HEART GRAPHIC DISPLAY SYSTEM

Abstract

Systems are provided for generating data representing electromagnetic states of a heart for medical, scientific, research, and/or engineering purposes. The systems generate the data based on source configurations such as dimensions of, and scar or fibrosis or pro-arrhythmic substrate location within, a heart and a computational model of the electromagnetic output of the heart. The systems may dynamically generate the source configurations to provide representative source configurations that may be found in a population. For each source configuration of the electromagnetic source, the systems run a simulation of the functioning of the heart to generate modeled electromagnetic output (e.g., an electromagnetic mesh for each simulation step with a voltage at each point of the electromagnetic mesh) for that source configuration. The systems may generate a cardiogram for each source configuration from the modeled electromagnetic output of that source configuration for use in predicting the source location of an arrhythmia.

Description

BACKGROUND

Many heart disorders can cause symptoms, morbidity (e.g., syncope or stroke), and mortality. Common heart disorders caused by arrhythmias include inappropriate sinus tachycardia (IST), ectopic atrial rhythm, junctional rhythm, ventricular escape rhythm, atrial fibrillation (AF), ventricular fibrillation (VF), focal atrial tachycardia (focal AT), atrial microreentry, ventricular tachycardia (VT), atrial flutter (AFL), premature ventricular complexes (PVCs), premature atrial complexes (PACs), atrioventricular nodal reentrant tachycardia (AVNRT), atrioventricular reentrant tachycardia (AVRT), permanent junctional reciprocating tachycardia (PJRT), and junctional tachycardia (JT). The sources of arrhythmias may include electrical rotors (e.g., ventricular fibrillation), recurring electrical focal sources (e.g., atrial tachycardia), anatomically based reentry (e.g., ventricular tachycardia), and so on. These sources are important drivers of sustained or clinically significant episodes. Arrhythmias can be treated with ablation using different technologies, including radiofrequency energy ablation, pulsed field ablation, cryoablation, ultrasound ablation, laser ablation, external radiation sources, directed gene therapy, and so on by targeting the source of the heart disorder. Since the sources of heart disorders and the locations of the source vary from patient to patient, even for common heart disorders, targeted therapies require the source of the arrhythmia to be identified.

Unfortunately, common methods for reliably identifying the sources and their source locations of a heart disorder can be complex, cumbersome, and expensive. For example, one method uses an electrophysiology catheter having a multi-electrode basket catheter that is inserted into the heart (e.g., left ventricle) intravascularly to collect from within the heart measurements of the electrical activity of the heart, such as during an induced episode of VF. The measurements can then be analyzed to help identify a possible source location. Presently, electrophysiology catheters are expensive (and generally limited to a single use) and may lead to serious complications, including cardiac perforation and tamponade. Another method uses an exterior body surface vest with electrodes to collect measurements from the patient's body surface, which can be analyzed to help identify an arrhythmia source location. Such body surface vests are expensive, are complex and difficult to manufacture, and may interfere with the placement of defibrillator pads needed after inducing VF to collect measurements during the arrhythmia. In addition, the vest analysis requires a computed tomography (CT) scan and is unable to sense the interventricular and interatrial septa where approximately 20% of arrhythmia sources may occur.

BRIEF DESCRIPTION OF THE DRAWINGS

The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

FIG. 1 illustrates an ICS graphic in some embodiments.

FIG. 2 illustrates an SL graphic in some embodiments.

FIG. 3 is a block diagram illustrating components of the HGD system in some embodiments.

FIG. 4 is a flow diagram that illustrates the processing of a generate ICS similarity map in some embodiments.

FIG. 5 is a flow diagram that illustrates processing of a calculate similarity scores component of the HGD system in some embodiments.

FIG. 6 is a flow diagram that illustrates the processing of an identify cycles component of the HGD system in some embodiments.

FIG. 7 is a block diagram that illustrates the processing of a normalize cycle component of the HGD system in some embodiments.

FIG. 8 is a flow diagram that illustrates the processing of an extract segment component of the HGD system in some embodiments.

FIG. 9 is a flow diagram that illustrates the processing of a generate source location graphic of the HGD system in some embodiments.

FIG. 10 is a flow diagram that illustrates processing of a calculate library similarity scores component of the HGD in some embodiments.

FIG. 11 is a flow diagram that illustrates the processing of a generate source location indicators component of the HGD the system in some embodiments.

FIG. 12 is a flow diagram that illustrates the processing of a set color of pixels component of the HGD system in some embodiments.

FIG. 13 is a flow diagram that illustrates the processing of a generate AF SL graphic component in some embodiments.

FIG. 14 is a flow diagram that illustrates the processing of an extract cycles from AF ECG component in some embodiments.

DETAILED DESCRIPTION

Methods and systems are provided for displaying electromagnetic (EM) source graphic information relating to sources and source locations of patterns (abnormal or normal) of electrical activity of an EM source within a body. A body may be, for example, a human body, and the EM source may be a heart, a brain, a liver, a lung, a kidney, a stomach, or another part of the body that generates an EM field that can be measured, preferably, from outside the body and represented by EM measurements such as a cardiogram represented by an electrocardiogram (ECG) or a vectorcardiogram (VCG) and an electroencephalogram (EEG). In the following, an EM graphic display system is described primarily in the context of an EM source that is a heart. The methods and systems of the EM graphic system described below for the heart can be employed and adapted to provide graphics of other EM sources of a body. For example, the identification of cycles of a cardiogram as described below can be adapted to identify cycles of an EEG or cycles of EM measurements representing the EM field of a stomach. A graphic provides visual indications that can be textual and non-textual (e.g., pictorial, animation, and computer-generated rendering).

In some embodiments, methods and systems are provided for displaying heart graphic information relating to sources and source locations of heart disorders to assist in evaluation of heart disorders. In some embodiments, a heart graphic display (“HGD”) system provides a cycle similarity (“CS”) graphic and a source location (“SL”) graphic. The ICS graphic provides information on the similarity between arrhythmia cycles of a cardiogram (e.g., of a patient). The HGD system may also generate metrics relating to similarity of the arrhythmia cycles. An arrhythmia cycle may include an AF cycle, a VF cycle, atrial flutter, ventricular tachycardia, and so on. An AF cycle is a cycle, often identified during a long T-Q interval on the ECG during AF, comprising the interval between activation of a group of cells in the atrium until the next activation of that group of cells. A VF cycle is a cycle, often identified as a QRS complex on the ECG during VF, between activation of a group of cells in the ventricle until the next activation of that group of cells.

In some embodiment, the HGD system generates information based on analysis of an arrhythmia cardiogram collected during an arrhythmia episode (e.g., AF or VF). The information relates to similarity of arrhythmia cycles of the arrhythmia cardiogram and/or based on source locations associated with arrhythmia cycles. The HGD system receives an arrhythmia cardiogram and extracts arrhythmia cycles from the arrhythmia cardiogram. The HGD system then groups the extracted arrhythmia cycles into collections of similar arrhythmia cycles. For example, when the arrhythmia is atrial flutter (AFL) or ventricular tachycardia (VT), most of the extracted arrhythmia cycles may be similar to each other resulting in one collection that contains a large fraction of the extracted arrhythmia cycles and possibly other collections that each contains a small fraction of extracted arrhythmia cycles that are similar to each other. When the arrhythmia is AF or VF, the extracted arrhythmia cycles may be mostly dissimilar resulting in many collections each with a small fraction of the extracted arrhythmia cycles exhibiting similarity. After generating the collections, the HGD system may generate an ICS graphic and a SL graphic and various metrics relating to similarity of the extracted arrhythmia cycles. The collections may be identified using various techniques such as clustering techniques (e.g., k-means clustering), machine learning techniques (e.g., neural network), and so on. With a machine learning technique, a neural network may be trained using arrhythmia cycles that is each labeled with a collection. For example, the labels may represent 10 collections of similar arrhythmia cycles. The training data may include patient arrhythmia cycles and/or simulated arrhythmia cycles.

FIG. 1 illustrates an example ICS graphic in some embodiments. The ICS graphic 100 includes a similarity map 110 that provides visual information on the similarity between pairs of arrhythmia cycles. The x-axis and the y-axis represent the arrhythmia cycles, and the intersections of a first arrhythmia cycle and a second arrhythmia cycle represent the similarity between the first arrhythmia cycle and the second arrhythmia cycle. The color in the area at the intersection of an arrhythmia cycle of the x-axis and an arrhythmia cycle of the y-axis represents a similarity score (e.g., based on a Pearson correlation coefficient or a cosine similarity) indicating the similarity between those arrhythmia cycles. The similarity scores may range from 0 to 1 with 1 being the similarity score of identical arrhythmia cycles. Legend 114 of the ICS graphic illustrates the color range representing the similarity scores with red representing the lowest similarity score and blue representing the highest similarity score. (The similarity scores may be represented by similarity indicators based on graphic characteristics other than color such as using a grayscale, a range polyhedron from a triangle to a circle, and so on.) As an example, the intersection 111 of cycle 35 and cycle 22 is blue indicating a relatively high similarity between those cycles. The intersection 112 of cycle 35 and cycle 20 is red indicating a relatively low similarity between those arrhythmia cycles. The intersections along the diagonal represent the similarly of each arrhythmia cycle to itself and is thus dark blue indicating a similarity score of 1. The similarity map may be considered to correspond to a symmetric matrix. Popup 113 identifies, when a pointer is moved over an intersection, the cycles of the x-axis and the y-axis of the intersection and the similarity score of those cycles as a z-axis value. When a pointer selects a window or range of cycles, a popup can also display statistical summaries of the similarity scores within the window such as similarity score average and standard deviation, number of significantly different groups of similar cycles, a stability index, or arrhythmia type classification. The ICS graphic may also be generated based on library cycles of a library cardiogram of a library of cardiograms. A cardiogram library of library cardiograms may include patient cardiograms collected from patients and simulated cardiograms generated by simulating electrical activity a heart. A cardiogram library may be generated using the techniques described in U.S. patent application Ser. No. 16/206,005, filed on Nov. 30, 2018 and entitled “Calibration of Simulated Cardiograms,” which is hereby incorporated by reference. The ICS graphic may be used by an electrophysiologist to differentiate the types of arrhythmias such as AF and atrial flutter.

The stability index (or stability feature) for a source is a measure of the cycle-to-cycle consistency of a dominant arrhythmia source localized to a particular region in the heart. The stability index may be based on the similarity scores of a sequence of consecutive cycles and represented by the standard deviation of all similarity scores either above or below the identity diagonal of the ICS graphic within a user-defined time interval. A high mean and a low standard deviation indicate a stable arrhythmia (e.g., focal tachycardia), and a low mean and high standard deviation indicates an unstable arrhythmia (e.g., rotor fibrillation). The measure may be a count of the number of off-diagonal clusters with high similarity to estimate the number of unique arrhythmia activation patterns and therefore sources. A technique for determining the stability of arrhythmia sources is described in Krummen, D., et. al., Rotor Stability Separates Sustained Ventricular Fibrillation From Self-Terminating Episodes in Humans, Journal of American College of Cardiology, Vol. 63, No. 23, 2014, which is hereby incorporated by reference.

The ICS graphic is useful to identify characteristic electrical activation dynamics to help eliminate the driving mechanisms of a clinical arrhythmia. An effective target source stability index or arrhythmia type classification may be indicated by the persistence of similar arrhythmia cycles. Such similar arrhythmia cycles may indicate consistent, periodic electrical activation dynamics within the heart. Electrical activation dynamics (i.e., how action potentials propagate through the cardiac tissue) are influenced by a number of factors including electrical properties of the cardiac tissue, source location, source type (e.g., rotor or focal, fibrillation, tachycardia, flutter, etc.), and disease substrate (e.g., scar). Given these factors, the similarity between arrhythmia cycles may not necessarily indicate a rotor-based arrhythmia source, but it may more generally indicate which electrical activation patterns in the cardiogram are more or less dominant or prevalent. Dominant electrical activation patterns that appear frequently in the cardiogram indicate driving, sustaining mechanisms of an arrhythmia.

Analysis of an ICS graphic may show multiple driving mechanisms as multiple groupings of similar arrhythmia cycles. When multiple groupings occur in a time-continuous sequence (consecutive groupings of high similarity scores close to the diagonal of identity of the similarity map), the analysis may suggest a transition of electrical activation dynamics from one grouping to the next. This transition may be due to movement of the source location or a change from one arrhythmia type to another (e.g. rotor-based VF can evolve into a focally driven VT). For example, referring to FIG. 1, the intersections of x-axis cycles 22-29 and y-axis cycles 22-29 represent a grouping of similar arrhythmia cycles and the intersections of x-axis cycles 33-40 and y-axis cycles 33-40 represent another grouping of similar cycles. The intersections of x-axis cycles 22-29 and y-axis cycles 33-40 indicate similarity of these groupings also referred to as off-diagonal clusters. It is possible to have multiple distant arrhythmia sources or types in the heart that do not represent movement of a single driving mechanism. An indication of multiple driving mechanisms may be present if multiple groupings of similar arrhythmia cycles do not occur in a time-continuous sequence and the estimated source locations from the localization analysis are distant. For example, the grouping represented by x-axis cycles 22-29 and y-axis cycles 22-29 and the grouping represented by x-axis cycles 49-56 and y-axis cycles 49-56 are not in a time-continuous sequence. In other cases, groupings not in a time-continuous sequence may turn out to be the same arrhythmia pattern and therefore suggest a common driving mechanism. Such a common driving mechanism may be indicated by a grouping of high similarity on the off-diagonal portions of the similarity map. For example, the grouping represented by x-axis cycles 22-29 and y-axis cycles 49-56 indicates the similarity of the grouping represented by x-axis cycles 22-29 and y-axis cycles 22-29 and the grouping representing by x-axis cycles 49-56 and y-axis cycles 49-56, which are not in a time-continuous sequence. If the similarity is low between groupings of similar cycles not in time-continuous sequence, the analysis may suggest multiple independent driving mechanisms.

The SL graphic provides information relating to source locations identified by comparing extracted arrhythmia cycles to library cycles of a library of cardiograms. Each library cycle is associated with a source location and type. FIG. 2 illustrates an example SL graphic in some embodiments. SL graphic 200 includes a representation of a heart (e.g., a ventricle) 210 along with source location indicators 221-224 derived from one or more arrhythmia cycles. To identify the source locations or types to include in the SL graphic, the HGD system calculates similarity scores indicating similarity between library cycles and the extracted arrhythmia cycles of a patient cardiogram of the patient. The HGD system then identifies some number (i.e., one or more) of source locations associated with library cycles (referred as target library cycles) based on their similarity scores. For example, the HGD system may identify source locations based on the highest similarity score of the library cycles associated with each source location, based on the average of the four highest similarity scores of the library cycles associated with each source location, and so on. Source locations associated with the library cycles having the highest similarity scores are likely to be more informative in assisting the evaluation of a heart disorder. The HGD system then generates the SL graphic that includes a representation of a portion of a heart and a source location and type indicator for each source location and type of a target library cycle.

In some embodiments, the source location and type indicators may have characteristics that vary based on similarity scores associated with the source location and type indicators. The source location and type indicators may be centered on the source locations. The source location indicators may be circles with characteristics such as size of the circle and color of the circle. The size of the circle may be based on a similarity score associated with the target library cycles. A circle with a small size (i.e., radius) that is centered on a source location may indicate that a target library cycle associated with that source location and type has a high similarity score. (A source location and type associated with a target library cycle is considered to be associated with the similarity score of that target library cycle.) A small size may indicate greater confidence in the relevance of the source location. The HGD system may assign a color to each source location and type indicator to indicate the relative ordering of the similarity scores of the source locations. For example, assuming four source location indicators are displayed, the source location indicator associated with the highest similarity score may be assigned the color red, and the other three source location indicators may be assigned the colors yellow, green, and blue in decreasing order of their associated similarity scores. The HGD system may set the center of the circle to the color assigned to a source location indicator and vary the color assigned to remainder of the circle linearly from the color assigned to the center to the color indicating the smallest similarity score at the circumference of the circle. For example, the coloring of the circles may vary from red to blue, yellow to blue, green to blue, and solid blue. The source location indicators may be in shapes other than circles and may be based on a grayscale rather than a range of colors. A source location indicator may be selected from a list of source location indicators 230. Each item in the list of source location indicators may be labeled by the cycle/beat number, cycle start and end times, a similarity statistic, or a description of the source type.

In some embodiments, the HGD system may include in the SL graphic source locations derived from multiple extracted arrhythmia cycles. For example, the HGD system may identify source locations for multiple extracted cycles and include source location indicators for the multiple extracted arrhythmia cycles on the SL graphic simultaneously. Alternatively, the HGD system may generate a source location indicator for each extracted arrhythmia cycle and output the source location indicators in sequence to present an animation of the source locations as they move from extracted arrhythmia cycle to extracted arrhythmia cycle. When the SL graphic includes source locations for multiple extracted arrhythmia cycle simultaneously, some of the source location indicators may overlap. In such a case, the HGD system may set a pixel of an overlapping portion to an average of the colors for that pixel from the overlapping source location indicators. For example, if the colors for a pixel is yellow in one source location indicator and blue in the other source location indicator, the color for the pixel of the LS graphic may be green. A cycle selection box 230 of FIG. 2 allows a user to select one or more extracted arrhythmia cycles to be used to identify the source locations for the SL graphic.

In some embodiments, the HGD system may identify a source location for a collection of extracted arrhythmia beats. The HGD system may identify library cycles of a cardiogram library that are similar to the extracted arrhythmia beats of the collection and retrieve the source locations associated with the identified library cycles. The HGD system then displays a SL graphic based on the retrieved source locations. Such an SL graphic allows an electrophysiologist to view one or more source locations that may be the driving force of an AF or VF. The HGD system may display an SL graphic based on source locations derived from multiple collections of extracted arrhythmia cycle to provide information on arrhythmias such as atrial flutter or ventricular tachycardia that may be helpful to an electrophysiologist.

In some embodiments, the HGD system extracts arrhythmia cycles from a VCG that is derived from an ECG. To extract the AF cycles, the HGD system removes noise from an ECG. An ECG may include noise generated from medical equipment in the room in which the ECG is collected. The HGD system then identifies the T-Q intervals in each heartbeat. FIG. 5 below illustrates the identification of T-Q intervals (segments) for AF and QRST intervals (segments) for VF. The HGD system processes the AF cycles in each T-Q interval. To process a T-Q interval, the HGD system first denoises the T-Q interval to remove baseline wander so that the beginning and ends of the AF cycles are near the x-axis, for example to remove offset bias). The HGD system then converts the T-Q interval of the ECG to a VCG and removes noise from the VCG. The HGD system then identifies the AF cycles of the T-Q interval based on the VCG.

To extract the AF cycles from a VCG, the HGD display system identifies the axis with the greatest amplitude. The HGD system then identifies where each VCG tracing crosses the x-axis a certain direction (e.g., from negative to positive) and designates that crossing as a start of an AF cycle. The HGD system then identifies when the VCG tracing then next crosses the x-axis in the certain direction. The HGD system may employ a blanking window (e.g., 100 milliseconds) after the start of an AF cycle during which crossings are ignored to account for jitter in the VCG tracing that may cause multiple crossing in a very short period. The crossing delimits the start of an AF cycle (except for the last crossing) and the end of the prior AF cycle (except for the first crossing). The HGD system may use a similar approach to identify VF cycles with crossings defined based on QRS complexes.

FIG. 3 is a block diagram illustrating components of the HGD system in some embodiments. The HGD system 300 includes components to generate an ICS graphic that include a generate ICS graphic component 310, a calculate similarity scores component 311, an identify cycles component 312, a normalize cycles component 313, and an extract cycles component 314. The generate ICS graphic component generates an ICS graphic for a patient cardiogram by invoking the calculate similarity scores component to calculate the similarity scores for pairs of patient cycles of a patient cardiogram, generating the ICS graphic based on the similarity scores, and outputting the ICS graphic. The calculate similarity scores component invokes the identify patient cycles component to identify patient cycles within the patient cardiogram, invokes the extract segments component to extract relevant segments from the patient cycles and discard not useful patient cycles, and then calculates the similarity score for each pair of patient cycles. The identify cycles component identifies patient cycles and invokes the normalize cycles component to normalize the patient cycles.

The HGD system also includes components to generate an SL graphic that include a generate SL graphic component 320, a calculate library similarity scores component 321, a generate source location indicators component 322, and a set color of pixels component 323. The generate SL graphics component invokes the calculate library similarity scores to calculate the similarity scores for each pair of a patient cycle and a library cycle, invokes the generate source library indicators component to generate the source library indicators for each patient cycle, receives a selection of patient cycles, invokes the set color of pixels component to populate the SL graphics based on the SL indicators associated with the received patient cycles, and then outputs the SL graphics. The calculate library similarity scores component calculates the similarity scores for each pair of a patient cycle and a library cycle and, for each patient cycle, sorts the source locations based on the similarity scores associated with the library cycles associated with the source locations. The generate source location indicators component generates, for each patient cycle, a source location indicator for each source location or a subset of the source locations such as those associated with the highest similarity scores. The set color of pixels component calculates the pixel values for overlapping regions of source location indicators.

The patient cardiogram data store 340 contains the patient cardiogram such as an electrocardiogram (“ECG”) or a vectorcardiogram (“VCG”). The cardiogram library 350 contains library cycles of library cardiograms. The library cardiograms may be simulated cardiogram generated based on a computational model of the heart and/or a collection of patient cardiograms. The library cycles may have been identified in a manner similar to that of identifying the patient cycles. Each library cycle is associated to a corresponding source location or type.

The computing systems (e.g., network nodes or collections of network nodes) on which the HGD system and the other described systems may be implemented may include a central processing unit, input devices, output devices (e.g., display devices and speakers), storage devices (e.g., memory and disk drives), network interfaces, graphics processing units, cellular radio link interfaces, global positioning system devices, and so on. The input devices may include keyboards, pointing devices, touch screens, gesture recognition devices (e.g., for air gestures), head and eye tracking devices, microphones for voice recognition, and so on. The computing systems may include high-performance computing systems, cloud-based servers, desktop computers, laptops, tablets, e-readers, personal digital assistants, smartphones, gaming devices, servers, and so on. For example, the simulations and training may be performed using a high-performance computing system, and the classifications may be performed by a tablet. The computing systems may access computer-readable media that include computer-readable storage media and data transmission media. The computer-readable storage media are tangible storage means that do not include a transitory, propagating signal. Examples of computer-readable storage media include memory such as primary memory, cache memory, and secondary memory (e.g., DVD) and other storage. The computer-readable storage media may have recorded on them or may be encoded with computer-executable instructions or logic that implements the HGD system and the other described systems. The data transmission media are used for transmitting data via transitory, propagating signals or carrier waves (e.g., electromagnetism) via a wired or wireless connection. The computing systems may include a secure cryptoprocessor as part of a central processing unit for generating and securely storing keys and for encrypting and decrypting data using the keys.

The HGD system and the other described systems may be described in the general context of computer-executable instructions, such as program modules and components, executed by one or more computers, processors, or other devices. Generally, program modules or components include routines, programs, objects, data structures, and so on that perform tasks or implement data types of the HGD system and the other described systems. Typically, the functionality of the program modules may be combined or distributed as desired in various examples. Aspects of the HGD system and the other described systems may be implemented in hardware using, for example, an application-specific integrated circuit (“ASIC”) or field programmable gate array (“FPGA”).

The component of the HGD system may be implemented using a client-server model. For example, a client system may support receiving from a user a request for information on a heart, send to a server the request, and receive from the server indications of similarity scores. The client may then display a graphic that can be generated by the client based on the received indications of similarity scores or generated by and received from the server. The server may be hosted in a data center (e.g., cloud-based data center).

FIG. 4 is a flow diagram that illustrates the processing of a generate ICS similarity map in some embodiments. A generate ICS similarity map component generates a similarity map for an ICS graphic based on similarity between either atrial segments or ventricular segments of patient cycles of a patient cardiogram (“CG”). In block 401, the component invokes a calculate similarity scores component to calculate the similarity scores between the patient cycles. In block 402, the component adds a label or identifier of the patient cycles to the x-axis and the y-axis. In similarity map 110 of FIG. 1, the labels “Beat 1,” “Beat 3,” and so on identify every other patient cycle rather than every cycle because of space limitations. In block 403, the component selects the next pair of patient cycles (“PC”). In decision block 404, if all the patient cycles have already been selected, then the component continues at block 407, else the component continues at block 405. In block 405, the component selects a color based on the similarity score for the pair of patient cycles. In block 406, the component fills in the intersections of the pair of patient cycles. For example, for the pair of patient cycle 22 and patient cycle 35, the component sets the intersections of (35, 22) and (22, 35) to blue indicating a relatively high similarity score for those patient cycles. The component then loops to block 403 to select the next pair of patient cycles. In block 407, the component may add a legend for the color range to the ICS graphic before outputting the ICS graphic. The component then completes.

The similarity map may also be displayed as a three-dimensional graph with the similarity score representing in the z-axis. In such a case, the HGD system allows the three-dimensional graph to be rotated to view the graph at different angles around the x-, y-, and z-axes.

FIG. 5 is a flow diagram that illustrates processing of a calculate similarity scores component of the HGD system in some embodiments. The calculate similarity scored component 500 is invoked to calculate the similarity scores between patient cycles of a patient cardiogram. In block 501, the component invokes an identify cycles component to identify the patient cycles in a patient cardiogram. In blocks 502-506, the component loops identifying segments of the patient cycles and discarding those cycles that do not satisfy a usefulness criterion. A patient cycle may be not useful based on various criteria such being too short and/or too long. For example, a criterion may be that the patient cycle is longer than 500 ms and shorter than 1000 ms. In block 502, the component selects the next patient cycle. In decision block 503, if all the patient cycles have already been selected, then the component continues at block 507, else the component continues at block 504. In block 504, the component invokes an extract segment component to extract the relevant segments of the patient cycle. In decision block 505, if the patient cycle is indicated as not being useful by the extract segment component, then the component continues at block 506, else the component loops to block 502 to select the next patient cycle. In block 506, the component discards the patient cycle and loops to block 502 to select the next patient cycle. In blocks 507-509, the component loops calculating the similarity score between pairs of patient cycles. In block 507, the component selects the next pair of patient cycles. In decision block 508, if all the patient cycles have already been selected, then the component completes, else the component continues at block 509. In block 509, the component calculates the similarity score for the pair of patient cycles and then loops to block 507 to select the next pair of patient cycles.

In some embodiments, the HGD system may identify cycles (periodic intervals of arrhythmic activity) within an ECG or VCG. A cycle may be delimited by successive crossings from a negative voltage to a positive voltage (“positive crossings”) or successive crossings from a positive voltage to a negative voltage (“negative crossings”) with respect to a spatial direction or set of directions comprising a reference frame or set of reference frames. A reference frame may coincide with anatomical axes (e.g. left-to-right with x, superior-to-inferior with y, anterior-to-posterior with z), imaging axes (e.g. CT, MR, or x-ray coordinate frames), body-surface lead vectors, principal axes computed by principal component analysis of measured or simulated EM source configurations and outputs, or user-defined directions of interest. For example, a three-second VCG may have three cycles, and each cycle may be delimited by the times of the positive crossings along the x-axis. Alternatively, the cycles may be delimited by crossings along the y-axis or z-axis. In addition, cycles may be defined by negative crossings. The cycles may also be identified by processing by identifying an R peak and setting the end of the cycle to an estimate of the end of the T segment.

FIG. 6 is a flow diagram that illustrates the processing of an identify cycles component of the HGD system in some embodiments. The identify cycles component 600 is invoked to identify the cycles within a VCG and provides the normalized VCGs (nVCG[ ]) for the cycles. The component of FIG. 6 identifies cycles based on negative crossings of the x-axis. In block 601, the component initializes an index j to 2 for indexing through the VCG and sets an index k to 0 for indexing through the identified cycles. In decision block 602, if index j is greater than the size of the VCG, then the component has identified all the cycles and the component completes, providing the normalized nVCG, else the component continues at block 603. In block 603, if the prior voltage of the x-axis of the VCG (VCG.V[j−1].x) is greater than or equal to zero and the indexed voltage of the x-axis of the VCG (VCG.V[j].x) is less than zero (i.e., a negative crossing of the x-axis), then the start of a possible cycle has been identified and the component continues at block 604 to identify the cycle, else the component continues at block 609. In block 604, the component sets the start of the indexed cycle (C[k].start) equal to index j. In decision block 605, if at least one cycle has already been identified, then the end of the prior cycle is known and the component increments index k and continues at block 606, else the component increments index k and continues at block 609. In block 606, the component sets the end of the prior cycle to index j−1. In block 607, the component extracts the VCG (eVCG) for the prior indexed cycle delimited by the start and the end of the prior cycle. In block 608, the component invokes a normalize cycle component, passing an indication of the extracted VCG (eVCG), and receives the normalized cycle (nVCG). In block 609, the component increments the index j for indexing through the VCG and loops to block 602.

FIG. 7 is a block diagram that illustrates the processing of a normalize cycle component of the HGD system in some embodiments. The normalize cycle component 700 is invoked, passing an indication of the VCG of a cycle, and normalizes the cycle. If the cycles are of an ECG, then the cycles can be normalized in time and voltage to value between 0 and 1. In block 701, the component identifies the maximum vector magnitude V′ of the vectors in the cycle. For example, a vector magnitude of a vector may be calculated by taking the square root of the sum of the squares of the x, y, and z values of the vector. In block 702, the component sets index i to index a next axis of the VCG. In decision block 703, if all the axes have already been selected, then the component completes, providing the normalized VCG, else the component continues at block 704. In block 704, the component initializes an index j to 1 for indexing through the vectors of a normalized cycle. In decision block 705, if index j is greater than the number of vectors of a normalized cycle, then the component loops to block 702 to select the next axis, else the component continues at block 706. In block 706, the component sets the normalized VCG for the indexed vector for the indexed axis to an interpolation of the passed VCG, the indexed vector, and the maximum vector magnitude V′. The interpolation effectively compresses or expands the VCG to the number of vectors in the normalized VCG and divides the x, y, and z values of the vector by the maximum vector magnitude V′. In block 707, the component increments the index j and then loops to block 705.

FIG. 8 is a flow diagram that illustrates the processing of an extract segment component of the HGD system in some embodiments. The extract segment component 800 extracts the relevant portion of a cycle depending on whether an atrial or ventricular analysis is being performed. In block 801, the component identifies the R peak of the cycle. For example, the component may apply a Pan-Tompkins-type algorithm. (See, Pan, J., Tompkins, W. J. “A Real-Time QRS Detection Algorithm”. IEEE Transactions on Biomedical Engineering (3): 230 (1985).) In block 802, the component identifies the QRST segment. Once the R peak is identified, various techniques may be used to identify the QRST segment (i.e., QT interval). One technique is to use a fixed window that is a certain time before the R peak and a certain time after the R peak such as 30 ms before and 250 ms after. Another technique may also be based on manual identification of the QRST segment in which case the R peak may not be separately identified. Another technique may algorithmically identify the QRST segment factoring the heart rate. Another technique may algorithmically identify the R onset or T offset factoring rates of voltage change. In decision block 803, if the atrial segment is to be identified, then the component continues at block 804, else the component continues at block 806. In block 804, the component removes the QRST segment from the cycle resulting in the atrial segments remaining. In decision block 805, if the atrial segments satisfy an atrial useful criterion, then the component continues indicating the cycle with the QRST segment removed, else the component completes indicating that the atrial segments are not useful. An atrial useful criterion may be that the atrial segments of the cycle have length that are within a minimum and maximum time such as 500 and 1000 ms. In block 806, the component removes the non-QRST segments from the cycle. In decision block 807, if the QRST segment satisfies a ventricular useful criterion, then the component completes indicating the cycle with the non-QRST segments removed, else the component completes indicating that the QRST segment is not useful. The component may remove a segment of a cycle represented by an ECG by setting the values corresponding to the segment to zero.

FIG. 9 is a flow diagram that illustrates the processing of a generate source location graphic of the HGD system in some embodiments. A generate SL graphic component 900 inputs a patient cycle and outputs SL graphics for selected sets of patient cycles. In block 901, the component invokes a calculate library similarity scores component to generate the similarity scores for pairs of patient cycles and library cycles. In block 902, the component invokes a generate source location indicators to generate the indicators of source locations for each patient cycle. In block 903, the component receives a selection of a set of one or more patient cycles. A user may make the selection using cycle selection box 230 of SL graphic 200. The selection may also be received from a software system that interfaces with the HGD system, a data store, and so on. In decision block 904, if no more sets of patient cycles are to be selected, then the component completes, else the component continues at block 905. In block 905, the component invokes a set color of pixels component to set the color for each pixel of the SL graphic for the selected patient cycles. In block 906, the component outputs the SL graphic. In block 907, the component displays popups as a cursor moves to different locations on the SL graphic. For example, when the cursor is moved over a source location indicator, a popup may be displayed that include statistics relating to patient cycles associated with the corresponding source location such as the average similarity score of the highest similarity scores of those cycles. The component then loops to block 903 to receive a selection of patient cycles. In some embodiments, the SL graphic may be three-dimensional, and the HGD system may support three-dimensional rotation of the SL graphic so that a person can view, for example, the inside, front, and back of the heart.

FIG. 10 is a flow diagram that illustrates the processing of a calculate library similarity scores component of the HGD system in some embodiments. The calculate library similarity score component 1000 calculates the similarity scores pairs of patient cycles and library cycles and, for each patient cycle, sorts the similarity scores of the patient cycles associated with each source location. In block 1001, the component selects the next patient cycle. In decision block 1002, if all the patient cycles have already been selected, then the component completes, else the component continues at block 1003. In block 1003, the component selects the next library cycle (“LC”). In decision block 1004, if all the library cycles have already been selected for the selected patient cycle, then the component continues at block 1006, else the component continues at block 1005. In block 1005, the component calculates the similarity score for the patient cycle and the library cycle and continues at block 1003. In blocks 1006-1008, the component loops sorting the library cycles for each source location. In block 1006, the component selects the next source location. In decision block 1007, if all the source locations have already been selected for the patient cycle, then the component loops to block at block 1001 to select the next patient cycle, else the component continues at block 1008. In block 1008, the component sorts the library cycles associated with the selected source location for the selected patient cycle based on the similarity scores for those library cycles and the selected patient cycle. The component then loops to block 1006 to select the next source location.

FIG. 11 is a flow diagram that illustrates the processing of a generate source location indicators component of the HGD the system in some embodiments. A generate source location indicators component generates the source location indicators for each patient cycle. In block 1101, the component selects the next patient cycle. In decision block 1102, if all the patient cycles have already been selected, then the component completes, else the component continues at block 1103. In block 1103, the component selects the next source location. In decision block 1104, if all the source locations have already been selected, then the component continues at block 1106, else the component continues at block 1105. In block 1105, the component generates a source location score for the selected patient cycle. For example, the source location score may be the highest similarity score for a library cycle (and the selected patient cycle) associated with that source location. As another example, the source location score may be the average of the five highest similarity scores. In block 1106, the component selects the next source location in source location score order. In decision block 1107, if the number of source locations whose source location indicators that are to be included in the SL graphic has already been selected, then the component continues at block 1101, else the component continues at block 1108. In block 1108, the component sets the color for the source location based on the source location ordering. In block 1109, the component sets the radius for the source location based on the source location score. In block 1110, the component fills a data structure representing the circle for the source location based on the color and the radius. The component then loops to block 1106 to select the next source location.

FIG. 12 is a flow diagram that illustrates the processing of a set color of pixels component of the HGD system in some embodiments. A set color of pixels component 1200 is provided patient cycles and calculates the color of each pixel for the SL graphics based on the source location indicators for the patient cycles. In block 1201, the component selects the next pixel of the SL graphic. In decision block 1202, if all the pixels of already been selected, then the component completes, else the component continues at block 1203. In block 1203, the component identifies the source location indicators of the patient cycles that contain the selected pixel. In block 1204, the component calculates an average of the identified source location indicators based on the data structures representing the circles for the source locations. In block 1205, the component sets the color of the pixel based on the average and then loops to block 1201 to select the next pixel.

FIG. 13 is a flow diagram that illustrates the processing of a generate AF SL graphic component in some embodiments. The AF SL graphic component 1300 inputs an AF ECG and displays an AF SL graphic that illustrates source locations associated with the AF cycles. In block 1301, the component invokes an extract cycles from AF ECG component to extract AF cycles. In block 1302, the component identifies collections of similar extracted AF cycles. In block 1303, the component selects a collection based on, for example, the collection with the greatest number of extracted AF cycles or input from an electrophysiologist. In block 1304, the component identifies library AF cycles that are similar to the extracted cycles of the collection. In block 1305, the component retrieves the source locations associated with each library cycles. In block 1306, the component generates an AF SL graphic (e.g., as illustrated in FIG. 2) based on the retrieved source locations. In block 1307, the component displays the AF SL graphic and completes.

FIG. 14 is a flow diagram that illustrates the processing of an extract cycles from AF ECG component in some embodiments. The extract cycles from AF ECG component 1400 inputs an AF ECG and outputs AF cycles. In block 1401, the component removes noise from the AF ECG. Techniques for removing noise from ECGs are described in Bhumika Chandrakar, O. P. Yadav, V. K. Chandra, “A Survey of Noise Removal Techniques for ECG Signals,” International Journal of Advanced Research in Computer and Communication Engineering, Vol. 2, Issue 3, March 2013. In block 1402, the component identifies T-Q intervals. In blocks 1403-1408, the component loop extracting AF cycles from each T-Q interval. In block 1403, the component selects the next T-Q interval. In decision block 1404, if all the T-Q intervals have already been selected, then the component completes returning the extracted AF cycles, else the component continues at block 1405. In block 1405, the component removes noise from the T-Q interval. In block 1406, the component converts the T-Q interval to a VCG. In block 1407, the component removes noise from the VCG. A technique for removing noise from a VCG is describes in S. J. Simske and D. R. Blakley, “Using the Vectorcardiogram to Remove ECG Noise,” 2012 19th IEEE International Conference on Image Processing, 2012, pp. 2301-2304, doi: 10.1109/ICIP.2012.6467356. In block 1408, the component extracts AF cycles from the VCG and loops to block 1403 to select the next T-Q interval.

The following paragraphs describe various embodiments of aspects of the HGD system. An implementation of the HGD system may employ any combination of the embodiments. The processing described below may be performed by a computing system with a processor that executes computer-executable instructions stored on a computer-readable storage medium that implements the HGD system.

Intra-Cardiogram Similarity

In some embodiments, a method performed by one or more computing systems is provided for presenting information to assist evaluation of an electromagnetic (“EM”) source of a body. The method accesses indications of cycles within EM measurements of an EM field of the EM source, each pair of cycles having a similarity score indicating similarity between the cycles of the pair. The method generates a graphic indicating the similarity score for each pair of cycles. The method outputs the graphic to an output device. In some embodiments, the cycles have an ordering, the graphic includes a map with first axis representing each cycle as ordered and a second axis representing each cycle as ordered, and an intersection of a pair of cycles provides an indication of the similarity score for the pair of cycles. In some embodiments, the indications of similarity scores are based on varying a graphic characteristic. In some embodiments, the graphic characteristic is based on color. In some embodiments, the graphic characteristic is based on intensity level. In some embodiments, the graphic further includes a third axis representing similarity score and wherein the indication of the similarity score for a pair of cycles is based on height along the third axis. In some embodiments, the method rotates the graphic. In some embodiments, the output device is a display device, and the method receives a selection of a pair of cycles and displaying information relating the selected pair of cycles. In some embodiments, the display information includes the similarity score for the selected pair of cycles. In some embodiments, the output device is a display device and further comprising receiving a selection of multiple pairs of cycles and displaying information relating the selected multiple pair of cycles. In some embodiments, the displayed information includes average similarity score, standard deviation of the similarity scores, a stability index, or a number of groups of pairs of similar cycles. In some embodiments, the EM source is a heart and the EM measurements are represented by a cardiogram. In some embodiments, the indications of similarity scores are based on varying a graphic characteristic. In some embodiments, the EM measurements are generated based on a computational model that models EM activations of the EM source.

In some embodiments, one or more computing systems are provided for presenting information to assist evaluation of a heart of a patient. The one or more computing systems comprises one or more computer-readable storage mediums for storing computer-executable instructions and one or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums. When executed, the instructions to access indications of cycles within a cardiogram of the patient. Each pair of cycles has a similarity score indicating similarity between the cycles of the pair. The instructions generate a graphic indicating the similarity score for each pair of cycles. The instructions display the graphic via a display device. In some embodiments, the cycles have an ordering within the cardiogram, the graphic includes a map with first axis representing each cycle as ordered and a second axis representing each cycle as ordered, and an intersection of a pair of cycles provides an indication of the similarity score for the pair of cycles. In some embodiments, the indications of similarity scores are based on varying a graphic characteristic. In some embodiments, the graphic characteristic is based on color. In some embodiments, graphic characteristic is based on intensity level. In some embodiments, the graphic further includes a third axis representing similarity score and wherein the indication of the similarity score for a pair of cycles is based on height along the third axis. In some embodiments, the instructions further receive a selection of a pair of cycles and display information relating the selected pair of cycles. In some embodiments, the display information includes the similarity score for the selected pair of cycles. In some embodiments, the instructions further receive a selection of multiple pairs of cycles and displaying information relating the selected multiple pair of cycles. In some embodiments, the displayed information includes average similarity score, standard deviation of the similarity scores, a stability index, or a number of groups of pairs of similar cycles. In some embodiments, the instructions further identify the cycles of the cardiogram. In some embodiments, the instructions further calculate the similarity score for each pair of cycles.

In some embodiments, one or more computing systems is provided for presenting information relating to a heart. The one or more computing systems comprise one or more computer-readable storage mediums storing computer-executable instructions and one or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums. When executed, the instructions receive from a user a request for the information relating to the heart. The instructions send to a server a request for the information. The instructions receive from the server the information indicating a similarity score for each pair of cycles of electrical activations of the heart. The similarity score for a pair of cycles indicating similarity between the cycles of the pair. The instructions display on a display device visual indications of the similarity scores of the received information. In some embodiment, the cycles have an ordering, the graphic includes a map with a first axis representing each cycle as ordered and a second axis representing each cycle as ordered, and an intersection of a pair of cycles provides the visual indication of the similarity score for the pair of cycles. In some embodiments, the instructions further generate the graphic. In some embodiments, the request is sent to and the received information is received from a server. In some embodiments, the request is sent to and the received information is received from a cloud-based system. In some embodiments, the received information includes the graphic. In some embodiments, a visual indication is non-textual.

Source Location

In some embodiments, a method performed by one or more computing systems is provided for presenting information to assist evaluation of an electromagnetic (“EM”) source of a patient. The method accesses a library having library cycles of library EM measurements of an EM field of the EM source. Each library cycle is associated with a source location of the EM source. The method, for each of a plurality of library cycles, calculates a similarity score for the library cycle indicating similarity between the library cycle and a patient cycle of patient EM measurements of the patient. The method identifies a source location associated with a target library cycle based on the similarity score of the target library cycle. The method generates a graphic that includes a representation of a portion of the EM source and a source location indicator located based on the identified source location. The source location indicator has a characteristic that is based on the similarity score associated with the target library cycle. The method outputs the graphic to an output device. In some embodiments, the characteristic is size of the source location indicator that is based on the similarity score. In some embodiments, the source location indicator is a circle and the size indicate the radius of the circle. In some embodiments, a smaller size indicates a higher similarity score. In some embodiments, the characteristic is color. In some embodiments, the source location indicator is a circle, and a characteristic is size of the circle and a characteristic is color. In some embodiments, the color varies from the center of the circle to the circumference of the circle. In some embodiments, the patient cycle is one of a plurality of patient cycles and for each patient cycle generating a graphic and outputting the graphics in sequence to animate the locations of the source locations. In some embodiments, multiple source locations are associated with target library cycles are identified based on the similarity scores of the target library cycles and the graphic includes source location indicators located based on the identified source locations, each source location indicator with a characteristic that is based on the similarity score associated with a target library cycle. In some embodiments, the characteristic is size of the source location indicators, and the sizes are based on the similarity scores of the identified library cycles. In some embodiments, a smaller size indicates a higher similarity score. In some embodiments, characteristic is color with each source location indicator having a different color. In some embodiments, when source location indicators overlap on the graphic, the color of the overlap is based on a combination of the color of the source location indicators that overlap. In some embodiments, the source location indicators are circles and a characteristic is size of the circles and a characteristic is color that indicates an ordering of a source location score that is based on the similarity scores of the target library cycles. In some embodiments, the color of a circle is at the center and the coloring of the circle varies from the center to the circumference. In some embodiments, the method receives a selection of the patient cycles specified by a person. In some embodiments, the method identifies the patient cycles. In some embodiments, the method outputs statistics relating to the identified target cycle. In some embodiments, the identifying identifies source locations associated with target library cycles based on the similarity scores of the target library cycles and the graphic includes source location indicators for the identified source locations. In some embodiments, the EM measurements are generated based on a computational model that models EM activation of the EM source. In some embodiments, the EM source is a heart and the EM measurements are represented by a cardiogram. In some embodiments, the source location relates to an abnormal electrical activation of the heart and further comprising displaying a source type indicator indicate the source type of the abnormal electrical activation. In some embodiments, the source type is selected from a group consisting of rotor, focal, fibrillation, tachycardia, and flutter.

In some embodiments, one or more computing systems are provided for presenting information to assist evaluation of a heart of a patient. The one or more computing systems comprise one or more computer-readable storage mediums for storing computer-executable instructions and one or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums. When executed, the instructions, for each of a plurality of library cycles of a library, calculate a similarity score for the library cycle indicating similarity between the library cycle and a patient cycle of patient cardiogram of the patient. Each library cycle is associated with a source location of the heart. The instructions identify a source location associated with a target library cycle based on the similarity score of the target library cycle. The instructions generate a graphic that includes a representation of a portion of a heart and a source location indicator located on the heart based on the identified source location. The source location indicator has a characteristic that is based on the similarity score associated with the target library cycle. The instructions display the graphic via a display device. In some embodiments, the characteristic is size of the source location indicator that is based on the similarity score. In some embodiments, the source location indicator is a circle, and the size indicates the radius of the circle. In some embodiments, a smaller size indicates a higher similarity score. In some embodiments, the characteristic is color. In some embodiments, the source location indicator is a circle, characteristic is size of the circle, and a characteristic is color. In some embodiments, the color varies from the center of the circle to the circumference of the circle. In some embodiments, the patient cycle is one of a plurality of patient cycles, and the instructions generates a graphic for each patient cycle and display the graphics in sequence to animate changes in the source locations of the patient cycles. In some embodiments, the instructions identify source locations associated with target library cycles based on the similarity scores of the target library cycles and to generate the graphic to include source location indicators located based on the identified source locations, each source location indicator having a characteristic that is based on the similarity score associated with a target library cycle. In some embodiments, the characteristic is size of the source location indicators, and the sizes being based on the similarity scores of the identified library cycles. In some embodiments, a smaller size indicates a higher similarity score. In some embodiments, the characteristic is color with each source location indicator having a different color. In some embodiments, when source location indicators overlap on the graphic, the color of the overlap is based on a combination of the color of the source location indicators that overlap. In some embodiments, the source location indicators are circles, a characteristic is size of the circles, a characteristic is color that indicates an ordering of a source location score that is based on the similarity scores of the target library cycles. In some embodiments, the color of a circle is at the center and the coloring of the circle varies from the center to the circumference. In some embodiments, the instructions display statistics relating to the identified target cycle. In some embodiments, the identifying identifies source locations associated with target library cycles based on the similarity scores of the target library cycles, and the graphic includes source location indicators for the identified source locations. In some embodiments, at least some of the library cycles are generated based on a computational model that models electrical activation of the heart. In some embodiments, at least some of the library cycles of cardiograms collected from people. In some embodiments, the source location relates to an abnormal electrical activation of the heart and the instructions display a source type indicator indicate the source type of the abnormal electrical activation. In some embodiments, the source type is selected from a group consisting of rotor, focal, fibrillation, tachycardia, and flutter.

Although the subject matter has been described in language specific to structural features and/or acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims. Accordingly, the invention is not limited except as by the appended claims.

Claims

1. A method performed by one or more computing systems for presenting information to assist evaluation of an electromagnetic (“EM”) source of a body, the method comprising: accessing indications of cycles within EM measurements of an EM field of the EM source;identifying a collection of cycles that are similar based on a similarity scores indicating similarity between cycles;generating a graphic that provides non-textual indications of the similarity scores for the cycles of the collection; andoutputting the graphic to an output device.

2. The method of claim 1 wherein the EM measurements are cardiograms.

3. The method of claim 1 wherein the EM measurements are generated based on a computational model that models EM activations of the EM source.

4. The method of claim 1 wherein the non-textual indications of similarity scores are based on varying a graphic characteristic.

5. The method of claim 1 wherein the graphic includes a map with a first axis representing cycles of the collection and a second axis representing cycles of the collection and wherein an intersection corresponding to a pair of cycles on the map provides the non-textual indication of the similarity score for the pair of cycles.

6. The method of claim 5 wherein the graphic further includes a third axis representing similarity score and wherein the non-textual indication of the similarity score for a pair of cycles is based on height along the third axis.

7. The method of claim 1 wherein the output device is a display device and further comprising: receiving a selection of a pair of cycles and displaying information relating to the selected pair of cycles.

8. The method of claim 1 further comprising: for each of a plurality of first cycles of the cycles, for each of a plurality of second cycles of the cycles, calculating a similarity score for a pair of cycles that includes that first cycle and that second cycle.

9. The method of claim 8 wherein the similarity score is based on a Pearson correlation.

10. The method of claim 1 wherein the identifying of the collection includes applying a clustering technique to identify clusters of similar cycles.

11. The method of claim 10 wherein the clustering technique is a k-means clustering technique.

12. One or more computing systems for presenting information relating to a heart of a patient, the one or more computing systems comprising: one or more computer-readable storage mediums storing computer-executable instructions for controlling the one or more computing systems to: access indications of cycles within a cardiogram of the patient, each pair of cycles having a similarity score indicating similarity between the cycles of the pair; andidentify a collection of similar cycles based on similarity scores;generate a pictorial representation that indicates the similarity score for each pair of cycles of the collection wherein the pictorial representation includes a map with a first axis representing each cycle of the collection and a second axis representing each cycle of the collection, and wherein an intersection of a pair of cycles on the map provides a visual indication of the similarity score for the pair of cycles; andoutput the pictorial representation; andone or more processors for executing the computer-executable instructions stored in the one or more computer-readable storage mediums.

13. The one or more computing systems of claim 12 wherein the visual indications of similarity scores are based on varying a graphic characteristic.

14. The one or more computing systems of claim 12 wherein the pictorial representation further includes a third axis representing similarity score and wherein the visual indication of the similarity score for a pair of cycles is based on height along the third axis.

15. The one or more computing systems of claim 12 wherein the instructions further control the one or more computing systems to receive a selection of a pair of cycles and display information relating to the selected pair of cycles.

16. One or more computing systems for displaying cardiac information on a driving force of an arrhythmia to assist in evaluation a heart disorder of a patient, the one or more computing systems comprising: one or more computer-readable storage mediums that store computer-executable instructions for controlling the one or more computing systems to: access source configurations of a heart, each source configuration specifying a source location of an arrhythmia;for each of a plurality of the source configurations, simulate electrical activity of a heart having that source configuration;generate a simulated cardiogram based on the simulated electrical activity; andadd to a library cycles in association with the source location of that source configuration, the library cycles derived from that simulated cardiogram;identify patient cycles of a patient cardiogram that are similar to each other;retrieve from the library cycles that are similar to the patient cycles; anddisplay a representation of a heart and source location indicators derived from the source locations associated with the retrieved library cycles; andone or more processors for controlling the one or more computing systems to execute one or more of the computer-executable instructions.

17. The one or more computing systems of claim 16 further comprising instructions to provide an indication of a source location to an ablation device for automatic positioning of the ablation device to target the source location.

18. The one or more computing systems of claim 17 wherein the ablation device is a stereotactic body radiation therapy device.

19. The one or more computing systems of claim 16 further comprising instructions to position an ablation device to target one of the source locations and activate energy of the ablation device to perform an ablation.

20. One or more computing systems for displaying cardiac information on a driving force of an arrhythmia to assist in evaluating a heart disorder a patient, the one or more computing systems comprising: one or more computer-readable storage mediums that store computer-executable instructions for controlling the one or more computing systems to: train a machine learning (ML) algorithm using training data that includes arrhythmia cycles labeled with an indication of a collection, each collection corresponding to similar arrhythmia cycles;identify patient arrhythmia cycles of a patient cardiogram;for each of the patient arrhythmia cycles, apply the trained ML algorithm to that patient arrhythmia cycle to generate an indication of a collection for that patient arrhythmia cycle;select a collection of patient arrhythmia cycles;retrieve from a library of library arrhythmia cycles a plurality of library arrhythmia cycles that are similar to the patient arrhythmia cycles of the selected collection, each library arrhythmia cycle associated with a source location of an arrhythmia; anddisplay a representation of a heart and source location indicators derived from the source locations associated with the retrieved library arrhythmia cycles; andone or more processors for controlling the one or more computing systems to execute one or more of the computer-executable instructions.

21. The one or more computing systems of claim 20 further comprising instructions to provide an indication of a source location to an ablation device for automatic positioning of the ablation device to target that source location.

22. The one or more computing systems of claim 21 wherein the ablation device is a stereotactic body radiation therapy device.

23. The one or more computing systems of claim 20 further comprising instructions to position an ablation device to target a source location and activate energy of the ablation device to perform an ablation targeting that source location.

24. One or more computing systems for determining a type of cardiac arrhythmia, the one or more computing systems comprising: one or more computer-readable storage mediums that store computer-executable instructions for controlling the one or more computing systems to: access a patient cardiogram representing a cardiac arrhythmia;extract cycles of the patient cardiogram based on successive crossing of a zero voltage axis of the patient cardiogram in a specified direction;for each of the extracted cycles, applying a machine learning algorithm to that extracted cycle to assign that extracted cycle to a collection out of a plurality of collections of extracted cycles, each collection being a collection of similar extracted cycles;identify the type of cardiac arrhythmia based on the number of extracted cycles in each collection; anddisplay an indication of the type of arrhythmia; andone or more processors for controlling the one or more computing systems to execute one or more of the computer-executable instructions.

25. The one or more computing systems of claim 24 wherein the identification of the type of cardiac arrhythmia is based on the largest number of extracted cycles in a collection relative to the number of extracted cycles in each of the other collections.

26. The one or more computing systems of claim 24 wherein the extracted cycles are extracted from one or more T-Q intervals of the patient cardiogram and wherein the identification of the type of cardiac arrhythmia identifies an atrial fibrillation when the largest number is less than a threshold number that is derived from the total number of extracted cycles.

27. The one or more computing systems of claim 24 wherein the extracted cycles are extracted from one or more T-Q intervals of the patient cardiogram and wherein the identification of the type of cardiac arrhythmia identifies an atrial flutter when the largest number is more than a threshold number that is derived from the total number of extracted cycles.

28. The one or more computing systems of claim 24 wherein the extracted cycles are extracted based on QRS complexes of the patient cardiogram and wherein the identification of the type of cardiac arrhythmia identifies an atrial fibrillation when the largest number is less than a threshold number that is derived from the total number of extracted cycles.

29. The one or more computing systems of claim 24 wherein the extracted cycles are extracted based on QRS complexes of the patient cardiogram and wherein the identification of the type of cardiac arrhythmia identifies an atrial tachycardia when the largest number is more than a threshold number that is derived from the total number of extracted cycles.