Intracardiac unipolar far field cancellation using multiple-electrode catheters and methods for creating ECG depth and radial lens

Abstract

To provide an electroanatomical mapping system for isolating intracardiac ECG (IECG) signals at various depths within myocardial tissue.SOLUTION: The invention provides embodiments of methods and systems for determining electrocardiogram (ECG) depth of electrical activity. Electrical activity from a plurality of electrodes of a catheter is received and, within the electrical activity, an electrical signal originating from a depth within cardiac tissue is identified. In some embodiments, spatial electrode signal analysis of the electrical activity may be performed for each electrode of the plurality of electrodes. In some embodiments, a linear and / or non-linear combination of signal components within the electrical activity may be calculated. In some embodiments, a neural network may be provided with the electrical signals and may determine an estimated distance to the nearest activation for at least one electrode of the plurality of electrodes. A visualization of the identified electrical signals at a specified depth within the cardiac tissue may be provided for display.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application claims the benefit of priority as a continuation-in-part of U.S. Provisional Application No. 18 / 072,793, entitled "Intracardiac Unipolar Far Field Cancellation Using Multiple Electrode Catheters," filed December 1, 2022, which claims priority to U.S. Provisional Application No. 63 / 288,966, filed December 13, 2021, and also claims priority to U.S. Provisional Application No. 63 / 523,976, entitled "Intracardiac Unipolar Far Field Cancellation Using Multiple Electrode Catheters and Methods for Creating an ECG Depth and Radial Lens," filed June 29, 2023. Each of the above-mentioned applications is incorporated herein by reference in its entirety.

[0002] FIELD OF THEINVENTION The systems and methods herein relate generally to the field of cardiac electrophysiology, and more specifically to an electroanatomical mapping system for isolating intracardiac ECG (IECG) signals at various depths within myocardial tissue. [Background technology]

[0003] In general, conventional cardiac mapping is a computer-processed operation that generates a three-dimensional map of the heart that medical professionals can use to determine the exact location of the source of an arrhythmia or when performing medical procedures such as cardiac ablation to treat atrial fibrillation (aFib).

[0004] As an example, in conventional cardiac mapping, a three-dimensional map of the heart is created as a medical professional (e.g., a physician) guides a catheter through a patient's blood vessels until the catheter enters the heart. The catheter senses electrical activity, and computer processing operations analyze the electrical activity to generate a three-dimensional map of the heart. A problem with conventional cardiac mapping is that far-field signals mask or interfere with local field signals in the electrical activity. Furthermore, conventional computer processing operations are not well suited to deal with far-field signals. For example, computer processing operations assume that all far-field signals are the same on all electrodes of the catheter. Summary of the Invention [Problem to be solved by the invention]

[0005] What is needed is a system and method for extracting and analyzing electrical activity from a catheter while reducing or canceling interference from the far field.

[0006] In addition, intracardiac ECGs collected during cardiac ablation procedures are two-dimensional and reflect electrical activity / voltage only at the surface of the intracardiac tissue. A need exists for a means to identify and visualize in three dimensions the ECG signal that reflects electrical conduction below the surface of the myocardial tissue (e.g., 0.5-10 mm deep within the cardiac tissue). [Means for solving the problem]

[0007] Exemplary embodiments, methods, and systems are provided. Methods and systems for performing these methods may be implemented by a mapping engine executed by one or more processors. The method may include receiving electrical activity from a plurality of electrodes of a catheter, and identifying within the electrical activity electrical signals originating from a depth within the cardiac tissue. The method may include performing a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes, and performing a linear and in some embodiments a non-linear combination of signal components within the electrical activity. In some embodiments, identifying electrical signals originating from a depth within the cardiac tissue may include receiving real-time ECG readings from the plurality of electrodes as input to a neural network, and determining, by the neural network, an estimated distance to a nearest activation for at least one of the plurality of electrodes. In some embodiments, identifying electrical signals originating from a depth within the cardiac tissue may include applying, for each of the plurality of electrodes, a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position. In some embodiments, the method may include visualizing on a display the identified electrical signals at the specified depth within the cardiac tissue. According to one or more embodiments, the example methods described above may be implemented as an apparatus, a system, and / or a computer program product.

[0008] In some aspects, the disclosure is directed to a method. The method includes receiving, by a mapping engine executed by one or more processors, electrical activity from a plurality of electrodes of a catheter and identifying, within the electrical activity, electrical signals originating from a depth within the cardiac tissue. In some implementations, the method includes identifying, within the electrical activity, electrical signals originating from an endocardial surface of the cardiac tissue. In some implementations, the method includes performing, by the mapping engine, a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes, where identifying the electrical signals originating from a depth within the cardiac tissue includes calculating a linear combination of signal components within the electrical activity. In a further implementation, the method includes calculating the linear combination of signal components within the electrical activity by multiplying each signal component within the electrical activity by a corresponding weight determined via the trained model. In another further implementation, the method includes identifying the electrical signals originating from a depth within the cardiac tissue by calculating a nonlinear combination of signal components within the electrical activity.

[0009] In some implementations, the method includes identifying electrical signals originating from a depth within the cardiac tissue by providing electrical activity from a plurality of electrodes as input to a neural network, and determining, via the neural network, for at least one of the plurality of electrodes, an estimated distance from a corresponding electrode to a nearest activation.

[0010] In some implementations, the method includes identifying, for each of a plurality of electrodes, electrical signals emanating from a depth within the cardiac tissue by applying a mathematical model and known electrode locations of the plurality of electrodes to calculate an expected signal at each known electrode location. In some implementations, the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes. In some implementations, identifying electrical signals emanating from a depth within the cardiac tissue includes running a machine learning model trained on simulated ECG data. In some implementations, the method includes generating, by a mapping engine, a visual display of the identified electrical signals at the specified depth within the cardiac tissue.

[0011] In another aspect, the present disclosure is directed to a system including a memory storing mapping engine software and one or more processors configured to execute the mapping engine to receive electrical activity from multiple electrodes of a catheter and to identify, within the electrical activity, electrical signals originating from a depth within the cardiac tissue.

[0012] In some implementations, the one or more processors are further configured to identify electrical signals within the electrical activity originating from an endocardial surface of the cardiac tissue. In some implementations, the one or more processors are further configured to perform a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes, and identifying the electrical signals originating from a depth within the cardiac tissue includes computing a linear combination of signal components within the electrical activity. In a further implementation, computing the linear combination of signal components within the electrical activity includes multiplying each signal component within the electrical activity by a corresponding weight determined via the trained model. In another further implementation, identifying the electrical signals originating from a depth within the cardiac tissue includes computing a non-linear combination of signal components within the electrical activity.

[0013] In some implementations, the one or more processors are further configured to provide the electrical activity from the plurality of electrodes as an input to the neural network and determine, for at least one of the plurality of electrodes, an estimated distance from a corresponding electrode to a nearest activation via the neural network. In some implementations, the one or more processors are further configured to, for each of the plurality of electrodes, apply the mathematical model and the known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position. In some implementations, the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes. In some implementations, identifying the electrical signal originating from a depth within the cardiac tissue includes running a machine learning model trained on the simulated ECG data.

[0014] In another aspect, the disclosure is directed to a method. The method includes receiving, by a mapping engine (101) executed by one or more processors (61), electrical activity from a plurality of electrodes (26) of a catheter (14) and identifying, within the electrical activity, electrical signals originating from a depth (820) within the cardiac tissue (12). In some implementations, the method includes identifying, within the electrical activity, electrical signals originating from an endocardial surface of the cardiac tissue (12). In some implementations, the method includes performing, by the mapping engine (101), a spatial electrode signal analysis of the electrical activity for each electrode (26) of the plurality of electrodes (26), where identifying the electrical signals originating from a depth (820) within the cardiac tissue (12) includes calculating a linear combination of signal components within the electrical activity. In further implementations, calculating the linear combination of signal components within the electrical activity includes multiplying each signal component within the electrical activity by a corresponding weight determined via the trained model. In another further implementation, identifying the electrical signal originating from a depth 820 within the cardiac tissue 12 includes computing a nonlinear combination of signal components within the electrical activity.

[0015] In some implementations, identifying the electrical signal originating from the depth (820) within the cardiac tissue (12) includes providing electrical activity from the plurality of electrodes (26) as input to a neural network, and determining, via the neural network, for at least one of the plurality of electrodes (26), an estimated distance to a nearest activation from a corresponding electrode (26). In some implementations, identifying the electrical signal originating from the depth (820) within the cardiac tissue (12) includes, for each electrode (26) of the plurality of electrodes (26), applying the mathematical model and known electrode positions of the plurality of electrodes (26) to calculate an expected signal at each known electrode position.

[0016] In some implementations, the received electrical activity includes real-time electrocardiogram (ECG) readings from the multiple electrodes (26). In some implementations, identifying the electrical signals originating from a depth within the cardiac tissue includes running a machine learning model trained on the simulated ECG data. In some implementations, the method includes generating, by the mapping engine (101), a visual representation (20) of the identified electrical signals at the specified depth within the cardiac tissue (12).

[0017] In another aspect, the present disclosure is directed to a system including a memory (62) storing mapping engine (101) software, and one or more processors (61) configured to execute the mapping engine (101) to receive electrical activity from a plurality of electrodes (26) of a catheter (14) and to identify, within the electrical activity, electrical signals originating from a depth (802) within cardiac tissue (12).

[0018] In some implementations, the one or more processors (61) are further configured to identify, within the electrical activity, electrical signals originating from an endocardial surface of the cardiac tissue (12). In some implementations, the one or more processors (61) are further configured to perform a spatial electrode signal analysis of the electrical activity for each electrode (26) of the plurality of electrodes (26), and identifying the electrical signals originating from a depth (820) within the cardiac tissue (12) includes computing a linear combination of signal components within the electrical activity. In a further implementation, computing the linear combination of signal components within the electrical activity includes multiplying each signal component within the electrical activity by a corresponding weight determined via the trained model. In another further implementation, identifying the electrical signals originating from a depth (820) within the cardiac tissue (12) includes computing a non-linear combination of signal components within the electrical activity.

[0019] In some implementations, the one or more processors (61) are further configured to provide the electrical activity from the plurality of electrodes (26) as an input to the neural network and determine, via the neural network, for at least one of the plurality of electrodes (26), an estimated distance to a nearest activation from a corresponding electrode (26). In some implementations, the one or more processors (61) are further configured to, for each electrode (26) of the plurality of electrodes (26), apply the mathematical model and the known electrode positions of the plurality of electrodes (26) to calculate an expected signal at each known electrode position.

[0020] In some implementations, the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes (26). In some implementations, identifying the electrical signals originating from a depth within the cardiac tissue includes running a machine learning model trained on the simulated ECG data. In some implementations, the one or more processors (61) are further configured to generate a visual representation (20) of the identified electrical signals at the specified depth within the cardiac tissue (12). [Brief description of the drawings]

[0021] A more detailed understanding may be had from the following description, given by way of example in conjunction with the accompanying drawings, in which like reference numbers indicate similar elements and in which: [Figure 1] FIG. 1 illustrates an example system in which one or more features of the subject matter of the present disclosure can be implemented, in accordance with one or more embodiments. [Diagram 2] 1 illustrates a diagram of a system in accordance with one or more embodiments. [Figure 3A] 1 illustrates a method according to one or more embodiments. [Figure 3B] 1 illustrates a graph according to one or more embodiments. [Figure 4] 1 illustrates a method according to one or more embodiments. [Diagram 5] 1 illustrates an exemplary catheter according to one or more embodiments. [Figure 6] 1 illustrates an exemplary catheter according to one or more embodiments. [Figure 7] 1 illustrates a method according to one or more embodiments. [Figure 8] 1 illustrates an example catheter of an example system configured to identify electrical activation signals at different distances from an electrode location, in accordance with one or more embodiments. [Figure 9] 1 illustrates an exemplary method for isolating electrical signals at a distance (e.g., depth) within cardiac tissue using ECG information collected from a multi-electrode mapping catheter, according to one or more embodiments. [Figure 10] An alternative exemplary method is shown for isolating electrical signals at a distance (e.g., depth) within cardiac tissue using ECG information collected from a multi-electrode mapping catheter, according to one or more embodiments. [Figure 11]FIG. 1 illustrates another alternative exemplary method for isolating electrical signals at a distance (e.g., depth) within cardiac tissue using ECG information collected from a multi-electrode mapping catheter, according to one or more embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] The systems and methods herein relate generally to signal processing. In accordance with one or more embodiments, the systems and methods herein provide for intracardiac unipolar far-field reduction or cancellation using a multiple electrode catheter. As an example, the systems and methods herein are described with respect to a mapping engine.

[0023] The mapping engine may be implemented as processor executable code or software that is necessarily based on the process operations by the medical device equipment and the processing hardware of the medical device equipment. For ease of explanation, the mapping engine is described herein with respect to mapping the heart, however, any anatomical structure, body part, organ, or portion thereof may be targeted for mapping by the mapping engine described herein. In accordance with one or more embodiments, the mapping engine obviates the traditional cardiac mapping assumption that all far-field signals are the same on all electrodes. In contrast to traditional cardiac mapping, the mapping engine intelligently implements a new assumption that the far-field signals are similar but not identical. As a result, the mapping engine weights the electrodes inversely proportional to their distance to identify common signal components between the electrodes. This common signal component is used by the mapping engine to precisely reduce or cancel far-field interference. Reducing or canceling far-field interference by the mapping engine produces a cleaner near-field signal for improved cardiac mapping. Additionally or alternatively, in some embodiments, the mapping engine may use information generated by weighting electrodes inversely proportional to distance to identify signals emitted at specific distances from the electrodes, including depths within the myocardial tissue.

[0024] One or more advantages, technical effects, and / or benefits of the mapping engine may include removing far-field interference and estimating pure local activity (e.g., especially for highly complex mapping cases) from unipolar signals instead of the need for bipolar signals, which are known for their disadvantages such as electrode distance and directional distortion. Thus, by utilizing and transforming specifically unipolar signals, the mapping engine doubles the number of electrodes analyzed to obtain or improve spatial behavior resolution. Additionally, in cases of late activation, the mapping engine may also remove persistent aFib near-far field signals to enhance detection of aFib targets.

[0025] Additional advantages, technical effects, and / or benefits include enabling an electroanatomical mapping system (interchangeably referred to herein as a "cardiac mapping" system) the ability to identify and / or visualize electrical activity in myocardial tissue beneath the inner surface of the heart (e.g., 0.5-10 mm deep within the cardiac tissue) in three dimensions. Furthermore, by identifying electrical activity by depth and location within the myocardial tissue, corresponding ECG information may be presented to a user of the electroanatomical mapping system in a variety of ways, including, but not limited to, constructed layer or "shell" surfaces of electrical maps (also referred to as "maps") at different depths of the electroanatomical model, and / or allowing a physician or system operator to view ECG waveforms at selectable depths within the tissue. The novel capabilities enabled by the methods and systems described herein that allow for identification, isolation, and / or visualization of electrical activity within the myocardial tissue may be described as the creation of ECG depth and radial lenses.

[0026] 1 is a diagram of an example system (e.g., medical device instrumentation and / or catheter-based electrophysiological mapping and ablation) shown as system 10 in which one or more features of the subject matter herein may be implemented according to one or more embodiments. All or a portion of system 10 may be used to collect information (e.g., biometric data and / or training data sets) and / or may be used to implement a mapping engine 101 as described herein. The system 10 includes a recorder 11, a heart 12, a catheter 14, a model or anatomical map 20, an electrogram 21, a spline 22, a patient 23, a physician 24 (or medical professional or clinician), a location pad 25, an electrode 26, a display device 27, a distal tip 28, a sensor 29, a coil 32, a patient interface unit (PIU) 30, an electrode skin patch 38, an ablation energy generator 50, and a workstation 55 (including at least one processor 61 and at least one memory 62 storing a mapping engine 101 therein), as shown. It should be noted that each element and / or item of the system 10 is representative of one or more of that element and / or item. The example system 10 shown in FIG. 1 can be modified to implement the embodiments disclosed herein. The embodiments of the present disclosure can be similarly applied using other system components and configurations. Additionally, system 10 may include additional components, such as elements for sensing electrical activity and / or physiological signals, wired or wireless connectors, processing devices, and display devices.

[0027] The system 10 includes multiple catheters 14 that are percutaneously inserted by a physician 24 through the vascular system of a patient 23 into a chamber or vasculature of the heart 12. Typically, a delivery sheath catheter (one example of a catheter 14) is inserted into the left or right atrium near a desired location in the heart 12. The multiple catheters 14 can then be inserted into the delivery sheath catheter to reach the desired location. The multiple catheters 14 may include catheters dedicated to sensing Intracardiac Electrogram (IEGM) signals, catheters dedicated to ablation, and / or catheters dedicated to both sensing and ablation. An exemplary catheter 14 configured to sense IEGMs is shown herein. The physician 24 brings a distal tip 28 of the catheter 14 into contact with the heart wall to sense a target site of the heart 12. For ablation, the physician 24 similarly brings a distal end of an ablation catheter to a target site for ablation.

[0028] Catheter 14 is an exemplary catheter that includes one, and preferably multiple, electrodes 26, optionally distributed across multiple splines 22 at distal tip 28, and configured to sense IEGM signals. Catheter 14 may additionally include a sensor 29 embedded in or near distal tip 28 for tracking the position and orientation of distal tip 28. Optionally and preferably, position sensor 29 is a magnetic-based position sensor including three magnetic coils for sensing three-dimensional (3D) position and orientation.

[0029] The sensor 29 (e.g., a position-based or magnetic-based position sensor) may operate in conjunction with a location pad 25 that includes a plurality of magnetic coils 32 configured to generate a magnetic field within a predetermined working volume. The real-time position of the distal tip 28 of the catheter 14 may be tracked based on the magnetic fields generated by the location pad 25 and sensed by the sensor 29. Details of magnetic-based position sensing technology are described in U.S. Patent Nos. 5,5391,199; 5,443,489; 5,558,091; 6,172,499; 6,239,724; 6,332,089; 6,484,118; 6,618,612; 6,690,963; 6,788,967; and 6,892,091.

[0030] The system 10 includes one or more electrode patches 38 placed on the patient 23 for skin contact to establish location references for the location pads 25 and impedance-based tracking of the electrodes 26. For impedance-based tracking, a current is directed to the electrodes 26 and sensed at the patches 38, so that the location of each electrode can be triangulated via the patches 38. Details of impedance-based location tracking technology are described in U.S. Patent Nos. 7,536,218, 7,756,576, 7,848,787, 7,869,865, and 8,456,182, which are incorporated herein by reference.

[0031] Recorder 11 displays electrograms 21 captured by electrodes 18 (e.g., surface electrocardiogram (ECG) electrodes) and IEGMs captured by electrodes 26 of catheter 14. Recorder 11 may include pacing capability for pacing the cardiac rhythm and / or may be electrically connected to a stand-alone pacer.

[0032] The system 10 may include an ablation energy generator 50 adapted to deliver ablation energy to one or more electrodes 26 at a distal tip 28 of a catheter 14 configured for ablation. The energy generated by the ablation energy generator 50 may include, but is not limited to, radiofrequency (RF) energy or pulsed-field ablation (PFA) energy, including monopolar or bipolar high-voltage DC pulses such as may be used to produce irreversible electroporation (IRE), or combinations thereof. The PIU 30 is an interface configured to establish electrical communication between the catheters, electrophysiology equipment, a power source, and a workstation 55 that controls the operation of the system 10. The electrophysiology equipment of the system 10 may include, for example, multiple catheters 14, location pads 25, body surface ECG electrodes 18, electrode patches 38, the ablation energy generator 50, and a recorder 11. Optionally, and preferably, PIU 30 additionally includes processing capabilities for implementing real-time calculations of catheter location and performing ECG calculations.

[0033] The workstation 55 includes a processor unit 61 having a memory 62, a memory 62 or storage device loaded with suitable operating software, and user interface functionality, as described further herein. The workstation 55 may optionally provide multiple functions, including: (1) modeling the endocardial anatomical structures in three dimensions (3D) and rendering a model or anatomical map 20 for display on the display device 27; (2) displaying activation sequences (or other data) compiled from recorded electrograms 21 with representative visual indicators or images superimposed on the rendered anatomical map 20 on the display device 27; (3) displaying real-time locations and orientations of multiple catheters within the heart chambers; and (5) displaying sites of interest, such as locations where ablation energy has been applied, on the display device 27. One commercially available product embodying elements of the system 10 is available as the CARTO® 3 system, available from Biosense Webster, Inc., 31A Technology Drive, Irvine, CA 92618.

[0034] The system 10 can be utilized to detect, diagnose, and / or treat cardiac conditions (e.g., using the mapping engine 101). Cardiac conditions, such as cardiac arrhythmias, persist as common and dangerous medical ailments, especially in the geriatric population. For example, the system 10 can be part of a surgical system (e.g., a CARTO® system sold by Biosense Webster) configured to obtain biometric data (e.g., anatomical and electrical measurements of a patient's organs as described herein, such as the heart 12) and perform cardiac ablation procedures. More specifically, the treatment of cardiac conditions, such as cardiac arrhythmias, often requires obtaining detailed mapping of cardiac tissue, chambers, veins, arteries, and / or electrical pathways. For example, as a prerequisite for successfully performing catheter ablation (as described herein), the source of the cardiac arrhythmia may be precisely localized in the chambers of the heart 12. Such localization may be performed by electrophysiological studies during which spatially resolved electrical potentials are detected by a mapping catheter (e.g., catheter 14) introduced into the chambers of the heart 12. This electrophysiological test, so-called electroanatomical mapping, provides 3D mapping data that can be displayed on the display device 27. Often, mapping and therapeutic functions (e.g., ablation) are provided by a single catheter or a group of catheters, with the mapping catheter also simultaneously acting as a therapeutic (e.g., ablation) catheter. According to one or more embodiments, the mapping engine 101 can be stored and executed directly by the catheter 14. According to one or more embodiments, the mapping engine 101 performs an electrocardiograph and / or an intracardiac electrogram (ECG / ICEG). The ECG / ICEG is a process of decomposing / analyzing / recording the electrical activity of the heart 12 using physiological signals of multiple electrodes 18 and electrodes 26 (e.g., in ICEG, at least one of the electrodes 26 is inside the heart 12).That is, one or more catheters 14 may be guided by a physician 24 into the heart 12 of a patient 23 as the system 100 assists in detecting, diagnosing, and / or treating a cardiac condition. The multiple electrodes 18 and 26 then detect and provide signals (aka physiological signals) that the mapping engine 101 uses to identify small electrical changes resulting from the myocardial electrophysiological pattern of depolarization between each heartbeat. The ECG / ICEG may be performed over a period of time, e.g., 10 seconds. This captures the entire magnitude and direction of the electrical depolarization of the heart 12 at each instant across the cardiac cycle. The ECG / ICEG may be recorded during a diagnostic or therapeutic procedure. The length of time of this procedure may vary from tens of minutes to hours. During each therapeutic procedure, there are typically dozens of ablation sessions, with each session lasting, e.g., from a few seconds to up to about a minute.

[0035] According to one or more embodiments, the mapping engine 101 can include and execute one or more algorithms, such as algorithms implementing mathematical multi-signal decomposition operations or mathematical dimensionality reduction techniques. The mapping engine 101 reduces the need for wavefront algorithms, as the mapping engine 101 provides accurate near-field components, activation times (e.g., using minimum derivatives), and voltage levels (e.g., max-min). Examples of the one or more algorithms include, but are not limited to, singular value decomposition (SVD), principal component analysis (PCA), and / or other matrix decompositions or factorizations.

[0036] According to one or more further embodiments, the mapping engine 101 may further utilize information generated by a mathematical dimensionality reduction algorithm, such as an SVD or PCA algorithm, to identify electrical signals originating from estimated depths within the cardiac tissue. When the mapping engine 101 generates orthogonal signals generated by SVD or PCA, each orthogonal signal component may correspond to electrical activity originating from a particular distance or depth relative to a particular electrode. A linear combination of the signal components is applied, and a weight assigned to each orthogonal signal determines the weight contribution to the final composite signal. Optionally, a non-linear combination of the signal components may also be performed to compensate for signal degradation or attenuation depending on the distance from the electrode. The weight applied to each signal may be adjusted in a myriad of ways, including by employing deep learning or other artificial intelligence (AI) or machine learning (ML) algorithms. By considering the spatial information contained in the orthogonal signals and manipulating their contributions through mathematical combinations (linear or non-linear), the system may distinguish and extract electrical signals originating from various distances around a particular electrode, including depths within the cardiac tissue.

[0037] Alternatively, in some embodiments, a deep learning or machine learning model may be employed independently, without the use of mathematical dimensional algorithms. In such embodiments, the ECG signal from the electrodes of the mapping catheter may be input into a trained model, which then processes the ECG signal directly to identify patterns or features associated with particular distances or depths from the electrodes, consistent with its training based on electrical mapping and corresponding information regarding electrical propagation within a range of tissue depths.

[0038] According to one or more alternative embodiments, the mapping engine 101 may implement a neural network configured to identify intracardiac ECG signals representative of electrical activity at different depths within the myocardial tissue. A trained machine learning model, specifically a neural network, may be used to estimate the distance to the nearest activation within the myocardial tissue based on the ECG readings. The model may be trained using simulated data encompassing various scenarios, such as different catheter locations, membrane potential locations, and activation states of the myocardial tissue (e.g., healthy, scarred, fibrotic, etc.). When employed as part of a cardiac mapping system 10, such as as part of the mapping engine 101, the model may receive input ECG signals from electrodes of a multi-electrode catheter and generate an output in the form of identifying electrical signals at specified distances or depths within the myocardial tissue.

[0039] In yet another alternative embodiment, the mapping engine 101 may implement models, field projection calculations, and optimization and inverse problem solving to identify ECG signals representing electrical activity at different depths within the myocardial tissue. In this embodiment, the electrical activity of the cardiac tissue is represented as a set of sources, such as dipoles, distributed on the surface of the heart chamber and at the depth of the tissue. Each such dipole has an activation time, and all run synchronously to generate a propagating activation wave. This has a field projection onto the location of the electrode that is compared with measurements from the electrode. By solving a mathematical minimization problem that finds the time and distance at which a grid of dipoles that resamples the measurements most is activated, the distance from the electrode to the nearest activation can be estimated. To estimate the electrical activity at a particular location within the myocardial tissue, a field projection is calculated based on the location of the electrode on the mapping catheter.

[0040] Thus, in some embodiments, the mapping engine 101 may comprise hardware, software, or a combination of hardware and software for implementing a neural network, a classifier, a machine learning engine, or a similar model. For example, the mapping engine 101 may include software executed by one or more processors of a device or system, such as a central processing unit, a graphics processing unit, a tensor processing unit, a vector processing unit, or other type of processor or co-processor. In many implementations, the mapping engine 101 may be executed by multiple computing devices (e.g., as a cluster of instruments, a cloud of virtual computing devices, a server farm, etc.). In many such implementations, the functionality of the mapping engine 101 may be distributed across multiple processors and / or computing devices. For example, in some implementations, a first estimation of electrical activity at a first location may be performed by a first processor and / or a first physical or virtual computing device, and a second estimation of electrical activity at the same first location (e.g., at a different time) or at a different second location (e.g., at the same time or a different time) may be performed by a second processor and / or a second physical or virtual computing device. This may allow for scalability and low latency production of results, even for highly complex estimations.

[0041] In a patient (e.g., patient 23) having normal sinus rhythm (NSR), the heart (e.g., heart 12), including the atria, ventricles, and excitable conduction tissue, is electrically excited to beat in a synchronous, patterned manner. Note that this electrical excitation can be detected as intracardiac electrocardiogram (IC ECG) data, or the like.

[0042] According to one or more embodiments, in patients (e.g., patient 23) with cardiac arrhythmias (e.g., aFib), the abnormal regions of cardiac tissue do not follow the synchronous beating cycle associated with normal conductive tissue, in contrast to patients with NSR. In contrast, the abnormal regions of cardiac tissue conduct abnormally to adjacent tissue, which disrupts the cardiac cycle and results in an asynchronous cardiac rhythm. It should be noted that this asynchronous cardiac rhythm can also be detected as IC ECG data. Such abnormal conduction has previously been known to occur in various regions of the heart 12, such as, for example, in the region of the sino-atrial (SA) node along the conduction pathways of the atrioventricular (AV) node, or in the myocardial tissue forming the walls of the ventricles and atria. Other conditions, such as atrial flutter, exist in which abnormal conductive tissue patterns lead to reentry pathways, causing the heart chambers to beat in a regular pattern that can be multiples of the sinus rhythm.

[0043] As an example, in assisting the system 10 in detecting, diagnosing, and / or treating a cardiac condition, a physician 24 may guide the catheter 14 into the heart 12 of a patient 23 lying in bed. For example, the physician 24 may insert the shaft through the sheath while manipulating the distal end of the shaft using a manipulator near the proximal end of the catheter 14 and / or a deflection from the sheath. According to one or more embodiments, the catheter 14 may be attached to the distal end of the shaft. The catheter 14 may be inserted through the sheath in a collapsed state and then expanded within the heart 12.

[0044] In general, electrical activity at a point within the heart 12 can typically be measured by advancing a catheter 14 containing an electrical sensor (e.g., sensor 29) at or near its distal tip (e.g., at least one electrode 26) to the point within the heart 12, contacting tissue with the electrical sensor, and acquiring data at the point. One drawback to mapping the ventricles using a type of catheter that includes only a single distal tip electrode is the long time required to collect point-by-point data over the necessary number of points required for a detailed map of the ventricle as a whole. Therefore, multi-electrode catheters (e.g., catheter 14) have been developed to simultaneously measure electrical activity at multiple points within the heart chambers.

[0045] The catheter 14, which may include at least one electrode 26 and a catheter needle coupled to its body, may be configured to obtain biometric data, such as electrical signals, of an internal organ (e.g., the heart 12) and / or ablate a tissue region thereof (e.g., a chamber of the heart 12). It is noted that the electrode 26 is representative of any similar element, such as a tracking coil, a piezoelectric transducer, an electrode, or a combination of elements configured to ablate a tissue region or obtain biometric data. According to one or more embodiments, the catheter 14 may include one or more position sensors that are used to determine trajectory information. This trajectory information may be used to infer motion characteristics, such as tissue contractility.

[0046] The biometric data (e.g., patient biometrics, patient data, or patient biometric data) may include one or more of local activation time (LAT), electrical activity, topology, bipolar mapping, baseline activity, ventricular activity, dominant frequency, impedance, and the like. LAT may be a time point of threshold activity corresponding to local activation calculated based on a normalized initial starting point. Electrical activity may be any applicable electrical signal that may be measured based on one or more thresholds and may be sensed and / or enhanced based on signal-to-noise ratio and / or other filters. Topology may correspond to the physical structure of the body part or a portion of the body part, and may correspond to changes in the physical structure for different parts of the body part or for different body parts. The dominant frequency may be a frequency or range of frequencies commonly found in a portion of the body part and may be different in different parts of the same body part. For example, the dominant frequency of the PVs of the heart may be different from the dominant frequency of the right atrium of the same heart. Impedance may be a resistance measurement in a particular region of the body part.

[0047] Examples of biometric data include, but are not limited to, patient identification data, IC ECG data, bipolar intracardiac reference signals, anatomical and electrical measurements, trajectory information, body surface (BS) ECG data, historical data, brain biometrics, blood pressure data, ultrasound signals, radio signals, audio signals, two-dimensional or three-dimensional (3D) image data, blood glucose data, temperature data, other electrical activity and / or physiological signals. The biometric data may generally be used to monitor, diagnose, and treat any number of different diseases, such as cardiovascular diseases (e.g., arrhythmias, cardiomyopathies, and coronary artery disease), and autoimmune diseases (e.g., diabetes mellitus type I and type II). It is noted that BS ECG data may include data and signals collected from electrodes on the patient's surface, IC ECG data may include data and signals collected from electrodes inside the patient's body, and ablation data may include data and signals collected from ablated tissue. Additionally, the BS ECG data, IC ECG data, and ablation data may be derived from one or more treatment records along with catheter electrode position data.

[0048] For example, the catheter 14 may use the electrodes 26 to implement intravascular ultrasound and / or MRI catheterization to image the heart 12 (e.g., acquire and process biometric data). The catheter 14 is shown in an enlarged view within a chamber of the heart 12. It will be understood that any shape that includes one or more electrodes 26 may be used to implement the embodiments disclosed herein.

[0049] Examples of catheter 14 include, but are not limited to, a straight catheter with multiple electrodes, a balloon catheter including electrodes distributed on multiple spines forming a balloon, a lasso, a catheter with multiple electrodes in a grid shape or a loop catheter with multiple electrodes, a high density catheter, or any other applicable shape or combination. The straight catheter may be fully or partially elastic so that it can twist, bend, and / or otherwise change its shape based on received signals and / or based on the action of an external force (e.g., cardiac tissue) on the straight catheter. The balloon catheter may be designed to hold its electrodes in intimate contact against the endocardial surface when deployed within the patient's body. As an example, the balloon catheter may be inserted into a lumen such as a pulmonary vein (PV). The balloon catheter may be inserted into the PV in a deflated state, such that the balloon catheter does not occupy its maximum volume while inserted into the PV. The balloon catheter may be expanded while inside the PV, such that the electrodes on the balloon catheter contact the entire circular portion of the PV. Such contact with the entire circular portion of the PV or any other lumen allows for efficient imaging and / or ablation. Other examples of catheters 14 include the PentaRay® catheter and the Constellation catheter. Examples of catheters 14 are also described with respect to Figures 5-6, as described in more detail below.

[0050] According to another example, a body patch and / or body surface electrodes (e.g., one or more electrode patches 38) may be placed on or adjacent to the body of the patient 23. A catheter 14 having one or more electrodes 26 may be placed within the body (e.g., within the heart 12), and the position of the catheter 14 may be determined by the 100 system based on signals transmitted and received between one or more electrodes 26 of the catheter 14 and the body patch and / or body surface electrodes. Additionally, the electrodes 26 may sense biometric data from within the body of the patient 23, such as within the heart 12 (e.g., the electrodes 26 sense tissue electrical potentials in real time). The biometric data may be correlated with the determined position of the catheter 14, thereby allowing a rendering of the patient's body part (e.g., the heart 12) to be displayed, showing the biometric data superimposed on the shape of the body part.

[0051] As a further example, the catheter 14 and other items of the system 10 can be connected to a workstation 55. The workstation 55 can include any computing device that employs ML / AI algorithms (which can be included in the mapping engine 101). According to an exemplary embodiment, the workstation 55 includes one or more processors 61 (any computing hardware) and memory 62 (any non-transitory tangible medium), where the one or more processors 61 execute computer instructions for the mapping engine 101 and the memory 62 stores these instructions for execution by the one or more processors 61. For example, the workstation 55 can be configured to receive and process biometric data to determine whether a particular tissue region conducts electricity. In some embodiments, the workstation 55 can be further programmed by the mapping engine 101 (in software) to perform functions of an ablation procedure guidance method. For example, an ablation procedure guidance method can include receiving input (e.g., including one or more images and a conduction velocity vector estimate), utilizing the images and the conduction velocity vector estimate to generate a digital twin of the anatomical structure, and presenting the digital twin to provide precise ablation guidance of the anatomical structure and to provide electrophysiology information of the anatomical structure.

[0052] In accordance with one or more embodiments, the mapping engine 101 may be external to the workstation 55, such as located in the catheter 14, in an external device, in a mobile device, in a cloud-based device, or may be a stand-alone processor. In this regard, the mapping engine 101 may be transferable / downloadable in electronic form over a network.

[0053] In one example, the workstation 55 may be any computing device described herein including software (e.g., mapping engine 101) and / or hardware (e.g., processor 61 and memory 62), such as a general-purpose computer with suitable front-end and interface circuitry for transmitting and receiving signals to and from the catheter 14 and controlling other components of the system 10. For example, the front-end and interface circuitry may include an input / output (I / O) communication interface that allows the workstation 55 to receive signals from and / or transmit signals to at least one electrode 26. The workstation 55 may include real-time noise reduction circuitry, typically configured as a field programmable gate array (FPGA). The real-time noise reduction circuitry is followed by an analog-to-digital (A / D) ECG or electrocardiogram / electromyogram (EMG) signal conversion integrated circuit. The workstation 55 may communicate signals from the A / D ECG or EMG circuitry to a separate processor and / or may be programmed to perform one or more of the functions disclosed herein.

[0054] A display device 27, which may be any electronic device for visually presenting biometric data, is connected to the workstation 55. According to an exemplary embodiment, during a procedure, the workstation 55 may facilitate the presentation of a rendering of the body part to the physician 24 on the display device 27 and store data representative of the rendering of the body part in memory 62. For example, a map indicative of motion characteristics may be rendered / constructed based on trajectory information sampled at a sufficient number of points within the heart 12. By way of example, the display device 27 may include a touch screen that may be configured to receive input from the physician 24 in addition to presenting the rendering of the body part.

[0055] In some embodiments, the physician 24 may use one or more input devices, such as a touchpad, mouse, keyboard, gesture recognizer, etc., to manipulate the elements of the system 10 and / or the rendering of the body part. For example, the input device may be used to change the position of the catheter 14 so that the rendering is updated. It should be noted that the display device 27 may be located at the same location or at a remote location, such as another hospital or another healthcare provider network. According to one or more embodiments, the system 10 may also obtain biometric data using ultrasound, computed tomography (CT), MRI, or other medical imaging techniques that utilize the catheter 14 or other medical equipment. For example, the system 10 may obtain ECG data and / or anatomical and electrical measurements (e.g., biometric data) of the heart 12 using one or more catheters 14 or other sensors. More specifically, the workstation 55 may be connected by a cable to BS electrodes, including adhesive skin patches, that are applied to the patient 23. The BS electrodes acquire / generate biometric data in the form of BS ECG data. For example, the processor 61 determines the position coordinates of the catheter 14 within the body part (e.g., the heart 12) of the patient 23. The position coordinates are based on impedance or electromagnetic fields measured between the body surface electrodes and the electrodes 26 of the catheter 14 or other electromagnetic components. Additionally or alternatively, the location pads generating the magnetic fields used for steering may be located on the surface of the bed (or table) and may be separate from the bed. The biometric data may be transmitted to the workstation 55 and stored in the memory 62. Alternatively or additionally, the biometric data may be transmitted to a server, which may be local or remote, using a network as further described herein.

[0056] According to one or more embodiments, the catheter 14 may be configured to ablate a tissue region of a chamber of the heart 12. For example, the catheter 14 is shown in an expanded view within a chamber of the heart 12. Additionally, an ablation electrode, such as at least one electrode 26, is configured to provide energy to a tissue region of the internal organ (e.g., the heart 12). The energy may be thermal energy and may cause damage to the tissue region starting from a surface of the tissue region and extending through a thickness of the tissue region. Biometric data related to the ablation procedure (e.g., ablated tissue, ablation location, etc.) may be considered ablation data.

[0057] According to one example, with respect to acquiring biometric data, a multi-electrode catheter (e.g., catheter 14) is advanced into a chamber of heart 12. Anteroposterior (AP) and lateral fluoroscopic photographs can be acquired to establish the position and orientation of each of the electrodes. An ECG can be recorded from each of the electrodes 26 in contact with the cardiac surface relative to a time reference, such as the occurrence of P waves in sinus rhythm from a BS ECG and / or a signal from an electrode 26 of catheter 14 positioned within the coronary sinus. The system further disclosed herein can distinguish between electrodes that record electrical activity and electrodes that do not record electrical activity due to not being in close proximity to the endocardial wall. After an initial ECG is recorded, the catheter can typically be repositioned and fluoroscopic photographs and ECGs can be recorded again. An electrical map (e.g., via cardiac mapping) can then be constructed from an iteration of the above process.

[0058] Cardiac mapping can be implemented using one or more techniques. In general, mapping of cardiac regions, such as cardiac regions, tissues, veins, arteries, and / or electrical pathways of the heart 12, can lead to identification of problem areas, such as scar tissue, arrhythmia sources (e.g., electrical rotors), healthy regions, and the like. The cardiac regions are mapped such that a visual rendering of the mapped cardiac regions is provided using a display, as further disclosed herein. Additionally, cardiac mapping (which is an example of cardiac imaging) can include mapping based on one or more modalities, such as, but not limited to, LAT, regional activation rate, electrical activity, topology, bipolar mapping, dominant frequency, or impedance. Data (e.g., biometric data) corresponding to multiple modalities may be acquired using a catheter (e.g., catheter 14) inserted into the patient's body and may be provided for rendering at the same time or at different times based on corresponding settings and / or physician 24 preferences.

[0059] As an example of the first technique, cardiac mapping may be implemented by sensing electrical properties of cardiac tissue, such as LAT, as a function of precise location within the heart 12. Corresponding data (e.g., biometric data) may be acquired by one or more catheters (e.g., catheter 14) advanced into the heart 12 and having electrical and location sensors (e.g., electrodes 26) at their distal tips. As a specific example, location and electrical activity may be initially measured at about 10 to about 20 points on the inner surface of the heart 12. These data points may generally be sufficient to generate a preliminary reconstruction or map of the cardiac surface with satisfactory quality. The preliminary map may often be combined with data measured at additional points to generate a more comprehensive map of the cardiac electrical activity. In clinical practice, it is not uncommon to collect data from 100 or more sites (e.g., several thousand) to generate a detailed comprehensive map of the electrical activity of the cardiac chambers. The detailed maps generated can then serve as the basis for making decisions about a course of therapeutic action, such as tissue ablation as described herein, to alter the propagation of the heart's electrical activity and restore normal cardiac rhythm.

[0060] Furthermore, cardiac mapping is generated based on the detection of intracardiac potential fields (e.g., which are an example of IC ECG data and / or bipolar intracardiac reference signals). Non-contact methods can be implemented to simultaneously acquire large amounts of cardiac electrical information. For example, a catheter type having a distal end portion can be provided with a series of sensor electrodes distributed over its surface and connected to insulated conductors for connection to signal sensing and processing means. The size and shape of the end portion can be such that the electrodes are spaced apart from the walls of the cardiac chambers by a large distance. The intracardiac potential fields can be detected during one cardiac beat. According to one example, the sensor electrodes can be distributed on a series of circumferences located in planes spaced apart from one another. These planes can be perpendicular to the long axis of the end of the catheter. At least two additional electrodes can be disposed adjacent to both ends of the long axis of the end. As a more specific example, the catheter can include four circumferences with eight electrodes equiangularly spaced apart on each circumference. Thus, in this particular implementation, the catheter can include at least 34 electrodes (32 circumferential electrodes and 2 end electrodes). As another more specific example, the catheter can include 5 soft flexible branches, 8 radial splines, or other multi-spline catheters such as a turner-type with parallel splines (e.g., any of which can have a total of 42 electrodes).

[0061] As an example of electrical or cardiac mapping, electrophysiological cardiac mapping systems and techniques based on non-contact and non-expandable multi-electrode catheters (e.g., catheter 14) can be implemented. An ECG may be acquired using one or more catheters 14 with multiple electrodes (e.g., 42-122 electrodes, etc.). This implementation allows knowledge of the relative geometry of the probe and endocardium to be obtained by an independent imaging modality, such as transesophageal echocardiography. After independent imaging, non-contact electrodes can be used to measure cardiac surface potentials and a map can be constructed from the surface potentials (e.g., possibly using a bipolar intracardiac reference signal). This technique can include the following steps (after the independent imaging step): (a) measuring potentials using multiple electrodes placed on a probe placed within the heart 12; (b) determining the geometric relationship of the probe surface to the endocardial surface and / or other fiducials; (c) generating a matrix of coefficients representing the geometric relationship of the probe surface to the endocardial surface; and (d) determining the endocardial potentials based on the electrode potentials and the matrix of coefficients.

[0062] As another example of electrical or cardiac mapping, techniques and devices can be implemented for mapping the electrical potential distribution of a heart chamber. An intracardiac multi-electrode mapping catheter assembly is inserted into heart 12. The mapping catheter (e.g., catheter 14) assembly preferably includes a multi-electrode array or companion reference catheter having one or more integral reference electrodes (e.g., one or more electrodes 26).

[0063] According to one or more embodiments, the electrodes may be deployed in a generally spherical array that may be spatially referenced to points on the endocardial surface by a reference electrode or by a reference catheter that is brought into contact with the endocardial surface. A preferred electrode array catheter may have a large number of individual electrode sites (e.g., at least 24). Additionally, this exemplary technique may be implemented with knowledge of the location of each of the electrode sites on the array and with knowledge of the cardiac geometry. These locations are preferably determined by impedance plethysmography.

[0064] From an electrical or cardiac mapping perspective, and according to another example, the catheter 14 is a cardiac mapping catheter assembly that includes an electrode array that defines multiple electrode sites. The cardiac mapping catheter assembly also includes a lumen for carrying a reference catheter having a distal tip electrode assembly that is used to probe the heart wall. The cardiac mapping catheter assembly includes a braid of insulated wires (e.g., having 24-64 wires in the braid), each of which may be used to form an electrode site. The cardiac mapping catheter assembly is readily positionable within the heart 12 to be used to obtain electrical activity information from a first set of non-contact electrode sites and / or a second set of contact electrode sites.

[0065] According to yet another example, a catheter 14 capable of implementing mapping of electrophysiological activity within the heart includes a distal tip adapted to deliver stimulation pulses for pacing the heart or an ablation electrode for ablating tissue in contact with the tip, and may further include at least a pair of orthogonal electrodes for generating a differential signal indicative of local cardiac electrical activity in the vicinity of the orthogonal electrodes.

[0066] As described herein, system 10 is utilized to detect, diagnose, and / or treat cardiac conditions. In an exemplary operation, a process for measuring electrophysiological data within a heart chamber is implemented by system 10. This process includes, in part, placing a set of active and passive electrodes within heart 12, supplying current to the active electrodes thereby generating an electric field within the heart chamber, and measuring the electric field at the passive electrode sites. The passive electrodes are included in an array disposed on an inflatable balloon of a balloon catheter. In a preferred embodiment, the array is said to have 60-64 electrodes.

[0067] As another exemplary operation, cardiac mapping is implemented by the system 10 using one or more ultrasound transducers. The ultrasound transducers may be inserted into the patient's heart 12 and may acquire multiple ultrasound slices (e.g., two-dimensional or 3D slices) at various locations and orientations within the heart 12. The location and orientation of a given ultrasound transducer is known, and the acquired ultrasound slices are stored so that they can be displayed at a later time. One or more ultrasound slices corresponding to the position of a catheter 14 (e.g., a treatment catheter) are later displayed, with the catheter 14 being overlaid onto the one or more ultrasound slices.

[0068] Considering the system 10, it can be seen that cardiac arrhythmias, including atrial arrhythmias, can be of the multiwavelet-reentrant type, characterized by multiple asynchronous loops of electrical impulses that scatter and often self-propagate around the atria (e.g., another example of IC ECG data). Alternatively or in addition to the multiwavelet-reentrant type, cardiac arrhythmias can also have focal excitation sources, such as when identified regions of tissue in the atria are excited autonomously in a rapidly repetitive manner (e.g., another example of IC ECG data). Ventricular tachycardia (V-tach or VT) is a tachycardia or fast heart rhythm that originates from one of the ventricles. It is a potentially fatal arrhythmia, as it can lead to ventricular fibrillation and sudden death.

[0069] For example, aFib occurs when the normal electrical impulses generated by the sinoatrial node (e.g., another example of IC ECG data) are overwhelmed by chaotic electrical impulses (e.g., signal interference) originating in the atrial veins and PVs, conducting irregular impulses to the ventricles. The result is an irregular heartbeat that may persist for minutes to weeks or even years. In many cases, aFib is a chronic condition that often carries a small increase in the risk of death from stroke. The treatment approach for aFib is medication to slow the heart rate or restore normal heart rhythm. Additionally, patients with aFib are often given anticoagulants to protect against the risk of stroke. The use of such anticoagulants carries its own risks, internal bleeding. In some patients, medication is not sufficient and the patient's aFib is deemed drug refractory, i.e., untreatable by standard pharmacological interventions. Synchronized cardioversion can also be used to convert aFib to a normal heart rhythm. Alternatively, patients with aFib are treated with catheter ablation.

[0070] Catheter ablation-based therapy may include mapping electrical properties of cardiac tissue, particularly the endocardium and cardiac volumes, and selectively ablating cardiac tissue by application of energy. Electrical or cardiac mapping (e.g., implemented by any of the electrophysiological cardiac mapping systems and techniques described herein) includes creating an electrical potential map (e.g., voltage map) of wave propagation along cardiac tissue, or a map of arrival times (e.g., LAT map) to points located within various tissues. Electrical or cardiac mapping (e.g., cardiac map) can be used to detect localized cardiac tissue dysfunction. Ablation, such as cardiac mapping-based ablation, can stop or change the propagation of unwanted electrical signals from one portion of the heart 12 to another.

[0071] The ablation process damages undesired electrical pathways by creating non-conductive lesions. Various energy delivery modalities have been previously disclosed for creating lesions, including the use of microwave, laser, and more generally radio frequency energy to create a conduction block along the cardiac tissue wall. Another example of an energy delivery method is irreversible electroporation (IRE), which applies a high electric field that damages cell membranes. In a two-stage procedure (mapping followed by ablation), electrical activity at points within the heart 12 is typically sensed and measured by advancing a catheter 14 containing one or more electrical sensors (e.g., electrodes 26) into the heart 12 and acquiring / capturing data (e.g., generally as biometric data or specifically as ECG data) at multiple points. The ECG data is then used to select a target region of the endocardium where ablation will be performed.

[0072] Cardiac ablation and other cardiac electrophysiology procedures are becoming increasingly complex as physicians treat challenging conditions such as atrial fibrillation and ventricular tachycardia. Treatment of refractory arrhythmias can now rely on the use of 3D mapping systems to reconstruct the anatomy of the targeted heart chamber. In this regard, the mapping engine 101 employed by the system 10 herein manipulates and evaluates biodata in general, or ECG data in particular, to generate improved tissue data that allows for more accurate diagnosis, images, scans, and / or maps to treat abnormal heart rhythms or arrhythmias. For example, cardiologists rely on software such as the Complex Fractionated Atrial Electrograms (CFAE) module of the CARTO® 3 3D mapping system manufactured by Biosense Webster, Inc. (Diamond Bar, Calif.) to generate and analyze ECG data. The mapping engine 101 of the system 10 enhances this software to generate and analyze improved biomedical data, which further provides multiple pieces of information regarding the electrophysiological properties of the heart 12 (including scar tissue) that represent the cardiac substrate (anatomical and functional) of the aFib.

[0073] Thus, the system 10 implements a 3D mapping system, such as the CARTO® 3 3D mapping system, to locate potential arrhythmogenic substrates of cardiomyopathies for detection of abnormal ECG. Substrates related to these cardiac conditions have been associated with the presence of delayed ECG and fragmentation of the endocardial and / or epicardial layers of the ventricular chambers (right and left). For example, areas of low or medium voltage may indicate ECG fragmentation and delayed activity. Furthermore, areas of low or medium voltage during sinus rhythm may correspond to critical isthmus identified in sustained coherent ventricular arrhythmias (e.g., in non-permissive ventricular tachycardia, as well as in the atrium). In general, abnormal tissue is characterized by low voltage ECG. However, early clinical experience with endocardial-epicardial mapping has shown that areas of low voltage are not always present as the only arrhythmogenic mechanism in such patients. Indeed, areas of low or medium voltage may show ECG fragmentation and delayed activity during sinus rhythm, which corresponds to the critical isthmus identified during sustained coherent ventricular arrhythmias (e.g., only applicable to non-permissive ventricular tachycardia). Moreover, in many cases, ECG fragmentation and delayed activity are observed in areas showing normal or near-normal voltage amplitudes (>1-1.5 mV). The latter areas can be evaluated according to voltage amplitudes, but are not considered normal according to the intracardiac signal and therefore represent true arrhythmogenic substrates. 3D mapping allows to localize arrhythmogenic substrates on the endocardial and / or epicardial layers of the right / left ventricle, whose distribution may vary depending on the evolution of the primary disease.

[0074] As another exemplary operation, cardiac mapping may also be implemented by the system 10 using one or more multi-electrode catheters (e.g., catheter 14). The multi-electrode catheters are used to stimulate and map electrical activity within the heart 12 and to ablate sites of abnormal electrical activity. In use, the multi-electrode catheter is inserted into a major vein or artery, such as the femoral vein, and then guided into a chamber of the heart 12 of interest. A typical ablation procedure involves inserting the catheter 14, having at least one electrode 26 at its distal end, into the heart chamber. A reference electrode is provided by taping to the patient's skin or by a second catheter placed in or near the heart, or by selecting from one or the other electrodes 26 of the catheter 14. Radio frequency (RF) current is applied to the tip electrode 26 of the ablation catheter 14, and current flows through the medium (i.e., blood and tissue) surrounding the tip electrode toward the reference electrode. The current distribution is determined by the amount of electrode surface in contact with the tissue compared to blood, which is more conductive than tissue. Heating of the tissue occurs due to its electrical resistance. When the tissue is heated sufficiently, cell destruction is induced in the cardiac tissue, resulting in the formation of electrically non-conductive lesions in the cardiac tissue. This process also results in heating of the tip electrode 26 by conduction from the heated tissue to the electrode itself. When the temperature of the electrode becomes high enough, possibly exceeding 60° C., a thin, transparent film of dehydrated blood proteins may form on the surface of the electrode 26. As the temperature continues to increase, this dehydrated layer may gradually thicken, causing blood to coagulate on the electrode surface. Because dehydrated biological material has a higher electrical resistance than endocardial tissue, the impedance to the flow of electrical energy into the tissue also increases. When the impedance becomes high enough, an impedance rise occurs and the catheter 14 must be withdrawn from the body and the tip electrode 26 must be cleaned.

[0075] Referring now to FIG. 2, a diagram of a system 200 capable of implementing one or more features of the subject matter of the present disclosure, according to one or more exemplary embodiments, is illustrated. The system 200 includes an apparatus 204, a local computing device 206, a remote computing system 208, a first network 210, and a second network 211 in association with a patient 202 (e.g., an example of the patient 23 of FIG. 1). Additionally, the apparatus 204 includes a biometric sensor 221 (e.g., an example of the catheter 14 of FIG. 1), a processor 222, a user input (UI) sensor 223, a memory 224, and a transceiver 225. It should be noted that the mapping engine 101 of FIG. 1 is reused in FIG. 2 for ease of explanation and brevity. According to one embodiment, the apparatus 204 is an example of the system 100 of FIG. 1, and the apparatus 204 may include components internal to the patient 202 and components external to the patient 202. According to another embodiment, the device 204 is a device external to the patient 202, including an attachable patch (e.g., attached to the patient's skin). According to another embodiment, the device 204 can be internal to the body of the patient 202 (e.g., subcutaneously implantable), and the device 204 can be inserted into the body of the patient 202 by any applicable method, including oral infusion, surgical insertion via a vein or artery, endoscopic surgery, or laparoscopic surgery. According to one embodiment, although a single device 204 is shown in FIG. 2, an exemplary system can include multiple devices.

[0076] Thus, the apparatus 204, the local computing device 206, and / or the remote computing system 208 may be programmed to execute computer instructions for the mapping engine 101. As an example, the memory 223 stores these instructions for execution by the processor 222 to enable the apparatus 204 to receive and process biometric data via the biometric sensor 201. In this manner, the processor 222 and the memory 223 are representative of the processor and memory of the local computing device 206 and / or the remote computing system 208.

[0077] The apparatus 204, the local computing device 206, and / or the remote computing system 208 can be any combination of software and / or hardware that individually or collectively stores, executes, and implements the mapping engine 101 and its functions. Furthermore, as described herein, the apparatus 204, the local computing device 206, and / or the remote computing system 208 can be an electronic computer framework that includes and / or employs any number and combination of computing devices and networks utilizing various communication technologies. The apparatus 204, the local computing device 206, and / or the remote computing system 208 are easily scalable, extensible, and modular, with the ability to change to different services or reconfigure some features independently of other features.

[0078] Networks 210 and 211 may be wired networks, wireless networks, or may include one or more wired and wireless networks. According to one embodiment, network 210 is an example of a short-range network (e.g., a local area network (LAN) or a personal area network (PAN)). Information can be transmitted between apparatus 204 and local computing device 206 over short-range network 210 using any one of a variety of short-range wireless communication protocols, such as Bluetooth, Wi-Fi, Zigbee, Z-Wave, near field communication (NFC), Ultraband, Zigbee, or infrared (IR). Further, network 211 is an example of one or more of an intranet, a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a direct connection or series of connections, a cellular telephone network, or any other network or medium capable of facilitating communication between local computing device 206 and remote computing system 208. Information may be transmitted over network 211 using any one of a variety of long-range wireless communication protocols (e.g., TCP / IP, HTTP, 3G, 4G / LTE, or 5G / New Radio). It is noted that the wired connections of networks 210 and 211 may be implemented using Ethernet, Universal Serial Bus (USB), RJ-11, or any other wired connection, and the wireless connections may be implemented using Wi-Fi, WiMAX, and Bluetooth, infrared, cellular networks, satellite communications, or any other wireless connection methodology.

[0079] In operation, the device 204 may continuously or periodically acquire, monitor, store, process, and communicate biometric data associated with the patient 202 via the network 210. Additionally, the device 204, the local computing device 206, and / or the remote computing system 208 communicate via the networks 210 and 211 (e.g., the local computing device 206 may be configured as a gateway between the device 204 and the remote computing system 208). For example, the device 204 may be an example of the system 100 of FIG. 1 configured to communicate with the local computing device 206 via the network 210. The local computing device 206 may be, for example, a fixed / standalone device, a base station, a desktop / laptop computer, a smartphone, a smartwatch, a tablet, or any other device configured to communicate with other devices via the networks 211 and 210. A remote computing system 208, implemented as a physical server on or connected to the network 211 or as a virtual server within a public cloud computing provider of the network 211 (e.g., Amazon Web Services (AWS)®), can be configured to communicate with the local computing device 206 over the network 211. Thus, biometric data related to the patient 202 can be communicated throughout the system 200.

[0080] The elements of the device 204 are now described. The biometric sensor 221 may include, for example, one or more transducers configured to convert one or more environmental conditions into electrical signals such that different types of biometric data may be observed / obtained / obtained. For example, the biometric sensor 221 may include one or more of an electrode (e.g., electrodes 18 and 26 of FIG. 1), a temperature sensor (e.g., a thermocouple), a blood pressure sensor, a blood glucose sensor, a blood oxygen sensor, a pH sensor, an accelerometer, and a microphone.

[0081] In executing the mapping engine 101, the processor 222 can be configured to receive, process, and manage biometric data acquired by the biometric sensor 221, and communicate the biometric data to the memory 224 for storage and / or across the network 210 via the transceiver 225. Biometric data from one or more other devices 204 may also be received by the processor 222 via the transceiver 225. Also, as described in more detail below, the processor 222 can be configured to selectively respond to different tapping patterns (e.g., single tap or double tap) received from the UI sensor 223 such that different tasks of the patch (e.g., acquiring, storing, or transmitting data) are initiated based on the detected pattern. In some embodiments, the processor 222 can generate audible feedback regarding detection of the gesture.

[0082] The UI sensor 223 includes a piezoelectric or capacitive sensor configured to receive user input, such as, for example, a tap or touch. For example, the UI sensor 223 may be controlled to implement capacitive coupling in response to the patient 202 tapping or touching the surface of the device 204. Gesture recognition may be implemented by any one of a variety of capacitive types, such as resistive capacitive, surface capacitive, projected capacitive, surface ultrasonic, piezoelectric, and infrared touch. The capacitive sensor may be positioned over a small area or length of the surface such that a tap or touch on the surface activates the monitoring device.

[0083] The memory 224 is any non-transitory tangible medium, such as magnetic, optical, or electronic memory (e.g., any suitable volatile and / or non-volatile memory, such as random access memory or a hard disk drive). The memory 224 stores computer instructions executed by the processor 222. The transceiver 225 may include a separate transmitter and a separate receiver. Alternatively, the transceiver 225 may include a transmitter and receiver integrated into a single device. In operation, the device 204 utilizing the mapping engine 101 observes / acquires biometric data of the patient 202 via the biometric sensor 221, stores the biometric data in memory, and shares this biometric data throughout the system 200 via the transceiver 225. The mapping engine 101 may then utilize models, neural networks, machine learning, and / or artificial intelligence to provide intracardiac monopolar far-field reduction or cancellation using multiple electrode catheters.

[0084] Referring now to FIG. 3A, a method 300 (e.g., implemented by the mapping engine 101 of FIG. 1 and / or FIG. 2) is shown according to one or more exemplary embodiments. The method 300 is an example of providing intracardiac unipolar far-field reduction or cancellation using a multiple electrode catheter (e.g., catheter 14). The method 300, in one example, performs atrial local activation detection and ventricular detection within the QRS (regardless of late local activation and annotation). It is noted that the use of multiple electrode catheters provides a spatial advantage, and the mapping engine 101 analyzes each electrode 18 and 26 and each corresponding intracardiac voltage signal based on surrounding spatial electrode information. In this regard, the mapping engine 101 reduces or removes far-field information from all electrodes while significantly emphasizing pure local unipolar signals. Because the mapping engine 101 is superior to known bipolar sensitivity to spatial electrode orientation and distance disadvantages, the mapping engine 101 also provides scar detection based on local activation amplitude in addition to accurate activation annotation.

[0085] The method 300 begins at block 310, where the mapping engine 101 performs spatial electrode signal analysis for each of the multiple electrodes. The spatial electrode signal analysis compensates for far-field characteristics, which may be an estimate of far-field information, which is a spatial location dependent on providing electrode weighting information and determining common signal components. At least one specific electrode is selected for the spatial electrode signal analysis. For each specific electrode, all other electrode information is considered and correlated with respect to that specific electrode. According to one or more embodiments, the spatial electrode signal analysis is repeated with each electrode being a specific electrode. According to one or more embodiments, for the sampling time window of each multi-electrode catheter stable site, all electrodes are analyzed one by one. Furthermore, for each specific electrode, information of all other electrodes is considered as further described herein.

[0086] According to one or more embodiments of spatial electrode signal analysis, electrodes surrounding a particular electrode are weighted inversely proportional to their distance from the particular electrode. For example, the catheter 14 includes three electrodes in a row separated by a distance, such as 2 millimeters. The mapping engine 101 may analyze the first electrode and assign a weighting function. Exemplary weighting functions include 1-0.25×D, where D is the distance from the first electrode to the other electrode. The mapping engine 101 then generates a weighting vector, where the weight decreases as the distance from the reference electrode increases for a particular electrode (e.g., the further away, the less weighted). According to one or more embodiments, electrodes beyond a certain distance may be excluded (note that the distance is catheter-dependent, and in some cases is 5 millimeters). The mapping engine 101 then analyzes the electrode weighting information using one or more algorithms (e.g., a mathematical multi-signal decomposition operation). Examples of the one or more algorithms include, but are not limited to, singular value decomposition (SVD), principal component analysis (PCA), and / or other matrix decompositions or factorizations. SVD can factorize a real or complex matrix (e.g., electrode weight matrix W) to generalize the eigenvalue decomposition of a square normal matrix with an orthonormal eigenbasis. The mapping engine 101 estimates the common signal components hidden in all signals by analyzing the electrode weight information. The common signal components are estimated by the mapping engine 101 by taking the maximum decomposed component / signal. Other examples of common signal components include the maximum common signal component and the estimation of far-field information.

[0087] At block 320, the mapping engine 101 scales the common signal components (identified by the mapping engine 101 and / or spatial electrode signal analysis). According to one or more embodiments, the mapping engine 101 automatically scales the common signal components by applying an inverse decomposition formula using only the common signal components. The far-field signal is then estimated by taking the scaled common signal.

[0088] According to one or more embodiments, the mapping engine 101 scales the common signal components to best fit the analyzed electrode signals. For example, as shown in graph 325 of FIG. 3B, the scaling is shown by a scalar multiplication. Graph 325 includes sample index on the x-axis and milli-voltages on the y-axis. The graph shows original signal 326, common signal 327, scaled common signal 328 (i.e., far-field), and residual local signal 329 (i.e., near-field). Furthermore, a residual signal is determined; i.e., once scaled and matched, the common signal components are subtracted, leaving only the residual signal. The residual signal estimates pure local information. Local information can include, but is not limited to, local field signals within the electrical activity detected by the electrodes.

[0089] In block 340, the mapping engine 101 analyzes the pure local information to identify / generate / provide local unipolar activation. The local unipolar activation may include complex cases such as dual activation patterns. According to one or more embodiments, for LAT mapping, the mapping engine 101 utilizes minimum pure local information derivative to detect momentary activation within a cardiac cycle. According to one or more embodiments, for voltage (V) mapping, the mapping engine 101 takes the local activation peak-to-peak within a cardiac cycle. For both maps, the mapping engine 101 averages the final local results of several cardiac cycles and adds spatial smoothing by combining all the different spatial local results. In block 350, the mapping engine 101 displays the results. According to one or more embodiments, the mapping engine provides the local unipolar activation (e.g., for each electrode, all local activation patterns) on a fixed zero background of the user interface of the display 165. The visual representation by the display 165 of the local activation patterns may include a long duration for consistency analysis.

[0090] 4, a method 400 (e.g., implemented by the mapping engine 101 of FIG. 1 and / or FIG. 2) is shown in accordance with one or more exemplary embodiments. Method 400 is an example of estimating far-field information, as described with respect to block 310 of FIG. 3A. Method 400 begins at block 410, where the mapping engine 101 receives an electrode signal s from a plurality of electrodes 18 and 26. According to one or more embodiments, the mapping engine 101 receives an electrode signal s from electrode 26 of catheter 14.

[0091] At block 420, the mapping engine 101 determines a reference electrode q of the plurality of electrodes 18 and 26. The reference electrode q may be selected from any of the catheter's electrodes 26. Additionally, more than one reference electrode q may be determined such that method 400 is performed for each reference electrode q.

[0092] At block 430, the mapping engine 101 determines weights according to an electrode weight matrix W based on the reference electrode q. The electrode weight matrix W can be based on the structure of the catheter 14. Note that the electrode weight matrix W stores and manages weights of electrodes surrounding a particular electrode inversely proportional to the distance from the particular electrode. Examples of high density mapping catheters can include, but are not limited to, the OCTARAY™ mapping catheter, the PENTARAY® NAV ECO high density mapping catheter, and / or the OPTRELL™ high density mapping catheter, all from Biosense Webster, Inc. FIGS. 5-6 show an exemplary catheter 14.

[0093] FIG. 5 illustrates a catheter 500 according to one or more embodiments. The catheter 500 may be a catheter having multiple electrodes 534, such as at least 20 electrodes. As shown, the multiple electrodes may include as many as 48 electrodes located across or distributed across multiple spines 537. Using such a large number of electrodes 534 distributed across a large area by spines 537 allows for capturing a large amount of electrical activity across a large area at once. According to one or more embodiments, the multiple spines 537 may be moved through a sheath in a contracted state and expanded once inside the patient 23 of FIG. 1. Electrical activity at any focal point within the heart 12 may be measured, typically by advancing the catheter 500, contacting the catheter 500 with cardiac tissue, and acquiring data related to electrical activity at that point. One or more of the electrodes 534 may be selected as a reference electrode q.

[0094] For example, electrode 560 is designated as the reference electrode q. Furthermore, as the distance from electrode 560 increases, the weight decreases. In this manner, because electrode 562 is closer to electrode 560 than electrode 564, electrode 562 has a heavier weight than electrode 564. Furthermore, because electrode 566 is farther from electrode 560 than electrode 562, electrode 566 has a lighter weight than electrode 562. Note that the distance between each of the electrodes 534 can be measured with respect to three-dimensional space (i.e., using x, y, z coordinates). As a result, electrodes 564 and 566 may or may not have the same weight.

[0095] FIG. 6 illustrates a catheter 500 according to one or more embodiments. The catheter 600 is a catheter having multiple electrodes 605 that provide one or more physiological signals. For example, the number of electrodes 605 can be at least three arranged in any combination of groups. As shown in FIG. 6, a group of three electrodes 612, 614, and 612 are on the same spline 620 as a group of two electrodes 622 and 624. According to one or more embodiments, the catheter 600 includes at least five arms (shown in FIG. 6), and in some cases eight or more arms, and the electrodes can include 20 or 48 monopolar electrodes positioned in pairs or sets one or two millimeters apart from each other (e.g., the higher the density, the better the performance).

[0096] Returning to the method 400, in block 440, the mapping engine 101 determines a dot product of weights according to the electrode weight matrix W and the electrode signal s.

[0097] At block 450, the mapping engine 101 processes the dot products using one or more algorithms to generate orthogonal signals. In general, the one or more algorithms of the mapping engine 100 include mathematical multi-signal decomposition operations. Examples of the one or more algorithms include, but are not limited to, singular value decomposition (SVD), principal component analysis (PCA), and / or other matrix decompositions or factorizations. SCV can factorize real or complex matrices (e.g., electrode weight matrix W) to generalize the eigenvalue decomposition of square normal matrices with orthonormal eigenbasis.

[0098] At block 460, the mapping engine 101 selects the largest signal of the orthogonal signals. At block 455, the mapping engine 101 performs a correlation of the signal of the reference electrode q against the largest signal to provide a projection value. At block 470, the mapping engine 101 takes a tensor product of the largest signal and the projection value to provide an estimate of the far field (e.g., a common signal component of the electrical activity estimated by spatial electrode signal analysis).

[0099] 7, a method 700 (e.g., implemented by mapping engine 101 of FIG. 1 and / or FIG. 2) in accordance with one or more exemplary embodiments is shown. Method 700 is an example of cardiac cycle specific electrode signal analysis, as described with respect to block 310 of FIG. 3A.

[0100] At block 708, the mapping engine 101 receives a first input. The first input includes a global P×P weight matrix. At block 709, a second input is received. The second input includes a specific electrode index K.

[0101] At block 710, the mapping engine 101 performs a first peaking operation. The first peaking operation includes peaking a row vector K of a matrix. At block 715, the mapping engine 101 converts the global P×P weighting matrix into a diagonal matrix. According to one or more embodiments, the mapping engine 101 outputs a specific P×P weighting matrix for electrode K, where Wkk=1.

[0102] At block 718, the mapping engine 101 receives a third input. The third input includes N time samples with P electrode information in an N×P matrix. At block 720, the mapping engine 101 performs a matrix multiplication. The matrix multiplication includes a multiplication between an N×P matrix and a diagonal matrix to output a weighted information N×P matrix. At block 725, the mapping engine 101 performs an orthogonal decomposition of the weighted information N×P matrix to generate an orthogonal component N×P matrix.

[0103] At block 730, the mapping engine 101 performs a second peak operation. The second peak operation peaks the orthogonal component N×P matrix to find the largest component. The largest component is the common component.

[0104] At block 735, the mapping engine 101 performs orthogonal combination of the common components. The mapping engine 101 may utilize the N×P matrix of orthogonal components from the orthogonal decomposition of block 725. The mapping engine 101 provides the common components based on the combined N×P matrix. At block 740, the mapping engine 101 selects a column K. At block 745, the mapping engine 101 provides an output. The output includes an estimated far-field signal for electrode K.

[0105] At block 755, the mapping engine 101 also selects column K to determine original information for electrode K. At block 760, the mapping engine 101 performs far-field subtraction on the original information using the output of block 745 to determine residual information. At block 765, the mapping engine 101 provides an output, which includes estimated local information for each electrode.

[0106] 8 illustrates an exemplary catheter of an exemplary system configured to identify electrical activation signals at different distances from an electrode location, according to one or more embodiments. As shown, an exemplary mapping catheter 500 is inserted into the left ventricle (shown as a cross section through the myocardial tissue of the left ventricle 804) of the heart 12 of the patient 23, as may be performed during a cardiac ablation procedure to address ventricular tachycardia. According to certain embodiments described further below, the mapping engine 101 of the cardiac mapping system 10 may be configured to identify and / or visualize in three dimensions electrical activity below the surface of the myocardial tissue (e.g., 0.5-10 mm deep within the cardiac tissue) for an electrode, such as electrode 810 or electrode 814, in contact with the endocardium of the left ventricle 804.

[0107] For example, the mapping engine 101 may be utilized to determine electrical activity at the surface and / or at one or more depths below the surface, rather than just detecting electrical activity at the inner surface of the myocardial tissue of the left ventricle 804. In the illustrated example, the tissue depths 820 are 1 mm, 2 mm, 3 mm, and 4 mm below the inner surface of the myocardial tissue of the left ventricle 804, however, it will be appreciated that application of the methods herein may be applied for any tissue depth at any surface of the heart relevant to the target arrhythmia and therapeutic treatment (e.g., the inner surface of another one or more ventricles of the heart, at a depth within the myocardial tissue of that one or more ventricles). As discussed above, electrical activity may be detected or identified as near-field components, activation times (e.g., using minimum derivatives), and / or voltage levels (e.g., max-min). Advantageously, the exemplary methods described herein with respect to FIGS. 9-11 allow for identifying different electrical signals at each specified depth (e.g., depth 820) within the myocardial tissue.

[0108] FIG. 9 illustrates an exemplary method for separating electrical signals at a distance (e.g., depth) within cardiac tissue using ECG information collected from a multi-electrode mapping catheter, as shown in the example of FIG. 8. At block 902, a deep learning or machine learning model is trained to calculate weights for discriminating ECG signals at different distances or depths from a particular electrode. The data collection process for training such a model may be performed in a variety of ways. In one example, data is collected through preclinical studies. The collected data may include samples of electrical activity (e.g., near-field components, activation times (e.g., using minimum derivatives), and / or voltage levels (e.g., max-min)) at a surface of the myocardial tissue and at multiple locations within the myocardial tissue. For example, the collected data may include electrical activity for each of multiple points (e.g., point clouds, voxel clouds, or other three-dimensional structures) at multiple times (e.g., before, during, and / or after activation). In some implementations, the collected data may be collected via electroanatomical mapping (EAM). For example, in some implementations, the study may include using an animal or human model that includes a myocardial infarction (MI) scar or channel within the body. In some implementations, EAM may be performed to capture electrical signals and map the model. Depth measurements of the scar or channel are then taken during post-study dissection to obtain ground truth depth information. In some implementations, the training data may include simulated ECG signals (e.g., signals that appear as if the ECG signal were coming from a diagnostic catheter and are similar to previously measured signals). Producing such ECG signals may be done by varying real data, mixing real data (e.g., mixing signals measured from multiple studies or patients), generating statistically representative data (e.g., by determining the mean of the input data set and varying the data within one or more standard deviations), etc. The input data may be in any suitable format, such as an array of voltage values.Such voltage values ​​are typically given in mV and in some implementations may range from 0 to 5 mV, or may span a wider range (e.g., 0 to 10 mV, 0 to 20 mV, etc.). The training data may also include simulations of tissue, ventricular anatomy, and conductivity in different parts. For example, training data simulating scar may have a lower conductivity compared to healthy tissue, which has a higher conductivity.

[0109] Additionally or alternatively, simulations of electrical waves originating from different tissue depths within the myocardial tissue may be performed, thereby generating an analysis signal with a known depth of electrical source. For example, cardiac anatomical structures and conductivities may be simulated, including different structural variations of cardiac tissue, with and without scarring, or with scarring of different depths and sizes. Variations around different scarring, structural sizes, transmurality, partial scarring, etc. may be simulated, as would be measured if a catheter were used to map the electrical activity of the region. In some implementations, anatomical structures may also be varied.

[0110] Various deep learning or machine learning model architectures, including but not limited to convolutional neural network (CNN), recurrent neural network (RNN), or transformer, may be trained on the collected data depending on the particular characteristics of the collected data. The training data may be fed to the selected model and trained using an appropriate optimization algorithm, such as stochastic gradient descent (SGD) or Adam optimization. Through the training process, the model learns to associate the input ECG signals with their respective depths or distances, effectively learning weights to linearly combine components of the signals. In some implementations, the input data may be split into training data and test data, allowing testing and validation of the accuracy and / or sensitivity of the model. In some implementations, the input data may be enhanced to increase generalization, for example, by generating additional data (e.g., pseudo or virtual data generated from the input data via scaling, rotation, translation, etc., and similar thereto, referred to as shadow data, generalized data, or similar terms). Once a model has been successfully trained and evaluated, it can be deployed in an embodiment of the mapping engine 101 to identify signals at various depths in tissue.

[0111] Next, at block 904, spatial electrode signal analysis may be performed for each of the multiple electrodes of the mapping catheter in contact with the cardiac tissue. The spatial electrode signal analysis may be performed according to the above method 300 or method 400, including a multi-signal decomposition operation, such as but not limited to SVD or PCA, to generate multiple orthogonal signal components. Specifically, in some implementations, local unipolar signals may be identified and far-field signals may be suppressed to map the electrical activity of cells on the inner surface of the ventricle. The ECG signal measured at a particular point is the sum of different electrical signals received at that point, some signals being far-field (generated by more distant cardiac tissue), others being local (generated by inner cardiac tissue), and still others being between the two (i.e., near, but not at the surface). When the far-field signals are removed, the local and nearby signals are preserved.

[0112] At block 906, a mathematical combination of different components is performed to identify ECG signals at different distances or depths from a particular electrode in the plurality of electrodes. The mathematical combination includes linear and in some embodiments non-linear combinations to compensate for signal degradation or attenuation that depends on the distance of the signal from the target electrode. For example, in some implementations, the signals may be weighted (e.g., aX+bY+cZ). In other implementations, the signals may be weighted based on the depth of the tissue (e.g., X(z x )+Y(z y ^2), z nwhere is the depth from below the surface to the signal source). In other implementations, the signals may be added based on the distance from the electrode (e.g., the x,y Cartesian distance between the signal source and the electrode). Various weighting and / or exponential functions may be utilized. The weights or weighting coefficients may be adjusted in a myriad of ways, including by employing deep learning or other artificial intelligence (AI) or machine learning (ML) algorithms trained using the data in the manner described above with reference to block 902. By considering the spatial information contained in the orthogonal signals (e.g., x,y coordinates, x,y,z coordinates, z depth measurements, etc.) and manipulating their contributions through mathematical combinations, the mapping engine 101 may distinguish and extract (i.e., identify) electrical signals originating from various distances around a particular electrode, including depths within the myocardial tissue.

[0113] Finally, at block 908, the ECG signals at multiple distances or depths from the electrodes are visualized on a display. The visualization may take the form of a conventional electrogram 21, where the electrograms shown on the display may be selectable based on multiple distances, e.g., tissue depths into the myocardial tissue (e.g., 1, 2, 3, 4, ..., 10 mm). Additionally or alternatively, the visualization may take the form of an electroanatomical 3D map, where the map includes multiple "shells," each corresponding to a particular tissue depth into the myocardial tissue (e.g., 1, 2, 3, 4, ..., 10 mm or more). Each shell may be constructed in the same manner as used by electroanatomical systems to represent signal propagation along the surface of the tissue in 3D cardiac maps, such as those available in the CARTO® 3 system from Biosense Webster. In some implementations, the visualization may provide a point cloud, voxel cloud, or other three-dimensional representation. In some implementations, the visualization may show a smooth surface at selectable depths within the tissue (e.g., showing a map at a depth of 2 mm). Different depths can be selected by the user or physician, and in some implementations the visualization can be animated to cycle through different depths (e.g., stepping through each depth repeatedly, either sequentially over time or one point at a time).

[0114] FIG. 10 illustrates an alternative exemplary method for separating electrical signals at distances (e.g., depths) within cardiac tissue using ECG information collected from a multi-electrode mapping catheter, as shown in the example of FIG. 8. Beginning at block 1002, simulated data is generated for various conditions. More specifically, a set of simulated data may be generated, including various catheter locations, membrane potential locations, and various conditions, including but not limited to activation states of myocardial tissue (e.g., healthy, scarred, fibrotic). The data may be in any suitable format, such as an array of potentials at a given time, an array of voxels with coordinates and voltage values ​​(e.g., {x, y, z, v}), a depth map, or any other such format. The simulation also preferably considers the transition point from an inactivated state to an activated state of membrane potential (e.g., a determined threshold voltage, such as 0.5 mV, 1 mV, 0.5 V, or any other suitable voltage threshold).

[0115] As the next block 1004, a neural network is trained using the simulated data, where the network takes as input ECG readings from a multi-electrode catheter (or simulated readings from a model of myocardial tissue at various activation states) and outputs an estimated distance to the nearest activation in the myocardial tissue. The training process includes adjusting the weights of the network using a cost function, specifically, the difference between the network output at each point and the distance of the activation from the electrodes. In some implementations, as discussed above, the input or simulated data may be split into a first set of training data and a second set of test data. The weights may be modified to increase the accuracy and / or sensitivity of the trained model to general data.

[0116] At block 1006, the mapping engine 101 may implement the trained model to analyze ECG readings from the multi-electrode catheter and output, in real time, an estimated distance to the nearest activation for each electrode. The nearest activation may be on the inner surface of the myocardial tissue or in deeper tissue and may be identified by depth or distance to the activation. In some implementations, this may be repeated for multiple depths (e.g., the mapping engine 101 may implement the trained model to analyze ECG readings from the multi-electrode catheter and output, for each electrode, an estimated distance to the nearest activation at a first depth, such as 1 mm, then repeat the process for a depth of 2 mm, a depth of 3 mm, etc.). In other implementations, the process need not be iterative, and instead the trained model may provide a list of depths and a voltage signal per each depth.

[0117] Finally, the electrical signals are visualized on a display for multiple distances or depths from the electrodes in block 1008. Exemplary visualizations may take the form as described above with respect to block 908 of FIG.

[0118] Figure 11 illustrates yet another alternative exemplary method for separating electrical signals at distances (e.g., depths) within cardiac tissue using ECG information collected from a multi-electrode mapping catheter as shown in the example of Figure 8. At block 1102, the mapping engine 101 may implement a mathematical model that includes a group of sources (e.g., dipoles) distributed on a surface and defined within the cardiac tissue, the model preferably including a representation of the spatial distribution, activation times, and other relevant parameters of the dipole sources.

[0119] Next, for each electrode, the mathematical model and the known electrode positions are applied to calculate the expected signal (i.e., field projection) at each electrode location, in block 1104. This calculation may include determining the electrical potential at each electrode based on the dipole activations and their positions.

[0120] Optimization and inverse problem solving techniques are then applied to determine the dipole activations that match (i.e., best align) the input ECG readings with the calculated field projections, at block 1106. In some embodiments, the optimization and inverse problem solving techniques may include optimization algorithms such as gradient descent, Levenberg-Marquardt, or genetic algorithms to iteratively adjust the dipole activations to minimize an objective function.

[0121] Finally, the electrical signals are visualized on a display for multiple distances or depths from the electrodes in block 1108. Exemplary visualizations may take the form as described above with respect to block 908 of FIG.

[0122] The flow diagrams and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flow diagrams or block diagrams may represent a module, segment, or portion of instructions that includes one or more executable instructions for implementing the depicted logical function. In some alternative implementations, the functions shown in the blocks may occur in an order other than that shown in the figures. For example, two blocks shown in succession may in fact be executed substantially simultaneously, or the blocks may sometimes be executed in the reverse order, depending on the functionality involved. It should also be noted that each block of the block diagrams and / or flow diagrams, as well as combinations of blocks in the block diagrams and / or flow diagrams, may be implemented by a dedicated hardware-based system that performs a particular function or operation, or may operate or execute a combination of dedicated hardware and computer instructions. Although features and elements have been described above in certain combinations, those skilled in the art will understand that each feature or element can be used alone or in combination with other features and elements. In addition, the methods described herein may be implemented in a computer program, software, or firmware embodied in a computer-readable medium for execution on a computer or processor. Computer-readable medium, as used herein, should not be construed as being a transitory signal in itself, such as an electric wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable), or an electrical signal transmitted through a current line.

[0123] Examples of computer readable media include electrical signals (transmitted over wired or wireless connections) and computer readable storage media. Examples of computer readable storage media include, but are not limited to, registers, cache memory, semiconductor memory devices, magnetic media such as internal hard disks and removable disks, magneto-optical media, optical media such as compact disks (CDs) and digital versatile disks (DVDs), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), and memory sticks. The processor may be used with software to implement a radio frequency transceiver for use in a terminal, base station, or any host computer.

[0124] The terms used herein are for the purpose of describing particular embodiments only and are not intended to be limiting. As used herein, the singular forms "a", "an" and "the" include the plural unless the context clearly dictates otherwise. It is further understood that the terms "comprise" and / or "comprising", as used herein, indicate the presence of stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or other features, integers, steps, operations, elements, components, and / or groups thereof. The description of different embodiments herein is provided for illustrative purposes, but is not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles, practical applications, or technical improvements of the embodiments compared to the art found in the market, or to enable the skilled artisan to understand the embodiments disclosed herein.

[0125] [Embodiment] (1) A method comprising: receiving, by a mapping engine executed by one or more processors, electrical activity from a plurality of electrodes of the catheter; and identifying within the electrical activity electrical signals originating from deep within the cardiac tissue. (2) The method of embodiment 1, further comprising identifying within the electrical activity an electrical signal originating from an endocardial surface of the cardiac tissue. (3) performing, by the mapping engine, a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes; 2. The method of embodiment 1, wherein identifying electrical signals originating from a depth within the cardiac tissue comprises calculating a linear combination of signal components within the electrical activity. (4) The method of embodiment 3, wherein calculating a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model. (5) The method of embodiment 3, wherein identifying electrical signals originating from depth within the cardiac tissue further comprises calculating a nonlinear combination of the signal components within the electrical activity.

[0126] (6) The method of embodiment 1, wherein identifying electrical signals originating from a depth within cardiac tissue includes providing the electrical activity from the plurality of electrodes as input to a neural network, and determining, via the neural network, an estimated distance to a nearest activation from a corresponding electrode for at least one of the plurality of electrodes. (7) The method of embodiment 1, wherein identifying electrical signals originating from a depth within the cardiac tissue includes, for each of the plurality of electrodes, applying a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position. (8) The method of embodiment 1, wherein the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes. (9) The method of embodiment 1, wherein identifying electrical signals originating from a depth within the cardiac tissue includes running a machine learning model trained on simulated ECG data. (10) The method of embodiment 1, further comprising generating, by the mapping engine, a visual display of the identified electrical signal at a specified depth within the cardiac tissue.

[0127] (11) A system comprising: a memory for storing mapping engine software; one or more processors executing the mapping engine, receiving electrical activity from the electrodes of the catheter; and one or more processors configured to identify, within the electrical activity, electrical signals originating from deep within the cardiac tissue. (12) The system of embodiment 11, wherein the one or more processors are further configured to identify, within the electrical activity, electrical signals originating from an endocardial surface of the cardiac tissue. (13) The one or more processors are further configured to perform a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes; 12. The system of claim 11, wherein identifying electrical signals originating from a depth within the cardiac tissue comprises calculating a linear combination of signal components within the electrical activity. (14) The system of embodiment 13, wherein computing a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model. (15) The system of embodiment 13, wherein identifying electrical signals originating from a depth within the cardiac tissue further comprises calculating a nonlinear combination of the signal components within the electrical activity.

[0128] (16) The system of embodiment 11, wherein the one or more processors are further configured to provide the electrical activity from the plurality of electrodes as an input to a neural network and determine, via the neural network, for at least one of the plurality of electrodes, an estimated distance from a corresponding electrode to a nearest activation. (17) The system of embodiment 11, wherein the one or more processors are further configured to, for each of the plurality of electrodes, apply a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position. (18) The system of embodiment 11, wherein the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes. (19) The system of embodiment 11, wherein the processor is further configured to execute a machine learning model trained on simulated ECG data. (20) The system of embodiment 11, wherein the one or more processors are further configured to generate a visual display of the identified electrical signal at a specified depth within the cardiac tissue.

[0129] (21) A method comprising the steps of: receiving, by a mapping engine (101) executed by one or more processors (61), electrical activity from a plurality of electrodes (26) of the catheter (14); and identifying within the electrical activity an electrical signal originating from a depth (820) within the cardiac tissue (12). (22) The method of embodiment 21, further comprising identifying within the electrical activity an electrical signal originating from an endocardial surface of the cardiac tissue (12). (23) performing, by the mapping engine (101), a spatial electrode signal analysis of the electrical activity for each electrode (26) of the plurality of electrodes (26); 22. The method of claim 21, wherein identifying the electrical signal originating from a depth (820) within the cardiac tissue (12) comprises calculating a linear combination of signal components within the electrical activity. (24) The method of embodiment 23, wherein computing a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model. (25) The method of embodiment 23, wherein identifying the electrical signal originating from a depth (820) within the cardiac tissue (12) further comprises calculating a nonlinear combination of the signal components within the electrical activity.

[0130] (26) The method of embodiment 21, wherein identifying an electrical signal originating from a depth (820) within the cardiac tissue (12) includes providing the electrical activity from the plurality of electrodes (26) as an input to a neural network, and determining, via the neural network, an estimated distance from a corresponding electrode (26) to a nearest activation for at least one of the plurality of electrodes (26). (27) The method of embodiment 21, wherein identifying an electrical signal originating from a depth (820) within the cardiac tissue (12) includes, for each electrode (26) of the plurality of electrodes (26), applying a mathematical model and known electrode positions of the plurality of electrodes (26) to calculate an expected signal at each known electrode position. (28) The method of embodiment 21, wherein the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes (26). (29) The method of embodiment 21, wherein identifying electrical signals originating from a depth within the cardiac tissue includes running a machine learning model trained on simulated electrocardiogram (ECG) data. (30) The method of embodiment 21, further comprising generating, by the mapping engine (101), a visual representation (20) of the identified electrical signal at a specified depth within the cardiac tissue (12).

[0131] (31) A system comprising: A memory (62) for storing software for the mapping engine (101); One or more processors (61) executing the mapping engine (101), receiving electrical activity from a plurality of electrodes (26) of the catheter (14); and one or more processors (61) configured to identify, within the electrical activity, electrical signals originating from a depth (802) within the cardiac tissue (12). (32) The system described in embodiment 31, wherein the one or more processors (61) are further configured to identify, within the electrical activity, electrical signals originating from an endocardial surface of the cardiac tissue (12). (33) The one or more processors (61) are further configured to perform a spatial electrode signal analysis of the electrical activity for each electrode (26) of the plurality of electrodes (26); 32. The system of embodiment 31, wherein identifying electrical signals originating from a depth (820) within the cardiac tissue (12) comprises calculating a linear combination of signal components within the electrical activity. (34) The system of embodiment 33, wherein computing a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model. (35) The system of embodiment 33, wherein identifying an electrical signal originating from a depth (820) within the cardiac tissue (12) further comprises calculating a nonlinear combination of the signal components within the electrical activity.

[0132] (36) The system of embodiment 31, wherein the one or more processors (61) are further configured to provide the electrical activity from the plurality of electrodes (26) as an input to a neural network and determine, via the neural network, for at least one of the plurality of electrodes (26), an estimated distance from a corresponding electrode (26) to a nearest activation. (37) The system of embodiment 31, wherein the one or more processors (61) are further configured to, for each electrode (26) of the plurality of electrodes (26), calculate an expected signal at each known electrode position by applying a mathematical model and known electrode positions of the plurality of electrodes (26). (38) The system of embodiment 31, wherein the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes (26). (39) The system of embodiment 31, wherein the processor is further configured to execute a machine learning model trained on simulated electrocardiogram (ECG) data. (40) The system described in embodiment 31, wherein the one or more processors (61) are further configured to generate a visual display (20) of the identified electrical signal at a specified depth within the cardiac tissue (12).

[0133] (41) A method comprising the steps of: receiving, by a mapping engine executed by one or more processors, electrical activity from a plurality of electrodes of the catheter; and identifying within the electrical activity electrical signals originating from deep within the cardiac tissue. (42) The method of embodiment 41, further comprising identifying within the electrical activity an electrical signal emanating from an endocardial surface of the cardiac tissue. (43) performing, by the mapping engine, a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes; 43. The method of embodiment 41 or 42, wherein identifying electrical signals originating from a depth within the cardiac tissue comprises calculating a linear combination of signal components within the electrical activity. (44) The method of embodiment 43, wherein computing a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model. (45) The method of embodiment 43, wherein identifying electrical signals originating from a depth within the cardiac tissue further comprises calculating a nonlinear combination of the signal components within the electrical activity.

[0134] (46) A method according to any of embodiments 41 to 45, wherein identifying electrical signals originating from a depth within cardiac tissue includes providing the electrical activity from the plurality of electrodes as input to a neural network, and determining, via the neural network, an estimated distance from a corresponding electrode to a nearest activation for at least one of the plurality of electrodes. (47) A method according to any of embodiments 41 to 46, wherein identifying electrical signals originating from a depth within the cardiac tissue includes, for each of the plurality of electrodes, applying a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position. (48) The method of any of embodiments 41 to 47, wherein the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes. (49) The method of any of embodiments 41 to 48, wherein identifying electrical signals originating from a depth within the cardiac tissue includes running a machine learning model trained on simulated electrocardiogram (ECG) data. (50) The method of any of embodiments 41 to 49, further comprising generating, by the mapping engine, a visual display of the identified electrical signal at a specified depth within the cardiac tissue.

[0135] (51) A system comprising: a memory for storing mapping engine software; one or more processors executing the mapping engine, receiving electrical activity from the electrodes of the catheter; and one or more processors configured to identify, within the electrical activity, electrical signals originating from deep within the cardiac tissue. (52) The system of embodiment 51, wherein the one or more processors are further configured to identify, within the electrical activity, electrical signals originating from an endocardial surface of the cardiac tissue. (53) The one or more processors are further configured to perform a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes; The system of embodiment 51 or 52, wherein identifying electrical signals originating from depth within the cardiac tissue includes calculating a linear combination of signal components within the electrical activity. (54) The system of embodiment 53, wherein calculating a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model. (55) The system of embodiment 53, wherein identifying electrical signals originating from a depth within the cardiac tissue further comprises calculating a nonlinear combination of the signal components within the electrical activity.

[0136] (56) The system described in any of embodiments 51 to 55, wherein the one or more processors are further configured to provide the electrical activity from the plurality of electrodes as an input to a neural network and determine, via the neural network, for at least one of the plurality of electrodes, an estimated distance from a corresponding electrode to a nearest activation. (57) A system described in any of embodiments 51 to 56, wherein the one or more processors are further configured to, for each of the plurality of electrodes, apply a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position. (58) A system described in any of embodiments 51 to 57, wherein the received electrical activity includes real-time electrocardiogram (ECG) readings from the plurality of electrodes. (59) The system of any of embodiments 51 to 58, wherein the processor is further configured to execute a machine learning model trained on simulated electrocardiogram (ECG) data. (60) A system described in any of embodiments 51 to 59, wherein the one or more processors are further configured to generate a visual display of the identified electrical signal at a specified depth within the cardiac tissue.

Claims

1. 1. A system comprising: a memory for storing mapping engine software; one or more processors executing the mapping engine, receiving electrical activity from the electrodes of the catheter; and one or more processors configured to identify, within the electrical activity, electrical signals originating from deep within the cardiac tissue.

2. The system of claim 1 , wherein the one or more processors are further configured to identify, within the electrical activity, electrical signals emanating from an endocardial surface of the cardiac tissue.

3. the one or more processors are further configured to perform a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes; The system of claim 1 , wherein identifying electrical signals originating from a depth within the cardiac tissue comprises calculating a linear combination of signal components within the electrical activity.

4. 4. The system of claim 3, wherein computing a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model.

5. The system of claim 3 , wherein identifying electrical signals originating from a depth within cardiac tissue further comprises calculating a nonlinear combination of the signal components within the electrical activity.

6. 2. The system of claim 1, wherein the one or more processors are further configured to provide the electrical activity from the plurality of electrodes as an input to a neural network and determine, via the neural network, for at least one of the plurality of electrodes, an estimated distance from a corresponding electrode to a nearest activation.

7. 2. The system of claim 1, wherein the one or more processors are further configured to: for each of the plurality of electrodes, apply a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position.

8. The system of claim 1 , wherein the received electrical activity comprises real-time electrocardiogram (ECG) readings from the plurality of electrodes.

9. The system of claim 1 , wherein the processor is further configured to execute a machine learning model trained on simulated ECG data.

10. The system of claim 1 , wherein the one or more processors are further configured to generate a visual display of the identified electrical signal at a specified depth within the cardiac tissue.

11. 1. A method comprising: receiving, by a mapping engine executed by one or more processors, electrical activity from a plurality of electrodes of the catheter; and identifying within the electrical activity electrical signals originating from deep within the cardiac tissue.

12. The method of claim 11 , further comprising identifying within the electrical activity electrical signals originating from an endocardial surface of the cardiac tissue.

13. performing, by the mapping engine, a spatial electrode signal analysis of the electrical activity for each electrode of the plurality of electrodes; The method of claim 11 , wherein identifying electrical signals originating from a depth within the cardiac tissue comprises calculating a linear combination of signal components within the electrical activity.

14. 14. The method of claim 13, wherein computing a linear combination of signal components in the electrical activity further comprises multiplying each signal component in the electrical activity by a corresponding weight determined via a trained model.

15. 14. The method of claim 13, wherein identifying electrical signals originating from a depth within cardiac tissue further comprises calculating a nonlinear combination of the signal components within the electrical activity.

16. 12. The method of claim 11, wherein identifying electrical signals originating from a depth within cardiac tissue comprises: providing the electrical activity from the plurality of electrodes as input to a neural network; and determining, via the neural network, for at least one of the plurality of electrodes, an estimated distance from a corresponding electrode to a nearest activation.

17. 12. The method of claim 11, wherein identifying electrical signals originating from a depth within cardiac tissue comprises, for each of the plurality of electrodes, applying a mathematical model and known electrode positions of the plurality of electrodes to calculate an expected signal at each known electrode position.

18. The method of claim 11 , wherein the received electrical activity comprises real-time electrocardiogram (ECG) readings from the multiple electrodes.

19. 12. The method of claim 11, wherein identifying electrical signals originating from a depth within the cardiac tissue comprises running a machine learning model trained on simulated ECG data.

20. The method of claim 11 , further comprising generating, by the mapping engine, a visual representation of the identified electrical signal at a specified depth within the cardiac tissue.

Images