Arrhythmia localization using data from electrodes within the heart

Abstract

Various embodiments include a method of using information from medical electrodes positioned in a patient's heart to identify and display arrhythmia localization. The method may include obtaining electrical signals and location information from electrodes positioned within the patient's heart, generating a patient-specific 3D heart model, generating a patient-specific electrical conduction map identifying arrhythmia initiation site localization, displaying the 3D localization for physician use, and conducting a cardiac ablation procedure using the displayed localization to identify ablation locations. In some embodiments, the medical electrodes may be electrodes on electrophysiology catheters temporarily positioned with the heart. In some embodiments, the medical electrodes may be electrodes of an implanted cardiac defibrillator (ICD) permanently positioned in the heart.

Description

Attorney Docket No.: 35313-0011WOARRHYTHMIA LOCALIZATION USING DATA FROM ELECTRODES WITHIN THE HEART CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to and benefit from U. S. Provisional Application No: 63 / 777,021 filed March 25, 2025, and U. S Provisional Application No: US 63 / 836,713 filed July 1, 2025, each of which is hereby incorporated in its entirety by this reference.FIELD OF INVENTION

[0002] The present disclosure relates to cardiac arrhythmia localization techniques, and more particularly to a method for identifying and displaying ventricular arrhythmia initiation sites using data from electrodes positioned within the heart combined with three-dimensional heart modeling.BACKGROUND

[0003] Cardiac arrhythmias are abnormal heart rhythms that can affect the heart's ability to pump blood effectively. These irregular heartbeats occur when there are disruptions in the heart's electrical conduction system, which normally coordinates the contraction of the heart's chambers. Arrhythmias can range from benign to lifethreatening, depending on their type, duration, and underlying cause.

[0004] A number of electrophysiology (EP) procedures involve positioning catheters with electrodes positioned near the distal end for sensing electrical potentials within the heart. Multiple EP catheters (e.g., two, three, or four) may be positioned in various portions of the heart to obtain information on the timing, magnitude, and direction of conduction waves passing through heart tissues as the heart beats. In a typical procedure, one EP catheter may be positioned in the coronary sinus, a second electrode may be positioned in the right ventricle, and a third electrode may be positioned in the left ventricle.15448153-1 1Attorney Docket No.: 35313-0011WO

[0005] Implantable cardioverter-defibrillators (ICDs) are medical devices designed to monitor heart rhythms and deliver electrical shocks to restore normal heart function when dangerous arrhythmias are detected. These devices are typically implanted in patients at risk for sudden cardiac arrest due to ventricular arrhythmias. In addition to providing antitachycardia therapies when necessary, ICDs continuously monitor the heart's electrical activity and record data on detected arrhythmic events, providing valuable information for healthcare providers in managing patients with heart rhythm disorders.

[0006] Recent advancements in cardiac imaging and three-dimensional (3D) modeling techniques have enabled more detailed visualization of heart structures and electrical activity. These technologies allow for the creation of patient-specific heart models that can be used to analyze cardiac functions during cardiac procedures.. For example, visualization of heart structures and electrical procedures have enabled ablative procedures to treat various cardiac tachyarrhythmias (e.g., ventricular tachyarrhythmias (VT) that include monomorphic and polymorphic ventricular tachycardias and premature ventricular contractions (PVC)). Additionally, improvements in electrophysiology (EP) mapping systems have enhanced the ability to locate and characterize arrhythmia sources within the heart during diagnostic and therapeutic procedures. However, locating and characterizing arrhythmia sources within the heart during diagnostic and therapeutic procedures can prolong the procedures and increase the mortality rate.

[0007] Accordingly, there is a desire to locate the origin of arrhythmia prior to the cardiac procedures to effectively manage therapy and to guide ablation procedures. SUMMARY

[0008] This summary is provided to introduce concepts of various aspects in a simplified form that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claims.15448153-1 2Attorney Docket No.: 35313-0011WO

[0009] In various aspects, a method of using information from electrodes positioned within a patient's heart to identify and display arrhythmia localization may include determining locations of one or more electrodes positioned within the heart, generating a patient-specific three-dimensional (3D) heart model including locations of the one or more electrodes, generating a patient-specific electrical conduction map of the patient's heart during an arrhythmia based on the patient-specific 3D heart model and data from the one or more intracardiac electrodes recorded during arrhythmia events. The patient-specific electrical conduction map of the patient's heart may identify an initiation site of the arrhythmia on the 3D heart model to graphically display locations of arrhythmia activationt. This 3D localization of the initiation site of the arrhythmia may be displayed on monitor for use by a physician preparing for a cardiac electrophysiology procedure, and may be displayed on a monitor in the operating room or catheterization lab while the physician conducts a cardiac ablation procedure. In this way, the physician may use the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for conducting an ablation.

[0010] In some aspects, the electrodes positioned within a patient's heart may be electrodes on electrophysiology (EP) catheters that are temporarily positioned in the heart during an EP procedure. In such aspects, the method may include obtaining locations of one or more electrodes in the heart may be obtained from a catheter electrode tracking system, and correlating the electrode locations with the 3D heart model as part of generating the patient-specific electrical conduction map.

[0011] In some aspects, the method may include using information from an implanted cardiac defibrillator (ICD) in a patient's heart to identify and display arrhythmia localization and may include downloading from the ICD information recorded during detected arrhythmia events in the patient's heart. Such aspects may include downloading from the ICD information recorded during detected arrhythmia events in the patient's heart, and obtaining medical imaging that identifies locations of one or more ICD electrodes in the patient's heart.15448153-1 3Attorney Docket No.: 35313-0011WO

[0012] In some aspects, the method may include obtaining 3D imagery of electrocardiogram (ECG) electrodes on the patient's torso during an ECG recording procedure, merging the 3D image of the patient's torso with the 3D heart model, generating a patient-specific electrical conduction map of a patient's heart of an arrhythmia based on the patient-specific 3D heart model, the downloaded ICD recorded information regarding detected arrhythmia events, ECG data from the ECG recording procedure and electrogram recordings from intracardiac electrodes during an EP procedure.

[0013] In some aspects, the method may include conducting point-by-point contact electrophysiology recordings during the cardiac ablation procedure, and updating the patient-specific electrical conduction map to display an updated 3D localization of the initiation site of the arrhythmia.

[0014] In some aspects, generating a patient-specific 3D model of the heart including a 3D internal surface model may include using magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient to generate the patientspecific 3D heart model including locations of the one or more ICD electrodes.

[0015] In some aspects, ICD information recorded during detected arrhythmia events in the patient's heart may be combined with the patient-specific 3D heart model to identify isochrones at time intervals of heartbeats to determine directions of depolarization wavefronts to reveal an initiation site of the arrhythmia in a heartbeat.

[0016] In some aspects, the method may include displaying heart structures including one or more of the aorta, aortic arch, pulmonary veins or coronary vessels on the displayed 3D heart model. In some aspects, the method may include displaying heart scar tissue indicative of ischemic heart disease on the displayed 3D heart model. In some aspects, the method may include displaying the localization of the arrhythmia as multiple points representative of multiple beats of ventricular tachycardia on the displayed 3D heart model.15448153-1 4Attorney Docket No.: 35313-0011WO

[0017] In some aspects, the arrhythmia may be an atrial arrhythmia. In some aspects, the arrhythmia may be a ventricular arrhythmia. In some aspects, the arrhythmia may be a pre-ventricular contraction (PVC). In some aspects, the arrhythmia may be a ventricular tachycardia. In some aspects, the arrhythmia may be a dysrhythmia between the two ventricles.

[0018] Further aspects may include a computing device having a processing system with processor-executable instructions to perform operations corresponding to any of the methods summarized above. Further aspects may include a non-transitory processor-readable storage medium having stored thereon processor-executable instructions to cause a processing system to perform operations corresponding to any of the methods summarized above. Further aspects may include a computing device having various means for performing functions corresponding to any of the method operations summarized above.BRIEF DESCRIPTION OF THE FIGURES

[0019] The accompanying drawings, which are incorporated herein and constitute part of this specification, illustrate exemplary embodiments of the claims and, together with the general description given and the detailed description, serve to explain the features herein. It is noted that, in accordance with the standard practice in the industry, various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or reduced for clarity of discussion.

[0020] FIG. 1 is a block diagram of a system for processing and analyzing cardiac data according to various embodiments.

[0021] FIGS. 2A-2C are orthogonal views of a human heart showing anatomical structures and locations of intracardiac electrodes suitable for use with three alternative embodiments.

[0022] FIG. 3 is a perspective view of a three-dimensional heart model showing a shock vector angle according to various embodiments.15448153-1 5Attorney Docket No.: 35313-0011WO

[0023] FIGS. 4A is a three-dimensional heart model showing a localization and cardiac activation map of a single arrhythmia event according to various embodiments.

[0024] FIG. 4B is a three-dimensional view of a heart model with a localization point displayed on its surface according to various embodiments.

[0025] FIG. 5 is a process flow diagram of a method for using electrophysiology data of a patient’s heart obtained from intracardiac EP catheters to identify arrhythmia locations and conduct an ablation procedure according to various embodiments.

[0026] FIG. 6 is a process flow diagram of a method for using ICD data to identify arrhythmia locations and conduct an ablation procedure according to various embodiments.

[0027] FIG. 7 is a process flow diagram of a method for using ICD data to identify arrhythmia locations and conduct an ablation procedure according to various embodiments.

[0028] FIG. 8 is a block diagram of another example system for processing cardiac data and supporting ablation procedures according to various embodiments.

[0029] FIG. 9A is an anatomical view of a heart model showing an example of anatomical structures and ICD electrode placements.

[0030] FIG. 9B is a cross-sectional view of a heart model showing an example of anatomical structures and EP catheter electrode placements.

[0031] FIG. 9C is a cross-sectional view of a heart model showing an example of anatomical structures, ICD and EP catheter electrode placements.

[0032] FIG. 10A is a set of graphs showing activation time measurements from implantable cardioverter-defibrillator electrograms.

[0033] FIG. 10B shows example EGM traces of electrical potentials from three combinations of ICD electrodes (RV tip-to-RV coil, can-to-RV coil, and RV tip-to-15448153-1 6Attorney Docket No.: 35313-0011WORV ring) illustrating a method of identifying a global activation time according to some embodiments.

[0034] FIG. 10C is an illustration of vectors from an electrode to locations on a heart of ICD electrodes for determining vectorcardiograms (VCG).

[0035] FIG. 11A is an illustration of multi-lead internal electrocardiographic (MLiECG) framework or system that is defined based on intracardiac leads (e.g., CEID leads).

[0036] FIG. 11B is an illustration of an example coordinate system to anatomically align vector leads in a MLiECG framework.

[0037] FIG. 11C is an illustration of another example coordinate system to anatomically align vector leads in a MLiECG framework.

[0038] FIG. 11D is an illustration of a bipolar internal vector lead triangle which forms a first, second, and third vector leads of the MLiECG framework.

[0039] FIG. 12A is an isochrone diagram illustrating simulated cardiac activation patterns from a right ventricle tip stimulation.

[0040] FIG. 12B is a heart model view showing isochrones and an activation search region identified via simulation based on RV lead signals only.

[0041] FIG. 12C is a heart model view showing isochrones and an activation search region identified via simulation based on LV lead signals only.

[0042] FIG. 12D is a heart model view showing isochrones and an activation search region identified via simulation based on using both RV and LV lead signals.

[0043] FIG. 13 is a process flow diagram of a method for localizing and treating cardiac arrhythmias using ICD data according to various embodiments.

[0044] FIG. 14 is a process flow diagram of a method for processing electrocardiogram data according to some embodiments.15448153-1 7Attorney Docket No.: 35313-0011WO

[0045] FIG. 15 is a process flow diagram of a method for localizing and treating cardiac arrhythmias according to some embodiments.

[0046] FIG. 16 is a process flow diagram of a method for determining and using a likely point of origin of cardiac arrhythmia according to some embodiments.

[0047] FIG. 17 is a process flow diagram of a method for localizing and treating cardiac arrhythmias using electrophysiology (EP) intracardiac catheter electrode data according to some embodiments.

[0048] FIG. 18 is diagram of a laptop computer, according to various embodiments.

[0049] FIG. 19 is diagram of a server, according to various embodiments.DETAILED DESCRIPTION

[0050] Various embodiments will be described in detail with reference to the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts. References made to particular examples and implementations are for illustrative purposes and are not intended to limit the scope of the claims.

[0051] Various embodiments include a method for identifying and displaying arrhythmia localization using information obtained from electrodes positioned within a patient’s heart. Such intracardiac electrodes may be electrode on electrophysiology (EP) catheters temporarily positioned in the heart during an EP procedure or as part of an ablation procedure, the pacing and / or sensing electrodes of cardiac electronic implantable devices (e.g., an implanted cardiac defibrillator (ICD)), or both. Various embodiment methods combine electrical signal and electrode position data from intracardiac electrodes, particularly such data recorded during arrhythmia events (e.g., ventricular tachyarrhythmias (VT) that could be polymorphic or monomorphic ventricular tachycardia, or premature ventricular contractions (PVC)), with medical imaging to generate a patient-specific three-dimensional (3D) heart model and electrical conduction map. In some variations, the electrical conduction map may15448153-1 8Attorney Docket No.: 35313-0011WOindicate how the electrical activity propagates in the heart. In some variations, the electrical conduction map can be generated by simulating the propagation of activation (i.e., depolarization) wave through the heart. For example, the electrical conduction map can be generated by simulating the propagation of activation wave through the heart based on an electrical conduction model. Additionally or alternatively, the electrical conduction map can be generated by using EGM data recorded during a cardiac procedure. Generating such a 3D heart model and electrical conduction map using electrogram data gathered within the heart during arrhythmia events or VT may provide precise localization of arrhythmia initiation sites for use in performing cardiac ablation procedures. The term measured or recorded EGM may refer to the intracardiac electrocardiograms that are measured or recorded during VT and the term simulated EGM may refer to the intracardiac electrocardiograms that are generated or simulated based on computational models of the heart (as described herein). Such computational models may be based on boundary element method (BEM), or finite element method (FEM) or hybrid element method (HEM) that combined BEM and FEM.

[0052] This approach integrates data from cardiac electronic implantable devices (CEIDs), such as for example ICDs, which continuously monitor heart rhythms, with advanced 3D modeling techniques to provide physicians with detailed visualizations of arrhythmia origins. By merging CEID data, medical imaging, and electrocardiogram (ECG) recordings, the method offers an improved tool for planning and guiding electrophysiology procedures, potentially enhancing the efficacy of arrhythmia treatments.

[0053] The terms “computing system” and “computing device” are used herein to refer to (but not limited to) any one or all of servers, workstations, desktop computers, laptop computers, tablet devices, cellular phones, virtual machines, implantable devices (e.g., ICDs or other CEIDs) and other similar computing systems that include a memory for storing documents and computational data, and a programmable processing system that may provide the functionality of various embodiments.15448153-1 9Attorney Docket No.: 35313-0011WO

[0054] The term “processing system” is used herein to refer to one or more processors, including multi-core processors, graphics processing units (GPU), neural network processing units (NPU), microprocessor units (MPU), arithmetic logic units (ALU), memory systems, etc., that are organized and perform computing functions of various embodiments as described herein.

[0055] The term “activation” or local activation of a given myocardial location, is used hereinafter to refer to the time point of myocardial depolarization or activation at that myocardial location. During arrhythmia, activation wave originates at a location where the arrhythmia, e.g., VT or PVC, is first initiated as in focal arrhythmias or is emerged from a protected isthmus as in reentrant arrhythmias, which is also referred to hereinafter as the point of origin of an arrhythmia.

[0056] The terms “global activation time” and “t=0” are used herein to refer to the time at which the activation wave during an arrhythmia beat or cardiac cycle first begins at the arrhythmia initiation site. As the activation wave travels at some finite speed(s) or conduction velocities through heart tissues, the activation time at which an electrode senses the initiation of depolarization of an arrhythmia beat will depend on the distance of the recording electrode location from the initiation site of the arrhythmia beat and the conduction velocities. The term local activation time (LAT) of a given myocardial location, is used hereinafter to refer to the time point interval from the very beginning of myocardial depolarization or activation, within either the atria or ventricles, to the time point when the front of the myocardial activation wave, or myocardial activation wavefront arrives at that myocardial location. LAT can be determined based on analysis of measured or recorded EGMs or be calculated or “simulated” based on mathematical model(s) or be approximated by neighboring recording electrodes in which case it represents the LAT of the myocardium immediately underneath or adjacent to the unipolar recording electrode or between the paired bipolar recording electrodes. The term “earliest activation time” is used herein to refer to the time at which the depolarization signal or activation wave is first measured or detected by any of a set of electrodes. If an electrode is located close to15448153-1 10Attorney Docket No.: 35313-0011WOthe arrhythmia activation site, then the LAT of that electrode may be the earliest activation time and a first approximation to the global activation time or t=0. Further methods that the system may use to identify the global activation time are described below with reference to FIG. 10B..

[0057] Current methods for localizing cardiac arrhythmias, particularly ventricular tachycardia (VT) and premature ventricular contractions (PVC), often rely on body surface electrocardiograms (ECGs) or invasive electrophysiology (EP) studies.However, these approaches may fail to identify clinical arrhythmia during ablation procedures, leading to potentially ineffective treatments and prolonged procedural time that may be associated with increased complications and mortality. Additionally, existing techniques for analyzing electrograms (EGMs) from CEID, e.g., ICD, are limited in their ability to precisely identify arrhythmia origins. Therefore, there is an unmet need for a method that can accurately localize arrhythmia sources using CEID EGM data (e.g., ICD EGM data), combined with advanced 3D heart modeling and simulation techniques, to improve the efficiency and efficacy of cardiac ablation procedures while reducing patient risk and procedure duration.

[0058] Various embodiments include systems and methods for localizing and treating cardiac arrhythmia using EGM data from intracardiac electrodes. In particular, the EGM data from intracardiac electrodes may be combined with three-dimensional heart modeling and simulation techniques to identify the location of arrhythmia. For example, the EGM data may be combined with a patient-specific three-dimensional (3D) heart model to identify the location of arrhythmia. In some embodiments, the systems and methods may combine the EGM data from intracardiac electrodes with three-dimensional heart modeling techniques based on a multi-lead internal electrocardiographic (MLiECG) framework or system. Herein, the terms MLiECG framework, MLiECG system and MLiECG vector system may be used interchangeably. The MLiECG framework enables the ability to precisely define the location of the intracardiac electrodes within a patient-specific three-dimensional (3D)15448153-1 11Attorney Docket No.: 35313-0011WOheart model. For instance, the location of the electrodes may be precisely defined with respect to the anatomical features of the heart.

[0059] In some embodiments, the systems and methods for localizing and treating cardiac arrhythmia may include simulating EGMs through the 3D heart model. The simulated EGMs may be EGMs that would be observed at each intracardiac electrode resulting from propagation of activation waves initiated at one or more locations within an area that is likely to be the origin of an arrhythmia. In some variations, the location of the electrodes may be precisely defined within the 3D heart model prior to simulating EGMs. The simulated EGMs can be compared with recorded EGMs (i.e., EGMs that are recorded by the intracardiac electrodes). The location of the most likely point of origin of arrhythmia may be identified based on the comparison.

[0060] Various embodiments improve the localization of critical isthmus sites, which may be particularly useful for identifying locations for ablation for treating ventricular tachycardia (VT). VT is a potentially life-threatening heart rhythm disorder characterized by a rapid irregular heartbeats originating in the ventricles. Critical isthmus(es) are locations in the heart where heart tissues enable sustained depolarization re-entrant circuits that can cause VT. Specifically, a critical isthmus is defined as a pathway within the reentrant circuit where there is no lateral propagation across the isthmus and conduction velocity is typically reduced. This slow conduction can facilitate the circuit’s persistence, as it allows the electrical wavefront to re-excite tissue that has just recovered from the previous excitation, thus sustaining the tachycardia. Critical isthmus sites are locations where an ablation procedure may be most effective by breaking the circuit.

[0061] According to an embodiment there is provided a method for localizing and treating cardiac arrhythmias, comprising:simulating ICD EGMs that would be observed at each ICD electrode resulting from propagation of activation waves initiated at each of multiple locations within a likely arrhythmia origin area;15448153-1 12Attorney Docket No.: 35313-0011WOcomparing the simulated ICD EGMs to the recorded ICD EGMs; identifying a location that is a most likely point of origin of the arrhythmia based on the comparison; andoutputting the identified most likely point of origin of the arrhythmia for use in conducting a cardiac ablation procedure.

[0062] In an embodiment, the method further comprises:generating a patient-specific three-dimensional (3D) heart model based on the medical imaging of a patient’s heart;obtaining implantable cardioverter-defibrillator (ICD) electrograms (EGMs) recorded from electrodes positioned within the patient’s heart;processing the ICD EGMs to select an arrhythmia beat;determining for each ICD electrode a local activation time (LAT) of the arrhythmia beat and identifying a global activation time (t=0);simulating time-shifted activation waves through the 3D heart model in response to pacing at one or more ICD electrodes, wherein simulations of each timeshifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding ICD electrode; anddetermining the likely arrhythmia origin area based on the simulated timeshifted activation waves.

[0063] In an embodiment, the method further comprises conducting the ablation procedure.

[0064] In an embodiment, simulating time-shifted activation waves through the 3D heart model uses the 3D heart model to simulate progression of the activation wave front through heart tissues from a pacing or initiation at an electrode location.

[0065] In an embodiment, the method further comprises localizing ICD electrodes within the 3D heart model based on the medical imaging, wherein simulations of timeshifted activation waves and simulations of EGMs that would be observed at each15448153-1 13Attorney Docket No.: 35313-0011WOICD electrode from initiation at multiple locations within the likely arrhythmia origin area uses the location of each ICD electrode in the 3D heart model.

[0066] In an embodiment, the method further comprises including locations and characteristics of scar tissue in the 3D heart model.

[0067] In an embodiment, determining the local activation time (LAT) of the ICD electrodes comprises identifying a time of earliest significant change in potential in one of the ICD EGMs.

[0068] In an embodiment, determining the likely arrhythmia origin area comprises identifying an intersection of isochrones at t=0 from the simulated time-shifted activation waves.

[0069] In an embodiment:comparing the simulated EGMs to the recorded ICD EGMs comprises using a correlation metric to identify a best match; andidentifying a location that is a most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated EGM as the most likely point of origin of the arrhythmia.

[0070] In an embodiment, the method further comprises:positioning electrocardiogram (ECG) electrodes on the patient’s thorax; obtaining electrograms from the ECG electrodes;time-synchronizing electrograms from the ECG electrodes to ICD EGMs; processing the ICD EGMs and ECG electrograms together to select the arrhythmia beat;simulating time-shifted activation waves from the at least one ECG electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding ECG electrode;15448153-1 14Attorney Docket No.: 35313-0011WOsimulating ECG signals that would be observed at each ECG electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area; andcomparing the simulated ECG signals to the recorded ECG signals, wherein identifying a location that is the most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated ICD EGM or ECG signals as the most likely point of origin of the arrhythmia.

[0071] In an embodiment, the method further comprises:positioning electrophysiology (EP) catheters in the patient’s heart; localizing EP catheter electrodes within the patient’s heart;obtaining electrograms from the EP catheter electrodes;processing the ICD EGMs and EP electrode electrograms together to select the arrhythmia beat;determining LAT of the EP electrodes;simulating time-shifted activation waves from the EP electrodes, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding EP electrode;simulating EP EGMs that would be observed at each EP electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area; andcomparing the simulated EP EGMs to the recorded EP EGMs,wherein identifying a location that is the most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated ICD EGM or EP EGM as the most likely point of origin of the arrhythmia.

[0072] In an embodiment the arrhythmia is a ventricular tachycardia (VT) or a premature ventricular contraction (PVC).15448153-1 15Attorney Docket No.: 35313-0011WO

[0073] In an embodiment simulating EGMs from multiple locations comprises simulating EGMs from all grid points in the 3D heart model within the likely arrhythmia origin area.

[0074] In an embodiment, the method further comprises updating the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during an EP procedure.

[0075] In an embodiment, the method further comprises calibrating the 3D heart model by:pacing the heart at multiple locations during an EP procedure;recording ICD EGMs during the pacing; andadjusting conduction velocity parameters in the 3D heart model based on measured activation times between the pacing location and ICD electrodes.

[0076] In an embodiment the ICD electrodes include one or more of a right ventricle (RV) tip electrode, an RV ring electrode, an RV coil electrode, a can electrode, and a left ventricle (LV) electrode.

[0077] In an embodiment determining the LAT of electrodes comprises identifying a minimum of a first derivative of an EGM of the electrode.

[0078] In an embodiment the medical imaging comprises magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient’s heart.

[0079] In an embodiment:the ICD EGMs are obtained from a cardiac resynchronization therapy (CRT) device having electrodes in both right and left ventricles; anddetermining the likely arrhythmia origin area is based on the simulated timeshifted activation waves from the electrodes in both right and left ventricles to reduce a size of the likely arrhythmia origin area.

[0080] In an embodiment outputting the identified most likely point of origin of the arrhythmia comprises displaying the 3D heart model with the identified most likely15448153-1 16Attorney Docket No.: 35313-0011WOpoint of origin of the arrhythmia highlighted for use during the cardiac ablation procedure.

[0081] In an embodiment there is provided a system for localizing cardiac arrhythmias, comprising:a memory configured to store a patient-specific three-dimensional (3D) heart model and electrogram data;an interface configured to receive implantable cardioverter-defibrillator (ICD) electrograms (EGMs) from electrodes positioned within a patient’s heart; anda processing system configured to:simulate, using the stored model, ICD EGMs that would be observed at each ICD electrode resulting from propagation of activation waves initiated at each of multiple locations within a likely arrhythmia origin area;compare the simulated ICD EGMs to the recorded ICD EGMs;identify a location that is a most likely point of origin of the arrhythmia based on the comparison; andoutput the identified most likely point of origin of the arrhythmia.

[0082] In an embodiment the processing system is further configured to generate the patient-specific 3D heart model based on medical imaging of the patient’s heart;process the received ICD EGMs to select an arrhythmia beat;determine for each ICD electrode a local activation time (LAT) of the arrhythmia beat and identify an earliest global activation time;simulate time-shifted activation waves through the 3D heart model in response to pacing at each ICD electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding ICD electrode;15448153-1 17Attorney Docket No.: 35313-0011WOdetermine the likely arrhythmia origin area based on the simulated time-shifted activation waves.

[0083] In an embodiment the processing system is further configured to simulate time-shifted activation waves through the 3D heart model by using the 3D heart model to simulate progression of an activation wave front through heart tissues from a pacing or initiation at an electrode location.

[0084] In an embodiment the processing system is further configured to include locations of ICD electrodes within the 3D heart model based on the medical imaging, and wherein simulations of time-shifted activation waves and simulations of EGMs that would be observed at each ICD electrode from initiation at multiple locations within the likely arrhythmia origin area uses the location of each ICD electrode in the 3D heart model.

[0085] In an embodiment the processing system is further configured to include locations and characteristics of scar tissue in the 3D heart model.

[0086] In an embodiment the processing system is configured to determine the local activation time (LAT) of the ICD electrodes by identifying a time of earliest significant change in potential in one of the ICD EGMs.

[0087] In an embodiment the processing system is configured to determine the likely arrhythmia origin area by identifying an intersection of isochrones at t=0 from the simulated time-shifted activation waves.

[0088] In an embodiment the processing system is configured to:compare the simulated EGMs to the recorded ICD EGMs by using a correlation metric to identify a best match; andidentify a location that is a most likely point of origin of the arrhythmia by identifying the initiation location of the best matching simulated EGM as the most likely point of origin of the arrhythmia.

[0089] In an embodiment:the interface is further configured to receive electrograms from electrocardiogram (ECG) electrodes positioned on the patient’s thorax; and15448153-1 18Attorney Docket No.: 35313-0011WOthe processing system is further configured to:process the ICD EGMs and ECG electrograms together to select the arrhythmia beat;determine LAT of at least one ECG electrode;simulate time-shifted activation waves from the at least one ECG electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding ECG electrode;simulate ECG signals that would be observed at each ECG electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area; andcompare the simulated ECG signals to the recorded ECG signals, wherein identifying a location that is the most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated ICD EGM or ECG signals as the most likely point of origin of the arrhythmia.

[0090] In an embodiment:the interface is further configured to receive electrograms from electrophysiology (EP) catheter electrodes positioned within the patient’s heart; and the processing system is further configured to:localize EP catheter electrodes within the patient’s heart;process the ICD EGMs and EP electrode electrograms together to select the arrhythmia beat;determine LAT of the EP electrodes;simulate time-shifted activation waves from the EP electrodes, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding EP electrode;15448153-1 19Attorney Docket No.: 35313-0011WOsimulate EP EGMs that would be observed at each EP electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area; andcompare the simulated EP EGMs to the recorded EP EGMs,wherein identifying a location that is the most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated ICD EGM or EP EGM as the most likely point of origin of the arrhythmia.

[0091] In an embodiment the arrhythmia is a ventricular tachycardia (VT) or a premature ventricular contraction (PVC).

[0092] In an embodiment the processing system is configured to simulate EGMs from multiple locations by simulating EGMs from all grid points in the 3D heart model within the likely arrhythmia origin area.

[0093] In an embodiment the processing system is further configured to update the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during an EP procedure.

[0094] In an embodiment the processing system is further configured to calibrate the 3D heart model by:outputting signals for controlling pacing of the heart at multiple locations during an EP procedure;recording ICD EGMs during the pacing; andadjusting conduction velocity parameters in the 3D heart model based on measured activation times between the pacing location and ICD electrodes.

[0095] In an embodiment the ICD electrodes include one or more of a right ventricle (RV) tip electrode, an RV ring electrode, an RV coil electrode, a can electrode, and a left ventricle (LV) electrode.

[0096] In an embodiment the processing system is configured to determine the LAT of electrodes by identifying a minimum of a first derivative of an EGM of the electrode.15448153-1 20Attorney Docket No.: 35313-0011WO

[0097] In an embodiment the medical imaging comprises magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient’s heart.

[0098] In an embodiment the ICD EGMs are obtained from a cardiacresynchronization therapy (CRT) device having electrodes in both right and left ventricles; and the processing system is configured to determine the likely arrhythmia origin area based on the simulated time-shifted activation waves from the electrodes in both right and left ventricles to reduce a size of the likely arrhythmia origin area.

[0099] In an embodiment the system further comprises a display, wherein the processing system is configured to output the identified most likely point of origin of the arrhythmia by displaying the 3D heart model with the identified most likely point of origin of the arrhythmia highlighted for use during the cardiac ablation procedure.

[0100] In an embodiment there is provided a method for localizing and treating cardiac arrhythmias using electrophysiology electrodes, comprising:obtaining medical imaging of a patient's heart;generating a patient-specific three-dimensional (3D) heart model based on the medical imaging and a selected representative 3D heart model;positioning electrophysiology (EP) electrode catheters in the patient's heart; localizing EP electrodes within the patient's heart and within the 3D heart model;obtaining EP electrode electrograms (EGMs) from an EP system; processing the EP electrode EGMs to select an arrhythmia beat; determining for each EP electrode a local activation time (LAT) of the arrhythmia beat and identifying an earliest LAT (t=0);simulating time-shifted activation waves through the 3D heart model in response to pacing at each EP electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding EP electrode;15448153-1 21Attorney Docket No.: 35313-0011WOdetermining a likely arrhythmia origin area based on the simulated time-shifted activation waves;simulating EP electrode EGMs that would be observed at each EP electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area;comparing the simulated EP electrode EGMs to recorded EP electrode EGMs; identifying a location that is a most likely point of origin of the arrhythmia based on the comparison; andoutputting the identified most likely point of origin of the arrhythmia for use in conducting a cardiac ablation procedure.

[0101] In an embodiment, the method further comprises conducting the ablation procedure.

[0102] In an embodiment simulating time-shifted activation waves through the 3D heart model uses the 3D heart model to simulate progression of the activation wave front through heart tissues from a pacing or initiation at an EP electrode location.

[0103] In an embodiment determining the LAT of each EP electrode comprises identifying a time of earliest significant change in potential in each EP electrode EGM.

[0104] In an embodiment determining the likely arrhythmia origin area comprises identifying an intersection of isochrones at t=0 from the simulated time-shifted activation waves.

[0105] In an embodiment:comparing the simulated EP electrode EGMs to the recorded EP electrode EGMs comprises using a correlation metric to identify a best match; and identifying a location that is a most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated EP electrode EGM as the most likely point of origin of the arrhythmia.15448153-1 22Attorney Docket No.: 35313-0011WO

[0106] In an embodiment the arrhythmia is a ventricular tachycardia (VT) or a premature ventricular contraction (PVC).

[0107] In an embodiment simulating EP electrode EGMs from multiple locations comprises simulating EP electrode EGMs from all grid points in the 3D heart model within the likely arrhythmia origin area.

[0108] In an embodiment the method further comprises updating the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during the EP procedure.

[0109] In an embodiment the method further comprises calibrating the 3D heart model by:pacing the heart at multiple locations during the EP procedure;recording EP electrode EGMs during the pacing; andadjusting conduction velocity parameters in the 3D heart model based on measured activation times between the pacing location and EP electrodes.

[0110] In an embodiment determining the LAT of each EP electrode comprises identifying a minimum of a first derivative of an EGM of each EP electrode.

[0111] In an embodiment the medical imaging comprises magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient's heart.

[0112] In an embodiment outputting the identified most likely point of origin of the arrhythmia comprises displaying the 3D heart model with the identified most likely point of origin of the arrhythmia highlighted for use during the cardiac ablation procedure.

[0113] In an embodiment there is provided a system for localizing and treating cardiac arrhythmias using electrophysiology electrodes, comprising:a memory configured to store a patient-specific three-dimensional (3D) heart model and electrogram data;an interface configured to receive electrophysiology (EP) electrode electrograms (EGMs) from an EP system; and15448153-1 23Attorney Docket No.: 35313-0011WOa processing system configured to:generate the patient-specific 3D heart model based on medical imaging of the patient's heart and a selected representative 3D heart model;localize EP electrodes within the patient's heart and within the 3D heart model; process the EP electrode EGMs to select an arrhythmia beat;determine for each EP electrode a local activation time (LAT) of the arrhythmia beat and identify an earliest LAT (t=0);simulate time-shifted activation waves through the 3D heart model in response to pacing at each EP electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding EP electrode;determine a likely arrhythmia origin area based on the simulated time-shifted activation waves;simulate EP electrode EGMs that would be observed at each EP electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area;compare the simulated EP electrode EGMs to recorded EP electrode EGMs; identify a location that is a most likely point of origin of the arrhythmia based on the comparison; andoutput the identified most likely point of origin of the arrhythmia for use in conducting a cardiac ablation procedure.

[0114] In an embodiment the processing system is further configured to output control signals suitable for controlling conducting the ablation procedure.

[0115] In an embodiment the processing system is configured to simulate timeshifted activation waves through the 3D heart model by using the 3D heart model to15448153-1 24Attorney Docket No.: 35313-0011WOsimulate progression of the activation wave front through heart tissues from a pacing or initiation at an EP electrode location.

[0116] In an embodiment the processing system is configured to determine the LAT of each EP electrode by identifying a time of earliest significant change in potential in each EP electrode EGM.

[0117] In an embodiment the processing system is configured to determine the likely arrhythmia origin area by identifying an intersection of isochrones at t=0 from the simulated time-shifted activation waves.

[0118] In an embodiment the processing system is configured to:compare the simulated EP electrode EGMs to the recorded EP electrode EGMs by using a correlation metric to identify a best match; andidentify a location that is a most likely point of origin of the arrhythmia by identifying the initiation location of the best matching simulated EP electrode EGM as the most likely point of origin of the arrhythmia.

[0119] In an embodiment the arrhythmia is a ventricular tachycardia (VT) or a premature ventricular contraction (PVC).

[0120] In an embodiment the processing system is configured to simulate EP electrode EGMs from multiple locations by simulating EP electrode EGMs from all grid points in the 3D heart model within the likely arrhythmia origin area.

[0121] In an embodiment the processing system is further configured to update the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during an EP procedure.

[0122] In an embodiment the processing system is further configured to calibrate the 3D heart model by:outputting signals suitable for controlling pacing of the heart at multiple locations during an EP procedure;recording EP electrode EGMs during the pacing; and15448153-1 25Attorney Docket No.: 35313-0011WOadjusting conduction velocity parameters in the 3D heart model based on measured activation times between the pacing location and EP electrodes.

[0123] In an embodiment the processing system is configured to determine the LAT of each EP electrode by identifying a minimum of a first derivative of an EGM of each EP electrode.

[0124] In an embodiment the medical imaging comprises magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient's heart.

[0125] In an embodiment the system further comprise a display, wherein the processing system is configured to output the identified most likely point of origin of the arrhythmia by displaying the 3D heart model with the identified most likely point of origin of the arrhythmia highlighted for use during the cardiac ablation procedure.

[0126] In an embodiment there is provided a computing system, comprising: a memory; anda processor coupled to the memory and configured with processor-executable instructions, the instructions, when executed by the processor, configured to:obtain electrical signals and location information from electrodes positioned within the patient’s heart;generate a patient-specific three-dimensional (3D) heart model;generate a patient-specific electrical conduction map of the patient’s heart during an arrhythmia based on the patient-specific 3D heart model and the electrical signals and location information from electrodes positioned within the patient’s heart, the patient- specific electrical conduction map of the patient’s heart identifying a 3D localization of an initiation site of the arrhythmia;display the 3D localization of the initiation site of the arrhythmia for use by a physician preparing for and / or during a cardiac electrophysiology procedure; and15448153-1 26Attorney Docket No.: 35313-0011WOoutput signals suitable for conducting a cardiac ablation procedure using the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for ablation.

[0127] In an embodiment the instructions, when executed, further configured to cause the processor to, when obtaining electrical signals and location information from electrodes positioned within the patient’s heart, receive signals from electrodes on one or more electrophysiology (EP) catheters temporarily positioned with the heart, determining locations of the electrodes within the heart based on information from an EP system

[0128] In an embodiment the instructions, when executed, further configured to cause the processor to, when obtaining electrical signals and location information from electrodes positioned within the patient’s heart,download from an implanted cardiac defibrillator (ICD) information recorded during detected arrhythmia events in the patient’s heart;obtain medical imaging that identifies locations of one or more ICD electrodes in the patient’s heart;generate the patient-specific three-dimensional (3D) heart model including locations of the one or more ICD electrodes on the heart.

[0129] In an embodiment the instructions, when executed, further configured to cause the processor to conduct point by point contact electrophysiology recordings during the cardiac ablation procedure and updating the patient-specific electrical conduction map to display an updated 3D localization of the initiation site of the arrhythmia.

[0130] In an embodiment generating a patient-specific 3D model of the heart includes a 3D internal surface model comprises using magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient to generate the patientspecific 3D heart model including locations of the one or more ICD electrodes.15448153-1 27Attorney Docket No.: 35313-0011WO

[0131] In an embodiment the instructions, when executed, further configured to cause the processor to combineICD information recorded during detected arrhythmia events in the patient’s heart with the patient-specific 3D heart model to identify isochrones at time intervals of heartbeats to determine directions of depolarization wavefronts to reveal an initiation site of the arrhythmia in a heartbeat.

[0132] In an embodiment the instructions, when executed, further configured to cause the processor to display heart structures including one or more of the aorta, aortic arch, pulmonary veins, or coronary vessels on the displayed 3D heart model.

[0133] In an embodiment the instructions, when executed, further configured to cause the processor to display heart scar tissue indicative of ischemic heart disease on the displayed 3D heart model.

[0134] In an embodiment the instructions, when executed, further configured to cause the processor to display the localization of the arrhythmia as multiple points representative of multiple beats of a ventricular tachycardia on the displayed 3D heart model.

[0135] In an embodiment the arrhythmia is one of an atrial arrhythmia, a ventricular arrhythmia, a pre-ventricular contraction (PVC), a ventricular tachycardia, or a dysrhythmia between the two ventricles.

[0136] In an embodiment the instructions, when executed, further configured to cause the processor to:obtain a 3D image of electrocardiogram (ECG) electrodes on the patient’s torso during an ECG recording procedure; andmerge the 3D image of the patient’s torso with the 3D heart image, wherein generating a patient-specific electrical conduction map of the patient’s heart during an arrhythmia is based on the patient-specific 3D heart model, the electrical signals and location information from electrodes positioned within the patient’s heart, and ECG data from the ECG recording procedure.15448153-1 28Attorney Docket No.: 35313-0011WO

[0137] In an embodiment there is provided a method of using information from medical electrodes positioned in a patient’s heart to identify and display arrhythmia localization, comprising:obtaining electrical signals and location information from electrodes positioned within the patient’s heart;generating a patient-specific three-dimensional (3D) heart model; generating a patient-specific electrical conduction map of the patient’s heart during an arrhythmia based on the patient- specific 3D heart model and the electrical signals and location information from electrodes positioned within the patient’s heart, the patient-specific electrical conduction map of the patient’s heart identifying a 3D localization of an initiation site of the arrhythmia;displaying the 3D localization of the initiation site of the arrhythmia for use by a physician preparing for and / or during a cardiac electrophysiology procedure; and conducting a cardiac ablation procedure using the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for ablation.

[0138] In an embodiment obtaining electrical signals and location information from electrodes positioned within the patient’s heart comprises receiving signals from electrodes on one or more electrophysiology (EP) catheters temporarily positioned with the heart, determining locations of the electrodes within the heart based on information from an EP system.

[0139] In an embodiment obtaining electrical signals and location information from electrodes positioned within the patient’s heart comprises downloading from an implanted cardiac defibrillator (ICD) information recorded during detected arrhythmia events in the patient’s heart, the method further comprising:obtaining medical imaging that identifies locations of one or more ICD electrodes in the patient’s heart;generating the patient-specific three-dimensional (3D) heart model including locations of the one or more ICD electrodes on the heart.15448153-1 29Attorney Docket No.: 35313-0011WO

[0140] In an embodiment the method further comprises conducting point by point contact electrophysiology recordings during the cardiac ablation procedure and updating the patient-specific electrical conduction map to display an updated 3D localization of the initiation site of the arrhythmia.

[0141] In an embodiment generating a patient-specific 3D model of the heart including a 3D internal surface model comprises using magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient to generate the patientspecific 3D heart model including locations of the one or more ICD electrodes.

[0142] In an embodiment ICD information recorded during detected arrhythmia events in the patient’s heart is combined with the patient- specific 3D heart model to identify isochrones at time intervals of heartbeats to determine directions of depolarization wavefronts to reveal an initiation site of the arrhythmia in a heartbeat.

[0143] In an embodiment the method further comprises displaying heart structures including one or more of the aorta, aortic arch, pulmonary veins, or coronary vessels on the displayed 3D heart model.

[0144] In an embodiment the method further comprises displaying heart scar tissue indicative of ischemic heart disease on the displayed 3D heart model.

[0145] In an embodiment the method further comprises displaying the localization of the arrhythmia as multiple points representative of multiple beats of a ventricular tachycardia on the displayed 3D heart model.

[0146] In an embodiment the arrhythmia is one of an atrial arrhythmia, a ventricular arrhythmia, a pre-ventricular contraction (PVC), a ventricular tachycardia, or a dysrhythmia between the two ventricles.

[0147] In an embodiment the method further comprises:obtaining a 3D image of electrocardiogram (ECG) electrodes on the patient’s torso during an ECG recording procedure; andmerging the 3D image of the patient’s torso with the 3D heart image, wherein generating a patient-specific electrical conduction map of the patient’s heart during an arrhythmia is based on the patient-specific 3D heart model, the15448153-1 30Attorney Docket No.: 35313-0011WOelectrical signals and location information from electrodes positioned within the patient’s heart, and ECG data from the ECG recording procedure.

[0148] Existing Methods for Localization of Arrhythmia

[0149] One method of locating isthmuses is known as wavefront curvature analysis. Wavefront curvature refers to the bending of the electrical activation wavefront as it propagates through the heart. In cardiac electrophysiology (EP), high curvature, particularly in convex shapes, can lead to a functional conduction block due to source-sink mismatch, where the activating current is insufficient to excite the downstream tissue. In reentrant circuits, the critical isthmus is where this slow conduction frequently occurs, further enabling the circuit to sustain itself.

[0150] One such existing method includes identifying a critical isthmus location using cardiac imaging (e.g., MRI or CT) to estimate wavefront curvature in the ventricles based on local gradients in the heart wall thickness. According to this existing methodology, regions with abrupt changes in wall thickness, leading to high wavefront curvature, are likely sites of slow conduction or block, which are characteristic of the critical isthmus.

[0151] Another existing method for identifying curvature in the depolarization wavefront comprises a computational method that is designed to enhance the accuracy of cardiac imaging and electrocardiographic analysis. It integrates multiple data sources, including 3D heart models derived from images from computed tomography (CT) or magnetic resonance imaging (MRI), and electrocardiogram (EGM) data that is obtained from intracardiac electrodes, to provide a three-dimensional (3D) graphical representation of the heart's electrical activity.

[0152] For example, this method begins by creating a 3D model of the heart (also referred to herein as “3D heart model”) and torso (also referred to herein as a “3D torso model”) from CT or MRI images. In some embodiments, the 3D heart model is initially matched to one of eight reference models of human hearts, with manual adjustments made to the models to better fit the patient's specific heart geometry. The15448153-1 31Attorney Docket No.: 35313-0011WOmethod then aligns the location of ECG leads on the patient’s torso to the 3D heart model and the 3D torso model. Thus, conventionally, ECG electrodes are positioned on the patient’s torso, and the existing method uses a 3D camera to capture surface images of electrode positions and merge the locations of the ECG electrodes identified in the images with the 3D torso model.

[0153] Isochrones, or lines of equal depolarization time, are calculated based on ECG data from each ECG lead starting from each heartbeat depolarization starting point in the 3D heart model. The average direction of the depolarization wave is then determined at three (or more) specific time points during the QRS complex of the ECG, such as 0.1, 0.5, and 0.8 times the QRS duration. These calculations result in "model vectors" that represent the direction of electrical activity associated with the depolarization wave moving through heart tissue.

[0154] The method further refines its analysis by using unipolar surface ECG leads and the Wilson central terminal as a reference. For each of the three-time points, it computes the average position of isochrones and adjusts the ECG lead vectors accordingly. The vertex that maximizes the dot products of the adjusted ECG vectors and model vectors is identified as the origin of the electrical activity. Mapping the ECG vectors to the patient-specific 3D heart model enables the method to generate a 3D rendering of how the depolarization wavefront passes through heart tissue. This visualization can reveal impediments and imperfections in the wavefront, such as locations of wavefront curvature and reentrant circuits.

[0155] However, initiation sites of ventricular tachyarrhtyhmias (VT), including monomorphic and polymorphic ventricular tachycardias and PVC, can be hard to locate using Electrophysiology (EP) measurements due to appearing episodically during VT, which may not coincide with an EP procedure.15448153-1 32Attorney Docket No.: 35313-0011WO

[0156] Overview of Example Systems and Methods for Precise Localization of Arrhythmia

[0157] Although various embodiments described herein disclose systems and methods that use ICDs, it should be readily understood that these systems and methods can include any suitable cardiac electronic implantable medical device.Various embodiments described herein may improve the identification of VT initiation sites by using EP data gathered by intracardiac electrodes on EP catheters during an EP procedure or obtained from a CEID that stores signal data from pacing electrodes as well as any CEID sensing electrodes. Either type of intracardiac electrode or both types of intracardiac electrodes may record electrical signals at different parts of the heart, particularly locations on the endocardium. Being located within the heart, intracardiac electrodes may gather electrical signals that are closer to arrhythmia initiation sites than surface ECG electrodes used in conventional methods such as those described above, thereby providing more detailed information for identifying wavefront curvature, critical isthmus sites, and reentrant circuits, as well as characterization of the heart’s conduction wavefront.

[0158] EP catheters described herein include a plurality of electrodes on the distal tip and along a length likely to be positioned within the heart during an EP procedure. In some embodiments, the EP catheter can be coupled to an EP system that collects signal data from each electrode on the EP catheter. CEID leads include a pacing electrode at the tip and may include sensing electrodes at various locations along the length of the lead within the heart, each of which may provide electrical signal data to a processor (e.g., CEID processor), for example for storage in memory.

[0159] In various embodiments, the electrical signal data from intracardiac electrodes such as electrodes on an EP catheter and / or electrodes in a CEID can be used to identify the timing of signals associated with contractions and tribulations of the heart. By combining the timing of such signals with the location in the heart of each electrode, a dynamic map of the conduction paths through the patient’s heart can be generated by a computing device. In various embodiments, such timing-location 15448153-1 33Attorney Docket No.: 35313-0011WOdata pairs from multiple intracardiac electrodes can be mapped to a 3D heart model of the patient’s heart to provide a 3D visualization of the patient’s heart’s heart conduction patterns.

[0160] For example, a 3D model of the heart can be generated from imaging data received from imaging systems (e.g., CT, MRI, PET images, or a combination thereof). In some embodiments, the 3D heart model is initially matched to one of a number of (e.g., eight) reference models of human hearts, with manual adjustments made to the models to better fit the patient's specific heart geometry. Thus, the 3D heart model that is generated may be patient-specific.

[0161] In some embodiments, the 3D heart model may include a 3D heart model that has been modeled using boundary element modeling, finite element modeling, hybrid boundary-finite element modeling, or a combination thereof. The location of the electrodes in the 3D heart model may be identified based on the imaging data from the imaging systems. Electrical potential for one or more points (e.g., vertices in a mesh geometrical model, or voxels in a discretized 3D volume) in the 3D heart model may be simulated based on EGM data from each intracardiac electrode. These electrical potentials may be used to generate simulated EGMs. Isochrones, or lines of equal depolarization time, may be calculated from each heartbeat depolarization starting point in the 3D heart model based on the simulated electrical potential. The average direction of the depolarization wave may then be determined at specific points during a ventricular EGM complex. This results in the generation of depolarization vectors or model vectors that represent the direction of the electrical activity associated with the depolarization wave through the heart tissue.

[0162] EGM data from each intracardiac electrode may be recorded. The depolarization vector may be calculated from this measured or recorded EGM data. In some embodiments, the simulated depolarization vector may be compared to the depolarization vector that was calculated from the recorded EGM data. For instance, the angles of the depolarization vectors for the simulated and recorded EGMs may be compared. The best match may provide an indication of the cardiac arrythmia origin.15448153-1 34Attorney Docket No.: 35313-0011WOAdditionally or alternatively, the simulated depolarization wave may be compared with a depolarization wave that is calculated from measured or recorded EGM data. The best match may provide an indication of the cardiac arrythmia origin.

[0163] In some embodiments, to further improve the accuracy and precision of localizing the origin of arrhythmia, the location of the electrodes may be precisely defined in the 3D heart model using the MLiECG framework and the anatomical alignment described herein. The recorded EGMs and the simulated EGMs may be defined based on the MLiECG framework and may be anatomically aligned. This may precisely align the location of the electrodes within the 3D heart model. A comparison of the simulated and the recorded vectors may be performed following such alignment to determine the cardiac arrythmia origin with further precision.

[0164] In some embodiments, the electrical signal and location data may be supplemented by signals captured by electrocardiogram (ECG) electrodes positioned on the patient’s thorax to supplement or improve the conduction map. Such visualization can reveal the initiation sites of VT and other arrhythmias, information that can be used to guide a physician performing an ablation procedure to cure or ameliorate the arrhythmia.

[0165] To map the conduction wave through the patient’s heart using electrical signals from electrodes within the heart, as well as on the thorax, a computing system (e.g., as described herein) obtains information on the location of each electrode from which signal data is obtained.

[0166] In the case of ECG electrodes on the patient’s thorax, an imaging system (e.g., a digital camera positioned at a known location in a coordinate system) may obtain digital images that can be processed by the computing system to identify the location of each electrode. The location of each electrode may be identified in a coordinate system that can be correlated to a frame of reference of the patient’s body (e.g., based on the images of the patient’s body with an imaging system positions at a15448153-1 35Attorney Docket No.: 35313-0011WOknown location in the coordinate system). Such localization techniques may identify the location of each ECG on the patient within a few millimeters.

[0167] In the case of CEID electrodes, the locations of electrodes within a patient’s heart may be recorded when the device and electrodes are implanted in the patient and / or determined prior to a procedure (e.g., an EP procedure and / or ablation procedure) using medical imaging, such as computed tomography (CT), that images both the heart and electrodes. Techniques to precisely identify the location of each electrode in a CEID are described herein.

[0168] In the case of EP catheter electrodes that are temporarily inserted in the patient’s heart during an EP procedure, medical imaging may be limited to fluoroscopy. Accordingly, there are generally two existing methodologies employed in EP systems to localize EP catheter electrodes positioned within a patient’s heart during intracardiac procedures. The first of these methodologies relies on magnetic field-based localization techniques. In such magnetic-based systems, one or more magnets or electromagnetic coils positioned externally relative to the patient’s body generate controlled magnetic fields. Conversely, the second methodology utilizes impedance-based electro-resistance localization, wherein surface electrodes externally affixed to the patient generate electric fields within the patient's thoracic cavity, and the resulting induced electrical signals are measured to ascertain the intracardiac electrode locations during intracardiac procedures (e.g., EP procedures).

[0169] Magnetic field-based localization operates by employing multiple magnetic field generators, typically arranged in a fixed configuration external to the patient's body. In practical implementations, such as Biosense Webster’s CARTO™ system, a pad containing at least three magnetic coils is placed beneath the patient, wherein each coil emits a unique ultralow magnetic field. EP catheters used in these intracardiac procedures incorporate miniature magnetic sensors, generally placed in or near the EP catheter tip, to measure magnetic field strength. The strength of each magnetic field measured at the sensor is inversely proportional to the distance between the sensor and each coil. By processing the measured magnetic field strengths from each coil, the 15448153-1 36Attorney Docket No.: 35313-0011WOsystem calculates the three-dimensional coordinates of the EP catheter tip via triangulation or trilateration techniques. In advanced implementations, EP catheters are equipped with multiple orthogonally oriented sensors, enabling the system to additionally determine the orientation and rotational positioning of the catheter, thereby providing six degrees of positional freedom.

[0170] In contrast, impedance-based electro-resistance localization methods rely on the creation of transient, low-amplitude electric fields within the patient's body using externally applied electrode patches. These patches, typically arranged in three orthogonal pairs positioned around the patient (for example, along lateral, anterior-posterior, and superior-inferior axes), sequentially emit an alternating current at a predetermined frequency, commonly around 8 kHz, thus establishing known voltage gradients within the body along each defined axis. Intracardiac catheter electrodes located within these gradients detect induced voltages whose magnitudes correlate directly with their positions along the gradients. By sequentially measuring the induced voltage signals from each orthogonal axis, the system computes the three-dimensional location of each electrode through interpolation or triangulation of voltage gradients. An advantage of this electro-resistance localization method is that it facilitates simultaneous tracking of multiple catheter electrodes since each electrode independently senses its local electric field strength.

[0171] Recent advanced embodiments of these EP catheter localization systems employ a hybrid approach integrating both magnetic and impedance-based localization methods to maximize positional accuracy and reliability. Specifically, the CARTO® 3 system from Biosense Webster, traditionally based on magnetic localization, now incorporates impedance-based localization capabilities utilizing surface electrode patches similar to those used by Abbott’s EnSite™ system, allowing the 3D mapping system to localize standard catheters (e.g., EP catheters) without embedded magnetic sensors. Conversely, Abbott’s EnSite™ Precision system, which traditionally relied on impedance-based localization, incorporates a magnetic field generator located beneath the patient to complement its impedance measurements.15448153-1 37Attorney Docket No.: 35313-0011WOSpecially adapted Sensor Enabled™ catheters containing magnetic sensors are utilized to provide additional real-time magnetic location data. This integration allows the EnSite™ Precision system to continuously calibrate and correct impedance-based localization measurements by referencing highly accurate magnetic sensor-derived positions, thus significantly enhancing the stability and accuracy of EP catheter electrode localization during intracardiac electrophysiology procedures.

[0172] In addition to locating the sensing and / or pacing electrodes, whether ECG, EP catheter, or CEID on the patient, so as to localize arrhythmia, the computing system may need to translate and transform the location coordinates of the electrodes to align the location with the 3D heart model that is selected and modified by the computing system to match the patient’s heart. Aligning the coordinates of electrodes to (e.g., a reference frame of the 3D heart model) enables the computing system to precisely define the location of the electrodes within the 3D heart model. This enables generation of a 3D activation map produced in various embodiments that can be used to guide a physician to an arrhythmia (e.g., VT) initiate site for ablation. In particular, the ability to precisely define the location of the electrodes within the 3D heart model may enable the computing system to precisely localize arrhythmia origin. For ECD electrodes imaged on the patient’s thorax, such alignment may be based on image data from multiple cameras or multiple angles in combination with dimensional and positional information regarding the patient’s heart gathered from medical imaging.

[0173] Framework to Align Location of Electrodes on an EP Catheter within a 3D Heart Model During an Intracardiac Procedure

[0174] In the case of EP catheter electrodes that are temporarily inserted in the patient’s heart during an intracardiac procedure (e.g., EP procedure), the computing system may translate the electrode position coordinates to a reference frame of the heart using existing coordinate transformation methods. For example, both the EP 3D mapping systems such as Biosense Webster’s CARTO® system and Abbott’s EnSite™ system utilize patient-specific coordinate systems to accurately represent catheter electrode positions within the heart. These coordinate systems are established 15448153-1 38Attorney Docket No.: 35313-0011WOrelative to the patient’s anatomical features to maintain consistent referencing throughout procedures.

[0175] The CARTO® system defines its coordinate frame based on the fixed arrangement of multiple external magnetic field generators, typically integrated into a locator pad placed beneath the patient. To anchor this reference frame to patient anatomy, CARTO® employs external body patches with embedded magnetic reference sensors positioned at predetermined locations, such as the patient’s chest or back. As EP catheter electrode positions are computed based on magnetic field measurements relative to the locator pad, concurrent measurement of the reference sensors’ positions allows the CARTO® system to define the catheter coordinates in relation to the patient’s body rather than to a fixed laboratory frame. Consequently, the CARTO® coordinate axes align anatomically (e.g., left-right, anterior-posterior, cranial-caudal), and the system dynamically compensates for patient movement or respiration by continuously tracking shifts in the reference sensor positions.

[0176] The EnSite™ system employs a coordinate system defined by externally positioned electrode patches placed directly on the patient’s skin. These patches, arranged in orthogonal pairs along the patient’s lateral (left-right), anterior-posterior (chest-back), and superior-inferior (head-foot) axes, sequentially generate electrical fields to form voltage gradients that intersect centrally within the patient’s thoracic region. This intersection point typically defines the origin of the EnSite™ coordinate system, from which electrode positions are computed and reported, typically in millimeters. Catheter electrode locations within the generated fields are determined by measuring voltage signals induced at each electrode along these gradients. Dedicated reference sensors, separate from the orthogonal patch electrodes and sometimes including embedded magnetic sensors, are affixed to specific stable locations on the patient’s torso. Data from these reference sensors enable the system to monitor and correct positional shifts due to patient movement or physiological variations, thereby maintaining stable alignment of the catheter coordinates to patient anatomy throughout the procedure.15448153-1 39Attorney Docket No.: 35313-0011WO

[0177] In the latest iterations, both CARTO® and EnSite™ have incorporated hybrid methodologies that leverage both magnetic and impedance-based localization techniques to optimize the stability and accuracy of their patient-specific coordinate systems. CARTO® enhances its magnetic localization capability by integrating impedance-based methods via external electrode patches, thus accommodating standard catheters lacking magnetic sensors. Conversely, EnSite™ augments its traditional impedance-based coordinate system by adding a magnetic field generator placed beneath the patient, combined with specialized catheters incorporating magnetic sensors. Both systems utilize these dual-reference frameworks to provide continuous calibration, improved spatial accuracy, and real-time compensation for patient movements, effectively merging the benefits of magnetic precision with the versatility and adaptability of impedance-based localization.

[0178] The CARTO® and EnSite™ EP 3D mapping system achieve alignment of EP catheter electrode positions with the patient’s cardiac anatomy primarily by generating real-time, three-dimensional geometric representations of the cardiac chambers. During an intracardiac procedure, the EP catheter electrodes are navigated within the heart chambers, and their positions at contact points on the endocardial surfaces are recorded. Collecting numerous such data points creates a three-dimensional surface geometry, commonly referred to as a "shell," representing the internal contours of the chamber. To maintain anatomical accuracy, the systems apply motion compensation techniques to reduce artifacts caused by respiratory movement and cardiac cycle variations. These techniques typically involve gating the data acquisition process to specific phases of the cardiac or respiratory cycles, or employing signal averaging methods, thus stabilizing the resulting geometry and ensuring that electrode coordinates accurately reflect anatomical locations.

[0179] Electrode location accuracy may be refined by using heart structure landmarks obtained prior to the procedure through CT, magnetic resonance imaging (MRI), intracardiac echocardiography (ICE), and fluoroscopy. CARTO ’s image integration software, CARTOMERGE®, performs a rigid registration of pre-acquired15448153-1 40Attorney Docket No.: 35313-0011WOCT or MRI cardiac images to the catheter-generated anatomical map using corresponding fiducial landmarks selected by the operator. EnSite’s corresponding system employs a two-step registration process, first performing a rigid alignment based on fiducial points, followed by an optional dynamic, non-rigid alignment to locally adjust the map geometry, thereby correcting minor anatomical discrepancies.

[0180] The CARTO® and EnSite™ 3D mapping systems utilize coordinate transformation algorithms to integrate and harmonize positional data from different measurement modalities and imaging sources into a unified anatomical coordinate system. Internally, both systems continuously align impedance-based measurements with magnetic sensor-based reference data through linear transformations that correct for variations such as scale differences or positional drift, thereby ensuring consistent and accurate electrode localization. For integration with external imaging, both systems initially employ a rigid transformation based on anatomical landmarks to register the catheter-generated maps with pre-acquired anatomical images from modalities such as CT or MRI. CARTO® typically employs only a global rigid transformation for this image registration. EnSite™ further enhances alignment accuracy by subsequently applying non-rigid, local deformation algorithms that adapt the electroanatomic geometry more closely to specific anatomical contours identified in imaging, thereby optimizing correspondence between catheter positions and patient anatomy.

[0181] Through the combinations of electrode localizing techniques and coordinate transformation operations, the latest versions of the CARTO® and EnSite™ electrophysiology 3D mapping systems achieve localization accuracy generally on the order of one millimeter or less. Specifically, the CARTO® system has demonstrated submillimeter precision, with typical reported accuracies around 0.5 mm under optimal conditions. CARTO® maintains this accuracy through continuous real-time calibration, employing reference sensors on the patient to detect and correct for patient movements and respiration-induced shifts. Abbott’s EnSite™ Precision15448153-1 41Attorney Docket No.: 35313-0011WOsystem has achieved accuracy levels comparable to CARTO®, typically reported as less than 1 mm error under stable clinical conditions.

[0182] The accuracies in EP catheter electrode locations now achievable with 3D mapping EP systems such as CARTO® and EnSite™ enable the computing system to track the timing of electrical pulses associated with beats and tribulation at precise locations on the inside of the heart, thereby improving the accuracy of the 3D heart conduction map the computing system can produce.

[0183] Framework to Align Location of Intracardiac Electrodes within a 3D Heart Model Prior to an Intracardiac Procedure

[0184] To enable locating initiation sites of VT events, the recording of electrical signals in the heart by intracardiac electrodes should occur during a VT event. As an CEID is implanted permanently in the patient, the CEID processor and memory can record electrical signal data from the CEID electrodes whenever a VT event occurs. This data can then be downloaded and processed when the patient is prepared for an ablation procedure. To enable the recording of EP data using temporarily inserted EP catheters, the patient’s heart may be induced into VT through electrical stimulation by one of the electrodes, such as to induce a series of irregular or fast heartbeats.

[0185] It may be desirable to align the location of intracardiac electrodes, and in particular, CEID electrodes within a 3D heart model prior to the intracardiac procedure (e.g., EP procedure). In particular, it may be desirable to identify and define the location of the CEID electrodes within the 3D heart model precisely.

[0186] Accordingly, various embodiments described herein include methods and systems for localizing and treating cardiac arrhythmias using EGM data from CEID, based on a multi-lead internal electrocardiographic framework (MLiECG framework), combined with advanced three-dimensional (3D) heart modeling and simulation techniques. The term “lead” here refers to the “vector lead” that indicates the directional path of the electrical signal in the heart between two electrodes that are disposed in the heart at least some distance away from each other, e.g., as nonlimiting15448153-1 42Attorney Docket No.: 35313-0011WOexample, 5 mm or more or even across heart chamber(s), as shown in FIG. 11A and will be referred to as “vector lead” to be clearly distinguished from the implantable lead of CEID or implantable lead.

[0187] In some embodiments, MLiECG framework may comprise a first vector lead (MLiECG lead I) between a first electrode (RA electrode or RAtip) disposed in the right atrium and a second electrode. The second electrode may be agenerator electrode or CEID-Can that may be disposed in the CEID generator or its header or alternatively the generator itself may act as the second electrode. The MLiECG framework may comprise a second vector lead (MLiECG lead II) between the first electrode ( RA electrode or RAtip), and a third electrode. The third electrode may be disposed in the right ventricle (RV electrode or RVtip), typically near the apex or intraventricular septum away from the ventricular base. The MLiECG framework may further comprise a third vector lead between the second electrode(generator electrode or CEID-Can), and the third electrode(RV electrode or RVtip). It should be readily understood that the designation of RAtip and RVtip electrodes reflect the distal disposition of the electrodes in their respective implantable leads and that the electrodes may be disposed as part of the active fixation mechanism at the very tip of the implantable lead or a fraction of millimeters or one or a few millimeters away from the very tip of the implantable lead. Furthermore, more than one electrode may be disposed distally in the implantable lead.

[0188] In some embodiments, a bipolar internal vector lead triangle is formed by the first, second and third vector leads (MLiECG leads I, II and III, FIG. 11D. The polarities of these vector leads may be defined as the following:

[0189] MLiECG lead I is RA electrode or RAtip (negative, -) to generator electrode or CEID-Can (positive, +),

[0190] MLiECG lead II is RA electrode or RAtip (-) to RV electrode or RVtip (+),

[0191] MLiECG lead III is generator electrode or CEID-Can (-) to RV electrode or RVtip (+)15448153-1 43Attorney Docket No.: 35313-0011WO

[0192] Hence, the EGM signals in MLiECG lead I, II and III have positive deflections when electrical activity moves towards the positive electrode and negative deflection in the opposite direction. It should be understood that the polarities of the vector leads defined above are meant to facilitate the standardization of MLiECG framework. Alternatively and additionally, the polarities of the vector leads may be defined in reverse (e.g., for one vector lead, or more than one vector lead or all of the vector leads) so long as these polarities are pre-defined and used in a consistent and invariable maimer.

[0193] In some embodiments, three augmented vector leads may be obtained from the first, second and third vector lead using a central terminal derived from two of first, second and third electrodes as the negative pole and remaining of the first, second and third electrodes as the positive pole.

[0194] aVRA = RAtip – (CEID-Can + RVtip) / 2,

[0195] aVGN = CEID-Can – (RAtip + RVtip) / 2, and

[0196] aVRV = RVtip – (RAtip + CEID-Can) / 2

[0197] In some embodiments, a virtual central terminal (VCT) is calculated by averaging the signals from CEID-Can, RAtip and RVtip, i.e,

[0198] VCT = (RAtip + RVtip + CEID-Can) / 3

[0199] In some embodiments, VCT provides the neutral reference point for unipolar EGM recordings, for example, from the electrodes disposed in the coronary sinus or from cardiac resynchronization therapy (CRT) leads in conjunction with the actively driven ground electrode (as described below), to reduce electrical interference noises.

[0200] Similarly, in some embodiments, there are may be additional recording electrodes (e.g., unipolar and / or bipolar). Such electrodes may be disposed in the implantable CEID leads and / or in the EP catheters during invasive EP procedures. These electrodes may include electrodes such as those disposed in the cardiac re synchronization therapy or coronary sinus leads and left bundle or His bundle pacing leads. Furthermore, such electrodes may include the proximal and distal15448153-1 44Attorney Docket No.: 35313-0011WOshocking coils, in the implantable leads placed for example, in the right atrium and ventricle. Such additional unipolar and / or bipolar recording electrodes can be configured into the MLiECG framework.

[0201] In some embodiments, the MLiECG framework may use an additional electrode. This electrode may be disposed at the proximal portion of the implantable RV lead, in the header of the CEID can or in the can itself. In some embodiments, this electrode may be the proximal ICD coil (if present) that serves as ground electrode or reference electrode and may be used to reduce common-mode noises (FIG. 11 A electrode 1157). It should be noted that other optional or existing electrodes disposed in the implantable CEID leads may serve the same function as a ground electrode which may serve as a passive ground reference or may be actively driven by an active circuitry similar to the driven right leg circuit for body surface electrocardiography.

[0202] In some embodiments, the MLiECG vector system may use an electrode model that integrates multiple points over the geometry of the CIED coil (e.g. in the RV blood cavity). When this coil is used as one of the two sensing electrodes, it electrically integrates the signal over the entire conducting region of the coil. When simulating electrograms involving the CIED coil, a number of electrograms may be simulated at a plurality of points that represent the geometry of the coil, which may be determined from imaging data and / or CIED device specifications, and integrated (or averaged) to simulate an average electrogram that better represents the electrical aspects of the device. When used in a simulated vector lead, the integrated coil signal may be converted to a single coil signal, and used as one of a pair of electrodes in simulated EGMs.

[0203] In some embodiments, further additional electrodes disposed in the implantable CEID lead(s) may be incorporated into the MLiECG framework. This may increase the accuracy of predicting arrhythmia origin. Such additional electrodes may be configured as unipolar electrode, such as for example, those disposed in the cardiac resynchronization therapy (CRT) or coronary sinus implantable lead (as noted above). In particular, these additional electrodes may include an additional electrode15448153-1 45Attorney Docket No.: 35313-0011WOdisposed near or at the coronary sinus (FIG. 11A electrode 1155). Additionally, these additional electrodes may be configured with any one of the other existing electrodes to form bipolar recording, either to record far- field EGMs when the paired electrodes are placed far from each other ( e.g., a fraction of a centimeter, a few centimeters apart across a large region of myocardium or even a heart chamber such as across the atrioventricular annulus or interventricular septum), or to record near-field EGMs when the paired electrodes are spaced a few millimeters apart from each other.

[0204] Alternatively, the MLiECG framework may be modified to have one of the electrodes disposed in the CIED as a reference electrode that weights all EGM from other internal electrodes equally. As a nonlimiting example, a dual chamber ICD comprises a generator can (CEID-Can) and electrodes that can be configured to record multiple bipolar EGMs including: (1) high energy coil in right ventricle (RVcoil) to distal electrode in the right atrial implantable lead (RAtip) or RVcoil-RAtip, (2) CEID-Can in the pectoral region (Can) to distal electrode disposed in the right ventricular implantable lead (RVtip) or CEID-Can - RVtip, and (3) CEID-Can -RVcoil. These vector leads may be converted to reference the CEID-Can electrode as follows:

[0205] RVtip - CEID-Can = - (CEID-Can - RVtip)

[0206] RAtip - CEID-Can = - (RVcoil - RAtip) - (CEID-Can - RVcoil)

[0207] RVcoil - CEID-Can = - (CEID-Can - RVcoil)

[0208] Alternatively, a “central reference” (CR) may be computed that is the center of another set of three bipolar vector leads by taking the mean of the three bipolar vector leads referenced to CEID-Can:

[0209] CR = mean[(RVtip - CEID-Can), (RAtip - CEID-Can),(RVcoil - CEID-Can)]

[0210] Bipolar vector leads with this reference point may then be computed respectively as:15448153-1 46Attorney Docket No.: 35313-0011WO

[0211] RVtip - CEID-Can - CR

[0212] RAtip - CEID-Can - CR

[0213] RVcoil - CEID-Can - CR

[0214] In some embodiments, in addition to its effect on noise reduction and signal stabilization, CR may be used to enable unipolar recordings. This process must be performed for both the measured and simulated EGMs (as further described herein). These vector leads referenced to CR may be used for both identification of the search area and for comparing the simulated and measured EGMs (as further described herein). This central reference configuration has the advantage of minimal hardware modification or no hardware modification on the existing CEIDs while maximizing the accuracy in predicting arrhythmia origin.

[0215] In some embodiments, it can be advantageous to align the MLiECG framework to 3D heart models. More specifically, it can be advantageous to anatomically align the MLiECG framework with the patient specific 3D heart models generated from medical imaging data, e.g., from imaging systems such as for example, cardiac CT, MRI, PET scans, a combination thereof, and / or the like.Anatomically aligning the MLiECG framework with patient specific 3D heart models enables the locations of the electrodes to be specified in their respective unique and precise coordinates within the specific 3D heart models. This in turn enables accurate localization of arrhythmia origin. In some variations, such alignment can be performed based on the specific anatomical features of the 3D heart model where such anatomic features may include the mitral annulus plane with its anteroposterior line (AP line) and inter-commissural line (IC line) defined as the x and y axis, respectively, and the perpendicular line passing the interception point between the AP and IC lines as the z axis (FIG 1 IB). Alternatively, the long axis (LA line) of the mitral annulus may serve as the x axis and its perpendicular line through the mid-point on the LA line may serve as the y-axis. Again, the perpendicular line passing the midpoint may serve as the z-axis (FIG 1 IC). These coordinate systems built upon patient specific 3D heart model are non-limiting examples to demonstrate the approach to15448153-1 47Attorney Docket No.: 35313-0011WOgenerate a coordinate system to enable precise location of electrodes within the 3D heart models so as to enable accurate prediction of arrhythmia origins. In this maimer, the MLiECG framework may be aligned with patient specific 3D heart models. Note, unlike most of the electrodes in the MLiECG framework, the precise locations and consequently the x, y z coordinates of the coil electrodes in the ICD leads and CEID-Can electrode can be defined by their respective center point or centroid as the best approximation because of their large physical dimensions in comparison to the other electrodes disposed in the implantable CEID leads.

[0216] Patient-Specific 3D Heart Models

[0217] Various embodiments include methods and systems for localizing and treating cardiac arrhythmias using CEID EGM data combined with advanced three-dimensional (3D) heart modeling and simulation techniques. Various embodiments utilize patient-specific 3D heart models generated from medical imaging data (e.g., imaging data received from imaging systems, such as for example, CT, MRI, PET, a combination thereof, and / or the like). Various embodiments further integrate CEID EGM data based on the MLiECG framework and subsequent anatomical alignment described above to simulate cardiac electrical activity and identify the likely origin of arrhythmias.

[0218] In some embodiments, the 3D heart model may be represented using boundary element method (BEM), finite element method (FEM), or a hybrid element method (HEM) that combines both BEM and FEM.

[0219] In some embodiments, electrical potentials generated by myocardial activation may be computed using BEM. There are existing methods for BEM modeling of cardiac electrical potentials or simulated EGMs. However, in contrast to the existing methods, the present disclosure employs BEM based computation in conjunction with CEID EGM data that has been integrated based on the MLiECG framework and anatomical alignment, thereby enabling precise localization of cardiac arrhythmia origin. In one embodiment, electrical sources associated with myocardial15448153-1 48Attorney Docket No.: 35313-0011WOactivation, for example equivalent dipole layer (EDL) sources defined across cardiac surfaces, may be used with BEM to compute electrical potentials at locations corresponding to CEID electrodes. The resulting simulated CEID EGMs (e.g., simulated CEID EGM data that has been integrated based on MLiECG framework and anatomical alignment) may be compared with measured or recorded CEID EGMs (e.g., recorded CEID EGM data that has been integrated based on MLiECG framework and anatomical alignment) during VT for arrhythmia localization using techniques such as, comparison between EGM-derived vectors and simulated anatomical activation vectors, morphology matching between simulated and measured or recorded electrograms, or a combined method integrating both vector similarity and electrogram morphology similarity.

[0220] In some embodiments, the heart model may be represented or discretized as a finite element mesh. There are existing methods for FEM-based computation of cardiac electrical potentials or simulated EGMs. However, in contrast to the existing methods, the present disclosure employs FEM modeling in conjunction with CEID EGM data that has been integrated based on MLiECG framework and anatomical alignment, thereby enabling precise localization of cardiac arrhythmia origin. In one embodiment, electrical activation propagation from candidate origin locations within the myocardium may be simulated using the cardiac FEM model, thereby generating activation times and corresponding activation isochrones representing propagation of the depolarization wavefront. Electrical sources associated with myocardial depolarization, such as, for example, activation-based volumetric (ABV) source model, in which the source is defined as a focal activation site within the myocardial volume, and / or equivalent dipole layer (EDL) sources model defined on cardiac surfaces, may be used in conjunction with the FEM model to compute electrical potentials or simulated EGMs at locations corresponding to CEID electrodes. The resulting simulated CEID EGMs (e.g., simulated CEID EGM data that has been integrated based on MLiECG framework and anatomical alignment) may then be compared with measured / recorded CEID EGMs (e.g., recorded CEID EGM data that15448153-1 49Attorney Docket No.: 35313-0011WOhas been integrated based on MLiECG framework and anatomical alignment) during VT for arrhythmia localization using techniques such as, comparison between EGM-derived vectors and simulated anatomical activation vectors, morphology matching between simulated and measured electrograms, or a combined method integrating both vector similarity and electrogram morphology similarity.

[0221] In some embodiments, the 3D heart model may be represented using a hybrid FEM-BEM model. There are existing methods for hybrid FEM-BEM modeling of bioelectric fields. However, in contrast to the existing hybrid FEM-BEM modeling methods, the present disclosure employs hybrid FEM-BEM modeling in conjunction with CEID EGM data that has been integrated based on MLiECG framework and anatomical alignment, thereby enabling precise localization of cardiac arrhythmia origin. In such embodiments, the myocardium may be represented using a FEM model that accounts for anisotropic electrical conduction within the myocardial tissue, while the surrounding torso and blood volume may be represented using a BEM model. This hybrid FEM-BEM model enables efficient computation of electrical potentials throughout the cardiac and torso volume conductor while preserving the ability of the FEM model to represent anisotropic conduction within the myocardium. In such configurations, sensing electrodes associated with a CEID may be incorporated into the volume conductor model, and electrical potentials at the electrode locations may be evaluated using the simulated activation sequence, the activation-based volumetric (ABV) source model, the EDL source model, and the hybrid FEM-BEM model, thereby generating simulated CEID EGMs corresponding to the implanted electrodes. These simulated EGMs (e.g., simulated EGMs that has been integrated based on MLiECG framework and anatomical alignment) may then be analyzed together with measured CEID EGM data (e.g., measured EGM data that has been integrated based on MLiECG framework and anatomical alignment) during VT for localization of cardiac arrhythmia origin using techniques such as, comparison between EGM-derived vectors and simulated anatomical activation vectors, morphology matching between simulated and measured electrograms, or a combined15448153-1 50Attorney Docket No.: 35313-0011WOmethod integrating both vector similarity and electrogram morphology similarity.

[0222] In some embodiments, the patient-specific 3D heart models comprise elements of myocardial tissue whose tissue characteristics may determine the conduction velocities across the myocardial elements. This must be accounted for during EGM simulation (as noted below). Such tissue characteristics may include the ventricular wall thickness and, as such, may influence the conduction velocity across the myocardial elements. In some embodiments, generating the simulated CEID EGMs that are integrated with the MLiECG framework and anatomical alignment described herein may further comprise computing a “distance matrix” that defines conduction velocity between any two vertices of a patient-specific 3D model (including heart and torso) (e.g., during the simulation of EGMs and / or depolarization waves). This distance matrix may assume anisotropic conduction through the ventricular myocardium in order to effectively model heterogeneous orientation and electrical conduction of cardiomyocytes. Specifically, vertex pairs belonging to the heart geometry and having a transmural connection (i.e. shortest path through the ventricular wall) are assigned a lower conduction velocity than other vertex pairs belonging to the heart geometry. In some embodiments, the thickness of the left ventricle (LV) walls may be used to further adjust conduction velocity. During creation of the 3D heart model, a user may identify scar (myocardial fibrosis) from imaging data (e.g., a CT image) and set the left ventricular wall thickness (LVWT) near the scar to a small value (e.g. < 5 mm), whereas healthy hearts may have LVWT between 5 and 11 mm. For example, a suitable application executed on the computing system may display the wall thickness measure to the user when creating the 3D heart model. Thus, in the present disclosure, when the distance matrix is calculated, the wall thickness measure may be utilized to automatically set the conduction velocity for vertex pairs within the smaller LVWT region to a lower conduction velocity (e.g.1 / 10 that of healthy tissue). This LVWT-adjusted distance matrix may then be utilized to simulate electrical potentials that account for scar. As noted above, such simulated electrical potentials are integrated with the MLiECG framework and15448153-1 51Attorney Docket No.: 35313-0011WOanatomical alignment described herein. Note the LVWT threshold at which to set a lower conduction velocity, and the conduction velocity value that is to be used may be refined based on clinical data. In some embodiments, this LVWT may be used alone or in combination(s) with additional tissue characteristics such as extent of edema, myocardial orientation and metabolic states as nonlimiting examples, the adiust the distance matrix.

[0223] EGM Simulations and Analyses

[0224] In some embodiments, localization of a cardiac arrhythmia origin may be performed by comparing an EGM-derived vector computed from measured CEID EGMs with an anatomical vector derived from simulated cardiac activation isochrones representing propagation of the myocardial depolarization wavefront. In some embodiments, the CEID EGMs may be integrated or aligned with the MLiECG framework and the anatomical vector may be derived based on simulated EGM integrated or aligned with MLiECG framework. In contrast to existing surface ECGbased approaches for localization of cardiac arrhythmia origin, the present disclosure obtains the EGM-derived vector from CEID EGM data that are analyzed within the MLiECG framework, thereby enabling localization based on CEID EGM data in reference to anatomic features of the heart instead of the features of the torso.Electrodes used for acquiring the CEID EGM data may include, for example, electrodes located in the right atrium (RA), right ventricle (RV), left ventricle (LV), coronary sinus (CS), or a device generator electrode (CAN). The measured EGMs from these electrodes may be referenced to a virtual reference electrode, thereby generating referenced CEID EGMs for subsequent signal analysis and arrhythmia localization within the MLiECG framework.

[0225] In one embodiment, a set of referenced EGM Et(t) may be obtained from N sensing electrodes according to^(0 = K(0 - Vref^15448153-1 52Attorney Docket No.: 35313-0011WOwhere Vj(t) represents the EGM measured at electrode i, and Vref(t) represents the virtual reference EGM derived from a combination of implanted electrode EGMs. The referenced signals Et(t) therefore represent unipolar CEID EGMs referenced to the virtual reference electrode.

[0226] In some embodiments, an EGM-derived vector may be computed from CEID EGM measurements obtained from multiple sensing electrodes. The EGM- derived vector may represent a dominant direction of cardiac electrical activation and may be used for localization of the cardiac arrhythmia origin.

[0227] In one embodiment, an EGM-derived vector VEGM(t) may be calculated using the referenced EGM signals and the spatial positions of the sensing electrodes according toN V EGM (0 — (0i=lwhere Et(t) represents the referenced EGM measured at electrode i, and rtrepresents a unit vector extending from a candidate cardiac origin location toward the spatial position of electrode i.

[0228] In parallel, a cardiac activation model may be used to simulate activation propagation within a patient-specific 3D heart model. For each candidate origin location within the 3D heart model, activation times may be computed and isochrone surfaces representing the propagation of the activation wavefront may be determined. From these isochrones, an anatomical vector VanatOmy(t) representing the dominant propagation direction of the activation wavefront may be derived.

[0229] The arrhythmia origin may then be estimated by comparing the direction of the EGM-derived vector VEGM(t) with the anatomical vector Vanatomy(t) derived from simulated activation isochrones. In one embodiment, the similarity between the two vectors may be quantified by computing the angle between the vectors:.. _ -1 | ^EGM(0 ’ ^anatomyCO \XII^EGMC II ||^ anatomyC 11 / 15448153-1 53Attorney Docket No.: 35313-0011WO

[0230] Candidate origin locations producing the smallest time-dependent angle, cr(t), between the EGM-derived vector and the anatomical vector may be identified as the most likely origin of the arrhythmia, wherein the angle a(t) is evaluated at one or more selected time points during the depolarization interval, and the candidate origin is determined based on a combined measure of the angles across the selected time points.

[0231] In some embodiments, localization of a cardiac arrhythmia origin may be performed using CEID EGM data. For each candidate arrhythmia location, an electrical source model, including but not limited to an ABV model and / or an EDL model, may be used to simulate myocardial electrical activity associated with depolarization. Using the patient-specific 3D anatomic model and electrode locations, simulated EGMs may be computed at the positions of the sensing electrodes. The simulated EGMs (e.g., simulated EGMs that has been integrated based on MLiECG framework and anatomical alignment )may then be compared with the measured or recorded CEID EGMs (e.g., recorded CEID EGMs that has been integrated based on MLiECG framework and anatomical alignment) during VT using one or more morphology similarity metrics, including but not limited to cross-correlation or other waveform similarity measures. Based on this comparison, a similarity score may be determined for each candidate location, and the candidate location producing simulated EGMs exhibiting the greatest similarity with the measured EGMs may be identified as the estimated origin of the arrhythmia.

[0232] In some embodiments, the vector comparison method and the electrogram morphology matching method described above may be combined to improve localization accuracy. For each candidate arrhythmia location i, a vector similarity measure and a morphology similarity measure may be computed according to the respective methods described herein. The two measures may then be combined to produce a composite localization score. In one embodiment, the composite score may be computed according toJi morph + ^2^i,vec15448153-1 54Attorney Docket No.: 35313-0011WOwhere Si morphrepresents the morphology similarity score between measured and simulated CEID EGMs for candidate location i, Si vecrepresents the vector similarity score between the EGM-derived vector and the anatomical vector, and w1and w2are weighting coefficients. The candidate location producing the optimal composite score i may be identified as the estimated origin of the arrhythmia.

[0233] In some embodiments, electrical measurements used for arrhythmia localization may include CEID EGMs and / or body surface ECG signals recorded from electrodes positioned on the torso of the patient. Such electrodes may be located outside the myocardial computational domain used to represent the myocardial tissue.

[0234] In one embodiment, the myocardium may be represented using a FEM model. In such configurations, body surface electrodes may be treated as observation locations having known spatial coordinates relative to the myocardial model, and the electrical potentials at the surface electrodes may be computed using forward modeling methods including, but not limited to, direct source evaluation.

[0235] In another embodiment, a hybrid FEM-BEM volume conductor model may be employed. In such embodiments, the myocardium may be represented using a FEM model that accounts for anisotropic electrical conduction within myocardial tissue, while the surrounding conductive media, including blood cavities and torso structures, may be represented using a BEM model. This hybrid FEM-BEM model enables computation of electrical potentials throughout the cardiac-torso volume conductor and may provide improved accuracy for electrodes located outside the myocardium, including body surface ECG electrodes.

[0236] The simulated electrical signals at the electrode locations may then be compared with corresponding measured signals, including CEID EGMs and / or bodysurface ECG signals, and the origin of the cardiac arrhythmia may be estimated based on the similarity between the simulated and measured signals.15448153-1 55Attorney Docket No.: 35313-0011WO

[0237] Prediction of Arrhythmia Origin

[0238] Various embodiments described herein obtain electrical signals and location information from electrodes positioned within the patient’s heart. A patientspecific 3D heart model may be generated based on medical imaging data received from imaging systems. A patient-specific electrical conduction map of the patient’s heart during arrhythmia may be generated based on the patient-specific 3D heart model and the electrical signals and location information from the electrodes positioned within the patient’s heart. The patient-specific electrical conduction map of the patient’s heart may identify a 3D localization of cardiac arrhythmia origin.

[0239] Some embodiments described herein process CEID EGMs to select arrhythmia beats, determine local activation times for each CEID electrode, and simulate time-shifted activation waves through the 3D heart model. By analyzing the intersection of isochrones at t=0 from these simulated waves, embodiments described herein identify a likely arrhythmia origin area. The embodiments then simulate EGMs from multiple locations within this area and compares them to the recorded CEID EGMs to pinpoint the most likely point of origin for the arrhythmia. This innovative approach offers significant improvements over existing methods for localizing cardiac arrhythmias, particularly ventricular tachycardia (VT) and premature ventricular contractions (PVC). By leveraging CEID data that is continuously recorded to capture EGMs during arrythmia events, even when arrhythmias are not present during clinical procedures, embodiments described herein provide more accurate and reliable localization of arrhythmia sources. This can lead to more efficient and effective cardiac ablation procedures, potentially reducing procedure duration, minimizing unnecessary tissue damage, and improving patient outcomes.

[0240] Furthermore, in some embodiments, the measured or recorded CEID EGMs may be integrated with the MLiECG framework and aligned with anatomical features of the heart to precisely locate the electrodes within the 3D heart model. CEID EGMs may be simulated at electrode locations and the simulated EGMs may be integrated with the MLiECG framework and subsequent anatomical alignment. The simulated15448153-1 56Attorney Docket No.: 35313-0011WOEMGs may be compared with the recorded CEID EGMs to precisely localize cardiac arrhythmias. The integration of the recorded and simulated CEID EGMs with the MLiECG framework and the anatomical alignment enables the embodiments described herein to precisely define the location of the electrodes within the 3D heart model. This in turn improves the accuracy of localizing cardiac arrhythmia origin.

[0241] Example Systems and Methods

[0242] In some embodiments, a computing system 100 for processing and analyzing cardiac data may include a computing system 102, as shown in FIG. 1. The computing system 102 may include a processing system 110 coupled to electronic storage 806 for storing data and a network transceiver 108 for communicating with other medical devices and systems.

[0243] The processing system 110 may be configured with processor-executable instructions 112, which may include modules such as for example, modules 114-124, to perform various functions including generating and processing cardiac modeling data and identifying cardiac arrhythmia origin.

[0244] An intracardiac electrode data download module 114 may include instructions to cause the processing system 110 to perform operations including receiving and processing data downloaded from EP systems coupled to intracardiac EP catheters during an EP procedure and / or downloaded from a CEID (e.g., an ICD). The intracardiac electrode data download module 114 may include instructions for obtaining or downloading the electrical signal data from EP systems, an ICD, or both. In addition to downloading electrical signal data, the intracardiac electrode data download module 114 may include instructions to cause the processing system 110 to receive information from an EP system regarding the location of each EP catheter electrode within the heart.

[0245] A medical imaging download module 116 may include instructions to cause the processing system 110 to perform operations including accessing medical imaging data for a patient, such as receiving image data from a medical imaging15448153-1 57Attorney Docket No.: 35313-0011WOsystem (e.g., X-ray, fluoroscopy, MRI, CT, etc.) and / or downloading image data from a medical records database (e.g., a hospital medical records system).

[0246] The processing system 110 may also be configured with instructions from a patient heart model generator module 118. The patient heart model generator module 118 may include instructions to cause the processing system 110 to perform operations including creating patient-specific heart models using data downloaded from a patient’s ICD by the intracardiac electrode data download module 114 and patient medical image data obtained by the medical imaging download module 116.

[0247] An electrical conduction map generator module 120 within the processorexecutable instructions 112 may include instructions to cause the processing system 110 to perform operations including producing maps of electrical activity in the heart based on ICD data and in some embodiments, on electrocardiogram (ECG) data received from an ECG system 132. This map may be based on a heart model and the ICD information recorded during detected arrhythmia events.

[0248] An ablation guidance module 122 within the processor-executable instructions 112 may include instructions to cause the processing system 110 to perform operations including providing guidance on arrhythmia initiation sites within the patient’s heart for use by physicians during ablation procedures. An arrhythmia model generator module 124 within the processor-executable instructions 112 may include instructions to cause the processing system 110 to perform operations including creating models of cardiac arrhythmias based on ICD data, and in some embodiments on electrocardiogram (ECG) data received from the ECG system 132. Such operations may include merging a 3D localization of an initiation site of the arrhythmia and the 3D heart model to form an arrhythmia activation model. This merged model may be used by the system 100 to provide a comprehensive visualization of the arrhythmia event within the context of the patient's specific cardiac anatomy and myocardial structure.15448153-1 58Attorney Docket No.: 35313-0011WO

[0249] The system 100 may interface with several external medical devices and system through dedicated connections or local and / or wide area networks via the network transceiver 108. An ECG system 132 may provide electrocardiogram data for the patient to the system 100. An ICD interface 134 may enable communication with implanted cardiac defibrillators to obtain downloads of recorded electric signals and pacing during arrhythmia events. A medical imaging system 136 may supply medical imaging data to the system 100. A display system 138 may present the processed information, models, and guidance to physicians both before procedures for diagnosis and planning use, and during procedures (e.g., during ventricular ablation treatments) such as to provide visual indications of arrythmia initiation sites overlayed on fluoroscopy of the patient’s heart.

[0250] FIG. 2A illustrates a cross-sectional view 200 of a human heart showing external and internal cardiac anatomy along with an example positioning of CEID (e.g., ICD) leads 202 and CEID (e.g., ICD) electrodes 204, 204 in an example of a dual-chamber ICD case.

[0251] The superior vena cava and inferior vena cava are shown connecting to the right atrium in FIG. 2A. These large veins may serve as potential pathways for introducing ICD leads into the heart. The right pulmonary arteries extend from the right ventricle, while the left pulmonary arteries and veins connect to the left side of the heart. The aortic arch emerges from the top of the heart, with the descending aorta continuing downward. In some embodiments, the 3D heart model generated by the computing system 102 (e.g., executing instructions from the patient-specific 3D heart model generator module 118) may include representations of these major vessels, which may be important for visualizing the overall cardiac structure and ICD electrode placement locations.

[0252] In some embodiments, the 3D heart model generated by the computing system 102 may include representations of ICD electrode locations 204 based on medical imagery data or physician notes on the ICD electrode implantation sites. For single-chamber ICDs, a lead 202 may be inserted through the subclavian vein, passing 15448153-1 59Attorney Docket No.: 35313-0011WOinto the right atrium, through the tricuspid valve, and into the right ventricle. The pacing / sensing electrode 204 may be positioned at or near the right ventricular apex as illustrated, while the defibrillation coil (not shown) may be positioned within the right ventricular cavity or right atrium. In dual-chamber ICDs, an additional lead and electrode 206 may be placed in the right atrium, often positioned in the right atrial appendage as illustrated. The right ventricular lead configuration may be similar to that of single-chamber ICDs. In some cases, a proximal defibrillation coil (not shown) may be located in the superior vena cava, although this may not be considered within the heart chambers. The patient-specific 3D heart model generated in various embodiments may be based upon electrical signals recorded by the ICD electrodes 204, 206, and incorporate these electrode locations in the displayed model to provide a visualization of the ICD system in relation to the patient's cardiac anatomy to aid in a treatment of ventricular arrythmia.

[0253] FIG. 2B illustrates a cross-sectional view 200 of a human heart showing external and internal cardiac anatomy along with an example positioning of EP catheters 212, 220 and electrodes 214, 216, 218, 222, 224, 226 within the left and right ventricles. As noted in the background, an EP procedure may involve a third EP catheter positioned within the coronary sinus, which is on the posterior of the heart and thus not visible in the view shown in FIG. 2B.

[0254] FIG. 2B shows how during an EP procedure a physician may maneuver one EP catheter 212 into the right ventricle and position the electrodes 214, 216, 218 so at least some of the electrodes contact the interior surface of the heart, and maneuver a second EP catheter 220 into the left ventricle and position the electrodes 222, 224, 2226 so at least some of the electrodes contact the interior surface of the heart. With the catheters so positioned, the locations of each electrode may be obtained by the catheter tracking system (e.g., Ensight® or CARTO®). The coordinate systems of the catheter tracking system and the 3D heart model may be correlated or aligned (e.g., through coordinate transformation calculations). Additionally, EP data may be collected from the electrodes 214, 216, 218, 222, 224, 226 during one or more heart15448153-1 60Attorney Docket No.: 35313-0011WObeats, particularly VT beats, and the EP data may be processed along with the electrode position information to generate a 3D activation model of the patient’s heart. As described herein, the EP data from the intracardiac electrodes 214, 216, 218, 222, 224, 226 may be combined with ECG data obtained by electrodes on the patient’s torso during the same heart beats to provide a better model of the heart’s conduction wave patterns and VT activation sites.

[0255] FIG. 2C illustrates an embodiment in which both signals obtained from both ICD electrodes and temporary EP catheter electrodes within the heart may be used. Thus, FIG. 2C shows the CEID (e.g., ICD) leads 202 and electrodes 204, 206 of FIG. 2A and the catheters 212, 220 and electrodes 214, 216, 218, 222, 224, 226 positioned within the heart.

[0256] FIGS. 2A-2C also reveal the internal structures of the heart, including the right atrium and right ventricle separated by the tricuspid valve. The pulmonary valve is positioned between the right ventricle and pulmonary arteries. On the left side, the left atrium and left ventricle are separated by the mitral valve. In some embodiments, the patient- specific 3D heart model generated by the computing system 102 may include detailed representations of these internal chambers and valves, which may be crucial for accurately localizing arrhythmia initiation sites and planning ablation procedures.

[0257] In some embodiments, the 3D heart model may include representations of heart structures such as the aorta, aortic arch, pulmonary veins, and coronary vessels. These structures, visible in FIGS. 2A-2C, may be important landmarks for navigating the heart during ablation and other procedures, as well as understanding the spatial relationships between different cardiac regions and the ICD electrodes when analyzing ICD data and planning for treatments.

[0258] The patient- specific 3D heart model generated in various embodiments may also incorporate information about heart scar tissue indicative of ischemic heart disease. While not shown in FIG. 2, areas of scarring may be identified through15448153-1 61Attorney Docket No.: 35313-0011WOmedical imaging techniques and integrated into the 3D model. This information may be crucial for understanding potential sources of arrhythmias and planning appropriate treatment strategies.

[0259] By incorporating these detailed anatomical features into the patient-specific 3D heart model, physicians may gain a comprehensive understanding of the individual patient's cardiac anatomy. This detailed model may aid in the precise localization of ICD electrodes, identification of arrhythmia initiation sites, and planning of cardiac ablation procedures.

[0260] FIG. 3 illustrates a non-limiting example of a 3D heart model 300 that the system 100 may generate according to various embodiments. The 3D heart model 300 may be displayed as a wireframe representation that shows the surface topology and contours of the heart structure. The 3D heart model may include a shock vector angle (not shown), which may be indicated by a curved line with an arrow at one end, demonstrating the direction of each electrical shock vector relative to the heart's surface.

[0261] With reference to FIGS. 1-3, the heart model generator module 112 may include instructions to cause the processing system 110 to generate a 3D heart model 300 based on medical imaging data, downloaded ICD data, EP catheter data, and / or ECG data, as well as other resources, such as a library of representative heart models that may be used as references. In some cases, but not limited to, the medical imaging data may be obtained using delayed enhanced magnetic resonance imaging (MRI) or delayed enhanced computed tomography (CT). These imaging techniques may provide detailed visualization of the heart's structures and may allow for accurate identification of ICD electrode locations within the patient's heart, which may be used in the calculation of electrical shock vectors relative to the heart's surface.

[0262] In some embodiments, when internal imaging data is not available for a specific patient, the heart model generator module 112 may include instructions to cause the processing system 110 to use a non-patient-specific heart model, which may15448153-1 62Attorney Docket No.: 35313-0011WOserve as a baseline for analysis and may be adapted and / or refined based on available patient data.

[0263] The 3D heart model with the shock vector angle 300 may be used by the processing system 110 executing the arrhythmia model generator module 124 to create models of cardiac arrhythmias. These models may take into account the specific geometry of the patient's heart, locations of scar tissue, and the positions of ICD electrodes, which may influence the path of electrical signals during arrhythmic events.

[0264] In some embodiments, the system 100 may display a localization point 402 and cardiac activation map of a single arrhythmia event on the 3D heart model 400 as shown in FIG. 4B. With reference to FIGS. 1-4A, in some embodiments, the arrhythmia model generator module 124 may include instructions to cause the processing system 110 to combine ICD information with the patient-specific 3D heart model 400 to identify isochrones or lines of equal depolarization at various time intervals per heartbeat to identify average directions of depolarization as function of time, and from such calculations determine model vectors of the depolarization wavefront across the patient’s heart. These calculations may identify a localization of the initiation site of the arrhythmia in a heartbeat, such as in a location of increased curvature in the wavefront, which may be represented by the localization point 402. The localization point 402 may indicate a critical isthmus associated with the arrhythmia, and thus a potential target for ablation.

[0265] In some embodiments, these calculations may take into account biophysical constraints, such as heart geometry, tissue conductivity, and anisotropic conduction properties, to refine the accuracy of localization. In this way, the system 100 may estimate the location of an arrhythmia’s initiation site that may be represented as the localization point 402 on the patient-specific 3D heart model 400 to provide a visual representation of the arrhythmia origin. In some cases, the localization point 402 may be displayed as a distinct marker on the surface of the heart model 400. This15448153-1 63Attorney Docket No.: 35313-0011WOinformation may assist physicians in targeting arrhythmogenic regions for diagnostic analysis, catheter mapping, or ablation planning.

[0266] In some embodiments, the display system 138 may present the 3D heart model 400 with the localization point 402 to users. The display system 138 may include instructions to cause the processing system 110 to display the localization of the arrhythmia as multiple points representative of multiple beats of a ventricular tachycardia on the displayed 3D heart model 400. This multi-point display may provide a more comprehensive view of the arrhythmia's behavior over time.

[0267] In some embodiments, the system 100 may be configured to display treatment guidance for physicians during a procedure, such as an ventricular ablation therapy. FIG. 4B illustrates an example of a rendered 3D heart model including a highlighted region 404 where ablation may be indicated. With reference to FIGS. 1-4B, in some embodiments the ablation guidance module 122 may include instructions to cause the processing system 110 to provide guidance, such as a highlighted region 404 superimposed on fluoroscopy to an indicated ablation location in a manner that may assist physicians in identifying and navigating to ablation sites by combining arrhythmia locations and myocardial structures.

[0268] FIG. 5 illustrates a method 500 for using electrical signal and electrode location data from electrodes on EP catheters temporarily positioned in a patient’s heart to identify arrhythmia locations. With reference to FIGS. 1-5 the method 500 may be performed in a processing system (e.g., 110) implementing software modules as described with reference to FIG. 1. Means for performing the functions of the operations in the method 500 may include a processing system including one or more processors and other components described herein. Further, one or more processors of a processing system may be configured with software or firmware modules (e.g., 114-124) to perform some or all of the operations of the method 500. To encompass the alternative configurations enabled in various embodiments, the hardware implementing any or all of the method 600 is referred to herein as a “processing system.”15448153-1 64Attorney Docket No.: 35313-0011WO

[0269] In block 502, the processing system may obtain medical imaging data that provides details on the location, shape and orientation of the patient’s heart within the patient’s body. Such medical imaging may be obtained using one or more imaging systems, such as for example CT, MRI, ICE, fluoroscopy, a combination thereof, and / or the like. In some embodiments, the processing system may download the medical imaging data from a medical records system, from a recorded medium storing the data, and / or directly from an imaging system.

[0270] In block 504, the processing system may generate a patient-specific 3D heart model based on the medical imaging and a select representative heart model. In such operations, the processing system may compare measurements of various heart structures to measurements of corresponding heart structures in a library of different representative heart models to select one model that best corresponds to the medical imaging. In some embodiments, the processing system may adjust or morph the selected representative heart model to better match the dimensions, orientation and size of the patient’s heart as determined based on the medical imaging. This generated 3D heart model may then be used as a framework onto which electrical conduction patterns can be mapped in subsequent operations.

[0271] In optional block 506, the clinicians may place ECG electrodes on the patient’s thorax. In some procedures, signals from the ECG electrodes may be used to assist the EP system in correlating EP signals from catheter electrodes to heartbeats and electrical patterns indicated by the ECG system. In optional embodiments, the signals from the ECG electrodes may be used in combination with electrical signals from intracardiac electrodes to generate a patient-specific electrical conduction map in subsequent operations.

[0272] In optional block 508, the processing system may obtain one or more images of the ECG electrodes on the patient’s thorax and process the images to determine the locations of each ECG electrode on the patent or within another frame of reference.15448153-1 65Attorney Docket No.: 35313-0011WO

[0273] In block 510, a physician may insert the EP catheters (e.g., three catheters) into the patient’s heart, and the processing system may obtain information on the location of each electrode on the catheters and receive EP signal, such as from an EP system. As described above, this may include obtaining the locations of electrodes in the heart within an accuracy of a few millimeters or less. As part of the data gathered in block 510, the processing system may obtain EP signals and match the signal information to timing data (e.g., timestamps) and correlate this information to the specific location of each electrode. In some procedures, the physician may pace the heart to induce an arrhythmia, such as a VT, to enable the system to collect the EP signals during one or a few beats during the arrhythmia event.

[0274] In block 512, the processing system may generate a patient-specific electrical conduction map of the patient’s heart based on the patient-specific 3D heart model and the electrode signal and location information that identifies a 3D localization of an initiation site of the arrhythmia. The patient-specific electrical conduction map may be dynamic, revealing the wavefronts of the conduction wave through the structures of the heart. In a VT event, the initiation site may be revealed as a distortion, loop, or break in the depolarization wavefront.

[0275] In block 514, the processing system may control the display system 138 to display the 3D localization of the initiation site of the arrhythmia for use by a physician preparing for a cardiac electrophysiology procedure. This visualization may assist in planning the procedure and identifying target areas for treatment.

[0276] In block 516, the processing system may execute the ablation guidance module 122 to provide visual guidance for physicians conducting a cardiac ablation procedure using the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for ablation. This guidance may help improve the precision and effectiveness of the ablation procedure.

[0277] In some embodiments, the method 500 may include additional operations for refining the arrhythmia localization during the procedure. In some embodiments15448153-1 66Attorney Docket No.: 35313-0011WOthe processing system may receive electrical signal information during an intracardiac procedure (e.g., during a cardiac ablation procedures) while the physician touches an electrode (e.g., an EP catheter electrode or an ablation electrode) on the myocardium to provide point-by-point contact electrophysiology recordings during the procedure. These recordings may be used by the processing system to update the patient-specific electrical conduction map and display an updated 3D localization of the initiation site of the arrhythmia.

[0278] The method 500 may be applicable to various types of arrhythmias. In some cases, the arrhythmia may be an atrial arrhythmia, originating in the upper chambers of the heart. In other cases, the arrhythmia may be a ventricular arrhythmia, such as a VT, occurring in the lower chambers. The method 500 may also be used to identify and localize pre-ventricular contractions (PVCs), which are early, abnormal heartbeats originating in the ventricles. In some embodiments, the method 500 may be particularly useful for localizing VT initiation sites. The method 500 may also be applied to identify dysrhythmias between the two ventricles, where there is an abnormal coordination of electrical activity between the left and right ventricles.

[0279] By providing detailed, patient-specific information about the location and characteristics of various arrhythmias, the method 500 may enable more precise and effective cardiac ablation procedures, potentially improving outcomes for patients with complex cardiac rhythm disorders.

[0280] FIG. 6 illustrates a method 600 for using ICD data to identify arrhythmia locations. With reference to FIGS. 1-6, the method 600 may be performed in a processing system (e.g., 110) implementing software modules as described with reference to FIG. 1. Means for performing the functions of the operations in the method 600 may include a processing system including one or more processors and other components described herein. Further, one or more processors of a processing system may be configured with software or firmware modules (e.g., 114-124) to perform some or all of the operations of the method 600. To encompass the alternative15448153-1 67Attorney Docket No.: 35313-0011WOconfigurations enabled in various embodiments, the hardware implementing any or all of the method 600 is referred to herein as a “processing system.”

[0281] In block 602, the processing system may execute the ICD data download module 114 to perform operations including downloading information from an ICD recorded during detected arrhythmia events in the patient's heart. This information may include electrogram data, timing of arrhythmic events, and other relevant cardiac parameters. The processing system 110 may download the ICD data through a secure wireless connection utilizing standardized communication protocols to ensure reliable and accurate data transfer from the implanted device to the system's storage and analysis components.

[0282] In block 604, the processing system may execute the medical imaging download module 116 to obtain medical imaging of the patient’s heart that also identifies locations of one or more ICD electrodes in the patient's heart. This imaging data may be acquired using techniques such as computed tomography (CT) or magnetic resonance imaging (MRI). In some cases, the medical imaging data may be downloaded from a database such as a hospital medical records system, allowing for efficient retrieval and integration of previously acquired patient imaging studies into the arrhythmia localization process.

[0283] In block 606, the processing system may execute the patient-specific 3D heart model generator module 118 to generate a 3D model that can be used to generate images of the patient's heart during an arrhythmia in subsequent operations in blocks 608 and 514. In some embodiments, the processing system may select from a library of heart models, a representative 3D heart model that is close in size, shape, orientation to that of the patient, and then adjust, reorient, and otherwise adjust the size, shape, and orientation to match medical imagery. Further, the processing system may include in the 3D heart model locations of pacing and sensing electrodes of the ICD, thereby correlating the recorded electrical signals downloaded from the ICD with the locations in the heart where the signals were obtained.15448153-1 68Attorney Docket No.: 35313-0011WO

[0284] In block 608, the processing system may execute the electrical conduction map generator module 120 to generate a patient-specific electrical conduction map of the patient's heart during an arrhythmia. This map may be based on the patientspecific 3D heart model generated in block 606, the locations of the ICD electrodes, and the downloaded ICD information recorded during detected arrhythmia events. In some embodiments, the map may be generated by combining the ICD information with the patient-specific 3D heart model to identify isochrones at time intervals of heartbeats from which direction vectors of depolarization wavefronts are generated to reveal the arrhythmia initiation site in a single heartbeat. The electrical conduction map may identify a localization of an initiation site of the arrhythmia. The electrical conduction map may provide a visual representation of the electrical activity in the heart during the arrhythmia event, potentially showing the propagation of electrical signals across the myocardium. In some aspects, the map may incorporate data from multiple arrhythmia events to provide a more comprehensive view of the patient's arrhythmia patterns. The electrical conduction map may also include information about the timing and sequence of depolarization in different regions of the heart, which may be useful for understanding the mechanism of the arrhythmia and planning treatment strategies.

[0285] In block 514, the processing system may control the display system 138 to display the 3D localization of the initiation site of the arrhythmia for use by a physician preparing for a cardiac electrophysiology procedure as described for the like number block with reference to FIG. 5.

[0286] In block 516, the processing system may execute the ablation guidance module 122 to provide visual guidance for physicians conducting a cardiac ablation procedure using the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for ablation as described for the like number block with reference to FIG. 5.

[0287] In some embodiments, the method 600 may include additional operations for refining the arrhythmia localization during the procedure. In some embodiments 15448153-1 69Attorney Docket No.: 35313-0011WOthe processing system 110 receive electrical signal information during an intra-cardiac procedure (e.g., during a cardiac ablation procedures) while the physician touches an electrode (e.g., a sensing electrode or an ablation electrode) on the myocardium to provide point-by-point contact electrophysiology recordings during the procedure. These recordings may be used by the processing system to update the patient-specific electrical conduction map and display an updated 3D localization of the initiation site of the arrhythmia.

[0288] The method 600 may be applicable to various types of arrhythmias. In some cases, the arrhythmia may be an atrial arrhythmia, originating in the upper chambers of the heart. In other cases, the arrhythmia may be a ventricular arrhythmia, occurring in the lower chambers. The method 600 may also be used to identify and localize pre-ventricular contractions (PVCs), which are early, abnormal heartbeats originating in the ventricles.

[0289] In some embodiments, the method 600 may be particularly useful for localizing VT, a potentially life-threatening arrhythmia characterized by a rapid heart rate originating in the ventricles. The method 600 may also be applied to identify dysrhythmias between the two ventricles, where there is an abnormal coordination of electrical activity between the left and right ventricles.

[0290] By providing detailed, patient-specific information about the location and characteristics of various arrhythmias, the method 600 may enable more precise and effective cardiac ablation procedures, potentially improving outcomes for patients with complex cardiac rhythm disorders.

[0291] FIG. 7 illustrates another embodiment method 700 that may be performed by the system 100 to identify arrhythmia locations using both intracardiac electrode data (e.g., EP catheter data and / or downloaded ICD data) and ECG data. The method 700 include operations of the method 600 described with reference to FIG. 6 plus operations to use ECG data gathered from electrodes located on the skin of the patient to provide more comprehensive information for cardiac procedures. With reference to15448153-1 70Attorney Docket No.: 35313-0011WOFIGS. 1-7, the method 700 may be performed in a processing system (e.g., 110) implementing software modules as described with reference to FIG. 1. Means for performing the functions of the operations in the method 700 may include a processing system including one or more processors and other components described herein. Further, one or more processors of a processing system may be configured with software or firmware modules (e.g., 114-124) to perform some or all of the operations of the method 700. To encompass the alternative configurations enabled in various embodiments, the hardware implementing any or all of the method 700 is referred to herein as a “processing system.”

[0292] In block 602, the processing system may execute the ICD data download module 114 to download data from the patient’s ICD as described for the like number block with reference to FIG. 6.

[0293] In block 604, the processing system may execute the medical imaging download module 116 to obtain medical imaging that identifies locations of one or more ICD electrodes in the patient's heart as described for the like number block with reference to FIG. 6.

[0294] In block 606, the processing system may execute the patient heart model generator 118 to generate a patient-specific 3D heart model as described for the like number block with reference to FIG. 6.

[0295] In block 702, the processing system may execute the ECG system 132 to obtain a 3D image of ECG electrodes on the patient's torso during an ECG recording procedure. This 3D image may provide spatial information about the ECG electrode positions relative to the patient's body

[0296] In block 704, the processing system may execute the heart model generator 112 to merge the 3D image of the patient's torso with the 3D heart model. This merging operation may include aligning the external ECG electrode positions with the internal cardiac structures. In some embodiments, this operation may include geometrically correlating or aligning (e.g., through coordinate transformation) the15448153-1 71Attorney Docket No.: 35313-0011WOcoordinate system of images of the ECG electrodes, and thus their locations in the image coordinate system, with the coordinate system of the 3D heart model of the patient. This may enable the processing system to combine electrophysiology information recorded on the outside of the patient’s torso with electrical signal information recorded by electrodes of the ICD in block 706.

[0297] In block 706, the processing system may execute the electrical conduction map generator 120 to generate a patient-specific electrical conduction map of a patient's heart of an arrhythmia. The operations performed in block 706 may be similar to those described for block 608 of the method 600 with reference to FIG. 6. This map may be based on the patient-specific 3D heart model, the downloaded ICD recorded information regarding detected arrhythmia events, and ECG data from the ECG recording procedure. The patient-specific electrical conduction map may identify a localization of an initiation site of arrhythmia events. This combination of internally recorded and externally recorded electrical data may provide a more comprehensive view of the arrhythmia's origin and propagation.

[0298] In block 514, the processing system may control the display system 138 to display the 3D localization of the initiation site of the arrhythmia for use by a physician as described for the like number block with reference to FIG. 5.

[0299] In block 516, the processing system may execute the ablation guidance module 122 to provide visual guidance for physicians conducting a cardiac ablation procedure using the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for ablation as described for the like number block with reference to FIG. 5.

[0300] In some embodiments, the electrical conduction map generator 120 may include instructions to cause the processing system 110 to perform operations including using the Wilson Central Terminal as a common reference point for ECG leads. This approach may help standardize the electrical measurements from different leads and facilitate integration with surface ECG data.15448153-1 72Attorney Docket No.: 35313-0011WO

[0301] The method 700 may provide a more comprehensive approach to arrhythmia localization by integrating multiple data sources and advanced processing techniques. By combining ICD data, ECG recordings, and detailed 3D imaging, the method 700 may offer improved accuracy in identifying arrhythmia initiation sites and guiding cardiac procedures.

[0302] Another example system 800 for processing cardiac data and supporting ablation procedures is illustrated in FIG. 8. The system 800 may include a computing system 802 that contains several interconnected components and modules for analyzing cardiac data and providing guidance during ablation procedures.

[0303] The computing system 802 may include an electronic storage 806 for storing data and a transceiver 808 for communicating with external devices and systems. A processing system 810 within the computing system 802 is configured with processor-executable instructions 812 that implement various functional modules for cardiac data analysis and procedure support.

[0304] The processor-executable instructions 812 may include an intracardiac electrode data download module 814 for obtaining electrode data from implanted devices (i,e., CEID, such as for example, ICD), a medical imaging download module 816 for receiving imaging data of a patient’s heart, a heart model generator module 818 for creating patient-specific 3D heart models, an electrical conduction simulation module 820 that simulates cardiac electrical activity, and a beat selection module 822 that identifies and selects relevant arrhythmia beats from recorded data. In some variations, the heart model generator module 818 may include further processor-executable instructions to execute MLiECG framework and anatomical alignment described herein to align the location of the intracardiac electrodes in the 3D heart model, so as to precisely define and specify the location of the intracardiac electrodes within the 3D heart model. In some variations, the heart model generator module 818 may include further processor-executable instructions to represent the 3D heart model using boundary element method (BEM), finite element method (FEM), or a hybrid element method (HEM) that combines both BEM and FEM as described herein. The 15448153-1 73Attorney Docket No.: 35313-0011WOheart generator module 818 may include instructions to utilize BEM modeling, FEM modeling, and / or hybrid element method modeling in conjunction with the MLiECG framework and anatomical alignment. This is described further in the above sections of this disclosure. In some variations, the electrical conduction simulation module 812 may simulate the cardiac electrical activity by accounting for tissue characteristics as discussed above (e.g, by computing a distance matrix as discussed above).

[0305] In some embodiments, the system 800 may include an origin area simulation module 824 that performs a reverse simulation of the propagation of depolarization / activation wavefronts across the heart to identify an initial target region for potential arrhythmia origins. This module may use the patient- specific 3D heart model and recorded electrogram data to simulate activation waves in reverse order from the relative recorded local activation times. In some embodiments, this reverseorder simulation of activation wave propagation may help to narrow down the search area for arrhythmia sources.

[0306] The system 800 also may include an ablation location selection module 828 that analyzes simulated and recorded data to identify optimal ablation sites. In some embodiments, this module performs correlation analysis of pacing simulations to match recorded ICD electrograms (EGMs). By comparing simulated EGMs from various potential arrhythmia origin sites to the actual recorded ICD EGMs, the system can identify the most likely location of the arrhythmia source. In some variations, the ablation location selection module 828 may include further instructions to compare the simulated EGMs that are integrated with the MLiECG framework and anatomical alignment with recorded CEID EGMs that are integrated with the MLiECG framework and anatomic alignment). For example, in some variations, the ablation location selection module 828 may analyze the simulated and recorded EGMs as described above under the sub-section of “EGM simulations and analyses” in the present disclosure.

[0307] A procedure support display module 830 may generate visual representations of the analysis results and cardiac data to assist clinicians during ablation procedures.15448153-1 74Attorney Docket No.: 35313-0011WO

[0308] The computing system 802 interfaces with several external systems through connections as illustrated in FIG. 8. These external systems may include an ECG system 832 for surface electrocardiogram recordings, an ICD interface 834 for communicating with implanted cardiac devices, a medical imaging system 836 for obtaining cardiac imaging data, and a 3D camera 838 for capturing spatial information regarding ECG electrodes on a patient. The system also may include an interface for displays and EP systems 840 for outputting determinations and receiving EGMs from implanted EP catheters.

[0309] The transceiver 808 facilitates communication between the computing system 802 and the various external systems and interfaces. The electronic storage 806 stores data used and generated by the processing system 810 and its modules, including patient-specific heart models, recorded and simulated electrograms, and analysis results.

[0310] In some embodiments, the system 800 may generate a heart model 900 based on medical imaging data, such as computed tomography (CT) and / or magnetic resonance imaging (MRI) scans of a patient’s heart. FIGs. 9A, 9B, and 9C illustrate different views of the heart model 900, showing various anatomical structures and ICD and EP electrode placements.

[0311] FIGS. 9A-9C also reveal the internal structures of the heart, including the right atrium and right ventricle separated by the tricuspid valve. The pulmonary valve is positioned between the right ventricle and pulmonary arteries. On the left side, the left atrium and left ventricle are separated by the mitral valve. The superior vena cava and inferior vena cava connect to the right atrium in FIG. 9A. These large veins may serve as potential pathways for introducing ICD leads and EP catheters into the heart. The right pulmonary arteries extend from the right ventricle, while the left pulmonary arteries and veins connect to the left side of the heart. The aortic arch emerges from the top of the heart, with the descending aorta continuing downward. In some embodiments, the 3D heart model generated by the computing system 802 (e.g., executing instructions from the patient-specific 3D heart model generator module 818) 15448153-1 75Attorney Docket No.: 35313-0011WOmay include representations of these major vessels, which may be important for visualizing the overall cardiac structure and ICD electrode placement locations, as well as indicating arrhythmia initiation sites and planning ablation procedures.

[0312] FIG. 9A illustrates a cross-sectional view 900a of a human heart showing external and internal cardiac anatomy along with an example positioning of an example CEID components. This example illustrates an example of a dual-chamber ICD case. FIG. 9A also illustrates major components of an example ICD, including a pulse generator 901, electrical leads 902, one or more defibrillation coils 903, and a number of sensor and / or pacing electrodes 904, 905, 906, 928, 930, 932, 934.

[0313] The pulse generator 901 is an integrated system that serves as the central control unit for the intracardiac defibrillator, processing cardiac signals detected by the lead electrodes, determining the need for therapy (pacing or defibrillation), and delivering the appropriate electrical impulses. Being contained in a hermetically sealed container, pulse generator 901 is commonly and referred to herein to as a “can.” The can 901 stores and manages the energy required for high-voltage defibrillation shocks and low-voltage pacing pulses. In some ICD models the can is configured as an active electrode, forming part of the defibrillation circuit, acting as the opposing pole to the ventricular coil to complete the electrical pathway through the heart.

[0314] The electrical leads 902 extend from the can 901 and are threaded through heart structures (e.g., the vena cava into the RV) and around the heart, delivering electrical signals of the heart from sensing electrodes 904, 905, 906 and pacing or electrical shocks to the defibrillation coil 903 and pacing electrodes.

[0315] The defibrillation coil 903 (“coil”) is a defibrillation electrode integrated into a lead 902 of the ICD. Typically, the coil resides within the blood pool of a cardiac ventricle, such as the RV, to ensure optimal energy delivery. The coil is configured to deliver high-energy electrical shocks to the heart tissue to terminate lifethreatening ventricular arrhythmias, such as VT or ventricular fibrillation (VF). The15448153-1 76Attorney Docket No.: 35313-0011WOcoil 903 serves as one pole of the defibrillation circuit, with the electrical energy passing through the cardiac tissue to the opposing pole, often the can or another coil. The large surface area of the coil facilitates efficient energy transfer, minimizing impedance and ensuring effective defibrillation while reducing the risk of thermal injury to surrounding tissues. Additionally, the coil’s placement in the ventricular blood pool allows for a broad electric field, enhancing the likelihood of capturing sufficient myocardial tissue to restore normal sinus rhythm.

[0316] The ICD may include multiple pacing / sensing electrodes. FIG. 9A illustrates an ICD implementation in which a sensor / pacing electrode 904 is anchored in the RV 904, a ring electrode 905 is positioned within the RV, and another senser / pacing electrode 906 is positioned in the right right atrium. In some embodiments, electrode 904 is another ring electrode disposed more proximally to electrode 905 in a RV lead with a passive fixation mechanism. In some embodiments, electrode 904 is part of an active fixation mechanism that anchors the ICD lead to the RV. In some embodiments, electrode 906 is disposed at the tip of the RA lead as part of the active fixation mechanism or as a ring electrode one of the closely spaced paired bipolar electrodes at the distal end of a RA lead with a passive fixation mechanism. The illustrated example ICD also includes electrodes 928, 930, 932, 934 touching or close to the exterior surface (i.e., epicardium) of the heart. Additionally, the can 901 and the coil 903 may function as electrical sensors for the ICD. Any ICD lead electrode can be used for pacing and / or sensing of EGMs. For pacing, an electrode needs to be close enough to or implanted into the myocardial tissue.Electrodes on ICD leads that are not on or implanted into myocardial tissue may be used to sense EGMs with accommodations for transmission delays through blood in which the electrode is suspended.

[0317] In some embodiments, the 3D heart model generated by the computing system 802 may include representations of ICD electrode locations based on medical imagery data or physician notes on the ICD electrode implantation sites. For singlechamber ICDs, a lead 902 may be inserted through the subclavian vein, passing into15448153-1 77Attorney Docket No.: 35313-0011WOthe right atrium, through the tricuspid valve, and into the right ventricle. The pacing / sensing electrode 904 may be positioned at or anchored into the RV apex as illustrated, while the defibrillation coil 203 may be positioned within the RV cavity (as illustrated) or right atrium. In dual-chamber ICDs, an additional lead with an electrode 906 may be placed in the right atrium, often positioned in the right atrial appendage as illustrated. The right ventricular lead configuration may be similar to that of single-chamber ICDs. In some cases, a proximal defibrillation coil (not shown) may be located in the superior vena cava, although this may not be considered within the heart chambers. Some ICDs may be configured with pacing electrodes and / or a coil to be positioned in the LV (not shown).

[0318] The patient- specific 3D heart model generated in various embodiments may incorporate the location of ICD electrodes 903, 904, 905, 906, 928, 930, 932, 934, process electrical signals recorded by the ICD electrodes, and in some embodiments, display locations of the ICD electrodes in 3-D heart model images to provide a visualization of the ICD system in relation to the patient's cardiac anatomy to aid in a treatment of an arrythmia (e.g., VT or PVC).

[0319] FIG. 9B illustrates a cross-sectional view 900b of a human heart showing external and internal cardiac anatomy along with an example positioning of EP catheters 912, 920 and electrodes 914, 916, 918, 922, 924, 926 within the left and right ventricles. An EP procedure may also involve positioning one or more EP catheters within the coronary sinus, which is on the posterior of the heart and thus not visible in the view shown in FIG. 9B.

[0320] FIG. 9B shows how during an EP procedure a physician may maneuver one EP catheter 912 into the right ventricle and position the electrodes 914, 916, 918 so at least some of the electrodes contact the interior surface of the heart, and maneuver a second EP catheter 920 into the left ventricle and position the electrodes 922, 924, 926 so at least some of the electrodes contact the interior surface of that chamber of the heart. With the catheters so positioned, the locations of each electrode may be obtained by the catheter tracking system (e.g., Ensight® or CARTO®). The 15448153-1 78Attorney Docket No.: 35313-0011WOcoordinate systems of the catheter tracking system and the 3D heart model may be correlated or aligned (e.g., through coordinate transformation calculations).Additionally, EP data may be collected from the electrodes 914, 916, 918, 922, 924, 926 during one or more heart beats, particularly arrythmia beats, and the EP data may be processed along with the electrode position information to generate a 3D activation model of the patient’s heart. In some embodiments, the EP data from the intracardiac electrodes 914, 916, 918, 922, 924, 926 may be combined with ECG data obtained by electrodes on the patient’s torso during the same heart beats to provide a better model of the heart’s conduction wave patterns and arrhythmia activation sites.

[0321] FIG. 9C illustrates a cross-sectional view 900c of a human heart showing external and internal cardiac anatomy with an implanted ICD and an example of EP catheters inserted within the left and right ventricles as may be implemented in an EP procedure. This figure illustrates an embodiment in which EGMs obtained from both ICD electrodes and temporary EP catheter electrodes within the heart may be used. Thus, FIG. 9C shows the ICD leads 902 and ICD electrodes 903, 904, 905, 906, 928, 930, 932, 934 as in FIG. 9A and the EP catheters 912, 920 and EP electrodes 914, 916, 918, 922, 924, 926 positioned within the heart.

[0322] In some embodiments, the 3D heart model generated by the computing system 802 (e.g., executing instructions from the patient-specific 3D heart model generator module 818) may include representations of heart structures such as the aorta, aortic arch, pulmonary veins, and coronary vessels. These structures, shown in FIGS. 9A-9C, may be important landmarks for navigating the heart during ablation and other procedures, as well as understanding the spatial relationships between different cardiac regions and the ICD electrodes when analyzing ICD data and planning for treatments. The intracardiac electrode data download module 814 may obtain data recorded from sampling and digitizing voltage signals measured across ICD electrodes 904, 906, 926, 928, 930, 932, which may be used by the beat selection module 822 to identify relevant arrhythmia beats. The origin area simulation module15448153-1 79Attorney Docket No.: 35313-0011WO824 may use the heart model 900 and the electrode data to simulate activation waves and identify potential arrhythmia origins.

[0323] In some embodiments, the procedure support display module 830 may generate visual representations of the heart model 900, including the electrode placement and activation path, to assist clinicians during ablation procedures. These visualizations may be presented on a display through the display interface 840, providing detailed anatomical context for the electrical activity data.

[0324] In some embodiments, the system determines local activation time from ICD EGMs. FIG. 10A illustrates activation time measurements from ICD EGMs. The figure includes two graphs showing electrical potential measurements over time.

[0325] The upper graph in FIG. 10A displays multiple overlaid waveform traces plotted as millivolts (mV) versus milliseconds (msec). Different colored traces represent signals from various ICD leads. The lower graph in FIG. 10A presents a simplified view of EGMs from two ICD leads, showing a relationship between two specific ICD lead measurements over the same time period.

[0326] In some embodiments, the activation time of an arrhythmia near the recording electrodes may be determined by identifying the minimum of the first derivative of the EGM signal. This point corresponds to the steepest downslope of the EGM waveform, which typically indicates the time of local tissue activation.

[0327] Referring to FIG. 10A, the vertical axis represents electrical potential difference in millivolts, while the horizontal axis shows time in milliseconds. The traces exhibit characteristic deflections and peaks that are typical of cardiac electrical activity recordings.

[0328] In some embodiments, the system (e.g., system 800) may analyze these EGM waveforms to identify the point of steepest negative slope for each lead. This point may be considered the local activation time for the cardiac tissue in proximity to that particular electrode.15448153-1 80Attorney Docket No.: 35313-0011WO

[0329] By determining the local activation times for multiple ICD leads, the system may construct a sequence of activation times across different regions of the heart. This information may be used in conjunction with the 3D heart model to simulate cardiac electrical activity and identify potential arrhythmia sources.

[0330] In some embodiments, the electrical conduction simulation module 820 may simulate pacing stimuli at implantable ICD electrode locations in reverse order of recorded activation sequence. The origin area simulation module 824 may use these simulations to generate isochrones and identify search regions for potential arrhythmia origins.

[0331] FIG. 10B shows example EGM traces from three combinations of ICD electrodes illustrating a method of identifying a global activation time according to some embodiments. Specifically, FIG. 10B shows non- limiting example traces of electrical potentials between the RV tip and RV coil 1002, between the can and RV coil 1004, and between the RV tip and the RV ring 1006.

[0332] Referring to FIG. 10B, the global activation time (t = 0) is the earliest time where any lead deviates from its corresponding isoelectric level. So to identify a global activation time for a particular arrhythmia beat, the system 800 may process the ICD EGM recording to identify an isoelectric level for multiple potential pairs using some or all of the ICD sensing electrodes. The figure shows examples of isoelectric levels 1006a, 1006b, 1006c identified for the electrical potentials between the RV tip and RV coil 1000, between the can and RV coil 1002, and between the RV tip and the RV ring 1004, respectively. In this example, the isoelectric levels 1006a, 1006b, 1006c are determined based on where the VT terminates and there is one PVC followed by one sinus beat, with plenty of time to return to baseline between beats. Observations of ICD EGMs in practice has revealed that there will typically be some beats before or after the arrhythmia recorded by the ICD that can be used to determine the isoelectric level.15448153-1 81Attorney Docket No.: 35313-0011WO

[0333] Note that the RVtip / RVring EGM 1004 does not show deviation from the isoelectric level 1006c until much later (about 80 msec), because it is a bipolar lead and relatively far from the pacing site. Conversely, the RVtip / RVcoil EGM 1000 never really returns to baseline during the VT. This is because the heart is still repolarizing when the next pacing pulse activates. The figure shows that where the VT terminates the T wave does eventually return to baseline. Thus, the best electrode-to-electrode potential EGM to use in this example for determining global activation time is the Can / RVcoil EGM 1002.

[0334] The vertical line 1008 illustrates an example of the global activation time for the beat following the line. The figure illustrates how typically there is some beat-to-beat variation, but this time appears to be the most common for all the beats. In this example, the beats were being paced, so the timing of the pacing provides the global activation time anyway.

[0335] FIG. 10C shows a cross section of a heart with locations of ICD electrodes 903, 904, 906, 928, 930, 932, 934 indicated as illustrated in FIG. 9A, and vector lines 1012-1024 to each of these electrodes from an ECG electrode 1010 positioned on the patient’s thorax.

[0336] Referring to FIG. 10B, electrograms from one or more electrodes of an external 12-lead can be used by the processing system 800 to calculate vectorcardiograms (VCG) in the heart. A vectorcardiogram (VCG) is a diagnostic tool that represents the heart’s electrical activity as a three-dimensional vector loop, capturing the magnitude and direction of the cardiac electrical dipole moment in a spatial coordinate system. In some variations, a VCG may be constructed based on the MLiECG framework described herein. For instance, the EGM data, both recorded and simulated EGMs, can be represented using the vector leads in the MLiECG framework to construct the vectorcardiograms. Unlike a conventional electrocardiogram (ECG), which records scalar potentials over time across multiple leads, the VCG provides a comprehensive spatial analysis by plotting vector loops corresponding to the cardiac cycle’s P, QRS, and T waves in frontal, sagittal, and 15448153-1 82Attorney Docket No.: 35313-0011WOtransverse planes. This enables detailed assessment of cardiac conditions, such as myocardial infarction, ventricular hypertrophy, and arrhythmias, by analyzing the orientation, shape, and timing of these loops. The VCG can be derived from standard 12-lead ECG data through mathematical transformations or recorded directly using specialized lead systems, offering enhanced diagnostic precision for spatial electrical dynamics in clinical and research applications.

[0337] The calculation of a vectorcardiogram from a 12-lead ECG may be performed using the equation:VCG(t)=(1 / N)2n=l: N fn ’ CCgn(t),where VCG(t) is the three-dimensional vector representing the heart’s electrical dipole at time (t), ecgn(t) is the scalar potential recorded by the nth ECG lead, and tnis the corresponding lead vector defining the spatial orientation of the nthelectrode relative to the heart’s electrical center.

[0338] FIG. 10B illustrates an instance in which there are seven ICD electrodes, and thus N=7. Using ICD electrodes, each ICD lead’s contribution is computed by scaling the corresponding ℓnlead vector 1012-1024 by the measured potential between them, which is possible when the ECG data is time-synchronized with the ICD EGM data. This time-synchronization may be accomplished by the processing system aligning pacing stimulus signals recorded by both the ECG system and the ICD. Each electrodes contribution may be computed by scaling its lead vector 1012-1024 by the measured potential, with the results summed and averaged to produce the VCG. This process may be repeated using more than one ECG electrode (e.g., one or more of the electrodes in a 12-lead ECG), with results further summed or averaged to generate what may be a more accurate VCG.

[0339] This method reconstructs the heart’s dipole by aggregating the spatial projections of electrical activity, enabling the generation of vector loops for diagnostic analysis. The approach assumes a linear relationship between surface potentials and the cardiac dipole, with lead vectors derived mathematically from standard ECG15448153-1 83Attorney Docket No.: 35313-0011WOconfigurations, providing a robust framework for translating scalar ECG data into a three-dimensional vector representation.

[0340] FIG. 12A illustrates a non-limiting example in which the system 800 may simulate pacing from the point of the ICD electrode(s), which may include the RV tip electrode 904 placed near the RV apex 1202. In such a simulation, the isochrones generated from this simulation may spread outward from the right ventricle tip 1202. A first search region 1204 may be identified based on locations of ICD electrodes, activation times in recorded EGMs, and time increments of the isochrones, with the identified search region representing locations where an arrhythmia could potentially originate given the activation timing from the right ventricle tip 1202 stimulation point.

[0341] The origin area simulation module 824 may analyze the isochrones generated from multiple simulated pacing sites. In some embodiments, the module 824 may identify a candidate region for potential arrhythmia origins by determining the intersection of isochrones at t=0 from these multiple simulated pacing sites. This approach may help narrow down the likely location of arrhythmia origin more effectively than using a single simulated pacing site.

[0342] The procedure support display module 830 may generate visual representations of these isochrone diagrams and search regions. These visualizations may be presented through the display interface 840, providing clinicians with detailed information about potential arrhythmia origins during ablation procedures.

[0343] In some embodiments, the ablation location selection module 828 may use the identified candidate region as a starting point for more detailed simulations and comparisons with recorded ICD electrograms. This approach may help optimize the process of identifying the most likely point of arrhythmia origin for ablation procedures.

[0344] In some embodiments, the system 800 may analyze activation regions and search areas to refine the localization of potential arrhythmia origins. Referring to15448153-1 84Attorney Docket No.: 35313-0011WOFIGs. 12B-12D, the heart model 900 may be used to visualize a search area 1210, 1212, 1214 based on EGM data from single or multiple ICD leads. FIG. 12B illustrates a non-limiting example of a search area 1210 that may be identified by the system 800 for a particular arrythmia initiation site when only signal from an ICD sensor electrode positioned in the RV (e.g., at the RV tip) is used to identify the search area. FIG. 12C illustrates a non-limiting example of a search area 1212 that may be identified by the system 800 for the same arrythmia initiation site when only signals from an ICD sensor electrode positioned in the LV (e.g., an LV mid electrode) is used to identify the search area. FIG. 12D illustrates a non-limiting example of a search area 1214 that may be identified by the system 800 for the same arrythmia initiation site when signals from ICD sensor electrodes positioned in both the RV (e.g., at the RV tip) and LV (e.g., an LV mid electrode) are used to identify the search area. As shown in the illustrated example, limiting the search area to the intersection of the search areas generated by the RV and LV electrodes removes the LV basal free wall from the potential origin area, reducing the area in the heart to be used in the method for identifying the most likely site of arrhythmia activation.

[0345] FIGs. 12B, 12C, and 12D are merely examples and are not generalizable to any patient nor reflective of what will be observed in any procedure. It should be appreciated that the “search area” identified by the system 800 in any procedure will be unique to the patient, depending on the heart’s geometry, electrode locations, and actual recorded EGMs.

[0346] The search area 1210, 1212 may be indicated with a coloring scheme, such as using red, green, and blue to represent different activation timing zones. In some embodiments, the electrical conduction simulation module 820 may generate a display of the search area for use during an EP or ablation procedure based on the simulated propagation of electrical signals through the heart model 800 as described herein.

[0347] The procedure support display module 830 may generate visual representations of the activation region 1210 and search area 1212 for presentation through the display interface 840. In some embodiments, these visualizations may 15448153-1 85Attorney Docket No.: 35313-0011WOprovide clinicians with valuable information for guiding ablation procedures, potentially improving the precision of treatment and reducing procedure duration.

[0348] In some embodiments, the beat selection module 822 may analyze ICD electrograms from multiple leads to identify characteristics that contribute to the refinement of the search area 1212. The intracardiac electrode data download module 814 may obtain all recorded data from ICD leads, allowing for a more comprehensive analysis of cardiac electrical activity.

[0349] The heart model generator module 818 may incorporate the refined search area 1212 into the patient-specific 3D heart model, providing a more accurate representation of potential arrhythmia origins. In some embodiments, this refined model may be used by the ablation location selection module 828 to identify optimal sites for cardiac ablation procedures.

[0350] In some embodiments, the system 800 may implement a method 1300 for localizing and treating cardiac arrhythmias using ICD lead EGM data, as illustrated in FIG. 13. The method 1300 may be implemented in a processing system 810 of a computing system 802 by implementing processor-executable instructions 814-830.

[0351] In step 1302, the processing system may obtain medical imaging of a patient’s heart. In some embodiments, the processing system may execute the medical imaging download module 816 to receive imaging data from the medical imaging system 836, which may include computed tomography (CT) and / or magnetic resonance imaging (MRI) scans.

[0352] In step 1304, the processing system may execute the heart model generator module 818 to generate a patient-specific three-dimensional (3D) heart model based on the medical imaging and a selected representative 3D heart model. This patientspecific heart model (e.g., model 900) may provide detailed anatomical information for subsequent analysis steps. In some variations, the processing system may generate a BEM model, FEM model, and / or a hybrid BEM-FEM model of the heart. These15448153-1 86Attorney Docket No.: 35313-0011WOmodels may be generated by incorporating the MLiECG framework and anatomical alignment as discussed above.

[0353] In some embodiments, the processing system may perform an optional step 1306 to incorporate locations and characteristics of scar tissue in the 3D heart model (e.g., heart model 900). For example, the processing system may compute the distance matrix that defines the conduction velocity between two vertices of the 3D heart model as described herein. This information may be based on medical imaging or other diagnostic tests.

[0354] In step 1308, the processing system may localize ICD leads within the 3D heart model (e.g., 3D heart model 900) based on the medical imaging data and / or information obtained from the ICD interface 834. In some variations, the processing system may implement instructions that define a framework for representing electrical signals from the electrodes. The framework may be aligned with anatomical features of the heart to produce a coordinate system that enables precise localisation of the intracardiac electrodes within the 3D heart model. For example, the processing system may implement the MLiECG framework and the anatomical alignment as described herein to precisely locate the intracardiac electrodes within the 3D heart model..

[0355] Additionally knowledge of the physician who implanted the ICD may be entered into the system 800, such as via a user interface (e.g., mouse) or touch-sensitive display screen that the implanting physician can use to indicate where the electrodes were implanted or suspended. Alternatively, physician notes made following ICD implantation may be input or otherwise used to provide the system with ICD electrode information. Location information from various sources (different imaging system, implanting physician notes or input, etc.) may be considered together to generate a most-likely location when there is overlap in locations determine by different localizing methods.

[0356] In step 1310, the processing system may execute the intracardiac electrode data download module 814 to obtain ICD EGMs from the ICD through the ICD15448153-1 87Attorney Docket No.: 35313-0011WOinterface 834. These EGMs may provide crucial information about cardiac electrical activity during arrhythmia events. In some variations, the processing system may integrate the ICD EGMs with the MLiECG framework and anatomical alignment described herein.

[0357] In step 1312, the processing system may execute the beat selection module 822 to process the CEID EGMs to select arrhythmia beat EGMs. This step may involve identifying specific beats that represent the arrhythmia of interest, such as VT or PVC.

[0358] In step 1314, the processing system may determine the local activation time (LAT) of CEID lead electrodes and identify the earliest LAT. In some embodiments, this may involve analyzing the EGM waveforms to determine the timing of electrical activation at each electrode location. In some implementations, this step may be performed manually by the physician reviewing the EGM printout or display. In some implementations, this step may be performed automatically by the processing system 802 based on mathematical analysis of the EGM data. In some implementations, this step may be performed automatically by the processing system 802 and checked manually by the physician.

[0359] In step 1316, the processing system may execute the electrical conduction simulation module 820 to simulate time-shifted activation waves from each ICD lead. These simulations may use the 3D heart model (e.g., 3D heart model 900) and the determined LATs to model the propagation of electrical activity through the cardiac tissue. This processing may incorporate the patient-specific geometry and the one-or-more 3D simulations of cardiac electrophysiology. In some variations, the simulated EGM data may be integrated with the MLiECG framework and the anatomical alignment as described herein.

[0360] In step 1318, the processing system may execute the origin area simulation module 824 to use the simulated activation waves to determine a likely arrhythmia origin area. This may include simulation of each time-shifted activation wave initiated15448153-1 88Attorney Docket No.: 35313-0011WOat a time equal to a negative of the LAT of the corresponding ICD electrode. In some embodiments, this may involve the processing system analyzing the intersection of isochrones at the global activation time t=0 from multiple simulated pacing sites to identify a candidate region for the arrhythmia source. In a non-limiting embodiment, the earliest pacing stimulus may be simulated at the negative of the LAT of the ICD electrode with the earliest observed LAT, the next pacing stimulus occurs at the negative of the next-earliest observed LAT, etc., and the intersection of isochrones at t = 0 may define the search area. For example, the global activation time t = 0 may be assumed by the system to be the time of the first pacing stimuli, in which case the identified search area may be the time equal to the LAT for that electrode. The system could start the simulation of the earliest activated electrode at t = 0 rather than the negative of the observed LAT. The system could then start the next simulation at the next earliest electrode at its observed LAT. This process may be repeated for each electrode for which there is EGM data. In this embodiment, the search area may be defined by the intersection of isochrones at the t = LAT of the earliest activated lead.

[0361] In step 1320, the processing system may simulate EGMs from multiple locations within the likely arrhythmia origin area. In some embodiments, the processing system may execute the electrical conduction simulation module 820 to perform pacing simulations from multiple points in the candidate region to estimate EGM signals that would be observed at each ICD electrode location. In an alternative embodiment, the processing system may simulate pacing from multiple locations over the entire heart and skip the reverse-order simulations to reduce the search area. In some variations, the simulated EGMs from multiple locations may be integrated with the MLiECG framework and the anatomical alignment as described herein.

[0362] In step 1322, the processing system may execute the ablation location selection module 828 to compare the simulated EGMs to the recorded ICD EGMs to identify the location that is the most likely point of origin of the arrhythmia. In some embodiments, this comparison may involve using correlation metrics or other signal analysis techniques to find the best match between simulated and recorded EGMs. In15448153-1 89Attorney Docket No.: 35313-0011WOsome variations, the processing system may execute the ablation location selection module 818 to compare the simulated EGMs to the recorded EGMs as further described under the section “EGM simulations and analyses” above.

[0363] In step 1324, the processing system may output the identified most likely point of origin of the arrhythmia or use this information to conduct a cardiac ablation procedure. In some embodiments, the processing system may execute the procedure support display module 830 to generate visual representations of the 3D heart model (e.g., 3D heart model 900) with the identified arrhythmia origin highlighted, which may be presented through the display interface 840 to guide clinicians during the ablation procedure.

[0364] In some embodiments, the processing system may use the electronic storage 806 to store intermediate and final results of the analysis, and use the transceiver 808 to facilitate communication with external systems such as the electrocardiogram (ECG) system 832, ICD interface 834, and medical imaging system 836 throughout the process.

[0365] In some embodiments, the system 800 may implement another method 1400 for processing ECG data, as illustrated in FIG. 14. The method 1400 may include multiple steps for obtaining and analyzing ECG data in conjunction with other cardiac information.

[0366] In step 1402, a clinician may position ECG electrodes on a patient’s thorax and capture images of the electrodes to provide location information that the processing system can use in processing surface ECG signals as described herein. In some embodiments, the processing system may use the medical imaging system 836 to capture the positions of the ECG electrodes relative to the patient’s anatomy.

[0367] In step 1404, the processing system may obtain electrograms from the ECG electrodes. In some embodiments, the processing system may execute the ECG system 832 to record the electrocardiograms and transmit the data to the computing system 802 through the transceiver 808.15448153-1 90Attorney Docket No.: 35313-0011WO

[0368] In step 1406, the processing system may time-synchronize electrograms from the ECG electrodes to electrograms from CEID EGMs. In some embodiments, this may be accomplished by the processing system by aligning pacing stimulus signals or heart beat fiducial points (such as QRS or CEID EGM onset), or sequences of pacing stimulus and / or heart beat signals recorded by both the ECG system and the CEID.

[0369] In step 1408, the processing system may execute the beat selection module 822 to process the ICD EGMs and ECG electrograms together to select an arrhythmia beat. This step may involve analyzing both the internal (ICD) and external (ECG) electrical signals to identify specific beats that represent the arrhythmia of interest.

[0370] Optionally, in some embodiments the processing system may determine the local activation time (LAT) of selected ECG electrodes in addition to the LAT of ICD electrodes as described regarding step 1314 of the method 1300. In some embodiments, this may involve analyzing the waveforms from both ICD and ECG recordings to determine the timing of electrical activation at each electrode location.

[0371] In step 1410, the processing system may execute the electrical conduction simulation module 820 to simulate time-shifted activation waves from ICD electrodes. As ECG electrodes are far-field, there should be no time-shifting of any ECG simulations. These simulations may use the heart model (e.g., heart model 900) and the determined LATs to model the propagation of electrical activity through the cardiac tissue, incorporating both internal and external electrode data. Simulation of ICD electrodes may be performed as described above regarding step 1318 and / or 1320 of the method 1300 with reference to FIG. 13.

[0372] In some embodiments, the processing system may execute the origin area simulation module 824 to use the simulated time-shifted activation waves to refine the search area 1212 for potential arrhythmia origins. By incorporating ECG data alongside ICD data, the system 800 may achieve a more comprehensive analysis of cardiac electrical activity.15448153-1 91Attorney Docket No.: 35313-0011WO

[0373] The processing system may execute the procedure support display module 830 to generate visual representations of the ECG electrode positions and the resulting activation waves in step 1324 of the method 1300 as described. In some embodiments, these visualizations may be presented through the display interface 840, providing clinicians with additional context for interpreting the combined ICD and ECG data during ablation procedures.

[0374] In some embodiments, the processing system may execute the intracardiac electrode data download module 814 to coordinate the integration of ICD EGM data with the ECG data obtained in step 1404. The processing system may store both the raw ECG data and the processed results using electronic storage 806 for further analysis and comparison.

[0375] The processing system may execute the heart model generator module 818 to incorporate the ECG electrode positions into the patient-specific 3D heart model (e.g., 3D heart model 900), allowing for a more accurate representation of the relationship between external ECG measurements and internal cardiac structures.

[0376] In some embodiments, the processing system may execute the ablation location selection module 828 to use the combined ICD and ECG data analysis to improve the accuracy of identifying the most likely point of origin of the arrhythmia. This enhanced localization may contribute to more precise targeting during cardiac ablation procedures.

[0377] In some embodiments, the processing system may implement a method 1500 for localizing and treating cardiac arrhythmias using multiple electrode types, as illustrated in FIG. 15. The method 1500 may incorporate data from various electrode sources, including ICD electrodes, an ECG system, and EP catheter electrodes, to provide a comprehensive analysis of cardiac electrical activity.

[0378] In step 1402, the method 1500 may begin with a clinician positioning and imaging ECG electrodes on a patient’s thorax. In some embodiments, the processing15448153-1 92Attorney Docket No.: 35313-0011WOsystem may execute the medical imaging system 836 to capture the positions of the ECG electrodes relative to the patient’s anatomy.

[0379] In step 1502, EP catheters may be inserted into the patient’s heart as part of an EP procedure. In some embodiments, these EP catheters may be inserted through blood vessels and guided into specific locations within the heart chambers.

[0380] In step 1504, the processing system may localize EP catheter electrodes within the patient’s heart. In some embodiments, the processing system may execute the heart model generator module 818 to incorporate the positions of these EP electrodes into the patient-specific 3D heart model (e.g., 3D heart model 900). In some variations, the positions of the EP electrodes may be localized based on the MLiECG framework and anatomical alignment described herein.

[0381] In step 1506, the processing system may execute the intracardiac electrode data download module 814 to obtain electrograms from implantable cardioverter-defibrillator (ICD) electrodes, ECG electrodes, and EP electrodes. In some embodiments, this step may involve coordinating data collection from multiple sources, including the ICD interface 834 and the ECG system 832.

[0382] In step 1508, the processing system may execute the beat selection module 822 to process electrograms from ICD electrodes, EP electrodes, and ECG electrograms to select an arrhythmia beat. In some embodiments, this step may involve analyzing electrical signals from all available electrode types to identify specific beats that represent the arrhythmia of interest.

[0383] In step 1510, the processing system may determine the local activation time (LAT) of ICD electrodes, EP electrodes, and optionally selected ECG electrodes. In some embodiments, this may involve analyzing the waveforms from all available electrode recordings to determine the timing of electrical activation at each electrode location.

[0384] In step 1512, the processing system may execute the electrical conduction simulation module 820 to simulate time-shifted activation waves from ICD electrodes,15448153-1 93Attorney Docket No.: 35313-0011WOEP electrodes, and optionally selected ECG electrodes. In some embodiments, these simulations may use the 3D heart model (e.g., heart model 900) and the determined LATs to model the propagation of electrical activity through the cardiac tissue, incorporating data from all available electrode types.

[0385] In step 1318, the processing system may execute the origin area simulation module 824 to use the simulated activation waves to determine a likely arrhythmia origin area. In some embodiments, this may involve analyzing the intersection of isochrones at t=0 from multiple simulated pacing sites to identify a candidate region for the arrhythmia source. Simulation of ICD electrodes may be performed as described above regarding the same numbered step of the method 1300 with reference to FIG. 13.

[0386] In step 1320, the processing system may simulate electrograms (EGMs) from multiple locations within the likely arrhythmia origin area. In some embodiments, the processing system may execute the electrical conduction simulation module 820 to perform pacing simulations from multiple points in the candidate region to estimate EGM signals that would be observed at each electrode location.

[0387] In step 1514, the processing system may execute the ablation location selection module 828 to compare the simulated electrograms to the recorded electrograms from the ICD electrodes, EP electrodes, and optionally selected ECG electrodes to identify the location that is the most likely point of origin of the arrhythmia. In some embodiments, this comparison may involve using correlation metrics or other signal analysis techniques to find the best match between simulated and recorded EGMs across all available electrode types.

[0388] In some embodiments, as part of the operations in step 1514, that processing system may also perform vectorgram analysis as described with reference to FIG. 10B based on ECG electrode position information determined from imaging of the patient’s thorax and ECG data from the ECG electrodes to identify a most likely point of origin of the arrhythmia. The processing system may then combine and / or correlate15448153-1 94Attorney Docket No.: 35313-0011WOthis ECG-based identified point of origin with the most likely point of origin determined based on ICD and / or EP electrode EGMs in order to reduce the uncertainty or positional error around the arrythmia origin point. Similarly, the processing system may determine the most likely origin point based solely on EP electrode EGMs, and combine and / or correlate that EP -based identified point of origin with the point of origin determine based on ICD and EP electrode EGMs in order to reduce the uncertainty or positional error around the arrythmia origin point. By correlating results of using the different electrode systems and geometries to identify the most likely point of arrythmia point of origin according to the methods described herein, the processing system may identify a more accurate and precise location for performing an ablation treatment.

[0389] In some embodiments, the processing system may execute the procedure support display module 830 to generate visual representations of the 3D heart model (e.g., heart model 900) with the identified arrhythmia origin highlighted in block 1324 of the method 1300 as described. These visualizations may be presented through the display interface 840 to guide clinicians during ablation procedures.

[0390] The processing system may store intermediate and final results of the analysis, including data from all electrode types in the electronic storage 806.

[0391] In some embodiments, the system 800 may implement a method 1600 that includes updating the 3D heart model based on results of the method operations as illustrated in FIG. 16. The method 1600 may include multiple steps for processing cardiac data and guiding ablation procedures.

[0392] In steps 1506-1514, the processing system may perform operations of the like numbered steps of the method 1500 as described above with reference to FIG. 15, and in steps 1318 and 1320, the processing system may perform operations of the like numbered steps of the method 1300 as described above with reference to FIG. 13.

[0393] In step 1602, the processing system may execute the heart model generator module 818 to update the 3D heart model (e.g., heart model 900) and / or activation15448153-1 95Attorney Docket No.: 35313-0011WOwave simulation model to account for ablated tissues or other scar tissue identified prior to or during the EP procedure. For instance, the processing system may compute a distance matrix as described above. In some embodiments, the processing may incorporate scar tissue information to alter distances and / or activation wave behavior in the heart model used for simulations. This updating process may occur during the EP procedure as new information about tissue characteristics becomes available. In some cases, the model may be updated during an EP procedure and used to further localize arrhythmia initiation locations by the processing system repeating the method, such as by repeating the operations in step 1506.

[0394] The processing system 810 may coordinate the integration and analysis of data from multiple electrode types, as well as the incorporation of scar tissue information and model updates, to provide a comprehensive and dynamic assessment of cardiac electrical activity and arrhythmia localization throughout the EP procedure.

[0395] In some embodiments, the system 800 may implement a method 1700 for localizing and treating cardiac arrhythmias using electrophysiology electrodes without EGM data from ICD leads, as illustrated in FIG. 18. This method 1800 may be useful in circumstances when a patient is undergoing an EP procedure to diagnose and / or treat an arrhythmia (e.g., a VT or PVC) without an ICD implanted or with an implanted ICD that is not capable of or did not record ICD EGMs. The method 1300 may be implemented in a processing system 810 of a computing system 802 by implementing processor-executable instructions 814-830.

[0396] In step 1302, the processing system may obtain medical imaging of a patient's heart. As described for the like numbered step in the method 1300, the processing system may execute the medical imaging download module 816 to receive imaging data from the medical imaging system 836, which may include CT and / or MRI scans.

[0397] In step 1304, the processing system may execute the heart model generator module 818 to generate a patient-specific 3D heart model based on the medical15448153-1 96Attorney Docket No.: 35313-0011WOimaging and a selected representative 3D heart model, as described for the like numbered step in the method 1300.

[0398] In step 1702, the processing system may execute the method 1700 to involve positioning electrophysiology (EP) electrode catheters in the patient's heart. In some embodiments, these EP catheters may be inserted through a main artery and guided into specific locations within the heart chambers.

[0399] In step 1704, the processing system may localize EP electrodes within the patient’s heart and then within the 3D heart model. In some embodiments, the processing system may execute the heart model generator module 818 to incorporate the positions of these EP electrodes into the patient-specific 3D heart model 900.

[0400] In step 1706, the processing system may execute the intracardiac electrode data download module 814 to obtain EP EGMs from the EP system. In some embodiments, this step may involve coordinating data collection from the EP system interface.

[0401] In step 1708, the processing system may execute the beat selection module 822 to process the EP electrode EGMs to select arrhythmia beat ventricular EGM complexes. This step may involve identifying specific beats that represent the arrhythmia of interest, such as VT or PVC.

[0402] In step 1710, the processing system may execute the processing system to determine the local activation time (LAT) of EP electrodes and identify the earliest LAT. In some embodiments, this may involve analyzing the EGM waveforms to determine the timing of electrical activation at each electrode location.

[0403] In step 1712, the processing system may execute the electrical conduction simulation module 820 to simulate time-shifted activation waves from each EP electrode. These simulations may use the 3D heart model 900 and the determined LATs to model the propagation of electrical activity through the cardiac tissue.Simulation of ICD electrodes may be performed as described above regarding step 1318 of the method 1300 with reference to FIG. 13.15448153-1 97Attorney Docket No.: 35313-0011WO

[0404] In step 1714, the processing system may execute the origin area simulation module 824 to use the simulated activation waves to determine a likely arrhythmia origin area. In some embodiments, this may involve analyzing the intersection of isochrones at t=0 from multiple simulated pacing sites to identify a candidate region for the arrhythmia source.

[0405] In step 1716, the processing system may simulate EGMs at each EP electrode from multiple locations within the likely arrhythmia origin area. In some embodiments, the processing system may execute the electrical conduction simulation module 820 to perform pacing simulations from multiple points in the candidate region to estimate EGM signals that would be observed at each EP electrode location.

[0406] In step 1718, the processing system may execute the ablation location selection module 828 to compare the simulated EGMs at each EP electrode to the recorded EP electrode EGMs to identify the location that is the most likely point of origin of the arrhythmia. In some embodiments, this comparison may involve using correlation metrics or other signal analysis techniques to find the best match between simulated and recorded EGMs.

[0407] In step 1720, the processing system may output the identified most likely point of origin of the arrhythmia or use this information to conduct a cardiac ablation procedure. In some embodiments, the processing system may execute the procedure support display module 830 to generate visual representations of the 3D heart model 900 with the identified arrhythmia origin highlighted, which may be presented through the display interface 840 to guide clinicians during the ablation procedure.

[0408] Additionally, the processing system may use information obtained from ECG electrodes on the patient’s thorax in operations of the method 1400 as described with reference to FIG. 14. Further, the processing system may update the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during the EP procedure in operations of step 1602 of the method 1600 as described with reference to FIG. 16.15448153-1 98Attorney Docket No.: 35313-0011WO

[0409] As with other methods, the processing system may store intermediate and final results of the analysis in the electronic storage 806.

[0410] In various embodiments, the system 800 may be implemented in any of a wide variety of various computing devices, such as, but not limited to, a laptop computer 1800 illustrated in FIG. 18, and / or a server 1900 illustrated in FIG. 19.

[0411] As an example of a standalone computing device suitable for implementing various embodiments, a laptop computer 1800 may include a circuit board 1802 that serves as a foundation for various internal components. A memory module 1808 and a processor 1812 may be mounted on the circuit board 1802 to provide computational capabilities. The laptop computer 1800 may also include a power supply 1813 for providing electrical power to the components, a storage device 1814 for data storage, and an expansion slot 1815 for adding additional functionality.

[0412] In some embodiments, the laptop computer 1800 may incorporate a cooling system 1816 to manage thermal output and maintain optimal operating temperatures for the internal components. User interface elements of the laptop computer 1800 may include a touchpad 1817, a keyboard 1818, and a display screen 1819, allowing for user interaction with the system 800.

[0413] A server 1900 may be used for performing for more intensive computational tasks and may include a multicore processor assembly 1901 to handle complex calculations required for cardiac arrhythmia localization and treatment analysis. The server 1900 may be equipped with a local network interface 1902 for connectivity, enabling communication with other components of the system 800.

[0414] In some embodiments, the server 1900 may include expansion ports 1903 for adding additional hardware capabilities and a disk drive 1904 for local data storage. The server 1900 may be connected to a local area network 1905, facilitating data exchange with other devices in the system 800. Large capacity storage 1906 may be incorporated into the server 1900 to accommodate extensive medical imaging data and patient records.15448153-1 99Attorney Docket No.: 35313-0011WO

[0415] In some implementations a local computer, such as the laptop computer 1800, and server 1900 may work in conjunction to implement various modules of the system 800, including the intracardiac electrode data download module 814, medical imaging download module 816, heart model generator module 818, electrical conduction simulation module 820, beat selection module 822, origin area simulation module 824, ablation location selection module 828, and procedure support display module 830.

[0416] In some embodiments, the laptop computer 1800 may serve as a user interface for clinicians, displaying the 3D heart model 900 and providing interactive features through the display interface 840. The server 1900 may handle the more computationally intensive tasks, such as generating the patient-specific 3D heart model 900 and performing electrical conduction simulations.

[0417] The system 800 may interface with electrophysiology (EP) systems to provide a rotatable 3D heart model 900 correlated to electrode positions. In some embodiments, the procedure support display module 830 may generate visual representations of the 3D heart model 900 that can be manipulated and rotated on the display screen 1819 of a display, such as on the laptop computer 1800. This rotatable model may be synchronized with real-time electrode position data from the EP system, allowing clinicians to visualize the spatial relationships between electrodes and cardiac structures during procedures.

[0418] In some embodiments, the system 800 may interface with robotic catheter systems for automated or semi-automated ablation procedures. The server 1900 may process data from the origin area simulation module 824 and ablation location selection module 828 to generate control signals for the robotic catheter system. These control signals may guide the positioning and activation of ablation catheters based on the identified most likely point of origin of the arrhythmia.

[0419] The electronic storage 806 of the system 800 may be distributed across the storage device 1814 of the laptop computer 1800 and the large capacity storage 190615448153-1 100Attorney Docket No.: 35313-0011WOof the server 1900. This distributed storage architecture may allow for efficient data management and rapid access to patient information and analysis results.

[0420] In some embodiments, the transceiver 808 of the system 800 may be implemented using the local network interface 1902 of the server 1900 and wireless communication capabilities of the laptop computer 1800. This configuration may enable seamless communication between the computing devices and external systems such as the ECG system 832, ICD interface 834, and medical imaging system 836.

[0421] The processing system 810 of the system 800 may leverage the computational power of both the processor 1812 in the laptop computer 1800 and the multicore processor assembly 1801 in the server 1800. This distributed processing approach may allow for real-time analysis of cardiac data and rapid generation of simulation results during EP procedures.

[0422] The computing devices described may be used individually or in combination to implement the system 800 and perform the operations of the methods 1300, 1400, 1500, 1600, and 1700. The specific hardware configuration may be adapted based on the computational requirements of the particular implementation and the needs of the users.

[0423] Implementation examples are described in the following paragraphs. While some of the following implementation examples are described in terms of example methods, further example implementations may include: the example methods discussed in the following paragraphs implemented by a computing system including a processing system (e.g., with processor-executable instructions) that may perform operations of the methods of the following implementation examples; and the example methods discussed in the following paragraphs may be implemented as a non-transitory processor-readable storage medium having stored thereon processorexecutable instructions to cause a processing system of a computing system to perform the operations of the methods of the following implementation examples.15448153-1 101Attorney Docket No.: 35313-0011WO

[0424] As used in this application, terminology such as “unit,” “component,” “module,” “system,” etc., is intended to encompass a software-implemented or computer-related entity. These entities may involve, among other possibilities, hardware, firmware, a blend of hardware and software, software alone, or software in an operational state. As examples, a component may encompass a running process on a processor, the processing system itself, an object, an executable file, a thread of execution, a program, or a computing device. To illustrate further, both an application operating on a computing device and the computing device itself may be designated as a component. A component might be situated within a single process or thread of execution or could be distributed across multiple processors or cores. In addition, these components may operate based on various non-volatile computer-readable media that store diverse instructions and / or data structures. Communication between components may take place through local or remote processes, function or procedure calls, electronic signaling, data packet exchanges, and memory interactions, among other known methods of network, computer, processor, or process-related communications.

[0425] Various embodiments illustrated and described are provided merely as examples to illustrate various features of the claims. However, features shown and described with respect to any given embodiment are not necessarily limited to the associated embodiment and may be used or combined with other embodiments that are shown and described. Further, the claims are not intended to be limited by any one example embodiment. For example, one or more of the operations of the methods may be substituted for or combined with one or more operations of the methods.

[0426] The foregoing method descriptions and the process flow diagrams are provided merely as illustrative examples and are not intended to require or imply that the operations of various embodiments must be performed in the order presented. As will be appreciated by one of skill in the art, the order of operations in the foregoing embodiments may be performed in any order. Words such as “thereafter,” “then,” “next,” etc. are not intended to limit the order of the operations; these words are15448153-1 102Attorney Docket No.: 35313-0011WOsimply used to guide the reader through the description of the methods. Further, any reference to claim elements in the singular, for example, using the articles “a,” “an,” or “the,” is not to be construed as limiting the element to the singular.

[0427] The various illustrative logical blocks, modules, circuits, and algorithm operations described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and operations have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the claims.

[0428] A number of different types of memories and memory technologies are available or contemplated in the future, any or all of which may be included and used in systems and computing devices that implement the various embodiments.

[0429] In one or more embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored as one or more instructions or code on a non-transitory computer-readable medium or non-transitory processor-readable medium. The operations of a method or algorithm disclosed herein may be embodied in a processor-executable software module, which may reside on a non-transitory computer-readable or processor-readable storage medium. Non-transitory computer-readable or processor-readable storage media may be any storage media that may be accessed by a computer or a processor. By way of example but not limitation, such non-transitory computer-readable or processor-readable media may include random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), FLASH memory, solid-state drives 15448153-1 103Attorney Docket No.: 35313-0011WO(SSD), non-volatile memory express (NVMe) drives, or any other medium that may be used to store program code in the form of processor-executable instructions or data structures and that may be accessed by a processor of a computing device. Modern technologies, such as cloud-based storage solutions, including infrastructure-as-a-service (IaaS) platforms, may offer scalable and distributed options for storing and accessing program code.

[0430] In addition, the operations of a method or algorithm may reside as one or any combination or set of codes and / or instructions on a non-transitory processor-readable medium and / or computer-readable medium, which may be incorporated into a computer program product. Emerging technologies, including quantum computing storage media and blockchain-based storage solutions, may further enhance data integrity and security. Artificial intelligence (Al) and machine learning (ML)-optimized hardware accelerators, such as graphical processing systems (GPUs) and tensor processing systems (TPUs), may be used to execute complex algorithms.

[0431] The preceding description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the claims. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the claims. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the following claims and the principles and novel features disclosed herein.15448153-1 104

Claims

Attorney Docket No.: 35313-0011WOCLAIMS1. A system for localizing cardiac arrhythmias, comprising:a memory configured to store a patient-specific three-dimensional (3D) heart model and electrogram data;an interface configured to receive implantable cardioverter-defibrillator (ICD) electrograms (EGMs) from electrodes positioned within a patient’s heart; anda processing system configured to:simulate, using the stored model, ICD EGMs that would be observed at each ICD electrode resulting from propagation of activation waves initiated at each of multiple locations within a likely arrhythmia origin area;compare the simulated ICD EGMs to the recorded ICD EGMs; identify a location that is a most likely point of origin of the arrhythmia based on the comparison; andoutput the identified most likely point of origin of the arrhythmia.

2. The system of claim 1, wherein the processing system is further configured to generate the patient-specific 3D heart model based on medical imaging of the patient’s heart;process the received ICD EGMs to select an arrhythmia beat;determine for each ICD electrode a local activation time (LAT) of the arrhythmia beat and identify an earliest global activation time;simulate time-shifted activation waves through the 3D heart model in response to pacing at each ICD electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding ICD electrode;determine the likely arrhythmia origin area based on the simulated time-shifted activation waves.15448153-1 105Attorney Docket No.: 35313-0011WO3. The system of either of claims 1 or 2, wherein the processing system is further configured to simulate time-shifted activation waves through the 3D heart model by using the 3D heart model to simulate progression of an activation wave front through heart tissues from a pacing or initiation at an electrode location.

4. The system of any of claims 1-3, wherein the processing system is further configured to include locations of ICD electrodes within the 3D heart model based on the medical imaging, and wherein simulations of time-shifted activation waves and simulations of EGMs that would be observed at each ICD electrode from initiation at multiple locations within the likely arrhythmia origin area uses the location of each ICD electrode in the 3D heart model.

5. The system of any of claims 1-4, wherein the processing system is configured to determine the local activation time (LAT) of the ICD electrodes by identifying a time of earliest significant change in potential in one of the ICD EGMs.

6. The system of any of claims 1-5, wherein the processing system is configured to determine the likely arrhythmia origin area by identifying an intersection of isochrones at t=0 from the simulated time-shifted activation waves.

7. The system of any of claims 1-6, wherein:the interface is further configured to receive electrograms from electrocardiogram (ECG) electrodes positioned on the patient’s thorax; andthe processing system is further configured to:process the ICD EGMs and ECG electrograms together to select the arrhythmia beat;determine LAT of at least one ECG electrode;15448153-1 106Attorney Docket No.: 35313-0011WOsimulate time-shifted activation waves from the at least one ECG electrode, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding ECG electrode;simulate ECG signals that would be observed at each ECG electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area; andcompare the simulated ECG signals to the recorded ECG signals, wherein identifying a location that is the most likely point of origin of the arrhythmia comprises identifying the initiation location of the best matching simulated ICD EGM or ECG signals as the most likely point of origin of the arrhythmia.

8. The system of any preceding claim, wherein:the interface is further configured to receive electrograms from electrophysiology (EP) catheter electrodes positioned within the patient’s heart; and the processing system is further configured to:localize EP catheter electrodes within the patient’s heart;process the ICD EGMs and EP electrode electrograms together to select the arrhythmia beat;determine LAT of the EP electrodes;simulate time-shifted activation waves from the EP electrodes, wherein simulations of each time-shifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding EP electrode;simulate EP EGMs that would be observed at each EP electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area; andcompare the simulated EP EGMs to the recorded EP EGMs, wherein identifying a location that is the most likely point of origin of the arrhythmia comprises identifying the initiation location of the best15448153-1 107Attorney Docket No.: 35313-0011WOmatching simulated ICD EGM or EP EGM as the most likely point of origin of the arrhythmia.

9. The system of any preceding claim, wherein the processing system is further configured to update the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during an EP procedure.

10. The system of claim 8, wherein the processing system is further configured to calibrate the 3D heart model by:outputting signals for controlling pacing of the heart at multiple locations during an EP procedure;recording ICD EGMs during the pacing; andadjusting conduction velocity parameters in the 3D heart model based on measured activation times between the pacing location and ICD electrodes.

11. The system of any preceding, wherein the processing system is configured to determine the LAT of electrodes by identifying a minimum of a first derivative of an EGM of the electrode.

12. The system of any preceding claim, wherein:the ICD EGMs are obtained from a cardiac resynchronization therapy (CRT) device having electrodes in both right and left ventricles; andthe processing system is configured to determine the likely arrhythmia origin area based on the simulated time-shifted activation waves from the electrodes in both right and left ventricles to reduce a size of the likely arrhythmia origin area.

13. The system of any preceding claim, further comprising a display, wherein the processing system is configured to output the identified most likely point of origin of15448153-1 108Attorney Docket No.: 35313-0011WOthe arrhythmia by displaying the 3D heart model with the identified most likely point of origin of the arrhythmia highlighted for use during the cardiac ablation procedure.

14. A system for localizing and treating cardiac arrhythmias using electrophysiology electrodes, comprising:a memory configured to store a patient-specific three-dimensional (3D) heart model and electrogram data;an interface configured to receive electrophysiology (EP) electrode electrograms (EGMs) from an EP system; anda processing system configured to:generate the patient-specific 3D heart model based on medical imaging of the patient's heart and a selected representative 3D heart model;localize EP electrodes within the patient's heart and within the 3D heart model;process the EP electrode EGMs to select an arrhythmia beat; determine for each EP electrode a local activation time (LAT) of the arrhythmia beat and identify an earliest LAT (t=0);simulate time-shifted activation waves through the 3D heart model in response to pacing at each EP electrode, wherein simulations of each timeshifted activation wave is initiated at a time equal to a negative of the LAT of the corresponding EP electrode;determine a likely arrhythmia origin area based on the simulated timeshifted activation waves;simulate EP electrode EGMs that would be observed at each EP electrode resulting from propagation of activation waves initiated at each of multiple locations within the likely arrhythmia origin area;compare the simulated EP electrode EGMs to recorded EP electrode EGMs;15448153-1 109Attorney Docket No.: 35313-0011WOidentify a location that is a most likely point of origin of the arrhythmia based on the comparison; andoutput the identified most likely point of origin of the arrhythmia for use in conducting a cardiac ablation procedure.

15. The system of claim 14, wherein the processing system is further configured to output control signals suitable for controlling conducting the ablation procedure.

16. The system of claims 14 or 15, wherein the processing system is configured to simulate time-shifted activation waves through the 3D heart model by using the 3D heart model to simulate progression of the activation wave front through heart tissues from a pacing or initiation at an EP electrode location.

17. The system of claims 14, 15 or 16, wherein the processing system is configured to determine the LAT of each EP electrode by identifying a time of earliest significant change in potential in each EP electrode EGM.

18. The system of any of claims 14-17, wherein the processing system is configured to determine the likely arrhythmia origin area by identifying an intersection of isochrones at t=0 from the simulated time-shifted activation waves.

19. The system of any of claims 14-18, wherein the processing system is configured to:compare the simulated EP electrode EGMs to the recorded EP electrode EGMs by using a correlation metric to identify a best match; andidentify a location that is a most likely point of origin of the arrhythmia by identifying the initiation location of the best matching simulated EP electrode EGM as the most likely point of origin of the arrhythmia.15448153-1 110Attorney Docket No.: 35313-0011WO20. The system of any of claims 14-19, wherein the processing system is configured to simulate EP electrode EGMs from multiple locations by simulating EP electrode EGMs from all grid points in the 3D heart model within the likely arrhythmia origin area.

21. The system of any of claims 14-20, wherein the processing system is further configured to update the 3D heart model and / or activation wave simulation model to account for ablated tissues or other scar tissue identified during an EP procedure.

22. The system of any of claims 14-21, wherein the processing system is further configured to calibrate the 3D heart model by:outputting signals suitable for controlling pacing of the heart at multiple locations during an EP procedure;recording EP electrode EGMs during the pacing; andadjusting conduction velocity parameters in the 3D heart model based on measured activation times between the pacing location and EP electrodes.

23. The system of any of claims 14-22, wherein the processing system is configured to determine the LAT of each EP electrode by identifying a minimum of a first derivative of an EGM of each EP electrode.

24. The system of any of claims 14-23, further comprising a display, wherein the processing system is configured to output the identified most likely point of origin of the arrhythmia by displaying the 3D heart model with the identified most likely point of origin of the arrhythmia highlighted for use during the cardiac ablation procedure.15448153-1 111Attorney Docket No.: 35313-0011WO25. A computing system, comprising:a memory; anda processor coupled to the memory and configured with processor-executable instructions, the instructions, when executed by the processor, configured to:obtain electrical signals and location information from electrodes positioned within the patient’s heart;generate a patient-specific three-dimensional (3D) heart model;generate a patient-specific electrical conduction map of the patient’s heart during an arrhythmia based on the patient-specific 3D heart model and the electrical signals and location information from electrodes positioned within the patient’s heart, the patient- specific electrical conduction map of the patient’s heart identifying a 3D localization of an initiation site of the arrhythmia;display the 3D localization of the initiation site of the arrhythmia for use by a physician preparing for and / or during a cardiac electrophysiology procedure; and output signals suitable for conducting a cardiac ablation procedure using the displayed 3D localization of the initiation site of the arrhythmia to identify one or more locations for ablation.

26. The system of claim 25, the instructions, when executed, further configured to cause the processor to, when obtaining electrical signals and location information from electrodes positioned within the patient’s heart, receive signals from electrodes on one or more electrophysiology (EP) catheters temporarily positioned with the heart, determining locations of the electrodes within the heart based on information from an EP system.

27. The system of any of claims 25 or 26, the instructions, when executed, further configured to cause the processor to, when obtaining electrical signals and location information from electrodes positioned within the patient’s heart, download from an15448153-1 112Attorney Docket No.: 35313-0011WOimplanted cardiac defibrillator (ICD) information recorded during detected arrhythmia events in the patient’s heart;obtain medical imaging that identifies locations of one or more ICD electrodes in the patient’s heart;generate the patient-specific three-dimensional (3D) heart model including locations of the one or more ICD electrodes on the heart.

28. The system of any of claims 25 to 27, the instructions, when executed, further configured to cause the processor to conduct point by point contact electrophysiology recordings during the cardiac ablation procedure and updating the patient-specific electrical conduction map to display an updated 3D localization of the initiation site of the arrhythmia.

29. The system of any of claims 25 to 28, wherein generating a patient- specific 3D model of the heart including a 3D internal surface model comprises using magnetic resonance imaging (MRI) or computed tomography (CT) images of the patient to generate the patient-specific 3D heart model including locations of the one or more ICD electrodes.

30. The system of any of claims 25 to 29, the instructions, when executed, further configured to cause the processor to combineICD information recorded during detected arrhythmia events in the patient’s heart with the patient-specific 3D heart model to identify isochrones at time intervals of heartbeats to determine directions of depolarization wavefronts to reveal an initiation site of the arrhythmia in a heartbeat.

31. The system of any of claims 25 to 30, the instructions, when executed, further configured to cause the processor to display heart structures including one or more of the aorta, aortic arch, pulmonary veins, or coronary vessels on the displayed 3D heart model.15448153-1 113Attorney Docket No.: 35313-0011WO32. The system of any of claims 25 to 31, the instructions, when executed, further configured to cause the processor to display heart scar tissue indicative of ischemic heart disease on the displayed 3D heart model.

33. The system of any of claims 25 to 32, the instructions, when executed, further configured to cause the processor to display the localization of the arrhythmia as multiple points representative of multiple beats of a ventricular tachycardia on the displayed 3D heart model.

34. The system of any of claims 25 to 33wherein the arrhythmia is one of an atrial arrhythmia, a ventricular arrhythmia, a pre-ventricular contraction (PVC), a ventricular tachycardia, or a dysrhythmia between the two ventricles.

35. The system of any of claims 25 to 34, the instructions, when executed, further configured to cause the processor to:obtain a 3D image of electrocardiogram (ECG) electrodes on the patient’s torso during an ECG recording procedure; andmerge the 3D image of the patient’s torso with the 3D heart image, wherein generating a patient-specific electrical conduction map of the patient’s heart during an arrhythmia is based on the patient-specific 3D heart model, the electrical signals and location information from electrodes positioned within the patient’s heart, and ECG data from the ECG recording procedure.15448153-1 114

Images