remove far field from intracardiac signals

Abstract

In one embodiment there is provided a method comprising: receiving a first intracardiac signal comprising a first far-field component captured by at least one first sensing electrode of a first catheter, the at least one sensing electrode being in contact with tissue of a heart chamber of a first living subject, and at least one far-field signal captured from at least one far-field electrode, the at least one far-field electrode being inserted into the heart chamber and not in contact with tissue of the heart chamber; training a neural network to remove far-field components from intracardiac signals in response to the first intracardiac signal and the at least one far-field signal; receiving a second intracardiac signal captured by at least one second sensing electrode of a second catheter inserted into a heart chamber of a second living subject; and applying the trained neural network to the second intracardiac signal to remove a corresponding second far-field component from the second intracardiac signal.

Description

[0001] Relevant application information

[0002] This patent application claims the benefit of U.S. Provisional Patent Application No. 63 / 070,897, filed August 27, 2020, and U.S. Provisional Patent Application No. 63 / 073,414, filed September 1, 2020, the disclosures of which are incorporated herein by reference. Technical Field

[0003] This invention relates to medical systems, and specifically, but not exclusively, to the processing of cardiac signals. Background Technology

[0004] Numerous medical procedures involve placing probes, such as catheters, inside a patient's body. Position sensing systems have been developed to track these probes. Magnetic position sensing is one method known in the art. In magnetic position sensing, a magnetic field generator is typically placed at a known location outside the patient's body. A magnetic field sensor within the distal end of the probe generates electrical signals in response to these magnetic fields; these signals are processed to determine the coordinate position of the distal end of the probe. These methods and systems are described in U.S. Patents 5,391,199, 6,690,963, 6,484,118, 6,239,724, 6,618,612, and 6,332,089, in PCT International Patent Publication WO 1996 / 005768, and in U.S. Patent Application Publications 2002 / 0065455, 2003 / 0120150, and 2004 / 0068178. Impedance- or current-based systems can also be used to track position.

[0005] Treatment of arrhythmias is a medical procedure in which these types of probes or catheters have proven extremely useful. Arrhythmias, and specifically atrial fibrillation, have always been a common and dangerous medical condition, especially in the elderly.

[0006] The diagnosis and treatment of cardiac arrhythmias involve mapping the electrical properties of cardiac tissue, particularly the endocardium, and selectively ablating cardiac tissue by applying energy. Such ablation can stop or alter unwanted electrical signals propagating from one part of the heart to another. Ablation methods disrupt unwanted electrical pathways by creating a non-conductive ablation focus. Various forms of energy delivery for creating ablation focuses have been disclosed, including the use of microwaves, lasers, and more commonly, radiofrequency energy to create conduction blocks along the cardiac tissue walls. In a two-step procedure (mapping followed by ablation), electrical activity at various points within the heart is typically sensed and measured by advancing a catheter containing one or more electrical sensors into the heart and acquiring data at multiple points. This data is then used to select the target endocardial region for ablation.

[0007] Electrode catheters have been widely used in medical practice for many years. They are used to stimulate and map electrical activity in the heart, as well as to ablate sites of abnormal electrical activity. In use, the electrode catheter is inserted into a major vein or artery, such as the femoral vein, and then guided to the cardiac chamber of interest. A typical ablation procedure involves inserting a catheter with one or more electrodes at its distal end into the cardiac chamber. A reference electrode can be provided, typically taped to the patient's skin, or a second catheter positioned in or near the heart can be used to provide the reference electrode. RF (radio frequency) current is applied between the catheter electrodes of the ablation catheter and an unrelated electrode (which may be one of the catheter electrodes), and the current flows through the medium between these electrodes (i.e., blood and tissue). The current distribution can depend on the amount of contact between the electrode surface and the tissue compared to blood, which has a higher conductivity than tissue. Heating of the tissue occurs due to its resistance. The tissue is sufficiently heated to destroy cells in the cardiac tissue, resulting in the formation of a non-conductive ablation focus within the cardiac tissue. In some applications, irreversible electroporation can be performed to ablate the tissue.

[0008] Sensors within the cardiac chambers can detect far-field electrical activity, i.e., peripheral electrical activity originating far from the sensor, which can distort or obscure local electrical activity, i.e., signals originating at or near the sensor. U.S. Patent Application Publication 2014 / 0005664, jointly assigned to Govari et al., discloses the differentiation between local components of intracardiac electrode signals arising from tissue in contact with the electrodes and their far-field contributions to the signal, and explains how therapeutic processes applied to the tissue can be controlled in response to the differentiated local components. Summary of the Invention

[0009] According to embodiments of this disclosure, a method for analyzing signals is provided, the method comprising: receiving a first intracardiac signal including a first far-field component captured by at least one first sensing electrode of a first catheter and at least one far-field signal captured from at least one far-field electrode, the at least one sensing electrode being in contact with tissue of a heart chamber of a first living subject, the at least one far-field electrode being inserted into the heart chamber but not in contact with the tissue of the heart chamber; training an artificial neural network to remove the far-field component from the intracardiac signal in response to the received first intracardiac signal and at least one far-field signal; receiving a second intracardiac signal captured by at least one second sensing electrode of a second catheter inserted into the heart chamber of a second living subject; and applying the trained artificial neural network to the second intracardiac signal to remove a corresponding second far-field component from the second intracardiac signal.

[0010] Further according to an embodiment of the present disclosure, the method includes calculating a first intracardiac signal with a corresponding first far-field component removed in response to the at least one far-field signal, wherein the training includes training the artificial neural network in response to the calculated first intracardiac signal with the corresponding first far-field component removed.

[0011] Furthermore, according to an embodiment of this disclosure, the training includes training an autoencoder, which includes an encoder and a decoder.

[0012] Additionally, according to embodiments of this disclosure, the method includes presenting to a display a representation of at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed.

[0013] Furthermore, according to embodiments of this disclosure, the method includes generating and presenting an electroanatomical map to a display in response to at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed from the second intracardiac signal.

[0014] Further according to an embodiment of this disclosure, the first conduit includes at least one far-field electrode.

[0015] According to another embodiment of this disclosure, a method for analyzing signals is also provided, the method comprising receiving intracardiac signals captured by at least one sensing electrode of a catheter inserted into the heart chamber of a living subject, and applying a trained artificial neural network to the intracardiac signals to remove corresponding far-field components from the intracardiac signals.

[0016] Furthermore, according to an embodiment of this disclosure, the method includes presenting to a display a representation of at least one intracardiac signal from which a corresponding far-field component of the far-field component has been removed.

[0017] Additionally, according to embodiments of this disclosure, the method includes generating and presenting an electroanatomical map to a display in response to at least one intracardiac signal in which a corresponding far-field component of the far-field component has been removed from the intracardiac signal.

[0018] According to another embodiment of this disclosure, a software product is also provided, comprising a non-transitory computer-readable medium storing program instructions that, when read by a central processing unit (CPU), cause the CPU to: receive a first intracardiac signal comprising a first far-field component captured by at least one first sensing electrode of a first catheter and at least one far-field signal captured from at least one far-field electrode, the at least one sensing electrode being in contact with tissue of a heart chamber of a first living subject, the at least one far-field electrode being inserted into the heart chamber but not in contact with the tissue of the heart chamber; and, in response to the received first intracardiac signal and at least one far-field signal, train an artificial neural network to remove the far-field component from the intracardiac signal.

[0019] Further according to an embodiment of the present disclosure, wherein when these instructions are read by the CPU, the CPU also causes the CPU to: receive a second intracardiac signal captured by at least one second sensing electrode of a second catheter inserted into the heart chamber of a second living subject, and apply a trained artificial neural network to the second intracardiac signal to remove a corresponding second far-field component from the second intracardiac signal.

[0020] Furthermore, according to an embodiment of the present disclosure, when these instructions are read by the CPU, the CPU also causes the CPU to: calculate a first intracardiac signal with the corresponding first far-field component removed in response to at least one far-field signal, and to train an artificial neural network in response to the calculated first intracardiac signal with the corresponding first far-field component removed.

[0021] Furthermore, according to embodiments of this disclosure, when these instructions are read by the CPU, the CPU also causes the CPU to present to the display a representation of at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed.

[0022] Furthermore, according to embodiments of this disclosure, when these instructions are read by the CPU, the CPU also causes the CPU to generate and present an electroanatomical map to the display in response to at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component is removed from the second intracardiac signal.

[0023] According to another embodiment of this disclosure, a software product is also provided, comprising a non-transitory computer-readable medium storing program instructions that, when read by a central processing unit (CPU), cause the CPU to: receive intracardiac signals captured by at least one sensing electrode of a catheter inserted into the heart cavity of a living subject, and apply a trained artificial neural network to the intracardiac signals to remove corresponding far-field components from the intracardiac signals.

[0024] Further according to an embodiment of the present disclosure, when these instructions are read by the CPU, the CPU also causes the CPU to present to the display a representation of at least one intracardiac signal in which a corresponding far-field component of the far-field component has been removed from the intracardiac signal.

[0025] Furthermore, according to an embodiment of this disclosure, when these instructions are read by the CPU, the CPU also causes the CPU to generate and present an electroanatomical map to a display in response to the removal of at least one of the corresponding far-field components of the intracardiac signal from the far-field components.

[0026] According to another embodiment of this disclosure, a medical system is also provided, the medical system including a first catheter and a processor; the first catheter includes at least one first sensing electrode configured to be inserted into a heart chamber of a first living subject; the processor is configured to: receive a first intracardiac signal including a first far-field component captured by the at least one first sensing electrode of the first catheter and at least one far-field signal captured from the at least one far-field electrode, and train an artificial neural network to remove the far-field component from the intracardiac signal in response to the received first intracardiac signal and at least one far-field signal, the at least one sensing electrode being in contact with tissue of the heart chamber of the first living subject, the at least one far-field electrode being inserted into the heart chamber and not in contact with the tissue of the heart chamber.

[0027] Furthermore, according to embodiments of this disclosure, the processor is configured to: calculate a first intracardiac signal with a corresponding first far-field component removed in response to at least one far-field signal, and train an artificial neural network in response to the calculated first intracardiac signal with the corresponding first far-field component removed.

[0028] Further according to an embodiment of the present disclosure, the artificial neural network includes an autoencoder, the autoencoder including an encoder and a decoder, the processor being configured to train the autoencoder to remove far-field components from the intracardiac signal in response to a received first intracardiac signal and at least one far-field signal.

[0029] Furthermore, according to an embodiment of the present disclosure, the system includes a second catheter including at least one second sensing electrode configured to be inserted into the heart chamber of a second living subject, wherein the processor is configured to: receive a second intracardiac signal captured by at least one second sensing electrode of the second catheter inserted into the heart chamber of the second living subject, and apply a trained artificial neural network to the second intracardiac signal to remove a corresponding second far-field component from the second intracardiac signal.

[0030] Additionally, according to embodiments of this disclosure, the trained artificial neural network includes an autoencoder comprising an encoder and a decoder, and the processor is configured to apply the autoencoder to the second intracardiac signal to remove a corresponding second far-field component from the second intracardiac signal.

[0031] Furthermore, according to an embodiment of this disclosure, the system includes a display, wherein the processor is configured to present to the display a representation of at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed.

[0032] Further according to an embodiment of the present disclosure, the system includes a display, wherein the processor is configured to generate and present an electroanatomical map to the display in response to at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed from the second intracardiac signal.

[0033] Furthermore, according to an embodiment of this disclosure, the first catheter includes at least one far-field electrode.

[0034] Additionally, according to an embodiment of this disclosure, the first catheter includes an expandable distal end basket assembly, at least one first sensing electrode disposed on the basket assembly, and at least one far-field electrode disposed in the basket assembly to prevent the at least one far-field electrode from contacting the cardiac cavity tissue of the first living subject.

[0035] According to another embodiment of this disclosure, a medical system is also provided, the medical system including a catheter and a processor; the catheter includes at least one sensing electrode configured to be inserted into the heart chamber of a living subject; the processor is configured to: receive intracardiac signals captured by the at least one sensing electrode inserted into the heart chamber, and apply a trained artificial neural network to the intracardiac signals to remove corresponding far-field components from the intracardiac signals.

[0036] Furthermore, according to embodiments of this disclosure, the trained artificial neural network includes an autoencoder comprising an encoder and a decoder, and the processor is configured to apply the autoencoder to the intracardiac signal to remove the corresponding far-field component from the intracardiac signal.

[0037] Further according to an embodiment of the present disclosure, the system includes a display, wherein the processor is configured to present to the display a representation of at least one intracardiac signal in which a corresponding far-field component of the far-field component has been removed.

[0038] Furthermore, according to an embodiment of the present disclosure, the system includes a display, wherein the processor is configured to generate and present an electroanatomical map to the display in response to at least one intracardiac signal in which a corresponding far-field component of the far-field component has been removed from the intracardiac signal. Attached Figure Description

[0039] The invention will be understood from the following detailed description taken in conjunction with the accompanying drawings, wherein:

[0040] Figure 1 A schematic diagram illustrating a system for performing catheter insertion surgery on the heart, constructed and operated according to an exemplary embodiment of the present invention;

[0041] Figure 2 For use with Figure 1 A perspective view of the conduits used in the system;

[0042] Figure 3 For use with Figure 1 Detailed schematic diagram of the electrode assembly used in the system;

[0043] Figure 4 For use Figure 3 The expected curve of the signal obtained from the electrode assembly;

[0044] Figure 5 For use with Figure 1 A schematic diagram of an artificial neural network used in a system;

[0045] Figure 6 To show the Figure 5 A schematic diagram illustrating the training of an artificial neural network;

[0046] Figure 7 For including training Figure 5 A flowchart of the steps in an artificial neural network method;

[0047] Figure 8 For catheters and by using Figure 1 A schematic diagram of the signals captured by the catheter in the system;

[0048] Figure 9 To illustrate the processing being done by a trained artificial neural network Figure 8 A schematic diagram illustrating the processing of the captured signals;

[0049] Figure 10 This includes processing using trained artificial neural networks. Figure 8 A flowchart of the steps in the method for capturing signals;

[0050] Figure 11 This is a schematic diagram showing the intracardiac signals; and

[0051] Figure 12 This is a schematic diagram of the electroanatomical illustration shown. Detailed Implementation

[0052] Overview

[0053] As previously mentioned, intracardiac electrode sensors detect far-field electrical activity, i.e., peripheral electrical activity originating far from the sensor. This far-field activity can distort or obscure local electrical activity, i.e., signals originating at or near the sensor. Removing far-field electrical activity from intracardiac signals is not a minor issue. This is because the far-field profile depends on many factors, including electrode shape and size, catheter location, anatomical structures, etc. Furthermore, the far and near fields share a common frequency, and therefore, using only low-pass, high-pass, or band-pass filters does not provide a satisfactory solution.

[0054] Some catheters can be used to provide signals that can be used to remove the far-field component from intracardiac signals. For example, a basket catheter may include a central electrode placed in the middle of the basket, which is maintained sufficiently far from the cardiac tissue to provide a reasonable estimate of the far-field component. The signal sensed by this central electrode can be used to remove the far-field component from the intracardiac signal sensed by the basket electrode. However, not all catheters have electrodes that measure only the far field. For example, balloon catheters or planar mesh catheters do not include electrodes that are always sufficiently far from the tissue. For such catheters, the problem of removing the far-field component from the sensed intracardiac signal remains.

[0055] Embodiments of the present invention address the aforementioned problem by training an artificial neural network (ANN) (such as an autoencoder) based on a set of training signals to remove far-field components from intracardiac signals. These training signals can be provided by a basket catheter that captures intracardiac signals within the heart chambers, and also simultaneously captures far-field signals using far-field electrodes (such as a central electrode positioned at the center of a basket assembly of the basket catheter).

[0056] An ANN can be trained by feeding the captured intracardiac signal into it. By using an iterative process, the parameters of the ANN are iteratively updated to reduce the difference between the ANN's actual output and the desired output (e.g., intracardiac signal with far-field components removed).

[0057] Once trained, the ANN can be applied to other intracardiac signals captured by another catheter (e.g., a catheter without a “far-field electrode”) in the same or different patients to remove the far-field component from the captured intracardiac signal.

[0058] In some implementations, the weights of the ANN can be sent to a cloud server, where the ANN can be executed to remove far-field components from the intracardiac signals sent to that server.

[0059] System Description

[0060] Now for reference Figure 1 This is a schematic diagram illustrating a medical system 10 constructed and operated according to an exemplary embodiment of the present invention for performing catheter insertion surgery on a heart 12. The medical system 10 can be configured to assess electrical activity on the heart 12 of a living subject and perform ablation surgery thereon. The system includes a catheter 14, which is inserted by an operator 16 through the skin across the patient's vascular system into a chamber or vascular structure of the heart 12. The operator 16 (typically a physician) contacts the distal end 18 of the catheter, for example, with the heart wall at the ablation target site. An electrical activity mapping can be prepared according to methods disclosed in U.S. Patents 6,226,542, 6,301,496, and 6,892,091. An article including elements of the system 10 may be... The system was purchased from Biosense Webster, Inc. (3333 Diamond Canyon Road, Diamond Bar, CA 91765). This system can be modified by those skilled in the art to implement the principles of the invention described herein.

[0061] Areas identified as abnormal, for example, by assessment using an electrical activity mapping technique, can be ablated by applying thermal energy. This is achieved, for instance, by conducting radiofrequency current through a catheter via wires to one or more electrodes at the distal end 18, which apply radiofrequency energy to the myocardium. The energy is absorbed in the tissue, heating it to a temperature at which it permanently loses its electrical excitability. Upon successful completion of this procedure, non-conductive ablation foci are formed in the cardiac tissue, disrupting abnormal electrical pathways leading to arrhythmias. The principles of this invention can be applied to different cardiac chambers to diagnose and treat a variety of different arrhythmias.

[0062] The catheter 14 typically includes a handle 20 with suitable controls to allow the operator 16 to manipulate, position, and orient the distal end 18 of the catheter 14 as required for ablation. To assist the operator 16, the distal portion of the catheter 14 includes a position sensor (not shown) that provides a signal to a processor 22 in a control console 24. The processor 22 can perform several processing functions as described below.

[0063] The lead connector 35 connects the console 24 to the surface electrode 30 and other components of the positioning subsystem for measuring the position and orientation coordinates of the catheter 14. The processor 22 or another processor (not shown) may be a component of the positioning subsystem. The catheter electrode (not shown) and the surface electrode 30 may be used to measure tissue impedance at the ablation site, as taught in U.S. Patent 7,536,218. A temperature sensor (not shown), typically a thermocouple or thermistor, may be mounted on the ablation surface of the distal portion of the catheter 14 as described below.

[0064] The console 24 typically includes one or more ablation power generators 25. The catheter 14 may be adapted to deliver ablation energy (e.g., radiofrequency energy, ultrasound energy, irreversible electroporation, and laser-generated light energy) to the heart using any known ablation technique. Such methods are disclosed in U.S. Patents 6,814,733, 6,997,924, and 7,156,816.

[0065] In one embodiment, the positioning subsystem includes a magnetic positioning tracking arrangement that utilizes a magnetic field generating coil 28 to determine the position and orientation of the catheter 14 by generating a magnetic field at a predetermined working volume and sensing these magnetic fields at the catheter. This positioning subsystem is described in U.S. Patents 7,756,576 and 7,536,218.

[0066] As described above, catheter 14 is coupled to console 24, allowing operator 16 to observe and control the function of catheter 14. Console 24 includes processor 22, typically a computer with appropriate signal processing circuitry. Processor 22 is coupled to drive display 29 (e.g., a monitor). This signal processing circuitry typically receives, amplifies, filters, and digitizes signals from catheter 14, including signals generated by sensors such as electrical sensors, temperature sensors, and contact force sensors, and multiple position sensing electrodes (not shown) located distal to catheter 14. The digitized signals are received by console 24 and used by the positioning system to calculate the position and orientation of catheter 14 and to analyze the electrical signals from the electrodes.

[0067] To generate electroanatomical maps, processor 22 typically includes an electroanatomical map generator, an image alignment program, an image or data analysis program, and a graphical user interface configured to present graphical information on display 29.

[0068] Typically, system 10 includes other components, but these are not shown in the figures for simplicity. For example, system 10 may include an electrocardiogram (ECG) monitor coupled to receive signals from one or more surface electrodes to provide ECG synchronization signals to console 24. As described above, system 10 typically also includes a reference position sensor, either located on an externally applied reference patch attached to the outside of the subject's body or on an internal catheter inserted into and held in a fixed position relative to heart 12. Conventional pumps and tubing may be provided to circulate fluid through catheter 14 to cool the ablation site. System 10 may receive image data from external imaging modalities such as MRI units and includes an image processor, which may be incorporated into or invoked by processor 22 for generating and displaying images.

[0069] In implementation, some or all of these functions of processor 22 may be combined in a single physical component, or alternatively implemented using multiple physical components. These physical components may include hardwired or programmable devices, or a combination of both. In some embodiments, at least some of the functions of processor 22 may be implemented by a programmable processor under the control of suitable software. This software may be downloaded electronically to the device via, for example, a network. Alternatively or otherwise, the software may be stored in a tangible, non-transitory computer-readable storage medium, such as optical, magnetic, or electronic memory.

[0070] Now for reference Figure 2 It is used for with Figure 1 A perspective view of the conduit 14 used in system 10.

[0071] The catheter 14 includes an elongated shaft 39 having a proximal end and a distal end, a control handle 20 at the proximal end of the catheter body, and an expandable distal end basket assembly 43 mounted at the distal end of the shaft 39.

[0072] Shaft 39 includes an elongated tubular structure having a single axial or central lumen (not shown), but optionally multiple lumens if desired. Shaft 39 is flexible, bendable, but substantially incompressible along its length. Shaft 39 can have any suitable construction and can be made of any suitable material. In some embodiments, the elongated shaft 39 includes an outer wall made of polyurethane or polyether block amide. This outer wall includes an embedded braided mesh made of stainless steel or the like to increase the torsional stiffness of shaft 39, such that the distal end of shaft 39 rotates accordingly when the control handle 20 is rotated.

[0073] The outer diameter of shaft 39 is not critical, but can be in the range of approximately 2 mm to 5 mm. Similarly, the thickness of the outer wall is not critical, but is generally thin enough that the central lumen can accommodate any one or more of the following: pull wires, lead wires, sensor cables, and any other wires, cables, or tubes. If desired, the inner surface of the outer wall can be lined with a reinforcing tube (not shown) to provide improved torsional stability. U.S. Patent 6,064,905 describes and illustrates an example of a catheter body construction suitable for use in conjunction with the present invention.

[0074] Assembly 43 is mounted to the distal end of shaft 39. For example... Figure 2 As shown, the basket assembly 43 includes five splines 45 or arms mounted generally evenly spaced around a contraction line 47 connected to the distal end of the assembly 43, which, depending on the specific application, contracts, retracts, and expands the assembly 43 when a traction or thrust is applied longitudinally to the contraction line 47. The contraction line 47 forms a longitudinal axis of symmetry of the assembly 43. All splines 45 are attached directly or indirectly to the contraction line 47 at their distal ends and to the shaft 39 at their proximal ends. As the contraction line 47 moves longitudinally to expand and contract the assembly 43, the splines 45 bend outwards in the expanded position and are generally straight in the contracted position. As those skilled in the art will recognize, depending on the specific application, the number of splines 45 can be varied as needed, such that the assembly 43 has at least two splines, typically at least three splines, and up to ten or more splines. The expandable distal end basket assembly 43 is not limited to the configuration shown, but may include other designs such as spherical or egg-shaped designs, which include multiple expandable arms connected directly or indirectly at their proximal and distal ends.

[0075] The assembly 43 includes at least one sensing electrode 49 disposed thereon. In some embodiments, each spline of the spline 45 may include a flexible wire with a non-conductive overlay, on which one or more sensing electrodes of the sensing electrodes 49 (e.g., annular spline electrodes) are mounted. For convenience, the electrode 49 is referred to as a “sensing electrode,” but it can also be used for ablation. In some embodiments, each flexible wire includes a flat nitinol wire, and each non-conductive overlay includes a biocompatible plastic tube such as polyurethane or polyimide tubing. Alternatively, the spline 45 may be designed without an inner flexible wire if a sufficiently rigid non-conductive material is used for the non-conductive overlay to allow the assembly 43 to expand, provided that the spline has an outer surface that is non-conductive on at least a portion of its surface for mounting the sensing electrode 49. In some embodiments, the spline may be formed of a flexible polymer spline circuit, wherein the electrode 49 is disposed on the outer surface of each flexible polymer spline circuit.

[0076] Each sensing electrode in the sensing electrodes 49 on the spline 45 is electrically connected to a suitable marking or monitoring system and / or ablation energy source by means of an electrode lead (not shown). The electrode lead extends through the control handle 20, through a lumen in the shaft 39, into the non-conductive cover of the corresponding spline 45, and is attached to its corresponding sensing electrode 49 by any suitable method. The catheter 14 includes a far-field electrode 51, such as a cylindrical electrode, disposed on the contraction line 47. The far-field electrode 51 is disposed in the expandable distal end basket assembly 43 to prevent contact between the far-field electrode 51 and the tissue of the heart chambers of the heart 12. In some embodiments, the catheter 14 may include more than one far-field electrode 51.

[0077] The function of the far-field electrode 51 is described below. In some embodiments, the far-field electrode 51 may be positioned on a different catheter that is inserted into the heart 12 simultaneously with the catheter 14. Additional details of the catheter 14 are described in U.S. Patent 6,748,255, which is cited above.

[0078] The catheter 14 typically has multiple electrodes 49 arranged on multiple flexible splines of the basket assembly 43. The catheter 14 is configured to be inserted into the heart 12 in a collapsed form. Figure 1 In the heart chambers of the heart 12, splines 45 are positioned close to each other. Once in the heart 12, splines 45 can be formed into their expanded basket shape by a contraction line 47, which holds the distal ends of splines 45 and pulls the distal ends of splines 45 in the proximal direction.

[0079] Now for reference Figure 3 , it is Figure 2 A detailed schematic diagram of the expandable distal end basket assembly 43. In the expanded form of the assembly 43, at least a portion of the sensing electrodes 49 of the spline 45 contact the endocardial surface 53 of the heart 12 and acquire signals corresponding to the electrode potentials generated at their contact points with the surface. However, since the sensing electrodes 49 are located in a conductive medium (blood), the acquired signals include far-field components from other regions of the heart 12 in addition to the electrode potentials from the contact points.

[0080] These far-field components constitute interference signals on the endocardial surface electrode potential. To counteract this interference, embodiments of the invention position the far-field electrode 51 on the systolic line 47. In the expanded configuration of the assembly 43, the far-field electrode 51 is positioned on the systolic line 47 to be approximately equidistant from all corresponding sensing electrodes 49 (i.e., sensing electrodes 49 equidistant from fixed reference points on the long axis of the catheter, such as reference point 55 at the proximal end of the assembly 43), and is prevented from contacting the cardiac surface by spline 45. For example, electrodes 57 and 59 are equidistant from reference point 55 and also from far-field electrode 51, as indicated by dashed lines 61 and 63, respectively. When the far-field electrode 51 is spaced at least 0.5 cm from the sensing electrode 49 in the expanded configuration of the assembly 43, the far-field electrode acquires far-field signals from the endocardial surface 53 but not near-field signals. However, the signal e(t) acquired by the sensing electrode 49 has both far-field and surface (near-field) components. The far-field component signal x(t) acquired by far-field electrode 51 is removed from the signal e(t) acquired by sensing electrode 49 in order to cancel out interference suffered by these electrodes, i.e., by signal subtraction: e(t) – x(t). Alternatively or additionally, any suitable method may be used to remove the far-field component.

[0081] In some embodiments, the conduit 14 is provided with a distal position sensor 65 and a proximal position sensor 67, the distal position sensor being mounted at or near the distal end of the connecting spline, and the proximal position sensor being mounted at or near the proximal end of the assembly 43, thereby allowing the coordinates of the position sensor 65 relative to the position sensor 67 to be determined in use, and combined with known information about the curvature of the spline 45 to locate the position of each sensing electrode in the sensing electrodes 49.

[0082] Now for reference Figure 4 It is usable Figure 3 The expected curves of the exemplary signals obtained by assembly 43. Graph 69 shows the intracardiac signal e(t) obtained from a monopolar or bipolar configuration of sensing electrode 49. Graph 71 is the signal recording line x(t) of far-field electrode 51, which may be a parallel recording line. Graph 73 is the recording line of the signal obtained when the signal of graph 71 is subtracted from the signal of graph 69 or when the far-field component in the electrogram e(t) is removed by applying the algorithms described in U.S. Patent Publication 2016 / 0175023 or U.S. Patent 9,554,718 with necessary modifications.

[0083] Now for reference Figure 5 It is used for with Figure 1 A schematic diagram of the artificial neural network 75 used in System 10.

[0084] A neural network is a network or circuit of neurons, or, in the modern sense, an artificial neural network composed of artificial neurons or nodes. The connections of biological neurons are modeled as weights. Positive weights reflect excitatory connections, while negative values ​​represent inhibitory connections. The input is modified by the weights and summed using a linear combination. Activation functions control the amplitude of the output. For example, an acceptable output range is typically between 0 and 1, or the range can be between -1 and 1.

[0085] These artificial networks can be used for predictive modeling, adaptive control, and applications, and can be trained on datasets. Experience-based self-learning can occur within the network, drawing conclusions from complex and seemingly unrelated groups of information.

[0086] For completeness, a biological neural network consists of one or more groups of chemically connected or functionally related neurons. A single neuron can connect to many other neurons, and the total number of neurons and connections in a network can be extensive. Connections (called synapses) typically form from an axon to a dendrite, but dendritic synapses and other connections are also possible. In addition to electrical signals, other forms of signaling, caused by the diffusion of neurotransmitters, exist.

[0087] Artificial intelligence, cognitive modeling, and neural networks are information processing paradigms inspired by how biological nervous systems process data. Artificial intelligence and cognitive modeling attempt to simulate some characteristics of biological neural networks. In the field of artificial intelligence, artificial neural networks have been successfully applied to speech recognition, image analysis, and adaptive control to build software agents or autonomous robots (in computers and video games).

[0088] In terms of artificial neurons, often referred to as artificial neural networks (ANNs) or simulated neural networks (SNNs), a neural network (NN) is a set of interconnected natural or artificial neurons that process information using mathematical or computational models based on computational connection methods. In most cases, an ANN is an adaptive system that adapts its structure based on the flow of external or internal information through the network. More practically, the term neural network is used for modeling nonlinear statistical data or as a decision-making tool. These terms can be used to model complex relationships between inputs and outputs or to find patterns in data.

[0089] In some implementation schemes, such as Figure 5 As shown, the artificial neural network 75 may include an autoencoder 77, which includes an encoder 79 and a decoder 81. In other embodiments, the artificial neural network 75 may include any suitable ANN. The artificial neural network 75 may be implemented in software and / or hardware.

[0090] Encoder 79 includes an input layer 83 where input is received. The encoder then includes one or more hidden layers 85 that progressively compress the input into code 87. Decoder 81 includes one or more hidden layers 89 that progressively decompress the code 87 up to an output layer 91 providing the output of autoencoder 77. Autoencoder 77 includes weights between these layers. Autoencoder 77 manipulates the data received at input layer 83 based on the values ​​of these weights between its layers.

[0091] During training of the autoencoder 77, the weights of the autoencoder 77 are updated, enabling the autoencoder 77 to perform the data manipulation tasks that were trained to perform. Figure 5 In the example, the autoencoder 77 is trained to remove far-field components from the intracardiac signal, as shown in the reference. Figure 6 and Figure 7 To describe in more detail.

[0092] The number and width of layers in the autoencoder 77 can be configured. As the number and width of layers increase, the accuracy with which the autoencoder 77 can manipulate data according to the task at hand also increases. However, a larger number of layers and wider layers typically require more training data, more training time, and training may not converge. For example, the input layer 83 may include 400 neurons (e.g., to compress a batch of 400 samples). The encoder 79 may include five layers compressed by a factor of 2 (e.g., 400, 200, 100, 50, 25). The decoder may include five layers decompressed by a factor of 2 (e.g., 25, 50, 100, 200, 400).

[0093] Now for reference Figure 6 and Figure 7 . Figure 6 To show the Figure 5 A schematic diagram of training an artificial neural network 75. Figure 7 For including training Figure 5 The flowchart 100 shows the steps in the method of artificial neural network 75.

[0094] Based on catheters such as Figures 1 to 3 The data captured by catheter 14 is used to train the artificial neural network 75. Electrodes 49 of catheter 14 ( Figure 3 ) and heart 12 ( Figure 1 The tissues of the endocardium (e.g., the endocardial surface 53) Figure 3Contact. Electrode 49 provides an intracardiac signal 93, including a far-field component. To provide high-quality training data, operator 16 typically confirms excellent contact between electrode 49, which provides the intracardiac signal 93, and the tissue. The far-field electrode 51 of catheter 14 ( Figure 3 It provides at least one far-field signal.

[0095] Processor 22 ( Figure 1 The catheter 14 is configured to receive (box 102) intracardiac signals 93 including far-field components captured by one or more electrodes in electrode 49 that are in contact with tissue of the heart chamber of a living subject. The catheter 14 can provide signals 93 from different electrodes 49 when it is located in a given position within the heart chamber, and / or can provide signals 93 from one or more electrodes 49 as the catheter 14 moves to different positions within the heart chamber. Intracardiac signals 93 can be provided from different heart chambers and even from different living subjects.

[0096] Processor 22 ( Figure 1 The catheter is also configured to receive (box 102) far-field signals captured by far-field electrode 51 (inserted into the heart chamber and not in contact with the heart chamber tissue), and simultaneously receive intracardiac signals 93 captured by electrode 49 of catheter 14 at any position of movement of catheter 14 during the capture of intracardiac signals 93. Therefore, each intracardiac signal 93 has a corresponding far-field signal captured within the same time interval for capturing that intracardiac signal 93. Multiple intracardiac signals 93 may have the same corresponding far-field signal captured simultaneously by multiple intracardiac signals 93.

[0097] In some implementations, processor 22 ( Figure 1 The system is configured to calculate (box 104) the intracardiac signal 93 with the corresponding far-field component removed in response to a far-field signal, thereby generating a cleaned intracardiac signal 95. The above reference can be used. Figure 3 One of the methods described or any suitable method for removing the far-field component from the intracardiac signal can be used to calculate the cleaned intracardiac signal 95.

[0098] Processor 22 is configured to train (box 106) an artificial neural network 75 (e.g., an autoencoder 77) to remove far-field components from the intracardiac signal in response to the received intracardiac signal 93 and the far-field signal captured by far-field electrode 51. In some embodiments, processor 22 is configured to train artificial neural network 75 in response to the intracardiac signal 93 and the calculated cleaned intracardiac signal 95 (i.e., the intracardiac signal 93 with the corresponding far-field components removed).

[0099] Training an artificial neural network 75 is typically an iterative process. One method for training an artificial neural network 75 will now be described below.

[0100] Processor 22 is configured to input the received intracardiac signal 93 (box 108, arrow 97) into artificial neural network 75. For example, the intracardiac signal 93 is input into input layer 83 of encoder 79. Processor 22 is configured to compare the output of artificial neural network 75 (e.g., the output of decoder 81 of autoencoder 77) with the desired output (i.e., the corresponding cleaned intracardiac signal 95) (box 110, arrow 99). For example, if there is a set of intracardiac signals A, B, C output by artificial neural network 75 and a corresponding set of cleaned intracardiac signals A', B', and C', processor 22 compares A with A', B with B', C with C', and so on. This comparison is typically performed using a suitable loss function that calculates the total difference between all outputs of artificial neural network 75 and all desired outputs (e.g., all corresponding cleaned intracardiac signals 95).

[0101] At decision box 112, processor 22 is configured to determine whether the difference between the output of artificial neural network 75 and the desired output is sufficiently small. If the difference between the output of artificial neural network 75 and the desired output is sufficiently small (branch 118), processor 22 is configured to save (box 120) the parameters (e.g., weights) of artificial neural network 75 (e.g., autoencoder 77) and / or send the parameters (e.g., weights) to a cloud processing server (not shown).

[0102] If the difference is not small enough (branch 114), processor 22 is configured to modify (box 116) the parameters (e.g., weights) of the artificial neural network 75 (e.g., autoencoder 77) to reduce the difference between the output of the artificial neural network 75 and the desired output of the artificial neural network 75. In the example above, the difference minimized is the total difference between all outputs of the artificial neural network 75 and all desired outputs (e.g., all corresponding cleaned intracardiac signals 95). Processor 22 is configured to modify the parameters using any suitable optimization algorithm (e.g., gradient descent algorithms such as the Adam optimization algorithm). Steps in boxes 108, 110, and 112 are then repeated.

[0103] Now for reference Figure 8 It is a catheter 200 and made of... Figure 1 A schematic diagram of the intracardiac signal 202 captured by the catheter 200 in system 10. The catheter 200 is a planar mesh catheter comprising a plurality of splines 204, wherein each spline 204 has an electrode 206 (only some electrodes are labeled for simplicity). The catheter 200 is configured to be inserted into the heart chamber of a living subject. The living subject can be: the same living subject with the catheter 14 inserted and the artificial neural network 75 trained thereon; or different living subjects.

[0104] The catheter 200 does not include electrodes that typically do not contact cardiac tissue, and therefore capturing signals that represent only the far field using the catheter 200 would be extremely difficult. (See below for reference.) Figure 9 and Figure 10 In more detail, the medical system 10 is configured to use a trained artificial neural network 75. Figure 6 Remove the far-field component from the intracardiac signal 202.

[0105] Catheter 200 is an example of a catheter that provides an intracardiac signal including the far-field component to be removed. Any suitable catheter (e.g., a balloon catheter) or even a catheter that includes a far-field electrode (e.g., a suitable basket catheter) can provide an intracardiac signal that is then processed by a trained artificial neural network 75 to remove the far-field component from the provided intracardiac signal.

[0106] Now for reference Figure 9 and Figure 10 . Figure 9 To illustrate the processing being done by the trained artificial neural network 75 Figure 8 A schematic diagram of the processing of the captured signal 202. Figure 10 This includes processing using trained artificial neural networks 75. Figure 8 The flowchart 250 shows the steps in the method for capturing signal 202. See also: Figure 8 .

[0107] Processor 22 ( Figure 1 The processor 22 is configured to receive (box 252) intracardiac signals 202 captured by sensing electrodes 206 of a catheter 200 inserted into the heart chamber of a living subject. The processor 22 is configured to apply (box 254, arrow 208) a trained artificial neural network 75 to the intracardiac signals 202 to remove corresponding far-field components from the intracardiac signals 202, thereby generating (arrow 212) a corresponding cleaned intracardiac signal 210.

[0108] In some embodiments, the trained artificial neural network includes a trained autoencoder 77. In these embodiments, the processor 22 ( Figure 1 The system is configured to apply the autoencoder 77 to the intracardiac signal 202 to remove the corresponding far-field component from the intracardiac signal 202.

[0109] Now for reference Figure 11 This is a schematic diagram representing the intracardiac signals shown in diagram 214. See also... Figure 10 Processor 22 ( Figure 1 ) is optionally configured to present (box 256) cleaned intracardiac signals 210 to display 29. Figure 9(i.e., the intracardiac signal 202 with the corresponding far-field component removed) is represented as 214.

[0110] Now for reference Figure 12 This is a schematic diagram of the electroanatomical illustration shown in Figure 216. See also: Figure 10 Processor 22 ( Figure 1 Optionally configured to generate and present (box 258) an electroanatomical illustration 216 to a display 29 in response to a cleaned intracardiac signal 210 (i.e., an intracardiac signal 202 with the corresponding far-field component removed).

[0111] As used herein, the term “about” or “approximately” for any numerical value or range indicates appropriate dimensional tolerances that allow a collection of parts or components to achieve the intended purpose as described herein. More specifically, “about” or “approximately” may refer to a range of values ​​±20% of the listed values; for example, “about 90%” may refer to a range of values ​​from 72% to 108%.

[0112] For clarity, the various features of the invention described in the context of individual embodiments may also be provided in combination in a single embodiment. Conversely, for brevity, the various features of the invention described in the context of individual embodiments may also be provided individually or in any suitable sub-combination.

[0113] The above embodiments are cited by way of example, and the invention is not limited to the specific examples shown and described above. Rather, the scope of the invention includes combinations and sub-combinations of the various features described above, as well as variations and modifications thereof, which will occur to those skilled in the art upon reading the above description and are not disclosed in the prior art.

Claims

1. A method for analyzing signals, the method comprising: Receiving: a first intracardiac signal including a first far-field component captured by at least one first sensing electrode of a first catheter, the at least one sensing electrode being in contact with the tissue of the heart chamber of a first living subject; and at least one far-field signal captured from at least one far-field electrode, said at least one far-field electrode being inserted into the heart chamber and not in contact with the tissue of the heart chamber; The artificial neural network is trained in response to the received first intracardiac signal and the at least one far-field signal to remove the far-field component from the intracardiac signal. Receive a second intracardiac signal captured by at least one second sensing electrode of a second catheter inserted into the cardiac chamber of a second living subject; as well as A trained artificial neural network is applied to the second intracardiac signal to remove the corresponding second far-field component from the second intracardiac signal.

2. The method of claim 1, further comprising calculating a first intracardiac signal with a corresponding first far-field component removed in response to the at least one far-field signal, wherein the training includes training the artificial neural network in response to the calculated first intracardiac signal with the corresponding first far-field component removed.

3. The method of claim 1, wherein the training includes training an autoencoder, the autoencoder comprising an encoder and a decoder.

4. The method of claim 1, further comprising presenting to a display a representation of at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed.

5. The method of claim 1, further comprising generating and displaying an electroanatomical map in response to at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed from the second intracardiac signal.

6. The method of claim 1, wherein the first conduit includes the at least one far-field electrode.

7. A software product comprising a non-transitory computer-readable medium storing program instructions that, when read by a central processing unit (CPU), cause the CPU to: Receiving: a first intracardiac signal including a first far-field component captured by at least one first sensing electrode of a first catheter, the at least one sensing electrode being in contact with the tissue of the heart chamber of a first living subject; and at least one far-field signal captured from at least one far-field electrode, said at least one far-field electrode being inserted into the heart chamber and not in contact with the tissue of the heart chamber; The artificial neural network is trained in response to the received first intracardiac signal and the at least one far-field signal to remove the far-field component from the intracardiac signal. Receive a second intracardiac signal captured by at least one second sensing electrode of a second catheter inserted into the cardiac chamber of a second living subject; as well as A trained artificial neural network is applied to the second intracardiac signal to remove the corresponding second far-field component from the second intracardiac signal.

8. The software product of claim 7, wherein the instruction, when read by the CPU, further causes the CPU to: Calculate the first intracardiac signal with the corresponding first far-field component removed in response to the at least one far-field signal; and The artificial neural network is trained in response to the calculated first intracardiac signal with the corresponding first far-field component removed.

9. The software product of claim 7, wherein the instruction, when read by the CPU, further causes the CPU to present to the display a representation of at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed from the second intracardiac signal.

10. The software product of claim 7, wherein the instruction, when read by the CPU, further causes the CPU to generate and present an electroanatomical diagram to a display in response to at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed from the second intracardiac signal.

11. A medical system comprising: A first catheter, the first catheter including at least one first sensing electrode configured to be inserted into the heart chamber of a first living subject; The second catheter includes at least one second sensing electrode configured to be inserted into the heart chamber of a second living subject. and Processor, the processor being configured to: Receiving: a first intracardiac signal including a first far-field component captured by at least one first sensing electrode of the first catheter, the at least one sensing electrode being in contact with the tissue of the heart chamber of the first living subject; and at least one far-field signal captured from the at least one far-field electrode, the at least one far-field electrode being inserted into the heart chamber and not in contact with the tissue of the heart chamber; The artificial neural network is trained in response to the received first intracardiac signal and the at least one far-field signal to remove the far-field component from the intracardiac signal. Receive a second intracardiac signal captured by at least one second sensing electrode of the second catheter inserted into the heart chamber of the second living subject; as well as A trained artificial neural network is applied to the second intracardiac signal to remove the corresponding second far-field component from the second intracardiac signal.

12. The system of claim 11, wherein the processor is configured to: Calculate the first intracardiac signal with the corresponding first far-field component removed in response to the at least one far-field signal; and The artificial neural network is trained in response to the calculated first intracardiac signal with the corresponding first far-field component removed.

13. The system of claim 11, wherein the artificial neural network includes an autoencoder, the autoencoder including an encoder and a decoder, and the processor is configured to train the autoencoder to remove the far-field component from the intracardiac signal in response to the received first intracardiac signal and the at least one far-field signal.

14. The system of claim 11, wherein the trained artificial neural network includes an autoencoder, the autoencoder including an encoder and a decoder, and the processor is configured to apply the autoencoder to the second intracardiac signal to remove the corresponding second far-field component from the second intracardiac signal.

15. The system of claim 11, further comprising a display, wherein the processor is configured to present to the display a representation of at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed.

16. The system of claim 11, further comprising a display, wherein the processor is configured to generate and present an electroanatomical map to the display in response to at least one second intracardiac signal in which a corresponding second far-field component of the second far-field component has been removed from the second intracardiac signal.

17. The system of claim 11, wherein the first conduit includes the at least one far-field electrode.

18. The system of claim 17, wherein the first catheter includes an expandable distal end basket assembly, the at least one first sensing electrode is disposed on the basket assembly, and the at least one far-field electrode is disposed in the basket assembly to prevent the at least one far-field electrode from contacting the tissue of the heart chamber of the first living subject.

Images