Improving mapping efficiency by suggesting mapping point locations

Abstract

To improve mapping efficiency.SOLUTION: A method and apparatus of mapping efficiency by suggesting map point's location includes: receiving data at a machine, the data including a plurality of signals received during the performance of a triangulation to locate a focal tachycardia; generating, by the machine, a prediction model as to the location of the focal tachycardia; and modifying, by the machine, the prediction model based upon additional data received by the machine.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] (CROSS-REFERENCE TO RELATED APPLICATIONS) This application claims the benefit of U.S. Provisional Application No. 63 / 046,948, filed July 1, 2020, the entire contents of which are incorporated herein by reference.

[0002] FIELD OF THE INVENTION The present invention relates to improved mapping efficiency in the context of artificial intelligence and machine learning by suggesting locations of mapping points. [Background technology]

[0003] While attempting to locate the origin of activation using triangulation, multiple acquired points are required, and no data exists regarding how those points should be spatially distributed. Summary of the Invention [Means for solving the problem]

[0004] A method and apparatus for improving mapping efficiency by suggesting locations of mapping points includes receiving data at a machine, the data including a plurality of signals received during a triangulation procedure to identify the location of a focal tachycardia, generating by the machine a predictive model regarding the location of the focal tachycardia, and modifying by the machine the predictive model based on additional data received by the machine. [Brief explanation of the drawings]

[0005] A more detailed understanding may be had from the following description, given by way of example in conjunction with the accompanying drawings, in which like reference numerals indicate similar elements and in which: [Figure 1] FIG. 1 is a block diagram of an exemplary system for remotely monitoring and communicating patient biometric indicators. [Figure 2]FIG. 1 is a system diagram of an example computing environment in communication with a network. [Figure 3] FIG. 1 is a block diagram of an example device capable of implementing one or more features of the present disclosure. [Figure 4] 4 shows a graphical depiction of an artificial intelligence system incorporating the exemplary device of FIG. 3. [Figure 5] 5 illustrates a method implemented in the artificial intelligence system of FIG. [Figure 6] Here is an example of a naive Bayes calculation probability: [Figure 7] 1 illustrates an exemplary decision tree. [Figure 8] 1 illustrates an exemplary random forest classifier. [Figure 9] 1 shows an exemplary logistic regression. [Figure 10] 1 illustrates an exemplary support vector machine. [Figure 11] An exemplary linear regression model is shown. [Figure 12] 1 illustrates an exemplary K-means clustering. [Figure 13] 1 illustrates an exemplary ensemble learning algorithm. [Figure 14] 1 illustrates an exemplary neural network. [Figure 15] 1 shows a hardware-based neural network. [Figure 16A] Examples of cardiomyopathies with different etiologies are shown below. [Figure 16B] Examples of cardiomyopathies with different etiologies are shown below. [Figure 16C] Examples of cardiomyopathies with different etiologies are shown below. [Figure 16D] Examples of cardiomyopathies with different etiologies are shown below. [Figure 17] FIG. 1 is a diagram of an example system in which one or more features of the presently disclosed subject matter may be implemented. [Figure 18] FIG. 1 is an exemplary diagram of a triangulation method according to one embodiment. [Figure 19]FIG. 1 is an exemplary flow diagram of an exemplary method for improving mapping efficiency by suggesting locations of mapping points, according to one embodiment. [Figure 20] 1 shows a graphical depiction of the system and information flow for a single query point. [Figure 21] 21 shows a depiction of using the system of FIG. 20 to infer the next region of interest. [Figure 22] 21 shows the architecture of the commutator network utilized in the system of FIG. 20. [Figure 23] An example of a Vnet network architecture is shown. [Figure 24] The integrated network architecture for use as the NN in FIG. 20 is shown. DETAILED DESCRIPTION OF THE INVENTION

[0006] The details of this application are described herein, but briefly, machine learning (ML) is used to train the system to improve mapping efficiency by guiding users to capture points at locations with informational value.

[0007] A method and apparatus for improving mapping efficiency by suggesting locations of mapping points includes receiving data at a machine, the data including a plurality of signals received during a triangulation procedure to identify the location of a focal tachycardia, generating by the machine a predictive model regarding the location of the focal tachycardia, and modifying by the machine the predictive model based on additional data received by the machine.

[0008] 1 is a block diagram of an exemplary system 100 for remotely monitoring and communicating patient biometrics (i.e., patient data). In the example shown in FIG. 1, the system 100 includes a patient biometric monitoring and processing unit 102 associated with a patient 104, a local computing device 106, a remote computing system 108, a first network 110, and a second network 120.

[0009] According to one embodiment, the monitoring and processing device 102 may be a device internal to the patient's body (e.g., subcutaneously implantable). The monitoring and processing device 102 may be inserted into the patient via any applicable method, including oral administration, surgical insertion via a vein or artery, an endoscopic procedure, or a laparoscopic procedure.

[0010] According to one embodiment, the monitoring and processing device 102 may be a device external to the patient. For example, as described in more detail below, the monitoring and processing device 102 may include an attachable patch (e.g., attached to the patient's skin). The monitoring and processing device 102 may also include a catheter with one or more electrodes, a probe, a blood pressure cuff, a weight scale, a bracelet or smartwatch biometric tracker, a glucose monitor, a continuous positive airway pressure (CPAP) machine, or virtually any device that can provide input regarding the patient's health or biometrics.

[0011] According to one embodiment, the monitoring and processing device 102 may include both components internal to the patient and components external to the patient.

[0012] A single monitoring and processing device 102 is shown in Figure 1. However, an exemplary system may include multiple patient biometric monitoring and processing devices. A patient biometric monitoring and processing device may be in communication with one or more other patient biometric monitoring and processing devices. Additionally or alternatively, a patient biometric monitoring and processing device may be in communication with a network 110.

[0013] One or more monitoring and processing devices 102 may acquire patient biometric data (e.g., electrical signals, blood pressure, body temperature, blood glucose levels, or other biometric data) and may receive at least a portion of the patient biometric data representing the acquired patient biometric indicators, as well as additional information associated with the acquired patient biometric indicators, from one or more other monitoring and processing devices 102. The additional information may be, for example, diagnostic information and / or additional information obtained from additional devices, such as wearable devices. Each monitoring and processing device 102 may process data including its own acquired patient biometric indicators as well as data received from one or more other monitoring and processing devices 102.

[0014] 1, network 110 is an example of a short-range network (e.g., a local area network (LAN) or a personal area network (PAN)). Information may be transmitted over short-range network 110 between monitoring and processing equipment 102 and local computing device 106 using any one of a variety of short-range wireless communication protocols, such as Bluetooth, Wi-Fi, Zigbee, Z-Wave, near field communication (NFC), Ultraband, Zigbee, or infrared (IR).

[0015] Network 120 may be a wired network, a wireless network, or may include one or more wired and wireless networks. For example, network 120 may be a long-range network (e.g., a wide area network (WAN), the Internet, or a cellular network). Information may be transmitted over network 120 using any one of a variety of long-range wireless communication protocols (e.g., TCP / IP, HTTP, 3G, 4G / LTE, or 5G / New Radio).

[0016] The patient monitoring and processing device 102 may include patient biometric sensors 112, a processor 114, user input (UI) sensors 116, memory 118, and a transmitter-receiver (i.e., transceiver) 122. The patient monitoring and processing device 102 may continuously or periodically monitor, store, process, and communicate any number of various patient biometric indicators over the network 110. Examples of patient biometric indicators include electrical signals (e.g., ECG signals and brain biometrics), blood pressure data, blood glucose data, and temperature data. The patient biometric indicators may be monitored and communicated for treatment across any number of various diseases, such as cardiovascular diseases (e.g., arrhythmias, cardiomyopathies, and coronary artery disease) and autoimmune diseases (e.g., type I and type II diabetes).

[0017] The patient biometric sensors 112 may include, for example, one or more sensors configured to sense a type of biometric patient biometric indicator. For example, the patient biometric sensors 112 may include electrodes configured to acquire electrical signals (e.g., cardiac signals, brain signals, or other bioelectrical signals), a temperature sensor, a blood pressure sensor, a blood glucose sensor, a blood oxygen sensor, a pH sensor, an accelerometer, and a microphone).

[0018] As described in more detail below, the patient biometric monitoring and processing device 102 may be an ECG monitor for monitoring cardiac ECG signals. The patient biometric sensor 112 of the ECG monitor may include one or more electrodes for acquiring the ECG signals. The ECG signals may be used in the treatment of various cardiovascular disorders.

[0019] In another example, the patient biometric device 102 may be a continuous glucose monitor for continuously monitoring a patient's blood glucose levels on an ongoing basis to treat various diseases, such as type 1 and type 2 diabetes. The CGM may include subcutaneously placed electrodes that can monitor blood glucose levels from the patient's interstitial fluid. The CGM may be a component of a closed-loop system in which blood glucose data is sent to an insulin pump, for example, for scheduled insulin delivery without user intervention.

[0020] The transceiver 122 may include a separate transmitter and receiver, or alternatively, the transceiver 122 may include a transmitter and receiver integrated into a single device.

[0021] The processor 114 may be configured to store patient data, such as patient biometric data acquired by the patient biometric sensors 112, in the memory 118 and to communicate the patient data over the network 110 via the transmitter of the transceiver 122. Data from one or more other monitoring and processing devices 102 may also be received by the receiver of the transceiver 122, as described in more detail below.

[0022] According to one embodiment, the monitoring and processing device 102 includes a UI sensor 116, which may include a piezoelectric or capacitive sensor configured to receive user input, such as a tap or touch. For example, the UI sensor 116 may be controlled to perform capacitive coupling in response to the patient 104 tapping or touching the surface of the monitoring and processing device 102. Gesture recognition can be implemented via any one of a variety of capacitive types, such as resistive capacitive, surface capacitive, projected capacitive, surface acoustic wave, piezoelectric, and infrared touch. The capacitive sensor may be positioned over a small area or length of the surface such that a tap or touch on the surface activates the monitoring device.

[0023] As described in more detail below, processor 114 may be configured to selectively respond to different tapping patterns (e.g., single tap or double tap) of a capacitive sensor, which may be UI sensor 116, so that different tasks of the patch (e.g., data acquisition, storage, or transmission) may be initiated based on the detected pattern. In some embodiments, audible feedback may be provided to the user from processing unit 102 when a gesture is detected.

[0024] The local computing device 106 of the system 100 may be configured to communicate with the patient biometric monitoring and processing device 102 and act as a gateway to the remote computing system 108 via a second network 120. The local computing device 106 may be, for example, a smartphone, smartwatch, tablet, or other portable smart device configured to communicate with other devices via the network 120. Alternatively, the local computing device 106 may be a fixed or stand-alone device, such as, for example, a fixed base station including modem and / or router capabilities, a desktop or laptop computer that uses an executable program to communicate information between the processing device 102 and the remote computing system 108 via a wireless module in a PC or a USB dongle. Patient biometric indicators may be communicated between the local computing device 106 and the patient biometric monitoring and processing device 102 via a short-range wireless network 110, 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 106 may also be configured to display the acquired patient electrical signals and information related to the acquired patient electrical signals, as described in more detail below.

[0025] In some embodiments, the remote computing system 108 may be configured to receive at least one of the monitored patient's biometric indicators and information associated with the monitored patient over the network 120, which is a long-range network. For example, if the local computing device 106 is a cellular phone, the network 120 may be a wireless cellular network, and information may be communicated between the local computing device 106 and the remote computing system 108 via a wireless technology standard, such as any of the wireless technologies mentioned above. As described in more detail below, the remote computing system 108 may be configured to provide (e.g., visually display and / or audibly provide) at least one of the patient's biometric indicators and associated information to a medical professional (e.g., a physician).

[0026] 2 is a system diagram of an example computing environment 200 in communication with network 120. In some examples, computing environment 200 is incorporated into a public cloud computing platform (such as Amazon Web Services or Microsoft Azure), a hybrid cloud computing platform (such as HP Enterprise OneSphere), or a private cloud computing platform.

[0027] As shown in FIG. 2, computing environment 200 includes a remote computing system 108 (hereinafter computer system), which is an example of a computing system in which embodiments described herein may be implemented.

[0028] The remote computing system 108 can perform various functions via the processor 220, which may include one or more processors. Functions may include analyzing monitored patient biometrics and related information and providing alerts, additional information, or instructions (e.g., via the display 266) according to physician-determined or algorithm-driven thresholds and parameters. As described in more detail below, the remote computing system 108 may also be used (e.g., via the display 266) to provide a dashboard of patient information to a medical professional (e.g., a physician), allowing such information to enable the medical professional to identify and prioritize patients with more significant needs than others.

[0029] 2, computer system 210 may include a communication mechanism, such as a bus 221, or other communication mechanism for communicating information within computer system 210. Computer system 210 further includes one or more processors 220 coupled with bus 221 for processing information. Processor 220 may include one or more CPUs, GPUs, or any other processors known in the art.

[0030] Computer system 210 also includes a system memory 230 coupled to bus 221 for storing information and instructions executed by processor 220. System memory 230 may include computer-readable storage media in the form of volatile and / or nonvolatile memory, such as read-only system memory (ROM) 231 and / or random access memory (RAM) 232. System memory RAM 232 may also include other dynamic storage device(s) (e.g., dynamic RAM, static RAM, and synchronous DRAM). System memory ROM 231 may also include other static storage device(s) (e.g., programmable ROM, erasable PROM, and electrically erasable PROM). Additionally, system memory 230 may be used to store temporary variables or other intermediate information during execution of instructions by processor 220. A basic input / output system (BIOS) 233 may include routines for transferring information, which may be stored in system memory ROM 231, between elements within computer system 210, such as during start-up. RAM 232 may contain data and / or program modules that are immediately accessible to and / or presently being operated on by processor 220. System memory 230 may also include, for example, an operating system 234, application programs 235, other program modules 236, and program data 237.

[0031] The illustrated computer system 210 also includes a disk controller 240 coupled to bus 221 to control one or more storage devices for storing information and instructions, such as a magnetic hard disk 241 and a removable media drive 242 (e.g., a floppy disk drive, a compact disk drive, a tape drive, and / or a solid state drive). Storage devices may be added to computer system 210 using an appropriate device interface (e.g., small computer system interface (SCSI), integrated device electronics (IDE), Universal Serial Bus (USB), or FireWire).

[0032] Computer system 210 may also include a display controller 265 coupled to bus 221 to control a monitor or display 266, such as a cathode ray tube (CRT) or liquid crystal display (LCD), for displaying information to a computer user. The illustrated computer system 210 includes a user input interface 260 and one or more input devices, such as a keyboard 262 and a pointing device 261, for interacting with a computer user and providing information to processor 220. Pointing device 261 may be, for example, a mouse, trackball, or pointing stick for communicating directional information and command selections to processor 220 and for controlling cursor movement on display 266. Display 266 may provide a touchscreen interface that may enable input that complements or replaces the communication of directional information and command selections by pointing device 261 and / or keyboard 262.

[0033] Computer system 210 may perform some or each of the functions and methods described herein in response to processor 220 executing one or more sequences of one or more instructions contained in a memory, such as system memory 230. Such instructions may be loaded into system memory 230 from another computer-readable medium, such as hard disk 241 or removable media drive 242. Hard disk 241 may include one or more data stores and data files used by the embodiments described herein. Data store contents and data files may be encrypted for improved security. Processor 220 may also be employed in a multi-processing configuration to execute one or more sequences of instructions contained in system memory 230. In alternative embodiments, hardwired circuitry may be used in place of or in combination with software instructions. Thus, embodiments are not limited to any specific combination of hardware circuitry and software.

[0034] As mentioned above, computer system 210 may include at least one computer-readable medium or memory for retaining programmed instructions according to the embodiments described herein and for containing the data structures, tables, records, or other data described herein. As used herein, the term “computer-readable medium” refers to any non-transitory, tangible medium that participates in providing instructions to processor 220 for execution. Computer-readable media may take many forms, including, but not limited to, non-volatile media, volatile media, and transmission media. Non-limiting examples of non-volatile media include optical disks, solid-state drives, magnetic disks, and magneto-optical disks, such as hard disk 241 or removable media drive 242. Non-limiting examples of volatile media include dynamic memory, such as system memory 230. Non-limiting examples of transmission media include coaxial cables, copper wire, and fiber optics, including the wires that comprise bus 221. Transmission media can also take the form of acoustic or light waves, such as those generated during radio wave and infrared data communications.

[0035] The computing environment 200 may further include a computer system 210 operating in a networked environment using logical connections to the local computing device 106 and one or more other devices, such as a personal computer (laptop or desktop), mobile device (e.g., a patient mobile device), server, router, network PC, peer device, or other common network node, and typically includes many or all of the elements described above with respect to the computer system 210. When used in a networked environment, the computer system 210 may include a modem 272 for establishing communications over the network 120, such as the Internet. The modem 272 may be connected to the system bus 221 via a network interface 270 or another appropriate mechanism.

[0036] Network 120 as shown in FIGS. 1 and 2 may be any network or system commonly known in the art, including the Internet, an intranet, a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a direct connection or series of connections, a cellular telephone network, or any other network or medium capable of facilitating communication between computer system 610 and other computers (e.g., local computing device 106).

[0037] 3 is a block diagram of an example device 300 capable of implementing one or more features of the present disclosure. The device 300 may be, for example, a local computing device 106. The device 300 may include, for example, a computer, a gaming device, a handheld device, a set-top box, a television, a mobile phone, or a tablet computer. The device 300 includes a processor 302, a memory 304, a storage device 306, one or more input devices 308, and one or more output devices 310. The device 300 may also optionally include an input driver 312 and an output driver 314. It is understood that the device 300 may include additional components not shown in FIG. 3 , including an artificial intelligence accelerator.

[0038] In various alternatives, processor 302 includes a central processing unit (CPU), a graphics processing unit (GPU), a CPU and a GPU located on the same die, or one or more processor cores, each of which may be a CPU or a GPU. In various alternatives, memory 304 is located on the same die as processor 302 or is located separate from processor 302. Memory 304 includes volatile or non-volatile memory, such as random access memory (RAM), dynamic RAM, or a cache.

[0039] The storage device 306 includes fixed or removable storage, such as a hard disk drive, solid state drive, optical disk, or flash drive. The input device 308 includes, but is not limited to, a keyboard, keypad, touch screen, touch pad, detector, microphone, accelerometer, gyroscope, biometric scanner, or network connection (e.g., a wireless local area network card for transmitting and / or receiving wireless IEEE 802 signals). The output device 310 includes, but is not limited to, a display, speakers, printer, haptic feedback device, one or more lights, antenna, or network connection (e.g., a wireless local area network card for transmitting and / or receiving wireless IEEE 802 signals).

[0040] Input driver 312 communicates with processor 302 and input device 308, allowing processor 302 to receive input from input device 308. Output driver 314 communicates with processor 302 and output device 310, allowing processor 302 to send output to output device 310. Note that input driver 312 and output driver 314 are optional components, and device 300 would operate similarly without input driver 312 and output driver 314. Output driver 316 includes an accelerated processing device ("APD") 316 coupled to display device 318. The APD accepts computational and graphic rendering commands from processor 302, processes those computational and graphic rendering commands, and provides pixel output to display device 318 for display. As described in further detail below, APD 316 includes one or more parallel processing units that perform computations according to the single-instruction-multiple-data ("SIMD") paradigm. Thus, although various functions are described herein as being performed by or in conjunction with APD 316, in various alternative examples, the functions described as being performed by APD 316 are additionally or alternatively performed by other computing devices having similar capabilities and that provide graphical output to display device 318 without being driven by a host processor (e.g., processor 302). For example, it is contemplated that any processing system that performs processing tasks according to the SIMD paradigm can perform the functions described herein. Computing systems that do not perform processing tasks according to the SIMD paradigm are contemplated to alternatively perform the functions described herein.

[0041] FIG. 4 shows a graphical depiction of an artificial intelligence system 200 incorporating the exemplary device of FIG. 3 . The system 400 includes data 410, a machine 420, a model 430, multiple outcomes 440, and underlying hardware 450. The system 400 operates by training the machine 420 using the data 410 while building a model 430 that enables the multiple outcomes 440 to be predicted. The system 400 may operate on the hardware 450. In such a configuration, the data 410 may be related to the hardware 450 and may originate from the device 102, for example. For example, the data 410 may be ongoing data or output data associated with the hardware 450. The machine 420 may operate as or be associated with a controller or collection of data associated with the hardware 450. The model 430 may be configured to model the operation of the hardware 450 as well as model the data 410 collected from the hardware 450 to predict outcomes achieved by the hardware 450. The hardware 450 may be configured to use the predicted outcome 440 to provide a predetermined desired outcome 440 from the hardware 450 .

[0042] Figure 5 illustrates a method 500 implemented in the artificial intelligence system of Figure 4. Method 500 includes collecting data from hardware at step 510. This data may include currently collected data, historical data, or other data from the hardware. For example, this data may include measurements taken during a surgical procedure and may be correlated with the outcome of the procedure. For example, cardiac temperature may be collected and correlated with the outcome of a cardiac procedure.

[0043] At step 520, method 500 includes training the machine on the hardware. Training may include analyzing and correlating the data collected at step 510. For example, in the case of the heart, temperature and outcome data may be trained to determine if a correlation or association exists between cardiac temperature during the procedure and the outcome.

[0044] Method 500 includes building a model based on the data associated with the hardware at step 530. Building the model may include modeling of physical hardware or software, modeling of algorithms, etc., as described below. The modeling may aim to represent the collected and trained data.

[0045] Method 500 includes predicting an outcome for a model associated with the hardware at step 540. This outcome prediction may be based on a trained model. For example, for the heart, an outcome may be predicted based on the temperature of the heart during the procedure, when a temperature between 97.7 and 100.2 during the procedure indicates a positive outcome from the procedure. This model is rudimentary and is provided for illustrative purposes to facilitate understanding of the present invention.

[0046] The present systems and methods operate to train machines, build models, and predict outcomes using algorithms. These algorithms may be used to solve the trained models and predict outcomes associated with the hardware. These algorithms can generally be categorized as classification algorithms, regression algorithms, and clustering algorithms.

[0047] For example, classification algorithms are used in situations where the dependent variable, or variable to be predicted, is divided into multiple classes and one class, i.e., the dependent variable, is predicted for a given input. Thus, classification algorithms are used to predict an outcome from among a number of fixed predefined outcomes. Classification algorithms may include naive Bayes algorithms, decision trees, random forest classifiers, logistic regression, support vector machines, and k-nearest neighbors.

[0048] In general, the Naive Bayes algorithm follows Bayes' Theorem and follows a probabilistic approach, it being understood that other probability-based algorithms may be used and generally operate using similar principles of probability theory as those described below for the exemplary Naive Bayes algorithm.

[0049] Figure 6 shows an example of a naive Bayes calculation of probability. The probabilistic approach of Bayes' theorem essentially means that instead of jumping directly to the data, the algorithm has a set of prior probabilities for each target class. After the data is input, the naive Bayes algorithm may update the prior probabilities to form posterior probabilities. This is done using the following formula:

[0050]

number

[0051] The Naive Bayes algorithm and the Bayes algorithm in general can be useful when one needs to predict whether an input belongs to a given list of n classes or not. Because the probabilities of all n classes are fairly low, a probabilistic approach can be used.

[0052] For example, as shown in FIG. 6, a person's decision to play golf depends on factors including the weather outside, as shown in a first data set 610. The first data set 610 lists the weather in a first column and the playing outcomes associated with that weather in a second column. A frequency table 620 generates the frequency with which a particular event occurs. The frequency table 620 determines how often a person plays or does not play golf in each weather condition. From there, a likelihood table is compiled to generate initial probabilities. For example, the probability that the weather is cloudy is 0.29, but the general probability of playing is 0.64.

[0053] Posterior probabilities may be generated from likelihood table 630. These posterior probabilities may be configured to answer questions about weather conditions and whether golf will be played under those weather conditions. For example, the probability that it is sunny outside and golf will be played can be calculated using the Bayesian formula: P(Yes|Sunny) = P(Sunny|Yes) × P(Yes) / P(Sunny) It may be represented by: According to likelihood table 630, P(sunny|yes) = 3 / 9 = 0.33, P(sunny)=5 / 14=0.36, P(yes) = 9 / 14 = 0.64 is. Therefore, P(yes|sunny) = .33 × .64 / .36 or approximately 0.60 (60%).

[0054] Generally, a decision tree is a tree structure similar to a flowchart, where each outer node represents a test of an attribute and each branch represents the result of that test. The leaf nodes contain the actual predicted label. A decision tree starts at the root of the tree and attribute values ​​are compared until a leaf node is reached. Decision trees can be used as classifiers when dealing with high-dimensional data and when little time is spent on data preparation. Decision trees can take the form of simple decision trees, linear decision trees, algebraic decision trees, deterministic decision trees, randomized decision trees, non-deterministic decision trees, and quantum decision trees. An exemplary decision tree is shown below in Figure 7.

[0055] 7 shows a decision tree that follows the same structure as the Bayesian example above when deciding whether to play golf. In the decision tree, a first node 710 examines whether the weather is sunny 712, cloudy 714, and rainy 716 as options for progressing down the decision tree. If the weather is sunny, the tree branch continues to a second node 720 that examines the temperature. In this example, the temperature at node 720 can be high 722 or normal 724. If the temperature at node 720 is high 722, a predicted outcome of "No" 723 for golf occurs. If the temperature at node 720 is normal 724, a predicted outcome of "Yes" 725 for golf occurs.

[0056] Furthermore, from the first node 710, the results Cloudy 714, Golf "Yes" 715 are generated.

[0057] From the first node, weather 710, a third node 730 checks the temperature (again) as a result of rain 716. If the temperature is normal 732 at the third node 730, the answer is "yes" 733 to play golf. If the temperature is low 734 at the third node 730, the answer is "no" 735 to not play golf.

[0058] From this decision tree, a golfer will play golf when it is cloudy 715, when it is sunny with normal temperature 725, and when it is rainy with normal temperature 733, but this golfer will not play when it is hot with sunny weather 723 or cold with rain 735.

[0059] A random forest classifier is a committee of decision trees, each of which is given a subset of the data's attributes and makes a prediction based on that subset. The mode of the decision trees' actual predictions is taken into account to provide the final random forest answer. Random forest classifiers generally mitigate the overfitting present in standalone decision trees, making them more robust and accurate classifiers.

[0060] FIG. 8 illustrates an exemplary random forest classifier for classifying clothing color. As shown in FIG. 8, the random forest classifier includes five decision trees 8101, 8102, 8103, 8104, and 8105 (collectively or generally referred to as decision tree 810). Each of the trees is designed to classify clothing color. Because the individual trees generally operate as the decision trees of FIG. 7, a discussion of each tree and each of the decisions made is not provided. In this illustration, three of the five trees (8101, 8102, 8104) determine that the clothing is blue, one determines that the clothing is green (8103), and the remaining tree determines that the clothing is red (8105). The random forest receives these actual predictions of the five trees and calculates the mode of the actual predictions to provide the random forest's answer that the clothing is blue.

[0061] Logistic regression is another algorithm for binary classification tasks. It is based on the logistic function, also known as the sigmoid function. This S-shaped curve can take any real-valued value and map it between 0 and 1, asymptotically approaching these bounds. Logistic models can be used to model the probability of a particular class or event being present, such as pass / fail, win / lose, alive / dead, or healthy / sick. This can be extended to model several classes of events, such as determining whether an image contains a cat, dog, lion, etc. Each object detected in an image is assigned a probability between 0 and 1, and these probabilities sum to 1.

[0062] In a logistic model, the log odds (log of the odds) of a value labeled "1" are a linear combination of one or more independent variables ("predictors"), each of which can be binary (two classes coded by indicator variables) or continuous (any real value). The corresponding probability of a value labeled "1" is labeled as such because it can vary between 0 (definitely a value of "0") and 1 (definitely a value of "1"). The logistic function is so named because it converts log odds to probabilities. The unit of measurement for the log odds scale is called a logit, another name for the logistic unit. Similar models, such as probit models, can also be used, but with a sigmoid function instead of the logistic function. A defining feature of the logistic model is that increasing one of the independent variables multiplicatively scales the odds of a given outcome by a constant factor, and each independent variable has its own parameters. For binary dependent variables, this generalizes the odds ratio.

[0063] In a binary logistic regression model, the dependent variable has two levels (categories). Outputs with more than two values ​​are modeled by multinomial logistic regression, and if the categories are ordered, they are modeled by ordinal logistic regression (e.g., proportional odds ordinal logistic model). Although a logistic regression model itself simply models the probability of the output with respect to the inputs and does not perform statistical classification (it is not a classifier), it can be used to create a classifier, for example, by selecting a cutoff value and classifying inputs with probabilities greater than the cutoff as one class and inputs with probabilities less than the cutoff as the other class. This is a common way to create a binary classifier.

[0064] FIG. 9 illustrates an exemplary logistic regression. This exemplary logistic regression allows for the prediction of an outcome based on a set of variables. For example, a school's acceptance outcome can be predicted based on an individual's grade point average. The relationship between past history of grade point average and acceptance allows for predictions. The logistic regression of FIG. 9 allows analysis of the grade point average variable 920 to predict an outcome 910 defined between 0 and 1. At the lower end of the S-curve 930, the grade point average 920 predicts an outcome 910 of not being accepted. At the upper end of the S-curve 940, the grade point average 920 predicts an outcome 910 of being accepted. Logistic regression can be used to predict home values, customer lifetime value in the insurance sector, etc.

[0065] A support vector machine (SVM) can be used to sort the data by making the margin between the two classes as far apart as possible. This is called margin-maximizing separation. Unlike linear regression, which uses the entire dataset for its purpose, SVM is able to take into account the support vectors while plotting the hyperplane.

[0066] FIG. 10 illustrates an exemplary support vector machine. In the exemplary SVM 1000, data can be classified into two distinct classes, represented as squares 1010 and triangles 1020. The SVM 1000 operates by drawing a random hyperplane 1030. This hyperplane 1030 is monitored by comparing the distance (shown by lines 1040) between the hyperplane 1030 and the nearest data points 1050 from each class. The data points 1050 closest to the hyperplane 1030 are known as support vectors. The hyperplane 1030 is drawn based on these support vectors 1050, with the optimal hyperplane having the greatest distance from each support vector 1050. The distance between the hyperplane 1030 and the support vectors 1050 is known as the margin.

[0067] The SVM 1000 may be used for data classification by using the hyperplane 1030 to maximize the distance between the hyperplane 1030 and the support vectors 1050. Such an SVM 1000 may be used, for example, to predict heart disease.

[0068] k Nearest Neighbors (KNN) refers to a set of algorithms that generally make no assumptions about the variance of the underlying data and perform a training phase in a reasonably short period of time. Generally, KNN uses a large number of data points divided into classes to predict the classification of a new sample point. Operationally, KNN specifies an integer N with the new sample. The N entries in the model of the system that are closest to the new sample are selected. The most common classification of these entries is determined, and that classification is assigned to the new sample. KNN generally requires increasing storage space as the training set grows. This also means that the estimation time increases linearly with the number of training points.

[0069] In regression algorithms, the output is continuous, so regression algorithms can be used when the target variable is a continuous variable. Linear regression is a common example of a regression algorithm. Linear regression can be used to measure true quality (such as housing costs, number of calls, or full trading) by considering consistent variables. The connection between the variables and the outcome is generated by fitting a line of best fit (hence the name linear regression). This line of best fit is known as the regression line and is referred to directly as the condition Y=a×X+b. Linear regression is most often used in approaches where the number of dimensions involved is low.

[0070] 11 shows an exemplary linear regression model, in which a predictor variable 1110 is modeled against a measurement variable 1120. Clusters of instances of the predictor variable 1110 and the measurement variable 1120 are plotted as data points 1130. The data points 1130 are then fitted to a best-fit line 1140. The best-fit line 1140 is then used in subsequent predictions, where the line 1140 is used to predict the predictor variable 1110 given the measurement variable 1120. Linear regression can be used to model and predict financial portfolios, salary forecasts, real estate, and estimated arrival times for transportation.

[0071] Clustering algorithms may be used to model and train datasets. In clustering, inputs are assigned to two or more clusters based on feature similarity. Clustering algorithms typically learn patterns and useful insights from data without guidance. For example, unsupervised learning algorithms such as K-means clustering can be used to cluster audiences into similar groups based on interests, age, geography, etc.

[0072] K-means clustering is generally considered a simple unsupervised learning approach. It allows similar data points to be clustered together and bound together in the form of clusters. One way to bind data points together is by calculating the centroid of the data points. To determine effective clusters, K-means clustering evaluates the distance of each point from the cluster centroid. Depending on the distance between the data point and the centroid, the data is assigned to the closest cluster. The goal of clustering is to determine the unique groupings of a set of unlabeled data. The "K" in K-means represents the number of clusters formed. The number of clusters (essentially the number of classes into which new instances of data can be classified) can be determined by the user. This determination can be made, for example, using feedback and looking at the size of the clusters during training.

[0073] K-means is often used when the dataset has distinct and well-separated points; otherwise, the modeling may render the clusters inaccurate if they are not separated. Also, K-means may be avoided if the dataset contains many outliers or is non-linear.

[0074] FIG. 12 illustrates K-means clustering. In K-means clustering, data points are plotted and assigned a K value. For example, for K=2 in FIG. 12, the data points are plotted as shown in depiction 1210. Next, in step 1220, the points are assigned to similar centers. Cluster centroids are identified as shown in 1230. Once the centroids are identified, the points are reassigned to clusters such that the distance between the data points and each cluster's centroid is minimized, as shown in 1240. New centroids of the clusters can then be determined, as shown in depiction 1250. Once the data points are reassigned to clusters, new centroids of the clusters formed and an iteration or series of iterations can occur to minimize the size of the clusters and determine the optimal centroid. Next, when a new data point is measured, the new data point can be compared to the centroids and clusters and identified as part of that cluster.

[0075] Ensemble learning algorithms may be used. These algorithms use multiple learning algorithms to achieve better predictive performance than would be possible from any of the constituent learning algorithms alone. Ensemble learning algorithms perform the task of searching a hypothesis space to find suitable hypotheses that make good predictions for a particular problem. Even if the hypothesis space contains hypotheses that are highly suitable for a particular problem, finding a suitable hypothesis can be very difficult. Ensemble algorithms combine multiple hypotheses to form a better one. The term ensemble typically refers to the method of generating multiple hypotheses using the same base learner. The broader concept of a multiple classifier system also encompasses the hybridization of hypotheses that are not derived from the same base learner.

[0076] Because evaluating the predictions of an ensemble typically requires more computation than evaluating the predictions of a single model, ensembles can be thought of as a way to compensate for poor learning algorithms by performing a lot of extra computation. Generally, fast algorithms such as decision trees are used in ensemble methods, for example, random forests, but slower algorithms can also benefit from ensemble methods.

[0077] Ensembles are themselves supervised learning algorithms because they can be trained and then used to make predictions. A trained ensemble therefore represents a single hypothesis. However, this hypothesis is not necessarily contained within the hypothesis space of the model from which it was constructed. Ensembles can therefore be shown to have more flexibility in the functions they can represent. This flexibility could theoretically lead to overfitting the training data more than a single model, although in practice some ensemble methods (especially bagging) tend to reduce problems associated with overfitting the training data.

[0078] Empirically, ensemble algorithms tend to perform better when there is a significant degree of diversity among the models. Therefore, many ensemble methods aim to promote diversity among the models they combine. While counterintuitive, more random algorithms (such as random decision trees) can be used to generate stronger ensembles than more well-thought-out algorithms (such as entropy-reduced decision trees). However, the use of a variety of powerful learning algorithms has been shown to be more effective than techniques that attempt to simplify models to promote diversity.

[0079] The number of component classifiers in an ensemble has a significant impact on the accuracy of prediction. For online ensemble classifiers, this becomes even more important due to the a priori determination of the ensemble size and the volume and velocity of the big data stream. Theoretical frameworks suggest that there is an ideal number of component classifiers for an ensemble, and that having more or fewer classifiers than this number will result in decreased accuracy. Theoretical frameworks suggest that using the same number of independent component classifiers as class labels will result in the highest accuracy.

[0080] Some common types of ensembles include Bayesian optimal classifiers, bootstrap aggregation (bagging), boosting, Bayesian model averaging, Bayesian model combination, and model bucketing and stacking. Figure 13 shows an exemplary ensemble learning algorithm in which bagging is performed in parallel (1310) and boosting is performed sequentially (1320).

[0081] A neural network is a network or circuit of neurons, or in the modern sense, an artificial neural network, composed of artificial neurons or nodes. The connections of biological neurons are modeled as weights. Positive weights reflect excitatory connections, while negative values ​​imply inhibitory connections. The inputs are modified by the weights and summed using a linear combination. An activation function can control the amplitude of the output. For example, the allowed range of the output is usually 0 to 1, but can also be -1 to 1.

[0082] These artificial networks can be used in predictive modeling, adaptive control, and other applications, and can be trained through datasets. Self-learning resulting from experience can occur within the network, allowing it to draw conclusions from complex and seemingly unrelated sets of information.

[0083] For completeness, biological neural networks consist of groups of chemically connected or functionally associated neurons. One neuron may be connected to many other neurons, and the total number of neurons and connections in the network may be large. Connections, called synapses, are usually formed from axons to dendrites, although dendritic synapses and other connections are also possible. Apart from electrical signaling, other forms of signaling result from the diffusion of neurotransmitters.

[0084] Artificial intelligence, cognitive modeling, and neural networks are information processing paradigms inspired by the way biological nervous systems process data. Artificial intelligence and cognitive modeling attempt to simulate some of the properties of biological neural networks. In the field of artificial intelligence, artificial neural networks have been successfully applied to speech recognition, image analysis, and adaptive control, and to build software agents or autonomous robots (in computer and video games).

[0085] A neural network (NN) is an interconnected group of natural or artificial neurons that uses a mathematical or computational model for information processing based on a connectionist approach to computation, in the case of artificial neurons, called an artificial neural network (ANN) or simulated neural network (SNN). In most cases, ANNs are adaptive systems that change their structure based on external or internal information flowing through the network. More practically, neural networks are nonlinear statistical data modeling or decision-making tools. They can be used to model complex relationships between inputs and outputs and to find patterns in data.

[0086] Artificial neural networks comprise networks of simple processing elements (artificial neurons) that can exhibit complex global behavior determined by the connections between the processing elements and element parameters.

[0087] One classic type of artificial neural network is the recurrent Hopfield network. The utility of artificial neural network models lies in the fact that they can be used to estimate and use functions from observations. Unsupervised neural networks can also be used to train representations of inputs that capture salient features of the input distribution, and more recently, deep learning algorithms that can implicitly learn distribution functions for observed data. Training with neural networks is particularly useful in applications where the complexity of the data or task makes manual design of such functions impractical.

[0088] Neural networks can be used in a variety of fields, and the tasks to which they are applied tend to fall into the following broad categories: function approximation or regression analysis, including time series prediction and modeling; classification, including pattern and sequence recognition, novelty detection and sequential decision making; data processing, including filtering, clustering, blind signal separation and compression.

[0089] Areas of application of ANNs include identification and control of nonlinear systems (vehicle control, process control), game playing and decision making (backgammon, chess, racing), pattern recognition (radar systems, face identification, object recognition), sequence recognition (gesture, speech, handwritten text recognition), medical diagnostics, financial applications, data mining (or knowledge discovery in databases, or "KDD"), visualization, and email spam filtering. For example, it is possible to create semantic profiles of user interests arising from photographs trained for object recognition.

[0090] 14 shows an exemplary neural network. The neural network has an input layer represented by multiple inputs, such as 14101 and 14102. The inputs 14101, 14102 are fed into a hidden layer, shown as including nodes 14201, 14202, 14203, and 14204. These nodes 14201, 14202, 14203, and 14204 are combined to produce output 1430 in the output layer. While neural networks perform simple processing via a hidden layer of simple processing elements, nodes 14201, 14202, 14203, and 14204, these nodes can exhibit complex global behavior determined by the coupling between the processing elements and element parameters.

[0091] The neural network of Figure 14 may be implemented in hardware. Referring to Figure 15, a hardware-based neural network is shown.

[0092] Cardiac arrhythmias, and atrial fibrillation in particular, remain common and dangerous conditions, particularly in the elderly population. In patients with normal sinus rhythm, the heart, consisting of atria, ventricles, and excitatory conduction tissue, is electrically excited to beat in a synchronous, patterned manner. In patients with cardiac arrhythmias, abnormal regions of cardiac tissue do not follow the synchronous beating cycle associated with normally conductive tissue, as in patients with normal sinus rhythm. In contrast, abnormal regions of cardiac tissue conduct abnormally to adjacent tissue, disrupting the cardiac cycle and resulting in asynchronous cardiac rhythms. Such abnormal conduction has previously been known to occur in various regions of the heart, such as the region of the sinoatrial (SA) node along the conduction pathways of the atrioventricular (AV) node and the bundle of His, or in the myocardial tissue forming the walls of the ventricles and atria.

[0093] Cardiac arrhythmias, including atrial arrhythmias, can be multiwavelet reentrant, characterized by multiple asynchronous loops of electrical impulses scattered around the atria, often self-propagating. Alternatively or in addition to multiwavelet reentrant, cardiac arrhythmias can also have a localized origin, such as when isolated regions of atrial tissue are spontaneously excited in a rapid, repetitive manner. Ventricular tachycardia (V-tach or VT) is a tachycardia or fast cardiac rhythm that occurs in one of the ventricles. It is a potentially fatal arrhythmia because it can lead to ventricular fibrillation and sudden death.

[0094] Atrial fibrillation, a type of arrhythmia, occurs when the normal electrical impulses generated by the sinoatrial node are overwhelmed by disorganized electrical impulses originating in the atria and pulmonary veins, causing irregular impulses to be conducted to the ventricles. The resulting irregular heartbeat can persist for minutes to weeks or even years. Atrial fibrillation (AF) is often a chronic condition that carries a small increased risk of death, often from stroke. Risk increases with age. Approximately 8% of people over the age of 80 have some degree of AF. While AF is often asymptomatic and generally not fatal in itself, it can lead to palpitations, weakness, fainting, chest pain, and congestive heart failure. The risk of stroke increases during AF because blood can pool in the insufficiently contracting atria and left atrial appendage, potentially forming a blood clot. The first-line treatment for AF is medication to slow the heart rate or restore normal heart rhythm. Additionally, patients with AF are often given anticoagulants to protect against the risk of stroke. The use of such anticoagulants carries its own risks: internal bleeding. In some patients, medication is not sufficient, and their AF is deemed drug-refractory, meaning it cannot be treated with standard pharmacological interventions. Synchronized electrical cardioversion can also be used to convert AF to a normal heart rhythm. Alternatively, patients with AF are treated with catheter ablation.

[0095] Catheter ablation-based treatments may involve mapping the electrical properties of cardiac tissue, particularly the endocardium and cardiac volumes, and selectively ablating the cardiac tissue through the application of energy. Cardiac mapping, which generates, for example, an electrical potential map (voltage map) of wave propagation along cardiac tissue or a map of arrival times to where various tissues are located (local activation time (LAT) map), can be used to detect local dysfunction in cardiac tissue. Ablation, such as that based on cardiac mapping, can stop or modify the propagation of unwanted electrical signals from one part of the heart to another.

[0096] Ablation techniques disrupt unwanted electrical pathways by creating non-conducting lesions. Various energy delivery modalities have been previously disclosed for creating lesions, including the use of microwave, laser, and more commonly, radiofrequency energy to create conduction blocks along cardiac tissue walls. In a two-stage procedure, mapping followed by ablation, electrical activity at points within the heart is typically sensed and measured by advancing a catheter containing one or more electrical sensors (or electrodes) into the heart and acquiring data at multiple points. These data are then used to select a target region of the endocardium where ablation will be performed.

[0097] Cardiac ablation and other cardiac electrophysiology procedures are becoming increasingly complex as clinicians treat challenging conditions such as atrial fibrillation and ventricular tachycardia. Treatment of refractory arrhythmias can now rely on the use of three-dimensional (3D) mapping systems to reconstruct the anatomy of the cardiac chamber of interest.

[0098] For example, cardiologists rely on software such as the Complex Fractionated Atrial Electrograms (CFAE) module of the CARTO® 3 3D mapping system manufactured by Biosense Webster, Inc. (Diamond Bar, California) to analyze intracardiac EGM signals and determine ablation points for treating various cardiac disorders, including atypical atrial flutter and ventricular tachycardia.

[0099] 3D maps can provide multiple pieces of information about the electrophysiological properties of tissues, representing the anatomical and functional substrates of these challenging arrhythmias.

[0100] Cardiomyopathy of different etiologies (e.g., ischemic, dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), arrhythmogenic right ventricular dysplasia (ARVD), left ventricular noncompaction (LVNC)) are characterized by areas of unhealthy tissue surrounded by areas of normally functioning cardiomyocytes with a distinguishable substrate.

[0101] 16A-16D show examples of cardiomyopathies with different etiologies. As a first example, FIGS. 16A and 16B show an exemplary rendering of a heart 1600 with post-ischemic ventricular tachycardia (VT) characterized by an endocardial-epicardial low- or intermediate-voltage region 1602 where signal conduction is slowed. This demonstrates that measuring late potentials within or around dense scar regions can help identify isthmuses that may sustain VT. The post-ischemic VT shown in FIG. 16A is characterized by an endocardial-epicardial low- or intermediate-voltage region where signal conduction is slowed. This demonstrates that measuring late potentials within or around dense scar regions can help identify isthmuses that may sustain VT. FIG. 16A shows the distribution of bipolar signal amplitude (Bi) in various parts of the heart 1600. FIG. 16A shows Bi ranging from 0.5 mV to 1.5 mV. Figure 16B shows the distribution of the Shortex complex interval (SCI) in various parts of the heart. As an example, the SCI ranges from 15.0 ms to 171.00 ms, with the SCI range of interest being 80 ms to 170 ms.

[0102] Figures 16C and 16D show exemplary renderings of a heart 1610 experiencing left ventricular noncompaction. More specifically, Figure 16C shows an epicardial voltage map, and Figure 16D shows a potential duration map (PDM). Three black circles 1612 in Figures 16C and 16D are marked as abnormally delayed potentials (potentials longer than 200 milliseconds).

[0103] Abnormal tissue is generally characterized by low-voltage EGMs. However, initial clinical experience with endocardial-epicardial mapping has shown that areas of low voltage are not always present as the sole arrhythmogenic mechanism in such patients. In fact, areas of low or intermediate voltages may exhibit EGM fragmentation and prolonged activity during sinus rhythm, which corresponds to the isthmus at risk identified during sustained and coherent ventricular arrhythmias, e.g., only in intolerant ventricular tachycardia. Furthermore, EGM fragmentation and prolonged EGM activity are often observed in areas showing normal or near-normal voltage amplitudes (>1–1.5 mV). Although the latter areas can be evaluated according to voltage amplitude, they cannot be considered normal according to the intracardiac signal and therefore represent a true arrhythmogenic substrate. 3D mapping may be able to identify the location of arrhythmogenic substrates on the endocardial and / or epicardial layers of the right / left ventricle, which may vary in distribution depending on the extent of the primary disease.

[0104] The substrates involved in these cardiac diseases are associated with subdivision of the endocardial and / or epicardial layers of the ventricular chambers (right and left) and the presence of delayed EGMs. 3D mapping systems such as the CARTO® 3 can identify the location of potential arrhythmogenic substrates for cardiomyopathies in terms of detecting abnormal EGMs.

[0105] Electrode catheters have been commonly used in medical practice for many years. They are used to stimulate and map electrical activity within the heart and to ablate sites of abnormal electrical activity. In use, an electrode catheter is inserted into a major vein or artery, such as the femoral artery, and then guided into a target heart chamber. A typical ablation procedure involves inserting a catheter with at least one electrode at its distal end into a heart chamber. A reference electrode is typically provided by a second catheter taped to the patient's skin or positioned within or near the heart. When RF (radio frequency) current is applied to the tip electrode of the ablation catheter, current flows through the medium (i.e., blood and tissue) surrounding the tip electrode toward the reference electrode. The distribution of the current depends on the amount of contact the electrode surface has with the tissue compared to the blood, which has a higher electrical conductivity than the tissue. Heating of the tissue occurs due to the electrical resistance of the tissue. Sufficient tissue heating can cause cell destruction in the cardiac tissue, resulting in lesions within the non-conductive cardiac tissue. During this process, the electrode also heats due to conduction from the heated tissue to the electrode itself. If the electrode temperature becomes high enough, perhaps above 60°C, a thin, transparent film of dehydrated blood proteins can form on the electrode's surface. As the temperature continues to rise, this dehydrated layer can gradually thicken, causing blood to coagulate on the electrode surface. Because dehydrated biological material has a higher electrical resistance than endocardial tissue, the impedance to the flow of electrical energy into the tissue also increases. If the impedance increases sufficiently, an impedance rise occurs, necessitating removal of the catheter from the body and cleaning of the tip electrode.

[0106] FIG. 17 is a diagram of an example system 1720 capable of implementing one or more features of the presently disclosed subject matter. All or a portion of the system 1720 can be used to collect information for a training dataset and / or to implement a trained model. The system 1720 can include a component, such as a catheter 1740, configured to injure a tissue region of an internal organ. The catheter 1740 can also be further configured to acquire biometric data. While the catheter 1740 is shown as being a point catheter, it will be understood that any shape of catheter including one or more elements (e.g., electrodes) can be used to implement the embodiments disclosed herein. The system 1720 includes a probe 1721 having a shaft that can be navigated by a physician 1730 into a body part, such as a heart 1726, of a patient 1728 reclining on a table 1729. However, multiple probes may be provided depending on the embodiment, and while a single probe 1721 is described herein for simplicity, it will be understood that the probe 1721 may represent multiple probes. 17 , a physician 1730 can insert a shaft 1722 through a sheath 1723 while manipulating the distal end of the shaft 1722 using a remote control 1732 near the proximal end of the catheter 1740 and / or a deflector from the sheath 1723. As shown in inset 1725, the catheter 1740 can be attached to the distal end of the shaft 1722. The catheter 1740 can be inserted through the sheath 1723 in a collapsed state and then expanded within the heart 1726. As further disclosed herein, the catheter 1740 can include at least one ablation electrode 1747 and a catheter needle 1748.

[0107] According to exemplary embodiments, catheter 1740 may be configured to ablate a tissue region of a chamber of heart 1726. Inset 1745 shows a close-up of catheter 1740 inside a chamber of heart 1726. As shown, catheter 1740 may include at least one ablation electrode 1747 coupled to the body of the catheter. According to other exemplary embodiments, multiple elements may be connected via splines that form the shape of catheter 1740. One or more other elements (not shown) may be provided and may be any element configured to perform ablation or acquire biometric data, such as an electrode, a transducer, or one or more other elements.

[0108] According to embodiments disclosed herein, an ablation electrode, such as electrode 1747, may be configured to deliver energy to a tissue region of a body organ, such as heart 1726. The energy may be thermal energy and may cause damage to the tissue region starting at the surface of the tissue region and extending through the thickness of the tissue region.

[0109] According to exemplary embodiments disclosed herein, the biometric data may include one or more of LAT, electrical activity, topology, bipolar mapping, dominant frequency, impedance, etc. The local activation time may be the time point of a threshold activation corresponding to local activation calculated based on a normalized initial starting point. The electrical activity may be any applicable electrical signal that can be measured based on one or more thresholds and may be detected and / or enhanced based on signal-to-noise ratio and / or other filters. The topology may correspond to the physical structure of a body part or portion of a body part, or may correspond to changes in the physical structure for different portions of the body part or for different body parts. The dominant frequency may be a frequency or range of frequencies that is common in a portion of a body part and may differ in different portions 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 resistance measurement in a given region of a body part.

[0110] 17 , the probe 1721 and catheter 1740 can be connected to a console 1724. The console 1724 can include a processor 1741, such as a general-purpose computer with suitable front-end and interface circuitry 1738, for transmitting and receiving signals to and from the catheter, as well as for controlling other components of the system 1720. In some embodiments, the processor 1741 can be further configured to receive biometric data, such as electrical activity, and determine whether a given tissue region conducts electricity. According to one embodiment, the processor can be external to the console 1724, for example, located in the catheter, an external device, a mobile device, a cloud-based device, or can be a stand-alone processor.

[0111] As noted above, the processor 1741 may include a general-purpose computer, which may be programmed in software to perform the functions described herein. The software may be downloaded to the general-purpose computer in electronic form, for example, over a network, or alternatively or additionally, may be provided and / or stored on non-transitory tangible media, such as magnetic, optical, or electronic memory. The exemplary configuration shown in FIG. 17 may be modified to implement embodiments disclosed herein. Embodiments of the present disclosure may be similarly applied using other system components and configurations. Additionally, the system 1620 may include additional components, such as elements for sensing electrical activity, wired or wireless connectors, processing and display devices, etc.

[0112] According to one embodiment, a display connected to a processor (e.g., processor 1741) may be located at a remote location, such as a separate hospital or a separate healthcare provider network. Additionally, system 1720 may be part of a surgical system configured to obtain anatomical and electrical measurements of a patient's organs, such as the heart, and to perform cardiac ablation procedures. One example of such a surgical system is the Carto® system sold by Biosense Webster.

[0113] System 1720 can also, and optionally, acquire biometric data, such as anatomical measurements of the patient's heart, using ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), or other medical imaging techniques known in the art. System 1720 can acquire electrical measurements using a catheter, an electrocardiogram (EKG), or other sensors that measure the electrical properties of the heart. The biometric data, including the anatomical and electrical measurements, may then be stored in memory 1742 of mapping system 1720, as shown in FIG. 17 . The biometric data may be transmitted from memory 1742 to processor 1741. Alternatively, or in addition, the biometric data may be transmitted to server 1760, which may be local or remote, using network 1662.

[0114] The network 1762 may be any network or system commonly known in the art, such as an intranet, a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a direct connection or series of connections, a cellular telephone network, or any other network or medium capable of facilitating communication between the mapping system 1720 and the server 1760. The network 1662 may be wired, wireless, or a combination thereof. Wired connections may be implemented using Ethernet, Universal Serial Bus (USB), RJ-11, or any other wired connection commonly known in the art. Wireless connections may be implemented using Wi-Fi, WiMAX, Bluetooth, infrared, cellular networks, satellite, or any other wireless connection method commonly known in the art. Additionally, several networks may operate alone or in communication with each other to facilitate communication within the network 1762.

[0115] In some cases, server 1762 may be implemented as a physical server. In other cases, server 1762 may be implemented as a virtual server, a public cloud computing provider (e.g., Amazon Web Services (AWS)).

[0116] The control console 1724 may be connected by a cable 1739 to body surface electrodes 1743, which may include adhesive skin patches that are affixed to the patient 1730. The processor, in conjunction with the current tracking module, may determine position coordinates of the catheter 1740 within the patient's body part (e.g., the heart 1726). The position coordinates may be based on impedance or electromagnetic fields measured between the body surface electrodes 1743 and electrodes 1748 or other electromagnetic components of the catheter 1740. Additionally or alternatively, the location pad may be located on the surface of the bed 1729 and may be separate from the bed 1729.

[0117] Processor 1741 may include real-time noise reduction circuitry, typically configured as a field programmable gate array (FPGA), followed by an analog-to-digital (A / D) ECG (electrocardiograph) or EMG (electromyogram) signal conversion integrated circuit. Processor 1741 may communicate signals from the A / D ECG or EMG circuitry to another processor and / or may be programmed to perform one or more functions disclosed herein.

[0118] The control console 1724 may also include an input / output (I / O) communication interface that allows the control console to communicate signals to and / or from the electrodes 1747 .

[0119] During a procedure, the processor 1741 facilitates the presentation of the body part rendering 1735 to the physician 1730 on the display 1727 and may store data representing the body part rendering 1735 in the memory 1742. The memory 1742 may comprise any suitable volatile and / or non-volatile memory, such as a random access memory or a hard disk drive. In some embodiments, the medical professional 1730 may be able to manipulate the body part rendering 1735 using one or more input devices, such as a touchpad, a mouse, a keyboard, a gesture recognizer, or the like. For example, the input device may be used to change the position of the catheter 1740 so that the rendering 1735 is updated. In an alternative embodiment, the display 1727 may include a touch screen, which may be configured to receive input from the medical professional 1730 in addition to presenting the body part rendering 1735.

[0120] As described herein, during the triangulation process, the acquired data indicates where the focal tachycardia location activation is likely to originate and which points are most informative for validating and finding the origin in the most data-efficient manner. The algorithm learns from previous maps, anatomy, and mapping sequences to recommend the best course of action at each point.

[0121] 18 is an exemplary illustration of a triangulation method 1800 according to one embodiment. A prediction of the LV predicted curve segment is calculated. The LV predicted curve segment includes curve segments 1810, 1820, and 1830. The intersection 1840 (circled) of curve segments 1810, 1820, and 1830 predicts the origin 1850 of the wavefront. The propagation of the wavefront is graphically represented by the progression from the earliest prediction to the latest prediction. The successful ablation site is considered to be the true wavefront origin 1850. As shown, the wavefront origin 1850 is located within the circle of intersection 1840.

[0122] The triangulation method 1800 uses a small number of mapping points (within curve segments 1810, 1820, and 1830 in this illustration) to identify the location of the focal arrhythmia source (wavefront origin 1850) for cases assumed by the physician to be focal (a single source that is not a fibrillatory arrhythmia). This mapping results in reduced treatment time for such cases. The algorithm receives the FAM mesh and a small number of LAT points. By inferring the propagation of the LAT on the FAM, the intersection 1840 of a small number of propagation curve segments (curve segments 1810, 1820, and 1830 in this illustration) is provided to estimate the source of the propagation (wavefront origin 1850). Disadvantages of the triangulation method 1800 include the physician having little knowledge of where to obtain informative LAT and the lack of consideration of geometric / electrical noise or differences in tissue conduction velocity.

[0123] 19 is an exemplary flow diagram of an exemplary method 1900 for improving mapping efficiency by suggesting locations of mapping points, according to one embodiment. At step 1910, data for machine learning (ML) is received at a machine / system. The data includes multiple signals received during a triangulation process to identify the location of a focal tachycardia. The received data is utilized to generate a predictive model for the location of the focal tachycardia (step 1920). The predictive model is based on additional data received by the machine (step 1930).

[0124] When determining local propagation to determine the focal point of ablation, some effects in the myocardium are stable and some are not. Therefore, local activation points (LAT) are used to estimate where the focal point is within the triangle. Data can be, for example, anatomical (FAM / CT) data and LAT points acquired by a Carto machine. In addition, ablation locations marked as focal / terminal by a physician or a coherent map can be used.

[0125] For mesh representations, MeshCNN or dense volumetric networks (3DCNNs) can be used. These points can be fed as a list, to an RNN, or as volumetric samples (volumes where each voxel represents one or more LATs, which are also 3D convolutions). A dense encoding layer(s) is used to combine these two modalities to provide a regression output for the next sample point and an estimate of the focal point. Input data is described (location and activation) at each stage for a single LAT point and added to the input, and the system / network output is the location of the next best point to sample across the entire anatomical structure, which can be expressed as coordinates in space, and the current estimate of the focal point. The system model receives the input LAT point-by-point and minimizes the cost by constraining the time period to have a minimum number of input points and a good estimate of the focal point (either by self-estimation, the original triangulation algorithm, or by using coherent or physician ablation points). Furthermore, the convolutional filter weights, fully connected layers, and GRU layers are all trainable. The output of the model represents the LAT of the focal point and a suggestion for the next point to obtain the estimate. Because statistical similarities may exist between patients, the similarities can be learned and used to further provide more accurate predictions to physicians. In addition, there may be differences in arrhythmias based on specific diseases, which can also be utilized in the prediction.

[0126] Despite advances in EP technology, locating the origin of a localized arrhythmia (e.g., VT, localized atrial tachycardia) in the ventricles or atria can be time-consuming, and activation mapping still has inherent limitations. Based on the algorithm in Figure 19, the following system and architecture predicts the origin of a localized wavefront using location and activation timing information from two pairs of sampled points. This system can be incorporated into an electroanatomical mapping (EAM) system to evaluate its accuracy in a three-dimensional clinical environment. The system described below includes an ML-based algorithm that integrates knowledge gained from pre-recorded cases to provide real-time guidance to physicians. This system overcomes the noted drawbacks and can achieve a faster and more robust solution. The system takes advantage of the dense nature of the map, allowing the focal point to be easily estimated, and allows the system to operate on a subset of inputs and use that data to estimate which unreached regions are most likely to locate the local source. Access to the entire data allows for local source identification.

[0127] A triangle mesh defines a 3D surface in space. Formally, space

[0128]

number

[0129] The anatomical activation maps can be represented in a uniform way. The input space is a 10-sampling matrix of voxels. 3 (cm 3 ) grid, each of which is assumed to contain 2 3 (mm 3) volume. Each voxel may contain a feature vector describing the electrical activation signal measured inside this volume. The feature vector can be trained to represent the reference annotation and the LAT annotation. The anatomical surface (the boundary between the blood pool and the tissue) is represented by a signed Euclidean distance (also known as a distance transform). For each voxel center, this measures the distance to the closest point on the anatomical surface. It is positive if the voxel is outside the anatomical structure and negative if it is inside. A feature vector can be trained that summarizes the anatomical structure.

[0130] Figure 20 shows a graphical depiction of the system and information flow 2000 for a single query point. The system 2000 provides representations of anatomical structures and signals and learns a map between these representations and local activation propagation assuming a single focus.

[0131] Inputs 2010 to the system 2000 may include recorded ECG signals and locations 2010.1, Carto annotations 2010.2a, and anatomical structure surfaces (e.g., FAM, mFAM, CT) 2010.3. Inputs 2010 may include the FAM mesh 2010.3, acquired LAT points (location in space (mm) 2010.2 and ECG (signal matrix described below) 2010.1, REF, and LAT values ​​(milliseconds)), focal points estimated from the algorithm, and a set of ablation points where arrhythmias terminate. This is used to verify the "true" focal points. The LAT / coherent map, which includes LAT mapping, is for all heart chambers and assigns activation time values ​​for each vertex, while the coherent map also indicates conduction velocity vectors (per vertex) and non-conducting regions. Focal points may be estimated on the LAT / coherent map where the mapping is sufficiently dense. All acquired points may be on the FAM. Verification is possible if an ablation point exists near the location where the arrhythmia terminates. Both intracardiac and body surface ECG signals 2010.1 can be used. The ECG signal 2010.1 is a matrix of signals for every point x in space sampled at time t (ECG sample rate) for all recorded channels c.

[0132]

number

[0133] To make the system 2000 useful for various intra-patient anatomical structures and activation propagation, the system 2000 can be trained to extract uniform, compact, and meaningful information about the anatomical and ECG signals. According to one embodiment, two networks can perform this operation in an unsupervised manner. The information is stored in the latent space R k(where K is constant and relatively small), the input is mapped from its original large vector dimension to a more compact vector dimension, and the input is reconstructed using only the latent representation. The latent space can capture important features by itself, or can be refined by additional tasks and constraints. Denoising / variable autoencoders, deep belief networks, and generative adversarial networks (GANs) are examples of networks that can perform this task.

[0134] The recorded ECG signals, locations 2010.1, and Carto annotations may be combined in a first NN 2020. This input may also include other inputs 2010. A translator network 2020 can be used to encode the ECG signals into feature vectors. A translator is the state of the art in sequence-to-sequence networks. Each ECG set of signals 2010.1 is acquired at some location in space for a fixed time interval. This results in a matrix of signals for each acquired point. A translator takes an input sequence and attempts to predict the next (or several) time steps of a given sequence. This is done unsupervised, by only looking at a training ECG signal set 2010.1, which may be taken from any case of the same heart chamber. Annotating the reference and LAT may be performed as an additional training option to fine-tune the latent space. Interpretable translation may be performed using an encoder-decoder network. A semantic autoencoder for zero-shot learning may also be used. The transformer itself is constructed from an encoder and a decoder, thus defining a latent vector as the input to the encoder.

[0135] 22 shows the architecture of a commutator network 2200 utilized in system 2000. The left side of network 2200 is the encoder, and the right side of network 2200 is the decoder.

[0136] The input 2220 to the converter 2200 may be a sequence of tokens, where a token is a vector that selects members of a discrete set. To tokenize the ECG signal 2230, the measured signal may be represented or approximated (a quantity or series) using discrete quantity(s). For example, the measured voltages in the ECG signal may be represented in the interval [0, V MAX ] and these signals may be represented by a set of discrete values. In one embodiment, 768 discrete values ​​are used. When representing the signals discretely, the closest value from the discrete space is selected for each ECG signal. In another embodiment, the discrete space may be divided on a logarithmic scale.

[0137] Here, each input is converted into a vector ("one-hot coded") with a 1 in the correct "token" and a 0 otherwise. A network may be trained to predict the value of the next time sample using a number of these vectors 2240, such as n = n, covering approximately 300-400 milliseconds based on the estimated tachycardia duration. According to one embodiment, time resampling of the vectors may be performed for efficiency. Values ​​may be predicted using softmax 2210 over the 768-dimensional output probability vector. The decoder receives the shifted encoder output 2250 and applies the regression to multiple samples 2260, allowing it to learn to predict the entire sequence. The network's internal representation is the input that is embedded, transformed, and operated on through the encoder's attention and projection layers. The final representation size may be selected by the implementation. A transformation of the entire sequence in n blocks (with or without overlap) may be performed by concatenation of the block vectors to generate the final representation vector of this signal 2270.

[0138] The second NN 2030 may operate on the anatomical surface information 2010.3 and other inputs 2010. The anatomical surface can be represented using a Vnet network, which can be a fully convolutional network with residual connections. This network accepts an input volume of a given size (50x50x50 in this case) and returns an output volume of the same size. The first part performs 3D convolutions followed by nonlinear (RELU) downsampling (max pooling) at each layer. Each 3D convolution has a filter of size 50x50x50xC, where C is the number of channels. The first three-dimensional weights represent spatial locations and are shared, while C is the current dimension of the input or feature data. The inner part (vector of size 512 in the network in the figure) represents the latent space with all residual links added (by concatenation). The concatenated latent vector is

[0139]

number

[0140] Figure 23 shows an example Vnet network architecture 2300. As shown, the Vnet architecture 2300 may include 323 resolutions, but can also be adapted to other resolutions. The Vnet architecture 2300 shows an example of a Vnet network, and the actual number of filters, initial volume size, etc. at each layer are subject to optimization. Since we can have several acquired points within the same voxel, we only take the average latent vector representing this voxel.

[0141] 20, the output of the first NN 2020 includes a spatial ECG feature transformation 2040 as described. The output of the second NN 2030 includes a spatial shape representation 2050 as described.

[0142] The spatial ECG feature transformations 2040 and spatial shape representation 2050 may be combined in a third NN 2060 along with the query points 2070 to generate an output, which may include a vector 2080 to the origin and a confidence level 2090.

[0143] The vector to source 2080 is a vector pointing in the direction of the source (across the surface shape), and the confidence level 2090 represents the reliability of the vector to source 2080 relative to the currently acquired point. The system 2000 may also use both classical triangulation algorithms and networks to output a focal point estimate and one or more points that can be sampled by the physician to improve the focal point estimate.

[0144] A vector direction 2080 to the source (close to the local inverse gradient). A confidence level 2090 measured by local consistency of the signal representation and the amount of meaningful data. By using the network to sample points across the anatomical surface, the output may include a point(s) in space that indicate the next most informative region to sample. These point(s) may be from regions with high confidence values ​​2090 and a point in space that represents a new origin that has been approximated using the intersection of a vector to the source or using a classical triangulation algorithm.

[0145] The system 2000 may incorporate a data flow as follows: The currently known signal and anatomical structure may be input 2010. The input 2010 may be converted to features 2040, 2050 via one or more NNs (shown as NNs 2020, 2030). The system 2000 may operate with NN 2060 across all voxels that contain the anatomical structure surface and do not contain adjacent ECG signals, querying the estimated confidence level of each voxel. A region is a defined area of ​​adjacent voxels. The region 2080 with the highest average confidence 2090 may be selected as the next suggested location for query 2070. The system 2000 may also output a current estimate of the focal point using a classical triangulation algorithm.

[0146] Figure 24 shows an integrated network architecture 2400 for use as the NN 2060. The network 2400 can receive an encoded representation of the ECG 2040, anatomical structures 2050, and query points 2070, and return an origin vector 2080 and a confidence level 2090. The network described in Figure 23 encodes the ECG signal in each voxel into a vector. An additional network can be used to convert this set of vectors into a single vector representation that summarizes all the spatial information. This converts a volume of 50x50x50xK (the first three dimensions are spatial, and K is the latent representation) into a latent vector

[0147]

number

[0148]

number

[0149] Additionally, system 2000 can provide an estimate using the intersection of source vectors for all currently known points using weighted majority voting. As shown in FIG. 21 , there is a depiction 2100 of using system 2000 to infer the next region of interest. Similar to FIG. 20 , a recorded ECG signal, location 2010.1, and an anatomical structure surface 2010.3 can be provided as input to system 2000. System 2000 can query all vertices of the anatomical structure surface that do not have information, shown as heart 2110. System 2000 can output the next region 2120 for data acquisition and vector 2080 to the source.

[0150] The unsupervised coding networks 2020, 2030, 2060 may be trained across a large corpus of cases. A network for each heart chamber may be created, each differing in anatomical structure and ECG signal phenomena. The second step in training the integrative network is to consider LAT / coherent / triangulation cases, where activations of only a portion of the heart chambers are used as inputs to the system. Then, the source vectors 2080 and confidences 2090 for each of the remaining points across the anatomy are calculated; this may be done multiple times per case, each time using a different initial point.

[0151] The confidence 2090 is calculated as a function of the following considerations: how close the point is to the actual source, how diverse the vectors to the source are in the vicinity of the point (lower is better), how diverse the latent representation of the ECG is in the vicinity of the point (lower is better, as long as there are enough measurements), and how much the estimate of the focal point improves (by using the current input point) and adding this point (the primary consideration).

[0152] The exact function for combining this can vary, and results with scores between 0 (no points improvement) and 1 (very good points to take) can be useful.

[0153] As briefly described above, data from previous patient cases is used to train the system 2000. For each such case, available data may include patient ID, data used as input to the system, and data used to calculate the desired output. The "gold standard," i.e., data used to calculate the desired output, may include a surface mesh from FAM or other models and a LAT / coherent map. A LAT / coherent map (determined experimentally) with enough acquired points may be selected as the training input. The earliest activation point may be designated as the input focal point. If the focal point can be validated by ablation data, it may be identified with a higher influence on the training process (by weighting).

[0154] A set of several (at least three) points S⊂V can be sampled as starting points. Each point is associated with a set of ECG signals within a certain neighborhood radius from the point. A new point can then be selected, and for this additional point, a vector to the source, a confidence, and an estimate of the new focal point can be calculated. This process finds the point p that best improves the focal estimate. * This process is repeated for all points to determine the pair (S,p * ) A new set S may be identified and this process may be repeated multiple times for each given case. Conduction velocity vectors (coherent) that will not deviate significantly from the back vector to the source may be used to provide additional clues as to the input point.

[0155] Each case may be weighted and ranked for training based on an "ablation index" calculated, for example, by CARTO®. Points may be discounted for ablation points with poor treatment quality (e.g., point instability, ablation that is not deep enough or does not use enough power), ablation over a large area, or multiple ablation sites. A particular patient's data set may be weighted according to the survival time of the treatment outcome. For example, a case in which arrhythmia recurs 3 months after treatment may have a lower weight than a case in which arrhythmia recurs only 12 months after treatment, while a case lacking ablation verification receives the lowest weight. Focal quality and mapping density estimated from different sets of points may also be included.

[0156] The goal of this process is to allow further training of the algorithm on cases where the foci are clear and clinically verified and the mapping is of high quality.

[0157] Typically, a dataset must be divided into a training set, a validation set, and a test set. The first two sets are used during system development, while the test set is used only to evaluate the accuracy of the system. Cross-validation may also be used to improve performance.

[0158] Patient parameters such as age, sex, and type of arrhythmia can affect the results. Separate models can be trained based on some of these parameters, and these parameters can be introduced into some of the layers in the NN architecture to help learn differences based on these parameters.

[0159] The proposed method combines unsupervised information and performs fully unsupervised signal and anatomical structure encoding, but activation mapping can be performed using existing mappings.

[0160] Although the network is relatively simple, the same knowledge representation method may be useful for other queries and more complex arrhythmias. This integrated framework combines information available in standard Carto cases, including ECG, maps, and activation mapping.

[0161] In some embodiments, a complete CT / MRI is not available at the time of case presentation. NNs may be able to estimate anatomical structure shape using catheter location information acquired during the scan. For example, Vnet can achieve this estimation by substituting inputs.

[0162] Even after the system is ready and deployed in hospitals, additional data may be accumulated. This additional data may be added to the training dataset, allowing the system to be retrained and its accuracy to be continually improved. Specifically, data from additional operations may provide feedback by factoring in the success assessment of ablations performed according to the system's recommendations.

[0163] While features and elements are described above in particular combinations, those skilled in the art will understand that each feature or element can be used alone or in any combination with the other features and elements. In addition, the methods described herein may be implemented in a computer program, software, or firmware embodied in a computer-readable medium for execution on a computer or processor. Examples of computer-readable media include electronic signals (transmitted via wired or wireless connections) and computer-readable storage media. Examples of computer-readable storage media include, but are not limited to, read-only memory (ROM), random-access memory (RAM), registers, cache memory, semiconductor memory devices, magnetic media such as internal hard disks and removable disks, magneto-optical media, and optical media such as CD-ROM disks and digital versatile disks (DVDs). A processor in conjunction with software can be used to implement a radio frequency transceiver for use in a WTRU, UE, terminal, base station, RNC, or any host computer.

[0164] [Embodiment] (1) A method for improving mapping efficiency by suggesting locations of mapping points, comprising: receiving data at the machine, the data including a plurality of signals received during a triangulation procedure to identify the location of a focal tachycardia; generating by the machine a predictive model for the location of the focal tachycardia; modifying, by the machine, the predictive model based on additional data received by the machine; A method comprising: (2) The method of embodiment 1, wherein a focal point within the triangle is estimated using local activation points (LAT). (3) The method of embodiment 1, wherein the data is anatomical structure (FAM / CT) data and LAT points acquired by a Carto machine. (4) The method of embodiment 1, wherein the data includes ablation locations indicated as focal / terminal by a physician or a coherent map. (5) The method of embodiment 1, wherein the data is described at each stage by a single LAT point (location and activation).

[0165] (6) The method of embodiment 1, wherein the location of the next best point for sampling across the anatomical structure is expressed as spatial coordinates. (7) The method of embodiment 1, further comprising learning and using patient affinities to further provide more accurate predictions to physicians. (8) The method of embodiment 1, wherein the data includes information based on a specific disease. (9) The method of embodiment 1, wherein the data in input space is divided into grid-sampled voxels, each voxel containing an electrical activation signal measured within the voxel. (10) A system for locating a focal point, comprising: Multiple inputs and a first converter that converts at least a first portion of the plurality of inputs into a spatial ECG feature vector; a second transformer that transforms at least a second portion of the plurality of inputs into a spatial shape representation; a neural network that operates on the spatial ECG feature vectors and the spatial shape representation to generate a plurality of outputs; A system comprising:

[0166] (11) The system of embodiment 10, wherein the input includes at least one of a FAM mesh, LAT points, an ECG signal, an estimated focal point, and a set of ablation points. (12) The system of claim 10, wherein the output includes a vector to the source and a confidence level. (13) The system of claim 10, wherein the output includes the following regions of interest: (14) The system of embodiment 10, wherein the first converter is a first neural network. (15) The system of embodiment 14, wherein the first neural network is a transformer network.

[0167] (16) The system of embodiment 10, wherein the second converter is a second neural network. (17) The system described in embodiment 16, wherein the second neural network is a Vnet network. (18) The system of embodiment 10, wherein the neural network inputs query points. (19) At least one of the plurality of outputs is a confidence level

number

Claims

1. 1. A system for locating a source of a focal arrhythmia, comprising: a plurality of inputs including recorded ECG signals and the locations where the signals were recorded, and anatomical surface information; a first converter that converts at least a first portion of the plurality of inputs into a spatial ECG feature vector; a second transformer that transforms at least a second portion of the plurality of inputs into a spatial shape representation; a neural network that receives the spatial ECG feature vector and the spatial shape representation as inputs and generates a plurality of outputs; Equipped with The system wherein the plurality of outputs include a vector to an origin of the local arrhythmia and a confidence level representing the reliability of the vector to the origin for the currently acquired sampling point.

2. The system of claim 1 , wherein the plurality of inputs includes at least one or more of a FAM mesh, LAT points, an ECG signal, an estimated local arrhythmia source, and a set of ablation points.

3. The system of claim 1 , wherein the plurality of outputs includes a next region of interest for obtaining the plurality of inputs.

4. The system of claim 1 , wherein the first transformer is a first neural network.

5. The system of claim 4 , wherein the first neural network is a commutator network.

6. The system of claim 1 , wherein the second transformer is a second neural network.

7. The system of claim 6 , wherein the second neural network is a Vnet network.

8. The system of claim 1 , wherein the neural network inputs a query point.

9. At least one of the plurality of outputs may be a confidence level [Equation 1] The system of claim 1 , wherein:

10. At least one of the plurality of outputs is R ３ The system of claim 1 , wherein the source vector is directed toward a source of the local arrhythmia within the vertex.

Images