RU2772201C2 - Iterative coherent mapping of electrophysiological activation of heart, including scar effects - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: group of inventions relates to medicine, namely to a method and a system for iterative coherent mapping of electrophysiological activation of the heart. In this case, an input grid representation of the cardiac chamber, a set of measured locations on the wall tissue of the cardiac chamber and a corresponding set of local activation time (hereinafter – LAT) values measured at these locations are received. The input grid is rearranged into a regular grid containing polygons of a regular shape. Data of the set of measured locations and corresponding LAT is coordinated with polygons of a regular shape. Iterative calculation is carried out for polygons of a regular shape of (i) corresponding LAT values and (ii) corresponding probabilities of scar tissue content in the wall tissue so that to obtain an activation electrophysiological (hereinafter – EF) wave on the regular grid, indicating scar tissue. A representation of an electroanatomic map containing the activation EF wave and scar tissue superimposed on the regular grid is provided. Binary marking of measured locations is carried out to indicate the presence of a scar and/or the presence of a double EF potential.

EFFECT: simplified and reliable mapping and interpretation of complex arrhythmias is achieved with an increase in the efficiency of subsequent invasive heart treatment, such as catheter ablation, by taking into account repeated wave inputs, which allows eliminating the resulting re-entry artifacts on the EF map.

10 cl, 6 dwg

Description

Technical field

The present invention relates essentially to electrophysiological mapping, and specifically to methods and systems for electrophysiological mapping of the heart.

State of the art

Electrophysiological mapping of the heart is often used to identify potential sources of cardiac arrhythmia in cardiac tissue. For example, U.S. Patent Application Publication 2017/0281031 describes electroanatomical mapping, which is performed by inserting a multi-electrode probe into the heart of a living subject, recording electrograms simultaneously from the electrodes at appropriate locations in the heart, delimiting the corresponding activation time intervals in the electrograms, generating a map of the propagation of electrical waves on based on activation time intervals, maximizing wave coherence by adjusting the local activation time of the electrograms and reporting the corrected local activation time.

As another example, U.S. Patent Application Publication 2016/0106376 describes the calculation of the local conduction velocity of the cardiac activation wavefront by collecting a plurality of electrophysiological (EP) data points using a multi-electrode catheter, with each EP data point including both position, and local activation time (LAT) data. For any EF data point, nearby EF data points can be defined, including the selected EF data point and at least two additional EF data points. Subsequently, position and LAT planes can be determined using the positions and LAT of the ESP data points respectively within the nearby area. The conduction velocity can be calculated from the intersection of the position and LAT planes. The resulting set of conduction velocities can be displayed as a graphical representation (eg, an electrophysiological map), for example, by displaying vector icons arranged in a uniform grid over a 3D model of the heart.

US Pat. No. 6,301,496 describes a method for diagnosing an abnormal condition in a biological structure such as a heart, comprising the steps of measuring a physiological response at at least three measured points on the surface of the biological structure, calculating a vector function related to the response, displaying a representation of the vector function, and inferring about the anomalous condition based on the representation. The present invention is particularly used for diagnosing cardiac arrhythmias, in which case the physiological response is the voltage from which the local activation time is inferred, and the vector function is the gradient of the local activation time, in particular the conduction velocity. It is expected that the magnitude of the conduction velocity in the scar tissue will be abnormally low.

The essence of the invention

In an embodiment of the present invention, a method is provided, including obtaining an input grid representation of a cardiac chamber, a set of measured locations on the tissue of the wall of the cardiac chamber, and a corresponding set of local activation time (LAT) values measured at those locations. The input mesh was modified into a regular mesh including regular polygons. The set of measured locations and corresponding LATs is data matched to regular polygons. The corresponding LAT values, and the corresponding probabilities of scar tissue incorporation into the wall tissue, are iteratively calculated for regular polygons so as to obtain an electrophysiological (EP) activation wave propagating along a regular grid that is indicative of scar tissue. An electroanatomical map is presented superimposed on a regular grid, and the map includes an EF-activation wave and scar tissue.

In some embodiments, regular polygons include regular triangles.

In some embodiments, the iterative calculation of LAT values and probabilities includes iteratively solving a set of equations for LAT values, slowness values, and ESP wave propagation.

In an embodiment, the iterative calculation of the probabilities includes reassigning each regular polygon a vector of slowness, recalculated using a scar multiplier.

In another embodiment, the method further includes binary marking at least some of the measured locations to indicate at least one of the presence of a scar and the presence of a dual ESP potential.

In some embodiments, the presentation of the electroanatomical map includes overlaying the electroanatomical map with conduction arrows indicating scar tissue.

According to an embodiment of the present invention, a system including an interface and a processor is further provided. The interface is configured to receive an input grid view of the cardiac chamber, a set of measured locations on the cardiac chamber wall tissue, and a corresponding set of local activation time (LAT) values measured at those locations. The processor is configured to rebuild the input mesh into a regular mesh including regular polygons, fit data representing a set of measured locations and corresponding LATs to regular polygons, and iteratively calculate (i) corresponding LAT values, and (ii) corresponding probabilities of incorporating scar tissue into wall tissue so as to obtain an electrophysiological (EP) activation wave propagating along the correct grid that indicates scar tissue. The processor is additionally configured to present an electroanatomical map that includes an EF-activation wave and scar tissue superimposed on a regular grid.

In another embodiment, the present invention provides a method including obtaining an input grid representation of a cardiac chamber, a set of measured locations on the tissue of the wall of the heart chamber, and a corresponding set of local activation time (LAT) values measured at the locations.

The input mesh was modified into a regular mesh including regular polygons. The set of measured locations and corresponding LATs is data matched to regular polygons. The corresponding LAT values are iteratively calculated for the regular polygons so that a round-robin solution for the EF-activation wave over a regular grid that takes into account the re-entry of the EF-wave. An electroanatomical map is presented, which includes a cyclic EF–activation wave superimposed on a regular grid.

In some embodiments, the iterative calculation of the LAT values includes iteratively solving a set of complex-valued equations, configured to describe the reentry of the ESP wave by including a value of the reentry cycle length.

In some embodiments, the iterative calculation of the LAT values includes iteratively solving a set of three linear equations for the LAT values, the slowness values, and the ESP wave propagation.

In an embodiment, iteratively calculating the LAT values includes reassigning each regular polygon a vector of slowness, recomputed using a proximity factor.

According to an embodiment of the present invention, a system including an interface and a processor is further provided. The interface is configured to receive an input grid view of the cardiac chamber, a set of measured locations on the cardiac chamber wall tissue, and a corresponding set of local activation time (LAT) values measured at those locations. The processor is configured to rebuild the input mesh into a regular mesh including regular polygons, fit data representing a set of measured locations and corresponding LATs to regular polygons, and iteratively calculate the corresponding LAT values for regular polygons such that to obtain a cyclic solution of the EF-activation wave on the correct grid, which takes into account the re-entry of the EF-wave. The processor is additionally configured to present an electroanatomical map that includes a cyclic EF-activation wave superimposed on a regular grid.

The present invention will become better understood from the following detailed description of the embodiments presented with reference to the accompanying drawings.

Brief description of the drawings

On FIG. 1 is a schematic representation of a three-dimensional (3D) cardiac navigation and electrophysiological (EP) signal analysis system in accordance with an embodiment of the present invention;

in FIG. 2 is a flowchart that schematically illustrates a method and algorithm for calculating a coherent ESP activation wave in accordance with an embodiment of the present invention;

in FIG. 3 is a flowchart that schematically illustrates the method and algorithm for calculating scar presence probabilities within the method described in FIG. 2 according to an embodiment of the present invention;

in FIG. 4 is a block diagram that schematically illustrates the method and algorithm for creating an EF activation wave map according to the method described in FIG. 2 according to an embodiment of the present invention;

in FIG. 5 is a schematic pictorial volumetric rendering of a left atrial coherent EF activation wave map that takes into account EF wave reentry and scar tissue, in accordance with an embodiment of the present invention; and

in FIG. 6 is a schematic pictorial volumetric rendering of a left atrial coherent EF activation wave map with conduction arrows superimposed, which illustrates the propagation of an EF activation wave, in accordance with an embodiment of the present invention.

Detailed description of embodiments of the invention

general description

Cardiac arrhythmia is a class of clinical conditions in which the heart beat is irregular. Within this class of conditions, supraventricular tachycardias (SVT) are an important group, which include atrial tachycardia, atrial fibrillation, atrial flutter, and paroxysmal supraventricular tachycardia. In SVT, as the names above suggest, the arrhythmia originates from the atrium of the heart.

Part of the SVT episodes is occupied by electrophysiological (EP) disturbances caused by scar tissue. For example, patients with ablation scars often experience episodes of atypical atrial tachycardia. In areas of scar tissue, EF activation may be completely blocked or partially blocked (areas with slow conduction). These areas are referred to herein as "block line" or scar areas.

In the embodiments of the present invention described herein, iterative coherent electroanatomical (ICEA) mapping methods and an arrhythmia cause detection system that belongs to the SVT group are provided. In the context of the present description, the term "coherent" means obtaining the most consistent description of the propagation of the EF-activation wave, in which the conduction velocity is cyclic and constant.

Thus, mathematically, no abrupt changes in the direction of the activation wave or abrupt changes in the velocity of the activation wave are allowed, except in non-conductive areas such as scars where activation is blocked by certain anatomical barriers or scar tissue. Thus, in addition to correctly capturing the continuous and cyclic (i.e., re-entry of the activation wave, as described below) signs of cardiac EF activation, the coherent mapping method described provides for the generation of an EF map that displays scar areas of any shape (e.g., line).

The described coherent mapping method detects, as noted above, the occurrence of reentrant activation waves, such as reciprocal tachycardia (RT), which is a physiological condition in which an EF-activation wave travels in a circle within the heart (i.e., a cyclic EF- wave of activation), rather than passing from one end of the heart to the other, and subsequently ends at a physiological barrier (non-conductive anatomical structure). As the EF activation wave continues to propagate with a certain cycle length (for example, the time between successive peaks on an ECG), the wave propagation in the heart chamber can cover the entire re-entry cycle, that is, the “late” wavefront in the cycle meets the “early” wavefront. front of the next cycle. As a result, regions may be assigned incorrect Local Activation Time (LAT) values that differ by one cycle length when these wavefronts should have similar time values.

By accounting for reentry waveforms, embodiments of the described ICEA method eliminate the resulting reentry artifacts on the EF map, and therefore ICEA is suitable for dealing with most types of arrhythmia, while in some embodiments, the described method is optimized for the case of atrial flutter (AFL) and others. cases of macroreciprocal states. The resulting ESP map of the cardiac chamber, which typically shows aberrant ESP activity, e.g. due to the presence of AFL, represents such aberrant ESP activity without re-entry rainbow artifacts (described below) that would otherwise would arise, i.e. when applying a model that does not take into account repeated entries of the EF wave. Thus, such “rainbow” artifacts are eliminated when using the described ICEA mapping method.

In some embodiments, the described model accepts an input grid representation of the cardiac chamber, a set of measured locations on the cardiac chamber wall tissue, and a corresponding set of LAT values measured at those locations. Collectively, the locations and corresponding LAT values are also referred to herein as "data points". The processor converts the input mesh to a regular mesh containing regular polygons such as triangles, also referred to herein as "faces". In some embodiments, the processor performs a fitting of a set of measured data points to regular polygon shapes.

After obtaining the initial iterative coherent solution, information is used that indicates interruptions in the rest of the ICEA-continuous map. For example, the processor uses user-provided data point labels as an indication of normal, scarred, or dual-potential areas of tissue. In the context of the present description, the double ESP potential is defined as a specific double peak that can be identified on fractionated and multicomponent electrograms recorded on the wall of cardiac chamber tissue, such as the atrium.

In some embodiments of the present invention, the processor further assigns each face (i.e., triangle) a probability that the region is electrically non-conductive, i. scar tissue. Probability is assigned (automatically or manually by the user) based on the measured points on the map, which are normal, scar or dual potential measurements. The assignment may also include taking into account LAT patterns and slowness values, where slowness is a vector whose magnitude is the reciprocal of the magnitude of the ESP velocity vector along a geodesic line connecting two adjacent faces and whose direction is the same as the velocity vector, as described below.

Once a face is assigned a high probability of being non-conductive, a relaxed slowness continuity constraint as well as a relaxed bonding velocity constraint is applied to create a map that takes into account the possible presence of scar tissue. Areas of low conductance (including blocking lines) are further identified by activation waves with opposite propagation vectors (i.e., opposite conductance arrows on the ICEA map pointing to the blocking line in between), where the propagation vector passes around the scar, indicating non-conductive tissue. In an embodiment, locations of low ESP signal amplitude are designated as the most likely parts of the conduction-impeding barrier.

The processor evaluates the ESP activation waveform on the correct grid, including indications of scar areas, by iteratively calculating LAT values and probabilities (ie, weights) of whether each of the faces is conductive or part of a scar area. Iteratively computed weights produce data that increase or decrease the likelihood that a particular face is a scar, and provide the calculated probabilities of a scar for each face as output. Subsequently, the scar weights (ie, scarring probabilities) are plugged back into the linear equations as factors that reduce the power of the continuity equations in the line of blocking areas. Slowness limits are also reduced by similar multipliers. This allows a more free change of solution (for example, abruptly) in the line of blocking regions. The processor performs an iterative calculation based on the assumption that the EF-activation wave is continuous until it hits the scar areas.

Subsequently, the processor generates a coherent activation map that describes an EF activation wave that is typically at least partially aberrant, for example due to the presence of scars. In an embodiment, the processor renders the final ICEA map using a cyclic color scale so as to de-dependency on the early-meets-late effect, which causes the rainbow artifact, as described below. In another embodiment, the coherent EF activation map contains conduction arrows that illustrate the normal or aberrant propagation of the EF activation wave.

Typically, the processor is programmed in software that contains a specific algorithm that allows the processor to perform each of the steps and functions associated with the processor described above.

The described method of iterative coherent EF mapping creates a mapping system in which clinicians can easily and reliably map and interpret complex arrhythmias. Thus, the disclosed systems and methods provide a tool that can enhance the success of subsequent invasive cardiac treatments, such as catheter ablation.

System Description

On FIG. 1 is a schematic representation of a three-dimensional (3D) navigation and electrophysiological (EP) signal analysis system 20 in accordance with an embodiment of the present invention. System 20 can be configured to analyze essentially any physiological parameter or combinations of such parameters. In the description presented herein, by way of example, it is assumed that the analyzed signals are intracardiac and/or extracardiac (body surface) electrocardiographic (ECG) potentials versus time. In order to fully characterize such dependencies, signals at different locations must be correlated with respect to each other in time, as is done when generating a LAT map. Timing is performed by measuring relative to a reference time (eg, any point in time), such as the start of each QRS complex of the ECG reference signal (ie, the start of each heartbeat). In an embodiment, the reference signal comes from a catheter placed in the coronary sinus. For SVT, the reference signal reflects atrial activity, which can be twice or three times the rate of ventricular activity. A method for generating a LAT map is described in US Pat. No. 9,050,011, cited above.

For simplicity and clarity, the following description, except where otherwise indicated, assumes an examination procedure in which the system 20 measures the actual electrical activity of the heart 34 using a probe 24. It is assumed that the distal end 32 of the probe has electrodes 22. The measured signals are used, among other things to create a LAT-map of at least part of the tissue of the wall of the heart 34 of the patient 26.

Typically, probe 24 includes a catheter that is inserted into the body of patient 26 during a mapping procedure performed by clinician 28 using system 20. During the procedure, it is assumed that a ground electrode 23 is connected to patient 26. It is also assumed that electrodes 29 are connected to the skin of the patient 26 in the region of the heart 34.

In an embodiment, probe 24 captures local intracardiac electrocardiograms (ECGs) as it travels across a portion of the chamber of the heart. Some of the features on the measured intracardiac ECG curves are annotated at the moment when an aberrant EF activation wave passes under the catheter electrode. In these cases, the location of the probe 24 is also recorded.

The system 20 may be controlled by a system processor 40 including a data processing unit 42 communicating with a memory 44. In some embodiments, memory 44 included in the system processor 40 stores a map 62 of LAT and/or stresses of at least a portion of the heart wall tissue. 34 patients 26. The processor 40 is typically mounted in a console 46 which contains controls 38, typically including a pointing device 39, such as a mouse or trackball, which the clinician 28 uses to interact with the processor.

The processor 40 (in particular, the data processing unit 42) uses software comprising a sensor tracking module 30, an ECG module 36, and an ESP activation analysis module 35 to control the system 20 and/or for the ESP activation analysis module 35 to perform at least a portion of described analysis (using, for example, LAT or corrected LAT-maps 62 stored in memory 44) so as to simulate arrhythmia.

The ECG module 36 is associated with enabling the actual electrical signals from the electrodes 22 and the electrodes 29 to be received. The module is configured to analyze the actual signals and can present the analysis results in the correct ECG format (typically a time-advance graphical representation) on the display 48.

The transducer tracking module 30 typically tracks the location of the distal end 32 of the probe 24 within the patient's heart 26. The tracking module may use any probe location tracking technique known in the art. For example, module 30 may use a magnetic field based location tracking subsystem. (For simplicity, the components of such a subsystem are not shown in Fig. 1).

In an alternative or additional embodiment, tracking module 30 may track probe 24 by measuring impedances between electrode 23, electrodes 29, and electrodes 22, as well as impedances with other electrodes that may be located on the probe. (In this case, electrodes 22 and/or electrodes 29 may generate both ECG signals and location tracking signals). The Carto3 ^® system, manufactured by Biosense Webster (Irvine, CA, USA), uses both magnetic field-based location tracking and impedance measurements to track location.

Using the tracking module 30, the processor 40 can measure the locations of the distal end 32. In addition, using both the tracking module 30 and the ECG module 36, the processor is able to measure the locations of the distal end as well as the LAT of the actual electrical signals detected at those specific locations. As noted above, electrical tracking signals from individual electrode 22 may be combined with magnetic tracking signals to record the location of each electrode. Such a hybrid (i.e., magnetic/electrical) tracking system and tracking method called Advanced Current Location (ACL) has been implemented in various medical devices, such as the CARTO™ system manufactured by Biosense Webster Inc., and is described in detail in a US patent. No. 8,456,182, the description of which is incorporated herein by reference.

The results of the operations performed by the processor 40 are presented to the clinician 28 on a display 48, which typically presents the clinician with a graphical user interface, a visual representation of the ECG signals recorded by the electrodes 22, and/or an image or map of the heart 34 during the examination.

The software used by the processor 40 may be downloaded to the processor 40 electronically, such as transmitted over a network, or in an alternative or additional embodiment, it may be provided and/or stored on a non-volatile tangible medium such as magnetic, optical, or electronic memory. In particular, the processor 40 uses a special algorithm that allows the processor 40 to perform the steps described below.

Iterative coherent ef-activation mapping including re-entry and scar effects

On FIG. 2 is a flowchart that schematically illustrates a method and algorithm for calculating a coherent ESP activation wave in accordance with an embodiment of the present invention. The algorithm according to the present embodiment is executed by the processor 40.

The following is a general description of steps 80-84 of the ICEA process followed by a detailed description of these steps using the flowchart shown in FIG. 2. In some embodiments, the described method includes receiving an input three-dimensional (3D) mesh (ie, shape) that can model, for example, the atrium of the heart. The input 3D mesh contains polygons such as triangles (i.e. triangular mesh). In some embodiments of the present invention, the processor rebuilds the input mesh into a regular mesh containing regular triangles (also referred to as "faces"). A proper mesh allows you to define and iteratively solve a series of linear equations that preserve the integrity of the ESP wave, as well as reduce the computational load, as described below.

The described methods further include receiving a set of data points containing measured locations on atrial wall tissue and a corresponding set of local activation time (LAT) values measured at those locations. In some embodiments, a set of data points are measured using an electroanatomical mapping catheter system such as CARTO™ using a mapping catheter such as PENTARAY® or LASSO® (system and catheters manufactured by Biosense Webster, Irvine, California, USA).

In some embodiments, the described methods use an anatomical map obtained using a fast anatomical mapping technique (FAM) as input. In the described method, more informative electroanatomical maps can be used as the input anatomical map, since the processor 40 discards any information on the map, except for the anatomical structure itself (for example, from the input electroanatomical map of the cardiac chamber, only the original grid is used to determine the geometry of the chamber).

In some embodiments of the present invention, the processor discards some of the received data points (i.e., measured locations and corresponding LATs) and/or performs data fitting on some of the data points such that the data points match a described correct grid.

In an embodiment, processor 40 discards data points determined to have been acquired when the mapping electrode was not in physical contact with tissue using a technique called Tissue Proximity Index (TPI).

In another embodiment, the processor adjusts the measured positions, which otherwise do not correspond to a realistic description of the anatomical structures provided by the correct grid. Thus, the described method of coherent mapping eliminates errors in catheter-measured locations, such as errors arising from catheter pressure on the chamber wall and shape distortion (pressure is applied to ensure good electrical contact between the EF-signal-receiving electrode and tissue), or from – for respiratory movement, which is not related to a change in the position of the catheter on the chamber wall.

The processor then uses the input, i.e. measured positions and the corresponding LAT, and the input mesh is rebuilt into triangular faces. In an embodiment, the described model assumes that the propagation of an EF-activation wave between two adjacent faces is carried out in the direction of its slowness vectors along a geodesic line in shape that connects the centers of the faces. Thus, the difference in time between the centers of the faces is the geodesic distance between them multiplied by the slowness vector.

Initially, the model additionally assumes that the EF activation wave is continuous, which means that adjacent faces must have similar slowness vectors. Moreover, the model also initially assumes that the conduction velocity of EF-activation waves is in the range between the minimum and maximum possible speed (i.e., within the speed range characteristic of normal cardiac muscle).

The described model formulates the above assumptions into a set of linear equations in boundary conditions. The processor first solves the equations non-iteratively to obtain an initial guess of a coherent EF activation wave superimposed on the shape. The ICEA model can be recalculated based on accumulating auxiliary data, such as local velocity vectors, to ensure that the ESP activation map continues to change as data is collected. In an embodiment, the calculation described by ICEA is initiated by the user, usually after receiving all the ESP data. If it is necessary to add new EF data, the user restarts the calculation from the very beginning, including the initial non-iterative solution.

Using the initial solution, the processor then iteratively solves a set of linear least-squares equations to find the "most consistent" ESP activation based on an additional assumption: the measured LAT values for each face affect the LAT values of their adjacent faces depending on the distance from the measurement point to center of the adjacent face. Iterative calculations make it possible to obtain the most consistent solution, consisting of the resulting EF-activation wave superimposed on the form. The number of iterations, which is usually several, is a predetermined parameter that is based on visual inspection of multiple EA maps during the development phase to confirm that there is no obvious change in the solution as the number of iterations increases above a predetermined value. The user initiates the calculation and usually receives a complete ICEA card within a few seconds.

, при трепетании предсердий, показанная во вставке 280, и (iv) местоположения с метками, такие как рубец и двойной потенциал. В некоторых вариантах осуществления стадии (iii) или (iv) не требуются, например, в случае, если следующий расчет не учитывает повторный вход или не включает в себя связанные с рубцом расчеты.The method begins at loading step 80, in which the processor 40 loads the input data for said method, including (i) a three-dimensional grid (triangular) representing the wall of the heart chamber, (ii) a set of measured data points (e.g., position and LAT), (iii) cycle lengths annotated on ECG traces such as cycle length,

, in atrial flutter, shown in box 280, and (iv) labeled locations such as scar and double potential. In some embodiments, steps (iii) or (iv) are not required, for example, if the next calculation does not account for reentry or does not include scar-related calculations.

Next, in equation building step 81, processor 40 assigns two variables (LAT value and slowness vector) to each face (ie, triangle) in the mesh. To create a set of equations for two variables, the method makes three assumptions that affect the two variables:

1. The EF activation wave is continuous, which means that adjacent faces must have similar slowness vectors.

2. The measured value of LAT in a given face influences the calculated neighboring LAT faces depending on the distance from the measurement site to the centroid of the neighboring face.

3. The activation wave corresponds to the described propagation equations, that is, the wave propagates between the faces in the direction of its slowness vectors, and the time difference between the centers of the faces is the geodesic distance between them multiplied by the slowness vector.

From the above assumptions, we obtain a set of linear equations:

Lv. one

, также называемый в настоящем документе «вес близости», отражает относительное влияние, которое измеренное значение LAT i оказывает на значение LAT, рассчитанное для грани j, а φ представляет собой периодическую сложную функцию, отражающую циклическую ЭФ–волну активации (для учета повторных входов), имеющую найденное значение LAT в качестве переменной.and in ur. 1 (I) weight

, also called "proximity weight" in this document, reflects the relative influence that the measured value of LAT i has on the value of LAT calculated for edge j , and φ is a periodic complex function reflecting the cyclic EF activation wave (to account for repeated entries) , which has the found LAT value as a variable.

, которое экспоненциально ослабляется расстоянием между любыми из используемых местоположений измерения и местоположением грани, в результате чего влияние уменьшается с увеличением расстояния от правильной формы. Так как

присваивается нулевое значение за пределами данного расстояния, которое, как правило, составляет порядка нескольких миллиметров, некоторые из этих уравнений представляют собой нулевые выражения. Таким образом, каждое ЭФ–измерение имеет относительный эффект, пропорциональный расстоянию от правильной формы, и каждое измерение влияет на соседние грани. Измеренные точки, наиболее близкие к правильной форме, с большей вероятностью будут точными.In ur. 1 ur. (I) relates the LAT measurement on face i to the LAT value of face j using

, which is exponentially attenuated by the distance between any of the measurement locations used and the location of the face, whereby the influence decreases with increasing distance from the correct shape. As

is assigned a value of zero beyond this distance, which is typically on the order of a few millimeters, some of these equations are null expressions. Thus, each EF measurement has a relative effect proportional to the distance from the regular shape, and each measurement affects adjacent faces. The measured points closest to the correct shape are more likely to be accurate.

и

, которые приведены в комплексных числах для учета повторного входа, как описано ниже, с помощью матрицы вращения

.Lv. (II) connects vectors of slowness

and

, which are given in complex numbers to account for reentry, as described below, using the rotation matrix

.

Lv. (III) is a propagation equation that indicates that a wave traveling between two adjacent faces must have slowness and LAT that are related by the distances between the connected faces. Because the equation will be written for any two adjacent faces, it reflects the relationship between all LATs and the slowness vectors associated with all the faces in the shape.

Lv. 1 (I)–(III) applies dependencies:

Lv. 2

и что предположения

и

верны.

в ур. 2 представляет собой длину цикла повторного входа.Dependencies Lv. 2 assume all timelines are short

and what assumptions

and

are true.

in ur. 2 is the length of the re-entry cycle.

The above form of describing the cyclic wave as complex numbers versus time takes into account the occurrence of reentry cases described above, in which the activation wave propagates continuously with a certain cycle length.

. Векторы медленности

находятся в

.

представляет собой матрицу вращения между двумя соседними гранями I и J.Lv. 1(II) is a continuous propagation (similarity) equation in which two adjacent faces must have similar slowness vectors

. Vectors of slowness

are situated in

.

is a rotation matrix between two adjacent faces I and J .

.Lv. 1 (III) is the equation for the propagation of a cyclic EF–activation wave between the faces. The distance that the wavefront has traveled is the projection of the geodesic distance, which is the distance between the two centroids of the faces,

.

In the original equation, solved in step 82, processor 40 solves Eq. 1 non-iteratively to obtain the initial assumption about the EF-wave of activation. In the embodiment, ur. 1 is solved by the principle of least squares. The solution is performed using linear algebra mathematical numerical methods containing the Cholesky decomposition followed by forward and backward substitution, as described in the book "Numerical Linear Algebra" by Trefethen, L. N. and Bau, D. (1997), pages 172–178, published by the publishing house SIAM.

In an embodiment, the terms (i) (ii) (iii) eq. 1 are included in one sparse matrix equation. Each sub-equation (ie term) can be assigned a different optimization weight to equalize the different influencing factors. These optimization weights can vary across iterations. These optimization weights are also predefined parameters such as the number of iterations.

Initially, it was assumed that the physical value of the slowness vector describes the change in the complex phase of time with distance, in contrast to the correct slowness vector, which describes the change in time with distance. Because slowness is complex, the physical definition is different and somewhat artificial.

Solution Lv. 1 in the complex domain describes cyclic activation in a natural way and removes the complexities associated with the "early meets late" effect. To obtain a physically meaningful expression for the slowness, additional stages are required, at which an iterative set of equations is built and solved.

преобразуют обратно в

. Для этого используют ур. 1 на основе реальных переменных

,

,

.It should be noted that solving complex functions provides a simple calculation of cyclic activation, but requires converting the complex slowness back to real values in order to preserve the physical value of the slowness vector. Thus, in the end

converted back to

. To do this, use ur. 1 based on real variables

,

,

.

,

:At stage 83 of constructing iterative equations for the iterative solution of Eq. 1 using the initial guess obtained in step 82, the described method linearizing Eq. 1 with application for

,

:

Lv. 3

To receive a linearized eq. one:

Lv. 4.(I)

Lv. 4.(II)

Lv. 4.(III)

,

. Константы

,

представляют собой решения, полученные при предыдущей итерации. При первой итерации для

,

задают начальные значения путем решений для комплексного LAT и реальной медленности начального предположения, полученного на стадии 82.In an embodiment, the real linearized eq. 1, i.e. ur. 4, solve for linear variables

,

. Constants

,

are the solutions obtained in the previous iteration. At the first iteration for

,

initial values are given by solutions for the complex LAT and the actual slowness of the initial guess obtained in step 82.

The conduction velocities for each facet are calculated based on the measured activation time, the calculated propagation vector, and the known distance between the centers of each facet. Under several conditions, the linear solution can form regions with slowness vectors close to zero (also called "sources" or "sinks"). Such areas are not realistic as they suggest locations with infinite velocity. To maintain EF conduction rates near some positive physiological constant a, a constraint is added:

Lv. 5

:Lv. 5 "punishes" deviations of high speeds (small slowness) from a. After the Taylor expansion, one obtains the following set of constraints for

:

Lv. 6

в связи с наличием в переменной выражений с квадратными корнями, таких как

,

. При записи уравнений в дельта–форме можно выполнять разложение Тейлора для

и записывать ограничения в линейной форме.Restrictions through ur. 6 are added only after finding a linear initial solution, since it is impossible to write the constraints in a linear form in

due to the presence of square root expressions in the variable, such as

,

. When writing equations in delta form, one can perform the Taylor expansion for

and write the constraints in linear form.

,

,

сначала посредством проецирования медленности на поверхность реконструкции, а затем записи уравнений в форме матрицы с применением способов, описанных на стадии 82 (т.е. изменяя линейный порядок на таковой, при котором ур. 4 легче решить).Then, in the iterative solution of step 84, the processor 40 solves Eq. 4 with variables

,

,

first by projecting the slowness onto the reconstruction surface, and then writing the equations in matrix form using the methods described in step 82 (ie, changing the linear order to such that Eq. 4 is easier to solve).

At stages 81–84 lvl. 4 is solved with the assumption of continuous propagation of the activation wave. Incorporating scar information into a solution requires a scaffold that detects scars and allows the development of ESP-activation wave interruptions (i.e., solutions) in blocking lines and non-conductive areas.

The coherent mapping method described assigns lower weights to constraints that make the solution incoherent (ie, inconsistent or discontinuous). Thus, the processor removes measured LAT values that the processor determines are invalid (ie, LAT outliers) during iterations. The method assigns new weights after receiving the iteration decision vector. For LAT constraints, this process filters inconsistent points. With respect to neighbor point constraints, this results in the removal of inconsistent edges.

на стадии 85 введения веса рубца, и итерационно рассчитывают на стадиях 87–100. Веса находятся в диапазоне

, где 0,05 указывает на рубец.The following is a general description of steps 85-90 of the ICEA process followed by a detailed description of these steps using the flowchart shown in FIG. 2. The specified frame is also included in the iterative method, and at this moment the weights of the scar (face) are introduced

at step 85 the introduction of the rumen weight, and iteratively calculated at steps 87-100. Weights are in the range

, where 0.05 indicates a scar.

включают обратно в стадию 87 в виде множителей, которые уменьшают мощность уравнений непрерывности в линии областей блокировки. Ограничения в отношении норм медленности, например установленные ур. 6, снижают также посредством множителей. Это позволяет более свободно изменять решение в линии областей блокировки.Iteratively computed weights produce data that increase or decrease the likelihood that a particular face is a scar, and provide the calculated probabilities of a scar for each face as output. Subsequently, scar weights (i.e., the likelihood of having scars)

are included back in step 87 as factors that reduce the power of the continuity equations in the line of blocking regions. Restrictions on the norms of slowness, for example, established by ur. 6 are also reduced by multipliers. This allows you to more freely change the solution in the line of blocking areas.

. для двух граней, смежных с краем

First, the propagation equation eq. 4 (II) are multiplied by at least two corresponding weighting factors

. for two faces adjacent to the edge

Lv. 7

).This reflects the assumption that the slow conduction between the faces is already indicated by a scar (i.e., of little importance).

).

Thus, in areas where a scar is found, the weights of the equations 40(ii) of continuous propagation are reduced. This reduces the requirement for similarity in slowness between adjacent faces, which allows for interruption in wave propagation in scarred areas.

Secondly, the normalized equations of slowness and speed (Equation 6) are multiplied by the calculated weights. Thus, reducing the requirement for a certain target speed value for the edges of the scar:

Lv. eight

After returning the scarring probabilities back to the equations in step 87 of introducing the scar weight, the processor 40 re-solves the modified eq. 4 and proposes a new solution at step 89 for deriving LAT and slowness values. Next, in step 100, new refined scarring probabilities are calculated up to a predetermined number of iterations that reliably result in a sufficiently accurate map, such as the coherent ESP activation map described in FIG. 6.

On FIG. 3 is a flowchart that schematically illustrates the method and algorithm for calculating scar presence probabilities within the method described in FIG. 2 in accordance with an embodiment of the present invention. On FIG. 3 describes the various modules (ie, calculation sub-steps) included in step 100 that allow the development of ESP-decision activation wave interrupts in blocking lines and non-conductive regions. The algorithm according to the present embodiment is executed by the processor 40.

, где один указывает на нормальную грань, а ноль указывает на рубец или «слабо проводящую» грань (в текущей версии нет различия между ними).Stage 100 includes calculation module (ie stages) 91-98 which outputs data for each face that increases or decreases the likelihood that a certain triangular mesh face is a scar. The output of each of steps 91-98 is a weight in the range of zero to one. Weights are in the range

, where one indicates a normal face and zero indicates a scar or "weakly conductive" face (there is no distinction between the two in the current version).

путем сведения к минимуму функции стоимости. На стадии 99 процессор 40 рассчитывает веса рубца

посредством введения информации, собранной на стадиях 91–98, в комбинированную вероятность для каждой грани.The weights calculated in steps 92-96 are arranged in the form of a cost function matrix as described below, and the rumen weights are calculated in step 99

by minimizing the cost function. In step 99, processor 40 calculates the rumen weights

by incorporating the information gathered in steps 91-98 into the combined probability for each facet.

, который указывает на наклон дисперсии LAT по отношению к близлежащим измерениям. Крутая дисперсия может указывать на области, граничащие с рубцом.In the boundary region weighting step 92, the processor 40 calculates the weight of the boundary region boundary

, which indicates the slope of the LAT variance with respect to nearby measurements. Steep dispersion may indicate areas bordering the scar.

, который указывает, имеет ли грань большой вектор медленности. Это повышает вероятность того, что грань является рубцом.In the step 93 of weighing a large slowness, the processor 40 calculates the weight of the slowness

, which indicates whether the face has a large slowness vector. This increases the likelihood that the face is a scar.

. Весь

отражает «прочность связи» между точкой измерения i и гранью j.In step 94, weighting the LAT link, processor 40 weights the difference between the measured LAT and the computed LAT of the edge. This is called the face measurement weight and is denoted

. Whole

reflects the "strength of the connection" between the measurement point i and face j.

, который отражает «прочность связи» между точкой измерения рубца i и гранью j. Вес уменьшается при увеличении пространственного расстояния между гранью

и гранью j.In step 95 of weighing the rumen connection, the processor 40 calculates the weight

, which reflects the "strength of connection" between the measurement point of the scar i and face j. The weight decreases as the spatial distance between the face increases.

and face j.

, которая указывает на область, в которой единичное измерение рубца окружено обычными (нерубцовыми) измерениями. Вес определяют с обеспечением уменьшения возможного наличия рубца для любой грани, в которой указание на рубец является слишком редким для рассмотрения в качестве действительно допустимого измерения. Это осуществляют путем анализа соотношения между близлежащими рубцовыми измерениями и близлежащими нерубцовыми измерениями. При наличии большого количества нерубцовых измерений и отдельного рубцового измерения вес, генерируемый этим модулем, уменьшает вероятность того, что эта грань представляет собой рубец.In the individual scar density weighting step 96, the processor 40 calculates the individual scar density weight

, which indicates the area where a single scar measurement is surrounded by normal (non-scar) measurements. The weight is determined to reduce the potential presence of a scar for any face where the indication of a scar is too rare to be considered a valid measurement. This is done by analyzing the relationship between nearby scar measurements and nearby non-scar measurements. When there are a large number of non-scar measurements and a single scar measurement, the weight generated by this module reduces the likelihood that this face is a scar.

The various weights calculated in steps 92-96 are processed by a scar information prediction and integration module 99 which calculates the possible presence of scars for each facet. As shown in FIG. 2, the scarring probabilities are subsequently included in an iterative calculation step 87 to reduce the power of the continuity equations in the line of blocking areas. In addition, as noted above, the relative power of the slowness norm and velocity norm equations also decreases. This allows you to more freely change the solution in the line of blocking areas.

In step 99, a scar probability weight is calculated for each facet by including the following rules in the cost function:

1. Scar/slow conduction areas are continuous.

2. A scar measurement means there is a scar nearby.

3. Areas with many nearby measurement points are conductive.

4. A large marginal area weight or measurements labeled with a dual potential indicate a scar margin.

5. Areas with little information about the surrounding space, with no nearby measurement points, are very likely to represent a scar.

6. Areas located in close proximity to the measurement with a scar mark and having a greater slowness are more likely to represent a scar.

The resulting cost is defined as the sum of the quadratic term and the linear term:

Lv. nine

представляет собой вектор весов для всех граней, а

H представляет собой симметричную матрицу членов стоимости, включающую пары весов граней:where

is the vector of weights for all faces, and

H is a symmetrical matrix of cost terms, including pairs of face weights:

Lv. ten

Quadratic cost term eq. 9 is defined as:

Lv. eleven

H reflects the continuity of scarring probabilities, which means that adjacent faces should have similar scarring probabilities. The purpose of this is to connect fragments of scar areas.

L is the linear cost term vector associated with each face. It consists of two contradictory terms, one of which includes all information that confirms the presence of a scar, and the other includes all information that indicates a normal face:

Subsequently, the linear cost vector is defined as

Lv. 12

,

для каждой грани подтверждают наличие либо рубцовой грани, либо нормальной грани и являются функциями весов, рассчитанных на стадиях 92–96. f представляет собой нормирующий фактор со значениями в диапазоне [0–1]. Если функция стоимости положительна для грани i, это означает, что оценка указывает на наличие рубца, а процессор 40 действует с обеспечением сведения к минимуму оценки посредством присвоения этой грани низкого веса. Если функция стоимости отрицательна, это указывает на нормальную грань, а процессор 40 действует с обеспечением максимального увеличения веса посредством присвоения этой грани большого веса. Это осуществляют путем записи линейного члена стоимости следующим образом:Line members

,

for each face confirm the presence of either a scar face or a normal face and are functions of the weights calculated in steps 92–96. f is a normalization factor with values in the range [0–1]. If the cost function is positive for face i, this means that the score indicates the presence of a scar, and processor 40 operates to minimize the score by assigning a low weight to that face. If the cost function is negative, this indicates a normal face, and processor 40 operates to maximize weight gain by giving that face a high weight. This is done by writing a linear cost term as follows:

Lv. thirteen

. Вес

, равный единице, указывает на проводящую грань, а вес, равный нулю, указывает на блокировку.This term will be added to the quadratic term (equation 11) to obtain the final cost function. For each face, an optimizing solution algorithm for the weights is determined

. The weight

A value of one indicates a conductive face, and a weight of zero indicates blockage.

равен 1, а

равен 0,2, которые представляют собой рубцовую и нормальную оценки грани,

будет равен 0,4, что больше нуля. Это побуждает оптимизирующий алгоритм присвоить вес

для этой грани, близкий к нулю для сведения к минимуму общей функции стоимости. Вес

, близкий к нулю, указывает на наличие рубца. В противном случае, если

равен 0, а

равен 1,

=–1, что дает отрицательную функцию стоимости, которая обеспечивает принудительное присвоение значения

, близкого к 1, оптимизирующему алгоритму, что указывает на нормальную грань. Применение отрицательных и положительных значений позволяет оптимизирующему алгоритму назначать значения, близкие к 1, для граней, которые являются нормальными, и при этом назначать значения, близкие к 0, для рубцовых граней.For example, assuming f is 0.5 if

equals 1, and

is 0.2, which are the cicatricial and normal facet scores,

will be equal to 0.4, which is greater than zero. This prompts the optimizing algorithm to assign a weight

for this face, close to zero to minimize the overall cost function. The weight

, close to zero, indicates the presence of a scar. Otherwise, if

is 0, and

equals 1,

=–1, which gives a negative cost function that enforces value assignment

, close to 1, by the optimizing algorithm, which indicates a normal face. Applying negative and positive values allows the optimization algorithm to assign values close to 1 for faces that are normal, while assigning values close to 0 for scarred faces.

путем сведения к минимуму общей функции стоимости

, что осуществляют с применением квадратичного программирования. Мы формулируем проблему следующим образом:Processor 40 receives vector weights

by minimizing the overall cost function

, which is carried out using quadratic programming. We formulate the problem as follows:

Lv. fourteen

которые представляют собой вероятность нормального включения грани i в стадию 87.This problem is solved using convex interior point optimization, as described in Florian A. Potra and Stephen J. Wright in, "Interior-point methods", Journal of Computational and Applied Mathematics, Volume 124, issues (1–2), pages 281–302, (2000). The output of this module is the weight of the cicatricial face

which are the probability of normal inclusion of face i in stage 87.

The block diagram example shown in Fig. 3 is provided solely for conceptual clarity. For example, additional implementation details are omitted for ease of presentation.

On FIG. 4 is a block diagram that schematically illustrates the method and algorithm for creating an EF activation wave map according to the method described in FIG. 2 in accordance with an embodiment of the present invention. The algorithm according to the present embodiment performs a process that starts with the processor 40 obtaining the 3D mesh representation 299a of the cardiac chamber at the step 299 of obtaining the input mesh. As shown, mesh 299aa contains irregular triangular faces. A mesh, such as mesh 299a, with triangles having large and small sides, which is typically generated using the aforementioned FAM technique, causes errors in geometric approximations performed in the described manner. Next, in the input mesh standardization step 300, the processor 40 rebuilds the input mesh 299 to produce a more correct mesh 300a. Re-meshing is also performed to optimize the computational steps implemented in the described method for estimating the EF-wave. In an embodiment, the grid resolution (number of triangles) determines the number of equations generated (eg, implemented in a given matrix size). In step 300, the design forces described in FIG. 2 and 3 by reducing this number of equations.

Next, in the mesh data fitting step 302, the processor 40 virtually transforms the triangular center fraction to better match nearby measured locations while maintaining the reconstruction state. Step 302 is required because, during meshing acquisition, the mapping catheter tends to press against the atrial walls, forming an "inflated" reconstruction geometry. Many of the measurements are far away from the reconstruction, which adds to the ambiguity when relating the location of the measurement to the camera geometry.

The processor 40 then performs the calculations described in FIG. 2 and FIG. 3, using a specific algorithm to obtain, for each face, optimized values of LAT, slowness, and scar probability of the triangular mesh, in step 90 of the interactive calculation. Typically, the scar weight is not updated every time the LAT and slowness values are updated.

Subsequently, at projection stage 304, processor 40 projects the values obtained at stage 302 onto the original grid 299a. As described above, the EF activation wave is calculated on a simplified grid. Nevertheless, the described method provides a final coherent map for the original grid. A map is defined at each face center for a low resolution mesh. Processor 40 plots the edges of the high resolution map (ie, input map 299a) onto points on the low resolution grid by "shooting" a ray from the high resolution grid in the direction normal to the face in and out.

When only one ray hits the low resolution grid, processor 40 selects that point as the associated point on the low resolution grid. When rays hit the grid in both directions, processor 40 selects the nearest point.

Subsequently, processor 40 interpolates the LAT value within the low resolution mesh face at the point hit by the ray and assigns this value to the high resolution mesh face. In an embodiment, processor 40 skips LAT information in areas of tissue in which the number of received data points is less than a predetermined desired value. Thus, an area of tissue in which EF mapping is too sparse is "gray" in any final visual representation.

In step 304, processor 40 generates an ESP activation coherence map 400, which is described in FIG. 6.

A key step in understanding arrhythmic flutter is being able to view the wave cycle across the entire grid. Visualization of 3D meshes on 2D screens involves some difficulties, including viewing the back side of the mesh, determining the anterior and distal portions of the mesh, etc. At step 410 of generating the isochronous view, the processor 40 converts the map 400 to the isochronous view 500.

The isochronous view 500 is a contour map 500a in which each contour line is a curve along which the LAT is constant. The isochronous view 500a is shown as an example of the left atrium in anteroposterior (AP) view. In practice, the map is color-coded.

Map 500a is transparent, allowing the user to view the entire range of the wave cycle in one view, while the frontal and distal components are easily distinguishable. Moreover, this view makes it easy to track the cycle throughout the structure, which allows a better understanding of the arrhythmia. The wave propagation velocity is easily determined from the density of the curves, since the contour interval has the same periodicity in time. An area with high contour density represents slow propagation, while low density reflects fast propagation. Scar areas are visible as homogeneous areas without contours.

The block diagram example shown in Fig. 4 is included solely for conceptual clarity. For example, illustrative graphics of some steps, such as step 302, have been omitted for ease of presentation.

On FIG. 5 is a schematic illustrative volumetric rendering of a left atrial coherent EF activation wave map 400 that takes into account EF wave reentry and scar tissue, in accordance with an embodiment of the present invention. As can be seen, map 400 shows a realistic continuous pattern of ESP activation in areas 210, since the rainbow artifact described above is eliminated in the manner described. As expected, there are interruptions between the EF conductive areas 210 and the non-conductive scar areas 220.

Insert 150 in FIG. 5 is a 3D interpolation of the left atrial EF activation color representation, calculated without the described methodology, for reference purposes. The example shown in box 150 does not take into account wave re-entry. As a result, and as shown in the inset, a “rainbow” artifact 155 appears. In the example shown in the inset, the propagation of the EF wave through the chamber covers the entire re-entry cycle, with the “late” wavefront in the cycle meeting the “early” wavefront next cycle. In this example, tissue regions are assigned LATs that are one cycle length apart. A narrow multi-tone rainbow-like trajectory which, when color-coded, indicates a transition of one full cycle length between adjacent areas, thus causing map interruptions that allow the embodiments of the present invention to be eliminated as explained above. Thus, the rainbow artifact described above is completely eliminated using the described ICEA method.

In some embodiments, a cyclic color scale is used to de-dependence on an early meets late effect in a window of interest (WOI). The colors reflect relative time differences with respect to the reference electrogram, eliminating the "early meets late" effect. The colored circle reflects the activation pattern. Thus, for example, red does not indicate an early wavefront, and purple does not indicate an early wavefront, as it does on a regular chart; the colored circle describes the proportional relationship in activation time in a natural, cyclic way, without creating rainbow-like transitions.

On FIG. 6 is a schematic pictorial volumetric rendering of a left atrial coherent EF activation wave map 440 with conduction arrows 330 superimposed, which illustrates the propagation of an EF activation wave, in accordance with an embodiment of the present invention. The conduction arrows 330 are vectors, each of a fixed length, and each pointing to a corresponding slowness vector at a location on the form, providing additional visualization of ESP activity. For example, conduction arrows 330 typically surround the scar area. In another embodiment (not shown in FIG. 6), the conduction arrows are of different lengths, reflecting the amount of slowness in addition to its direction.

It should be understood that the embodiments described above are by way of example only and that the present invention is not limited to the embodiments shown and described in detail above herein. On the contrary, the scope of the present invention includes both combinations and subcombinations of the various features described above, as well as variations and modifications thereof, which will be apparent to those skilled in the art upon reading the above description and which have not been described in the prior art. Documents incorporated into this patent application by reference are to be considered an integral part of the application, except that if the definitions of terms in these incorporated documents conflict with definitions made explicitly or implicitly in the present description, only the definitions of the present description should be considered.

Claims (22)

1. Method for iterative coherent mapping of electrophysiological activation of the heart, including: receiving an input grid representation of the cardiac chamber, a set of measured locations on the tissue of the wall of the cardiac chamber, and a corresponding set of local activation time (LAT) values measured at said locations; rebuilding the input mesh into a regular mesh containing regular polygons; matching the set of measured locations and corresponding LATs to regular polygons; iterative calculation for regular polygons of (i) respective LAT values and (ii) respective probabilities of scar tissue content in wall tissue so as to obtain an electrophysiological (EP) activation wave on a regular grid indicative of scar tissue; presentation of an electroanatomical map containing an EF activation wave and scar tissue superimposed on a regular grid; and binary marking at least some of the measured locations to indicate at least one of the presence of a scar and the presence of a dual ESP potential. 2. The method of claim 1, wherein the regular polygons are regular triangles. 3. The method according to claim 1, wherein the iterative calculation of LAT values and probabilities is an iterative solution of a set of equations for LAT values, slowness values, and ESP wave propagation. 4. The method of claim 1, wherein the iterative calculation of the probabilities is to reassign each regular polygon to a slowness vector recalculated using a scar multiplier. 5. The method of claim 1, wherein the representation of the electroanatomical map is an overlay of conduction arrows on the electroanatomical map indicating scar tissue. 6. System for iterative coherent mapping of electrophysiological activation of the heart, comprising: an interface that is configured to receive an input grid view of the cardiac chamber, a set of measured locations on the tissue of the cardiac chamber wall, and a corresponding set of local activation time (LAT) values measured at said locations, wherein the interface is configured to receive a binary marking of at least , some of the measured locations that indicate at least one of the presence of a scar and the presence of a dual ESP potential; and a processor that is configured to: - rebuilding the input grid into a regular grid containing regular polygons; - matching the data of the set of measured locations and the corresponding LATs to regular polygons; - iterative calculation for regular polygons of (i) respective LAT values and (ii) respective probabilities of scar tissue content in wall tissue so as to obtain an electrophysiological (EP) activation wave on a regular grid indicative of scar tissue; and - presentation of an electroanatomical map containing an EF–activation wave and scar tissue superimposed on a regular grid. 7. The system of claim 6 wherein the regular polygons are regular triangles. 8. The system of claim 6, wherein the processor is configured to iteratively calculate LAT values and probabilities by iteratively solving a set of equations for LAT values, slowness values, and ESP wave propagation. 9. The system of claim 6, wherein the processor is configured to iteratively calculate the probabilities by reassigning each regular polygon to a slowness vector recalculated using a scar multiplier. 10. The system of claim 6, wherein the processor is configured to present the electroanatomical map by superimposing conduction arrows on the electroanatomical map indicating scar tissue.

Images