atrial fibrillation

Abstract

This invention is entitled "Atrial Fibrillation". It provides a method. The method is implemented by a detection engine embodied in processor-executable code stored in memory and executed by at least one processor. The method includes modeling a vector velocity field that measures and quantifies the velocity of an electrocardiogram (ECG) data signal over a local activation time. The method further includes: determining a code for each point in a plane to provide a color-coded vector field image; detecting lesion indications and rotor indications by scanning the color-coded vector field image using a kernel; and classifying the lesion indications and rotor indications into maintenance foci.

Description

[0001] Priority Statement

[0002] This patent application claims priority to U.S. Provisional Patent Application Serial No. 63 / 039,291 entitled “ATRIAL FIBRILLATION”, which is incorporated herein by reference as if fully illustrated in this patent application for all purposes. Technical Field

[0003] This invention relates to artificial intelligence and machine learning in relation to atrial fibrillation. More specifically, this invention relates to systems and methods for implementing machine learning / artificial intelligence algorithms for detecting and terminating atrial fibrillation. Background Technology

[0004] Atrial fibrillation (“aFib”) is a tremor or irregular heartbeat (arrhythmia) that can lead to blood clots, stroke, heart failure, and other heart-related complications. aFib is the most common arrhythmia diagnosed in clinical practice. The prevalence of aFib in the United States is estimated to be between approximately 2.7 million and 6.1 million, and is projected to rise to 12.1 million by 2030. According to a 2013 study, global estimates put the number of individuals with aFib at approximately 33.5 million in 2010. This represents approximately 0.5% of the world's population.

[0005] Currently, physicians cannot successfully determine whether and when a specific aFib case has been terminated based on a set of electrocardiogram (ECG) signals. For example, conventional mechanisms discuss phase mapping (i.e., calculating the delay between electrodes and finding phase singularities) and vector analysis without using velocity vector fields, machine learning, deep learning, and / or other sophisticated algorithms. At best, conventional institutions can use the velocity characteristics of the aFib source determined by electrocardiographic current mapping before and after catheter ablation; however, these conventional institutions still cannot distinguish between active and passive lesion sources. Consequently, conventional mechanisms cannot determine aFib termination. Summary of the Invention

[0006] According to one exemplary embodiment, a method is provided. The method is implemented by a detection engine embodied in processor-executable code stored in memory and executed by at least one processor. The method includes modeling a vector velocity field that measures and quantifies the velocity of an electrocardiogram data signal over a local activation time. The method further includes: determining one or more codes for each point in a plane to provide a color-coded vector field image; detecting lesion indications and trochanter indications by scanning the color-coded vector field image using one or more kernels; and classifying the lesion indications and trochanter indications as maintenance foci.

[0007] According to one or more embodiments, the above exemplary method embodiments can be implemented as apparatus, system and / or computer program products. Attached Figure Description

[0008] A more detailed understanding can be obtained through the following specific embodiments provided by way of example and in conjunction with the accompanying drawings, wherein similar reference numerals in the drawings indicate similar elements, and wherein:

[0009] Figure 1 A diagram illustrating an exemplary system that can implement one or more features of the subject matter of this disclosure.

[0010] Figure 2 A block diagram of an exemplary system for anatomically correct reconstruction of the atria according to one or more embodiments is shown;

[0011] Figure 3 A method according to one or more implementation schemes is shown;

[0012] Figure 4 A graphical depiction of an artificial intelligence system according to one or more exemplary embodiments is shown;

[0013] Figure 5 An example block diagram of a neural network according to one or more implementation schemes and methods executed in the neural network is shown;

[0014] Figure 6 A method according to one or more implementation schemes is shown;

[0015] Figure 7 A graph is shown according to one or more implementation schemes;

[0016] Figure 8 A graph of the surface (x, y) according to one or more embodiments is shown; and

[0017] Figure 9 The following are illustrated according to one or more implementation schemes. Figure 8 A curve of the velocity vector field of the surface (x,y). Detailed Implementation

[0018] This document discloses machine learning and / or artificial intelligence methods and systems implemented by a detection engine. More specifically, the present invention relates to a detection engine including machine learning / artificial intelligence algorithms for detecting aFib and aFib termination.

[0019] One or more advantages, technical effects, and / or benefits of the detection engine may include addressing active versus passive issues (e.g., distinguishing between lesion sources and rotors) using feedback loops based on actual ablation activity, which is not available through conventional mechanisms. In this regard, the detection engine can model vector velocity fields while simultaneously distinguishing between lesion sources and rotors (e.g., active and passive). That is, the detection engine can define the location for calculating the Local Activation Time (LAT), calculate the derivative of the LAT to obtain the velocity, identify lesion sources (e.g., within a 1 mm box), classify active or passive lesion sources (e.g., within a 1 mm box), provide retrospective analysis (e.g., investigating past cases and determining where ablation occurred), and provide prospective analysis (e.g., determining where to ablate).

[0020] For ease of explanation, this paper describes the detection engine in relation to the identification and treatment of the heart's aFib; however, any anatomical structure, body part, organ, or portion thereof may be a target for mapping using the detection engine described herein. Furthermore, the detection engine and / or machine learning / artificial intelligence algorithms are processor-executable code or software that must be derived from the processing operations of the medical device equipment and its processing hardware.

[0021] According to one or more embodiments, a method is provided implemented by a detection engine embodied in processor-executable code stored in memory and executed by at least one processor. The method includes: modeling a vector velocity field by the detection engine, the vector velocity field measuring and quantifying the velocity of an electrocardiogram data signal after a local activation time; determining one or more codes for each point in a plane by the detection engine to provide a color-coded vector field image; detecting lesion indications and rotor indications by the detection engine using one or more kernels to scan the color-coded vector field image; and classifying the lesion indications and rotor indications into maintenance lesions by the detection engine.

[0022] According to one or more embodiments or any of the method embodiments described herein, the electrocardiogram data signal may be detected by a catheter within an anatomical structure and in communication with the detection engine.

[0023] According to one or more embodiments or any of the method embodiments described herein, the detection engine can detect one or more segments of the local activation time relative to the first activation time.

[0024] According to one or more embodiments or any of the method embodiments described herein, the detection engine can model the velocity vector field by calculating the direction of the radio wave at each x,y point and providing the velocity vector field using the derivative of the polynomial surface.

[0025] According to one or more embodiments or any of the method embodiments described herein, the detection engine can classify the maintenance foci by utilizing the velocity vector field as input to a machine learning or artificial intelligence algorithm.

[0026] According to one or more embodiments or any of the method embodiments described herein, the machine learning or artificial intelligence algorithm may include a deep convolutional neural network or a recurrent neural network to detect the location of the gold standard maintenance foci in the maintenance foci.

[0027] According to one or more embodiments or any of the method embodiments described herein, the machine learning or artificial intelligence algorithm can determine whether the ablation result for a specific case is successful regarding the electrocardiogram data signal.

[0028] According to one or more embodiments or any of the method embodiments described herein, the lesion indication and the rotor indication can be detected when all directions are sequential within the one or more cores.

[0029] According to one or more embodiments or any of the method embodiments described herein, when the maintenance focus is labeled, the detection engine automatically identifies and annotates the atrial fibrillation maintenance focus based on vector velocity and ablation information.

[0030] According to one or more embodiments or any of the method embodiments described herein, the detection engine may use region of interest annotation relative to active, passive, and unknown categories to at least indicate whether to ablate the region of interest or terminate atrial fibrillation.

[0031] According to one or more embodiments, the system includes a memory storing processor-executable code for a detection engine. The system also includes at least one processor that executes the processor-executable code such that the system: models a vector velocity field by the detection engine, the vector velocity field measuring and quantifying the velocity of an electrocardiogram data signal after a local activation time; determines one or more codes for each point in a plane by the detection engine to provide a color-coded vector field image; detects lesion indications and rotor indications by the detection engine using one or more kernels to scan the color-coded vector field image; and classifies the lesion indications and rotor indications into maintenance lesions by the detection engine.

[0032] According to one or more embodiments or any of the system implementations herein, the electrocardiogram data signal may be detected by a catheter within the anatomical structure and in communication with the detection engine.

[0033] According to one or more embodiments or any of the system embodiments described herein, the detection engine can detect one or more segments of the local activation time relative to the first activation time.

[0034] According to one or more embodiments or any of the system embodiments described herein, the detection engine can model the velocity vector field by calculating the direction of the radio wave at each x,y point and providing the velocity vector field using the derivative of the polynomial surface.

[0035] According to one or more embodiments or any of the system implementations herein, the detection engine can classify the maintenance foci by utilizing the velocity vector field as input to a machine learning or artificial intelligence algorithm.

[0036] According to one or more embodiments or any of the system implementations herein, the machine learning or artificial intelligence algorithm may include a deep convolutional neural network or a recurrent neural network to detect the location of the gold standard maintenance foci in the maintenance foci.

[0037] According to one or more embodiments or any of the system implementations herein, the machine learning or artificial intelligence algorithm can determine whether the ablation result for a specific case is successful regarding the electrocardiogram data signal.

[0038] According to one or more embodiments or any of the system embodiments described herein, the lesion indication and the rotor indication can be detected when all directions are sequential within the one or more cores.

[0039] According to one or more embodiments or any of the system implementations herein, when the maintenance focus is labeled, the detection engine automatically identifies and annotates the atrial fibrillation maintenance focus based on vector velocity and ablation information.

[0040] According to one or more embodiments or any of the system implementations herein, the detection engine may use region of interest annotation relative to active, passive, and unknown categories to at least indicate whether to ablate the region of interest or terminate atrial fibrillation.

[0041] Figure 1This is a diagram of an exemplary system (e.g., a medical device apparatus) shown as system 100, wherein one or more features of the subject matter herein may be implemented according to one or more embodiments. All or part of system 100 may be used to collect information (e.g., biometric data and / or training datasets) and / or to implement detection engine 101 (e.g., machine learning and / or artificial intelligence algorithms), as described herein. Detection engine 101 may be defined as a deep learning optimization for detecting maintenance foci of aFib to be ablated for treating subjects with persistent aFib and classifying velocity vector field images and raw data into maintenance foci.

[0042] As shown in the figures, system 100 includes a probe 105 having a catheter 110 (including at least one electrode 111), a shaft 112, a sheath 113, and a manipulator 114. As also shown, system 100 includes a physician 115 (or a medical professional or clinician), a heart 120, a patient 125, and a bed 130 (or table). Note that illustrations 140 and 150 show the heart 120 and catheter 110 in more detail. As also shown, system 100 includes a console 160 (including one or more processors 161 and memory 162) and a display 165. It should also be noted that each element and / or item of system 100 represents one or more of that element and / or item. Figure 1 The example of system 100 shown can be modified to implement the embodiments disclosed herein. The embodiments disclosed herein can be applied similarly using other system components and setups. Additionally, system 100 may include additional components such as elements for sensing electrical activity, wired or wireless connectors, processing and display devices, etc.

[0043] System 100 can be used to detect, diagnose, and / or treat cardiac conditions (e.g., using monitoring engine 101). Cardiac conditions such as arrhythmias have always been common and dangerous medical conditions, especially among the elderly. Additionally, System 100 can be used with surgical systems (e.g., those sold by Biosense Webster). As part of a system, this surgical system is configured to acquire biometric data (e.g., anatomical and electrical measurements of a patient's organ, such as heart 120) and to perform cardiac ablation procedures. More specifically, treatment of cardiac conditions such as arrhythmias often requires detailed mapping of cardiac tissue, chambers, veins, arteries, and / or electrical pathways. For example, a prerequisite for successful catheter ablation (as described herein) is that the cause of the arrhythmia is accurately located within a chamber of heart 120. Such localization can be accomplished via an electrophysiological study, during which a mapping catheter (e.g., catheter 110) introduced into the chamber of heart 120 is used to spatially resolve potentials. This electrophysiological study (so-called electroanatomical mapping) thus provides 3D mapping data that can be displayed on a monitor. In many cases, mapping and therapeutic functions (e.g., ablation) are provided by a single catheter or a group of catheters, such that the mapping catheter also operates simultaneously as a therapeutic (e.g., ablation) catheter. In this case, the detection engine 101 can be directly stored and executed by catheter 110.

[0044] In patients with normal sinus rhythm (NSR) (e.g., patient 125), the heart, including the atria, ventricles, and excitatory conduction tissues (e.g., heart 120), is electrically stimulated to beat in a synchronized, patterned manner. It should be noted that this electrical stimulation can be detected as intracardiac electrocardiogram (ICECG) data, etc.

[0045] Generally, the heart consists of four chambers—two upper chambers (atria) and two lower chambers (ventricles). The coronary sinuses (CS) are a collection of large vessels that join together to form the heart muscle, collecting blood and delivering less oxygenated blood to the right atrium. The rhythm of the heart is normally controlled by the sinoatrial node (not shown), located in the right atrium. The sinoatrial node generates an electrical impulse that typically begins each heartbeat and acts as a natural pacemaker. The electrical impulse travels from the sinoatrial node through the atrium, causing the atrial muscles to contract and pump blood into the ventricles. The electrical impulse then reaches a cluster of cells called the atrioventricular node (AV node) (not shown). The AV node is usually the only path for the signal to travel from the atrium to the ventricle. The AV node slows down the electrical signal before sending it to the ventricle. This delay (even slightly) allows the ventricle to fill with blood. When the electrical impulse reaches the ventricular muscles, the muscles contract, causing the muscles to pump blood to the lungs or the rest of the body. In a healthy heart, this process is typically smooth, resulting in a normal resting heart rate of 60 to 100 beats per minute. In a heart exhibiting one of the aforementioned disease states, erroneous electrical connections or abnormal areas of electrical activity in the heart trigger and maintain an abnormal rhythm. When this occurs, the heart rate accelerates too rapidly and does not allow the heart enough time to inflate before it contracts again. These ineffective contractions of the heart can cause mild dizziness or lightheadedness because the brain may not receive enough blood and oxygen.

[0046] In patients with arrhythmias (e.g., atrial fibrillation or aFib) (e.g., patient 125), abnormal areas of cardiac tissue do not follow the synchronous beating cycle associated with normal conduction tissue, in contrast to patients with NSR. Instead, the abnormal areas of cardiac tissue conduct abnormally to adjacent tissues, thus disrupting the cardiac cycle into an asynchronous rhythm. It should be noted that this asynchronous rhythm can also be detected in IC ECG data. Such aberrant conduction is previously known to occur in various regions of the heart 120, such as the sinoatrial (SA) node region, the conduction pathway along the atrioventricular (AV) node, or in the myocardial tissue that forms the walls of the ventricles and atria. Other conditions exist (such as fibrillation) in which the pattern of abnormal conduction tissue leads to a reentrant path, causing the chambers to beat in a regular pattern, which can be multiples of the sinus rhythm.

[0047] To support the detection, diagnosis, and / or treatment of cardiac conditions by system 100, probe 105 can be guided by physician 115 to the heart 120 of patient 125 lying on bed 130. For example, physician 115 can insert shaft 112 through sheath 113 while manipulating the distal end of shaft 112 using manipulator 114 near the proximal end of catheter 110 and / or deflecting from sheath 113. As shown in illustration 140, basket catheter 110 can be fitted at the distal end of shaft 112. Basket catheter 110 can be inserted through sheath 113 in a collapsed state and then deployed within heart 120.

[0048] Typically, a catheter 110 containing an electrical sensor (e.g., at least one electrode 111) at or near its distal tip is advanced to a point in the heart 120, the sensor contacts the tissue, and data is acquired at that point, thereby measuring the electrical activity at that point in the heart 120. A disadvantage of using a catheter containing only a single distal tip electrode to map the heart chambers is the long time required to acquire data point-by-point at the necessary number of points for a detailed map of the overall chambers. Therefore, multi-electrode catheters (e.g., catheter 110) have been developed to simultaneously measure electrical activity at multiple points in the heart chambers.

[0049] A catheter 110, which may include at least one electrode 111 and a catheter needle coupled to its body, may be configured to acquire biometric data, such as electrical signals from an in vivo organ (e.g., heart 120), and / or ablate a tissue region thereof (e.g., a ventricle of heart 120). It should be noted that electrode 111 refers to any similar element, such as a tracking coil, piezoelectric transducer, electrode, or combination of elements configured to ablate a tissue region or acquire biometric data. According to one or more embodiments, catheter 110 may include one or more position sensors for determining trajectory information. Trajectory information can be used to infer motion characteristics, such as the contractility of the tissue.

[0050] Biometric data (e.g., patient biometrics, patient data, or patient biometric data) may include one or more of the following: Local Activation Time (LAT), electrical activity, topology, bipolar mapping, reference activity, ventricular activity, dominant frequency, impedance, etc. LAT can be a time point corresponding to a threshold activity of local activation, calculated based on a normalized initial starting point. Electrical activity can be any applicable electrical signal that can be measured based on one or more thresholds and can be sensed and / or amplified based on signal-to-noise ratio and / or other filters. Topology can correspond to the physical structure of a body part or a portion of a body part, and can correspond to variations in the physical structure relative to different parts of the body part or different body parts. Dominant frequency can be a frequency or frequency range that is prevalent in a part of a body part and can differ in different parts of the same body part. For example, the dominant frequency of the PV of the heart can be different from the dominant frequency of the right atrium of the same heart. Impedance can be a resistance measurement at a given region of a body part.

[0051] Examples of biometric data include, but are not limited to, patient identification data, intracardiac ECG data, bipolar intracardiac reference signals, anatomical and electrical measurements, trajectory information, body surface (BS) ECG data, historical data, brain biometrics, blood pressure data, ultrasound signals, radio signals, audio signals, two-dimensional or three-dimensional image data, blood glucose data, and temperature data. Biometric data can generally be used to monitor, diagnose, and treat any number of various diseases, such as cardiovascular diseases (e.g., arrhythmias, cardiomyopathy, and coronary artery disease) and autoimmune diseases (e.g., type I and type II diabetes). It should be noted that BS ECG data may include data and signals collected from electrodes on the patient's surface, BS ECG data may include data and signals collected from electrodes inside the patient's body, and ablation data may include data and signals collected from ablated tissue. Additionally, BS ECG data, BS ECG data, and ablation data, along with catheter electrode location data, may be derived from one or more protocol records.

[0052] For example, catheter 110 may be used with electrodes 111 to achieve intravascular ultrasound and / or MRI catheter insertion for imaging of the heart 120 (e.g., acquiring and processing biometric data). Illustration 150 shows catheter 110 in a magnified view within a cardiac chamber of heart 120. Although catheter 110 is shown as a pointed catheter, it should be understood that any shape including one or more electrodes 111 may be used to implement the exemplary embodiments disclosed herein.

[0053] Examples of catheter 110 include, but are not limited to, linear catheters with multiple electrodes, balloon catheters including electrodes distributed on multiple ridges shaping the balloon, lasso or loop catheters with multiple electrodes, contact force sensing catheters, or any other suitable shape or type. The linear catheter may be fully or partially elastic, allowing it to twist, bend, and / or otherwise change its shape based on received signals and / or based on external forces (e.g., cardiac tissue) applied to the linear catheter. The balloon catheter may be designed such that its electrodes remain in close contact with the endocardial surface when deployed into a patient. For example, the balloon catheter may be inserted into a lumen, such as a PV. The balloon catheter may be inserted into the PV in a contracted state such that the balloon catheter does not occupy its maximum volume when inserted into the PV. The balloon catheter may inflate within the PV such that those electrodes on the balloon catheter contact the entire circular segment of the PV. Such contact with the entire circular segment of the PV or any other lumen enables effective imaging and / or ablation.

[0054] According to other examples, body patches and / or surface electrodes may also be positioned on or near the body of patient 125. A catheter 110 having one or more electrodes 111 may be positioned within the body (e.g., within the heart 120), and the location of the catheter 110 may be determined by system 100 based on signals transmitted and received between the one or more electrodes 111 of the catheter 110 and the body patch and / or surface electrodes. Additionally, the electrodes 111 may sense biometric data from within patient 125, such as within the heart 120 (e.g., the electrodes 111 sense the electrical potential of tissue in real time). The biometric data may be correlated with the determined location of the catheter 110, enabling the display of a rendering of the patient's body part (e.g., the heart 120) and the display of biometric data overlaid on the shape of the body part.

[0055] Probe 105 and other items of system 100 can be connected to console 160. Console 160 may include any computing device employing machine learning and / or artificial intelligence algorithms (represented as detection engine 101). According to an exemplary embodiment, console 160 includes one or more processors 161 (any computing hardware) and memory 162 (any non-transitory tangible medium), wherein the one or more processors 161 execute computer instructions relative to detection engine 101, and memory 162 stores these instructions for execution by the one or more processors 161. For example, console 160 may be configured to receive and process biometric data and determine whether a given tissue region is conductive.

[0056] In some implementations, console 160 may also be programmed by detection engine 101 (in software) to perform the following functions: modeling a vector velocity field that measures and quantifies the velocity of ECG data signals after local activation time; determining one or more codes for each point in the plane to provide a color-coded vector field image; detecting lesion indications and rotor indications by scanning the color-coded vector field image using one or more kernels (e.g., space within a mapping); and classifying the lesion indications and rotor indications into maintenance lesions. For example, detection engine 101 may include deep learning optimizations (as opposed to...) Figure 3 and Figure 6 The deep learning optimization receives biometric data acquired by the catheter 110 while it is manipulated within an anatomical structure. Once a mapping is generated, the detection engine 101 can receive user-modified input representing the mapping, such as through an existing user interface and / or a dedicated user interface for the detection engine 101. Generally, the detection engine 101 may provide one or more user interfaces, such as representing an operating system or other application and / or provided directly as needed. User interfaces include, but are not limited to, internet browsers, graphical user interfaces (GUIs), window interfaces, and / or other visual interfaces for applications, operating systems, folders, etc. According to one or more embodiments, the detection engine 101 may be located outside the console 160 and may be located, for example, within the catheter 110, in an external device, in a mobile device, in a cloud-based device, or may be a standalone processor. In this regard, the detection engine 101 may be transmitted / downloaded electronically via a network.

[0057] In one example, console 160 may be any computing device (such as a general-purpose computer) as described herein, including software (e.g., detection engine 101) and / or hardware (e.g., processor 161 and memory 162), having suitable front-end and interface circuitry for transmitting and receiving signals to and from probe 105, and for other components of the control system 100. For example, the front-end and interface circuitry may include an input / output (I / O) communication interface that enables console 160 to receive signals from and / or transmit signals to at least one electrode 111. Console 160 may include a real-time noise reduction circuitry system typically configured as a field-programmable gate array (FPGA), followed by an analog-to-digital (A / D) ECG or electrocardiograph or electromyography (EMG) signal conversion integrated circuit. Console 160 may pass signals from A / D or EMG circuitry to another processor and / or may be programmed to perform one or more functions disclosed herein.

[0058] Display 165 may be any electronic device used for the visual presentation of biometric data, and is connected to console 160. According to an exemplary embodiment, during a procedure, console 160 may facilitate the presentation of body part renderings to physician 115 on display 165, and the storage of data representing the body part renderings in memory 162. For example, a mapping depicting motion characteristics may be rendered / constructed based on trajectory information sampled at a sufficient number of points in the heart 120. As an example, display 165 may include a touchscreen that can be configured to accept input from medical professional 115 in addition to presenting body part renderings.

[0059] In some implementations, physician 115 may use one or more input devices (such as a touchpad, mouse, keyboard, gesture recognition device, etc.) to manipulate the rendering of components and / or body parts of system 100. For example, the input device may be used to change the position of catheter 110, causing the rendering to be updated. It should be noted that display 165 may be located in the same location or in a remote location, such as in a separate hospital or a separate healthcare provider network.

[0060] According to one or more embodiments, system 100 may also use ultrasound, computed tomography (CT), MRI, or other medical imaging techniques utilizing catheter 110 or other medical devices to acquire biometric data. For example, system 100 may use one or more catheters 110 or other sensors to acquire ECG data and / or anatomical and electrical measurements (e.g., biometric data) of heart 120. More specifically, console 160 may be connected via cable to BS electrodes comprising an adhesive skin patch attached to patient 125. BS electrodes may acquire / generate biometric data in the form of BS ECG data. For example, processor 161 may determine the positional coordinates of catheter 110 within a body part (e.g., heart 120) of patient 125. Positional coordinates may be based on impedance or electromagnetic field measured between the surface electrode and electrode 111 or other electromagnetic components of catheter 110. Additionally or alternatively, a positioning pad that generates a magnetic field for navigation may be located on the surface of bed 130 and may be detachable from bed 130. Biometric data may be transmitted to console 160 and stored in memory 162. Alternatively or otherwise, network 1762 can be used to transmit biometric data to server 1760, which may be local or remote.

[0061] According to one or more exemplary embodiments, catheter 110 may be configured to ablate a tissue region of a cardiac chamber of heart 120. Illustration 150 shows catheter 110 in an enlarged view within a cardiac chamber of heart 120. For example, ablation electrodes such as at least one electrode 111 may be configured to deliver energy to a tissue region of an organ (e.g., heart 120) in vivo. The energy may be thermal and may cause damage to the tissue region by starting from the surface of the tissue region and extending into the thickness of the tissue region. Biometric data relative to the ablation protocol (e.g., ablated tissue, ablation location, etc.) may be considered ablation data.

[0062] According to one example, a multi-electrode catheter (e.g., catheter 110) can be advanced into the chamber of heart 120 relative to the acquisition of biometric data. Anterior-posterior (AP) and lateral fluorescein maps can be obtained to establish the position and orientation of each electrode. ECG can be recorded relative to a time reference (such as the P wave in a sinus rhythm from a BS ECG and / or the start of a signal from the electrodes 111 of catheter 110 placed in the coronary sinus) from each of the electrodes 111 in contact with the cardiac surface. As further disclosed herein, the system can distinguish which electrodes record electrical activity from those that do not record electrical activity due to their lack of close proximity to the endocardial wall. After recording the initial ECG, the catheter can be repositioned, and fluorescein maps and ECG can be recorded again. An electrical mapping (e.g., via cardiac mapping) can then be constructed iteratively according to the above process.

[0063] Cardiac mapping can be achieved using one or more techniques. Generally, mapping of cardiac regions, such as the cardiac area of ​​heart 120, tissues, veins, arteries, and / or electrical pathways, can lead to the identification of problem areas such as scar tissue, sources of arrhythmia (e.g., telopolarization), healthy areas, etc. Cardiac regions can be mapped so that a visual rendering of the mapped cardiac region can be provided using a display, as further disclosed herein. Additionally, cardiac mapping (which is an example of cardiac imaging) can include mapping based on one or more modalities, such as, but not limited to, LAT, local activation velocity, electrical activity, topology, bipolar mapping, dominant frequency, or impedance. A catheter inserted into the patient's body (e.g., catheter 110) can be used to capture data corresponding to multiple modalities (e.g., biometric data), and this data can be provided simultaneously or at different times for rendering based on the physician 115's corresponding settings and / or preferences.

[0064] As an example of the first technique, cardiac mapping can be achieved by sensing the electrical properties of cardiac tissue (e.g., LAT) at a precise location within the heart 120. The corresponding data (e.g., biometric data) can be acquired via one or more catheters (e.g., catheter 110) advanced into the heart 1120 and having electrical and position sensors (e.g., electrode 111) at its distal tip. As a specific example, position and electrical activity can initially be measured at approximately 10 to approximately 20 points on the inner surface of the heart 120. These data points are typically sufficient to generate a preliminary reconstruction or mapping map of satisfactory quality of the cardiac surface. The preliminary map can be combined with data taken from additional points to produce a more comprehensive mapping map of cardiac electrical activity. In clinical settings, it is not uncommon to accumulate data at 100 or more sites (e.g., several thousand) to generate a detailed and comprehensive mapping map of cardiac chamber electrical activity. The resulting detailed map can then serve as a basis for determining therapeutic actions, such as tissue ablation as described herein, to alter the propagation of cardiac electrical activity and restore normal heart rhythm.

[0065] Additionally, cardiac mapping can be generated based on the detection of intracardiac electrical potential fields (e.g., examples of IC ECG data and / or bipolar intracardiac reference signals). Non-contact techniques for simultaneously acquiring large amounts of cardiac electrical information can be implemented. For example, a catheter type with a distal end portion can be provided with a series of sensor electrodes distributed on its surface and connected to an insulating electrical conductor for connection to a signal sensing and processing device. The size and shape of the end portion can allow the electrodes to be substantially spaced from the walls of the heart chambers. The intracardiac electrical potential field can be detected during a single heartbeat. According to one example, the sensor electrodes can be distributed on a series of circumferences located in planes spaced apart from each other. These planes can be perpendicular to the long axis of the end portion of the catheter. At least two additional electrodes can be provided adjacent to each other at the ends of the long axis of the end portion. As a more specific example, the catheter can include four circumferences, with eight electrodes spaced equally angularly on each circumference. Thus, in this specific embodiment, the catheter can include at least 34 electrodes (32 circumferential electrodes and 2 end electrodes). As another more specific example, the catheter may include other multi-spindle catheters, such as those with five flexible branches, eight radial splines, or parallel spline spatula type (e.g., any of which may have a total of 42 electrodes).

[0066] As an example of electrophysiological mapping or cardiac mapping, an electrophysiological cardiac mapping system and technique based on a non-contact, non-expanding multi-electrode catheter (e.g., catheter 110) can be implemented. ECG can be obtained using one or more catheters 110 having multiple electrodes (e.g., such as between 42 and 122 electrodes). According to this specific implementation, an understanding of the relative geometry of the probe and the endocardium can be obtained through independent imaging modalities such as transesophageal echocardiography. After independent imaging, non-contact electrodes can be used to measure the cardiac surface potential and construct a mapping map from it (e.g., in some cases, using a bipolar intracardiac reference signal). The technique may include the following steps (after the independent imaging step): (a) measuring the potential using multiple electrodes disposed on a probe positioned in the heart 120; (b) determining the geometric relationship between the probe surface and the endocardial surface and / or other references; (c) generating a coefficient matrix representing the geometric relationship between the probe surface and the endocardial surface; and (d) determining the endocardial potential based on the electrode potentials and the coefficient matrix.

[0067] According to another example of electrical mapping or cardiac mapping, techniques and apparatus for mapping the potential distribution of cardiac chambers can be realized. An intracardiac multi-electrode mapping catheter assembly can be inserted into the heart 120. The mapping catheter (e.g., catheter 110) assembly may include a multi-electrode array or a mating reference catheter having one or more integral reference electrodes (e.g., one or more electrodes 111).

[0068] According to one or more exemplary embodiments, the electrodes can be deployed in the form of a substantially spherical array, spatially referenced to points on the endocardial surface via a reference electrode or a reference catheter in contact with the endocardial surface. A preferred electrode array catheter may carry multiple individual electrode sites (e.g., at least 24). Furthermore, this exemplary technique is achieved by understanding the location of each electrode site in the array and understanding the cardiac geometry. These locations are preferably determined using techniques of impedance plethysmography.

[0069] Given electrical mapping or cardiac mapping, and according to another example, catheter 110 may be a cardiac mapping catheter assembly that may include an electrode array defining a plurality of electrode sites. The cardiac mapping catheter assembly may also include a lumen to receive a reference catheter having a distal tip electrode assembly that can be used to probe the heart wall. The cardiac mapping catheter assembly may include a braid of insulating wire (e.g., having 24 to 64 wires in the braid), and each wire may be used to form an electrode site. The cardiac mapping catheter assembly can be readily positioned in the heart 120 for acquiring electrical activity information from a first set of non-contact electrode sites and / or a second set of contact electrode sites.

[0070] Furthermore, according to another example, the catheter 110, which enables mapping electrophysiological activity within the heart, may include a distal tip adapted to deliver stimulation pulses for pacing the heart or an ablation electrode for ablating tissue in contact with the tip. The catheter 110 may also include at least one pair of orthogonal electrodes to generate a differential signal indicative of local cardiac electrical activity adjacent to the orthogonal electrodes.

[0071] As described herein, system 100 can be used to detect, diagnose, and / or treat heart conditions. In exemplary operation, a process for measuring electrophysiological data within the heart chambers may be implemented by system 100. The process may include, in part, positioning a set of active and passive electrodes within the heart 120, supplying current to the active electrodes to generate an electric field within the heart chambers, and measuring the electric field at the sites of the passive electrodes. The passive electrodes are contained in an array positioned on an inflatable balloon of a balloon catheter. In a preferred embodiment, the array reportedly has 60 to 64 electrodes.

[0072] As another exemplary operation, cardiac mapping can be performed by system 100 using one or more ultrasound transducers. The ultrasound transducers can be inserted into a patient's heart 120 and can collect multiple ultrasound slices (e.g., two-dimensional or three-dimensional slices) at various locations and orientations within the heart 120. The location and orientation of a given ultrasound transducer can be known, and the collected ultrasound slices can be stored so that they can be displayed at a later time. One or more ultrasound slices can be displayed corresponding to the location of probe 105 (e.g., a treatment catheter shown as catheter 110) at a later time, and probe 105 can overlay one or more ultrasound slices.

[0073] Given System 100, it should be noted that arrhythmias, including atrial arrhythmias, can be multi-wave reentrant, characterized by multiple asynchronous loops of electrical impulses dispersed around the atrial chambers and typically self-propagating (e.g., another example from IC ECG data). Alternatively, or in addition to multi-wave reentrant arrhythmias, arrhythmias can also have a focal source, such as when isolated areas of tissue within the atria beat spontaneously in a rapid, repetitive manner (e.g., another example from IC ECG data). Ventricular tachycardia (V-tach or VT) is a tachycardia or rapid rhythm originating in one ventricle. This is a potentially life-threatening arrhythmia because it can lead to ventricular fibrillation and sudden cardiac death.

[0074] For example, aFib occurs when the normal electrical impulses generated by the sinoatrial node (e.g., another example of IC ECG data) are overwhelmed by disordered electrical impulses originating from the atrial veins and PV, causing irregular impulses to travel to the ventricles. Irregular heartbeats are generated and can last from minutes to weeks, or even years. aFib is typically a chronic condition with a small increased risk of death, usually due to stroke. The first line of treatment for aFib is medication to slow the heart rate or restore a normal rhythm. Additionally, people with aFib are often given anticoagulants to prevent them from having a stroke. The use of such anticoagulants carries the inherent risk of internal bleeding. For some patients, medication is insufficient, and their aFib is considered drug-resistant, meaning it cannot be treated with standard medical interventions. Synchronized cardioversion can also be used to convert aFib to a normal rhythm. Alternatively, patients with aFib can be treated with catheter ablation.

[0075] Catheter-based ablation therapy may include mapping the electrical properties of cardiac tissue (particularly the endocardium and cardiac volume) and selectively ablating cardiac tissue by applying energy. Electrical mapping or cardiac mapping (e.g., implemented by any electrophysiological cardiac mapping system and techniques described herein) includes creating a mapping map of the potential propagating along cardiac tissue (e.g., a voltage mapping map) or a mapping map of the time of arrival (LAT) to various tissue locations (e.g., a LAT mapping map). Electrical mapping or cardiac mapping (e.g., cardiac mapping) can be used to detect localized cardiac tissue dysfunction. Ablation, such as cardiac mapping-based ablation, can stop or alter unwanted electrical signals propagating from one part of the heart 120 to another.

[0076] The ablation process disrupts 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 microwave, laser, and more commonly radiofrequency energy to create conduction blocks along the cardiac tissue wall. Another example of energy delivery technology includes irreversible electroporation (IRE), which provides a high electric field that damages cell membranes. In a two-step procedure (e.g., mapping followed by ablation), electrical activity at various points within the heart 120 is typically sensed and measured by advancing a catheter 110 containing one or more electrical sensors (or electrodes 111) into the heart 120 and acquiring / obtaining data (e.g., generally such as biometric data, or specifically such as ECG data) at multiple points. This ECG data is then used to select the endocardial target region to be ablated.

[0077] As clinicians treat increasingly challenging conditions such as atrial fibrillation and ventricular tachycardia, cardiac ablation and other cardiac electrophysiological procedures become increasingly complex. Treatment of complex arrhythmias may rely solely on the use of three-dimensional (3D) mapping systems to reconstruct the anatomy of the cardiac chambers of interest. In this regard, the detection engine 101 employed by the system 100 of this paper generally manipulates and evaluates biometric data or specifically ECG data to generate improved tissue data that enables more accurate diagnosis, imaging, scanning, and / or mapping for the treatment of abnormal heartbeats or arrhythmias. For example, cardiologists rely on software such as that produced by Biosense Webster, Inc. (Diamond Bar, California). The 3D mapping system includes a complex fragmented atrial electrocardiography (CFAE) module to generate and analyze ECG data. The detection engine 101 of system 100 enhances the software to generate and analyze improved biometric data, which further provides multiple pieces of information about the electrophysiological characteristics of the heart 120 (including scar tissue), representing the cardiac matrix (anatomy and function) of aFib.

[0078] Therefore, system 100 can realize 3D mapping systems such as 3. A 3D mapping system is used to locate the potential arrhythmogenic matrix of cardiomyopathy in the detection of abnormal EGM. The matrix associated with these cardiomyopathy is related to the presence of fragmented and prolonged ECGs in the endocardial and / or epicardial layers of the ventricular chambers (right and left). For example, low-voltage or intermediate-voltage regions can exhibit fragmented and prolonged ECG activity. Furthermore, during sinus rhythm, low-voltage or intermediate-voltage regions may correspond to key isthmuses identified during sustained and tissue ventricular arrhythmias (e.g., applicable to intolerance ventricular tachycardia, as well as in the atria). Generally, abnormal tissue is characterized by low-voltage ECGs. However, initial clinical experience in endocardial-epicardial mapping indicates that low-voltage regions are not always present as the sole arrhythmogenic mechanism in such patients. In fact, low-voltage or intermediate-voltage regions can exhibit fragmented and prolonged EGM activity during sinus rhythm, which corresponds to key isthmuses identified during sustained and tissue ventricular arrhythmias, e.g., only applicable to intolerance ventricular tachycardia. Furthermore, in many cases, EGM fragmentation and prolongation activity are observed in areas displaying normal or near-normal voltage amplitudes (>1-1.5 mV). While the latter areas can be assessed based on voltage amplitude, they cannot be considered normal based on intracardiac signals and thus represent the true arrhythmogenic matrix. 3D mapping can localize the arrhythmogenic matrix on the endocardial and / or epicardial layers of the right / left ventricle, which can vary in distribution depending on the extent of the primary disease.

[0079] As another exemplary operation, cardiac mapping can be performed by system 100 using one or more multi-electrode catheters (e.g., catheter 110). The multi-electrode catheter is used to stimulate and map electrical activity in the heart 120 and to ablate sites of abnormal electrical activity. In use, the multi-electrode catheter is inserted into a major vein or artery, such as the femoral vein, and subsequently guided into the chamber of the heart 120 of interest. A typical ablation procedure involves inserting catheter 110, having at least one electrode 111 at its distal end, into the heart chamber. A reference electrode is provided either by adhesive to the patient's skin or by a second catheter positioned in or near the heart, or selected from one of catheter 110 or another electrode 111. Radiofrequency (RF) current is applied to the tip electrode 111 of the ablation catheter 110, and the current flows to the reference electrode through a medium surrounding it (e.g., blood and tissue). The current distribution depends 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 the cells in the heart tissue, resulting in the formation of a non-conductive ablation foci within the heart tissue. During this process, heating of the tip electrode 111 also occurs due to conduction from the heated tissue to the electrode itself. If the electrode temperature becomes sufficiently high, possibly above 60°C, a thin, transparent coating of dehydrated hemoglobin can form on the surface of electrode 111. If the temperature continues to rise, this dehydrated layer can become increasingly thick, causing blood clotting on the electrode surface. Because the dehydrated biomaterial has a higher resistance than endocardial tissue, the impedance to the flow of electrical energy into the tissue also increases. If the impedance increases sufficiently, an impedance rise occurs, and the catheter 110 must be removed from the body and the tip electrode 111 cleaned.

[0080] Now go to Figure 2 A diagram is shown of a system 200 according to one or more exemplary embodiments, wherein one or more features of the subject matter of this disclosure may be implemented. Relative to patient 202 (e.g., Figure 1 (Example of patient 125), system 200 includes device 204, local computing device 206, remote computing system 208, first network 210, and second network 211. Additionally, device 204 may include biometric sensor 221 (e.g., ...). Figure 1 Examples include conduit 110, processor 222, user input (UI) sensor 223, memory 224, and transceiver 225. Note that for ease of explanation and brevity, Figure 1 The detection engine 101 in Figure 2 It is reused.

[0081] According to one embodiment, device 204 may be Figure 1An example of system 100, wherein device 204 may include both internal and external components. According to another embodiment, device 204 may be an external device including an attachable patch (e.g., attached to the patient's skin). According to another embodiment, device 204 may be internal to the patient 202 (e.g., subcutaneously implanted), wherein device 204 may be inserted into the patient 202 via any suitable means, including oral injection, surgical insertion via vein or artery, endoscopic procedure, or laparoscopic procedure. According to one embodiment, although in Figure 2 A single device 204 is shown, but the exemplary system may include multiple devices.

[0082] Therefore, device 204, local computing device 206, and / or remote computing system 208 can be programmed to execute computer instructions relative to detection engine 101. For example, memory 223 stores these instructions for execution by processor 222, enabling device 204 to receive and process biometric data via biometric sensor 201. Thus, processor 222 and memory 223 represent the processor and memory of local computing device 206 and / or remote computing system 208.

[0083] Device 204, local computing device 206, and / or remote computing system 208 can be any combination of software and / or hardware that individually or jointly store, execute, and implement the detection engine 101 and its functions. Additionally, device 204, local computing device 206, and / or remote computing system 208 can be an electronic computer framework, including and / or employing any number and combination of computing devices and networks utilizing various communication technologies, as described herein. Device 204, local computing device 206, and / or remote computing system 208 can be easily scaled, expanded, and modularized, with the ability to change for different services or reconfigure some features independently of others.

[0084] Networks 210 and 211 can be wired networks, wireless networks, or include one or more wired and wireless networks. According to one implementation, network 210 is an example of a short-range network (e.g., a local area network (LAN) or a personal area network (PAN)). Information can be transmitted between device 204 and local computing device 206 via network 210 using any of a variety of short-range wireless communication protocols such as Bluetooth, Wi-Fi, Zigbee, Z-Wave, Near Field Communication (NFC), ultra-band, Zigbee, or infrared (IR)). Additionally, network 211 is an example of one or more of the following: an intranet, a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a direct connection or a series of connections, a cellular telephone network, or any other network or medium capable of facilitating communication between local computing device 206 and remote computing system 208. Information can be transmitted via network 211 using any of a variety of long-range wireless communication protocols such as TCP / IP, HTTP, 3G, 4G / LTE, or 5G / new radio. It should be noted that for either network 210 or 211, wired connections can be implemented using Ethernet, Universal Serial Bus (USB), RJ-11, or any other wired and wireless connections can be implemented using Wi-Fi, WiMAX, Bluetooth, infrared, cellular networks, satellite, or any other wireless connection method.

[0085] In operation, device 204 can continuously or periodically acquire, monitor, store, and process biometric data associated with patient 202, and transmit this biometric data via network 210. Additionally, device 204, local computing device 206, and / or remote computing system 208 communicate via networks 210 and 211 (e.g., local computing device 206 may be configured as a gateway between device 204 and remote computing system 208). For example, device 204 may be... Figure 1 The system 100 is an example configured to communicate with a local computing device 206 via network 210. The local computing device 206 may be, for example, a fixed / standalone device, a base station, a desktop / laptop computer, a smartphone, a smartwatch, a tablet, or other device configured to communicate with other devices via networks 211 and 210. It may also be a physical server implemented on or connected to network 211, or a public cloud computing provider (e.g., Amazon Web Services) of network 211. The remote computing system 208 of the virtual server in the system can be configured to communicate with the local computing device 206 via network 211. Therefore, biometric data associated with the patient 202 can be transmitted throughout the system 200.

[0086] The elements of device 204 will now be described. Biometric sensor 221 may include, for example, one or more transducers configured to convert one or more environmental conditions into electrical signals, enabling the observation / acquisition / collection of different types of biometric data. For example, biometric sensor 221 may include one or more of the following: electrodes (e.g., Figure 1 Electrodes 111), temperature sensors (e.g., thermocouples), blood pressure sensors, blood glucose sensors, blood oxygen sensors, pH sensors, accelerometers, and microphones.

[0087] When the detection engine 101 is executed, the processor 222 may be configured to receive, process, and manage biometric data acquired by the biometric sensor 221, and transmit the biometric data to the memory 224 via the transceiver 225 for storage and / or across the network 210. Biometric data from one or more other devices 204 may also be received by the processor 222 via the transceiver 225. As described in more detail below, the processor 222 may be configured to selectively respond to different tap patterns (e.g., single tap or double tap) received from the UI sensor 223, such that different tasks of the patch (e.g., data acquisition, storage, or transmission) can be activated based on the detected pattern. In some embodiments, the processor 222 may generate audible feedback relative to the detected gesture.

[0088] UI sensor 223 includes, for example, a piezoelectric or capacitive sensor configured to receive user input such as a tap or touch. For example, in response to a patient 202 tapping or touching the surface of device 204, UI sensor 223 can be controlled to achieve capacitive connection. Gesture recognition can be achieved via any of a variety of capacitance types, such as resistive capacitance, surface capacitance, projected capacitance, surface acoustic waves, piezoelectricity, and infrared touch. The capacitive sensor can be positioned over a small area or along the length of the surface, such that a tap or touch on the surface activates the monitoring device.

[0089] Memory 224 is any non-transitory tangible medium, such as magnetic memory, optical memory, or electronic memory (e.g., any suitable volatile and / or non-volatile memory, such as random access memory or hard disk drive). Memory 224 stores computer instructions for execution by processor 222.

[0090] Transceiver 225 may include a separate transmitter and a separate receiver. Alternatively, transceiver 225 may include a transmitter and receiver integrated into a single device.

[0091] In operation, device 204 uses detection engine 101 to observe / acquire biometric data of patient 202 via biometric sensor 221, stores the biometric data in memory, and shares the biometric data across system 200 via transceiver 225. Detection engine 101 can then utilize models, algorithms (e.g., deep learning optimization), neural networks, machine learning, and / or artificial intelligence to generate and provide mappings to the physician, thereby reducing the processing load on system 100 and transforming the operation of system 100 into a more accurate mapping machine.

[0092] Now go to Figure 3 The diagram illustrates a method 300 according to one or more embodiments. Generally, method 300 demonstrates one or more operations of a detection engine 101 that implement deep learning optimizations for detecting maintenance foci of aFib to be ablated for treating persistent aFib subjects and classifying velocity vector field images and raw data into maintenance foci.

[0093] Method 300 begins at block 307, where detection engine 101 detects one or more segments of the LAT (e.g., relative to a first activation time). In this regard, the input to detection engine 101 includes biometric data in the form of ECG data signals. Based on the ECG data signals, one or more of the LATs are determined within at least one time interval (e.g., segment). Specifically, detection engine 101 determines a time point (e.g., relative to the first activation time) corresponding to a threshold activity of local activation, calculated based on a normalized initial starting point.

[0094] At block 315, detection engine 101 calculates the field, for example, by modeling a vector velocity field. For instance, the vector velocity field is measured and quantified as the ECG data signal passes through the LAT (e.g., by calculating the direction of the wave at each x,y point and utilizing the derivative of the polynomial surface to provide the velocity vector field). According to one or more embodiments, catheter 110 may be located on a surface in the atrium (x,y) plane, and detection engine 101 may use a scatter plot to describe the LAT (e.g., relative to the first activation time). Detection engine 101 may estimate the coefficients of the scatter plot and determine its derivatives to provide the velocity vector field.

[0095] At box 321, detection engine 101 detects and classifies maintenance foci. In this respect, detection engine 101 utilizes the velocity vector field of box 315 as input to machine learning and / or artificial intelligence algorithms (e.g., deep convolutional neural networks) to detect the location of the "gold standard" maintenance foci. One or more advantages, technical effects, and / or benefits of box 321 include a large-scale effort leading to an understanding of specific case outcomes, including whether the ablation outcome was successful (e.g., whether the ablation treatment had one or more positive and negative outcomes and to what extent). Therefore, detection engine 101 provides automated understanding based on a set of ECG signals that can be displayed via a GUI. Detection engine 101 performing method 300 can provide retrospective analysis to investigate past cases and determine where ablation occurred, as well as prospective analysis to determine where ablation will be performed in future cases.

[0096] According to one or more embodiments, the detection engine 101 utilizes machine learning algorithms (such as neural networks as described herein) to determine cues in the ECG data signal regarding the outcome (e.g., which part of the data indicates when and whether the procedure had a positive outcome). Cues may also include, but are not limited to, inputs such as system status, ablation parameters, ablation location, ablation duration, applied force, power, and temperature. Once the tachycardia has been altered by ablation and maintained, including, for example, prolongation and then termination of the tachycardia, the machine learning algorithm of the detection engine 101 can automatically flag the event and / or location leading to termination. The machine learning algorithm of the detection engine 101 can also flag the location and event of ablation resolved in termination predicted / expected by the machine learning algorithm based on physician-provided indications. Thus, the detection engine 101 informs the user of the clinical outcome of the electrophysiological procedure. Additionally, the operation of the detection engine 101 can be applied to other features of interest, such as the HIS bundle (e.g., including specific local signals) and diaphragmatic capture (e.g., including data fragment stimulation during an AI session), and the understanding of the appearance of successful termination.

[0097] Combined with box 321 of method 300, Figure 4 A graphical depiction of an artificial intelligence system 400 according to one or more embodiments is shown. The artificial intelligence system 400 includes data 410 (e.g., biometric data), a machine 420, a model 430, results 440, and (low-level) hardware 450. For ease of understanding where appropriate, reference is made to... Figures 1 to 3 conduct Figures 4 to 5 The description. For example, machine 410, model 430, and hardware 450 can be represented. Figures 1 to 2 The detection engine 101 (e.g., the machine learning and / or artificial intelligence algorithms therein) and the hardware 450 can also represent various aspects. Figure 1 Catheter 110 Figure 1The console 160 and / or Figure 2 Device 204. Generally speaking, the machine learning and / or artificial intelligence algorithms of the artificial intelligence system 400 (e.g., such as...) Figures 1 to 2 The detection engine 101 implements the operation of data 410 relative to hardware 450 to train the machine 420, build the model 430, and predict the results 440.

[0098] For example, machine 420 may act as or be associated with or related to a controller or data collection operation of hardware 450. Data 410 (e.g., biometric data as described herein) may be ongoing or output data associated with hardware 450. Data 410 may also include currently collected data, historical data, or other data from hardware 450; may include measurements taken during and associated with the results of the surgical procedure; and may include data collected and associated with the results of the cardiac procedure. Figure 1 The temperature of the heart 140; and can be associated with hardware 450. Data 410 can be divided into one or more subsets by machine 420.

[0099] Furthermore, machine 420 is trained relative to hardware 450. This training may also include analyzing and correlating the collected data 410. According to one or more embodiments, detection engine 101 may train machine learning algorithms relative to determining acute arrhythmia termination, aFib termination, or termination for any tachycardia and / or relative to results identified after a several-day disappearance period, and notifications that long-term follow-up use may be employed.

[0100] For example, in the case of the heart, data 410 on trainable temperature and outcomes can be used to determine the effects during cardiac procedures. Figure 1 Is there a correlation or relationship between the temperature of the heart 140 and the results? According to another embodiment, the training machine 420 may include a [missing information - likely a component or device]. Figure 1 The detection engine 101 is self-trained using one or more subsets. In this respect, Figure 1 The detection engine 101 learns to classify cases point by point.

[0101] Furthermore, model 430 is built upon data 410 associated with hardware 450. Building model 430 may include physical hardware or software modeling, algorithmic modeling, etc., attempting to represent the data 410 (or a subset thereof) that has been collected and trained. In some aspects, the construction of model 430 is part of a self-training operation performed by machine 420. Model 430 may be configured to model the operation of hardware 450 and to model the data 410 collected from hardware 450 to predict the outcome 440 achieved by hardware 450. The predicted outcome 440 (of model 430 associated with hardware 450) may utilize the trained model 430. For example, and to enhance understanding of this disclosure, in the case of the heart, if temperatures between 36.5°C and 37.89°C (i.e., 97.7°F and 100.2°F) during a protocol produce a positive outcome from a cardiac protocol, that outcome 440 may be predicted using those temperatures in a given protocol. Therefore, using the predicted results 440, the machine 420, model 430, and hardware 450 can be configured accordingly.

[0102] Therefore, in order for the artificial intelligence system 400 to use data 410 to operate relative to hardware 450 to train machine 420, build model 430, and predict results 440, the machine learning and / or artificial intelligence algorithms therein may include neural networks. 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.

[0103] Artificial neural networks involve networks of simple processing elements (artificial neurons) that can exhibit complex global behavior determined by the connections between these elements and their parameters. These connections in a network or circuit of 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 it could be between -1 and 1. In most cases, ANNs are adaptive systems based on information flowing through the network's external or internal structures to change their structure.

[0104] In more practical terms, neural networks are nonlinear statistical data modeling or decision-making tools that can be used to model complex relationships between inputs and outputs or to find patterns in data. Thus, ANNs can be used for predictive modeling and adaptive control applications, while being trained on datasets. Experience-based self-learning can occur within the network, drawing conclusions from complex and seemingly unrelated sets of information. The utility of artificial neural network models lies in their ability to infer functions from observations and also to use those functions. Unsupervised neural networks can also be used to learn representations of inputs that capture salient features of the input distribution, and recent deep learning algorithms can implicitly learn the distribution function of observed data. Learning in neural networks is particularly useful in applications where the complexity of the data (e.g., biometric data) or tasks (e.g., monitoring, diagnosing, and treating any number of diseases) makes manually designing such functions impractical.

[0105] Neural networks can be used in various fields. Therefore, for an artificial intelligence system 400, the machine learning and / or artificial intelligence algorithms may include neural networks, which are typically categorized according to the task they are applied to. These categorizations tend to fall into the following categories: regression analysis (e.g., function approximation), including time series forecasting and modeling; classification, including pattern and sequence recognition; novelty detection and sequential decision-making; data processing, including filtering; clustering; and blind signal separation and compression. For example, application areas of ANNs include nonlinear system recognition and control (vehicle control, process control), game playing and decision-making (backgammon, chat, competition), pattern recognition (radar systems, facial recognition, object recognition), sequence recognition (gesture, speech, handwritten text recognition), medical diagnosis and treatment, financial applications, data mining (or knowledge discovery in databases, "KDD"), visualization, and email spam filtering. For example, semantic features can be created from patient biometric data emerging from medical protocols.

[0106] According to one or more implementations, the neural network can implement a long short-term memory neural network architecture, a convolutional neural network (CNN) architecture, or a recurrent neural network (RNN) architecture, etc. The neural network can be configured with respect to multiple layers, multiple connections (e.g., encoder / decoder connections), regularization techniques (e.g., differential pressure), and optimized features.

[0107] Long Short-Term Memory (LSTM) neural network architectures include feedback connections and can process single data points (e.g., images) as well as entire sequences of data (e.g., speech or video). A unit in an LSM neural network architecture may consist of a cell, an input gate, an output gate, and a forget gate, where the cell remembers a value over any time interval, and the gates regulate the flow of information into and out of the cell.

[0108] A CNN (Neural Network Array) architecture is a shared-weight architecture with translation invariance, where each neuron in one layer is connected to all neurons in the next layer. Regularization techniques in CNN architectures can leverage hierarchical patterns in the data and assemble more complex patterns using smaller, simpler patterns. Other configurable aspects of a neural network implementing a CNN architecture can include the number of filters at each level, the kernel size, and the number of kernels per layer.

[0109] Now go to Figure 5 This diagram illustrates an example of a neural network 500 according to one or more embodiments and a block diagram of a method 501 performed in the neural network 500. The neural network 500 operates to support the implementation of the machine learning and / or artificial intelligence algorithms described herein (e.g., as described by...). Figures 1 to 2 The detection engine 101 is implemented in the neural network 500 (which can be implemented in hardware such as...). Figure 4 Implemented in machine 420 and / or hardware 450. As indicated herein, for ease of understanding where appropriate, refer to Figures 1 to 3 conduct Figures 4 to 5 The description.

[0110] In the exemplary operation, Figure 1 The detection engine 101 includes data collection 410 from hardware 450. In the neural network 500, the input layer 510 consists of multiple inputs (e.g., Figure 5 Inputs 512 and 514 are represented. Relative to block 520 of method 501, input layer 510 receives inputs 512 and 514. Inputs 512 and 514 may include biometric data. For example, data collection 410 may involve aggregating biometric data (e.g., BS ECG data, IC ECG data, and ablation data, along with catheter electrode position data) from one or more procedures recorded from hardware 450 into a dataset (as represented by data 410).

[0111] At block 525 of method 501, neural network 500 encodes inputs 512 and 514 using any portion of data 410 (e.g., datasets and predictions generated by artificial intelligence system 400) to produce a latent representation or data encoding. The latent representation includes one or more intermediate data representations derived from multiple inputs. According to one or more embodiments, the latent representation is... Figure 1 The detection engine 101 generates element-level activation functions (e.g., sigmoid functions or trimmed linear units). For example... Figure 5As shown, inputs 512 and 514 are provided to a hidden layer 530, depicted as including nodes 532, 534, 536, and 538. The neural network 500 performs processing via the hidden layer 530 with nodes 532, 534, 536, and 538 to exhibit complex global behavior determined by the connections between processing elements and element parameters. Therefore, the transition between layers 510 and 530 can be considered as an encoder stage, which takes inputs 512 and 514 and feeds them into the deep neural network (within layer 530) to learn some smaller representation of the inputs (e.g., the resulting latent representation).

[0112] The deep neural network can be a CNN, a long short-term memory neural network, a fully connected neural network, or a combination thereof. Inputs 512 and 514 can be intracardiac ECGs, surface ECGs, or both intracardiac and surface ECGs. This encoding provides dimensionality reduction for inputs 512 and 514. Dimensionality reduction is the process of reducing the number of random variables considered (inputs 512 and 514) by obtaining a set of principal variables. For example, dimensionality reduction can be feature extraction that transforms data (e.g., inputs 512 and 514) from a high-dimensional space (e.g., more than 10 dimensions) to a low-dimensional space (e.g., 2-3 dimensions). The technical effects and benefits of dimensionality reduction include reducing the time and storage space requirements of data 410, improving the visualization of data 410, and improving the interpretation of parameters for machine learning. This data transformation can be linear or non-linear. The operations of receiving (box 520) and encoding (box 525) can be considered as the data preparation part of the multi-step data manipulation performed by the detection engine 101.

[0113] At box 545 of method 510, neural network 500 decodes the latent representation. The decoding stage takes the encoder output (e.g., the resulting latent representation) and attempts to reconstruct some form of the inputs 512 and 514 using another deep neural network. In this regard, nodes 532, 534, 536, and 538 are combined to produce output 552 in output layer 550, as shown in box 560 of method 510. That is, output layer 590 reconstructs inputs 512 and 514 in a reduced dimension, but without signal interference, signal artifacts, and signal noise. Examples of output 552 include cleaned biometric data (e.g., a clean / denoised version of IC ECG data, etc.). The technical effects and benefits of cleaned biometric data include the ability to more accurately monitor, diagnose, and treat any number of various diseases.

[0114] return Figure 3Method 300 continues at box 330, where detection engine 101 determines / calculates the code for each point in the (x,y) plane relative to one of four (4) directions (e.g., left can be red, right can be green, top can be blue, and bottom can be yellow) to provide a color code vector field image. The color code vector field image enables the localization of lesion sources or maintenance lesions.

[0115] At frame 345, detection engine 101 detects a lesion / rotor indicator, for example, by scanning the color-coded vector field image of frame 330 using a kernel (e.g., typically a circle with a radius of 1 mm). Generally, the kernel can be a specified space within the mapping and can represent the desired size (e.g., a circle or a square) and dimensions. In some cases, the kernel can be the address of a point within the mapping, a computational segment of the mapping, the computational boundary of the mapping, etc. In cardiac tissue, a spiral wave reentry occurs when an electrically propagating wavefront encounters functionally inexcitable tissue and rotates around it in a vortex-like manner. Subsequently, the rotor indicator or rotor can be the center of rotation from which a 2D spiral excitation wave rotates outward. Alternatively, the lesion indicator or lesion can be an arrhythmia in which the electrical impulse originates from and is confined within the atrium. If it is within a kernel and all directions are ordered (e.g., clockwise or counterclockwise), the lesion / rotor indicator can be detected. Therefore, the detection engine 101 distinguishes between lesion sources and rotors (e.g., active and passive) on top of modeling the vector velocity field.

[0116] At box 360, detection engine 101 marks maintenance foci. According to one or more embodiments, a maintenance foci are tissue triggers and / or initiators that ensure the continued presence of aFib. Detection engine 101 may automatically identify or annotate aFib maintenance foci based on vector velocity and ablation information. More specifically, detection engine 101 may determine one or more gold-standard annotations (e.g., the most accurate annotation) for aFib maintenance foci. Alternatively or in combination, physicians may mark suspicious maintenance foci that can be ablated during the ablation procedure.

[0117] At box 375, detection engine 101 classifies lesions / trochanter indicators as maintenance foci. Classifying lesions / trochanter indicators as maintenance foci involves classifying active or passive regions of interest (ROIs) based on engineered features and vector field mapping using machine learning and / or artificial intelligence (e.g., logistic regression classifiers, support vector machines, deep learning, etc.). For example, a vector velocity image with maintenance foci gold standard annotations can be used as input and / or target for a neural network (e.g., an RNN or CNN as described herein) to predict whether pixels in the vector velocity image are maintenance foci. Furthermore, classifying one or more lesions and / or trochanter indicators as maintenance foci can utilize the velocity vector field as input to a neural network (e.g., an RNN or CNN as described herein) for detecting the location of gold standard maintenance foci.

[0118] According to one or more implementations, ROI annotation can be used by the detection engine 101 relative to active, passive, and unknown categories. For example, an active ROI can indicate that the ROI was ablated and that aFib termination or cycle length (CL) increased due to the ablation treatment. Active ROIs include maintenance foci with “clinical values” that include evidence of cycle length prolongation or atrial fibrillation termination resulting from ablation treatment near the lesion / rotor. A passive ROI can indicate that, in the absence of any visible changes in atrial fibrillation characteristics, there are no visible changes in aFib characteristics, including ablation treatment near the lesion source. An unknown annotation for an ROI can indicate that there is no visible ablation treatment near the lesion / rotor. In this way, one or more advantages, technical effects, and / or benefits of the detection engine include addressing most of the mapped points, which are not addressed by the physician because the physician is unaware of the active or passive nature of the ROI. The detection engine 101 can also assign probabilities across active, passive, and unknown categories (e.g., a scale of 0 to 100 indicating the probability of the category).

[0119] It should be noted that, relative to the operations of boxes 321, 360, and 375, detector engine 101 can use the result of any operation as feedback or input to another operation, as shown by the double arrows.

[0120] Now go to Figure 6 The diagram illustrates a method 600 according to one or more embodiments. Generally, method 600 demonstrates an implementation optimization of detection engine 101 to detect one or more atrial fibrillation maintenance foci to be ablated for treatment.

[0121] At block 605, method 600 begins, wherein detection engine 101 receives one or more inputs from conduit 110, such as IC ECG data signals. At block 610, detection engine 101 detects segments of LAT. Figure 7 A graph 700 is shown according to one or more embodiments. (e.g.) Figure 7 As shown, the segment of LAT is detected relative to the activation time relative to the first activation time (shown as circle 710).

[0122] At box 625, detection engine 101 determines / computes / models the vector velocity field. In this respect, detection engine 101 assumes that catheter 110 lies on the surface of the atrium (x,y) plane and uses a scatter plot to describe the activation time (e.g., relative to what is described as...). Figure 7 (The first activation time of the circle). Figure 8A graph 800 is shown for a surface (x,y) according to one or more embodiments. Graph 800 provides a graph including the surface (x,y) and the activation time in the segment in Z milliseconds (msec). Given graph 800, detection engine 101 can estimate the coefficients of the surface (x,y) that best fits its points. The polynomial surface T(x,y) can be used to estimate N and a relative to the cost function (as seen in Equation 1) using gradient descent. i,j At position x s ,y s The local activation time of L measured at (x) s ,y s Estimated surface T(x) in the plane s ,y s The mean squared error between the parameters is minimized by taking into account a regularization term (ρ is typically equal to 0.1) that accounts for the number of parameters estimated in the model. More specifically, the polynomial surface can be defined by Equation 2.

[0123]

[0124]

[0125] From T(x, y) to the vector field, the model can be used to calculate the direction of the radio wave at each point in (x, y) as defined by Equation 3.

[0126]

[0127] T x Defined by Formula 4.

[0128]

[0129] Figure 9 The following are illustrated according to one or more implementation schemes. Figure 8 The velocity vector field of the surface (x, y) is shown in Figure 900. That is to say, Figure 8 The derivative of the polynomial surface provides Figure 9 The velocity vector field shown.

[0130] At box 630, each point in the (x, y) plane is encoded into one of four (4) directions to provide a color-coded vector field image that allows for the localization of lesion sources or maintenance lesions.

[0131] At frame 645, the lesion / rotor can be detected by detection engine 101. For example, detection engine 101 can use a kernel (typically a circle with a radius of 1 mm) to scan the color-coded vector field image obtained in frame 630. If the rotor / lesion point appears within the kernel and in all four directions in sequence (clockwise or counterclockwise), then the lesion / rotor point is detected.

[0132] At box 660, the detection engine 101 can mark the maintenance lesion. At box 675, the detection engine 101 can classify the lesion / rotor indication as a maintenance lesion.

[0133] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible specific implementations of systems, methods, and computer program products according to various embodiments of the invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of instructions comprising one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions indicated in the blocks may not occur in the order shown in the drawings. For example, depending on the function involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a system based on dedicated hardware that performs the specified function or action, or by a combination of dedicated hardware and computer instructions.

[0134] While features and elements have been specifically described above, those skilled in the art will recognize that each feature or element can be used alone or in any combination with other features and elements. Furthermore, the methods described herein can be implemented in a computer program, software, or firmware incorporated in a computer-readable medium for execution by a computer or processor. As used herein, a computer-readable medium should not be construed as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0135] Examples of computer-readable media include electronic signals (transmitted via wired or wireless connections) and computer-readable storage media. Examples of computer-readable storage media include, but are not limited to, registers, cache memory, semiconductor memory devices, magnetic media such as internal hard disks and removable disks, magneto-optical media, optical media such as optical discs (CDs) and digital versatile discs (DVDs), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), and memory sticks. A processor associated with software can be used to implement a radio frequency transceiver used in a terminal, base station, or any host computer.

[0136] It should be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms “an,” “a,” and “the / described” include plural references. It should also be understood that the terms “comprising” and / or “including”, when used in this specification, specify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0137] The descriptions of various embodiments herein are presented for illustrative purposes and are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles of the embodiments, their practical application or improvement relative to commercially available technologies, or to enable others skilled in the art to understand the disclosed embodiments.

Claims

1. A method implemented by a detection engine, said detection engine being embodied in processor-executable code stored in memory and executed by at least one processor, said method comprising: The detection engine models the vector velocity field, which measures and quantifies the velocity of the electrocardiogram data signal after the local activation time. The detection engine encodes each point in the plane into one of four directions to provide a color-coded vector field image, wherein the four directions are each equal to a different color. The detection engine detects lesion indications and rotor indications by using one or more kernels to scan the color-coded vector field image; and The detection engine classifies the lesion indication and the rotor indication into maintenance lesions.

2. The method according to claim 1, wherein, The electrocardiogram data signal is detected by a catheter within the anatomical structure and in communication with the detection engine.

3. The method according to claim 1, wherein, The detection engine detects one or more segments of the local activation time relative to the first activation time.

4. The method according to claim 1, wherein, The detection engine models the velocity vector field by calculating the direction of the radio wave at each x, y point and using the derivative of the polynomial surface to provide the velocity vector field.

5. The method according to claim 1, wherein, The detection engine classifies the maintenance foci by using the velocity vector field as input to a machine learning algorithm.

6. The method according to claim 5, wherein, The machine learning algorithm includes a deep convolutional neural network or a recurrent neural network to detect the location of the gold standard maintenance foci in the maintenance foci.

7. The method according to claim 5, wherein, The machine learning algorithm determines whether the ablation result for a specific case based on the electrocardiogram data signal was successful.

8. The method according to claim 1, wherein, The lesion indication and the rotor indication are detected when all directions are sequential within one or more cores.

9. The method according to claim 1, wherein, When the maintenance foci are labeled, the detection engine automatically identifies and annotates atrial fibrillation maintenance foci based on vector velocity and ablation information.

10. The method according to claim 1, wherein, The detection engine uses region of interest annotation relative to the active, passive, and unknown categories to at least indicate whether the region of interest was ablated or atrial fibrillation terminated.

11. The method according to claim 1, wherein, The detection engine classifies the maintenance foci by using the velocity vector field as input to an artificial intelligence algorithm.

12. The method according to claim 11, wherein, The artificial intelligence algorithm includes a deep convolutional neural network or a recurrent neural network to detect the location of the gold standard maintenance foci in the maintenance foci.

13. The method according to claim 12, wherein, The artificial intelligence algorithm determines whether the ablation result for a specific case based on the electrocardiogram data signal was successful.

14. A system for detecting, diagnosing, and / or treating cardiac conditions, comprising: A memory that stores processor-executable code for the detection engine; and At least one processor, said at least one processor executing processor-executable code to cause the system to: The detection engine models the vector velocity field, which measures and quantifies the velocity of the electrocardiogram data signal after the local activation time. The detection engine encodes each point in the plane into one of four directions to provide a color-coded vector field image, wherein the four directions are each equal to a different color. The detection engine detects lesion indications and rotor indications by using one or more kernels to scan the color-coded vector field image; and The detection engine classifies the lesion indication and the rotor indication as maintenance lesions.

15. The system according to claim 14, wherein, The electrocardiogram data signal is detected by a catheter within the anatomical structure and in communication with the detection engine.

16. The system according to claim 14, wherein, The detection engine detects one or more segments of the local activation time relative to the first activation time.

17. The system according to claim 14, wherein, The detection engine models the velocity vector field by calculating the direction of the radio wave at each x, y point and using the derivative of the polynomial surface to provide the velocity vector field.

18. The system according to claim 14, wherein, The detection engine classifies the maintenance foci by using the velocity vector field as input to a machine learning algorithm.

19. The system according to claim 18, wherein, The machine learning algorithm includes a deep convolutional neural network or a recurrent neural network to detect the location of the gold standard maintenance foci in the maintenance foci.

20. The system according to claim 18, wherein, The machine learning algorithm determines whether the ablation result for a specific case based on the electrocardiogram data signal was successful.

21. The system according to claim 14, wherein, The lesion indication and the rotor indication are detected when all directions are sequential within one or more cores.

22. The system according to claim 14, wherein, When the maintenance foci are labeled, the detection engine automatically identifies and annotates atrial fibrillation maintenance foci based on vector velocity and ablation information.

23. The system according to claim 14, wherein, The detection engine uses region of interest annotation relative to the active, passive, and unknown categories to at least indicate whether the region of interest was ablated or atrial fibrillation terminated.

24. The system according to claim 14, wherein, The detection engine classifies the maintenance foci by using the velocity vector field as input to an artificial intelligence algorithm.

25. The system according to claim 24, wherein, The artificial intelligence algorithm includes a deep convolutional neural network or a recurrent neural network to detect the location of the gold standard maintenance foci in the maintenance foci.

26. The system according to claim 24, wherein, The artificial intelligence algorithm determines whether the ablation result for a specific case based on the electrocardiogram data signal was successful.

Images