System and method of operation thereof

Abstract

To automatically identify a location of a focal arrhythmogenic activity.SOLUTION: A method includes receiving, via a plurality of electrodes in a heart, a collection of acquisitions, where each acquisition comprises a set of electrophysiological (EP) signals measured by the electrodes. Respective directions of arrival (DOA) and respective distances relative to the electrodes from which the set of EP signals is originated are estimated for each acquisition. The acquisitions are aggregated to form a statistical distribution of the acquisitions as a function of the estimated DOA and distance. Using a statistical test, it is checked whether the statistical distribution is consistent, in accordance with a predefined consistency criterion. If the statistical distribution is found consistent, an estimated location in the heart of a focal source of an arrhythmogenic activity that generated the received EP signals is derived from the statistical distribution. The estimated location of the focal source is overlaid on an anatomical map of at least a portion of the heart.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application is related to U.S. patent application filed on even date, entitled "Error Estimation of Local Activation Times (LAT) measured by Multiple Electrode Catheter," attorney docket number 1002-1729.1, the disclosure of which is incorporated herein by reference.

[0002] FIELD OF THEINVENTION The present invention relates generally to electrophysiological mapping, and more particularly to electrophysiological mapping of the heart. [Background technology]

[0003] Invasive cardiac techniques for mapping electrophysiological (EP) properties of cardiac tissue have been previously proposed in the patent literature. For example, US Patent Application Publication No. 2017 / 0042449 describes a system and method for regional EP characterization of cardiac substrates using a multi-electrode catheter. The system selects at least one subset of electrodes from a plurality of electrodes and derives at least one orientation-independent signal from the at least one subset of electrodes from information content corresponding to a weighted portion of the electrogram signal. The system displays or outputs the catheter orientation-independent electrophysiological information to a user or process.

[0004] As another example, US Patent Publication No. 2015 / 0366476 describes a system and method for mapping cardiac electrical activity. The system may include a catheter shaft with multiple electrodes. A processor of the system may be capable of collecting a set of signals from at least one of the multiple electrodes. The set of signals may be collected over a period of time. The processor may also be capable of calculating at least one propagation vector from the set of signals, generating a data set from the at least one propagation vector, generating a statistical distribution from the data set, and generating a visual representation of the statistical distribution, such as a circular histogram of angles. By comparing the excitation times sensed by the target electrode from which the propagation vector is determined and adjacent electrodes, the direction (e.g., propagation angle) and velocity of the cell wavefront propagation may be determined.

[0005] US Patent Application Publication No. 2017 / 0202470 describes a system and method for identifying focal sources, which may include detecting electrocardiogram (ECG) signals over time via sensors, each ECG signal having a location in the heart and detected via one of the sensors indicative of electrical activity of the heart, each signal including at least an R wave and an S wave, creating an RS map comparing an RS ratio of each of the ECG signals, the RS ratio including a ratio of an absolute magnitude of the R wave to an absolute magnitude of the S wave, identifying a local activation arrival time (LAT) for each of the ECG signals, correlating the RS ratio of the ECG signals on the RS map with the identified LAT, and using the correlation to identify focal sources.

[0006] US Patent Application Publication No. 2017 / 0281031 describes electroanatomical mapping performed by inserting a multi-electrode probe into the heart of a living subject, recording electrograms from the electrodes simultaneously at each location in the heart, defining delimitations of each respective excitation time interval in the electrograms, generating a map of electrical propagation waves from the excitation time intervals, maximizing wave coherence by adjusting local excitation times within the excitation times of the electrograms, and reporting the adjusted local excitation times.

[0007] U.S. Patent Application Publication No. 2004 / 0243012 describes a method and system for identifying and localizing a reentry circuit tachycardia isthmus in a subject's heart during sinus rhythm. The method may include (a) receiving electrogram signals from the heart during sinus rhythm via electrodes, (b) generating a map based on the electrogram signals, (c) determining a location of the reentry circuit tachycardia isthmus in the heart based on the map, and (d) displaying the location of the reentry circuit tachycardia isthmus. Summary of the Invention [Means for solving the problem]

[0008] An embodiment of the present invention provides a method that includes receiving, via a plurality of electrodes in a heart, a collection of captured contents, each captured content including a set of electrophysiological (EP) signals measured by an electrode. A respective direction of arrival (DOA) and a respective distance relative to the electrode from which the set of EP signals originated are estimated for each of the captured contents. The captured contents are tallied to form a statistical distribution of the captured contents as a function of the estimated DOA and distance. A statistical test is used to check whether the statistical distribution of the captured contents is consistent according to a predefined consistency criterion. If the statistical distribution of the captured contents is found to be consistent, an estimated location within the heart of a proarrhythmic focal source that generated the received EP signal is derived from the statistical distribution. The estimated location of the focal source is overlaid on an anatomical map of at least a portion of the heart.

[0009] In some exemplary embodiments, estimating the DOA and distance for a given acquisition includes extracting a respective set of relative arrival times from a set of EP signals in the given acquisition and estimating the DOA and distance using the extracted relative arrival times.

[0010] In some demonstrative embodiments, aggregating the captured content includes pre-filtering the captured content according to a respective set of relative arrival times extracted from each captured content and including only the pre-filtered captured content in the statistical distribution of the captured content.

[0011] In one exemplary embodiment, pre-filtering the acquisition content according to the extracted set of relative arrival times includes: (a) using the estimated DOA and distance, calculating, for each acquisition content, a modeled set of relative arrival times that would result from EP waves emanating from a focal source at the estimated DOA and distance; and (b) for each acquisition content, determining a similarity between the extracted set of relative arrival times and the modeled set by applying a predefined geometric test.

[0012] In some exemplary embodiments, estimating the similarity includes computing a geometric test of cosine similarity between the two sets, hi other embodiments, estimating the similarity includes computing an estimation error for each relative arrival time and comparing the estimation error to a given threshold.

[0013] In an exemplary embodiment, the method further includes adjusting the time values ​​of the annotations for the entire EP signal where the voltage-time slope of the EP signal is less steep than a predefined slope, using the modeled set of relative arrival times.

[0014] In another exemplary embodiment, pre-filtering the acquisitions includes discarding one or more acquisitions determined to have a set of dissimilar arrival times.

[0015] In some exemplary embodiments, deriving the estimated location includes fitting a curve to the statistical distribution and finding the maximum of that curve as a function of the estimated DOA and distance.

[0016] In some exemplary embodiments, estimating the DOA and distance includes minimizing a cost function. In other exemplary embodiments, minimizing the cost function includes minimizing a weighted cost function. In further embodiments, minimizing the cost function includes iteratively minimizing the cost function by eliminating, at each iteration, the EP signal value with the largest estimation error.

[0017] In one exemplary embodiment, deriving the estimated location includes applying a k-means analysis to the statistical distribution, projecting the estimated location onto the anatomical structure, and selecting locations having a projection distance less than a given value.

[0018] According to an exemplary embodiment of the present invention, there is further provided a system including an interface and a processor, the interface configured to receive a collection of captured content captured by a plurality of electrodes in the heart, each captured content including a set of electrophysiological (EP) signals, the processor configured to (a) estimate for each captured content a respective direction of arrival (DOA) and respective distance relative to the electrodes from which the set of EP signals originated, (b) aggregate the captured content to form a statistical distribution of the captured content as a function of the estimated DOA and distance, (c) check using a statistical test whether the statistical distribution of the captured content is consistent according to a predefined consistency criterion, (d) if the statistical distribution of the captured content is found to be consistent, derive from the statistical distribution an estimated location within the heart of a proarrhythmic focal source that generated the received EP signal, and (e) overlay the estimated location of the focal source on an anatomical map of at least a portion of the heart.

[0019] Another exemplary embodiment of the invention provides a method that includes receiving a set of captured content via a plurality of electrodes in a heart, each captured content including a set of electrophysiological (EP) signals measured by an electrode. A respective direction of arrival (DOA) and a respective distance relative to the electrode from which the set of EP signals originated are estimated for at least some of the captured content. A timing error is estimated in at least one of the EP signals based on the estimated DOA and distance. A timing of the EP signal is adjusted to match the estimated DOA and distance and to correct the error. An EP map of at least a portion of the heart is generated using the set of EP signals including the adjusted EP signal.

[0020] In some exemplary embodiments, generating an EP map includes generating a local activation time (LAT) map.

[0021] In some exemplary embodiments, estimating the DOA and the range includes deriving the DOA and the range that minimizes a cost function.

[0022] In one exemplary embodiment, adjusting the timing of the EP signal includes (a) selecting an initial annotation in the EP signal; (b) determining a corrected annotation corresponding to the initial annotation based on the estimated DOA and distance; and (c) adjusting the timing of the EP signal upon determining that the corrected annotation meets a predefined condition.

[0023] In another exemplary embodiment, verifying that the corrected annotation satisfies the predefined condition includes verifying that the corrected annotation falls within a dip between adjacent peaks of the EP signal. In yet another exemplary embodiment, verifying that the corrected annotation satisfies the predefined condition includes verifying that the corrected annotation and the initial annotation are in the same monotonically decreasing interval of the EP signal.

[0024] In some exemplary embodiments, verifying that the corrected annotation satisfies a predefined condition includes verifying that the corrected annotation and the initial annotation are in the same monotonic interval of the EP signal and that the slope of that interval is lower than a predefined threshold slope.

[0025] According to an exemplary embodiment of the present invention, there is further provided a system including an interface and a processor, the interface configured to receive a collection of acquisitions acquired by a plurality of electrodes in a heart, each acquisition including a set of electrophysiological (EP) signals, the processor configured to (a) estimate, for at least some of the acquisitions, a respective direction of arrival (DOA) and a respective distance relative to an electrode from which the set of EP signals originated, (b) estimate a timing error in at least one of the EP signals based on the estimated DOA and distance, (c) adjust a timing of the EP signal to match the estimated DOA and distance and to correct the error, and (d) generate an EP map of at least a portion of the heart using the set of EP signals including the adjusted EP signal. [Brief description of the drawings]

[0026] The invention will be more fully understood from consideration of the following detailed description taken in conjunction with the drawings, in which: [Figure 1] FIG. 1 is a schematic, pictorial illustration of an electrophysiological (EP) mapping system, in accordance with an exemplary embodiment of the present invention. [Diagram 2] 1 is a flow chart that generally illustrates a method for automatic localization of an arrhythmia focal source, in accordance with an exemplary embodiment of the present invention. [Figure 3A] 2A-2C are two plots showing graphs of EP signals captured by the system of FIG. 1 in accordance with an exemplary embodiment of the present invention. [Figure 3B] 2A-2C are two plots showing graphs of EP signals captured by the system of FIG. 1 in accordance with an exemplary embodiment of the present invention. [Figure 4A] 4A and 4B are plots illustrating relative arrival times extracted and modeled using the EP signals of the graphs of FIGS. 3A and 3B, respectively, in accordance with an exemplary embodiment of the present invention. [Figure 4B]4A and 4B are plots illustrating relative arrival times extracted and modeled using the EP signals of the graphs of FIGS. 3A and 3B, respectively, in accordance with an exemplary embodiment of the present invention. [Diagram 5] 4C is a flow chart that generally illustrates a method for deriving direction of arrival (DOA) and distance according to the steps illustrated in FIGS. 4A and 4B, in accordance with an exemplary embodiment of the present invention. [Figure 6] 5 is a flow chart that conceptually illustrates a method for deriving a direction of arrival (DOA) from a focal source, in accordance with another exemplary embodiment of the present invention. [Figure 7A] 2A and 2B are plots showing (a) a graph of a unipolar EP signal captured by the system of FIG. 1 , (b) the catheter position, and (c) an isochronal map showing the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention. [Figure 7B] 2A and 2B are plots showing (a) a graph of a unipolar EP signal captured by the system of FIG. 1 , (b) the catheter position, and (c) an isochronal map showing the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention. [Figure 7C] 2A and 2B are plots showing (a) a graph of a unipolar EP signal captured by the system of FIG. 1 , (b) the catheter position, and (c) an isochronal map showing the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention. [Figure 8A] 2A-2C are plots illustrating a graph of a unipolar EP signal captured by the system of FIG. 1 and isochronal maps illustrating the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention; [Figure 8B] 2A-2C are plots illustrating a graph of a unipolar EP signal captured by the system of FIG. 1 and isochronal maps illustrating the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention; [Figure 9A]11A-11C are plots showing a graph of a unipolar EP signal with an estimation error higher than a given threshold, and a graph of an initially estimated position of a focal source in XY space, respectively, in accordance with an exemplary embodiment of the present invention. [Figure 9B] 11A-11C are plots showing a graph of a unipolar EP signal with an estimation error higher than a given threshold, and a graph of an initially estimated position of a focal source in XY space, respectively, in accordance with an exemplary embodiment of the present invention. [Figure 10] 9C is a graph showing the estimated position of the focal source of FIG. 9B for nine iterations of an iterative DOA model, in accordance with an exemplary embodiment of the present invention. [Figure 11A] 4 is a histogram of direction of arrival (DOA) and distance from a focal source in accordance with an exemplary embodiment of the present invention. [Figure 11B] 4 is a histogram of direction of arrival (DOA) and distance from a focal source in accordance with an exemplary embodiment of the present invention. [Figure 12] 1 is a plot illustrating DOA clusters analyzed by a k-means clustering model, in accordance with an exemplary embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027] Overview In focal arrhythmias, abnormal electrophysiological (EP) wave impulses are abnormally propagated from ectopic foci in the heart. Focal arrhythmias may be caused by localized abnormal cardiac tissues that trigger abnormal EP waves or by localized abnormal cardiac tissues that form small reentry pathways that cause erroneous propagation of existing EP waves. In some patients, the localized arrhythmogenic tissues may be ablated to eliminate focal arrhythmias. Thus, locating focal arrhythmogenic tissues may be clinically useful.

[0028] Exemplary embodiments of the invention described below provide EP mapping systems and methods that automatically identify the location of focal pro-arrhythmic effects in the heart. Additionally or alternatively, some exemplary embodiments provide methods for estimating annotation errors in measured EP values ​​and correcting the annotation errors according to their causes. In some exemplary embodiments, a processor generates an EP map (e.g., a LAT map) of at least a portion of the heart based on a set of annotations, including the corrected annotations.

[0029] The EP mapping system uses a multi-electrode catheter, such as the Pentaray® catheter (Biosent-Webster, Irvine, California), to obtain multiple acquisitions from the area of ​​the heart covered by the electrodes. Each acquisition includes a set of EP signals measured by the electrodes, which is sized according to the number of electrodes. However, other multi-electrode catheters may be used as desired with the disclosed techniques.

[0030] In some exemplary embodiments, the processor then determines the direction of arrival (DOA) and distance R relative to the electrode from which the set of captured EP signals originated. DOA For each acquisition, a processor aggregates the acquisitions to form a statistical distribution of the acquisitions (e.g., a histogram or a cluster map in XY space) as a function of the estimated DOA and distance, and uses statistical tests to check whether the statistical distribution of the acquisitions is consistent.

[0031] Consistency is checked according to a predefined consistency criterion. Examples of relevant consistency tests include, but are not limited to, the use of stationarity estimates or confidence intervals. As another example, a cluster map in XY space may be consistent if one or more clusters in the map each contain at least a given percentage (e.g., 10%) of the data points, as described below.

[0032] If the statistical distribution of the captured content is found to be consistent, the processor derives from the statistical distribution an estimated location within the heart of the pro-arrhythmic focal source that generated the received EP signal. Finally, the processor overlays the estimated location of the focal source on an anatomical map of at least a portion of the heart.

[0033] In some exemplary embodiments, to estimate DOA and distance, the processor annotates each EP signal with the time of arrival of the EP wavefront, hereinafter also referred to as the initial annotation. The processor extracts a respective set of relative arrival times from the set of initially annotated signals (i.e., from a given acquisition). Using a geometric model, the processor analyzes the extracted set of relative times to indicate the nature of the EP wave in question, as described below. This model assumes that each acquisition is uniquely associated with a single moving, widespread EP wavefront that has a constant velocity throughout the region in which the EP signal is acquired.

[0034] In some exemplary embodiments, when aggregating acquisitions, the processor pre-filters the acquisitions and includes only the pre-filtered acquisitions in the statistical distribution of acquisitions. The processor pre-filters the acquisitions by applying the following steps: (a) using the estimated DOA and distance to calculate, for each acquisition, a modeled set of relative arrival times resulting from EP waves originating at a focal source with the estimated DOA and distance relative to the catheter; and (b) applying a test for each acquisition to determine the degree to which the extracted set of relative arrival times is similar to the modeled set, discarding acquisitions that result in dissimilar sets of timings. Examples of such tests include tests of geometric similarity and comparison of the estimated error to a given threshold (hereinafter also referred to as timing error).

[0035] In some exemplary embodiments, the geometric similarity test includes applying a geometric test of cosine similarity between the extracted set and the modeled set of relative arrival times. The similarity may range between zero, representing perfect dissimilarity, and one, representing perfect similarity. In an alternative exemplary embodiment, a least squares method is used as the geometric test.

[0036] Abnormal EP waves are not necessarily of focal origin nature, which may be indicated by a similarity check. In an exemplary embodiment, regardless of the nature of the EP wave, i.e. focal type or abnormality, the derived modeled relative time may be used to adjust the time values ​​of the initial annotations that are not well-defined, i.e. the potential-time slope of the wavefront is less steep than a predefined slope. The adjusted annotations are also referred to as corrected annotations in the following.

[0037] Typically, the processor is programmed with software that contains specific algorithms that enable the processor to carry out each of the processor-related steps and functions outlined above.

[0038] The disclosed techniques for automatically identifying focal sources of pro-arrhythmic activity within the heart may improve clinical outcomes of associated catheter-based treatment of arrhythmias.

[0039] System Description 1 is a schematic, pictorial illustration of an electrophysiological (EP) mapping system 10, in accordance with an exemplary embodiment of the present invention. System 10 includes a catheter 14 that is inserted by a physician 32 through a patient's vascular system into a chamber or vasculature of a heart 12. The physician 32 brings a distal tip 18 of the catheter into contact with a heart wall, e.g., at an EP mapping target site. Catheter 14 typically includes a handle 20 having suitable controls that enable physician 32 to steer, position, and orient the distal end of catheter 14 as desired for EP mapping.

[0040] Catheter 14 is a multi-electrode catheter, such as the aforementioned Pentaray® catheter shown in inset 37. Pentaray catheter 14 includes five flexible arms 15, each carrying four electrodes 16. Thus, a total of 20 EP signals are acquired by system 10 at each instance of EP signal acquisition, as further described in FIG.

[0041] The catheter 14 is connected to a console 24, which allows a physician 32 to observe and adjust the function of the catheter 14. To assist the physician 32, the distal portion of the catheter 14 may include various sensors, such as a contact force sensor (not shown) and a magnetic sensor 33, which provide position, direction, and orientation signals to a processor 22 located in the console 24. The processor 22 may perform several processing functions, as described below. In particular, electrical signals may be transmitted to and from the heart 12 from electrodes 16 located at or near the distal tip 18 of the catheter 14 through cables 31 to the console 24. Pacing signals and other control signals may be transmitted from the console 24 through the cables 31 and the electrodes 16 to the heart 12.

[0042] The console 24 includes a monitor 29 that is driven by the processor 22. Signal processing circuitry in the electrical interface 34 typically receives, amplifies, filters, and digitizes signals from the catheter 14, including signals generated by the sensors and plurality of sensing electrodes 16 described above. The digitized signals are received and used by the console 24 and the positioning system to calculate the position and orientation of the catheter 14, and to analyze the EP signals from the electrodes 16, as described in more detail below.

[0043] During the disclosed procedure, the position of each of the electrodes 16 is tracked. This tracking may be performed, for example, using the CARTO® 3 system from Biosense Webster. Such a system measures the impedance between the electrodes 16 and a number of external electrodes 30 coupled to the patient's body. For example, three external electrodes 30 may be coupled to the patient's chest and another three external electrodes may be coupled to the patient's back. (For ease of illustration, only the chest electrodes are shown in FIG. 1.) Wire connections 35 link the console 24 to the body surface electrodes 30 and other components of the positioning subsystem to measure the coordinates of the position and orientation of the catheter 14. A method of tracking the position of the electrodes 16 based on electrical signals, termed Excitation Current Location (ACL), has been implemented in various medical applications, for example, the aforementioned CARTO® 3 system. Details of the ACL subsystem and process are presented in U.S. Pat. No. 8,456,182, which is assigned to the assignee of the present patent application and whose disclosure is incorporated herein by reference.

[0044] In some exemplary embodiments, the system 10 includes, in addition to or instead of the ACL tracking subsystem, a magnetic position tracking subsystem that uses a magnetic field generating coil 28 to generate magnetic fields at a predetermined fluence and senses these fields at the catheter to determine the position and orientation of a magnetic sensor 33 at the distal end of the catheter 14. Because the electrodes 16 know their position on the arm 15 and in relation to one another, when the catheter 14 is magnetically tracked within the heart, the position of each of the electrodes 16 within the heart will be known. A suitable magnetic position tracking subsystem is described in U.S. Patent Nos. 7,756,576 and 7,536,218, which are assigned to the assignee of the present patent application and whose disclosures are incorporated herein by reference.

[0045] Based on the EP signals from the position-tracked electrodes 16, electrical activation maps can be generated according to the methods disclosed in U.S. Pat. Nos. 6,226,542, 6,301,496, and 6,892,091, which are assigned to the assignee of the present patent application and whose disclosures are incorporated herein by reference.

[0046] Processor 22 operates system 10 using software stored in memory 25. The software may be downloaded to processor 22 in electronic form, for example over a network, or may alternatively or additionally be provided and / or stored on a non-transitory tangible medium, such as magnetic, optical, or electronic memory. In particular, processor 22 executes dedicated algorithms disclosed herein, including FIG. 2, that enable processor 22 to perform the steps of the present disclosure, as described further below.

[0047] The exemplary diagram shown in Figure 1 has been chosen purely for conceptual clarity, other types of sensing geometries may also be employed, such as the Lasso® catheter (manufactured by Biosense Webster, Inc.).

[0048] Automatic Identification of Focal Source Location in Atrial Fibrillation(AF). 2 is a flow chart that generally illustrates a method for automatic localization of focal sources of arrhythmias, according to an exemplary embodiment of the present invention. The algorithm executes a process that, according to the presented embodiment, begins with a physician 32 inserting a catheter 14 having multiple sensing electrodes 16 into a patient's heart 12 at a catheter insertion step 100.

[0049] Next, the system 10 receives a collection of acquisitions of a set of EP signals from the electrodes 16 that have been brought into contact with cardiac tissue by the physician 32, at an EP signal acquisition step 102. In a typical diagnostic interval of 30 seconds utilized by some of the disclosed embodiments, the system collects 100-200 acquisitions that include ECG segments, each having a typical duration of 100-200 milliseconds. In some exemplary embodiments, an automatic segmentation of the 30 second window is performed by the processor 22 to generate 100-200 millisecond segments, each corresponding to a single excitation propagating through the atrium.

[0050] Processor 22 then derives an estimate of the DOA and the estimated focal source distance from each capture in a DOA derivation step 104. Processor 22 then pre-filters each capture to discard captures that are not suitable for inclusion in a subsequent statistical analysis in a pre-filtering step 106. In some exemplary embodiments, processor 22 pre-filters the captures by comparing the error in estimated relative time between the extracted EP values ​​and the modeled EP values. This stage may include further processing attempts, such as including iterative calculations, to improve the DOA estimate and thereby reduce the estimation error, as described below.

[0051] In another exemplary embodiment, the processor pre-filters the captured content by performing a geometric test of consistency between the modeled EP values ​​and the extracted EP values ​​(e.g., a cosine similarity test).

[0052] In any event, in a counting step 108, processor 22 counts the verified DOA and distance values. Processor 22 then performs statistical tests in a statistical analysis step 110 to find one or more candidates for DOA and distance (i.e., candidate focus positions), if they exist.

[0053] The processor verifies whether the candidate locations are valid and which of the candidate locations are valid locations by projecting the locations on the anatomical structures in a projection verification step 112 .

[0054] Additionally, the processor verifies for the verified candidate locations in a direct verification step 114 that at least a minimum number of focus-indicating ECG signals, described below, have been captured at the verified locations.

[0055] Finally, in a focal source presentation step 116 , processor 22 superimposes one or more identified locations of pro-arrhythmic focal sources onto an anatomical map of at least a portion of heart 12 .

[0056] The exemplary flow chart shown in Figure 2 has been selected solely for the purposes of conceptual clarity. Further details and specific embodiments of the steps briefly described above are provided below, including the flow charts of Figures 5 and 6.

[0057] Deriving and validating direction of arrival (DOA) and distance by the first method 3A and 3B are two plots showing a graph 42 of an EP signal acquired by the system of FIG. 1, according to an exemplary embodiment of the present invention. The illustrated EP signal was acquired, for example, by the EP mapping system of FIG. 1 using catheter 14. The two acquisitions are part of a plurality of acquisitions numbered between tens and hundreds. Such a collection may include acquisitions acquired at different intracardiac placements of the catheter and / or repetitive acquisitions acquired during the same placement. Using a tracking system, each of the plurality of electrodes has an intracardiac location.

[0058] The graph shows the first annotated times 44 when an EP wave "hit" each of the 20 electrodes of the catheter 14. These annotations were made by methods known in the art, such as those described in U.S. Patent No. 8,700,136, which is assigned to the assignee of the present patent application, the disclosure of which is incorporated herein by reference.

[0059] In Figure 3A, the EP wave first strikes electrodes "13" and "14", then electrodes "15" and "7", etc. In Figure 3B, the EP wave first strikes electrodes "5" and "6", then electrodes "7" and "8", etc.

[0060] As can be seen in Figures 3A and 3B, some of the potential-time gradients of the initially annotated EP signals are poorly defined, i.e., gentler (e.g., in graphs 10 and 11 of Figure 3B). In an exemplary embodiment, the disclosed technique improves the accuracy of such annotations by deriving corrected annotations, as shown in Figures 7A, 8A, and 9A, even when the technique does not identify a focal source of proarrhythmia. In an exemplary embodiment, the modeled times derived below are based on the known geometry of the catheter 14 and are used to adjust the time values ​​of the initial annotations that are poorly defined, i.e., have potential-time gradients that are gentler than the specified gradients, as seen in Figures 3A and 3B.

[0061] 3A and 3B are provided by way of example only: if another catheter with multiple electrodes is used, such as a basket or lasso catheter, the size of the capture (e.g., the number of graphs in a set) and the number of times they are annotated will reflect the geometry of the given catheter, but will be used by the disclosed techniques as well.

[0062] Figures 4A and 4B are plots showing relative arrival times extracted and modeled using the EP signals of the graphs of Figures 3A and 3B, respectively, in accordance with an embodiment of the present invention. Processor 22 pre-filters the acquisition content according to each extracted set of arrival times by performing the steps described below.

[0063] The extracted times 66 in Figures 4A and 4B are derived by processor 22 calculating the difference between the original annotated times 44 in Figures 3A and 3B, respectively. The modeled relative times 68 in each of Figures 4A and 4B are derived by processor 22 using corrected annotated times (not shown), as described below.

[0064] A color scale 48 at the top of Figures 4A and 4B encodes the relative arrival times by color-coding each of the 20 illustrated electrodes 16 of the catheter 14. The top of Figures 4A and 4B further illustrate the tracked positions of the electrodes 16 (on the roughly demarcated arm 15 of the Pentaray catheter 14) in two examples where the electrodes captured EP signals. The position of each electrode 16 in 3D space is tracked, for example, using the ACL tracking technique previously described. The XYZ axes (Z not shown) belong to a fixed reference axis system, such as the position tracking system 20 that applies the ACL method.

[0065] Based on the extracted relative arrival time 66, the processor 22 estimates geometrically (e.g., as indicated by arrows 40a and 40b) an estimated focal source 50 from which the EP wave is likely to emerge, as further marked at the distance indicated by a line 60 connecting the common location where the arrows 40a and 40b originate from the distal tip 18 of the catheter 14.

[0066] 4A and 4B, the bottom portion, shows on the same graph a set of extracted arrival times 66 and a set of respective modeled arrival times 68. The relative arrival times 66 are derived by the processor 22 from the first annotated time that the actual EP wave "hits" the electrode. The modeled times 68 are calculated by the processor 22 using the estimated DOA 55 and distance 60 of the simulated EP wave, assuming that it originates at a focal source 50 with that estimated DOA and distance to the electrode 16.

[0067] In one exemplary embodiment, the captured content is determined by the processor as indicative of a focus source only if the cosine similarity timing match derived from the captured content exceeds a value of 0.9. Geometric tests (e.g., metrics) other than cosine similarity that check the degree to which the extracted and modeled sets are similar, such as Hamming distance, may also be utilized.

[0068] The processor may be referred to herein as a vector S EX and S MD We estimate the degree to which each pair of such sets for each capture, denoted as , are similar using the cosine similarity formula.

[0069]

number

[0070] Cosine similarity is where the normalized dot product of two ordered sets is calculated and can give any value between -1 and 1. In practice, cosine similarity is specifically used in the positive space where the result falls within [0,1). For example, a value of 1 corresponds to perfect similarity, while a value of zero or any negative value indicates perfect dissimilarity. In one exemplary embodiment, if the calculated cosine similarity gives a value above a predefined minimum value, such as greater than 0.9, the processor determines that the sets are similar.

[0071] The processor 22 performs a similarity check across all sets derived from the collection of captured content, and discards captured content having a cosine similarity below a predefined minimum value (eg, <0.9).

[0072] The processor 22 then calculates the phase of the vector (r 50 -r 18 ), and the distance 60 from which the set of signals originated are calculated using Equations 2 and 3.

[0073]

number

[0074] Figure 5 is a flow chart that generally illustrates a method for deriving direction of arrival (DOA) and distance following the steps shown in Figures 4A and 4B, in accordance with an exemplary embodiment of the present invention. The process begins in a relative time extraction step 200 with processor 22 extracting relative times of arrival 66 from the first annotated EP signal.

[0075] Next, based on the assumption that the EP signal is generated by (a) a single EP wave having a wide wavefront and (b) an EP wave whose wavefront propagates at a constant velocity, and based on the known geometry of catheter 14, processor 22 derives an estimate of DOA 55 and distance 60 of the EP wave for each capture in DOA and distance estimation step 202.

[0076] Next, in a relative time modeling step 204, based on the estimated DOA 55 and distance 60 values, processor 22 calculates the relative times generated by focal waves with the provisional DOA and distance from step 106. Processor 22 then checks, in a similarity checking step 206, the extent to which the sets of extracted and modeled relative times are similar, for example using a cosine similarity test.

[0077] If the sets are found to be dissimilar, then in acquisition discard step 208, processor 22 stores or discards the bad DOA and distance values ​​as not indicative. In aggregation step 210, all acquisitions having extracted and modeled relative time sets found to be similar (i.e., passed pre-filtering) are aggregated by processor 22 into separate distributions as a function of DOA and distance (e.g., into histograms 70 and 72 in Figures 11A and 11B below). As shown in Figures 11A and 11B below, the aggregated DOAs and distances are statistically analyzed to find the location of a focal source, if one is believed to be present from the histograms.

[0078] The exemplary flow chart shown in FIG. 5 has been selected solely for the purpose of conceptual clarity. Typically, additional steps may be performed, such as the physician 32 first anatomically mapping the relevant parts of the heart 12 (e.g., using a fast anatomical mapping (FAM) procedure) and obtaining an anatomical map. The criteria may vary depending on the type of statistical tool used. In an exemplary embodiment, the discarded set of modeled timings may still be used to adjust each initially annotated time that is not clearly contoured, as described below in “LAT Improvement.”

[0079] Deriving and validating direction of arrival (DOA) and range by the second method FIG. 6 is a flow chart that conceptually illustrates a method for deriving a direction of arrival (DOA) from a focal source, in accordance with another exemplary embodiment of the present invention.

[0080] 6 begins with processor 22 extracting relative times of arrival 66 from the first annotated EP signal in relative time extraction step 300. Next, the algorithm checks whether to apply a 3D or 2D weighted DOA model estimation in projection check step 304 by checking whether it is valid to project the catheter into 2D space (i.e., projection step 302), as described below.

[0081] Next, depending on whether the projection step 304 is found to be an invalid or valid step, the processor 22 executes either a 3D weighted DOA discovery model in a 3D modeling step 306 or a 2D weighted DOA discovery model in a 2D modeling step 308, respectively.

[0082] Using either the 3D or 2D model, the processor 22 then checks, in an estimation error step 310, whether the estimation error between the modeled relative time and the extracted relative time is less than a given threshold.

[0083] If the estimation error is within a given threshold, the processor 22 applies a LAT refinement calculation to make the DOA estimation more accurate, in a LAT refinement step 312. LAT refinement is described further below.

[0084] On the other hand, if the estimation error is higher than the given threshold, the processor 22 executes a DOA iterative model in a DOA iterative estimation step 314 .

[0085] In a follow up error step 316, the processor 22 checks whether the recalculated estimation error using the iterative model is below a given threshold. If not, the processor 22 stores or discards the bad DOA and range values ​​as not indicative in a discard acquisition step 318. However, if the iterative model is successful, the processor 22 applies the LAT refinement step 312 to the results.

[0086] In any event, in an aggregation step 320, the successfully prefiltered LAT-improved estimates of DOA and range are aggregated by processor 22. As will be shown below, the aggregated DOA values ​​are statistically analyzed to find one or more locations of a focal source, if one is deemed to be present by the statistical model.

[0087] In an exemplary embodiment, the disclosed method described in FIG. 6 for deriving DOA and direction utilizes a cost function in 3D space, as described in FIGS. 7A-7C.

[0088] 7A-7C are, respectively, (a) plots illustrating a graph of a unipolar EP signal captured by the system of FIG. 1, (b) the position of catheter 14, and (c) isochronal maps illustrating the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention. Specifically, FIG. 7B illustrates the position of distal tip 18 of catheter 14 in XY space, and the actual position 338 of the catheter on the anatomy of the left atrium 340.

[0089] The estimated error 550 (ie, timing error 550) is shown in Figures 7A, 8A and 9A as the time difference between the original annotation and the corrected annotation.

[0090] FIG. 7A shows the measured and initially annotated local activation times 332 (t i -circular) and the corresponding estimated local activation time 334 (

[0091]

number

[0092]

number

[0093] The cost function-based model of DOA is based on the assumption that there are at least 10 local atrial activations and that the time t of the local atrial activation of the i-th electrode, i=1,...,m, 10≦m≦N, is the time of local atrial activation of the i-th electrode. i where N is the number of active electrodes in the catheter, e.g., N=20 for the PentaRay® catheter. If a single EP wave is assumed to start at any point in 3D space and travel towards the catheter with constant conduction velocity (CV), then a cost function J(θ) can be defined for the “total cost” of the model:

[0094]

number

[0095] In Equation 4,

[0096]

number

[0097]

number

[0098]

number

[0099] The illustration at the top of FIG. 7B shows the resulting estimated focal activation. In FIG. 7B, a color scale (same as scale 48 in FIG. 4A and FIG. 4B) encodes the relative arrival times by color-coding each of the illustrated 20 electrodes 16 of the catheter 14. FIG. 7B further illustrates the tracked positions of the electrodes 16 (on the roughly demarcated arm 15 of the Pentaray catheter 14) in two separate examples where the electrodes captured EP signals. The position of each electrode 16 in 3D space is tracked, for example, using the ACL tracking technique described above. The XYZ axes (Z not shown) belong to a fixed reference axis system, such as the position tracking system 20 that applies the ACL method.

[0100] Finally, the DOA 55 and distance at which the set of signals 330 were emitted, derived from the cost function model, are also shown.

[0101] Figure 7C (isochronous map) shows that the analyzed EP wave propagates from the focal source location 336 within a circular line 344 that represents the arrival time in milliseconds according to the color bar on the right side of Figure 7C. The circles 350 represent the electrode locations, and the numbers within the circles represent the individual (i.e., per electrode) estimation errors 550 in milliseconds derived from the cost function in Figure 7A.

[0102] LAT Improvements In some exemplary embodiments of the present invention, the processor adjusts the first annotated time of the EP signal by selecting a first annotation in the EP signal, determining a corrected annotation that corresponds to the first annotation based on the estimated DOA and distance, and adjusting the timing of the EP signal upon determining that the corrected annotation meets predefined conditions.

[0103] The estimation error 550 is calculated by (a) minimizing a cost function set of location and conduction velocity, θ=(x 0 ,y 0 ,z 0 ,t 0 , v) to calculate the collision time

[0104]

number

[0105]

number

[0106] In one embodiment, the LAT value t i To improve the LAT estimation, if one of the predefined conditions 1 to 3 is satisfied,

[0107]

number

[0108] 1.

[0109]

number

[0110] 2.

[0111]

number

[0112] 3.t i and

[0113]

number

[0114] This discussion is followed by another subject which describes a simplified implementation of the cost function.

[0115] In some exemplary embodiments, a cost function in 2D space may be applied. In the 2D model, the catheter is projected onto the surface, which is done by taking the two eigenvectors with the highest eigenvalues. If the energy conserved by the two eigenvectors is more than the 95% that the model assumes, then the projection from the 3D space onto the surface is valid and the set of equations becomes the simpler θ=(x 0 ,y 0 ,t 0 ,v).

[0116] In some exemplary embodiments, an alternative DOA estimation process is used by the algorithm, which involves estimating the DOA using a weighted cost function, and is described in Figures 8A and 8B.

[0117] 8A and 8B are, respectively, a plot showing a graph of a unipolar EP signal 440 captured by the system of FIG. 1 and an isochronal map showing the respective estimated errors in the extracted EP values, in accordance with an exemplary embodiment of the present invention. The key concept behind the weighted cost function DOA model described below is that "sharp" excitations are "more reliable" than shallow excitations, and the level of sharpness is determined by the t i It is defined based on the dv / dt of a unipolar signal at each t i weight w from 0 to 1 based on its dv / dt i In FIG. 8A, the numbers near each circle represent the weight of the gradient.

[0118] Also note in Figure 8A that some EP signals contain the earliest S-wave pattern, such as signal 444 (both 448) sensed by electrodes E19 and E20. Such a negative slope pattern without an initial rise in signal amplitude as the EP wave approaches the electrode is indicative of an abnormal focal EP wave propagating away from the electrode. This condition indicates that catheter 14 is "on target" and that some of the electrodes are near the focal source of the arrhythmia (e.g., distance 60 is less than the length of arm 15).

[0119] As FIG. 8B shows, the estimated location 446 of the focal source derived using the weighted cost function is at least partially surrounded by the measured positions of the electrodes 16.

[0120] The necessary rewrite in the set of equations for the 2D cost function model (i.e., Eq. 5 without the z-dependence) is as follows:

[0121]

number

[0122] In some exemplary embodiments, if the relative time estimation error is greater than a given threshold, an iterative DOA estimation process is applied, which is described in Figures 9A and 9B, 10A and 10B, and 11A and 11B.

[0123] 9A and 9B are plots showing graphs of a unipolar EP signal with an estimation error 550 higher than a given threshold, and the respective initially estimated positions 560 of the focal source in XY space, according to an embodiment of the present invention. Furthermore, some EP values ​​are found within the double potential LAT (such as the corrected annotation value 333 of the EP signal of electrode 15 in FIG. 9A). Also, some EP values ​​(both measure estimates) are found between the start and end of the unipolar negative deflection (such as within the negative deflection 555 of the EP signal of electrode 16 in FIG. 9A).

[0124] As can be seen in FIG. 9B, location 560 is very close to the location of the distal tip of the catheter, but in light of the above observations, this location is likely incorrect.

[0125] In an exemplary embodiment, if the average estimation error is above a given threshold (e.g., 7 ms), the processor 22 performs an iterative calculation to estimate the DOA. The average estimation error in FIG. 9A is 12.4 ms. In each iteration, the local excitation time with the highest estimation error is eliminated from the DOA estimation. This process is repeated as long as there are more than 10 valid local excitation time values.

[0126] FIG. 10 is a graph showing the estimated location of the focal source 560 of FIG. 9B at nine iterations of the iterative DOA model according to an exemplary embodiment of the present invention. At the zeroth and first iterations, the focal source location 560 is approximately aligned with the location of the catheter's distal tip 18, however, from the second to the ninth iterations, the location of the focal activation shifts and is eventually located near electrode "1" (570). In FIG. 10, the solid circles represent valid electrodes for DOA estimation and the ring-shaped circles represent invalid electrodes. The tagging "iteration x" next to the ring electrodes informs that the particular electrode was excluded from the DOA estimation at iteration x. The percentage of invalid segments is a good measure of the "complexity" of AF in this subject.

[0127] Between the zeroth and ninth iterations, the maximum estimation error drops from approximately 25 ms to less than 5 ms. Conduction velocity CV, which also serves as the estimate of cost in Equation 4, drops from over 100 mm / ms to a minimum of 0.5 mm / ms.

[0128] The disclosed recursive model is used when dealing with acquisitions that have noisy waveforms or when multiple waves propagate towards the catheter.

[0129] First and second statistical testing methods A typical acquisition has a duration of 100-200 ms. A typical recording has a unipolar signal of 30 seconds and therefore contains approximately 120-200 acquisitions. All valid DOA estimates from approximately 120-200 acquisitions are stored until all acquisitions have been processed, and then statistical methods are applied to the corpus of valid DOA estimates.

[0130] First statistical method As described above and in this section, the processor 22 histograms the aggregated DOA and distance values ​​and statistically analyzes the histogram. In one exemplary embodiment, the processor is configured to derive an estimated position from the histogram by fitting a curve to the histogram and finding the maximum of the curve as a function of the estimated DOA and distance.

[0131] 11A and 11B are histograms 70A and 70B of direction of arrival (DOA) from a focal source and histograms 72A and 72B of distance, respectively, in accordance with an embodiment of the present invention. As shown, the DOA distribution consists of the number of acquisitions per DOA value, and the distance distribution consists of the number of acquisitions per distance value. Typically, processor 22 compiles (i.e., aggregates) histograms 70A and 70B and histograms 72A and 72B from a number of acquisitions ranging from tens to hundreds that have passed through a pre-filtering stage, as shown in FIG. 2 and analyzed in FIG. 3.

[0132] Using statistical tests, processor 22 first checks whether the DOA distribution produces consistent DOA values, as shown for example in the DOA histograms of Figures 11A and 11B. Examples of consistency testing tools include, but are not limited to, the use of stationarity estimates and confidence intervals.

[0133] If the DOA is found to be inconsistent, for example by a distribution showing two or more persistently different DOA values, the processor 22 terminates the disclosed focal source identification process. In one embodiment, the processor 22 presents a notification to the user that the process did not identify a pro-arrhythmic focal source.

[0134] If processor 22 derives consistent DOA values ​​as a function of DOA from the capture content distribution, processor 22 best estimates the DOA and s-distance from the distribution to the focal source of interest. Processor 22 then uses the best estimated DOA and distance to identify the location of the pro-arrhythmic focal source within heart 12 that generated the received EP signal.

[0135] As can be seen in Figure 11A, the most prevalent DOA values ​​of the clinical cases analyzed by histogram 70A of Figure 11A are near the DOA of 0.85π. The respective most common distance shown by histogram 72A is approximately 300 mm. Thus, processor 22 can identify the presence of a focal arrhythmia location for that patient at an angle of 0.85π with respect to the X-axis, at a distance of approximately 300 mm from the location of distal tip 18.

[0136] Figure 11B shows that the most prevalent DOA values ​​of the clinical cases analyzed by histogram 70B of Figure 11B are near a DOAπ of 0.5. The respective most common distance shown by histogram 72B is approximately 300 mm. Thus, processor 22 can identify the presence of a focal arrhythmia location for that patient at an angle of 0.5π with respect to the X-axis, at a distance of approximately 300 mm from the location of distal tip 18.

[0137] Second statistical method 12 is a plot showing DOA clusters analyzed by a k-means clustering model, according to an embodiment of the present invention. The K-means clustering model was applied by processor 22 to a set of DOA values ​​that had passed pre-filtering and were aggregated in XY space.

[0138] In Fig. 12, the circles represent DOA estimates from LAT value captures among many captures included in a single recording session. In the recording results shown, there are two clusters of DOAs that "explain" the data. The first cluster 80 (red circle at (-3.7mm, -0.2mm)) contains 80.5% of the DOAs, and the second cluster 88 contains 19.5% of the DOAs in the recording.

[0139] If the estimated location of one of the dominant clusters (more than 10% of DOA segments) can be projected onto the anatomical structure, i.e., the distance from the estimated location, such as estimated location 84, to the anatomical structure is less than a given value, e.g., 6 mm (configurable), then a focal source is identified. As seen in FIG. 12, the distance from the k-means clustering estimated location 84 to the anatomical location 90, given by the length of the arrow 85, is about 3 mm, well below the upper limit of 6 mm. Thus, location 90 has been confirmed as a focal source by the disclosed technique.

[0140] A focal source may also be confirmed if at least 10 signs of the earliest S-wave pattern (configurable) are found in electrodes located within a 6 mm radius from the focal point. It is important to note that focal detection-based DOA may appear at anatomical locations without placing the catheter within the region of focal activation, so the confirmation process is optional.

[0141] Although the embodiments described herein are primarily directed to cardiac applications, the methods and systems described herein can also be used in other applications, such as neurology, etc. The disclosed methods may also be applied to any data set with spatiotemporal "cues" of focal excitation, and a processor needs to find this focal excitation, for example for focal estimation in epilepsy patients using EEG / MEG.

[0142] It will therefore be understood that the above-described embodiments are cited by way of example, and that the present invention is not limited to what has been particularly shown and described above. Rather, the scope of the present invention includes both combinations of the various features described above and subcombinations thereof, as well as variations and modifications thereof that would occur to those skilled in the art upon reading the above description and that are not disclosed in the prior art. Documents incorporated by reference into this patent application are to be considered as part of this application, except that if any term is defined in these incorporated documents in a manner that is inconsistent with the definition expressly or impliedly given herein, then only the definition in this specification shall be considered.

[0143] [Embodiment] (1) A method for automatically identifying the location of focal arrhythmogenic activity, comprising: receiving a collection of capture content via a plurality of electrodes within the heart, each capture content including a set of electrophysiological (EP) signals measured by the electrodes; For each of the acquisitions, estimating a respective direction of arrival (DOA) and a respective distance relative to the electrodes from which the set of EP signals originated; aggregating the acquisition content to form a statistical distribution of the acquisition content as a function of estimated DOA and distance; using a statistical test to check whether the statistical distribution of the captured content is consistent according to a predefined consistency criterion; if the statistical distribution of the captured content is found to be consistent, deriving from the statistical distribution an estimated location within the heart of a pro-arrhythmic focal source that produced the received EP signal; and overlaying the estimated location of the focal source on an anatomical map of at least a portion of the heart. (2) The method of embodiment 1, wherein estimating the DOA and distance for a given capture content includes extracting respective sets of relative arrival times from the set of EP signals in the given capture content and estimating the DOA and distance using the extracted relative arrival times. (3) The method of embodiment 2, wherein aggregating the captured content includes pre-filtering the captured content according to the respective sets of relative arrival times extracted from each captured content, and including only the pre-filtered captured content in the statistical distribution of the captured content. (4) pre-filtering the acquisition content according to the extracted set of relative arrival times, using the estimated DOA and range to calculate, for each acquisition, a modeled set of relative arrival times resulting from EP waves emanating from a focal source at the estimated DOA and range; and determining, for each capture, a degree of similarity between the extracted set and the modeled set of relative arrival times by applying a predefined geometric test. (5) The method of claim 4, wherein estimating the similarity includes calculating a cosine-similarity geometrical test between the two sets.

[0144] (6) The method of embodiment 4, wherein estimating the similarity includes calculating an estimation error for each relative arrival time and comparing the estimation error with a given threshold. (7) The method of embodiment 4, comprising using the modeled set of relative arrival times to adjust annotation time values ​​for the entirety of the EP signal where the voltage-time slope of the EP signal is less than a predefined slope. (8) The method of embodiment 4, wherein pre-filtering the acquisition content includes discarding one or more acquisition contents determined to have a dissimilar set of arrival times. (9) The method of embodiment 1, wherein deriving the estimated position includes fitting a curve to the statistical distribution and finding a maximum of the curve as a function of the estimated DOA and distance. (10) The method of embodiment 1, wherein estimating the DOA and the distance includes minimizing a cost function.

[0145] (11) The method of embodiment 10, wherein minimizing the cost function comprises minimizing a weighted cost function. (12) The method of embodiment 10, wherein minimizing the cost function includes iteratively minimizing the cost function by eliminating, at each iteration, the EP signal value having the largest estimation error. (13) The method of embodiment 1, wherein deriving the estimated location includes applying a k-means analysis to the statistical distribution, projecting the estimated location onto an anatomical structure, and selecting locations having a projection distance less than a given value. (14) A system for automatically identifying the location of focal proarrhythmic effects, comprising: an interface configured to receive a collection of captured content captured by a plurality of electrodes within the heart, each captured content including a set of electrophysiological (EP) signals; 1. A processor comprising: For each of the acquisitions, estimate a respective direction of arrival (DOA) and a respective distance relative to the electrodes from which the set of EP signals originated; aggregating the acquisition content to form a statistical distribution of the acquisition content as a function of the estimated DOA and distance; using a statistical test to check whether the statistical distribution of the captured content is consistent according to a predefined consistency criterion; if the statistical distribution of the captured content is found to be consistent, deriving from the statistical distribution an estimated location within the heart of a pro-arrhythmic focal source that produced the received EP signal; and a processor configured to overlay the estimated location of the focal source on an anatomical map of at least a portion of the heart. (15) The system of embodiment 14, wherein for a given acquisition content, the processor is configured to estimate the DOA and distance by extracting a respective set of relative arrival times from the set of EP signals in the given acquisition content and estimating the DOA and distance using the extracted relative arrival times.

[0146] (16) The system of embodiment 15, wherein when aggregating the capture content, the processor is configured to pre-filter the capture content according to the respective set of relative arrival times extracted from each capture content, and to include only the pre-filtered capture content in the statistical distribution of the capture content. (17) The processor, using the estimated DOA and range to calculate, for each acquisition, a modeled set of relative arrival times resulting from EP waves emanating from a focal source at the estimated DOA and range; determining, for each capture, how similar the extracted set of relative arrival times and the modeled set are by applying a predefined geometric test; 17. The system of claim 16, further comprising: (18) The system of embodiment 17, wherein the processor is configured to estimate how similar the extracted set of relative arrival times and the modeled set are by calculating a geometric test of cosine similarity between the two sets. (19) The system of embodiment 17, wherein the processor is configured to estimate a similarity between the extracted set and the modeled set of relative arrival times by calculating an estimation error for each relative arrival time and comparing the estimation error to a given threshold. (20) The system of embodiment 17, wherein the processor is further configured to use the modeled set of relative arrival times to adjust annotation time values ​​for the entirety of the EP signal where the voltage-time slope of the EP signal is less than a predefined slope.

[0147] (21) The system of embodiment 20, wherein when pre-filtering the captured content, the processor is configured to discard one or more captured contents determined to have a dissimilar set of arrival times. (22) The system of embodiment 14, wherein the processor is configured to derive the estimated location by fitting a curve to the statistical distribution and finding a maximum of the curve as a function of estimated DOA and distance. (23) The system of embodiment 14, wherein the processor is configured to estimate the DOA and the distance by minimizing a cost function. (24) The system of embodiment 23, wherein the processor is configured to minimize the cost function by minimizing a weighted cost function. (25) The system of embodiment 23, wherein the processor is configured to iteratively minimize the cost function by eliminating, at each iteration, the EP signal value having the largest estimation error.

[0148] (26) The system of embodiment 14, wherein the processor is configured to derive the estimated location by applying a k-means analysis to the statistical distribution, projecting the estimated location onto an anatomical structure, and selecting locations having a projection distance less than a given value.

Claims

1. 1. A method of operation of a system for automatically identifying a location of focal pro-arrhythmic effects based on a collection of captured content, the collection including a set of electrophysiological signals, EP signals, captured and received by a plurality of electrodes in a heart, the method comprising: the processor estimating, for each of the acquisitions, a respective direction of arrival, DOA, and respective distance relative to the electrodes from which the set of EP signals was emitted; aggregating the acquired content to form a statistical distribution of the acquired content as a function of estimated DOA and distance; the processor checking whether the statistical distribution of the captured content is consistent according to a predefined consistency criterion using a statistical test; if the processor finds the statistical distribution of the captured content to be consistent, deriving from the statistical distribution an estimated location within the heart of a pro-arrhythmic focal source that produced the received EP signal; and the processor overlaying the estimated location of the focal source on an anatomical map of at least a portion of the heart.

2. A method of operating the system described in claim 1, wherein estimating the DOA and distance for the captured content includes extracting respective sets of relative arrival times from a set of EP signals in the captured content and estimating the DOA and distance using the extracted relative arrival times.

3. 3. A method of operating a system as described in claim 2, wherein aggregating the captured content includes pre-filtering the captured content according to the respective set of relative arrival times extracted from each captured content, and including only the pre-filtered captured content in the statistical distribution of the captured content.

4. Pre-filtering the acquisition content according to the extracted set of relative arrival times includes: using the estimated DOA and range to calculate, for each acquisition, a modeled set of relative arrival times resulting from the EP signal emanating from the focal source at the estimated DOA and range; and for each capture, determining a similarity between the extracted set and the modeled set of relative arrival times by applying a predefined geometric test.

5. The method of claim 4 , wherein determining the similarity comprises calculating a geometric test of cosine similarity between two of the sets.

6. A method according to claim 4 , wherein determining the similarity comprises calculating an estimation error for each relative arrival time and comparing the estimation error to a given threshold.

7. The method of claim 4 , wherein pre-filtering the acquisitions comprises discarding one or more acquisitions determined to have a dissimilar set of arrival times.

8. 2. The method of claim 1, wherein the statistical distribution is a histogram, and deriving the estimated location comprises fitting a fitted curve to the histogram and finding a maximum of the fitted curve as a function of estimated DOA and distance.

9. The method of claim 1 , wherein estimating the DOA and the distance comprises minimizing a cost function.

10. The method of claim 9 , wherein minimizing the cost function comprises minimizing a weighted cost function.

11. 10. The method of claim 9, wherein minimizing the cost function includes using a gradient descent estimation procedure and iteratively minimizing the cost function by eliminating EP signal values ​​having a maximum estimation error at each iteration in the gradient descent estimation procedure.

12. 2. The method of claim 1, wherein deriving the estimated location comprises applying a k-means analysis to the statistical distribution, projecting the estimated location onto an anatomical structure, and selecting locations having a projection distance less than a given value.

13. A system for automatically identifying a location of focal proarrhythmic effects based on a collection of captured content, including a set of electrophysiological signals (EP signals) captured and received by multiple electrodes within the heart, comprising: an interface configured to receive the collection of captured content; 1. A processor comprising: For each of the acquisitions, estimate a respective direction of arrival (DOA) and a respective distance relative to the electrodes from which the set of EP signals originated; aggregating the captured content to form a statistical distribution of the captured content as a function of the estimated DOA and distance; using a statistical test to check whether the statistical distribution of the captured content is consistent according to a predefined consistency criterion; if the statistical distribution of the captured content is found to be consistent, deriving from the statistical distribution an estimated location within the heart of a pro-arrhythmic focal source that produced the received EP signal; and a processor configured to overlay the estimated location of the focal source on an anatomical map of at least a portion of the heart.

14. The system of claim 13, wherein for the captured content, the processor is configured to estimate the DOA and distance by extracting respective sets of relative arrival times from a set of EP signals in the captured content and estimating the DOA and distance using the extracted relative arrival times.

15. 15. The system of claim 14, wherein when aggregating the captured content, the processor is configured to pre-filter the captured content according to the respective set of relative arrival times extracted from each captured content, and to include only the pre-filtered captured content in the statistical distribution of the captured content.

16. The processor, using the estimated DOA and range to calculate, for each acquisition, a modeled set of relative arrival times resulting from the EP signal emanating from the focal source at the estimated DOA and range; determining, for each capture, how similar the extracted set of relative arrival times and the modeled set are by applying a predefined geometric test; 16. The system of claim 15, configured to pre-filter the acquisition content according to the extracted set of relative arrival times by performing:

17. 17. The system of claim 16, wherein the processor is configured to determine how similar the extracted set of relative arrival times and the modeled set are by computing a geometric test of cosine similarity between the two sets.

18. 17. The system of claim 16, wherein the processor is configured to determine a similarity between the extracted set and the modeled set of relative arrival times by calculating an estimation error for each relative arrival time and comparing the estimation error to a given threshold.

19. 20. The system of claim 16, wherein in pre-filtering the acquisitions, the processor is configured to discard one or more acquisitions determined to have a dissimilar set of arrival times.

20. 14. The system of claim 13, wherein the statistical distribution is a histogram, and the processor is configured to derive the estimated location by fitting a fitted curve to the histogram and finding a maximum of the fitted curve as a function of estimated DOA and distance.

21. The system of claim 13 , wherein the processor is configured to estimate the DOA and the distance by minimizing a cost function.

22. 22. The system of claim 21, wherein the processor is configured to minimize the cost function by minimizing a weighted cost function.

23. 22. The system of claim 21, wherein the processor is configured to iteratively minimize the cost function using a gradient descent estimation procedure and configured to eliminate, at each iteration in the gradient descent estimation procedure, an EP signal value having a maximum estimation error.

24. 14. The system of claim 13, wherein the processor is configured to derive the estimated location by applying a k-means analysis to the statistical distribution, projecting the estimated location onto an anatomical structure, and selecting locations having a projection distance less than a given value.

Images