternatively or cumulatively to the methods according to claim 1 and/or 2, a space curve of the electrical potential is determined during the propagation and regression of the ventricular excitation of the heart in the space spanned by the leads, the space curve of the electrical potential is subjected to a shape analysis, with a projection of the space curve of the electrical potential in the main plane of the space curve or in at least one plane in space being carried out for the shape analysis and the projection of the space curve in the respective plane being subjected to a shape analysis, with preferably a comparison with a reference space curve is carried out. Method according to Claim 10, characterized in that the shape analysis of the projection of the space curve is carried out using a simplified space curve generated by means of Fourier descriptors, the projection of the space curve of the electric potential being represented by the complex value c(m] in the selected plane, such that the following applies: c(m) = x(m) + iy(m) , [1] where x(m], y(m] is the mth measured value of the electric potential in the 29 represent the respective projection and the following applies to the Fourier descriptors in the plane for N measured values:

Method according to Claim 10 or 11, characterized in that the roundness of the projection of the space curve is determined as the form analysis in the respective plane, the roundness corresponding to the ratio of the greatest width and greatest height of the projection of the space curve. Method according to one of Claims 10 to 12, characterized in that the compactness of the projection of the space curve in the respective plane results from the ratio of the square of the circumference of the projection of the space curve to the area enclosed by the projection of the space curve. Method according to one of Claims 10 to 13, characterized in that the shape analysis is carried out alternatively or cumulatively using the determined Fourier descriptors by forming a feature vector with the determined Fourier descriptors as components and comparing it with at least one reference feature vector. Method according to one of the preceding claims, characterized in that the signal processing comprises at least smoothing and/or filtering of the measured electrical potentials of the at least three leads. 30 Method according to one of the preceding claims, characterized in that the signal processing (block 1) includes a classification (block 3, block 7) of the measured electrical potentials of the at least three derivatives, so that ventricular excitation of the heart which deviates from the normal excitation complex, in particular ventricular extrasystoles, not be considered for further analysis of the arousal complex. Device for carrying out a method according to one of Claims 1 to 16, with

-means for correcting the processed excitation potential signal of the leads in order to obtain an excitation potential signal corrected with an offset value,

- means for determining the main orientation of the corrected excitation potential signals, preferably by determining an electric potential vector RDir of the QRS excitation complex of the heart of maximum length and/or an electric potential vector TDir of the T-wave of the heart's excitation recovery of maximum length and/or

- Means for determining the space curve of the electrical potential, preferably from the corrected excitation potential signal in the space spanned by the electrodes,

- Means for determining a measure of the spatial orientation of the 31 corrected excitation potential signals, preferably a measure of the spatial orientation of the electrical potential vector RDir and/or the electrical potential vector TDir and/or

- Means for analyzing the shape of the projection of the space curve of the electrical potential in a main plane or in at least one plane in space, the means for analyzing the shape preferably carrying out a comparison with a reference space curve and means for outputting the determined extent of the alignment of the electrical potential vectors and/or the shape analysis results.