Identifying activation in atrial fibrillation electrograms

Abstract

The present invention is entitled "Identifying Activation in an Atrial Fibrillation Electrogram." The present invention provides a system comprising: an electrical interface and a processor, the processor being configured to: receive, via the electrical interface, a signal sensed by at least one electrode in contact with cardiac tissue of a subject, the signal spanning consecutive time periods each having a length T1 and comprising a plurality of signal points; calculate respective thresholds for the time periods; select a set of points comprising, for each of the time periods, the signal point of maximum magnitude within the time period, provided that the maximum magnitude is greater than the threshold for the time period; remove from the set one of any pair of selected points that are within an interval T2 of each other, T2 being less than T1; and generate an output after the removal, the output being based on the remaining points in the set that correspond to respective electrical activations of the tissue.

Description

Technical Field

[0001] The present invention relates generally to the field of electrophysiology and, in particular, to cardiac arrhythmias such as atrial fibrillation. Background Art

[0002] Atrial fibrillation is an abnormal heart rhythm characterized by rapid and irregular beating of the atria. Typically, the shape of an atrial fibrillation electrocardiogram varies at different anatomical sites and / or over the duration of the recording. Therefore, conventional methods of annotating electrogram signals to determine local activation time (LAT) are not suitable for this type of arrhythmia.

[0003] U.S. Patent Application Publication 2015 / 0208942, the disclosure of which is incorporated herein by reference, describes a cardiac catheterization procedure performed by inserting a probe having electrodes into the heart of a living subject, recording a bipolar electrogram and a unipolar electrogram from one of the electrodes at a location in the heart, and defining a window of interest where the rate of change of the potential of the bipolar electrogram exceeds a predetermined value. An annotation is established in the unipolar electrogram, wherein the annotation represents the maximum rate of change of the potential of the unipolar electrogram within the window of interest. A quality value is assigned to the annotation, and a 3-dimensional map of a portion of the heart is generated that includes the annotation and its quality value. Summary of the Invention

[0004] We have developed technical solutions to specific real-world technical problems associated with atrial fibrillation, such as those described above in the background section. Our technical solutions, which cannot be performed by humans, provide consistent annotation results (e.g., in real time) to physicians performing electrophysiology (EP) procedures, such as when the procedures are applied to bipolar signals read from a mapping catheter (which may be positioned, for example, within a living heart) during atrial fibrillation. The technical features of the present invention, in the form of the embodiments described herein, are designed to improve the functionality of EP systems and, thereby, the health of subjects undergoing EP procedures.

[0005] We note that our technical solution cannot be directly implemented on a general-purpose computer because our solution requires the implementation of various components designed specifically for EP procedures, such as an EP catheter for mapping electrogram signals and sensors on the EP catheter that allow the system to determine the location on the living, beating heart from which the electrogram signals originate. For example, as described and illustrated herein, our technical solution addresses the problem of mapping and measuring atrial fibrillation (a real-life phenomenon) and annotating intracardiac signals originating therefrom. Thus, our technical solution advances EP technology to achieve the inventive concepts and technical features necessary to map and annotate intracardiac signals originating from atrial fibrillation.

[0006] Thus, according to some embodiments of the present invention, there is provided a system comprising an electrical interface and a processor. The processor is configured to receive, via the electrical interface, a signal sensed by at least one electrode in contact with cardiac tissue of a subject, the signal spanning continuous time periods each having a length T1 and comprising a plurality of signal points. The processor is further configured to calculate a corresponding threshold for the time period, and to select a set of points that, for each time period, includes the signal point of the maximum magnitude within the time period, provided that the maximum magnitude is greater than the threshold for the time period. The processor is further configured to remove from the set one of any pair of selected points that are within an interval T2 of each other, T2 being less than T1, and to generate an output after the removal that is based on the remaining points in the set that correspond to the corresponding electrical activation of the tissue.

[0007] In some embodiments, the processor is configured to receive a signal when the subject's heart is experiencing atrial fibrillation.

[0008] In some embodiments, T1 is between about 80 ms and about 120 ms.

[0009] In some embodiments, T2 is between about 80 ms and about 120 ms.

[0010] In some embodiments,

[0011] The interval is the first interval,

[0012] The time period includes a first time period, a second time period immediately after the first time period, and a third time period immediately after the second time period,

[0013] The point set includes a first time period signal point in the first time period and a second time period signal point in the second time period,

[0014] The processor is further configured to:

[0015] identifying a first global extremum of the signal within a second interval T3 from the first period signal point, and

[0016] identifying a second global extremum of the signal within a second interval from the second period signal point, and

[0017] The processor is configured to calculate the threshold for the third time period as the product of: (a) a coefficient less than one, and (b) a minimum of (i) a first global extreme value of the first global extreme value and (ii) a second global extreme value of the second global extreme value.

[0018] In some embodiments, the coefficient is between about 0.1 and about 0.5.

[0019] In some embodiments, T3 is from about 250 ms to about 350 ms.

[0020] In some embodiments, the processor is configured to remove, from any pair of selected points that are within T2 of each other and not within T2 of any other selected point, one point in the pair that has a smaller magnitude than the other point in the pair.

[0021] In some embodiments, the processor is configured to generate an output by noting signals to indicate that the remaining points in the set correspond to respective electrical activations.

[0022] In some embodiments, the processor is configured to generate output by:

[0023] calculating the length of the electrical activation period from the remaining points in the set by calculating at least one time interval between two of the remaining points in the set, and

[0024] Displays the length of the period.

[0025] According to some embodiments of the present invention, there is further provided a method comprising receiving a signal provided by at least one electrode in contact with cardiac tissue of a subject, the signal spanning consecutive time periods each having a length T1 and comprising a plurality of signal points. The method further comprises calculating a corresponding threshold for the time period, and selecting a set of points that, for each time period, comprises the signal point of maximum magnitude within the time period, provided that the maximum magnitude is greater than the threshold for the time period. The method further comprises removing from the set one of any pair of selected points that are within an interval T2 of each other, T2 being less than T1, and generating an output based on the remaining points in the set that correspond to corresponding electrical activations of the tissue.

[0026] It should be noted that a person experienced in interpreting intracardiac signals may not be able to provide the technical features described herein, at least due to the inherent difficulties in interpreting atrial fibrillation signals and / or due to the fact that the solutions described herein can be provided in near real time. For example, the signal may be annotated within 10 seconds, five seconds, or one second of receipt.

[0027] The present disclosure will be more fully understood through the following detailed description of embodiments of the present invention in conjunction with the accompanying drawings, in which: BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a schematic diagram of a system for annotating electrogram signals according to some embodiments of the present invention;

[0029] Figure 2 is a schematic diagram of identification of activation points in an electrogram signal according to some embodiments of the present invention; and

[0030] Figure 3is a flow chart of a method for annotating electrogram signals according to some embodiments of the present invention. DETAILED DESCRIPTION

[0031] Overview

[0032] In some cases, a physician may wish to understand the manner in which a wavefront of electrical activation propagates through the cardiac tissue of a subject during each cardiac cycle. In such cases, multiple electrodes may be used to sense or record electrogram signals at different corresponding locations on the cardiac tissue. For each signal, the physician may attempt to identify any points of electrical activation, i.e., any points in the signal that correspond to instances of the wavefront passing through the location where the signal was acquired. Unfortunately, however, in cases of atrial fibrillation or other arrhythmias, it may be difficult even for experienced EP physicians to identify any points of activation due to the irregularity and / or variability of the electrogram signal. For example, a conventional electrogram includes regularly spaced spikes that clearly indicate activation, whereas an irregular electrogram may exhibit a variety of different forms and may include many small peaks in close succession, with no peak clearly indicating activation.

[0033] To address this challenge, embodiments of the present invention provide techniques for automatically identifying activation points even in electrograms that exhibit atrial fibrillation or other arrhythmic behavior. According to these techniques, a processor divides the signal into continuous time segments, each having a predefined length T1, which is typically slightly greater than the minimum expected cycle length (i.e., the minimum expected length of an electrical activation cycle). In each time segment, the processor identifies a positive or negative peak whose magnitude is greater than the magnitude of any other point in the time segment and is also greater than a dynamically calculated threshold (assuming such a peak exists). These peaks are referred to herein as "candidate activation points" because each peak can correspond to an activation.

[0034] Next, the processor iterates over the identified candidate activation points. For any pair of these points that are within a spacing T2 of each other (T2 is typically around the minimum expected cycle length), the processor discards one of the two points. The remaining candidate points are then assumed to be activation points corresponding to different corresponding activations.

[0035] Finally, the processor annotates signals to indicate that the remaining candidate points correspond to different corresponding activations and / or displays other outputs based on the points corresponding to the different corresponding activations. For example, the processor can calculate the cycle length based on the corresponding times of the points and then display the cycle length, for example by including the cycle length in the electroanatomical map. Alternatively or additionally, by processing multiple signals acquired from different tissue regions, the processor can calculate the activation sequence of the tissue, i.e., the sequence of activated regions during each cardiac cycle.

[0036] As used herein, the term "about" or "approximately" for any numerical value or range indicates a suitable dimensional tolerance that allows the part or collection of elements to achieve its intended purpose as described herein. More specifically, "about" or "approximately" can refer to a range of ±10% of the recited value, for example, "about 90%" can refer to a range of values ​​from 81% to 99%. Additionally, as used herein, the terms "patient," "host," "user," and "subject" refer to any human or animal subject and are not intended to limit systems or methods to human use, although use of the subject invention in human patients represents a preferred embodiment.

[0037] System Description

[0038] First reference Figure 1 , which is a schematic diagram of a system 21 for annotating an electrogram signal 22 according to some embodiments of the present invention.

[0039] Figure 1 An EP physician 27 is shown operating an EP catheter 29, the distal end 31 of which is positioned within the heart 23 of a subject 25. As the physician 27 moves the distal end 31 of the catheter 29 along the endocardial or epicardial surface of the heart, one or more electrodes at the distal end of the catheter, which are in contact with the heart tissue, sense electrogram signals 22 generated by the tissue. For example, such signals may be sensed when the heart 23 is experiencing atrial fibrillation or any other arrhythmia, such as atrial flutter or atrial tachycardia. (In some cases, the arrhythmia may have been induced by the physician prior to the procedure.)

[0040] Processor (PROC) 28 receives signal 22 via electrical interface 35 (such as a socket or port) and as described below with reference to Figure 2 In response to processing the signals, the processor 28 generates an output, which typically includes a visual output displayed on the display 26. For example, the processor 28 may annotate at least one signal 22 to show the activation points of the signal and then display the annotated signal on the display 26. When annotating the signal, the processor 28 may, for example, place a marker 24 at each activation point.

[0041] In general, the electrodes at the distal end 31 may be arranged in any suitable configuration, such as a circular, linear, or multi-spline configuration. Typically, each signal 22 is a bipolar signal in that the signal represents the voltage between a corresponding pair of electrodes at the distal end 31. Alternatively, however, at least one signal may be a unipolar signal in that the signal represents the voltage between one electrode and a reference electrode externally coupled to the subject.

[0042] Generally speaking, processor 28 can be implemented as a single processor or a group of collaborative networked or clustered processors. In some embodiments, as described herein, the functions of processor 28 can be implemented in hardware using, for example, only one or more application-specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs). In other embodiments, the functions of processor 28 are implemented at least in part in software. For example, in some embodiments, processor 28 is embodied as a programmed digital computing device, which includes at least a central processing unit (CPU) and a random access memory (RAM). Program code (including software programs and / or data) is loaded into RAM for CPU execution and processing. For example, program code and / or data can be downloaded to processor 28 in electronic form over a network. Alternatively or additionally, program code and / or data can be provided and / or stored on a non-transitory tangible medium, such as a magnetic, optical or electronic memory. After being provided to processor 28, such program code and / or data generates a machine or special-purpose computer configured to perform the tasks described herein.

[0043] Now refer to Figure 2 , which is a schematic diagram of the identification of activation points in an electrogram signal according to some embodiments of the present invention.

[0044] Typically, when processing each signal 22, the processor 28 divides the total period spanned by the signal into successively smaller time periods, each having a length T1 (e.g., 100 ms). (Thus, the first time period spanned by the signal is from 0 to T1, the second time period is from T1 to 2T1, and so on.) The processor 28 also calculates corresponding thresholds for the time periods, as further described below. (In other words, it can be said that the processor 28 calculates a single threshold that is updated for each time period.) The processor 28 also selects a set of candidate activation points that, for each time period, includes the point of maximum magnitude (or "peak") in that time period, provided that the maximum magnitude is greater than the threshold for that time period and is also greater than the threshold for the time period. Figure 2 , a predefined noise threshold (e.g., 0.05 mV) indicated by noise threshold line 36. (Typically, the set of candidate activation points does not include any points other than the maximum magnitude point.) Finally, as described further below, processor 28 removes from the set of candidate activation points one of any pair of candidate points that are within a spacing T2 of each other, where T2 is less than T1 (e.g., 80 ms). The remaining points in the set are then assumed to be activation points.

[0045] Typically, processor 28 processes time periods sequentially. For each time period, processor 28 calculates a threshold based on the global extremum of the signal in or near that time period. For example, for any three consecutive time periods, after selecting a corresponding candidate activation point from each of the first two time periods, the threshold T for the third time period can be calculated as T=C*min(GE1, GE2), where C is a coefficient less than one (e.g., approximately 0.3), GE1 is the global extremum of the signal within an interval T3 from the first candidate activation point, T3 is greater than T1 (e.g., approximately 250 ms), and GE2 is the global extremum of the signal within T3 from the second candidate activation point. Processor 28 then selects another candidate activation point from the third time period, provided that the magnitude of the point is greater than both threshold T and a predefined noise threshold.

[0046] For example, in processing Figure 2 The example signals shown are:

[0047] (i) The processor 28 identifies a global extremum 30 of the signal within T3 from the start time of the signal (i.e., 0). In other words, the processor 28 identifies the positive or negative peak (i.e., maximum or minimum) of the signal within the interval 0-T3 that has the greatest magnitude relative to all other peaks within the interval.

[0048] (ii) The processor 28 calculates a first threshold by multiplying the magnitude of the extreme value 30 by a coefficient C.

[0049] (iii) Processor 28 identifies a maximum magnitude point within the first time period (0-T1) whose magnitude is greater than the first threshold and also greater than a predefined noise threshold. In this case, since extreme value 30 is reached at time T0a within the first time period (0-T1), processor 28 identifies extreme value 30 as a candidate activation point for the first time period.

[0050] (iv) The processor 28 identifies a global extremum 32 of the signal within T3 from T0a (ie between T0a and T0a+T3). The global extremum 32 is obtained at time T0c.

[0051] (v) Processor 28 calculates a second threshold by multiplying the magnitude of extreme value 32 by coefficient C.

[0052] (vi) Processor 28 calculates a threshold for the second time period (T1-2T1) by taking the minimum of the first threshold and the second threshold. Processor 28 then selects point 34, the highest magnitude point obtained at time T0b in the second time period, as a candidate activation point, given that the magnitude of point 34 is greater than both the threshold for the second time period and the predefined noise threshold.

[0053] Processor 28 then repeats steps (iv)-(vi) for a third time period (2T1-3T1). Specifically, processor 28 identifies a global extreme value of the signal within T3 from T0b (i.e., between T0b and T0b+T3), calculates a third threshold by multiplying the magnitude of the extreme value by C, sets the threshold for the third time period to the minimum of the second and third thresholds, and then selects the highest magnitude point in the third time period whose magnitude is greater than the threshold for the third time period and the predefined noise threshold. This point is extreme value 32.

[0054] The processor 28 then repeats steps (iv)-(vi) for the fourth time period and thereby selects another candidate activation point 38 obtained at time T0d. The remaining time periods are processed similarly.

[0055] As described above, after selecting a set of candidate activation points, the processor 28 prunes the set by removing one of any pair of candidate points that are within a spacing T2 of each other. For example, for any pair of candidate activation points that are within T2 of each other and are not within T2 of any other candidate point, the processor 28 may remove one point in the pair that has a smaller magnitude than the other point in the pair. If a particular candidate point is within T2 of both a previous candidate point and a subsequent candidate point, the processor 28 may remove the particular candidate point even if the magnitude of the particular candidate point is greater than the corresponding magnitudes of the two adjacent candidate points. Thus, for example, in Figure 2 In the case shown, if T0c is within T2 of both T0b and T0d, the extreme value 32 can be removed from the set of candidate points. The candidate points left after pruning are Figure 2 These star marks and / or any other suitable marks (e.g., circles or arrows) can be provided to the EP physician on a graphical display as "annotations" to the intracardiac signals obtained via the EP catheter, as described above with reference to Figure 1 As stated.

[0056] Typically, T1 is slightly larger than the minimum expected cycle length; for example, T1 can be between 80 ms and 120 ms. If T1 is smaller than this value, processor 28 may select a relatively large number of erroneous candidate activation points, i.e., candidate points that do not actually correspond to activations. Conversely, if T1 is larger than this value, processor 28 may miss some activations.

[0057] Typically, T2 is approximately equal to the minimum expected cycle length; for example, T2 may be between 80 ms and 120 ms. Such a T2 value helps remove false candidate activation points without a significant risk of removing true activation points.

[0058] Typically, T3 is approximately equal to the maximum expected cycle length; for example, T3 may be between 250 ms and 350 ms. Such a T3 value facilitates calculation of a threshold for each time period that reflects the behavior of the signal around that time period.

[0059] Typically, a suitable value for the coefficient C can be obtained by experiment. The inventors have found that a value between 0.1 and 0.5 is generally effective, at least for bipolar signals.

[0060] After pruning the set, the processor 28 generates an output based on the remaining points in the set corresponding to the corresponding tissue activations. For example, as described above with reference to Figure 1 Said and marked with a star Figure 2 As further shown in , processor 28 may annotate the signal to indicate that the remaining points in the set correspond to respective tissue activations.

[0061] Alternatively or additionally, the processor 28 may calculate the cycle length based on the remaining points in the set by calculating at least one time interval between two remaining points in the set. For example, the processor 28 may calculate the time interval between two consecutive points, or the average or median of a plurality of such intervals. (For example, in Figure 2 In the case shown, the processor 28 may calculate the period length of T0b-T0a or (T0c-T0a) / 2. ) The processor 28 may then display the calculated period length on the display 26.

[0062] In some embodiments, processor 28 performs the above-described activation point selection technique for each of a plurality of signals sensed by different corresponding electrodes located at different corresponding regions of tissue. In such embodiments, processor 28 can calculate the cycle length for each tissue region as described above and then display these cycle lengths on the electroanatomical map of the region. For example, processor 28 can color regions having different corresponding cycle lengths with different corresponding colors according to a predefined color scale.

[0063] Alternatively or additionally, the processor 28 may identify an activation sequence for different regions based on the order in which the activation signals occur. For example, the processor 28 may receive (i) a signal such as Figure 2As shown, (i) a first signal sensed by a first electrode in contact with a first tissue region, and (ii) a second signal sensed by a second electrode in contact with a second tissue region. If, by processing the second signal as described above, the processor 28 identifies activations that occur after T0a, T0b, and T0d, respectively (e.g., less than 80 ms later), the processor 28 can determine that the second region is activated after the first region during each cycle. Conversely, if the processor 28 identifies activations that occur slightly before T0a, T0b, and T0d, respectively, the processor 28 can determine that the second region is activated before the first region during each cycle.

[0064] As an alternative or in addition to identifying an activation sequence, processor 28 can calculate the propagation time required for the activation wavefront to propagate through multiple tissue regions by calculating the interval between the first and last activations identified for a particular cardiac cycle, or calculating the average or median of this interval over multiple cardiac cycles. For example, given (i) a first activation occurs at T0a in a first region, (ii) a second activation occurs 15 ms after T0a in a second region, and (iii) a third activation occurs 30 ms after T0a in a third region, processor 28 can calculate a propagation time of 30 ms. Processor 28 can also compare the cycle length of one or more signals to the propagation time. If the fraction of the propagation time to the cycle length is greater than a predetermined threshold, processor 28 can generate an output indicating a possible rotational activation.

[0065] Now refer to Figure 3 , which is a flow chart of a method 40 for annotating electrogram signals according to some embodiments of the present invention. The method 40 may be performed by the processor 28 ( Figure 1 ) is performed by executing any suitable software program.

[0066] Method 40 begins with the initialization of two variables: a first threshold variable TH1 and a second threshold variable TH2. To perform this initialization, at a first global extremum identification step 42, processor 28 first identifies a global extremum of the signal within T3 from the start time of the signal. Next, at a TH1 calculation step 44, processor 28 calculates TH1 based on the identified global extremum. For example, as described above with reference to Figure 2 As described, the processor 28 may calculate TH1 by multiplying the identified global extremum by a predefined coefficient. Next, at a TH2 setting step 46, the processor 28 sets TH2 equal to TH1.

[0067] The processor 28 then establishes a set of candidate activation points by iteratively processing the time periods spanned by the signal. At the beginning of each iteration, at a time period selection step 48, the processor 28 selects the next time period of the signal. Then, at a first check step 50, the processor 28 checks whether the maximum magnitude (i.e., absolute value) of the signal within the selected time period is greater than both the predefined noise threshold and the minimum values ​​of TH1 and TH2. If so, at a point addition step 52, the processor 28 adds the maximum magnitude point to the set of candidate activation points. Then, at a TH2 setting step 46, the processor 28 sets the second threshold TH2 to be equal to the first threshold TH1. Then, at a second global extreme value identification step 56, the processor 28 identifies the global extreme value of the signal within T3 from the time of the maximum magnitude point. Then, at a TH1 calculation step 44, the processor 28 updates TH1 using the newly identified global extreme value.

[0068] After updating TH1, or if processor 28 does not identify a suitable candidate activation point at first check step 50, processor 28 checks whether any time segments remain at a second check step 60. If so, processor 28 returns to time segment selection step 48 and then processes the next time segment of the signal.

[0069] After processing the final time period, processor 28 removes the additional candidate activation points from the set at a remove step 62, as described above with reference to Figure 2 Finally, at a marking step 64, the processor 28 marks the remaining candidate activation points as corresponding to the corresponding activations. Figure 2 As mentioned, processor 28 may alternatively or additionally generate any other type of output that assumes that each of the remaining candidate activation points corresponds to a respective activation.

[0070] It will be understood by those skilled in the art that the present invention is not limited to what has been specifically shown and described above. Rather, the scope of the embodiments of the present invention includes both combinations and subcombinations of the various features described above, as well as variations and modifications not within the prior art that may occur to those skilled in the art upon reading the foregoing description. For example, where methods and steps are described above as being performed in a particular order, it is not intended that the methods and steps must be performed in the order described; rather, any order that allows the embodiment to function for its intended purpose is within the scope of the present invention.

[0071] Documents incorporated by reference into this patent application are considered to be an integral part of this application, except that if any term defined in these incorporated documents conflicts with a definition explicitly or implicitly given in this specification, only the definition in this specification shall prevail.

Claims

1. A system for identifying activation in an atrial fibrillation electrogram, comprising: Electrical interface; and a processor configured to: receiving, via the electrical interface, a signal sensed by at least one electrode in contact with cardiac tissue of a subject, the signal spanning consecutive time periods each having a length T1 and comprising a plurality of signal points, Calculate the corresponding threshold value for the time period, selecting a set of points, said set of points including, for each said time period, a signal point with a maximum magnitude within said time period, provided that said maximum magnitude is greater than said threshold value for said time period, remove from the set one of any pair of selected points that are within a distance T2 from each other, T2 being less than T1, and An output is generated based on the remaining points in the set corresponding to respective electrical activations of the tissue. 2 . The system of claim 1 , wherein the processor is configured to receive the signal when the subject's heart is experiencing atrial fibrillation. The system of claim 1 , wherein T1 is between about 80 ms and about 120 ms. The system of claim 1 , wherein T2 is between about 80 ms and about 120 ms.

5. The system according to claim 1, wherein the interval is a first interval, wherein the time period includes a first time period, a second time period immediately after the first time period, and a third time period immediately after the second time period, wherein the point set includes a first time period signal point in the first time period and a second time period signal point in the second time period, wherein the processor is further configured to: identifying a first global extreme value of the signal within a second interval T3 from the first time period signal point, and identifying a second global extremum of the signal within the second interval from the second period signal point, and wherein the processor is configured to calculate the threshold for the third time period as the product of: (a) a coefficient less than one, and (b) the minimum of (i) a first global extreme value of the first global extreme value and (ii) a second global extreme value of the second global extreme value. The system of claim 5 , wherein the coefficient is between about 0.1 and about 0.

5. The system of claim 5 , wherein T3 is about 250 ms to about 350 ms.

8. The system of claim 1 , wherein the processor is configured to remove, from any pair of selected points that are within T2 of each other and not within T2 of any other selected point, one point in the pair that has a smaller magnitude than the other point in the pair.

9. The system of claim 1 , wherein the processor is configured to generate the output by annotating the signal to indicate that the remaining points in the set correspond to the respective electrical activations.

10. The system of claim 1 , wherein the processor is configured to generate the output by: The length of the electrical activation cycle is calculated based on the remaining points in the set by calculating at least one time interval between two points in the remaining points in the set, and the length of the cycle is displayed.

Images