Improved computer heart simulation

Abstract

To provide systems, methods, and computer program products for enhanced computational heart simulations.SOLUTION: Methods to enhance the computational localization of cardiac arrhythmia sources are provided. A method may include receiving, from a first user, clinical data associated with a clinical case. The clinical data may include a patient anatomic information, diagnostic and / or treatment modalities, treatment parameters, treatment outcome, and medical literature. The clinical case may be indexed based on a first plurality of characteristics associated with the clinical data. The indexing may include associating at least a portion of the clinical data with a computational simulation of cardiac arrhythmia having a second plurality of characteristics matching the first plurality of characteristics. At least a portion of the clinical data associated with the indexed clinical case may be provided to a second user in response to a query from the user. Related systems and articles of manufacture are also provided.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] Related Applications This application claims priority to U.S. Provisional Patent Application No. 62 / 786,973, filed Dec. 31, 2018, entitled “HEART RELATED SYSTEMS AND METHODS”. The entire disclosure thereof is incorporated herein by reference. This application also incorporates by reference U.S. Patent 10,319,144, “Computational Localization of Fibrillation Sources”, regarding computer models and libraries of computer simulations.

[0002] The inventive subject matter described herein relates to computer modeling and simulation, and more particularly to improving computer modeling and simulation for identifying the location of the origin of cardiac arrhythmias in order to enable targeted therapy.

Background Art

[0003] Cardiac arrhythmia is a common medical disorder in which abnormal electrical signals in the heart cause the heart to contract in a suboptimal manner. As a result, abnormal heartbeats, or arrhythmias, may occur in the atria of the heart (e.g., atrial fibrillation (AF)), and / or in the ventricles of the heart (e.g., ventricular tachycardia (VT) or ventricular fibrillation (VF)). Treatments for cardiac arrhythmia attempt to address mechanisms that cause persistent and / or clinically significant symptoms, including, for example, stable electrical whirls, recurrent electrical focal origins, reentrant electrical circuits, and the like. If left untreated, cardiac arrhythmia can lead to serious complications including pathological conditions (e.g., loss of consciousness, stroke, etc.) and death (e.g., sudden cardiac death (SCD)).

Summary of the Invention

[0004] Products including systems, methods, and computer program products are provided for improved computer cardiac simulation. In some exemplary embodiments, a system is provided which includes at least one processor and at least one memory. The at least one memory may contain program code that provides operation when executed by the at least one processor. The operation may include: receiving clinical data associated with a clinical case from a first user; indexing the clinical case based on at least a first set of features associated with the clinical data, wherein the indexing includes associating at least a portion of the clinical data with computer simulations of cardiac arrhythmias having a second set of features that match the first set of features; and responding to an inquiry from a second user by sending at least a portion of the clinical data associated with the indexed clinical case to the second user.

[0005] In some variations, one or more of the features disclosed herein, including the following, may be included in any feasible combination. Clinical data may include patient anatomical information, diagnostic and / or treatment modalities, treatment parameters, treatment outcomes, and medical literature.

[0006] In some variations, the first and second sets of features may include the patient's demographics, medical history, and treatment plan.

[0007] In some variations, indexing may involve calculating a similarity score for each of the multiple computer simulations of cardiac arrhythmias contained in the library, indicating the degree of agreement between a first set of features associated with the clinical data and a second set of features associated with the individual multiple computer models and / or simulations. Indexing may further involve associating at least a portion of the data with the one of the multiple computational models and / or simulations that has the highest similarity score.

[0008] In some variations, at least some of the clinical data, including its association with computer simulations of cardiac arrhythmias, may be stored in a data store.

[0009] In some variations, the query may include the patient's vector electrocardiogram (VCG). Responding to the query may involve identifying the computer model of cardiac arrhythmia that most closely matches the patient's vector electrocardiogram, and retrieving at least a portion of the clinical data associated with the indexed clinical case for sending at least a portion of the clinical data to a second user.

[0010] In another embodiment, a method for improved computer cardiac simulation is provided. The method may include: receiving clinical data associated with a clinical case from a first user; indexing the clinical data based on at least a first set of features associated with the clinical case, wherein the indexing includes associating at least a portion of the clinical data with computer simulations of cardiac arrhythmias having a second set of features that match the first set of features; and responding to an inquiry from a second user by sending at least a portion of the clinical data associated with the indexed clinical case to the second user.

[0011] In some variations, one or more of the features disclosed herein, including the following, may be included in any feasible combination. Clinical data may include patient anatomical information, diagnostic and / or treatment modalities, treatment parameters, treatment outcomes, and medical literature.

[0012] In some variations, the first and second sets of features may include the patient's demographics, medical history, and treatment plan.

[0013] In some variations, indexing may include calculating a similarity score for each of several computer simulations of cardiac arrhythmias included in the library, indicating the degree of agreement between a first set of features associated with the clinical data and a second set of features associated with each of the computer simulations. Indexing may further include associating at least a portion of the data with the one of the computer simulations that has the highest similarity score.

[0014] In some variations, the method may further include storing at least some clinical data, including its association with computer simulations of cardiac arrhythmias, in a data store.

[0015] In some variations, the query may include the patient's vector electrocardiogram (VCG). Responding to the query may involve identifying the computer model of cardiac arrhythmia that most closely matches the patient's vector electrocardiogram, and retrieving at least a portion of the clinical data associated with the indexed clinical case for sending at least a portion of the clinical data to a second user.

[0016] In other embodiments, a computer program product is provided which includes a non-temporary computer-readable medium for storing instructions. The instructions may result in an operation which may be performed by at least one data processor. The operation may include: receiving clinical data associated with a clinical case from a first user; indexing the clinical data based on at least a first set of features associated with the clinical case, wherein the indexing includes associating at least a portion of the clinical data with computer simulations of cardiac arrhythmias having a second set of features that match the first set of features; and responding to an inquiry from a second user by sending at least a portion of the clinical data associated with the indexed clinical case to the second user.

[0017] In other embodiments, an apparatus for improved computer cardiac simulation is provided. The apparatus may include: means for receiving clinical data associated with clinical cases from a first user; means for indexing clinical data based on at least a first set of features associated with clinical cases, wherein the indexing includes associating at least a portion of the clinical data with computer models and simulations of cardiac arrhythmias having a second set of features that match the first set of features; and means for responding to inquiries from a second user by sending at least a portion of the indexed clinical data associated with clinical cases to the second user.

[0018] In other embodiments, a system is provided comprising at least one processor and at least one memory. The at least one memory may contain program code that provides an operation when executed by the at least one processor. The operation may include: receiving patient data collected during an electrophysiological processing procedure; modifying one or more computer models and / or simulations of an arrhythmia based on at least the patient data; determining the origin location of the arrhythmia based on the modified at least one or more computer models and / or simulations of the cardiac arrhythmia; and providing a display of the origin location of the cardiac arrhythmia to notify an action based on the patient data.

[0019] In some variations, one or more of the features disclosed herein, including the following, may be optionally included in any feasible combination: Patient data may include at least one of the following features: sustained action potential recovery data, conduction velocity recovery data, patient anatomical geometry, voltage mapping, intracardiac ultrasound data, transthoracic ultrasound data, cone-beam computed tomography data, fluoroscopy data, patient demographics, cardiac activation patterns, local conduction velocities, and electrographic features.

[0020] In some variations, modifications may include applying patient-specific enhancements to one or more computer models and / or simulations, including at least one of geometrically deforming and / or rotating them; imposing voltage and / or electrographic information onto one or more computer simulations; indicating activation information; adding global and / or local information regarding the thickness of cardiac structural walls; and incorporating global and / or geographic information regarding the location and morphology of papillary muscles, pulmonary veins, and / or left and right atrial appendages.

[0021] In some variations, the corrections can be made in real time or near real time. One or more computer simulations of the corrected cardiac arrhythmias can be returned to the user for clinical use.

[0022] In some variations, one or more computer models and / or simulations may be part of a library of non-patient-specific computer simulations of cardiac arrhythmias.

[0023] In some variations, arrhythmia simulations may be initiated based on at least one arrhythmia solution associated with one or more modified computer simulations of cardiac arrhythmias, thereby generating a patient-specific arrhythmia vector electrocardiogram library for use in computer arrhythmia mapping processing.

[0024] In some variations, arrhythmia simulations may be performed for each of several origin locations, based on at least one or more modified computer simulations of cardiac arrhythmias. The arrhythmia simulations may be based on the assumptions of the origin location. Multiple origin locations and corresponding arrhythmia simulations may form a patient-specific arrhythmia library for use in computer arrhythmia mapping processing.

[0025] In another aspect, a method for improved computer heart simulation is provided. The method may include: receiving patient data collected during an electrophysiological procedure; modifying at least one computer model and / or simulation of an arrhythmia based at least on the patient data; determining an origin location of the arrhythmia based at least on the modified at least one computer model and / or simulation of the cardiac arrhythmia; and providing a display of the origin location of the cardiac arrhythmia to notify a treatment based on the patient data.

[0026] In some variations, one or more of the features disclosed herein, including the following features, may optionally be included in any practicable combination. The patient data may include at least one of action potential duration recovery data, conduction velocity recovery data, the patient's anatomical geometry, voltage mapping, intracardiac ultrasound data, transthoracic ultrasound data, cone beam computed tomography data, fluoroscopy data, the patient's demographics, cardiac activation patterns, local conduction velocity, and electrogram features.

[0027] In some variations, the modifying step may include applying patient-specific improvements including at least one of geometrically deforming and / or rotating the at least one computer model and / or simulation; imposing voltage and / or electrogram information on the at least one computer simulation; indicating activation information; adding global and / or local information regarding the thickness of the cardiac structure walls; and incorporating global and / or geographical information regarding the position and morphology of the papillary muscles, pulmonary veins, and / or left and right atrial appendages.

[0028] In some variations, the modification may be performed in real time or near real time. The modified at least one computer simulation of the cardiac arrhythmia may be returned to the user for clinical use.

[0029] In some variations, one or more computer models and / or simulations can be part of a library of non-patient-specific computer simulations of cardiac arrhythmias.

[0030] In some variations, the method can further include initiating an arrhythmia simulation to generate a patient-tailored arrhythmia vector electrocardiogram library for use in computer arrhythmia mapping processing, based on at least one arrhythmia solution associated with one or more computer simulations of a corrected cardiac arrhythmia.

[0031] In some variations, the method can further include, for each of a plurality of origin locations, performing an arrhythmia simulation based at least on one or more computer simulations of a corrected cardiac arrhythmia. The arrhythmia simulation can be performed based on an assumption of the origin location. The plurality of origin locations and corresponding arrhythmia simulations can form a patient-tailored arrhythmia library for use in computer arrhythmia mapping processing.

[0032] In other aspects, a computer program product is provided that includes a non-transitory computer-readable medium storing instructions. The instructions can cause operations to be performed by at least one data processor. The operations can include receiving patient data collected during an electrophysiological procedure, modifying one or more computer simulations of an arrhythmia based at least on the patient data, determining an origin location of the arrhythmia based at least on the one or more modified computer simulations of the cardiac arrhythmia, and providing a display of the origin location of the cardiac arrhythmia to notify of a treatment based on the patient data.

[0033] In other embodiments, an apparatus for improved computer cardiac simulation is provided. The apparatus may include: means for receiving patient data collected during an electrophysiological processing procedure; means for modifying one or more computer simulations of arrhythmias based on at least the patient data; means for determining the origin location of an arrhythmia based on at least one or more modified computer models and simulations of the cardiac arrhythmia; and means for providing a display of the origin location of the cardiac arrhythmia to notify the patient of an action based on the patient data.

[0034] In other embodiments, a system is provided comprising at least one processor and at least one memory. The at least one memory may contain program code that provides an operation when executed by the at least one processor. The operation may include: determining the location of each of n pacing sites in an electroanatomical map where a catheter, pacemaker lead, or implantable cardioverter-defibrillator lead is placed when one or more pacing stimuli are applied; identifying a computer model and simulation of arrhythmias associated with a vector electrocardiogram that matches a patient's vector electrocardiogram collected while pacing at each of the n pacing sites, and selecting one or more corresponding pacing sites in the computer model; aligning the electroanatomical map and the computer model based at least on the t-position of each of the n pacing sites in the electroanatomical map and the computer model; and generating, based at least on the alignment, a representation of the origin of a clinically important cardiac arrhythmia in the computer model associated with the location of each of the n pacing sites.

[0035] In some variations, one or more of the features disclosed herein, including the following features, may be optionally included in any feasible combination: n pacing sites may include at least three pacing sites.

[0036] In some variations, the alignment process may include applying a transformation matrix to align the first reference coordinate system of the electroanatomical map with the second reference coordinate system of the computer simulation.

[0037] In some variations, the location of the origin of clinically significant cardiac arrhythmias can be further transformed into an electroanatomical mapping system based at least on a flattened ellipsoid coordinate system.

[0038] In some variations, the procedure may be applied to the site of origin of a clinically significant cardiac arrhythmia, at least based on indications. The procedure includes at least one of ablation, targeted gene therapy, radiotherapy, and surgical intervention.

[0039] In some variations, computer models, electroanatomical maps, and mapping results with n pacing sites can be displayed in a aligned state.

[0040] In other embodiments, a method for improved computer cardiac simulation is provided. The method may include: determining the location of each of n pacing sites in an electroanatomical map where a catheter, pacemaker lead, or implantable cardioverter-defibrillator lead is placed when one or more pacing stimuli are applied; identifying a computer model and a simulation of arrhythmias associated with a vector electrocardiogram matching a patient's vector electrocardiogram collected while pacing at each of the n pacing sites, and selecting one or more corresponding pacing sites in the computer model; aligning the electroanatomical map and the computer model based at least on the locations of each of the n pacing sites in the electroanatomical map and the computer model; and generating a representation of the origin of clinically important cardiac arrhythmias in the computer model associated with the location of each of the n pacing sites, based at least on the alignment.

[0041] In some variations, one or more of the features disclosed herein, including the following features, may be optionally included in any feasible combination: n pacing sites may include at least three pacing sites.

[0042] In some variations, the alignment process may include applying a transformation matrix to align the first reference coordinate system of the electroanatomical map with the second reference coordinate system of the computer simulation.

[0043] In some variations, the location of the origin of clinically significant cardiac arrhythmias can be further transformed into an electroanatomical mapping system based at least on a flattened ellipsoid coordinate system.

[0044] In some variations, the procedure may be applied, at least on indication, to the site of origin of a clinically significant cardiac arrhythmia, where the procedure may include at least one of ablation, targeted gene therapy, radiotherapy, and surgical intervention.

[0045] In some variations, the method may further include displaying a computer model, an electroanatomical map, and a mapping result comprising n pacing sites in a aligned manner.

[0046] In another embodiment, a computer program product is provided which includes a non-temporary computer-readable medium for storing instructions. The instructions may result in an action which may be performed by at least one data processor. The action may involve determining the position of each of n pacing sites in an electroanatomical map where a catheter, pacemaker lead, or implantable cardioverter-defibrillator lead is placed when one or more pacing stimuli are applied. This may include identifying, for each of the n pacing sites, a computer model and arrhythmia simulation associated with a vector electrocardiogram that matches the patient's vector electrocardiogram collected during pacing at each of the n pacing sites, and selecting one or more corresponding pacing sites in the computer model; aligning the electroanatomical map and the computer model based at least on the location of each of the n pacing sites in the electroanatomical map and the computer model; and generating, based at least on the alignment, a representation of the location of the origin of clinically important cardiac arrhythmias in the computer model associated with the location of each of the n pacing sites.

[0047] In another embodiment, an apparatus for improved computer cardiac simulation is provided. The apparatus may include: means for determining the location of each of n pacing sites in an electroanatomical map where a catheter, pacemaker lead, or implantable cardioverter-defibrillator lead is placed when one or more pacing stimuli are applied; means for identifying a computer model and simulation of arrhythmias associated with a vector electrocardiogram matching a patient's vector electrocardiogram collected while pacing at each of the n pacing sites, and for selecting one or more corresponding pacing sites in the computer model; means for aligning the electroanatomical map and the computer model based at least on the t-position of each of the n pacing sites in the electroanatomical map and the computer model; and means for generating a representation of the origin of clinically important cardiac arrhythmias in the computer model associated with the location of each of the n pacing sites, based at least on the alignment.

[0048] The implementation of the inventive features of the present invention may include a consistent system and method, and includes the description of one or more features, and an article including a substantially embodied machine-readable medium on which one or more machines (e.g., a computer) are capable of operating to produce the operations described herein. Similarly, a computer system may include one or more processors and one or more memories coupled to one or more processors. The memory, which may include a computer-readable storage medium, may include coding and storing one or more programs that cause one or more processors to perform one or more of the operations described herein. A method executed by a computer, consistent with the implementation of one or more features of the present invention, may be executed by one or more data processors present in a single computing system or in a number of computing systems. Such complex computing systems may be linked together via one or more connections, exchanging data and / or commands or other instructions, etc., such connections include, for example, connections on a network (e.g., the Internet, a wireless wide area network, a local area network, a wide area network, a wired network, etc.), direct connections between one or more complex computing systems, etc.

[0049] Details of one or more variations of the inventive features described herein are made apparent in the accompanying drawings and description below. Other features and advantages of the inventive features described herein may be evident from the description, drawings, and claims. Certain features of the inventive features currently disclosed are described for illustrative purposes relating to computer cardiac simulation, but it should be readily apparent that such features are not intended to be limiting. The claims following this disclosure are intended to clarify the scope of the inventive features that are protected.

[0050] The accompanying drawings incorporated herein and constituting part thereof illustrate certain aspects of the inventive features disclosed herein and, together with the description, help to explain some of the principles associated with the disclosed practices. In those drawings, [Brief explanation of the drawing]

[0051] [Figure 1] This diagram illustrates an example of a cardiac arrhythmia management system in several exemplary embodiments. [Figure 2] The following diagram illustrates an example of data flow in a cardiac arrhythmia management system, based on several exemplary embodiments. [Figure 3A] Examples of data associated with a clinical case are shown in several exemplary embodiments. [Figure 3B] Examples of user interfaces are shown through several exemplary embodiments. [Figure 3C] This illustrates another example of data associated with a particular clinical case, using several exemplary embodiments. [Figure 3D] This illustrates another example of data associated with a particular clinical case, using several exemplary embodiments. [Figure 4] A flowchart illustrates an example of the process for modifying a library of non-patient-specific computer simulations, using several exemplary embodiments. [Figure 5A] Examples of data collected during a certain electrophysiological procedure, according to several exemplary embodiments, are shown. [Figure 5B] This illustrates another example of data collected during a certain electrophysiological procedure, according to several exemplary embodiments. [Figure 5C] This illustrates another example of data collected during a certain electrophysiological procedure, according to several exemplary embodiments. [Figure 6A] This section illustrates some computer models of cardiac arrhythmias using several exemplary embodiments. [Figure 6B]Examples of electroanatomical mapping are shown by several exemplary embodiments. [Figure 6C] This illustrates examples of electroanatomical mapping, including pacing sites, using several exemplary embodiments. [Figure 6D] This illustrates an example of a computer model of a cardiac arrhythmia with a specified pacing site, using several exemplary embodiments. [Figure 7A] The geometry of an oblate ellipsoid coordinate system is represented by several exemplary embodiments. [Figure 7B] This shows a flattened ellipsoid coordinate system according to several exemplary embodiments. [Figure 8A] This illustrates examples of electroanatomical mapping, including the geometric shape and ablation sites of the heart, according to several exemplary embodiments. [Figure 8B] This describes examples of stereotactic ablative radiotherapy using several exemplary embodiments. [Figure 9A] A flowchart illustrates an example of a process for improving a library of computer simulations using data associated with clinical cases, based on several exemplary embodiments. [Figure 9B] A flowchart illustrating an example of the process for modifying a computer simulation library, using several exemplary embodiments, is shown. [Figure 9C] A flowchart illustrating examples of processes for aligning computer simulations with electroanatomical mappings, according to several exemplary embodiments, is shown. [Figure 10] This shows a block diagram illustrating a computer system in several exemplary embodiments.

[0052] Where useful, similar reference numbers indicate similar structures, features, or elements. [Modes for carrying out the invention]

[0053] Cardiac arrhythmias (e.g., atrial fibrillation, ventricular tachycardia, ventricular fibrillation) may be treated by targeting persistent and / or clinically significant symptom-inducing mechanisms, including, for example, stable electrical rotation, recurrent focal electrical origins, and reentrant electrical circuits. Ablation is an example of cardiac arrhythmia treatment in which radiofrequency, cryotherapy, ultrasound, and / or radiation (e.g., stereotactic resective radiotherapy (SAbR)) may be applied to the origin of the cardiac arrhythmia. As a result, the lesion may alleviate the cardiac arrhythmia by interrupting and / or eliminating the unstable electrical signals that lead to abnormal cardiac activation. Nevertheless, the outcome of ablation can depend on various factors, including the precise localization of the arrhythmia origin. Precisely localizing the arrhythmia origin using existing techniques remains a challenge. Furthermore, insufficient access to prior clinical case data, including relevant patient anatomy, treatment parameters, and treatment outcomes, can be a further disadvantage for certain practitioners treating patients with cardiac arrhythmias. Therefore, various embodiments of the features of the present invention include techniques for improving the localization of the origin of cardiac arrhythmias and enhancing computer cardiac simulations in order to facilitate diagnosis and targeted treatment.

[0054] In some exemplary embodiments, a library containing multiple computer models and / or simulations of cardiac arrhythmias (as described in U.S. Patent 10,319,144, “Computational Localization of Fibrillation Sources”) may be enhanced with clinical data associated with clinical cases. For example, a first user may send clinical data associated with a clinical case, including patient anatomical information, data such as voltage map or electrographic features, diagnostic and / or treatment modalities, treatment parameters, treatment outcomes, relevant medical literature, etc., to a data controller combined with the library. The contents of the library may be indexed based on the specific features of the computer models and simulations of cardiac arrhythmias. For example, upon receiving clinical data from a first user, the controller may be configured to index the clinical case by at least identifying computer models and simulations that match the clinical case in the library and associating the corresponding clinical data with the matching computer models and simulations. A second user treating a patient's cardiac arrhythmia can, by querying the library based on patient data, not only gain access to computer simulations of matching cardiac arrhythmias, but also to relevant clinical data, such as the patient's anatomical information, diagnostic and / or treatment modalities, treatment parameters, treatment outcomes, and relevant medical literature.

[0055] In some exemplary embodiments, non-patient-specific computer models and simulations of cardiac arrhythmias included in a library may be enhanced using patient data collected during electrophysiological (EP) investigations, including, for example, sustained action potential (APD) recovery data, conduction velocity recovery data, patient anatomical geometry, voltage mapping, intracardiac ultrasound data, transthoracic ultrasound data, conventional computed tomography (CT) data, cone-beam computed tomography (CT) data, positron emission tomography (PET) scan data, fluoroscopy, magnetic resonance imaging data, patient demographics, cardiac activation patterns, local conduction velocity, electrogeographic analysis, etc. For example, a controller coupled to the library may be configured to modify one or more non-patient-specific computer simulations included in the library based on at least patient data. Such modifications may be made in real time (or near real time) so that the modified library of computer simulations is available when the patient undergoes treatment for the arrhythmia. For example, the localization of the origin of a cardiac arrhythmia may be performed based on the modified library of computer simulations before ablation is performed on the origin of the arrhythmia.

[0056] A computer model of a patient's anatomical structure, such as a computer representation of the patient's heart, can be used to provide supplementary information for procedures such as targeted ablation of the origin of cardiac arrhythmias in the patient's heart. While the results of computer mapping can visually identify the location of the origin of cardiac arrhythmias within the computer model, the precise relationship between the computer model, the electroanatomical map of the heart, and the patient's actual anatomy may be unclear. Therefore, in some exemplary embodiments, a computer model of a patient's anatomical structure can first be aligned to an electroanatomical map by tracking the location of one or more catheters, pacemaker leads, or implantable cardioverter-defibrillator (ICD) leads relative to the patient's anatomy. Next, the locations of n pacing sites where the catheter, pacemaker leads, or ICD leads are positioned when one or more pacing stimuli are applied can be identified in both the computer model and the electroanatomical map to provide n reference locations for aligning the computer model with the electroanatomical map. Finally, ablation may be performed on the origin of cardiac arrhythmias (e.g., ventricular fibrillation determined by the mapping results of a computer model) using a positional reference to the locations of n pacing sites identified in an anatomical computer simulation of the patient.

[0057] Figure 1 shows a diagram illustrating an example of a cardiac arrhythmia management system (100) according to several exemplary embodiments. According to Figure 1, the cardiac arrhythmia management system (100) may include a data controller (110) and a data store (120). As shown in Figure 1, the data controller (110) and the data store (130) may be communicably coupled via a network (140). Furthermore, Figure 1 shows a data controller (110) communicably coupled via the network (140) to one or more clients, including, for example, a first client (140a) associated with a first user (145a), a second client (140b) associated with a second user (145b), and so on. A first user (145a) in a first client (140a), and a second user (145b) in a second client (140b), may access the contents of a data store (120) via a data controller (110), which may include a library (125) of computer simulations of cardiac arrhythmias. It will be recognized that various technologies, including, for example, access control, encryption, and blockchain, may be applied to secure and / or anonymize data stored and / or transmitted within the cardiac arrhythmia management system (100).

[0058] In some exemplary embodiments, computer models and libraries (125) for computer simulation of cardiac arrhythmias may be enhanced with clinical data associated with clinical cases. For further explanation, Figure 2 shows a flowchart illustrating an example of data flow (200) in a cardiac arrhythmia management system (100). According to Figures 1 and 2, a first user (145a) of a first client (140a) may send clinical data associated with a clinical case to a data controller (110), including patient anatomical information (210a), diagnostic and / or treatment modalities (210b), treatment parameters (210c), clinical outcomes (210d), relevant medical literature (210e), etc., as shown in Figure 2.

[0059] As shown again in Figure 2, examples of patient anatomical information (210a) may include the geometric shape of the heart, the location of scars and fibrosis, the anatomy and pathophysiology of the chest, etc. Patient anatomical information (210a) may be captured during and / or prior to the clinical case in imaging studies. Alternatively and / or additionally, patient anatomical information (210a) may be captured during electroanatomical mapping processing procedures. Patient anatomical information (210) may be loaded into the library (120) by loading raw imaging information, for example, including text files containing patient information data (e.g., output data files containing electrophysiological information from an electrophysiological recording system), intracardiac ultrasound images, transthoracic ultrasound images, computed tomography (CT) images, 4D computed tomography video, magnetic resonance imaging (MRI) images, myocardial perfusion imaging (MIBI), positron emission tomography (PET) images, radiographs, etc. Tomographic images can utilize a spectrum of interpretations, ranging from human interpretation to automated three-dimensional image generation and analysis.

[0060] Furthermore, loading patient anatomical information (210) into the library (120) may include importing digital information such as geometric shape, catheter position, voltage maps, activation data, and analysis data from three-dimensional electroanatomical mapping. Figure 3A shows an example of a voltage map (310) showing the relationship between voltage and scar / fibrosis density at various locations across the left and right ventricles of the patient's heart. It will be recognized that at least a portion of the patient anatomical information (210a) may include annotations provided by a first user (145a). Figure 3B shows an example of a user interface (320) generated by the data controller (110). The user interface (320) may be displayed, for example, in a first client (140a) to receive one or more inputs from a first user (145a) corresponding to geometric shape interpretation, orientation, voltage, activation, and analysis information from the electroanatomical mapping system.

[0061] Examples of diagnostic and / or treatment modalities (210b) may include imaging techniques (e.g., fluoroscopy, ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), etc.), sheaths (e.g., pre-formed sheaths, maneuverable sheaths, etc.), mapping catheters (e.g., multi-electrode catheters), and ablation catheters (e.g., solid or perfusion type, 8 mm or 3.5 mm tip size, etc.). In some exemplary embodiments, the data controller (110) may generate a user interface including a dropdown menu (or another type of graphical user interface element) configured to allow a first user (145a) to input the diagnostic and / or treatment modality applied to the clinical case. Alternatively, and / or additionally, the data controller (110) may receive from the first user (145a) scans that identify one or more products used during the clinical case, including, for example, a product barcode (e.g., a barcode from a box containing the ablation catheter used in the clinical case), an image, etc.

[0062] Examples of procedure parameters (210c) may include parameters associated with ablation, such as ablation power, location, duration of lesion placement, and dimensions and / or shape of the lesion. For example, digital information including one or more procedure parameters (210c) may be exported from the electroanatomical mapping system and uploaded to the library (125). Figure 3C shows an example of an electroanatomical map showing the location of ablated lesions as dots. Alternatively and / or additionally, the data controller (110) may generate a user interface that may be displayed to a first client (140a) to receive one or more inputs from a first user (145a) corresponding to interpretations of ablation power, location, duration of lesion placement, and dimensions and / or shape of the lesion.

[0063] Additional examples of treatment parameters (210c) may include parameters relating to stereotactic radiotherapy (SAbR), such as target contour data, internal treatment volume (ITV), planned treatment volume (PTV), radiation dose, radiant energy / irradiation time, avoidance structures, respiratory and cardiac motion gating parameters, patient positioning and / or restraint devices, use of anesthetic anesthetics during treatment, parameters for programming pacemakers or implantable cardioverter-defibrillators (ICDs), cardiac rhythm during treatment, arrhythmia mapping techniques, relevant computed tomography (CT) imaging data, relevant magnetic resonance imaging (MRI) imaging data, relevant ultrasound imaging and tracking data, medication, antiarrhythmic therapy, anticoagulant therapy, clinical outcomes, complications, and adverse events. For example, digital information containing one or more treatment parameters (210c) may be exported from the stereotactic radiotherapy (SAbR) planning system and uploaded to the library (125). Figure 3D shows an example of planning software for stereotactic resection radiotherapy (SAbR), illustrating the target volume, avoidance structures, and calculated radiotherapy dose. Alternatively and / or additionally, the data controller (110) may generate a user interface in a first client (140a) to receive one or more inputs from a first user (145a), the one or more inputs corresponding to the interpretation of target contour data, internal treatment volume (ITV), planned treatment volume (PTV), radiation dose, radiant energy / irradiation time, avoidance structures, respiratory and cardiac motion gating parameters, patient positioning and / or restraint devices, use of anesthetic anesthetics during treatment, parameters for programming a pacemaker or implantable cardioverter-defibrillator (ICD), cardiac rhythm during treatment, arrhythmia mapping techniques, associated computed tomography (CT) imaging data, associated magnetic resonance imaging (MRI) imaging data, associated ultrasound diagnostic and tracking data, medication, antiarrhythmic therapy, anticoagulant therapy, clinical outcomes, complications, and adverse events.

[0064] Clinical outcomes (210d) may include results associated with an ablation having treatment parameters (210c), such results may include, for example, short-term ablation success (e.g., ablation that terminates arrhythmia, ablation that makes arrhythmia non-inducible, 6-month clinical outcome, etc.) and complications. Examples of relevant medical literature (210e), shown in Figure 2, include guidelines, clinical trials, expert opinions, and case reports related to clinical cases, and are indexed according to parameters of a library of computer models and arrhythmia simulations (e.g., arrhythmia type, geometric shape of the patient's heart and scar composition, arrhythmia origin location, etc.). The data controller (110) may generate a user interface including a dropdown menu (or another type of graphical user interface element) configured to allow a first user (145a) to select one or more inputs of clinical outcomes.

[0065] In some exemplary embodiments, the data controller (110) may be configured to index clinical data received from a first user (145a), so that the clinical data may be accessible, for example, to a second user (145b) in a second client (140b). The data controller (110) may also be configured to index clinical data received from a first user (145a) based at least on computer simulations contained in a library (125), so that the clinical data is associated with the computer simulation that most closely matches the corresponding clinical case. For example, each computer simulation contained in the library (125) may be associated with one or more features, such as the size and shape of the heart, the presence or absence of structural cardiac disease, the type of arrhythmia, etc. The data controller (110), upon receiving clinical data associated with a clinical case from a first user (145a), may be configured to calculate a similarity score for each computer simulation in the library (125), indicating the degree of agreement between the clinical case, the computer model, and the respective features of the simulation library (125). The clinical data associated with a clinical case may be indexed based on the computer model with the highest similarity score. That is, the clinical data associated with a clinical case may be associated with the computer model and / or simulation whose features (e.g., heart size, shape, presence or absence of structural cardiac disease, arrhythmia type, etc.) most closely match the features in the clinical case.

[0066] The contents of a simulation library (125), including computer models and / or simulations enhanced with clinical data associated with matching clinical cases, may be accessible to a second user (145b) of a second client (140b). For example, the second user (145b) may query the library (125) to identify important clinical cases. In some exemplary embodiments, the second user (145b) may be treating a patient with cardiac arrhythmia and therefore may query the library (125) based on patient data, including, for example, the patient's age, medical history, arrhythmia type, proposed treatment plan, etc. The data controller (110) may respond to a query from the second user (145b) by identifying at least one or more clinical cases contained in the library (120) that match the parameters of the query from the second user (145b). Alternatively, and / or additionally, instead of one or more specific clinical cases, a second user (145b) may apply a collection of clinical cases from Library (120) as training data to train a machine learning model to perform a variety of cognitive tasks, including, for example, determining the statistical probability of the origin of an arrhythmia, or performing a probabilistic analysis of potential clinical outcomes associated with various treatment approaches for arrhythmias, etc. (e.g., the location and / or volume, number, and pattern of ablation lesions).

[0067] In some exemplary embodiments, the machine learning model may include, for example, a neural network such as an autoencoder. The machine learning model may be trained on training data that includes clinical data from many patient cases that may be collected and fed into the machine learning model as input. The training data may include ground truth labels that include patient demographic information, electrocardiogram (ECG) and vector electrocardiogram (VCG) tracing, and identified arrhythmia origin locations. The arrhythmia origin locations may be further labeled with several rankings of ablation site, size, volume, and technique (e.g., catheter ablation or stereotactic radiotherapy), and outcome (e.g., arrhythmia termination, short-term ablation success, long-term ablation success, etc.). In addition, the machine learning model may be trained to consider features in the treatment approach for each patient (number, size, volume, composition, therapeutic dose, etc.). Furthermore, the machine learning model may be trained to seek similarity metrics between different clinical cases based on demographics, arrhythmia type, cardiac anatomy, etc., to determine relevance to both other training case data and / or future cases to be compared. When a user wishes to utilize a trained machine learning model, they may provide the patient's electrocardiogram (ECG) or vector electrocardiogram (VCG), as well as one or more patient and arrhythmia characteristics, as input to the trained machine learning model. Based on the input, the trained machine learning model may seek statistical probabilities of the arrhythmia origin location and probabilistic analyses of potential clinical outcomes associated with various treatment approaches for the arrhythmia (e.g., location, number, volume, composition, therapeutic dose, etc. of ablation lesions). In some examples, a data controller (110) may be configured to identify a selection of key clinical cases for case reference and procedural planning, based at least on the output of the trained machine learning model.

[0068] In some exemplary embodiments, non-patient-specific computer models and arrhythmia simulations included in the library (125) may be enhanced using patient data collected during electrophysiological investigations (EPS) in either an electrophysiology laboratory, radiology suite, or operating room (OR), such patient data including, for example, sustained action potential (APD) recovery data, conduction velocity recovery data, patient anatomical geometry, voltage mapping, intracardiac ultrasound data, transthoracic ultrasound data, conventional computed tomography data, cone-beam computed tomography data, four-dimensional computed tomography (4-DCT) data, magnetic resonance imaging (MRI) data, positron emission tomography (PET) data, patient demographics, cardiac activation patterns, local conduction velocity, electrogeographic analysis, and the like. For example, a data controller (110) may be configured to modify one or more non-patient-specific computer models and arrhythmia simulations included in the library (125) based on at least patient data. In this example, the geometric and voltage maps of the patient's left ventricle, generated by the electroanatomical mapping system during the ablation case, are exported to a USB memory stick and uploaded to the algorithm. The cardiac model is updated to include information on the size, orientation, and location of normal tissue, scar tissue, and fibrosis of the left ventricle. The previously calculated voltage solutions for cardiac arrhythmias are then incorporated into the updated cardiac model, and the voltage solutions are advanced in time to compute a vector electrocardiogram (VCG) library for the patient, with one or more VCG loops associated with each of the possible locations of cardiac arrhythmia origin. The adjusted VCG library and associated locations, and other associated metadata, are then returned to the clinical user to assist the clinical case being performed. The modifications may be made in real time (or near real time) so that the modified library (125) of the computer simulation is available when the patient is being treated for the arrhythmia.For example, the location of the origin of ventricular fibrillation can be determined based on a modified library of computer simulations (125) before ablation is performed at the origin of ventricular fibrillation.

[0069] Figure 4 represents a flowchart illustrating an example of the process (400) for modifying a library of non-patient-specific computer simulations by several exemplary embodiments. As shown in box A, the patient is led into an electrophysiology laboratory, radiology suite, or operating room, placed on a table for the procedure, and the procedure is initiated (the electrophysiological investigation (EPS) environment is represented by the bottom of Figure 4). Next, as shown in box B, the geometric shape of the patient's heart is made using a combination of non-invasive techniques (e.g., transthoracic ultrasound, fluoroscopy, cone-beam tomography, magnetic resonance imaging, etc.) and / or invasive techniques (e.g., an invasive electrophysiological catheter is placed in the heart and manipulated throughout the heart). The geometric shape of the heart is supplemented by APD recovery data, conduction velocity recovery data, voltage mapping data, intracardiac ultrasound data, patient demographic data [age, weight, height, ejection fraction], cardiac activation patterns, local conduction velocity, and electrographic analysis. Next, this data is collected and securely exported to high-performance computing resources for analysis (the area represented by the top of Figure 4). In this environment, an existing non-patient-specific library of computer simulations of cardiac arrhythmias (Box 1) is rapidly compared and adjusted according to the exported data (Box 2). In one example, the cardiac model is updated to include information on the size and orientation of the left ventricle, and the location of normal tissue, scar tissue, and fibrosis. Then, the voltage solution calculated previously for cardiac arrhythmias is incorporated into the updated cardiac model, and the solution is then advanced in time to compute a vector electrocardiogram (VCG) library for the patient, with one or more VCG loops associated with each of the possible locations of cardiac arrhythmia origin. As shown in Box 3, the adjusted library, voltage solution, and arrhythmia origin locations are returned to a local electrophysiology laboratory, radiology suite, or operating room (OR) mapping system for the patient's arrhythmia mapping. Meanwhile, in the clinical case (Box C), arrhythmia induction is attempted if necessary.The arrhythmia electrophoresis is saved and exported for analysis. The diagnostic catheter is removed (if any), and either an ablation catheter is placed in the heart, the plan for stereotactic radiotherapy (SAbR) is reviewed, or a surgical plan is evaluated. In box D, arrhythmia origin mapping is performed using the modified VCG library from box 3. The arrhythmia origin location (result of the computer mapping process) is displayed for the physician's interpretation. The mapping results inform the initiation of catheter ablation, stereotactic radiotherapy, or surgical intervention for the arrhythmia origin (box E).

[0070] As shown in Figure 4, the data controller (110) may be configured to modify a library (125) based on patient data collected during an electrophysiological (EP) study, as noted in Figure 4, the patient data including patient demographics and information obtained by placing one or more catheters in the patient's heart or from the urge signals of an implanted pacemaker or implantable cardioverter-defibrillator (ICD), thereby allowing the collection of patient-specific data such as sustained action potential (APD) recovery data, conduction velocity recovery data, the patient's anatomical and geometric shape, voltage mapping, intracardiac ultrasound data, cardiac activation patterns, local conduction velocity, and electrographic analysis. The data controller (110) may modify the library (125) by applying one or more patient-specific corrections so that one or more of the computer simulations contained in the library (125) better match the patient's specific characteristics.

[0071] To further illustrate, Figures 5A–C represent data collected during a certain electrophysiological procedure according to several exemplary embodiments. For example, Figure 5A represents an example of the outcome of a single extrastimulus pacing in the atrium, which may illustrate atrial action potential sustained (APD) recovery and activation latency. The action potential sustained (APD) recovery and activation latency shown in Figure 5A may be used to determine precise parameters for a more accurate simulation of atrial arrhythmias in a patient's heart.

[0072] Figure 5B shows an example of endocardial geometry and voltage maps in a patient with non-ischemic cardiomyopathy and ventricular arrhythmias. A considerable amount of data related to the arrhythmia simulation process can be generated during electrophysiological mapping using a three-dimensional electroanatomical mapping system. For example, the geometric shape and orientation of the heart can be obtained by moving an electrophysiological catheter within the heart. The set of points occupied by such a catheter may be used to generate the endocardial and epicardial surfaces of the heart, for example, shown in the geometric shapes of the endocardium of the left ventricle (geometric shape on the left side of the figure) and right ventricle (geometric shape on the right side of the figure) shown in Figure 5B.

[0073] Figure 5C shows an example of an intracardiac echocardiography (ICE) image of the left ventricle, where the endocardial surface (down arrow) and the basket of catheter splines (up arrow) are shown. Intracardiac echocardiography (ICE) systems or transthoracic echocardiography systems may collect dynamic high-resolution data regarding the thickness of the cardiac wall, the location and thickness of various structures (e.g., papillary muscles, pulmonary veins, left and right atrial appendages), and the location of other mapping and ablation catheters. The resulting echocardiographic images, as shown in Figure 5C, may then be used to further refine one or more non-patient-specific computer simulations that match patient-specific cardiac features, as described by the process (400) in Figure 4 and as described above.

[0074] In some exemplary embodiments, patient data may be exported from the electroanatomical mapping system and transferred directly to a data controller (110). For example, the patient's geometric shape, voltage map, activation map, and electromorphic map may be saved as data files (e.g., Universal Serial Bus (USB) memory stick, Compact Disc (CD), Digital Versatile Disc (DVD), etc.) before being uploaded to the data controller (110). Alternatively, when direct export is not feasible, the data controller (110) may provide a user-editable cardiac model and customizable tools, for example, via a graphical user interface, for geometric deformation and rotation, the imposition of voltage and / or electromorphic information onto computer models and arrhythmia simulations, and the display of activation information. Global and / or local information regarding the thickness of cardiac structural walls, as well as the location and morphology of the papillary muscles, pulmonary veins, and left and right atrial appendages, may be incorporated into the model by deforming the geometric shape or setting the wall thickness. For example, it will be recognized that various technologies, including access control, encryption, and blockchain, can be applied to secure and / or anonymize data transmitted to and from the data controller (110).

[0075] In some exemplary embodiments, upon receiving patient data, the data controller (110) may be configured to modify one or more computer models and arrhythmia simulations in the library (125) in real time or near real time, based on at least the patient data. Note that such modification may include one or more patient-specific corrections so that the computer simulations in the library (125) better match patient-specific features. For example, the data controller (110) may be configured to fit geometric data to a computer mesh, and as a result, mesh relationships may be utilized to compute arrhythmia simulations for the patient of interest. The data controller (110) may also introduce previously simulated voltage solutions and local origins into a more patient-specific model. The computer simulation may then begin to allow arrhythmia maturation and permutations to be recorded (for example, for a few seconds of pseudo-time). From the computational voltage solutions, the data controller (110) may compute and record vector electrocardiogram (VCG) data, which may be indexed to the origin locations of the arrhythmias. Alternatively, and / or additionally, computer renderings of voltage solutions can be performed and recorded as resources for mapping and verifying arrhythmias (e.g., the technology described in U.S. Patent 10,319,144, “Computational Localization of Fibrillation Sources”).

[0076] As noted, the data controller (110) may modify the library (125) in real time (or near real time) so that the modified library (125) of computer simulations is available when a patient is being treated for ventricular fibrillation. For example, a computer-generated vector electrocardiogram and associated origin location, along with a represented voltage solution, may be encrypted and sent to a first client (140a) and / or a second client 140b so that at least the first user 145a and / or the second user (145b) can determine the location of the patient's ventricular fibrillation before performing a procedure such as ablation at the origin of the ventricular fibrillation. It will be understood that the modified library (125) may enable the first user (145a) and / or the second user (145b) to perform higher-fidelity localization of the origin of cardiac arrhythmias, thereby improving the clinical outcomes of procedures targeting the origin of cardiac arrhythmias.

[0077] In some exemplary embodiments, the data controller (110) may be configured to align a computer model used for arrhythmia simulation and computer arrhythmia mapping with the patient's anatomy, using a three-dimensional electroanatomical map that tracks the location of one or more catheters. This can be achieved through the following workflow: First, n pacing procedures are performed in the patient's heart using either a directional catheter or pacing electrodes of a pacemaker or implantable cardioverter-defibrillator. Next, the pacing sites are recorded in the patient's heart using a three-dimensional electroanatomical mapping system. Then, the n pacing sites can be identified in the computer arrhythmia mapping system by analyzing each of the paced QRS complexes (e.g., calculating a vector electrocardiogram from the pacing QRS complexes) and comparing it to a library of simulated pacing vector electrocardiograms. The vector electrocardiogram with the highest similarity score may provide information about the location of the pacing site by heartbeat. Next, the geometric shapes of the computer model and the electroanatomical mapping system are combined to best superimpose the locations of n pacing sites, either by exporting the geometric shapes of the electroanatomical mapping into the computer model arrhythmia mapping system, or by exporting the geometric shapes of the computer model into the electroanatomical mapping system, or conceptually, for example, by using a least-squares fitting algorithm. Then, ablation can be performed at the origin of the cardiac arrhythmia, referencing the locations of the n pacing sites identified in the computer simulation of the patient's anatomy.

[0078] Figure 6A shows an example of a computer model (600) illustrating the origin of ventricular fibrillation in several exemplary embodiments. The computer simulation (600) shown in Figure 6A may be a “heatmap” showing the location of the origin of cardiac arrhythmias in a patient. The example of the computer model and mapping solution (600) shown in Figure 6A may be generated, for example, from clinical 12-lead electrocardiogram (ECG) data of the arrhythmia of interest and its calculated vector ECG, and matched to a vector ECG library of simulated arrhythmia simulations. While the computer model (600) shows the location of the origin of cardiac arrhythmias, it may lack a precise relationship between the patient’s anatomy and the geometric shape of the computer simulation (600). Therefore, the computer model and mapping solution (600) alone may not provide sufficient practical data for clinicians treating cardiac arrhythmias in patients.

[0079] To provide the precise location of the origin of cardiac arrhythmias in relation to the patient's anatomy, the data controller (110) may be configured to align a computer simulation of the patient's anatomy, for example, the computer model and mapping output (600) shown in Figure 6A may be aligned to an electroanatomical mapping (610) shown in Figure 6B. According to Figure 6B, the electroanatomical mapping (610) may track the position of one or more catheters (indicated by the upper arrows) or pacing electrodes of a pacemaker or implantable cardioverter-defibrillator (ICD) relative to the patient's anatomy (e.g., the left ventricle), as well as low-voltage areas (indicated by the lower arrows). Thus, in one exemplary embodiment, the data controller (110) may determine, based at least on the electroanatomical mapping, the positions of n (e.g., 3 or more) pacing sites where the catheter or pacing electrodes of a pacemaker or implantable cardioverter-defibrillator are placed when one or more pacing stimuli are applied. To further illustrate, Figure 6C shows an example of an electroanatomical map (620) including pacing sites (indicated by arrows). A data controller (110) can further determine the locations of the same n pacing sites in a computer mapping solution (600). Figure 6D shows a computer model (600) with the n pacing sites indicated by small white dots (indicated by arrows). Notably, as shown in Figure 6D, the location of the origin of a cardiac arrhythmia can be referenced in relation to the locations of the n pacing sites. As described above, the geometric shape of the computer model can be aligned to the geometric shape of a three-dimensional electroanatomical mapping system (or vice versa) using a three-dimensional least-squares fitting algorithm that references the locations of the n pacing sites. This alignment process may result in the generation and display of an updated Figure 6D to the user, and targeted therapy can be delivered more accurately from the site of interest (labeled "Origin of Arrhythmia") in Figure 6D.

[0080] In some exemplary embodiments, the location of the origin of a cardiac arrhythmia may be translated into an electroanatomical mapping system using a flattened ellipsoid coordinate system, which serves as a reference system for the ventricles. Figures 7A and 7B show the flattened ellipsoid coordinate system (700) in some exemplary embodiments. As shown in Figures 7A and 7B, the locations in the flattened ellipsoid coordinate system (700) may be denoted as σ, τ, and φ, where σ = cosh(μ) and τ = cos(ν).

[0081] Once the locations of the n pacing sites are known in computer simulations (600) and electroanatomical maps (620), the locations of the n pacing sites can be used to align the respective reference coordinate systems of the computer simulations (600) with the electroanatomical maps (620) using a deformation matrix A. The location of the origin of cardiac arrhythmias is further clarified based on the locations of the n pacing sites and represented by a tuple σ in the associated flattened ellipsoid coordinate system (700). source , τ source , and φ sourceThe locations of the origins of cardiac arrhythmias for n pacing sites can be practical data for clinicians treating patients with cardiac arrhythmias. For example, the computer model in Figure 6D can be aligned with the electroanatomical mapping geometric shape in Figure 6C using least-squares fitting. The fitted geometric shape could be transformed into a reference geometric shape through a process combining rotation, scaling, and translation (e.g., within a flattened ellipsoid coordinate system). The new image of the combined and aligned data ("updated" Figure 6D) can be generated and displayed to enable the user to precisely target the origin of the arrhythmia. Specifically, procedures including ablation, targeted gene therapy, stereotactic radiotherapy (e.g., gamma rays, proton beams), and surgical interventions can be performed at the locations identified as the origins of the cardiac arrhythmias. For example, Figure 8A shows the geometric shapes of the left and right ventricles, which have multiple ablation sites to which radiofrequency, cryotherapy, ultrasound, and / or stereotactic radiotherapy may be applied to alleviate cardiac arrhythmias by eliminating and / or interrupting unstable electrical signals that cause asynchronous myocardial contractions associated with cardiac arrhythmias. Figure 8B shows an example of stereotactic radiotherapy (SAbR) irradiation in a patient with refractory ventricular arrhythmias.

[0082] Figure 9A shows a flowchart illustrating an example of a process (900) for improving a library of computer simulations with data associated with clinical cases, according to several exemplary embodiments. Referring to Figures 1, 2, 3A-C, and 9A, the process (900) may be performed by a data controller (110) to supplement one or more computer simulations contained in a library (125) along with clinical data associated with clinical cases.

[0083] In (902), the data controller (110) may receive clinical data associated with a clinical case from a first user (145a). For example, the data controller (110) may receive clinical data associated with a clinical case from a first user (145a) in a first client (140a), including, for example, patient anatomical information, diagnostic and / or treatment modality, treatment parameters, treatment outcomes, relevant medical literature, etc.

[0084] In (904), the data controller (110) may store in the library (125) at least a portion of the clinical data, including computer simulations having features that most closely match those of the clinical case, by associating them with the clinical data. For example, when the data controller (110) receives clinical data associated with a clinical case, it may calculate a similarity score for each computer simulation in the library (125) that indicates the degree of agreement between the features of the clinical case and each feature of the computer simulation in the library (125). The clinical data associated with a clinical case may be indexed based on the computer simulation having the highest similarity score. For example, the clinical data associated with a clinical case may be associated with computer simulations where features (e.g., heart size, shape, presence or absence of structural heart disease, type of arrhythmia, etc.) most closely match those in the clinical case.

[0085] In (906), the data controller (110) may respond to a query from a second user (145b) by sending the second user (145b) at least data from a library (125) that includes at least a portion of the clinical data associated with the clinical case. For example, the second user (145b) may be treating a patient for cardiac arrhythmia and therefore may query the library (125) based on patient data, including, for example, the patient's age, medical history, and proposed treatment plan. The data controller (110) may respond to a query from a second user (145b) by identifying at least one or more clinical cases in the library (120) that match the parameters of the query from the second user (145b). Alternatively, and / or additionally, instead of one or more specific clinical cases, a second user (145b) may apply a collection of clinical cases from Library (120) as training data to train a machine learning model to perform a variety of cognitive tasks, including, for example, determining the statistical probability of the origin of an arrhythmia, or performing a probabilistic analysis of potential clinical outcomes associated with various treatment approaches for arrhythmias (e.g., location, number, and pattern of ablation lesions).

[0086] In some exemplary embodiments, the machine learning model may include, for example, a neural network such as an autoencoder. The machine learning model may be trained on training data that includes clinical data from a large number of patient cases that can be collected and input into the machine learning model. The training data may include ground truth labels that include patient demographic information, electrocardiogram (ECG) and vector electrocardiogram (VCG) traces, and identified arrhythmia origin locations. The arrhythmia origin locations may be further labeled with the ablation site, size, technique, internal targeting volume (ITV), planned targeting volume (PTV), ablation energy dose, and some ranking of the outcome (e.g., arrhythmia termination, short-term ablation success, long-term ablation success, etc.). Furthermore, the machine learning model may be trained to consider features present in the treatment approach for each patient (ablation lesion, number, size, composition, internal targeting volume (ITV), planned targeting volume (PTV), ablation energy dose, etc.). Additionally, the machine learning model may determine similarity metrics between different clinical cases based on demographics, arrhythmia type, cardiac anatomy, etc., to determine relevance to both other training case data and / or future cases to be compared. When a user wishes to utilize a trained machine learning model, the user may provide the patient's electrocardiogram (ECG) or vector electrocardiogram (VCG), as well as one or more patient and arrhythmia features, as input to the trained machine learning model. Based on at least the aforementioned inputs, the trained machine learning model may determine the statistical probability of the arrhythmia origin location and the probability analysis of potential clinical outcomes associated with various treatment approaches for the arrhythmia (e.g., location, number, and pattern of ablation lesions). Accordingly, the data controller (110) may be configured to identify a selection of relevant clinical cases for case reference and procedural planning, based at least on the output of the trained machine learning model.

[0087] Figure 9B shows a flowchart illustrating an example of a process (920) for modifying a library of computer simulations according to several exemplary embodiments. According to Figures 1, 4, 5A-C, and 9B, the process (920) may be performed by a data controller (110) to modify one or more computer simulations in the library (125) to better fit patient-specific features.

[0088] In (922), the data controller (110) may receive patient data collected during an electrophysiological investigation in either an electrophysiology laboratory, a radiology suite, or an operating room. In some exemplary embodiments, the data controller (110) may receive, and may include, patient data collected during an electrophysiological (EP) investigation, such as sustained action potential (APD) recovery data, conduction velocity recovery data, patient anatomical and geometric shape, voltage mapping, intracardiac ultrasound data, patient demographics, cardiac activation patterns, local conduction velocity, electrographic analysis, etc.

[0089] (924) The data controller (110) may modify one or more computer simulations contained in the library (125) based on at least patient data. In some exemplary embodiments, the data controller (110) may modify one or more non-patient-specific computer simulations contained in the library (125) based on at least patient data, so that one or more non-patient-specific computer simulations better match the patient's specific characteristics. These modifications may be made in real time (or near real time) so that the modified library (125) of computer simulations is available when the patient is being treated for an arrhythmia.

[0090] In (926), the data controller (110) may send a modified computer simulation to a first client (140a) and / or a second client (140b) to enable a first user (145a) and / or a second user (145b) to determine the location of the origin of the arrhythmia and to perform one or more actions at the location of the origin of the cardiac arrhythmia, based at least on the modified computer simulation. For example, the first user (145a) and / or the second user (145b) may perform a higher-fidelity localization of the origin of the cardiac arrhythmia based at least on the modified computer simulation. Thus, the outcome of subsequent actions performed at the origin of the cardiac arrhythmia may be improved by the higher-fidelity localization of the origin of the cardiac arrhythmia.

[0091] Figure 9C shows a flowchart illustrating an example of a process (930) for aligning computer simulations and electroanatomical mappings according to several exemplary embodiments. According to Figures 1, 6A-D, 7A-B, 8, and 9C, the process (930) may be performed by a data controller (110) to further localize the origin of cardiac arrhythmias.

[0092] In (932), the data controller (110) may identify the locations of n pacing sites on the electroanatomical map where the catheter is placed when one or more pacing stimuli are applied. For example, as shown in Figure 6C, the data controller (110) may identify one or more pacing sites on the electroanatomical map (620).

[0093] In (934), the data controller (110) can identify the locations of n pacing sites in a computer simulation of the patient's anatomy. For example, as shown in Figure 6D, the data controller (110) can further identify the locations of the same n pacing sites in a computer simulation (600).

[0094] In (936), the data controller (110) may align the electroanatomical map and the computer simulation of the patient's anatomy based on at least the locations of n pacing sites, so that the origin of the cardiac arrhythmia is displayed by the locations of the n pacing sites. In some exemplary embodiments, the data controller (110) may align the electroanatomical map (620) and the computer simulation (600) based on the locations of n pacing sites. For example, once the locations of n pacing sites are known in the computer simulation (600) and the electroanatomical map (620), the locations of the n pacing sites can be used to align the electroanatomical map (620) with the respective reference coordinate systems of the computer simulation (600) using a deformation matrix A. For example, the alignment of the electroanatomical map (620) and the computer simulation (600) may be achieved using least-squares fitting algorithms, rotation, translation, and scaling. Thus, the origin of cardiac arrhythmias can be further indicated based on the location of n pacing sites.

[0095] In (938), the data controller (110) may generate a user interface that displays the location of the origin of cardiac arrhythmias in relation to the locations of n pacing sites. As should be noted, the location of the origin of cardiac arrhythmias in relation to n pacing sites can be actionable data for clinicians treating patients with cardiac arrhythmias. Thus, the data controller (110) may provide this information to a first user (145a) and / or a second user (145b), for example, by generating a user interface that displays the location of the origin of cardiac arrhythmias in relation to the locations of n pacing sites. For example, procedures including ablation, targeted gene therapy, stereotactic radiotherapy (e.g., gamma rays, proton beams), and surgical interventions may be performed at locations identified as the origin of cardiac arrhythmias. For example, as shown in Figure 8, a first user (145a) and / or a second user (145b) may perform a procedure at the site of origin of the cardiac arrhythmia to alleviate the arrhythmia by interrupting and / or eliminating the unstable electrical signals that cause the abnormal myocardial contractions associated with the arrhythmia.

[0096] Figure 10 shows a block diagram illustrating a computer system (1000) in several exemplary embodiments. According to Figures 1 and 10, the computing system (1000) can be used to implement a data controller (110) and / or any components therein.

[0097] As shown in Figure 10, the computing system (1000) may include a processor (1010), memory (1020), storage devices (1030), and input / output devices (1040). The processor (1010), memory (1020), storage devices (1030), and input / output devices (1040) may be interconnected via a system bus (1050). The processor (1010) can process instructions for execution within the computing system (1000). Such instructions for execution may, for example, implement one or more components of a data controller (110). In some implementations of the particulars of the present invention, the processor (1010) may be a single-threaded processor. Alternatively, the processor (1010) may be a multi-threaded processor. The processor (1010) can process instructions stored in memory (1020) and / or storage device (1030) for displaying graphical information for a user interface provided via input / output device (1040).

[0098] Memory (1020) is a computer-readable medium, such as a volatile or non-volatile medium, that stores information within the computing system (1000). Memory (1020) may store, for example, a data structure representing a configuration object database. A storage device (1030) can provide persistent storage for the computing system (1000). The storage device (1030) may be a floppy disk device, a hard disk device, an optical disk device, or a tape device, or other suitable persistent storage means. An input / output device (1040) provides input / output operations for the computing system (1000). In some embodiments of the particulars of the present invention, the input / output device (1040) includes a keyboard and / or a pointing device. In various embodiments, the input / output device (1040) includes a display device for displaying a graphical user interface.

[0099] According to some embodiments of the specific features of the present invention, the input / output device (1040) may provide input / output operations for a network device. For example, the input / output device (1040) may include an Ethernet port or other networking port for communicating with one or more wired and / or wireless networks (e.g., a local area network (LAN), a wide area network (WAN), the Internet).

[0100] In some specific embodiments of the present invention, the computing system (1000) may be used to run various interactive computer software applications that can be used to organize, analyze, and / or store data in various formats (e.g., tables). Alternatively, the computing system (1000) may be used to run any type of software application. These applications may be used to perform various functions, e.g., planning functions (e.g., generating, managing, and editing spreadsheet documents, word processor documents, and / or other objects), calculation functions, communication functions, etc. The application may include various add-in functions or may be a standalone computing product and / or function. When activated within the application, the function may be used to generate a user interface provided via an input / output device (1040). The user interface may be generated by the computing system (1000) and presented to the user (e.g., on a computer screen monitor).

[0101] One or more embodiments or features of the inventive features described herein may be realized in digital electronic circuits, integrated circuit designs, specially designed application-specific integrated circuits (ASICs), field-programmable gate array (FPGA) computer hardware, firmware, software, and / or combination thereof. These various embodiments or features may include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, the processor may be special or multipurpose and may be coupled to a storage system, at least one input device, and at least one device for receiving and transmitting data and instructions. The programmable system or computing system may include a client and a server. The client and the server are typically far apart from each other and interact typically via a communication network. The client-server relationship is generated by computer programs running on each computer and having a client-server relationship with each other.

[0102] These computer programs, which may also be called programs, software, software applications, applications, components, or code, contain machine language instructions for a programmable processor and may be implemented in high-level procedural and / or object-oriented programming languages ​​and / or assembly / machine language. As used herein, the term “machine-readable medium” means any computer program product, apparatus, and / or device, such as magnetic disks, optical disks, memory, and programmable logic circuits (PLDs), which are used to provide machine language instructions and / or data to a programmable processor and which receive machine language instructions as machine-readable signals. The term “machine-readable signals” means any signals used to provide machine language instructions and / or data to a programmable processor. Machine-readable medium may store such machine language instructions non-temporarily, such as non-temporarily stored solid-state memory, or magnetic hard drives, or any equivalent storage medium. Machine-readable media may, alternatively or additionally, store such machine language instructions in a temporary manner, such as a processor cache or other random access memory associated with one or more physical processor cores.

[0103] The inventive features described herein may be embodied in systems, apparatus, methods, and / or articles depending on the desired configuration. The practices revealed in the preceding description do not represent all practices that are consistent with the inventive features described herein. Rather, they are merely some examples that are consistent with embodiments relating to the described inventive features. While some variations are described in detail above, other improvements or additions are possible. In particular, further features and / or variations may be provided in addition to those revealed herein. For example, the practices described above may cover various combinations and subcombinations of the disclosed features, and / or combinations and subcombinations of some of the further features disclosed above. Furthermore, the logic flows shown in the accompanying drawings and / or described herein do not necessarily require a specific or sequential order to achieve the desired result. Other practices may fall within the scope of the following claims.

Claims

1. A system, said system: At least one processor, A memory containing program code, and a system that includes, when the program code is executed by at least one processor: Accessing multiple computer simulations of cardiac arrhythmias having voltage solutions, wherein each of the multiple computer simulations of cardiac arrhythmias is based on a computer model which is a non-patient specific computer model, and each of the computer models is related to cardiac characteristics. Receiving patient data collected during electrophysiological investigations, Modify the computer model, including the cardiac features, based at least on patient data. Incorporating the solution for cardiac arrhythmia voltages previously calculated in the uncorrected computer model into the corrected computer model, The computer simulation of the modified computer model is advanced in time, After advancing the aforementioned computer simulation in time, the location of the origin of the arrhythmia is determined based on the aforementioned computer simulation. A system that provides operations including displaying the location of the origin of cardiac arrhythmias in order to notify the patient of treatment based on patient data.

2. The system according to claim 1, wherein the patient data includes at least one of action potential sustained recovery data, conduction velocity recovery data, patient anatomical geometry, voltage mapping, intracardiac ultrasound data, transthoracic ultrasound data, cone-beam computed tomography data, fluoroscopy data, patient demographics, cardiac activation pattern, local conduction velocity, and electrographic features.

3. The system according to claim 1, wherein the modifications include applying patient-specific enhancements to one or more computer simulations, which include at least one of geometrically deforming and / or rotating them; imposing voltage and / or electrographic information onto one or more computer simulations; indicating activation information; adding global and / or local information regarding the thickness of the cardiac structural walls; and incorporating global and / or geographic information regarding the location and morphology of the papillary muscles, pulmonary veins, and / or left and right atrial appendages.

4. The system according to claim 1, wherein the modification is performed in real time or near real time, and thereafter, for clinical use, a computer simulation of one or more modified cardiac arrhythmias is returned to the user.

5. The system according to claim 1, characterized in that one or more computer simulations are part of a library of non-patient-specific computer simulations of cardiac arrhythmias.

6. The system according to claim 1, further comprising initiating arrhythmia simulations to generate a patient-adjusted arrhythmia vector electrocardiogram library for use in computer arrhythmia mapping processing, based on at least one arrhythmia solution associated with one or more computer simulations of modified cardiac arrhythmias.

7. The system according to claim 1, further comprising performing an arrhythmia simulation for each of a plurality of origin locations, based on at least one modified computer simulation of cardiac arrhythmias, wherein the arrhythmia simulation is performed based on assumptions of the origin location, and the plurality of origin locations and corresponding arrhythmia simulations form a patient-adjusted arrhythmia library for use in computer arrhythmia mapping processing.

8. A method performed by a computer, the method performed by the computer is: A step of accessing multiple computer simulations of cardiac arrhythmias having a voltage solution, wherein each of the multiple computer simulations of cardiac arrhythmias is based on a computer model which is a non-patient specific computer model, and each of the computer models is related to cardiac characteristics, The process of receiving patient data collected during electrophysiological investigations, A step of modifying the computer model, which includes the characteristics of the heart, based at least on patient data, A step of incorporating the solution of cardiac arrhythmia voltages previously calculated in the uncorrected computer model into the corrected computer model, A step of advancing the computer simulation of the modified computer model in time, After advancing the aforementioned computer simulation in time, the process of determining the location of the origin of the arrhythmia based on the aforementioned computer simulation is performed. A computer-based method comprising the step of providing a display of the origin of a cardiac arrhythmia in order to notify a patient of a course of action based on patient data.

9. The method according to claim 8, wherein the patient data includes at least one of action potential sustained recovery data, conduction velocity recovery data, patient anatomical geometry, voltage mapping, intracardiac ultrasound data, transthoracic ultrasound data, cone-beam computed tomography data, fluoroscopy data, patient demographics, cardiac activation pattern, local conduction velocity, and electrogeographic features.

10. The method according to claim 8, wherein the modification step includes applying patient-specific enhancements to one or more computer simulations, which include at least one of geometrically deforming and / or rotating them; imposing voltage and / or electrographic information onto one or more computer simulations; indicating activation information; adding global and / or local information regarding the thickness of the cardiac structural walls; and incorporating global and / or geographic information regarding the location and morphology of the papillary muscles, pulmonary veins, and / or left and right atrial appendages.

11. The method according to claim 8, wherein the modification step is performed in real time or near real time, and thereafter, one or more computer simulations of the modified cardiac arrhythmias are returned to the user for clinical use.

12. The method according to claim 8, characterized in that the one or more computer simulations are part of a library of non-patient-specific computer simulations of cardiac arrhythmias.

13. The method according to claim 8, further comprising the step of initiating arrhythmia simulations to generate a patient-adjusted arrhythmia vector electrocardiogram library for use in computer arrhythmia mapping processing, based on at least one arrhythmia solution associated with one or more computer simulations of modified cardiac arrhythmias.

14. The method according to any one of claims 8 to 13, further comprising the step of performing an arrhythmia simulation for each of a plurality of origin locations, based on at least one computer simulation of a modified cardiac arrhythmia, wherein the arrhythmia simulation is performed on assumptions of the origin location, and the plurality of origin locations and corresponding arrhythmia simulations form a patient-adjusted arrhythmia library for use in computer arrhythmia mapping processing.

15. A non-temporary computer-readable medium for storing instructions, wherein when the instructions are executed by at least one data processor, the result is: Accessing multiple computer simulations of cardiac arrhythmias having voltage solutions, wherein each of the multiple computer simulations of cardiac arrhythmias is based on a computer model which is a non-patient specific computer model, and each of the computer models is related to cardiac characteristics. Receiving patient data collected during electrophysiological investigations, Modify the computer model, including the cardiac features, based at least on patient data. Incorporating the solution for cardiac arrhythmia voltages previously calculated in the uncorrected computer model into the corrected computer model, The computer simulation of the modified computer model is advanced in time, After advancing the aforementioned computer simulation in time, the origin of the arrhythmia is determined based on the aforementioned computer simulation. A non-temporary computer-readable medium that provides a display of the origin location of cardiac arrhythmias to notify the patient of treatment based on patient data, and brings about the necessary actions.

16. A device, the device is: Means for accessing multiple computer simulations of cardiac arrhythmias having a voltage solution, wherein each of the multiple computer simulations of cardiac arrhythmias is based on a computer model which is a non-patient specific computer model, and each of the computer models is related to cardiac characteristics, Means for receiving patient data collected during electrophysiological procedures, Means for modifying the computer model, including the characteristics of the heart, based at least on patient data, Means for incorporating the solution of cardiac arrhythmia voltages previously calculated in the uncorrected computer model into the corrected computer model, Means for advancing the computer simulation of the modified computer model in time, After advancing the aforementioned computer simulation in time, means for determining the location of the origin of the arrhythmia based on the aforementioned computer simulation, An apparatus including means for providing a display of the origin of cardiac arrhythmia in order to notify a patient of a course of action based on patient data.

17. A computer program for causing a computer to perform the method described in any one of claims 8 to 14.

Images