CN113812957A

CN113812957A - Ventricular far-field estimation using an auto-encoder

Info

Publication number: CN113812957A
Application number: CN202110694726.XA
Authority: CN
Inventors: Y·A·阿摩司; M·巴-塔尔; S·戈德堡; G·D·马勒基; M·阿米特; L·特索夫
Original assignee: Biosense Webster Israel Ltd
Current assignee: Biosense Webster Israel Ltd
Priority date: 2020-06-19
Filing date: 2021-06-21
Publication date: 2021-12-21
Also published as: IL284224A; IL284224B1; JP2022001256A

Abstract

The invention provides ventricular far-field estimation using an automatic encoder. The invention provides a method. The method includes receiving an input intracardiac signal from a monitoring and processing device. Each of these input intracardiac signals includes an artifact. The method includes encoding, by an auto-encoder, the input intracardiac signals with an intracardiac data set to produce a potential representation. The method also includes decoding, by the auto-encoder, the potential representation to produce an output intracardiac signal. These output intracardiac signals comprise the input intracardiac signals reconstructed without these signal artifacts.

Description

Ventricular far-field estimation using an auto-encoder

Technical Field

The present invention relates to artificial intelligence and machine learning automatic encoders associated with ventricular far-field estimation and identification and decomposition of near-field and far-field signals in cardiac electrical activity.

Background

Treatment of cardiac disorders, such as arrhythmias, typically requires cardiac mapping (i.e., mapping cardiac tissue, chambers, veins, arteries, and/or passageways). An Electrocardiogram (ECG) is an example of cardiac mapping. An ECG is generated from electrical signals from the heart that describe the activity of the heart.

The ECG is used during a cardiac procedure to identify potential locations of origin of a cardiac disorder. Generally, when a physician uses an ECG to study cardiac activity, signal interference, signal artifacts, and signal noise associated with the underlying electrical signals of the ECG can particularly obscure the accuracy of the ECG. Signal interference may also result from processing signal regions (including high frequency and harmonic regions) with sharp changes, peaks, and/or pacing signals. Because of these disturbances, artifacts, and noise, the physician is unable to distinguish the originating location of the ventricles and atria in real time (e.g., during a cardiac procedure), which increases the difficulty of diagnosing/treating cardiac conditions. Accordingly, there is a need to provide improved methods of cardiac mapping for removing such interference, artifacts, and noise.

A unipolar signal is a combination of a near-field signal and a far-field signal. During an ablation procedure, it is important to identify and isolate the near field signal. When electrodes are inserted into a muscle, such as a heart muscle, each activation of the muscle generates an electric field. Each electrode captures all sources of electric fields in the location where it is placed, including near-field signals near the electrode and far-field signals away from the electrode.

Disclosure of Invention

According to one embodiment, a method is provided. The method includes receiving an input intracardiac signal from a monitoring and processing device. Each of these input intracardiac signals may include at least an artifact. The method includes encoding, by an auto-encoder, the input intracardiac signals with an intracardiac data set to produce a potential representation. The method also includes decoding, by the auto-encoder, the potential representation to produce an output intracardiac signal. These output intracardiac signals may comprise input intracardiac signals reconstructed without artifacts.

According to one embodiment, a method of resolving a near-field signal and a far-field signal is provided. The measured signal may be received. The measured signal may be encoded by an auto-encoder to produce a potential representation. The potential representation may be decoded by an auto-encoder to resolve near-field and far-field components from the measured signal. Far-field ventricular measurements may be acquired. Measurements may be acquired using a multi-electrode catheter and a body surface ECG signal. A synthetic local field signal may be added. The resulting far-field signal and the residual near-field signal may be detected. Decoding the potential representation may be based on the detected resulting far-field signal and the residual near-field signal.

The above-described method embodiments may be implemented as an apparatus, system, and/or computer program product in accordance with one or more embodiments.

Drawings

A more particular understanding can be obtained by reference to the following detailed description, which is provided by way of example in connection with the accompanying drawings, wherein like reference numbers refer to like elements, and in which:

fig. 1 is an illustration of an exemplary system in which one or more features of the disclosed subject matter may be implemented.

Fig. 2 illustrates a block diagram of an exemplary system for remotely monitoring and communicating patient biometrics in accordance with one or more embodiments;

FIG. 3 depicts a graphical depiction of an artificial intelligence system in accordance with one or more embodiments;

FIG. 4 illustrates a block diagram of a method performed in the artificial intelligence system of FIG. 3, in accordance with one or more embodiments;

fig. 5 illustrates an example of a neural network in accordance with one or more embodiments;

FIG. 6 shows a block diagram of a method in accordance with one or more embodiments;

FIG. 7 shows a graphical depiction of a signal in accordance with one or more embodiments;

FIG. 8 shows a graphical depiction of a signal in accordance with one or more embodiments;

FIG. 9 shows a graphical depiction of a signal in accordance with one or more embodiments;

FIG. 10 depicts a graphical depiction of a signal process according to one or more embodiments;

FIG. 11 shows a block diagram of a method in accordance with one or more embodiments; and is

Fig. 12 is an exemplary flow diagram of an exemplary method of decomposing a near-field signal and a far-field signal according to an embodiment.

Detailed Description

An artificial intelligence and machine learning autoencoder (generally referred to herein as an autoencoder) is disclosed herein. The auto-encoder may be processor-executable code or software that is necessarily derived from the processing operations performed by the medical device apparatus and the processing hardware of the medical device apparatus to provide an improved ECG for treating a cardiac condition. According to one embodiment, the autoencoder may provide specific encoding and decoding methods for the medical device apparatus. This particular encoding and decoding method may involve multi-step data manipulation of the electrical signal of the heart that removes signal interference, signal artifacts, and signal noise from the electrical signal.

In this regard and in operation, an automated encoder may receive an input intracardiac signal (e.g., an electrical signal of the heart that includes signal interference, signal artifacts, and signal noise). The intracardiac signals may be recorded and processed in real-time by a monitoring and processing device (e.g., a catheter having an automated encoder therein), and/or recorded and transmitted by the monitoring and processing device to a computing device having an automated encoder therein.

The auto-encoder may encode the input intracardiac signals using an intracardiac data set (e.g., a predetermined and approved electrical signal of the heart without signal interference, signal artifacts, and signal noise). This encoding by the auto-encoder produces a potential representation from the input intracardiac signals. The auto-encoder may further decode the potential representation to produce an output intracardiac signal. The output intracardiac signals may be input intracardiac signals reconstructed without signal interference, signal artifacts, and signal noise. An improved ECG for treating the cardiac condition is then generated from the output intracardiac signals.

Technical effects of the automated encoder include the generation of output intracardiac signals in real-time, further enabling the generation of improved ECGs for physicians (such as during cardiac protocols) that use these improved ECGs to study cardiac activity to identify potential locations of origin of cardiac disorders. The improved ECG is not obscured by signal interference, signal artifacts, and signal noise of the original input intracardiac signals, as these artifacts have been removed during decoding. Further, technical effects of the autoencoder include producing an improved ECG with improved accuracy in which signal interference, signal artifacts, and signal noise are removed to allow for real-time provision of the location of origin of the ventricles and atria, respectively.

In one embodiment, an auto-encoder may be utilized to train the system to decompose near-field and far-field signals detected by the electrodes by analyzing a large number of data points. The bits may be selected as part of a training set for use in training the system to identify far-field signal components.

A signal may be provided and an attempt to regenerate the signal may be made by providing a signal having a large number of far-field signals and a signal having a large number of both far-field and near-field signals. A signal may be provided to the auto-encoder to reconstruct the far-field signal. Once the network is trained, the network may output far-field components from the provided signals.

Fig. 1 shows a diagram of an exemplary system 100 (e.g., a medical device apparatus) that can implement one or more features of the presently disclosed subject matter. All or part of system 100 may be used to collect information for intracardiac data sets (e.g., training data sets), and/or all or part of system 100 may be used to implement an auto-encoder (e.g., trained model) as described herein.

System 100 may include a component, such as catheter 105, configured to damage a tissue region of an organ within the body. The catheter 105 may also be further configured to obtain biometric data including electrical signals of the heart (e.g., intracardiac signals). Although the catheter 105 is shown as a pointed catheter, it should be understood that any shape of catheter including one or more elements (e.g., electrodes) may be used to implement embodiments disclosed herein.

The system 100 includes a probe 110 having an axis that can be navigated by a doctor or medical professional 115 to a body part of a patient 125, such as a heart 120, lying on a bed (or table) 130. According to an embodiment, a plurality of probes may be provided. However, for simplicity, a single probe 110 is described herein. However, it should be understood that the probe 110 may represent a plurality of probes.

The exemplary system 100 may be used to detect, diagnose, and treat cardiac conditions (e.g., using intracardiac signals). Cardiac disorders, such as cardiac arrhythmias (specifically atrial fibrillation), have long been a common and dangerous medical condition, especially in the elderly population. In a patient with a normal sinus rhythm (e.g., patient 125), a heart (e.g., heart 120) including atria, ventricles, and excitatory conducting tissue is electrically stimulated to beat in a synchronized, patterned manner. The electrical excitation may be detected as an intracardiac signal.

For patients with arrhythmias (e.g., patient 125), the abnormal region of cardiac tissue does not follow the synchronous beat cycle associated with normal conductive tissue as does a patient with a normal sinus rhythm. In contrast, abnormal regions of cardiac tissue are abnormally conducted to adjacent tissue, thereby disrupting the cardiac cycle to an asynchronous rhythm. Such unsynchronized heart rhythms may also be detected as intracardiac signals. Such abnormal conduction is previously known to occur in various regions of the heart (e.g., heart 120), such as in the Sinoatrial (SA) junction region, along the conduction pathway of the Atrioventricular (AV) node, or in the myocardial tissue forming the ventricular and atrial chamber walls.

In addition, cardiac arrhythmias, including atrial arrhythmias, may be of the multi-wavelet reentrant type characterized by multiple asynchronous loops of electrical impulses dispersed around the atrial chamber and generally self-propagating (e.g., another example of an intracardiac signal). Alternatively, or in addition to the multi-wavelet reentry type, arrhythmias may also have a focal source, such as when isolated tissue regions within the atrium beat autonomously in a rapidly repeating manner. (another example of intracardiac signals, for example) ventricular tachycardia (V-tach or VT) is a tachycardia or tachyarrhythmia that originates in one of the ventricles. This is a potentially life-threatening arrhythmia, as it can lead to ventricular fibrillation and sudden death.

One type of arrhythmia, atrial fibrillation, occurs when normal electrical impulses (e.g., another example of intracardiac signals) generated by the sinoatrial node are overwhelmed by disorganized electrical impulses (e.g., signal disturbances) originating in the atria and pulmonary veins that cause irregular impulses to be transmitted to the ventricles. Thereby producing an irregular heartbeat and may last from minutes to weeks, or even years. Atrial Fibrillation (AF) is generally a chronic condition that slightly increases the risk of death usually caused by stroke. The first line treatment for AF is a medication that can slow or normalize heart rhythm. In addition, people with AF are often given anticoagulants to prevent them from being at risk for stroke. The use of such anticoagulants is associated with its own risk of internal bleeding. For some patients, drug treatment is insufficient and their AF is considered drug refractory, i.e. no treatment is available with standard drug intervention. Synchronous electrical cardioversion may also be used to restore AF to a normal heart rhythm. Alternatively, AF patients are treated by catheter ablation.

Catheter ablation-based treatments may include mapping electrical properties of cardiac tissue (particularly endocardium and heart volume), and selectively ablating the cardiac tissue by applying energy. Cardiac mapping includes creating maps of electrical potentials, e.g., wave propagation along cardiac tissue (e.g., voltage maps) or arrival times to various tissue sites (e.g., local time activation (LAT) maps). Cardiac mapping can be used to detect local cardiac tissue dysfunction. Ablation, such as ablation based on cardiac mapping, may stop or alter the propagation of unwanted electrical signals from one portion of the heart to another.

The ablation process damages the unwanted electrical pathways by forming electrically non-conductive lesions. Various forms of energy delivery for forming lesions have been disclosed and include the use of microwaves, lasers and more commonly radio frequency energy to form conduction blocks along the walls of cardiac tissue. In a two-step procedure of mapping and then ablation, electrical activity at various points in the heart is typically sensed and measured by inserting a catheter (e.g., catheter 105) containing one or more electrical sensors (e.g., at least one ablation electrode 134 of catheter 105) into the heart (e.g., heart 120) and acquiring data at the various points. This data (e.g., biometric data including intracardiac signals) is then utilized to select an endocardial target area for which ablation is to be performed. Due to the use of the automated encoders employed by the exemplary system 100 (e.g., medical device apparatus), this data is more accurate and better able to support the selection of endocardial target areas for ablation than the underlying electrical signals of the ECG, including signal interference, signal artifacts, and signal noise. Signal interference, signal artifacts, and signal noise may be collectively referred to herein as artifacts. Examples of artifacts include, but are not limited to, power noise (e.g., electrostatic and electromagnetic coupling between the circuit and 50 or 60Hz power supply lines), Fluro noise (e.g., fluorescent lamps), contact noise (e.g., collisions between conduit electrodes), and deflection noise (e.g., discharge of static electricity during conduit deflection).

Cardiac ablation and other cardiac electrophysiology protocols become increasingly complex as clinicians treat increasingly challenging conditions such as atrial fibrillation and ventricular tachycardia. Treatment of complex arrhythmias currently relies on the use of three-dimensional (3D) mapping systems in order to reconstruct the anatomy of the heart chamber of interest. In this regard, the automated encoder employed by the exemplary system 100 herein (e.g., medical device apparatus) provides underlying output signals such that an improved 3D map and/or ECG for treating a cardiac disorder can be generated.

For example, cardiologists rely on software, such as that produced by Biosense Webster, Inc. (Diamond Bar, Calif.)

A Complex Fractionated Atrial Electrogram (CFAE) module of the 33D mapping system to generate and analyze an intracardiac Electrocardiography (EGM). The automated encoder of exemplary system 100 (e.g., medical device apparatus) enhances the software to generate and analyze improved intracardiac Electrograms (EGMs) so that ablation points may be determined for treating a range of cardiac disorders, including atypical atrial flutter and ventricular tachycardia.

The improved 3D map supported by the automated encoder may provide a plurality of pieces of information about the electrophysiological properties of tissue that represent these challenging arrhythmic anatomical and functional substrates.

Cardiomyopathies of different etiologies (hypoxia, Dilated (DCM), Hypertrophic Cardiomyopathy (HCM), Arrhythmogenic Right Ventricular Dysplasia (ARVD), left ventricular incompetence (LVNC), etc.) have identifiable substrates characterized by regions of unhealthy tissue surrounded by normal functioning cardiomyocytes.

Abnormal tissue is typically characterized by low pressure EGMs. However, initial clinical experience in intracardiac-epicardial mapping indicates that the low voltage region is not always present as the only arrhythmogenic mechanism in such patients. Indeed, low or medium pressure regions may exhibit EGM fragmentation and prolonged activity during sinus rhythms corresponding to critical isthmuses identified during persistent and organized ventricular arrhythmias, e.g., applicable only to intolerant ventricular tachycardia. In addition, in many cases, EGM fragmentation and prolonged activity were observed in regions exhibiting normal or near normal voltage amplitudes (> 1-1.5 mV). While a region can then be evaluated in terms of voltage amplitude, they cannot be considered normal in terms of intracardiac signals, and thus represent a true arrhythmogenic substrate. The 3D mapping can localize the proarrhythmic matrix on the endocardial and/or epicardial layers of the right/left ventricle, which can vary in distribution according to the spread of the primary disease.

The stroma associated with these cardiac conditions is associated with the presence of fragmented and prolonged EGMs in the endocardial and/or epicardial layers of the ventricular chambers (right and left). 3D mapping systems, such as

3, the potential proarrhythmic substrates of cardiomyopathy can be localized in terms of abnormal EGM detection.

Electrode catheters (e.g., catheter 105) are used in medical practice. The electrode catheter is used to stimulate and map electrical activity in the heart, and to ablate sites of abnormal electrical activity. In use, an electrode catheter is inserted into a main vein or artery, such as the femoral artery, and then introduced into the heart chamber of interest. A typical ablation procedure involves inserting a catheter having at least one electrode at its distal end into a heart chamber. A reference electrode is provided, typically taped to the patient's skin, or may be provided using a second catheter placed in or near the heart. A radio frequency (RE) current is applied to the tip electrode of the ablation catheter, and the current flows to the reference electrode through the medium (i.e., blood and tissue) surrounding it. The distribution of the current depends on the amount of contact of the electrode surface with the tissue compared to blood, which has a higher conductivity than the tissue. Heating of the tissue occurs due to the electrical resistance of the tissue. The tissue is heated sufficiently to cause cell destruction in the heart tissue, resulting in the formation of non-conductive foci within the heart tissue. During this process, heating of the electrode also occurs due to conduction from the heated tissue to the electrode itself. If the electrode temperature becomes high enough, possibly above 60 degrees celsius, a thin transparent coating of dehydrated blood proteins can form on the surface of the electrode. If the temperature continues to rise, the dehydrated layer may become thicker, resulting in blood clotting on the electrode surface. Because dehydrated biological material has a higher electrical resistance than endocardial tissue, the impedance to the flow of electrical energy into the tissue also increases. If the impedance increases sufficiently, an impedance rise occurs and the catheter must be removed from the body and the tip electrode cleaned.

Treatment of cardiac disorders, such as arrhythmias, typically requires obtaining a detailed map of the cardiac tissue, chambers, veins, arteries, and/or electrical pathways. For example, a prerequisite for successful catheter ablation is that the cause of the arrhythmia is located precisely in the heart chamber. Such localization can be accomplished via electrophysiological studies during which electrical potentials are spatially resolved detected with a mapping catheter introduced into the heart chamber. This electrophysiological study (so-called electroanatomical mapping) thus provides 3D mapping data that can be displayed on a monitor. In many cases, mapping and treatment functions (e.g., ablation) are provided by a single catheter or a set of catheters, such that the mapping catheter also operates simultaneously as a treatment (e.g., ablation) catheter. In this case, the autoencoder may be stored and executed directly by catheter 105.

Mapping of cardiac regions, such as electrical pathways of cardiac regions, tissue, veins, arteries, and/or the heart (e.g., 120), can result in identifying problem regions such as scar tissue, arrhythmia sources (e.g., electrical rotors), healthy regions, and the like. The cardiac region may be mapped such that a visual rendering of the mapped cardiac region is provided using a display, as further disclosed herein. Additionally, cardiac mapping may include mapping based on one or more modalities such as, but not limited to, Local Activation Time (LAT), electrical activity, topology, bipolar mapping, dominant frequency, or impedance. Data corresponding to the plurality of modalities may be captured using a catheter inserted into a patient and may be provided for rendering at the same time or at different times based on corresponding settings and/or preferences of a medical professional.

Cardiac mapping may be accomplished using one or more techniques. As an example of the first technique, cardiac mapping may be achieved by sensing electrical properties (e.g., LAT) of cardiac tissue from precise locations within the heart. Corresponding data may be acquired by one or more catheters advanced into the heart using a catheter having an electrical sensor and a position sensor in its distal tip. Specifically, for example, the location and electrical activity may be initially measured at about 10 to about 20 points on the inner surface of the heart. These data points may generally be sufficient to generate a satisfactory quality preliminary reconstruction or map of the cardiac surface. The preliminary map may be combined with data taken from additional points to produce a more comprehensive map of cardiac electrical activity. In a clinical setting, it is not uncommon to accumulate data at 100 or more sites to generate a detailed and comprehensive map of the heart chamber electrical activity. The detailed map generated can then be used as a basis for deciding on the course of therapeutic action, such as tissue ablation, to alter the propagation of cardiac electrical activity and restore a normal heart rhythm.

As shown in fig. 1, medical professional 115 may insert shaft 137 through sheath 136 while manipulating the distal end of shaft 137 using manipulator 138 near the proximal end of catheter 105 and/or deflecting from sheath 136. As shown in inset 140, the catheter 105 may be fitted at the distal end of the shaft 137. Catheter 105 may be inserted through sheath 136 in a collapsed state and may then be deployed within heart 120. As further described herein, the catheter 105 may include at least one ablation electrode 134 and a catheter needle.

According to an embodiment, the catheter 105 may be configured to ablate a tissue region of a heart chamber of the heart 120. The inset 150 shows the catheter 105 in an enlarged view within a heart chamber of the heart 120. As shown, the catheter 105 may include at least one ablation electrode 134 coupled to the body of the catheter. According to other embodiments, multiple elements may be connected via an elongated strip forming the shape of the conduit 105. One or more other elements (not shown) may be provided, which may be any element configured to ablate or obtain biometric data, and may be an electrode, a transducer, or one or more other elements.

In accordance with embodiments disclosed herein, an ablation electrode, such as at least one ablation electrode 134, may be configured to provide energy to a tissue region of an internal body organ, such as the heart 120. The energy may be thermal energy and may cause damage to the tissue region starting at a surface of the tissue region and extending into a thickness of the tissue region.

According to embodiments disclosed herein, the biometric data may include one or more of LAT, electrical activity, topology, bipolar maps, dominant frequency, impedance, and the like. The LAT may be a point in time corresponding to threshold activity of local activation calculated based on a normalized initial starting point. The electrical activity may be any suitable electrical signal that may be measured based on one or more thresholds and may be sensed and/or enhanced based on a signal-to-noise ratio and/or other filters. The topology may correspond to the physical structure of a body part or a portion of a body part, and may correspond to a change in the physical structure relative to a different portion of the body part or relative to a different body part. The dominant frequency may be a frequency or range of frequencies that are ubiquitous at one part of the body part and may be different in different parts of the same body part. For example, the dominant frequency of the pulmonary veins of a heart may be different from the dominant frequency of the right atrium of the same heart. The impedance may be a measure of the resistance at a given region of the body part.

As shown in FIG. 1, the probe 110 and catheter 105 may be connected to a console 160. The console 160 may include a computing device 161 employing an autoencoder as described herein. According to one embodiment, the console 160 and/or computing device 161 includes at least a processor that executes computer instructions for the autoencoder described herein and a memory that stores instructions for execution by the processor.

Computing device 161 may be any computing device, such as a general purpose computer, including software and/or hardware, with suitable front end and interface circuitry 162 for transmitting signals to and receiving signals from catheter 105, as well as for controlling other components of system 100. Computing device 161 may include real-time noise reduction circuitry, typically configured as a Field Programmable Gate Array (FPGA), followed by an analog-to-digital (a/D) electrocardiogram or Electromyogram (EMG) signal conversion integrated circuit. The computing device 161 can transfer signals from the a/D ECG or EMG circuitry to another processor and/or can be programmed to perform one or more of the functions disclosed herein.

For example, the one or more functions include receiving an input intracardiac signal, encoding the input intracardiac signal with an intracardiac data set to produce a potential representation, and decoding the potential representation to produce an output intracardiac signal. The front end and interface circuitry 162 includes an input/output (I/O) communication interface that enables the console 160 to receive signals from and/or transmit signals to the at least one ablation electrode 134.

In some embodiments, the computing device 161 may be further configured to receive biometric data, such as electrical activity, and determine whether a given tissue region is conductive. According to one embodiment, the computing device 161 may be located external to the console 160, and may be located, for example, in a catheter, in an external device, in a mobile device, in a cloud-based device, or may be a stand-alone processor.

As noted above, computing device 161 may comprise a general purpose computer that is programmable with software to perform the functions of the autoencoder described herein. The software may be downloaded to the general purpose computer in electronic form, over a network, for example, or it may alternatively or additionally be provided and/or stored on a non-transitory tangible medium, such as magnetic, optical, or electronic memory (e.g., any suitable volatile and/or non-volatile memory, such as random access memory or hard disk drive). The exemplary configuration shown in fig. 1 may be modified to implement the embodiments disclosed herein. The disclosed embodiments of the invention may be similarly applied using other system components and arrangements. Additionally, the system 100 may include additional components, such as elements for sensing electrical activity, wired or wireless connectors, processing and display devices, and the like.

According to one embodiment, the display 165 is connected to a computing device 161. During the procedure, computing device 161 may facilitate presenting the body part rendering to medical professional 115 on display 165 and storing data representing the body part rendering in memory. In thatIn some embodiments, medical professional 115 may be able to manipulate the body part rendering using one or more input devices (such as a touchpad, mouse, keyboard, gesture recognition device, etc.). For example, an input device may be used to change the position of catheter 105 so that the rendering is updated. In an alternative embodiment, display 165 may include a touch screen that may be configured to accept input from medical professional 115 in addition to presenting body part renderings. The display 165 may be located at the same location or at a remote location, such as a separate hospital or in a separate healthcare provider network. Additionally, the system 100 may be part of a surgical system configured to obtain anatomical and electrical measurements of a patient organ (such as the heart 120) and perform a cardiac ablation procedure. An example of such a surgical system is marketed by Biosense Webster

Provided is a system.

The console 160 can be connected by a cable to a body surface electrode, which can include an adhesive skin patch that is attached to the patient 125. The processor, in conjunction with the current tracking module, may determine positional coordinates of the catheter 105 within a body part of the patient 125 (e.g., the heart 120). The location coordinates may be based on impedance or electromagnetic fields measured between the body surface electrodes and the electrodes or other electromagnetic components of the catheter 105 (e.g., the at least one ablation electrode 134). Additionally or alternatively, the placemat may be located on the surface of the bed 130 and may be separate from the bed 130.

The system 100 may also and optionally use ultrasound, Computed Tomography (CT), Magnetic Resonance Imaging (MRI), or other medical imaging techniques known in the art to obtain biometric data, such as anatomical measurements of the heart 120. The system 100 may use a catheter or other sensor that measures electrical characteristics of the heart 120 to obtain an ECG or electrical measurement. The biometric data, including the anatomical and electrical measurements, may then be stored in a non-transitory tangible medium of the console 160. The biometric data may be transmitted from a non-transitory tangible medium to the computing device 161. Alternatively or in addition, the biometric data may be transmitted to a server 1760, which may be local or remote, using a network 1762.

According to one or more embodiments, a catheter including a position sensor may be used to determine a trajectory of points on the surface of the heart. These trajectories can be used to infer motion characteristics, such as the contractility of tissue. When trajectory information is sampled at a sufficient number of points in the heart 120, a map depicting such motion characteristics may be constructed.

The electrical activity at a point in the heart 120 may be measured by generally advancing the catheter 105 containing an electrical sensor at or near its distal tip (at least one ablation electrode 134) to the point in the heart 120, contacting the tissue with the sensor and acquiring data at that point. One disadvantage of using a catheter 105 containing only a single distal tip electrode to map a heart chamber is the long time required to acquire data point-by-point over the necessary number of points required for a detailed map of the chamber population. Accordingly, multi-electrode catheters have been developed to measure electrical activity at multiple points in the heart chamber simultaneously.

The multi-electrode catheter may be implemented using any suitable shape, such as a linear catheter with multiple electrodes, a balloon catheter including electrodes dispersed over multiple ridges that shape the balloon, a lasso or loop catheter with multiple electrodes, or any other suitable shape. The linear catheter may be fully or partially elastic such that it may twist, bend, or otherwise change its shape based on the received signals and/or based on the application of an external force (e.g., cardiac tissue) on the linear catheter. The balloon catheter may be designed such that its electrodes may remain in intimate contact with the endocardial surface when it is deployed into the patient. For example, the balloon catheter may be inserted into a lumen, such as a Pulmonary Vein (PV). The balloon catheter may be inserted into the PV in a deflated state such that the balloon catheter does not occupy its maximum volume when inserted into the PV. The balloon catheter may be inflated inside the PV such that those electrodes on the balloon catheter are in contact with the entire circular segment of the PV. Such contact with the entire circular portion of the PV or any other lumen may enable effective mapping and/or ablation.

According to one example, a multi-electrode catheter may be advanced into a chamber of the heart 120. Anteroposterior (AP) and lateral fluorescence maps can be obtained to establish the position and orientation of each electrode. The EGM may be recorded by each of the electrodes in contact with the surface of the heart relative to a time reference, such as from a P-wave in a sinus rhythm from a body surface ECG. As further disclosed herein, the system can distinguish between those electrodes that record electrical activity and those electrodes that do not record electrical activity due to not being in close proximity to the endocardial wall. After recording the initial EGM, the catheter may be repositioned, and the fluorogram and EGM may be recorded again. The electrical map may then be constructed according to iterations of the above process.

According to one example, a cardiac map may be generated based on the detection of the intracardiac potential field. Non-contact techniques for simultaneously acquiring large amounts of electrical information of the heart may be implemented. For example, a catheter having a distal end portion may be provided with a series of sensor electrodes distributed over its surface and connected to insulated electrical conductors for connection to a signal sensing and processing device. The end portion may be sized and shaped such that the electrode is substantially spaced from the wall of the heart chamber. The intracardiac potential field may be detected during a single heartbeat. According to one example, the sensor electrodes may be distributed over a series of circumferences lying in planes spaced apart from each other. These planes may be perpendicular to the long axis of the end portion of the catheter. At least two additional electrodes may be provided adjacent at the ends of the long axis of the end. As a more specific example, the catheter may include four circumferences, with eight electrodes equiangularly spaced on each circumference. Thus, in this implementation, the catheter may include at least 34 electrodes (32 circumferential electrodes and 2 end electrodes).

According to another example, electrophysiology cardiac mapping systems and techniques based on non-contact and non-expanding multi-electrode catheters may be implemented. EGMs may be obtained through a catheter having multiple electrodes (e.g., 42 to 122 electrodes). According to this implementation, knowledge of the relative geometry of the probe and endocardium may be obtained, such as by a separate imaging modality (such as transesophageal echocardiography). After independent imaging, the non-contact electrodes can be used to measure cardiac surface potentials and construct a map therefrom. The technique may include the following steps (after the separate imaging step): (a) measuring the electrical potentials with a plurality of electrodes disposed on a probe positioned on the heart 120; (b) determining a geometric relationship of the probe surface and the endocardial surface; (c) generating a coefficient matrix representing a geometric relationship of the probe surface and the endocardial surface; and (d) determining the endocardial potential based on the electrode potential and the coefficient matrix.

According to another example, techniques and devices for mapping electrical potential distributions of a heart chamber may be implemented. An intracardiac multi-electrode mapping catheter assembly may be inserted into a patient's heart 120. The mapping catheter assembly may include a multi-electrode array with an integral reference electrode, or preferably, a mating reference catheter. The electrodes may be deployed in a substantially spherical array. The electrode array may be spatially referenced to a point on the endocardial surface by a reference electrode or by a reference catheter in contact with the endocardial surface. Preferred electrode array catheters can carry a plurality of individual electrode sites (e.g., at least 24). In addition, the exemplary technique can be implemented by knowing the location of each of the electrode sites on the array and knowing the geometry of the heart. These locations are preferably determined by the technique of impedance plethysmography.

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

According to another example, another catheter for mapping electrophysiological activity within the heart may be implemented. The catheter body may include a distal tip adapted to deliver stimulation pulses for cardiac pacing or an ablation electrode for ablating tissue in contact with the tip. The catheter may also include at least one pair of orthogonal electrodes to generate a difference signal indicative of local cardiac electrical activity proximate the orthogonal electrodes.

According to another example, a process for measuring electrophysiological data in a heart chamber may be implemented. The method may include, in part, positioning a set of active and passive electrodes into the heart 120, supplying current to the active electrodes, thereby generating an electric field in a heart chamber, and measuring the electric field at the location of the passive electrodes. The passive electrodes are included in an array positioned on an inflatable balloon of the balloon catheter. In a preferred embodiment, the array is said to have 60 to 64 electrodes.

According to another example, cardiac mapping may be accomplished using one or more ultrasound transducers. An ultrasound transducer may be inserted into a patient's heart 120 and a plurality of ultrasound slices (e.g., two-dimensional or three-dimensional slices) may be collected at various locations and orientations within the heart 120. The position and orientation of a given ultrasound transducer may be known, and the collected ultrasound slices may be stored so that they may be displayed at a later time. One or more ultrasound slices corresponding to the position of the probe (e.g., treatment catheter) over a period of time may be displayed, and the probe may be overlaid on the one or more ultrasound slices.

According to other examples, the body patch and/or the body surface electrodes may be positioned on or near the patient's body. A catheter having one or more electrodes may be positioned within a patient's body (e.g., within the patient's heart 120), and the position of the catheter may be determined by the system based on signals transmitted and received between the one or more electrodes of the catheter and the body patch and/or body surface electrodes. In addition, the catheter electrodes may sense biometric data (e.g., LAT values) from within the patient's body (e.g., within the heart 120). The biometric data may be associated with the determined position of the catheter such that a rendering of a body part of the patient (e.g., the heart 120) may be displayed and the biometric data overlaid on the body shape may be displayed.

Referring now to fig. 2, a block diagram of an exemplary system 200 for remotely monitoring and transmitting biometric data (i.e., patient biometric, patient data, or patient biometric data) is shown. In the example shown in fig. 2, the system 200 includes a monitoring and processing device 202 (i.e., a patient data monitoring and processing device) associated with a patient 204, a local computing apparatus 206, a remote computing system 208, a first network 210, and a second network 211. In accordance with one or more embodiments, the monitoring and processing device 202 may be an example of the catheter 105 of fig. 1, the patient 204 may be an example of the patient 125 of fig. 1, and the local computing apparatus 206 may be an example of the console 160 of fig. 1.

The monitoring and processing device 202 includes a patient biometric sensor 212, a processor 214, a User Input (UI) sensor 216, a memory 218, and a transmitter-receiver (i.e., transceiver) 222. In operation, the monitoring and processing device 202 acquires biometric data (e.g., electrical signals, blood pressure, temperature, blood glucose level, or other biometric data) of the patient 204 and/or receives at least a portion of the biometric data representing any acquired patient biometrics and additional information associated with the acquired patient biometrics from one or more other patient biometric monitoring and processing devices. The additional information may be, for example, diagnostic information and/or additional information obtained from an additional device, such as a wearable device.

The monitoring and processing device 202 may employ the automated encoder described herein to process data, including acquired biometric data and any biometric data received from one or more other patient biometric monitoring and processing devices. For example, in this regard, in processing data, the autoencoder may include a neural network for learning potential representations (or data encodings) from biometric data in an unsupervised manner. Furthermore, the auto-encoder can learn to detect particular data by training the neural network to ignore signal interference, signal artifacts, and signal noise by considering clean data sets without pre-programming with particular rules.

The monitoring and processing device 202 may continuously or periodically monitor, store, process, and transmit any number of various patient biometrics (e.g., acquired biometric data) via the network 210. As described herein, examples of patient biometrics include electrical signals (e.g., ECG signals and brain biometrics), blood pressure data, blood glucose data, and temperature data. Patient biometrics can be monitored and communicated in order to treat any number of a variety of diseases, such as cardiovascular diseases (e.g., arrhythmias, cardiomyopathy, and coronary artery disease) and autoimmune diseases (e.g., type I and type II diabetes).

Patient biometric sensor 212 may include, for example, one or more transducers configured to convert one or more environmental conditions into electrical signals such that different types of biometric data are acquired. For example, patient biometric sensor 212 may include one or more electrodes configured to acquire electrical signals (e.g., cardiac signals, brain signals, or other bioelectric signals), a temperature sensor (e.g., a thermocouple), a blood pressure sensor, a blood glucose sensor, a blood oxygen sensor, a pH sensor, an accelerometer, and a microphone.

As described in greater detail herein, the monitoring and processing device 202 may be an ECG monitor for monitoring an ECG signal of a heart (e.g., the heart 120 of fig. 1). In this regard, the patient biometric sensor 212 of the ECG monitor may include one or more electrodes (e.g., electrodes of the catheter 105 of fig. 1) for acquiring ECG signals. The ECG signal can be used to treat various cardiovascular diseases.

In another example, the monitoring and processing device 202 may be a continuous blood glucose monitor (CGM) for continuously monitoring the blood glucose level of a patient for continuous treatment of various diseases, such as type I and type II diabetes. In this regard, the patient biometric sensor 212 of the CGM may include a subcutaneously disposed electrode (e.g., the electrode of catheter 105 of fig. 1) that may monitor blood glucose levels from the patient's interstitial fluid. The CGM may be a component of, for example, a closed loop system, where blood glucose data is sent to an insulin pump for calculating the delivery of insulin without user intervention.

The processor 214 may be configured to receive, process, and manage biometric data acquired by the patient biometric sensor 212, and to transfer the biometric data to the memory 218 via the transceiver 222 for storage and/or across the network 210. Data from one or more other monitoring and processing devices 202 may also be received by processor 214 via transceiver 222, as described in more detail herein. As described in greater detail herein, the processor 214 may be configured to selectively respond to different tap patterns (e.g., single or double tap) received from the UI sensor 216 (e.g., a capacitive sensor therein) such that different tasks (e.g., acquisition, storage, or transmission of data) of the patch may be activated based on the detected patterns. In some implementations, the processor 214 may generate audible feedback with respect to detecting the gesture.

The UI sensor 216 may include, for example, a piezoelectric sensor or a capacitive sensor configured to receive user input such as a tap or touch. For example, in response to the patient 204 tapping or touching a surface of the monitoring and processing device 202, the UI sensor 216 may be controlled to achieve a capacitive coupling. Gesture recognition may be accomplished via any of a variety of capacitance types, such as resistive-capacitive, surface-capacitive, projected-capacitive, surface acoustic wave, piezoelectric, and infrared touch. The capacitive sensor may be disposed at a small area or over the length of the surface such that a tap or touch of the surface activates the monitoring device.

The memory 218 is any non-transitory tangible medium, such as magnetic memory, optical memory, or electronic memory (e.g., any suitable volatile memory and/or non-volatile memory, such as random access memory or a hard drive). According to one or more embodiments, memory 218 may store processor-executable code, software, or instructions for training algorithms and autoencoders.

The transceiver 222 may include a separate transmitter and a separate receiver. Alternatively, the transceiver 222 may include a transmitter and a receiver integrated into a single device.

According to one embodiment, the monitoring and processing device 202 may be a (e.g., subcutaneously implantable) device within the body of the patient 204. The monitoring and processing device 202 may be inserted into the body of the patient 204 via any suitable means, including oral injection, surgical procedure insertion via veins or arteries, endoscopic or laparoscopic procedures.

According to one embodiment, the monitoring and processing device 202 may be a device external to the patient 204. For example, as described in more detail herein, the monitoring and processing device 202 may include an attachable patch (e.g., that attaches to the skin of the patient). The monitoring and processing device 202 may also include a catheter having one or more electrodes, a probe, a blood pressure cuff, a weight scale, a bracelet or smart watch biometric tracker, a glucose monitor, a Continuous Positive Airway Pressure (CPAP) machine, or virtually any device that can provide input related to the health or biometric identification of a patient.

According to one embodiment, the monitoring and processing device 202 may include both patient-internal components and patient-external components.

Although fig. 2 shows a single monitoring and processing device 202, an exemplary system may include multiple patient biometric monitoring and processing devices. For example, the monitoring and processing device 202 may communicate with one or more other patient biometric monitoring and processing devices. Additionally or alternatively, one or more other patient biometric monitoring and processing devices may be in communication with the network 210 and other components of the system 200.

The local computing device 206 and/or the remote computing system 208, along with the monitoring and processing device 202, may be any combination of software and/or hardware that individually or collectively store, execute, and implement an autoencoder and its functions. Further, the local computing device 206 and/or the remote computing system 208, along with the monitoring and processing equipment 202, may 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. The local computing device 206 and/or the remote computing system 208, along with the monitoring and processing apparatus 202, may be readily scalable, extensible, and modular, with the ability to change to different services or reconfigure some features independently of other features.

According to one embodiment, the local computing device 206 and the remote computing system 208, along with the monitoring and processing apparatus 202, may include at least a processor that executes computer instructions for an auto-encoder and a memory that stores instructions for execution by the processor.

The local computing device 206 of the system 200 is in communication with the monitoring and processing apparatus 202 and may be configured to act as a gateway to the remote computing system 208 through the second network 211. For example, the local computing device 206 may be a smart phone, smart watch, tablet, or other portable smart device configured to communicate with other devices via the network 211. Alternatively, the local computing device 206 may be a fixed or stand-alone device, such as a fixed base station including, for example, modem and/or router capabilities, a desktop or laptop computer that uses executable programs to transfer information between the processing apparatus 202 and the remote computing system 208 via the PC's radio module, or a USB dongle. Biometric data may be communicated between the local computing device 206 and the monitoring and processing device 202 via a short-range wireless network 210, such as a Local Area Network (LAN) (e.g., a Personal Area Network (PAN)), using short-range wireless technology standards (e.g., bluetooth, Wi-Fi, ZigBee, Z-wave, and other short-range wireless standards). In some embodiments, the local computing device 206 may also be configured to display the acquired patient electrical signals and information associated with the acquired patient electrical signals, as described in more detail herein.

In some embodiments, the remote computing system 208 may be configured to receive at least one of the monitored patient biometric and information associated with the monitored patient via the network 211 as a remote network. For example, if the local computing device 206 is a mobile telephone, the network 211 may be a wireless cellular network and information may be communicated between the local computing device 206 and the remote computing system 208 via a wireless technology standard, such as any of the wireless technologies described above. As described in greater detail herein, the remote computing system 208 may be configured to provide (e.g., visually display and/or audibly provide) at least one of patient biometric and related information to a medical professional, doctor, or healthcare professional.

In fig. 2, network 210 is an example of a short-range network, such as a Local Area Network (LAN) or a Personal Area Network (PAN). Information may be sent between the monitoring and processing device 202 and the local computing arrangement 206 via the short-range 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), ultraband, Zigbee, or Infrared (IR).

The network 211 may be a wired network, a wireless network, or a network including one or more wired and wireless networks, such as an intranet, a Local Area Network (LAN), a Wide Area Network (WAN), a Metropolitan Area Network (MAN), a direct or serial connection, a cellular telephone network, or any other network or medium capable of facilitating communication between the local computing device 206 and the remote computing system 208. Information may be transmitted via network 211 using any of a variety of long-range wireless communication protocols (e.g., TCP/IP, HTTP, 3G, 4G/LTE, or 5G/new radio). The wired connection may be implemented using ethernet, Universal Serial Bus (USB), RJ-11, or any other wired connection known in the art. The wireless connection may be implemented using Wi-Fi, WiMAX, and Bluetooth, infrared, cellular networks, satellite, or any other wireless connection method. In addition, several networks may operate independently or in communication with each other to facilitate communication in network 211. In some cases, remote computing system 208 may be implemented as a physical server on network 211. In other cases, the remote computing system 208 may be implemented as a public cloud computing provider (e.g., Amazon Web Services) on the network 211

) The virtual server of (1).

FIG. 3 illustrates an artificial intelligence system 300 in accordance with one or more embodiments. The artificial intelligence system 300 can include data 310, a machine 320, a model 330, a plurality of results 340, and underlying hardware 350. FIG. 4 illustrates a method 400 performed in the artificial intelligence system of FIG. 3. For ease of understanding, fig. 3 and 4 are described with reference to fig. 2.

In general, the artificial intelligence system 300 operates the method 400 by using the data 310 to train a machine 320 (e.g., the local computing device 206 of FIG. 2) while building a model 330 to achieve multiple results 340 (which can be predicted). In this configuration, the artificial intelligence system 300 may operate with respect to hardware 350 (e.g., the monitoring and processing device 202 of FIG. 2) to train the machine 320, build the model 330, and predict results using algorithms. These algorithms may be used to solve the trained model 330 and predict the results 340 associated with the hardware 350. These algorithms can generally be divided into classification, regression and clustering algorithms.

At block 410, the method 400 may include collecting data 310 from the hardware 350. The machine 320 may operate as and/or be associated with a controller or data collection element associated with the hardware 350. The data 310 (e.g., biometric data that may be derived from the monitoring and processing device 202 of fig. 2) may be associated with hardware 350. For example, data 310 may be data being generated or output data associated with hardware 350. The data 310 may also include data currently collected from the hardware 350, historical data, or other data. For example, the data 310 may include measurements during a surgical procedure and may be associated with the results of the surgical procedure. For example, the temperature of the heart (e.g., the heart of patient 204) may be collected and correlated with the results of a cardiac procedure.

At block 420, the method 400 includes training the machine 320, such as with respect to the hardware 350. The training may include analysis and correlation of the data 310 collected in block 410. For example, with respect to the heart, the temperature and result data 310 may be trained to determine whether there is a correlation or link between the temperature of the heart (e.g., the heart of the patient 204) and the results during a cardiac procedure.

At block 430, the method 400 may include building the model 330 based on the data 310 associated with the hardware 350. Building the model 330 may include physical hardware or software modeling, algorithmic modeling, and the like. The modeling may attempt to represent the collected and trained data 310. According to one embodiment, the model 330 may be configured to model the operation of the hardware 350 and model the data 310 collected from the hardware 350 in order to predict the results achieved by the hardware 350. In accordance with one or more embodiments, the model 330 can distinguish between ventricular far-field and atrial-based activation relative to an auto-encoder, and generate distinct maps for atrial and ventricular activation.

At block 440, the method 400 may include predicting a plurality of outcomes 340 of the model 330 associated with the hardware 350. Such prediction of the plurality of results 340 may be based on the trained model 330. For example, to increase the understanding of the present disclosure, with respect to the heart, if the temperature during the procedure is between 36.5 degrees celsius and 37.89 degrees celsius (i.e., 97.7 degrees fahrenheit and 100.2 degrees fahrenheit), a positive result is produced from the cardiac procedure, which can be predicted in a given procedure based on the temperature of the heart during the cardiac procedure. Thus, using the predicted result 340, the hardware 350 may be configured to provide some desired result 340 from the hardware 350.

Turning now to fig. 5, an example of a neural network 500 is shown in accordance with one or more embodiments. The neural network 500 may operate as an implementation of an autoencoder. The neural network 500 may be implemented in hardware, such as the machine 320 (e.g., the local computing device 206 of fig. 2) and/or the hardware 350 (e.g., the monitoring and processing apparatus 202 of fig. 2). A neural network is a network or circuit of neurons, or in a modern sense, an artificial neural network consisting of artificial neurons or nodes.

For example, an ANN may involve a network of processing elements (artificial neurons) that may exhibit complex global behavior determined by the connections between the processing elements and the element parameters. These connections of the network or circuit of neurons can be modeled as weights. Positive weights may reflect excitatory connections, while negative values may indicate inhibitory connections. The inputs may be modified by weights and summed using linear combinations. The activation function may control the amplitude of the output. For example, an acceptable output range is typically between 0 and 1, or the range may be between-1 and 1.

In most cases, an ANN is an adaptive system that changes its structure based on external or internal information flowing through the network. In more practical terms, neural networks are non-linear statistical data modeling or decision tools that can be used to model complex relationships between inputs and outputs or to find patterns in data. Thus, the ANN can be used for predictive modeling and adaptive control applications while training via a data set. Empirically generated self-learning can occur within the ANN, which can draw conclusions from complex and seemingly unrelated sets of information. Artificial neural network models are useful in that they can be used to infer a function from observations and can also be used to use that function. 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 the observed data. Learning in neural networks is particularly useful in applications where the complexity of the data or tasks makes it impractical to manually design such functions.

Neural networks are used in different fields. The tasks applied by ANN tend to fall into the following broad categories: functional approximation or regression analysis, including time series prediction and modeling; classification, including pattern and sequence recognition, novelty detection, and order decision; and data processing, including filtering, clustering, blind signal separation, and compression.

Application areas for ANN may include nonlinear system recognition and control (vehicle control, process control), game play and decision making (checkers, chat, competition), pattern recognition (radar systems, facial recognition, object recognition), sequence recognition (gesture, voice, handwritten text recognition), medical diagnostics, financial applications, data mining (or knowledge discovery in databases, "KDD"), visualization, and email spam screening. For example, a semantic feature map of user interest may be created from a picture trained for object recognition.

Turning now to fig. 6, a block diagram of a method 600 in accordance with one or more embodiments is shown. The method 600 depicts the operation of the neural network 500 (e.g., an autoencoder). Turning to fig. 5, in the neural network 500, an input layer 510 is represented by a plurality of inputs, such as 512 and 514. With respect to block 610 of fig. 6, the input layer 510 may receive the plurality of inputs (e.g., input intracardiac signals) as an initial operation. The plurality of inputs may be ultrasound signals, radio signals, audio signals, or two-dimensional pictures. More specifically, the plurality of inputs may be represented as input data (X), which is raw data recorded from the atrium. The desired information may be present in high frequency regions of the heart (e.g., atria), and the auto-encoder provides better construction of the input intracardiac signals. According to one or more embodiments, the plurality of inputs may be a combination of intracardiac and body surface ECGs (to remove far-field noise from intracardiac signals).

At block 620 of fig. 6, the neural network 500 may encode the input intracardiac signals with the intracardiac data set to produce a potential representation. The potential representation may include one or more intermediate images derived from the input intracardiac signals. According to one or more embodiments, the potential representation is generated by an element-wise activation function (e.g., an sigmoid function or a rectifying linear unit) of the auto-encoder that applies a weight matrix to the input intracardiac signals and adds a bias vector to the result. The weights and biases of the weight matrix and bias vector may be initialized randomly and then updated iteratively during training.

The intracardiac data set may be a training data set or clean data that includes a predetermined and approved signal (i.e., a clean example) that is free of interference, artifacts, and noise. In one embodiment, an expert medical professional, doctor, or the like may examine, edit to remove signal interference, signal artifacts, and signal noise, and approve each electrical signal of the intracardiac data set. In one embodiment, intracardiac data may have electrical signals on the order of thousands or greater, with template matching and blanking used to examine the signal morphology of each electrical signal. For example, with an intracardiac data set (e.g., a database of "clean versions" of intracardiac ECG signals), noise reduction of any IC-ECG artifacts may be performed. Given the number of electrical signals and the complexity of inspection, editing, and approval, the creation of an intracardiac data set may be considered a data training portion of a multi-step data manipulation by an automated encoder.

As shown in fig. 5,

inputs

512 and 514 are provided to hidden layer 530, depicted as including nodes 532, 534, 536, and 538. The encoding provides input intracardiac signals with reduced dimensionality. Dimensionality reduction is the process of reducing the number of random variables (of multiple inputs) under consideration by obtaining a set of primary variables. For example, the dimension reduction may be feature extraction that converts data (e.g., multiple inputs) from a high-dimensional space (e.g., more than 10 dimensions) to a low-dimensional space (e.g., 2-3 dimensions). Technical benefits of reducing dimensionality include reducing time and storage space for data, improving visualization of data, and improving interpretation of parameters for machine learning. The data transformation may be linear or non-linear. The operations of receiving (block 610) and encoding (block 620) may be considered a data preparation portion of a multi-step data manipulation by an auto-encoder.

According to one embodiment, data preparation may also include intracardiac electrocardiogram (IC-ECG) data collection of the atrium (the upper chamber through which blood enters the ventricles of the heart), where it is recorded simultaneously from the ventricles (the two lower chambers of the heart).

At block 630 of fig. 6, the neural network 500 may decode the potential representation to generate output intracardiac signals. In the case of IC-ECG, the output intracardiac signals may be ventricular far-field estimates. As shown in fig. 5, nodes 532, 534, 536, and 538 may combine to produce an output 552 in an output layer 550, where the output layer 550 may reconstruct the

inputs

512 and 514 over a reduced dimension but without signal interference, signal artifacts, and signal noise. The neural network 500 may perform processing via the hidden layer 530 of nodes 532, 534, 536, and 538 to exhibit complex global behavior determined by the connections between processing elements and element parameters. The target data of the output layer 550 may include a target data type one ventricular activity (Y1) and include a far-field attenuated (Y2) target data type two input data. The far field may cause problems with generating and navigating the 3D map (e.g., the ventricular far field may interfere with atrial activation). Thus, technical effects and benefits of an automatic encoder employing neural network 500 include improving the accuracy of 3D maps due to artifact (relative to the far field) removal.

In accordance with one or more embodiments, a model employing an autoencoder of the neural network 500 can distinguish between ventricular far-field and atrial-based activation, and generate distinct maps for atrial and ventricular activation.

According to one embodiment, the auto-encoder may be a de-noising auto-encoder to look up the mapping function (f, g) such that f (x) -Y1 and g (x) -Y2. In this regard, the task of the auto-encoder may be to learn a mapping from X to X by inputting some dimension reduction of X (e.g., constructing two neural networks (F, G) such that h ═ F (X) and X ═ G (h)). The dimension of h is less than the dimension of X. In a denoised auto-encoder, although similar in architecture, the denoised auto-encoder learns a mapping from X to Y, where Y is a denoised version of X.

Referring to fig. 7, a graphical depiction of a signal 700 is shown in accordance with one or more embodiments. As shown in signal 700, the ECG signal contains a P wave 710 (due to atrial depolarization), a QRS complex 720 (due to atrial repolarization and ventricular depolarization), and a T wave 730 (due to ventricular repolarization). The ECG signal is produced by contraction (depolarization) and relaxation (repolarization) of the atrial and ventricular muscles of the heart. To record the ECG signal, the electrodes may be placed at specific locations on the human body or may be positioned within the human body via a catheter. Artifacts (e.g., noise) are unwanted signals that are combined with electrical signals, such as ECG signals, and can sometimes hinder the diagnosis and/or treatment of cardiac conditions. Artifacts in the electrical signal may be baseline wander, power line interference, EMG noise, power line noise, and the like. That is, examples of artifacts include, but are not limited to, power noise (e.g., electrostatic and electromagnetic coupling between the circuit and 50 or 60Hz power supply lines), Fluro noise (e.g., fluorescent lamps), contact noise (e.g., collisions between conduit electrodes), and deflection noise (e.g., discharge of static electricity during conduit deflection).

Baseline wander may occur at the base axis (x-axis) position of the signal, which appears to "wander" or move up and down rather than straight. This can result in the entire signal being offset from its normal base. In an ECG signal, baseline drift may be caused by improper electrode contact (e.g., electrode-skin impedance), patient movement, and periodic movement (e.g., breathing).

Fig. 8 illustrates a graphical depiction of a signal 810 shown in a graph 800 in accordance with one or more embodiments. In this regard, the signal 800 is a typical ECG signal that is affected by the baseline shift 820. The frequency content of the baseline drift is in the range of 0.5 Hz. An increase in body movement during exercise or stress testing increases the frequency content of baseline drift. According to a specific implementation, a Finite Impulse Response (FIR) high-pass zero-phase forward backward filter with a cutoff frequency of 0.5Hz may be used to estimate and remove the baseline wander 820 in the ECG signal 810, given that the baseline signal is a low frequency signal.

The electromagnetic field caused by the power lines represents a common noise source in electrical signals such as the ECG, as well as any other bioelectric signals recorded from the patient's body. Such noise is characterized by sinusoidal interference, e.g., 50 or 60Hz, possibly accompanied by multiple harmonics. Such narrow-band noise makes analysis and interpretation of the ECG more difficult because the delineation of low amplitude waveforms becomes unreliable and can introduce spurious waveforms. When ECG signals superimpose low frequency ECG waves, such as P-wave 710 and T-wave 730, it may be desirable to remove power line interference from these ECG signals.

The presence of muscle noise can interfere with many electrical signal applications, such as ECG applications, as low amplitude waveforms can become blurred. Compared to baseline drift 820 and 50/60Hz interference, muscle noise is not removed by narrow band filtering, but presents a different filtering problem because the spectral content of muscle activity significantly overlaps with the spectral content of the PQRST complex 720. Since the ECG signal 810 is a repetitive signal, techniques are available to reduce muscle noise in a manner similar to the processing of evoked potentials. Fig. 9 illustrates a graphical depiction 900 of a signal 905 shown in accordance with one or more embodiments. In this regard, the signal 905 is an ECG signal disturbed by EMG noise 910.

Instruments for measuring electrical signals, such as ECG signals, typically detect electrical interference corresponding to line or trunk frequencies. Although the nominal setting is 50Hz or 60Hz, the line frequency in most countries can differ from these nominal values by a few percent.

Various techniques for removing electrical interference from electrical signals may be implemented. Several of these techniques use one or more low pass filters or notch filters. For example, a system for variable filtering of noise in an ECG signal may be implemented. The system may have a plurality of low pass filters including one filter having a 3dB point at, for example, about 50Hz and a second low pass filter having a 3dB point at, for example, about 5 Hz.

According to another example, a system for rejecting line frequency components of an electrical signal may be implemented by passing the signal through two notch filters connected in series. A system may be implemented having a notch filter that may have either or both low-pass coefficients and high-pass coefficients for removing line frequency components from the ECG signal. The system may also support removing burst noise and calculating heart rate from the notch filter output.

According to another example, a system having several units for removing interference may be implemented. The units may comprise an averaging unit for generating an averaged signal over several cardiac cycles, a subtraction unit for subtracting the averaged signal from the input signal to generate a residual signal, a filter unit for providing a filtered signal from the residual signal, and/or an addition unit for adding the filtered signal to the averaged signal.

According to another example, an analog-to-digital (a/D) converter may provide noise suppression by synchronizing the converter's clock to a phase locked loop set to the line frequency.

Additionally, a biometric (e.g., biopotential) patient monitor may use surface electrodes to make biopotential measurements, such as an ECG or electroencephalogram (EEG). The fidelity of these measurements is limited by the effectiveness of the connection of the electrodes to the patient. The resistance of the electrode system to the flow of current (referred to as the electrical impedance) characterizes the effectiveness of the connection. Generally, the higher the impedance, the lower the fidelity of the measurement. Several mechanisms may result in lower fidelity.

The signal from an electrode with a high impedance is affected by thermal noise (or so-called johnson noise), which is a voltage that increases with the square root of the impedance value. Furthermore, the voltage noise of biopotential electrodes tends to exceed what johnson predicts. In addition, amplifier systems that measure from biopotential electrodes may have reduced performance at higher electrode impedances. The impairment is characterized by poor common mode rejection, which increases the contamination of bioelectrical signals by noise sources, such as patient motion and electronics that may be used on or around the patient. These sources of noise are particularly prevalent in procedure rooms and may include equipment such as electrosurgical units (ESUs), cardiopulmonary bypass pumps (CPBs), electric motor driven surgical saws, lasers, and other sources.

During a cardiac protocol, it is often desirable to continuously measure the electrode impedance in real time as the patient is monitored. For this reason, impedance is usually established using ohm's law by injecting a very small current through the electrodes and measuring the resulting voltage. The current may be injected using a DC or AC source. Due to the electrode impedance, it is generally not possible to separate the voltage from the voltage artifacts caused by the interference. The interference may increase the measured voltage and thus the measured apparent impedance, causing the biopotential measurement system to falsely detect an impedance that is higher than the impedance that actually exists. Typically, such monitoring systems have maximum impedance threshold limits that can be programmed to prevent their operation when they detect impedances exceeding these limits. This is particularly true for systems that make measurements of very small voltages, such as EEG. Such systems require very low electrode impedance.

Cardiac ablation procedures can be guided using high-resolution intracardiac Electrograms (EGMs). Cardiac ablation, etc., may be used to treat Ventricular Tachycardia (VT), where rapid and irregular heartbeats are caused by complex Electrophysiological (EP) circuitry and reentry in one of the ventricles. Thus, catheter ablation is aimed at targeting the origin of VT. The mapping of VT circuits and the identification of their sources is critical to the success of VT ablation. The main challenge in interpreting intracardiac EGMs in the presence of VT is that the EGM signal can have a complex morphology, making it difficult to extract the Local Activation Time (LAT) with a sufficiently high spatial resolution. This in turn makes it difficult to accurately map complex electrical circuits in the heart chamber, which is critical for locating relevant ablation targets.

Unipolar EGM signals typically have a much lower signal-to-noise ratio than bipolar signals, so bipolar signals are currently the primary tool for extracting LATs. However, unipolar signals may provide better spatial and temporal resolution, which may significantly improve mapping of VT circuits. Thus, advanced Digital Signal Processing (DSP) methods can be applied to extract accurate LATs from noisy unipolar signals. Advanced digital signal processing methods or systems are intended to reduce or attenuate noise from a signal and may include a variety of digital filters. A linear smoothing filter (e.g., a low pass filter or a high pass filter) or any other smoothing operator that can be convolved with the signal can be used to reduce or attenuate noise. A non-linear filter (e.g., a median filter for noise reduction) may be used to reduce or attenuate the noise. A wavelet transform may be used that achieves both noise reduction and feature retention. Statistical noise reduction methods may be used which may use ambient or adjacent signals or any other pattern to reduce unwanted components in the signal.

The bipolar signal comes from two adjacent unipolar electrodes. The unipolar signal comes from the unipolar electrode and the reference electrode and is a combination of the far field contribution and the near field contribution. The primary source of noise in unipolar signals is the far-field signal generated due to voltage depolarization of distant tissue. Because of the large distance between the monopolar and reference electrodes, the far-field signal is typically not fully accounted for and is not completely removed from the signal compared to the bipolar signal. Since the two unipolar signals forming a unipolar pair have very similar far fields, the difference between them is almost zero except in the case where there is local activity at each of the unipolar signals. This local activity is referred to as a near-field signal and may be indicated as a small peak on the bipolar signal.

In the case of a unipolar electrode located under scar tissue that does not produce electrical activity, the near field may have a lower amplitude than the far field than would be caused by healthy tissue. This makes it particularly difficult to apply classical DSP methods to distinguish far-field signals from near-field signals. In this case, the near field may be very low and negligible. Thus, the bipolar signal may not have any activation and may effectively be zero. This type of monopole can be represented as a pure far-field signal because there is no significant local activity. In this case, the bipolar signal may appear flat, and the two unipolar signals may be nearly identical. These types of signals can be obtained by placing the catheter in a position out of contact with the heart muscles or as body surface ECG signals, which are essentially far-field signals. These types of unipolar signals may be training data sets to enable the neural network to learn this type of activity and to be able to distinguish the far-field components from the mixed unipolar signals.

While in the current context, the far-field contribution is considered to be the noise to be removed, in other contexts, the far-field contribution itself may contain useful information. This may provide additional motivation to separate these two contributions.

Deep Neural Network (DNN) based Deep Learning (DL) has become a subversive technique in the application of computer algorithms to various fields, such as computer vision and DSP. DL allows complex patterns and data to be extracted from signals and images, which typically occurs if such extraction was not previously possible or only possible through time-consuming manual analysis. Therefore, the application of DL to intracardiac EGMs is particularly attractive, where reducing protocol time and increasing clinical success rate are key goals.

Machine Learning (ML) is a set of algorithms and statistical models used for data analysis to perform a specific task. The DL is a subset of machine learning algorithms that set model parameters during the training process to allow accurate prediction of the expected output of unseen data. ML and DL techniques allow analysis of highly complex spatiotemporal information that classical algorithms have difficulty analyzing. While machine learning is typically based on feature extraction using a list of heuristics for the data, DL is based on learning from examples and typically does not require feature extraction from the data. The main difference between DL and traditional ML is that the training process requires a large amount of data. Given a sufficient amount of data, the performance of DL-based algorithms is generally superior to traditional ML algorithms.

Thus, DL is a useful tool for decomposing near-field and far-field components in ECG signals, and in particular VT signals. This would allow activation detection only for near field activity. This is useful because when mapping ventricular activity, the far field can be strong and mask near field activity, misleading the annotation mechanism. This is also useful in the case of Atrial Fibrillation (AFIB), because the ventricular signals are strong, which may be incorrectly annotated as atrial activity.

Therefore, DL methods are desired to shorten the time of the entire clinical procedure by basing insight to medical personnel (e.g., cardiologists and electrophysiologists) that is currently only available through manual data analysis by trained clinicians, and to identify depth data patterns that are currently not identifiable manually or using classical algorithms (e.g., DSP and computer vision), and thus allow for identification of ablation targets in more complex situations that are currently untreatable.

DL training may be unsupervised. That is, while there is a large body of pre-recorded unipolar EGM signals, and additional signals can be collected if needed, the main challenge of applying DL to remove far-field noise is the lack of baseline true phase data for training the DL model. Any far-field and near-field signal decomposition is an estimate and does not necessarily correspond to the true far-field and near-field signals at a particular electrode. Thus, the DL method may be unsupervised rather than supervised. A body surface ECG may be used with the remote electrodes as a baseline true phase for the far field components.

Fig. 10 shows a graphical depiction (10A, 10B, 10C, 10D, 10E, and 10F) of a far-field removed signal process 1000 in accordance with one or more embodiments. The signals in fig. 10A-10E are Intracardiac (IC) ECG signals recorded from different locations along the Coronary Sinus (CS). In fig. 10A, signal 1021 represents a body surface IC ECG signal. A dividing line 1032 indicates a Local Activation Time (LAT). A boundary 1032 also exists in fig. 10C, 10D, 10E, and 10F. Boundary 1043 indicates the QRS location. A boundary line 1043 is also present in fig. 10A, 10C, 10D, 10E, and 10F. In fig. 10, the X-axis represents time, and the Y-axis represents mV.

As shown in fig. 10, 1054 represents the far-field component of the IC ECG signal. Fig. 10C, 10D, 10E, and 10F illustrate the progression of signal 1054 in which the amount of far-field removal is increased, where signal 1065 represents the IC ECG signal after far-field removal. Far-field removal may be accomplished, for example, by creating a blanking period, e.g., the IC ECG signal 1065 may be zero during the far-field period.

Fig. 11 shows a block diagram of a method 1100 according to one or more embodiments. According to one embodiment, method 1100 may be implemented by a noise reduction auto-encoder. Any combination of software and/or hardware (e.g., the local computing device 206 and the remote computing system 208 along with the monitoring and processing apparatus 202) may store, execute, and implement the noise reduction autoencoder and its functions, either individually or collectively. The noise reduction auto-encoder may train the auto-encoder to reconstruct the input from its own corrupted version to force the hidden layer (e.g., hidden layer 530 of fig. 5) to discover more robust features (i.e., useful features that will constitute a better higher-level representation of the input) and prevent it from learning characteristics (i.e., always returning to the same value). In this regard, the noise-reducing auto-encoder may encode the input (e.g., to retain information about the input) and may reverse the effects of the corruption process randomly applied to the input of the auto-encoder.

In accordance with one or more embodiments, the noise-reducing autoencoder may implement a long-short term memory neural network architecture, a convolutional neural network architecture, or other similar architecture. The architecture of the noise reduction auto-encoder may be configured with respect to multiple layers, multiple connections (e.g., encoder/decoder connections), regularization techniques (e.g., differential pressure or BN); and optimizing the features.

The long-short term memory neural network architecture may include feedback connections and may process a single data point (e.g., such as an image) as well as an entire data sequence (e.g., such as voice or video). The cells of the long-short term memory neural network architecture may consist of cells, input gates, output gates, and forgetting gates, where cells remember values at arbitrary time intervals, and gates regulate the flow of information into and out of the cells.

The convolutional neural network architecture may be a shared weight architecture with translation invariance features, where each neuron in one layer is connected to all neurons in the next layer. Regularization techniques of convolutional neural network architectures can take advantage of hierarchical patterns in the data and assemble more complex patterns using smaller and simpler patterns. Other configurable aspects of the architecture may include the number of filters at each stage, the kernel size, the number of kernels per layer, if the noise reduction auto-encoder implements a convolutional neural network architecture.

Method 1100 begins at block 1105, where a noise reduction auto-encoder may receive a "clean and approved" intracardiac data set from a plurality of electrical signals. As referred to herein, an expert medical professional, physician, or the like can examine and edit the data set to remove signal interference, signal artifacts, and signal noise, and approve each electrical signal of the intracardiac data set. At block 1110, the noise reduction auto-encoder builds a model (e.g., model 330 of fig. 3) from the clean and approved intracardiac data set.

At block 1115, a noise reduction auto encoder may receive an input intracardiac signal that includes at least far-field artifacts. The input intracardiac signals may be recorded by one or more monitoring and processing devices (e.g., a five-ray catheter with twenty electrodes, a basket catheter with sixty-four electrodes, a plurality of body surface leads, etc.). The far field may cause problems with generating and navigating the 3D map (i.e., the ventricular far field may interfere with atrial activation).

At block 1120, the noise reduction auto-encoder may encode the input intracardiac signal using the model (from block 1110). The encoding provides a dimensionally reduced input intracardiac signal that removes at least far-field artifacts, depending on how the model indicates the reduction. The result of the encoding is a potential representation. At block 1130, the noise reduction auto-encoder may decode the potential representation to produce an output intracardiac signal.

At block 1135, the noise reduction auto encoder may map out the intracardiac signals. For example, the denoising auto-encoder (with its underlying architecture) finds the mapping function (f, g) such that f (x) Y1 and g (x) Y2.

At block 1140, an ECG may be generated from the mapped output intracardiac signals. The ECG may be generated by the computing device that is executing the noise reduction auto-encoder or by another device. The medical professional may then be shown an improved ECG resulting from the removal of signal interference, signal noise, and signal artifacts. An improved ECG may significantly reduce the time spent on each cardiac case.

As referred to herein, during intracardiac electrogram mapping, the mapping catheter may record both atrial and ventricular activation. In some cases, the ventricular far-field may interfere with atrial activation (e.g., signal interference), which may affect clinical understanding and interpretation of the Carto map. In accordance with one or more embodiments, technical effects and benefits of the noise reduction auto-encoder may include distinguishing between ventricular far-fields and atrial-based activations and generating distinguishing maps for atrial and ventricular activations (e.g., the noise reduction auto-encoder uses a model during decoding to distinguish between ventricular far-fields and atrial-based activations within one or more output intracardiac signals).

Fig. 12 is an exemplary flow diagram of a method 1200 of decomposing a near-field signal and a far-field signal according to an embodiment. During the training phase, far-field ventricular measurements may be acquired (1210). These may be unipolar signals. Measurements may be acquired using a multi-electrode catheter and/or a body surface ECG. There may be many far-field measurements. The far-field measurement may be a pure far-field signal. In one implementation, the pure far-field signal may result from a recording where the bipolar signal is zero or nearly zero. Thus, the near field in a unipolar signal may be absent or very small. In one embodiment, the simulation may be used for pure far-field measurements by using, for example, specialized simulation software that can generate pure far-field signals. This can be done by controlling the source that generates the ECG signal and using only the far source. In one embodiment, an expert may determine a pure far field degree. In one embodiment, the body surface ECG may contain primarily the far field. In one embodiment, measurements from areas of scar tissue may be used for pure far-field measurements that may not contain local activity, so the near-field is ignored.

A synthetic local field signal may be added 1220. The synthesized local field signal may, for example, be from an analog of an ECG signal. A large number of unipolar signals can be introduced in the training phase so the algorithm can learn to recognize unipolar signal morphology. Further, the algorithm may be exposed to far-field signals and may be able to learn to detect these far-field signals.

The algorithm may be configured to evaluate or learn both a pure far-field signal and a combination or mix (real or synthetic mixed signal) of far-field and near-field signals. The algorithm can detect or predict a far-field component from the hybrid signal, which is a common part of all electrodes (far-field). The near field is unique to an electrode because it has local activity that affects only a small area of tissue, while the far field has a much larger contribution (both in terms of signal amplitude and dispersion within the tissue), and is therefore referred to as the common part because it is common to a large number of electrodes. By subtracting the far-field component from the original signal, the common portion of the electrodes may be a pure near-field signal.

Data may be conventionally collected from multiple patients at the EP protocol (1240) and provided to the system. The data may include regular unipolar signals that are a combination of far field signals and near field signals. These signals can be collected in any VT protocol using a multi-electrode catheter. Another approach may be to use a composite signal that combines the far-field and near-field components. For example, analog or synthetic data may be used to generate a pure far-field signal that may be used as a gold standard. These signals may be generated using specialized simulation software or any other simulation program that can control the source of the ECG signals. The data may include unique unipolar signals that include only far-field contributions and no near-field components. The unipolar signal may be obtained by placing the catheter in a position that does not contact the heart muscle. These signals may include ECG values and 3D positions (for each electrode, such that signal V (x, y, z, t). the data may include body surface ECG signals, which are substantially far-field signals.

Prior to processing the unipolar data to extract the near field contributions in the training phase (1230), a pre-processing filtering step may be performed to remove uncorrelated signals and artifacts. Manual annotations may be provided for unipolar signals. The pre-processing filtering may be evaluated by the user (semi-automatically), while at a later stage the annotation data may be used to train a conventional classification Convolutional Neural Network (CNN) to automatically perform the filtering.

The neural network training obtains far-field signal estimates (1230). By knowing the far-field contribution, the near-field signal is the residual signal remaining after the far-field signal is removed from the regular unipolar signal. The bipolar signal can then be reconstructed between the two unipolar signal sets.

In the far-field attenuation model (1250), the auto-encoder can automatically decompose the near-field and far-field from the measured signals (1240) since it is known from the neural network training (1230) what the far-field and near-field signals are.

The neural network may be implemented by a variety of exemplary methods, including an autoencoder and a conjoined network, which may be applied to the decomposition or separation of the far-field and near-field contributions (1250).

Auto-encoders (AEs) are a class of unsupervised DNNs that learn reduced-dimensional representations of a given data set so that they can then produce new data that is statistically similar to the original data set.

In the context of the current approach, if an appropriately selected AE is trained using EGM signals that contain only far-field contributions and no near-field contributions, the AE can be used to extract the far-field contributions from any arbitrary EGM signal. The training phase aims to equalize the input X and output X' (pure far-field), while the prediction phase maps any arbitrary EGM signal to the far-field signal that includes it. There are several AE types (e.g., various AEs, reconstructed AEs, noise-reduced AEs, counter AEs) that include, for example, available elements.

A connected network comprising two identical parts is trained in a fully supervised manner to distinguish pairs of similar and dissimilar features. Then, when the reference feature and the new feature are presented, the network predicts whether the new feature is authentic (i.e., similar to the reference). In recent years, the concept of a connected network has been generalized to DNN and successfully applied to face recognition and face verification. More recently, disjunctive NN has been applied to unsupervised learning for visual representation and medical diagnosis. Since the two unipolar signals from very close electrodes (which form a bipolar pair) typically have very similar far field components, a conjoined network may be utilized. Thus, if two unipolar signals are fed to each section of the conjoined network, a cost function can be constructed that tends to equalize the outputs of the two sections. To avoid obtaining trivial solutions (such as the same zero signal), a constraint term may be added to the cost function. This is a term that tends to minimize the difference between the results and the mean of the two input signals, for example.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Although features and elements are described above with particularity, those of ordinary skill in the art will recognize that each feature or element can be used alone or in any combination with the other features and elements. Furthermore, the methods described herein may 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 understood as a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse traveling through a fiber optic cable), or an electrical signal transmitted through a wire.

Examples of computer readable media include electronic signals (transmitted over a wired or wireless connection) 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 magnetic disks, magneto-optical media, optical media such as Compact Disks (CDs) and Digital Versatile Disks (DVDs), Random Access Memories (RAMs), Read Only Memories (ROMs), erasable programmable read only memories (EPROMs or flash memories), Static Random Access Memories (SRAMs), and memory sticks. A processor associated with software may be used to implement a radio frequency transceiver for use in a WTRU, UE, terminal, base station, RNC, or any host computer.

It is to 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, the singular forms "a", "an" and "the" include plural references unless the context clearly dictates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, element components, and/or groups thereof.

The description of the various embodiments herein is presented for purposes of illustration, but is not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill 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, the practical application or technical improvements over commercially available technology, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method, comprising:

receiving one or more input intracardiac signals from a monitoring and processing device, wherein each of the one or more input intracardiac signals includes one or more signal artifacts;

encoding, by an auto-encoder, the one or more input intracardiac signals with an intracardiac data set to produce a potential representation;

decoding, by the auto-encoder, the potential representation to produce one or more output intracardiac signals comprising the one or more input intracardiac signals reconstructed without the one or more signal artifacts.

2. The method of claim 1, wherein decoding the potential representation to produce one or more output intracardiac signals comprises reconstructing the one or more output intracardiac signals from the potential representation including reduced dimensionality.

3. The method of claim 1, wherein the one or more input intracardiac signals are recorded by a patient biometric sensor of the monitoring and processing device.

4. The method of claim 1, wherein the intracardiac data set comprises a predetermined and approved signal free of the one or more signal artifacts.

5. The method of claim 1, further comprising generating an electrocardiogram from the one or more output intracardiac signals, the electrocardiogram being free of the one or more signal artifacts.

6. The method of claim 1, wherein the auto-encoder comprises a model that distinguishes between ventricular far-fields and atrial-based activation within the one or more output intracardiac signals during decoding.

7. A system, comprising:

a memory storing processor-executable instructions of an autoencoder; and

a processor configured to execute the processor-executable instructions of the autoencoder to cause the system to:

encoding the one or more input intracardiac signals with an intracardiac data set to produce a potential representation;

decoding the potential representation to produce one or more output intracardiac signals comprising the one or more input intracardiac signals reconstructed without the one or more signal artifacts.

8. The system of claim 7, wherein decoding the potential representation to produce one or more output intracardiac signals comprises reconstructing the one or more output intracardiac signals from the potential representation including reduced dimensionality.

9. The system of claim 7, wherein the one or more input intracardiac signals comprise biometric data.

10. The system of claim 7, wherein the one or more input intracardiac signals are recorded by a patient biometric sensor of the monitoring and processing device.

11. The system of claim 7, wherein the intracardiac data set includes a predetermined and approved signal free of the one or more signal artifacts.

12. The system of claim 7, wherein the processor is further configured to execute the processor-executable instructions of the auto-encoder to cause the system to generate an electrocardiogram from the one or more output intracardiac signals, the electrocardiogram being free of the one or more signal artifacts.

13. The system of claim 7, wherein the auto-encoder comprises a noise reduction auto-encoder.

14. The system of claim 7, wherein the auto-encoder comprises a model that distinguishes between ventricular far-fields and atrial-based activation within the one or more output intracardiac signals during decoding.

15. A method of resolving near-field and far-field signals, the method comprising:

receiving the measured signal;

encoding the measured signal by an auto-encoder to produce a potential representation; and

decoding, by the auto-encoder, the potential representation to decompose a near-field component and a far-field component from the measured signal.

16. The method of claim 15, wherein the signal is an Electrocardiogram (ECG) signal.

17. The method of claim 15, wherein the measured signal is a unipolar signal.

18. The method of claim 15, further comprising:

obtaining a far-field ventricular measurement result through a training algorithm;

adding the synthesized local field signal; and

detecting the resulting far-field signal and the residual near-field signal by the training algorithm.

19. The method of claim 15, wherein the far-field ventricular measurements are acquired using a multi-electrode catheter or a body-surface ECG signal.

20. The method of claim 18, wherein decoding the potential representation to decompose near-field and far-field components is based on the detected resulting far-field and residual near-field signals.