Noise reduction in recording devices

Abstract

To reduce noise in the potentiometer signal. [Solution] In one embodiment, the method includes receiving a cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with tissue, introducing the cardiac signal segment into a cable extending to a recording device, wherein the cable outputs a corresponding noise-compensated cardiac signal segment in response to electrical noise acquired within the cable, training an artificial neural network to at least partially compensate for electrical noise added to the signal within the cable in response to the received cardiac signal segment and the corresponding noise-compensated cardiac signal segment, receiving a cardiac signal in response to electrical activity sensed by a second sensing electrode, and applying the trained artificial neural network to the cardiac signal to generate a noise-compensated cardiac signal that at least partially compensates for electrical noise that is not yet in the cardiac signal but is added to the cardiac signal within the cable.

Description

Technical Field

[0001] (Cross - reference to Related Applications) This application claims the benefit of U.S. Provisional Patent Application No. 63 / 108,998, filed on November 3, 2020, the disclosure of which is incorporated herein by reference.

[0002] (Field of the Invention) The present invention relates to medical devices, and more particularly, but not exclusively, to reducing noise in electrocardiogram signals.

Background Art

[0003] A wide range of medical procedures involve placing probes, such as catheters, within a patient's body. To track such probes, position sensing systems have been developed. Magnetic position sensing is one of the known methods in the art. In magnetic position sensing, a magnetic field generator is typically placed at a known position outside the patient's body. Magnetic field sensors within the distal end of the probe generate electrical signals in response to these magnetic fields, and 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. Pat. Nos. 5,391,199, 6,690,963, 6,484,118, 6,239,724, 6,618,612, and 6,332,089, International Publication No. WO 1996 / 005768, and U.S. Patent Application Publication Nos. 2002 / 0065455, 2003 / 0120150, and 2004 / 0068178. Position may also be tracked using impedance or current - based systems.

[0004] One medical procedure in which these types of probes or catheters have proven to be extremely useful is in the treatment of cardiac arrhythmias. Cardiac arrhythmias, and particularly atrial fibrillation, persist as a common and dangerous medical condition, especially in the elderly population.

[0005] The diagnosis and treatment of cardiac arrhythmias involve mapping the electrical properties of cardiac tissue, particularly the endocardium, and selectively ablating the cardiac tissue by applying energy. Such ablation can stop or modify the propagation of unwanted electrical signals from one part of the heart to another. The ablation process destroys unwanted electrical pathways by forming non-conductive damaged areas. Various modes of energy delivery have been disclosed to date for the purpose of forming damaged areas, 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 that involves mapping followed by ablation, a catheter containing one or more electrical sensors is typically advanced into the heart to sense and measure electrical activity at each point within the heart by obtaining data at multiple points. These data are then used to select the target region of the endocardium to be ablated.

[0006] Electrode catheters have been commonly used in medical practice for many years. Electrode catheters are used to stimulate and map electrical activity within the heart and to ablate areas exhibiting abnormal electrical activity. During use, the electrode catheter is inserted into a major vein or artery, such as the femoral vein, and then guided into the cardiac chambers of the heart 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 may be provided, typically taped to the patient's skin, or by a second catheter positioned within or near the heart. RF (radio frequency) current is applied between the catheter electrodes (one or more) of the ablation catheter and an indifferent electrode (which may be one of the catheter electrodes), and the current flows through the medium between the electrodes, i.e., between blood and tissue. The distribution of the current may depend on the amount of electrode surface in contact with the tissue compared to blood, which has higher conductivity than tissue. Tissue heating occurs due to the electrical resistance of the tissue. When the tissue is sufficiently heated, cell destruction is induced in the cardiac tissue, resulting in the formation of damaged areas within the non-conductive cardiac tissue. In some applications, irreversible electroporation can be performed to ablate tissue. [Overview of the Initiative] [Means for solving the problem]

[0007] According to one embodiment of the present disclosure, a method for analyzing a signal is provided, which includes receiving a first cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with first biological tissue, introducing the received first cardiac signal segment into a recording device cable extending to a recording device, wherein the cable outputs a corresponding noise-enhanced cardiac signal segment in response to electrical noise acquired within the cable, training an artificial neural network to at least partially compensate for electrical noise added to the cardiac signal within the cable in response to the received first cardiac signal segment and the corresponding noise-enhanced cardiac signal segment, receiving a second cardiac signal in response to electrical activity sensed by a second sensing electrode in contact with second biological tissue, applying the trained artificial neural network to the second cardiac signal to generate a noise-compensated second cardiac signal that at least partially compensates for electrical noise not yet present in the second cardiac signal but added to the second cardiac signal within the cable, and outputting the noise-compensated second cardiac signal to a recording device via the cable.

[0008] Furthermore, according to embodiments of the present disclosure, training includes inputting a noise-enhanced cardiac signal segment into an artificial neural network and iteratively adjusting the parameters of the artificial neural network to reduce the difference between the output of the artificial neural network and the received first cardiac signal segment.

[0009] Furthermore, according to embodiments of the present disclosure, the method includes converting a first cardiac signal segment from digital to analog format, introducing the first cardiac signal segment in analog format into a cable, further including converting a noise-added cardiac signal segment to digital format, and training an artificial neural network in response to a received first cardiac signal segment in digital format and a corresponding noise-added cardiac signal segment in digital format.

[0010] In addition, according to embodiments of the present disclosure, training includes training an autoencoder, which includes an encoder and a decoder.

[0011] Furthermore, according to embodiments of the present disclosure, the method includes inserting a first catheter, including a first sensing electrode, into a cardiac chamber of a first living organism, and inserting a second catheter, including a second sensing electrode, into a cardiac chamber of a second living organism.

[0012] Furthermore, according to embodiments of this disclosure, the first catheter includes a second catheter.

[0013] Furthermore, according to embodiments of the present disclosure, introduction includes introducing a received first cardiac signal segment to a first end of a recording device cable, the method further includes electrically connecting the first end of a shielded cable to a second end of the recording device cable, the method further includes the second end of the shielded cable outputting a noise-enhanced cardiac signal segment, and training includes training an artificial neural network in response to the received first cardiac signal segment and the corresponding noise-enhanced cardiac signal segment output by the second end of the shielded cable.

[0014] According to another embodiment of the present disclosure, a software product comprising a non-temporary computer-readable medium in which program instructions are stored, wherein, when read by a central processing unit (CPU), the CPU receives a first cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with first biological tissue, and introduces the received first cardiac signal segment into a recording device cable extending to a recording device, wherein the cable outputs a corresponding noise-added cardiac signal segment in response to electrical noise acquired within the cable, and the received first cardiac signal segment and the corresponding noise-added cardiac signal segment A software product is provided that trains an artificial neural network to at least partially compensate for electrical noise added to the heart signal within the cable in response to electrical activity sensed by a second sensing electrode in contact with the tissue of a second living organism, receives a second heart signal in response to electrical activity sensed by a second sensing electrode in contact with the tissue of a second living organism, applies the trained artificial neural network to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for electrical noise that is not yet present in the second heart signal but is added to the second heart signal within the cable, and outputs the noise-compensated second heart signal to a recording device via the cable.

[0015] Furthermore, according to yet another embodiment of the present disclosure, a medical system is provided, the medical system comprising: a first sensing electrode configured to contact the tissue of a first living organism; a second sensing electrode configured to contact the tissue of a second living organism; a recording device cable extending to a recording device; and a processing circuit that receives a first cardiac signal segment in response to electrical activity sensed by the first sensing electrode in contact with the tissue; introduces the received first cardiac signal segment into the recording device cable, the cable being configured to output a corresponding noise-added cardiac signal segment in response to electrical noise acquired within the cable; and the received first cardiac signal segment and the corresponding noise-added The system comprises a processing circuit configured to: train an artificial neural network to at least partially compensate for electrical noise added to the cardiac signal in the cable in response to a cardiac signal segment; receive a second cardiac signal in response to electrical activity sensed by a second sensing electrode in contact with the tissue of a second living organism; apply the trained artificial neural network to the second cardiac signal to generate a noise-compensated second cardiac signal that at least partially compensates for electrical noise that is not yet in the second cardiac signal but is added to the second cardiac signal in the cable; and output the noise-compensated second cardiac signal to a recording device via the cable.

[0016] Furthermore, according to embodiments of the present disclosure, the processing circuit is configured to input a noise-added cardiac signal segment to an artificial neural network and iteratively adjust the parameters of the artificial neural network to reduce the difference between the output of the artificial neural network and the received first cardiac signal segment.

[0017] Furthermore, according to embodiments of the present disclosure, the processing circuit further comprises a digital-to-analog converter configured to convert a first heart signal segment from a digital format to an analog format, wherein the processing circuit is configured to introduce the first heart signal segment in analog format into a cable; and an analog-to-digital converter configured to convert a noise-added heart signal segment to a digital format, wherein the processing circuit is configured to train an artificial neural network to at least partially compensate for electrical noise added to the heart signal in the cable in response to a received first heart signal segment in digital format and a corresponding noise-added heart signal segment in digital format.

[0018] Furthermore, according to embodiments of the present disclosure, the artificial neural network includes an autoencoder including an encoder and a decoder, and the processing circuit is configured to train the autoencoder to at least partially compensate for electrical noise added to the heart signal in the cable in response to a first heart signal segment received and a corresponding noise-added heart signal segment.

[0019] Furthermore, according to one embodiment of the present disclosure, the processing circuit is configured to apply an autoencoder to the second heart signal to generate a noise-compensated second heart signal.

[0020] In addition, according to embodiments of the present disclosure, the system includes a first catheter equipped with a first sensing electrode and configured to be inserted into a first cardiac chamber of a living organism, and a second catheter equipped with a second sensing electrode and configured to be inserted into a second cardiac chamber of a living organism.

[0021] Furthermore, according to embodiments of this disclosure, the first catheter includes a second catheter.

[0022] Furthermore, according to an embodiment of the present disclosure, the system further includes a shielded cable having a first end and a second end, the recording device cable having a first end electrically connected to the processing circuit and a second end electrically connected to the recording device and the first end of the shielded cable, the second end of the shielded cable being electrically connected to the second end of the recording device cable, the processing circuit being configured to introduce the received first heart signal segment to the first end of the recording device cable, the second end of the shielded cable being configured to output the noise-added heart signal segment, and the processing circuit being configured to train an artificial neural network in response to the received first heart signal segment and the corresponding noise-added heart signal segment output by the second end of the shielded cable.

Brief Description of the Drawings

[0023] The present invention will be understood from the following detailed description in conjunction with the accompanying drawings. [Figure 1] It is a pictorial view of a system for performing a catheter procedure on the heart, the system being configured and operative in accordance with an embodiment of the present invention. [Figure 2] It is a perspective view of a catheter for use in the system of FIG. 1. [Figure 3] It is a detailed schematic view of an electrode assembly for use with the system of FIG. 1. [Figure 4] It is a detailed view of the processing circuit of the system of FIG. 1. [Figure 5] It is a flowchart including steps in the method of operating the system of FIG. 1. [Figure 6] It is a schematic view of an artificial neural network for use in the system of FIG. 1. [Figure 7] It is a schematic view showing the training of the artificial neural network of FIG. 6. [Figure 8] It is a flowchart including sub-steps in the steps of the method of FIG. 5. [Figure 9]This is a schematic diagram illustrating the processing of capture signals by a trained artificial neural network. [Figure 10] This is a flowchart showing the steps involved in processing the capture signal in Figure 9 using a trained artificial neural network. [Modes for carrying out the invention]

[0024] overview As mentioned above, during cardiac electrophysiological (EP) examinations or ablation procedures, the leads between the patient and the electrical system used in the procedure, such as the Carto® system (Biosense Webster, Inc., Irvine, California) and / or external recording devices, can pick up noise. Even if the noise originates from a known device, it may be difficult or even impossible to remove the device. For example, the Carto system may use an external uninterruptible power supply (UPS) that cannot be removed but generates electrical noise. The Carto system usually uses filtering techniques to clean up the noise signal. However, signals recorded by external recording devices can be very noisy, which can be a significant problem due to the very small voltages associated with the electrophysiological signal. Also, external recording devices are generally analog devices that do not have noise filtering capabilities and therefore cannot filter out noise from the received cardiac signal.

[0025] Embodiments of the present invention reduce problems associated with electrical noise pickup in the cable between a catheter (and / or body surface electrode) and a recording device by routing the cardiac signal captured by the catheter (and / or body surface electrode) through a processing circuit (e.g., part of a Carto system). The processing circuit compensates for expected noise pickup before outputting the cardiac signal to an external recording device via the recording device cable. The compensation is performed using an artificial neural network (ANN), which is trained to at least partially compensate for electrical noise that is not yet in the cardiac signal but is added to the cardiac signal in the cable from the processing circuit to the recording device.

[0026] An ANN may be trained as follows: The processing circuit receives a cardiac signal segment (which may be the same signal segment) in response to electrical activity sensed by one or more sensing electrodes (e.g., catheters and / or body surface electrodes) in contact with biological tissue. The processing circuit is electrically connected to a recording device via one end of a recording device cable, which is generally unshielded and therefore picks up ambient electrical noise. The other end of the recording device cable is electrically connected to a shielded cable that returns to the processing circuit. As used herein or in the claims, the term “shielded cable” is defined as an electrical cable of one or more insulated conductors surrounded by a common conductive layer, which may consist of braided strands of copper (or other metals such as aluminum), unbraided spiral windings of copper tape, or layers of conductive polymer.

[0027] The processing circuit introduces the received heart signal segment into the recording device cable, where the heart signal segment picks up noise, and the recording device cable outputs the noise-added heart signal segment. The noise-added heart signal segment then passes through the shielded cable and returns to the processing circuit.

[0028] The processing circuit trains an artificial neural network to at least partially compensate for electrical noise added to the heart signal within the cable in response to the received heart signal segment and the corresponding noise-enhanced heart signal segment. In some embodiments, the processing circuit inputs the noise-enhanced heart signal segment to the artificial neural network and iteratively adjusts the parameters of the artificial neural network (e.g., weights) to reduce the difference between the output of the artificial neural network and the received heart signal segment.

[0029] System Description Hereinafter, we refer to Figure 1, a pictorial illustration of a medical system 10 configured according to an embodiment of the present invention for performing a catheter insertion procedure on a working heart 12. The medical system 10 may be configured to evaluate electrical activity and perform an ablation procedure on the living heart 12. The system 10 comprises a catheter 14 that is percutaneously inserted by an operator 16 into a lumen or vascular structure of the heart 12 through the patient's vascular system. The operator 16, usually a physician, brings the distal end 18 of the catheter into contact with the heart wall, for example, at the ablation target site. An electrical activity map may be prepared according to the methods disclosed in U.S. Patents No. 6,226,542, No. 6,301,496, and No. 6,892,091. One commercially available product embodying the elements of system 10 is available as the CARTO® 3 system (available from Biosense Webster, Inc., Irvine, CA). This system may be modified by those skilled in the art to embody the principles of the invention described herein.

[0030] Areas identified as abnormal by evaluation of the electrical activity map can be ablated by applying thermal energy, for example, by passing a high-frequency current through a wire inside the catheter to one or more electrodes at the distal end 18 that apply high-frequency energy to the myocardium. This energy is absorbed into the tissue and heats it until the tissue permanently loses its electrical excitability. If performed successfully, this procedure creates non-conductive damaged areas in the cardiac tissue, which block the abnormal electrical pathways that cause arrhythmias. The principle of the present invention can be applied to different cardiac chambers to diagnose and treat a number of different cardiac arrhythmias.

[0031] The catheter 14 typically includes a handle 20 having a control unit suitable for a handle, which enables the operator 16 to manipulate, position, and orient the distal end 18 of the catheter 14 as desired for ablation. To assist the operator 16, the distal portion of the catheter 14 houses a position sensor (not shown) that supplies signals to a processing circuit 22 located in a console 24. The processing circuit 22 can perform several processing functions, as described below.

[0032] The wire connection section 35 may connect the console 24 to the body surface electrodes 30 and other components of the positioning subsystem for measuring the position and orientation coordinates of the catheter 14. A processing circuit 22 or another processor (not shown) may be an element of the positioning subsystem. As taught in U.S. Patent No. 7,536,218, the catheter electrodes (not shown) and the body surface electrodes 30 may be used to measure tissue impedance at the ablation site. A temperature sensor (not shown), typically a thermocouple or thermistor, may be placed on the ablation surface of the distal portion of the catheter 14, as described below.

[0033] The console 24 typically houses one or more ablation power generators 25. The catheter 14 can be adapted to conduct ablation energy to the heart using any known ablation technique, such as radiofrequency energy, ultrasonic energy, irreversible electroporation, and laser-generated light energy. Such methods are disclosed in U.S. Patents 6,814,733, 6,997,924, and 7,156,816.

[0034] In one embodiment, the positioning subsystem includes a magnetic position tracking arrangement that uses a magnetic field generating coil 28 to generate a magnetic field within a predetermined working volume and senses these magnetic fields in the catheter to determine the position and orientation of the catheter 14. The positioning subsystem is described in U.S. Patents No. 7,756,576 and No. 7,536,218.

[0035] As described above, the catheter 14 is coupled to the console 24, which allows the operator 16 to observe and adjust the function of the catheter 14. The console 24 includes a processing circuit 22, which is typically a computer with appropriate signal processing circuits. The processing circuit 22 is coupled to drive a display 29 (e.g., a monitor). The signal processing circuit receives, amplifies, filters, and digitizes signals from the catheter 14, typically including signals generated by sensors such as electrical sensors, temperature sensors, and contact force sensors, as well as multiple position-sensing electrodes (not shown) located distally within the catheter 14. The digitized signals are used by the console 24 and the positioning system to calculate the position and orientation of the catheter 14 and to analyze the electrical signals from the electrodes.

[0036] To generate an electroanatomical map, the processing circuit 22 typically includes a mapping module that includes an electroanatomical map generator, an image registration program, an image or data analysis program, and a graphical user interface configured to display graphical information on a display 29.

[0037] Although not shown in the diagram for simplification, system 10 typically includes other elements. For example, system 10 may include an electrocardiogram (ECG) monitor, which is coupled to receive signals from one or more surface electrodes 30 to supply an ECG synchronization signal to console 24. As mentioned above, system 10 also typically has a reference position sensor, either on an externally attached reference patch attached to the outside of the patient's body or on an internally-placed catheter inserted into the heart 12 and maintained in a fixed position relative to the heart 12. Conventional pumps and lines may be provided for circulating fluid through the catheter 14 to cool the ablation site. System 10 may also receive image data from an external imaging modality, such as an MRI unit or similar, and includes an image processor that is incorporated into, or can be activated by, a processing circuit 22 for generating and displaying images.

[0038] In practice, some or all of the functions of the processing circuit 22 can be combined into a single physical component, or alternatively implemented using multiple physical components. These physical components may include hardwired devices, programmable devices, or a combination of the two. In some embodiments, at least some of the functions of the processing circuit 22 may be executed by a programmable processor under the control of suitable software. This software may be downloaded to the device in electronic form, for example, via a network. Alternatively or additionally, this software may be stored in a tangible, non-temporary computer-readable medium such as optical memory, magnetic memory, or electronic memory.

[0039] Here, see Figure 2, a perspective view of the catheter 14 for use in the system 10 of Figure 1.

[0040] The catheter 14 comprises 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 attached to the distal end of the shaft 39.

[0041] The shaft 39 has an elongated tubular structure having a single axial or central lumen (not shown), but may optionally have multiple lumens. The shaft 39 is flexible, i.e., bendable, but substantially incompressible along its length. The shaft 39 may be of any preferred structure and may be made of any preferred material. In some embodiments, the elongated shaft 39 includes an outer wall made of polyurethane or polyether block amide. The outer wall has an embedded braided mesh, such as stainless steel, to increase the torsional rigidity of the shaft 39, so that when the control handle 20 is rotated, the distal end of the shaft 39 rotates in a corresponding manner.

[0042] The outer diameter of the shaft 39 is not critical, but may 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 so that the central lumen can accommodate any one or more of the puller wire, lead wire, sensor cable, and any other wire, cable, or tube. Optionally, the inner surface of the outer wall may be lined with a stiffening tube (not shown) to improve torsional stability. An example of a catheter body configuration suitable for use in connection with the present invention is described and illustrated in U.S. Patent No. 6,064,905.

[0043] The assembly 43 is mounted on the distal end of the shaft 39. As shown in Figure 2, the basket assembly 43 comprises five splines 45 or arms mounted generally at roughly equal intervals around a shrink wire 47, which is connected to the distal end of the assembly 43. Depending on the circumstances, applying a tensile or compressive force longitudinally to the shrink wire 47 causes the assembly 43 to shrink, retract, and expand. The shrink wire 47 forms the longitudinal axis of symmetry of the assembly 43. All splines 45 are directly or indirectly attached to the shrink wire 47 at their distal ends and to the shaft 39 at their proximal ends. When the shrink wire 47 is moved longitudinally to expand and contract the assembly 43, the splines 45 are curved outward in an arc shape in the expanded position and are generally straight in the contracted position. As will be understood by those skilled in the art, the number of splines 45 can be varied as needed depending on the specific application, so that the assembly 43 has at least two splines, generally at least three splines, and as many as ten or more splines. The expandable distal end basket assembly 43 is not limited to the illustrated configuration and may include other designs, such as a spherical or oval design, which includes multiple expandable arms connected directly or indirectly at the proximal and distal ends. In other embodiments, the basket assembly may be replaced by any suitable distal end assembly, such as a balloon assembly, a focal catheter assembly, a flat grid assembly, a multiple spline assembly, or a focal catheter assembly.

[0044] In some embodiments, the assembly 43 includes at least one sensing electrode 49 positioned thereon. Each spline 45 may comprise a non-conductive coated flexible wire on which one or more sensing electrodes 49 (e.g., ring spline electrodes) are mounted. The electrodes 49 are conveniently referred to as “sensing electrodes” but may be used to perform ablation. In some embodiments, each flexible wire includes a flat nitinol wire, and each non-conductive coating includes a biocompatible plastic tube, such as a polyurethane or polyimide tube. Alternatively, the spline 45 can be designed without an internal flexible wire, as long as the spline has a non-conductive outer surface on at least a portion of its surface for mounting the sensing electrodes 49, provided that a sufficiently rigid non-conductive material is used for the non-conductive coating to allow for expansion of the assembly 43. In some embodiments, the spline may be formed from a flexible polymer strip circuit, with the electrodes 49 positioned on the outer surface of each of the flexible polymer strip circuits.

[0045] Each of the sensing electrodes 49 on the spline 45 is electrically connected to a suitable mapping or monitoring system and / or ablation energy source by an electrode lead wire (not shown). The electrode lead wires pass through the control handle 20, through the lumen in the shaft 39, and extend into the non-conductive coating of the corresponding spline 45, and are attached to their corresponding sensing electrodes 49 by any preferred method. The catheter 14 optionally includes a far-field electrode 51, such as a cylindrical electrode, positioned on a retractable wire 47. The far-field electrode 51 is positioned in an expandable distal end basket assembly 43 to prevent the far-field electrode 51 from coming into contact with the tissue of the cardiac chambers of the heart 12. The function of the far-field electrode 51 is described below with reference to Figure 3. Further details of the catheter 14 are described in U.S. Patent No. 6,748,255, referenced above. The catheter 14 typically has multiple sensing electrodes 49 positioned on multiple flexible splines of the basket assembly 43. The catheter 14 is configured to be introduced into the cardiac chambers of a living heart 12 (Figure 1) in a collapsed shape, and the splines 45 are relatively close to each other. One or more of the sensing electrodes 49 are configured to be in contact with living tissue. Once inside the heart 12, the splines 45 hold their distal ends, and the distal ends of the splines 45 can be formed into their expanded basket shape by a shrinkable wire 47 that is pulled proximal.

[0046] Referring now to Figure 3, this figure is a detailed schematic diagram of the expandable distal end basket assembly 43 of Figure 2. In the expanded shape of the assembly 43, at least a portion of the sensing electrode 49 of the spline 45 is in contact with the endocardial surface 53 of the heart 12, and a signal corresponding to the electrode potential generated at the contact point with the surface is obtained. However, since the sensing electrode 49 is in a conductive medium (blood), in addition to the electrode potential from the contact point, the obtained signal also includes a far-field component from other areas of the heart 12.

[0047] The far-field component constitutes an interference signal at the endocardial surface electrode potential. To cancel out the interference, some embodiments position the far-field electrode 51 on the shrinkable wire 47. In the expanded configuration of the assembly 43, the far-field electrode 51 is positioned on the shrinkable wire 47 at approximately equidistant from all corresponding sensing electrodes 49, i.e., sensing electrodes 49 that are equidistant from fixed reference points on the long axis of the catheter, such as the reference point 55 at the proximal end of the assembly 43, and is shielded from contact with the cardiac surface by the spline 45. For example, electrodes 57 and 59 are equidistant from the reference point 55 and also equidistant from the far-field electrode 51, as indicated by the dashed lines 61 and 63, respectively. If the far-field electrode 51 is at least 0.5 cm away from the sensing electrodes 49 in the expanded configuration of the assembly 43, a far-field signal is obtained, but no near-field signal from the endocardial surface 53 is obtained. However, the signal e(t) obtained by the sensing electrodes 49 may have both far-field and surface (near-field) components. The far-field component signal x(t) obtained by the far-field electrode 51 may be removed from the signal e(t) obtained by the sensing electrode 49, i.e., by signal subtraction: e(t)-x(t), in order to cancel out the interference experienced by these electrodes. In addition, or alternatively, the removal of the far-field component may be achieved using any preferred method, such as the algorithm described in U.S. Patent Publication No. 2016 / 0175023 or U.S. Patent No. 9,554,718. In some embodiments, the far-field component of the signal captured by the sensing electrode 49 is not removed.

[0048] In some embodiments, the catheter 14 includes a distal position sensor 65 mounted at or near the location where the distal end of the spine is connected, and a proximal position sensor 67 mounted at or near the proximal end of the assembly 43, thereby determining the coordinates of the position sensor 65 relative to the coordinates of the position sensor 67 during use, and using this together with known information regarding the curvature of the spline 45 to find the respective positions of the sensing electrodes 49.

[0049] Next, refer to Figures 4 and 5. Figure 4 is a more detailed diagram of the processing circuit 22 in system 10 of Figure 1. Figure 5 is a flowchart 100 including the steps in the operation method of system 10 of Figure 1.

[0050] The processing circuit 22 includes an analog-to-digital converter 70, a digital signal filtering circuit 72, a synchronization circuit 74, a neural network training circuit 76, a noise compensation circuit 78, a digital-to-analog converter 80, and an analog-to-digital converter 82. In some embodiments, the analog-to-digital converter 70 and the analog-to-digital converter 82 may be implemented in a single multi-channel analog-to-digital converter.

[0051] The synchronization circuit 74, the neural network training circuit 76, and / or the noise compensation circuit 78 can be implemented using software running on a processor and / or using hardwired processing circuits. This software can be downloaded, for example, in electronic form to a computer or processor via a network. Alternatively or additionally, the software may be provided on a non-transient tangible medium such as an optical, magnetic, or electronic storage medium.

[0052] The neural network training module 76 is configured to train an artificial neural network 75, as will be described in more detail below with reference to Figures 6-8. The artificial neural network 75 may include an autoencoder 77, as will be described in more detail with reference to Figure 6.

[0053] The medical system 10 includes a recording device cable 84 extending to the recording device 86. The medical system 10 also includes a shielded cable 90.

[0054] One end of the recording device cable 84 is connected to the processing circuit 22 via a digital-to-analog converter 80. The other end of the recording device cable 84 is connected to the recording device 86. The recording device cable 84 includes individual insulated wires that are generally unshielded or not adequately shielded, and therefore pick up electrical noise 88 from the surrounding electrophysiological (EP) laboratory environment.

[0055] The recording device cable 84 generally includes separate insulated wires to carry the respective signals captured by the sensing electrodes 49 (Figure 3) and optionally the body surface electrodes 30 (Figure 1) from the processing circuit 22 to the recording device 86. The recording device cable 84 may also include one or more calibration wires to carry the respective calibration signals from the digital-to-analog converter 80 to the end of the recording device cable 84 adjacent to the recording device 86. The calibration wire(s) of the recording device cable 84 are electrically connected to the wires of the shielded cable 90 (the end of the shielded cable 90 closest to the recording device 86) (at the end of the recording device cable 84 connected to the recording device 86) and return to the processing circuit 22 via the analog-to-digital converter 82.

[0056] The artificial neural network 75 is trained based on data captured from a catheter, such as the catheter 14 shown in Figures 1-3 (block 102), which is inserted into the cardiac chambers of a living organism, and / or from body surface electrodes 30 (Figure 1) applied to the skin of the living organism, and from noise picked up by a recording device cable 84, as described in more detail below, from noise 88 in the EP laboratory. For example, one or more electrodes 49 (Figure 3) of the catheter 14 are in contact with the tissue of the cardiac chambers of the heart 12 (Figure 1) (e.g., the endocardial surface 53 (Figure 3)), providing a cardiac signal segment used to train the artificial neural network 75. To provide high-quality training data, the operator 16 generally ensures that there is good contact between the tissue of the living organism and the electrodes 49 (and / or body surface electrodes 30) that provide the cardiac signal segment.

[0057] The processing circuit 22 (Figure 1) is configured to receive cardiac signal segments (block 104) in response to electrical activity sensed by one or more electrodes 49 (and / or one or more body surface electrodes 30) in contact with biological tissue. The catheter 14 can provide signals from different electrodes 49 while in a given position within the cardiac chamber, and / or from one or more electrodes 49 while the catheter 14 is moving to different positions within the cardiac chamber. The cardiac signal segments may be provided from different cardiac chambers, and even from different biological sources.

[0058] The processing circuit 22 is configured to receive cardiac signal segments from the sensing electrode 49 of the catheter 14 and / or from the body surface electrode 30 via an analog-to-digital converter 70 and optionally a digital signal filtering circuit 72. The analog-to-digital converter 70 is configured to convert the received cardiac signal segments from analog format to digital format (block 106).

[0059] In some embodiments, the digital signal filtering circuit 72 is coupled to receive cardiac signal segments (in digital form) from one or more of the sensing electrodes 49 and / or body surface electrodes 30, and is configured to filter noise from the received signal or signal segment. The digital signal filtering device 72 may include various filtering circuits, for example, but not limited to, a low-pass filter for removing signals with frequencies higher than a threshold frequency (e.g., 60 Hz or 100 Hz), and / or a band-stop filter for removing signals with frequencies within a frequency range (e.g., 100-200 Hz). The cardiac signal may include frequencies similar to noise in the range of, for example, 50 Hz, and therefore, simply filtering out the 50 Hz component using a low-pass filter or band-stop filter may not yield acceptable results. Therefore, other filtering methods may be applied to remove noise associated with an external source without adversely affecting the cardiac signal. At least some of the functions of the digital signal filtering device 72 may optionally be performed by one or more computers or processors running software.

[0060] In some embodiments, the processing circuit 22 generally receives continuous signals from the sensing electrode 49 and the body surface electrode 30. When the neural network training circuit 76 is being trained, the training is performed using individual signal segments, as described in more detail below. Thus, the received signal(s) may be logically segmented using synchronization signals generated by the synchronization circuit 74 or using cross-correlations (within the neural network training circuit 76), as described in more detail below. In some embodiments, the signals may be physically segmented by the synchronization circuit 74. In other embodiments, the synchronization circuit 74 may segment the received cardiac signals by adding markers to the cardiac signals to identify the segments to be used as training data.

[0061] The processing circuit 22 is configured to introduce the received cardiac signal segment (received via the analog-to-digital converter 70 and optionally the digital signal filtering circuit 72) into the recording device cable 84 via a digital-to-analog converter 80, which is configured to convert the received cardiac signal segment from digital to analog format (block 108). Thus, the processing circuit 22 is configured to introduce the received cardiac signal segment in analog format into one or more calibration wires at one end of the recording device cable 84 (block 110), and to output the corresponding noise-added cardiac signal segment to the end of the shielded cable 90 closest to the recording device 86 in response to electrical noise acquired in the recording device cable 84. For example, noise is added to cardiac signal segment A to obtain cardiac signal segment A', and noise is added to cardiac signal segment B to obtain cardiac signal segment B'. The end of the shielded cable 90 closest to the processing circuit 22 is configured to output the noise-added cardiac signal segment to the analog-to-digital converter 82 of the processing circuit 22. The processing circuit 22 is also configured to introduce the cardiac signal segment to the neural network training circuit 76 via the synchronization circuit 74.

[0062] The analog-to-digital converter 82 is configured to receive a noise-added heart signal segment output from the shielded cable 90 (block 112) and convert the noise-added heart signal segment from analog to digital format (block 114).

[0063] The received cardiac signal and the noise-added cardiac signal segment are received by the neural network training circuit 76. The neural network training circuit 76 of the processing circuit 22 is configured to train an artificial neural network 75 (e.g., an autoencoder 77) (block 116) to at least partially compensate for the electrical noise added to the cardiac signal in the recording device cable 84, in accordance with the received cardiac signal segment in digital format and the corresponding noise-added cardiac signal segment output by the end of the shielded cable 90 (closest to the processing circuit 22), which is now in digital format. The process of block 116 will be described in more detail with reference to Figures 6 to 8.

[0064] The processing circuit 22 trains the artificial neural network 75 based on the use of the cardiac signal segments corresponding to the noise-added cardiac signal segments. When a continuous cardiac signal is input to the neural network training circuit 76 and the same continuous cardiac signal is input to the recording device cable 84 to generate a continuous noise-added cardiac signal, the corresponding segments of the signal need to be identified for use in the training process. As described above, the synchronization circuit 74 can generate a synchronization signal (e.g., including periodic pulses), which is introduced into its own calibration wire and shield cable 90 within the recording device cable 84, as well as transmitted to the neural network training circuit 76, so that the neural network training circuit 76 can identify the segments of the received cardiac signal and the noise-added cardiac signal based on the synchronization signal. Alternatively, the signal may be divided into separate physical segments by the synchronization circuit 74 for transmission to the neural network training circuit 76 and introduction into the recording device cable 84. Alternatively, the neural network training circuit 76 may apply a cross-correlation between the cardiac signal segment received from the digital signal filtering circuit 72 and the noise-enhanced cardiac signal segment received from the shielded cable 90 in order to accurately match the cardiac signal segment with the noise-enhanced cardiac signal segment.

[0065] In some embodiments, the artificial neural network 75 may include an ANN used for training and a different ANN used for generation (to provide noise-compensated cardiac signals). Parameters (e.g., weights of the artificial neural network 75) may be intermittently copied from the training ANN to the generation ANN. In other embodiments, a single ANN may be used. Training of ANN 75 may be performed during a medical procedure so that the accuracy of ANN 75 is improved and so that ANN 75 responds to new electrical noises generated in the EP room during the medical procedure.

[0066] Referring now to Figure 6, this figure is a schematic diagram of the artificial neural network 75 for use in system 10 of Figure 1.

[0067] 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 ​​indicate inhibitory connections. Inputs are modified by the weights and summed using linear combinations. An activation function can control the amplitude of the output. For example, the acceptable range of the output is usually 0 to 1, but it can also be -1 to 1.

[0068] These artificial networks can be used for predictive modeling, adaptive control, and applications, and can be trained via datasets. Self-learning derived from experience can occur within the network, allowing it to draw conclusions from complex and seemingly unrelated sets of information.

[0069] For completeness, a biological neural network consists of groups of chemically connected or functionally associated neurons. One neuron may be connected to many other neurons, and the total number of neurons and connections in the network can be wide-ranging. Connections called synapses are typically formed from axons to dendrites, but interdendritic synapses and other connections are also possible. Apart from electrical signaling, there are other forms of signaling resulting from the diffusion of neurotransmitters.

[0070] 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 of the 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, and are used to build software agents or autonomous robots (in computer and video games).

[0071] A neural network (NN), also known as an artificial neural network (ANN) or simulated neural network (SNN), is an interconnected group of natural or artificial neurons that uses mathematical or computational models based on an accessibility-theoretic approach to information processing. In most cases, an ANN is an adaptive system that modifies its structure based on external or internal information flowing through the network. More practically, neural networks are nonlinear statistical data modeling or decision-making tools. They can be used to model complex relationships between inputs and outputs and to find patterns in data.

[0072] In some embodiments, as shown in Figure 6, the artificial neural network 75 may include an autoencoder 77 including an encoder 92 and a decoder 94. In other embodiments, the artificial neural network 75 may include any suitable ANN. The artificial neural network 75 may also include software executed by the processing circuit 22 (Figure 4) and / or hardware modules configured to perform the functions of the artificial neural network 75.

[0073] The encoder 92 includes an input layer 96 where the input is received. The encoder then includes one or more hidden layers 97 that progressively compress the input into code 98. The decoder 94 includes one or more hidden layers 99 that progressively decompress the code 98 to an output layer 95 where the output of the autoencoder 77 is provided. The autoencoder 77 includes interlayer weights of the autoencoder 77. The autoencoder 77 manipulates the data received in the input layer 96 according to the values ​​of the various interlayer weights of the autoencoder 77.

[0074] The weights of the autoencoder 77 are updated during training so that the autoencoder 77 can perform the data manipulation task that it is trained to perform. In the example in Figure 6, the autoencoder 77 is trained to remove future noise from the heart signal, as will be explained in more detail with reference to Figures 7 and 8.

[0075] The number and width of layers within the autoencoder 77 may be configurable. As the number and width of layers increase, the autoencoder 77 can manipulate the data more precisely according to the task. However, a larger number of layers and wider layers generally require more training data, i.e., more training time, and training may not converge. For example, the input layer 96 may contain 400 neurons (e.g., to compress a batch of 400 samples). The encoder 92 may contain five layers compressed by a coefficient of 2 (e.g., 400, 200, 100, 50, 25). The decoder 94 may contain five layers decompressed by a coefficient of 2 (e.g., 25, 50, 100, 200, 400).

[0076] Next, refer to Figures 7 and 8. Figure 7 is a schematic diagram showing the training of the artificial neural network 75 in Figure 6. Figure 8 is a flowchart including the sub-steps of the process in block 116 in Figure 5.

[0077] The neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to train an artificial neural network 75 (e.g., an autoencoder 77) to at least partially compensate for electrical noise added to the heart signal within the cable 84 (Figure 4) in response to the received heart signal segment (Graph 152) and the corresponding noise-added heart signal segment (Graph 150).

[0078] Training the artificial neural network 75 is largely an iterative process. One method of training the artificial neural network 75 is described below. The processor 76 of the processing circuit 22 (Figure 4) is configured to iteratively adjust the parameters of the artificial neural network 75 (block 120) to reduce the difference between the output of the artificial neural network 75 and the desired output (e.g., a received heart signal segment).

[0079] The following describes the sub-processes of block 120.

[0080] The neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to input a noise-added cardiac signal segment (Graph 150) to the artificial neural network 75 (block 122, arrow 154). For example, the noise-added cardiac signal is input to the input layer 96 of the encoder 92. The neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to compare the output of the artificial neural network 75 (for example, the output of the decoder 94 of the autoencoder 77) with a desired output, i.e., the corresponding received cardiac signal segment (Graph 152) (block 124, arrow 156). For example, if there is a set of intracardiac signal segments A, B, and C output by the artificial neural network 75, and corresponding sets of received intracardiac signal segments A', B', and C', the processor 76 of the processing circuit 22 (Figure 4) compares A with A', B with B', and C with C'. The comparison is generally performed using a suitable loss function that calculates the overall difference between all outputs of the artificial neural network 75 and all desired outputs (e.g., all corresponding intracardiac signal segments (Graph 152)).

[0081] In the determination block 126, the neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to determine whether the difference between the output of the artificial neural network 75 and the desired output is sufficiently small. If the difference between the output of the artificial neural network 75 and the desired output is sufficiently small (branch 132), the neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to save the parameters (e.g., weights) of the artificial neural network 75 (e.g., autoencoder 77) (block 134) and / or send the parameters (e.g., weights) to a cloud processing server (not shown).

[0082] If the difference is not sufficiently small (Branch 128), the neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to correct the parameters (weights) of the artificial neural network 75 (e.g., autoencoder 77) (Block 130) to reduce the difference between the output of the artificial neural network 75 and the desired output of the artificial neural network 75. The difference to be minimized in the above embodiment is the overall difference between all outputs of the artificial neural network 75 and all desired outputs (e.g., all received intracardiac signal segments (Graph 152)). The neural network training circuit 76 of the processing circuit 22 (Figure 4) is configured to modify the parameters using any suitable optimization algorithm, for example, a gradient descent algorithm such as Adam optimization. The process in Blocks 122-126 is then repeated.

[0083] Next, refer to Figures 9 and 10. Figure 9 is a schematic diagram showing the processing of the captured cardiac signal 202 as processed by a trained artificial neural network 75. Figure 10 is a flowchart 250 that includes the steps in a method of processing the captured signal 202 in Figure 9 using the trained artificial neural network 75.

[0084] The catheter 200 is configured to be inserted into the heart chambers of a living organism. The catheter 200 includes one or more sensing electrodes 206 configured to contact the tissue of the living organism. The living organism may be the same organism into which the catheter 14 (Figure 1) was inserted and the artificial neural network 75 was trained accordingly, or it may be a different living organism.

[0085] The medical system 10 is configured to generate noise-compensated cardiac signals from cardiac signals 202 using a trained artificial neural network 75 (Figure 7), as will be described in more detail below.

[0086] Catheter 200 is an example of a catheter that provides a cardiac signal to which noise compensation is added. Any suitable catheter (e.g., balloon, basket, or focal catheter) may provide a cardiac signal which is later processed using a trained artificial neural network 75 to generate a noise-compensated cardiac signal. In some embodiments, catheter 14 can be used to provide a cardiac signal which is processed by the artificial neural network 75 to provide a noise-compensated cardiac signal. In other words, the same catheter used to train the artificial neural network 75 can provide a cardiac signal which is processed by the trained artificial neural network 75 to provide a noise-compensated cardiac signal. In some embodiments, the same catheter provides a cardiac signal to be processed and at the same time provides a cardiac signal segment which is used as training data to train the artificial neural network 75 to generate a noise-compensated cardiac signal from the trained artificial neural network 75 which is continuously trained with the training data.

[0087] Similarly, one or more cardiac signals provided by one or more body surface electrodes 30 (Figure 1) may be processed by an artificial neural network 75 to add noise compensation to the provided cardiac signals.

[0088] The catheter 200 is inserted into a cardiac chamber of the living body (block 252), and / or the body surface electrode 30 is applied to the skin surface of the living body.

[0089] The noise compensation circuit 78 of the processing circuit 22 (Figure 4) is configured to receive the cardiac signal 202 (block 254) in response to electrical activity sensed by the sensing electrode 206 (and / or body surface electrode 30) in contact with the tissue of the living body (for example, while the catheter 200 is inserted into the cardiac chamber of the living body). The cardiac signal 202 is received via an analog-to-digital converter 70 (Figure 4) configured to convert the cardiac signal 202 from analog to digital. The digital signal filtering circuit 72 is optionally configured to remove noise from the cardiac signal 202.

[0090] The noise compensation circuit 78 of the processing circuit 22 is configured to apply a trained artificial neural network 75 to the heart signal 202 (block 256) to generate a noise-compensated heart signal 210 that at least partially compensates for electrical noise that is not yet present in the heart signal 202 but is added to the heart signal 202 in the cable 84 (Figure 4).

[0091] In some embodiments, the trained artificial neural network includes a trained autoencoder 77. In these embodiments, the noise compensation circuit 78 of the processing circuit 22 is configured to apply the trained autoencoder 77 to the heart signal 202 to generate a noise-compensated heart signal 210 that at least partially compensates for electrical noise that is not yet in the heart signal 202 but is added to the heart signal 202 in the cable 84 (Figure 4).

[0092] The processing circuit 22 is configured to output the noise-compensated heart signal 210 to the recording device 86 (Figure 4) (block 258) via a digital-to-analog converter 80 (configured to convert the heart signal 202 into an analog-format noise-compensated heart signal 210) and a cable 84.

[0093] The terms “about” or “approximately” used herein with respect to any numerical value or range of numerical values ​​indicate a suitable tolerance for dimensions that enables a part or set of components to function in accordance with its intended purpose as described herein. More specifically, “about” or “approximately” may refer to a range of values ​​of ±20% of the listed values; for example, “about 90%” may refer to a range of values ​​from 72% to 108%.

[0094] Various features of the present invention are described in the context of separate embodiments for clarity, and these may be provided in combination in a single embodiment. Conversely, various features of the present invention described in the context of a single embodiment for brevity may be provided separately or in any preferred partial combination.

[0095] The embodiments described above are by reference only, and the present invention is not limited to those specifically illustrated and described in the above specification. Rather, the scope of the present invention includes both combinations and partial combinations of the various features described in the above specification, as well as variations and modifications thereof that a person skilled in the art would likely conceive upon reading the foregoing description, but which are not disclosed in the prior art.

[0096] [Implementation Method] (1) A method for analyzing a signal, Receiving a first cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with first biological tissue, The received first cardiac signal segment is introduced into a recording device cable extending to a recording device, wherein the cable outputs a corresponding noise-added cardiac signal segment in response to electrical noise acquired within the cable. Training an artificial neural network to at least partially compensate for electrical noise added to the cardiac signal within the cable in response to the received first cardiac signal segment and the corresponding noise-added cardiac signal segment, Receiving a second cardiac signal in response to electrical activity detected by a second sensing electrode in contact with second biological tissue, Applying the trained artificial neural network to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for the electrical noise that is not yet present in the second heart signal but is added to the second heart signal within the cable, The noise-compensated second heart signal is output to the recording device via the cable, Methods that include... (2) The training described above is The noise-added cardiac signal segment is input to the artificial neural network, The parameters of the artificial neural network are iteratively adjusted to reduce the difference between the output of the artificial neural network and the received first cardiac signal segment. The method according to Embodiment 1, including the method described above. (3) The method according to Embodiment 1, further comprising converting the first cardiac signal segment from digital to analog format, wherein the introduction comprises introducing the first cardiac signal segment in analog format into the cable, the method further comprises converting the noise-added cardiac signal segment into digital format, and the training comprises training the artificial neural network in response to the received first cardiac signal segment in digital format and the corresponding noise-added cardiac signal segment in digital format. (4) The method according to Embodiment 1, wherein the training includes training an autoencoder comprising an encoder and a decoder. (5) Inserting the first catheter, which includes the first sensing electrode, into the first cardiac chamber of the living body, Inserting the second catheter, which includes the second sensing electrode, into the second cardiac chamber of the living body, The method according to Embodiment 1, further comprising:

[0097] (6) The method according to embodiment 5, wherein the first catheter includes the second catheter. (7) The introduction includes introducing the received first cardiac signal segment to the first end of the recording device cable, The method further includes electrically connecting the first end of the shielded cable to the second end of the recording device cable, The method further includes the second end of the shielded cable outputting the noise-added heart signal segment, The method according to Embodiment 1, wherein the training includes training the artificial neural network in response to the received first cardiac signal segment and the corresponding noise-enhanced cardiac signal segment output by the second end of the shielded cable. (8) A software product including a non-temporary computer-readable medium on which program instructions are stored, wherein, when the instructions are read by a central processing unit (CPU), the CPU Receiving a first cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with first biological tissue, The received first cardiac signal segment is introduced into a recording device cable extending to a recording device, wherein the cable outputs a corresponding noise-added cardiac signal segment in response to electrical noise acquired within the cable. Training an artificial neural network to at least partially compensate for electrical noise added to the cardiac signal within the cable in response to the received first cardiac signal segment and the corresponding noise-added cardiac signal segment, Receiving a second cardiac signal in response to electrical activity detected by a second sensing electrode in contact with second biological tissue, Applying the trained artificial neural network to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for the electrical noise that is not yet present in the second heart signal but is added to the second heart signal within the cable, The noise-compensated second heart signal is output to the recording device via the cable, A software product that enables the following action. (9) A medical system, A first sensing electrode configured to contact the tissue of a first living organism, A second sensing electrode configured to contact the tissue of a second living organism, Recording device cable extending to the recording device, A processing circuit, wherein the processing circuit is In response to electrical activity sensed by the first sensing electrode in contact with the tissue, a first cardiac signal segment is received. The received first cardiac signal segment is introduced into the recording device cable, wherein the cable is configured to output a corresponding noise-added cardiac signal segment in response to electrical noise acquired within the cable. Training an artificial neural network to at least partially compensate for electrical noise added to the cardiac signal within the cable in response to the received first cardiac signal segment and the corresponding noise-added cardiac signal segment, Receiving a second cardiac signal in response to electrical activity sensed by the second sensing electrode in contact with the tissue of the second living organism, Applying the trained artificial neural network to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for the electrical noise that is not yet present in the second heart signal but is added to the second heart signal within the cable, A processing circuit is configured to output the noise-compensated second heart signal to the recording device via the cable, A medical system equipped with these features. (10) The processing circuit The noise-added cardiac signal segment is input to the artificial neural network. The parameters of the artificial neural network are iteratively adjusted to reduce the difference between the output of the artificial neural network and the received first cardiac signal segment. The system according to embodiment 9, configured as described above.

[0098] (11) The processing circuit A digital-to-analog converter configured to convert the first cardiac signal segment from a digital format to an analog format, wherein the processing circuit is configured to introduce the first cardiac signal segment in analog format into the cable, An analog-to-digital converter configured to convert the noise-added cardiac signal segment into a digital format, wherein the processing circuit is configured to train the artificial neural network to at least partially compensate for electrical noise added to the cardiac signal in the cable in response to the received first cardiac signal segment in digital format and the corresponding noise-added cardiac signal segment in digital format, The system according to embodiment 9, further comprising the above. (12) The system according to Embodiment 9, wherein the artificial neural network comprises an autoencoder including an encoder and a decoder, and the processing circuit is configured to train the autoencoder to at least partially compensate for electrical noise added to the heart signal in the cable in response to the received first heart signal segment and the corresponding noise-added heart signal segment. (13) The system according to embodiment 12, wherein the processing circuit is configured to apply the autoencoder to the second heart signal to generate the noise-compensated second heart signal. (14) A first catheter equipped with the first sensing electrode and configured to be inserted into the first cardiac chamber of a living organism, A second catheter comprising the second sensing electrode and configured to be inserted into the second cardiac chamber of a second living organism, The system according to embodiment 9, further comprising the above. (15) The system according to embodiment 14, wherein the first catheter includes the second catheter.

[0099] (16) Further comprising a shielded cable having a first end and a second end, The recording device cable has a first end electrically connected to the processing circuit and a second end electrically connected to the recording device and the first end of the shielded cable. The second end of the shielded cable is electrically connected to the second end of the recording device cable. The processing circuit is configured to introduce the received first heart signal segment to the first end of the recording device cable. The second end of the shielded cable is configured to output the noise-added heart signal segment. The system according to embodiment 9, wherein the processing circuit is configured to train the artificial neural network in response to the received first cardiac signal segment and the corresponding noise-enhanced cardiac signal segment output by the second end of the shielded cable.

Claims

1. A software product including a non-temporary computer-readable medium on which program instructions are stored, wherein when the program instructions are read by a central processing unit (CPU), the CPU... Receiving a first cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with first biological tissue, The received first cardiac signal segment is input to a noise compensation circuit configured to apply an artificial neural network to the input signal, and the artificial neural network is applied to the received first cardiac signal segment. The first cardiac signal segment to which the artificial neural network is applied is introduced to a first end of a recording device cable extending to a recording device, wherein the recording device cable outputs a corresponding noise-added cardiac signal segment via a second end of the recording device cable in response to electrical noise acquired within the recording device cable. The noise-added cardiac signal segment is input to the neural network training circuit via a shielded cable connected to the second end of the recording device cable, The received first cardiac signal segment is input to the neural network training circuit, and the artificial neural network is trained to at least partially compensate for electrical noise estimated to be added to the cardiac signal within the recording device cable by iteratively adjusting parameters based on the difference between the received first cardiac signal segment and the corresponding noise-added cardiac signal segment. Receiving a second cardiac signal in response to electrical activity detected by a second sensing electrode in contact with second biological tissue, The artificial neural network trained by the noise compensation circuit is applied to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for the electrical noise that is not yet present in the second heart signal but is estimated to be added to the second heart signal within the recording device cable, The noise-compensated second heart signal is output to the recording device via the recording device cable, A software product that enables the following action.

2. It is a medical system, A first sensing electrode configured to contact the tissue of a first living organism, A second sensing electrode configured to contact the tissue of a second living organism, Recording device cable extending to the recording device, A processing circuit comprising a noise compensation circuit configured to apply an artificial neural network to an input signal, and a neural network training circuit configured to train the artificial neural network, wherein the processing circuit is In response to electrical activity sensed by the first sensing electrode in contact with the tissue, a first cardiac signal segment is received. The received first cardiac signal segment is input to the noise compensation circuit, and the artificial neural network is applied to the received first cardiac signal segment. The method involves introducing a first cardiac signal segment to which the artificial neural network is applied to a first end of the recording device cable, wherein the recording device cable is configured to output a corresponding noise-added cardiac signal segment via a second end of the recording device cable in response to electrical noise acquired within the recording device cable. The noise-added cardiac signal segment is input to the neural network training circuit via a shielded cable connected to the second end of the recording device cable, The received first cardiac signal segment is input to the neural network training circuit, and the artificial neural network is trained to at least partially compensate for electrical noise estimated to be added to the cardiac signal within the recording device cable by iteratively adjusting parameters based on the difference between the received first cardiac signal segment and the corresponding noise-added cardiac signal segment. Receiving a second cardiac signal in response to electrical activity sensed by the second sensing electrode in contact with the tissue of the second living organism, The artificial neural network trained by the noise compensation circuit is applied to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for the electrical noise that is not yet present in the second heart signal but is estimated to be added to the second heart signal within the recording device cable, A processing circuit is configured to output the noise-compensated second heart signal to the recording device via the recording device cable, A medical system equipped with these features.

3. The processing circuit described above A digital-to-analog converter configured to convert a first cardiac signal segment to which the artificial neural network is applied from a digital format to an analog format, wherein the processing circuit is configured to introduce the first cardiac signal segment to which the artificial neural network is applied in analog format into the recording device cable, An analog-to-digital converter configured to convert the noise-added cardiac signal segment into a digital format, wherein the processing circuit is configured to input the received first cardiac signal segment in digital format to the neural network training circuit and train the artificial neural network to at least partially compensate for electrical noise estimated to be added to the cardiac signal within the recording device cable by iteratively adjusting parameters based on the difference between the received first cardiac signal segment in digital format and the corresponding noise-added cardiac signal segment in digital format, The medical system according to claim 2, further comprising the above.

4. The medical system according to claim 2, wherein the artificial neural network comprises an autoencoder including an encoder and a decoder, and the processing circuit is configured to input the received first cardiac signal segment to the neural network training circuit and train the autoencoder to at least partially compensate for electrical noise estimated to be added to the cardiac signal in the recording device cable by iteratively adjusting parameters based on the difference between the received first cardiac signal segment and the corresponding noise-added cardiac signal segment.

5. The medical system according to claim 4, wherein the processing circuit is configured to apply the autoencoder to the second heart signal to generate the noise-compensated second heart signal.

6. A first catheter equipped with the first sensing electrode and configured to be inserted into the first cardiac chamber of a living organism, A second catheter comprising the second sensing electrode and configured to be inserted into the second cardiac chamber of the second living organism, The medical system according to claim 2, further comprising the above.

7. The medical system according to claim 6, wherein the first catheter includes the second catheter.

8. It is a program that analyzes signals, On the computer, Receiving a first cardiac signal segment in response to electrical activity sensed by a first sensing electrode in contact with first biological tissue, The received first cardiac signal segment is input to a noise compensation circuit configured to apply an artificial neural network to the input signal, and the artificial neural network is applied to the received first cardiac signal segment. The first cardiac signal segment to which the artificial neural network is applied is introduced to a first end of a recording device cable extending to a recording device, wherein the recording device cable outputs a corresponding noise-added cardiac signal segment via a second end of the recording device cable in response to electrical noise acquired within the recording device cable. The noise-added cardiac signal segment is input to the neural network training circuit via a shielded cable connected to the second end of the recording device cable, The received first cardiac signal segment is input to the neural network training circuit, and the artificial neural network is trained to at least partially compensate for electrical noise estimated to be added to the cardiac signal within the recording device cable by iteratively adjusting parameters based on the difference between the received first cardiac signal segment and the corresponding noise-added cardiac signal segment. Receiving a second cardiac signal in response to electrical activity detected by a second sensing electrode in contact with second biological tissue, The artificial neural network trained by the noise compensation circuit is applied to the second heart signal to generate a noise-compensated second heart signal that at least partially compensates for the electrical noise that is not yet present in the second heart signal but is estimated to be added to the second heart signal within the recording device cable, The noise-compensated second heart signal is output to the recording device via the recording device cable, A program that causes something to happen.

9. The program according to claim 8, wherein the program further causes the computer to convert the first cardiac signal segment to which the artificial neural network is applied from a digital format to an analog format, the introduction includes introducing the first cardiac signal segment to which the artificial neural network is applied in analog format into the recording device cable, the program further causes the computer to convert the noise-added cardiac signal segment to a digital format, and the training includes inputting the received first cardiac signal segment in digital format into the neural network training circuit and training the artificial neural network by iteratively adjusting parameters based on the difference between the received first cardiac signal segment in digital format and the corresponding noise-added cardiac signal segment in digital format.

10. The program according to claim 8, wherein the training includes training an autoencoder comprising an encoder and a decoder.

11. Inserting the first catheter, which includes the first sensing electrode, into the first cardiac chamber of the living body, Inserting the second catheter, which includes the second sensing electrode, into the second cardiac chamber of the living body, The program according to claim 8, further comprising:

12. The program according to claim 11, wherein the first catheter includes the second catheter.

Images