Intracardiac pattern matching

Abstract

To provide intra-cardiac pattern matching.SOLUTION: Methods, apparatuses and systems for intra-cardiac pattern matching are disclosed. A unipolar pattern intra-cardiac (IC) electromyography (EGM) signals is received for an area of a heart from a plurality of activity channels corresponding to a plurality of electrodes of a catheter. A window of interest (WOI) of the IC EGM signals is received representing an entire cycle length for a single heartbeat. A pattern of interest (POI) is selected to include a portion of the WOI corresponding to an arrhythmia activation. A template POI is generated representative of the arrhythmia activation. Subsequent electrical activity is received, weights are applied and the subsequent electrical activity is compared with the template POI. A correlation score is generated and compared with a threshold correlation score.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] (CROSS-REFERENCE TO RELATED APPLICATIONS) This application claims the benefit of U.S. Provisional Patent Application No. 62 / 944,557, filed December 6, 2019, for Intra-Cardiac Pattern Matching, which is incorporated herein by reference as if set forth in its entirety.

[0002] FIELD OF THE INVENTION This application relates generally to intracardiac (IC) signals, and more particularly to detecting IC signals having similar morphologies. [Background technology]

[0003] To accurately map the region of the cardiac chamber that is producing an arrhythmia, it is important to capture only the signal (i.e., beat) that represents that specific arrhythmia. Signals due to effects such as ectopic beats, mechanical stimulation of the tissue, and variations in arrhythmia morphology due to alternative activation patterns of the same cycle length should be ignored. Introducing signals generated by such effects into the map of the cardiac chambers creates inaccuracies in the regional activation map and distorts the visualization of the arrhythmia, making it difficult for medical professionals to clearly identify the arrhythmia mechanism. Summary of the Invention [Means for solving the problem]

[0004] Disclosed herein are methods, devices, and systems for medical treatment and signal mapping of cardiac chambers. In one embodiment, a unipolar pattern of intracardiac (IC) electrogram (EGM) signals is received from a region of the heart. The IC EGM signals are received from multiple activity channels corresponding to multiple electrodes of one or more catheters. A pattern of interest (POI) of the IC EGM signals is selected to encompass at least a portion of a cycle length of the IC EGM signals corresponding to cardiac activity of interest, e.g., a cardiac condition such as arrhythmia activation. A template POI is generated. The template POI is created from the multiple template channels based on the IC EGM signals within a window of interest (WOI). Subsequent electrical activity, including the IC EGM signals, is received from the multiple activity channels corresponding to multiple electrodes of one or more catheters. Weights are applied to the multiple activity channels to generate weighted subsequent electrical activity. The weights are based on a derivative of the pattern in each template channel. A correlation score is calculated by comparing the weighted subsequent electrical activity to the template POI. A determination is then made as to whether the correlation score exceeds a threshold correlation value. Subsequent electrical activity beats having a maximum correlation value greater than the threshold correlation value may be determined to be problematic electrical activity similar to the problematic electrical activity identified at the POI of the template pattern. [Brief explanation of the drawings]

[0005] A more detailed understanding may be had from the following description, given by way of example in conjunction with the accompanying drawings, in which: [Figure 1] 1 is a diagram of an exemplary system in which one or more features of the present invention may be implemented. [Figure 2] 10 is a flowchart for determining correlation between a template and subsequent electrical activity. [Figure 3] A finite impulse response (FIR) chart, and [Figure 4]FIG. 1 is a schematic diagram illustrating operational blocks of a phase shifting operation, according to one embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0006] Cardiac electrical signals can be classified into different groups. The first group may include sinus rhythm, which is present when the heart rate is within its normal range. The second group may include different types of arrhythmias or irregular heart beats, which can occur when the electrical impulses regulating the heart rhythm do not travel normally. There are two main types of arrhythmias: tachycardia, in which the heart beats too quickly, and bradycardia, in which the heart beats too slowly. The third group of cardiac electrical signals may be mechanically induced, resulting from catheters manipulated within the heart chambers of patients undergoing cardiac procedures. Clinically relevant beats are often arrhythmias. Therefore, the objective of the present invention is to provide a tool that helps physicians focus on clinically relevant heart beats while filtering out cardiac beats that are not of clinical interest, such as sinus rhythm and mechanically induced beats. One method of identifying an electrical signal representing an arrhythmia is to compare the current group of the electrical signal with electrical signals previously identified as problematic arrhythmias. However, currently available conventional filtering modules have the disadvantage of only being able to distinguish atrial activation based on cycle length. Multiple activation sources (local or periodic) may have similar period lengths and therefore may be distinguishable only based on the unique sequence of activation defined by their wave propagation. Such differences may not be captured utilizing such existing conventional filtering modules.

[0007] The techniques disclosed herein enable discrimination between atrial activations captured on an IC reference signal based on their unique morphology and activation sequence using the correlation techniques disclosed herein. Activations of similar cycle lengths can be identified along with unwanted mechanically induced beats caused by catheter tissue irritation and other sources. In addition, changes in the IC signal can frequently be caused by reference catheter movement, resulting in unstable references and potentially inaccurate data acquisition. The techniques disclosed herein incorporate sensitivity to such changes and can prevent acquisition when reference catheter movement occurs.

[0008] In particular, the techniques provided herein enable automated differentiation of IC activations based on the similarity of the activation sequence to a previously determined template sequence. The techniques provided herein enable sensitivity to both clinical and mechanical changes (e.g., reference catheter movement) that may not be easily identified by the user.

[0009] According to an exemplary embodiment, unipolar intracardiac (IC) signals from within a heart chamber are recorded using a catheter inserted into the heart. The unipolar signal may mitigate or eliminate the effects of local activation changes, which may not reflect actual changes in heart chamber activation. Furthermore, the unipolar signal may exhibit greater temporal stability compared to a bipolar signal. A pattern of interest (POI) is selected within a window of interest (WOI). The WOI captures the entire cycle length of a single heart beat and collects all data related to that particular activation. Additional WOIs are utilized to capture the entire cycle lengths of subsequent heart beats and collect all data related to those heart beats. The POI captures a specific segment within the WOI of a heart beat of clinical interest, such as when arrhythmia activation is present. The selection of the POI may be based on cardiac activity that a medical professional may identify as arrhythmia activation. Alternatively, the selection of the POI may be performed automatically by a processor (e.g., processor 40 of FIG. 1 ) forming part of the present invention. Arrhythmic activation may be identified by observing a representation of the arrhythmia (e.g., an IC EGM signal) on a monitor or other display. A POI associated with the arrhythmia may be captured and stored and may include a signal or signals corresponding to the arrhythmia. As disclosed herein, problematic cardiac activity identified via a POI may be used to generate a template of a given problematic cardiac activity so that subsequent cardiac activity correlated to the problematic cardiac activity may be identified.

[0010] Subsequent cardiac activity (e.g., subsequent cardiac beats) is compared to a template created based on the identified problematic cardiac activity to determine whether the subsequent cardiac activity correlates to the template. As disclosed herein, correlation is determined based on a comparison of the multi-channel unipolar template cardiac signal and the multi-channel unipolar subsequent cardiac signal (e.g., via a Pearson correlation calculation). A weighted correlation score is assigned to the comparison of the multi-channel cardiac signal based on the multi-channel unipolar signal to determine whether the subsequent cardiac activity correlates to the template. One or more filters may be applied to remove baseline drift, a low-frequency artifact in the cardiac signal caused by respiratory motion and other patient movement. Based on the techniques disclosed herein, subsequent cardiac activity that correlates to the template is presented to a medical professional. Such activity may be presented on an image or rendering of the heart such that regions of the heart exhibiting such correlated activity are highlighted via a display or other visual medium.

[0011] 1 illustrates an exemplary system for performing an invasive medical procedure including an apparatus 20 according to an exemplary embodiment of the present invention. The medical procedure is performed by a medical professional 22, and by way of example, the medical procedure described herein is assumed to include the acquisition of intracardiac electromyogram (IC EGM) signals from a heart 24 of a human patient 26. Generally, IC EGM signals are biomedical signals that measure electrical currents generated in muscle (e.g., cardiac tissue) during activity such as contraction.

[0012] To acquire IC EGM signals, medical professional 22 may insert probe 28 into sheath 30 that has been previously positioned within a lumen, such as an artery or vein, of human patient 26. Sheath 30 is positioned such that distal end 32 of probe 28 may enter heart 24 of patient 26 and contact tissue of heart 24 after exiting distal end 34 of sheath 30.

[0013] Probe 28 may comprise any type of catheter 29 that can be inserted into heart 24 of patient 26 and tracked, typically using a magnetic tracking system and / or an impedance measurement system. For example, probe 28 may comprise a lasso catheter, a shaft catheter, or a PentaRay™ catheter manufactured by Biosense Webster® of Diamond Bar, Calif., or catheters generally similar to these catheters. Biosense Webster also manufactures magnetic tracking and impedance measurement systems that may be used in an embodiment of the present invention.

[0014] Probe 28 includes one or more electrodes 36 and / or one or more catheters 29, each in combination with one or more electrodes 36, used to acquire IC EGM signals used by a processor 40 included in device 20 in performing the techniques described herein. In addition to acting as a central processing unit, processor 40 may include real-time noise reduction circuitry 44 (typically configured as a field programmable gate array (FPGA) followed by an analog-to-digital (A / D) signal conversion integrated circuit 46). Processor 40 may pass signals from A / D circuitry 46 to another processor and may be programmed to execute the algorithms disclosed herein.

[0015] Processor 40 is located within an operator console 60 of device 20. Operator console 60 may be used by medical professional 22 and includes controls 62 that communicate with processor 40. During treatment, processor 40 communicates with an EGM module 66 in module bank 70 to acquire IC EGM signals and execute the algorithms disclosed herein.

[0016] EGM module 66 receives IC EGM signals from electrodes 36. In one embodiment, the IC EGM signals are forwarded within EGM module 66 through a low-noise preamplifier 68, through one or more low-pass filters 71A, and through one or more high-pass filters 71B to a main amplifier 72. EGM module 66 also includes an analog-to-digital converter (ADC) 74, which forwards digitized values ​​of the IC EGM module signals to processor 40 for execution by processor 40 of the algorithms described herein. Such components of an EGM module are standard and are provided to clean the IC EGM signals by removing noise and compensating for interference and interruptions. Typically, processor 40 controls the operation of preamplifier 68, low- and high-pass filters 71A and 71B, amplifier 72, and ADC 74.

[0017] 1 illustrates the EGM module 66 as having one channel for receiving signals from the electrodes 36. However, it should be understood that the EGM module 66 typically includes multiple channels substantially similar to that illustrated. For example, the EGM module 66 may include ten or more channels corresponding to the relative electrical activity sensed by ten different electrodes 36.

[0018] EGM module 66 enables processor 40 to acquire and analyze electrical signals received by electrodes 36, including the IC EGM module signals referred to herein. The IC EGM signals are typically presented to medical professional 22 as voltage-time graphs, updated in real time on display screen 80.

[0019] The software for processor 40 and module bank 70 may be downloaded electronically, for example, over a network, to processor 40. Alternatively or additionally, the software for processor 40 and module bank 70 may be provided on a non-transitory tangible medium, such as an optical, magnetic, or electronic storage medium.

[0020] To operate the device 20, the module bank 70 typically includes modules in addition to the EGM module 66 described above. For example, the module bank 70 may include one or more tracking modules (not shown) that enable the processor 40 to track the distal end of the probe 28. For simplicity, such other modules are not shown in FIG. 1. All modules utilized to operate the device 20 may include hardware and software elements.

[0021] In addition to display screen 80 being utilized to present IC EGM signals acquired by electrodes 36, display screen 80 may also be utilized to present the results of the techniques described herein. For example, the results of the techniques described herein may be incorporated into a map 82 of heart 24 displayed on display screen 80.

[0022] Referring now to FIG. 2, steps of a process 200 for determining whether subsequent electrical activity correlates with a previously generated template for a given problematic cardiac activity, or whether the correlation value exceeds a threshold correlation value, are shown.

[0023] In step 210 of process 200 of FIG. 2, catheter 29 may be inserted into patient's heart 24 and may be configured to detect IC EGM signals at multiple points on heart 24. The IC EGM signals may be sensed by electrodes 36 that are part of catheter 29. The IC EGM signals may be unipolar IC signals. The signals may be provided via a display, such as display 80 of FIG. 1. As shown in step 210, the unipolar pattern IC EGM signals from points on heart 24 may be provided to a user (e.g., medical professional 22 shown in FIG. 1) or processor 40 (FIG. 1). The unipolar pattern IC EGM signals may include patterns corresponding to cardiac conditions, such as arrhythmias, mapped within a given time window.

[0024] According to one exemplary embodiment, the unipolar pattern IC EGM signal provided in step 210 may be pre-processed such that one or more filters may be applied to the IC EGM signal prior to template creation. It will be appreciated that the pre-processing techniques disclosed in step 240a may be applied to the IC EGM signal in step 210 prior to template creation. Such filters may include median and FIR filters, which are described in more detail below. Such median and FIR filters are conventional and well known in the art and are utilized to remove baseline fluctuations caused by patient breathing and motion that often occurs during the medical procedures described herein.

[0025] In step 220, medical professional 22 or processor 40 may define a WOI that includes a pattern of interest (POI) corresponding to a cardiac condition such as cardiac arrhythmia, e.g., a time segment on the ECG / EGM signal where the activation pattern is observed. When electrodes 36 of catheter 29 detect IC EGM signals at multiple points within heart 24, the IC EGM signals are presented to device 20 as an IC EGM signal or data stream. In a known manner, the stream of IC EGM signals may be divided into sections or segments corresponding to single contractions or beats of the heart, referred to as WOIs.

[0026] A POI is a time segment on the ECG / EGM signal stream where an activation pattern is observed and captured on a reference catheter. Correlation is calculated only for the pattern selected by the POI. Medical professional 22 or processor 40 may provide reference annotations indicating the onset and offset of activation corresponding to observed problematic cardiac activity, such as arrhythmia. POIs may be defined based on these annotations. Furthermore, POIs may be extracted from patterns exhibited by multi-channel unipolar IC EGM signals contained within the WOI. A template pattern may be generated based on the POI, such that the template pattern may be the extracted POI, or may be a filtered or otherwise adjusted version of the POI. The POI may then be used for correlation comparison, as described further below. The WOI may include IC EGM signals provided by multiple channels, each corresponding to a signal received by a respective electrode of catheter 29. The multiple IC EGM signals may exhibit a pattern on which a template can be based. When a pattern is captured, the WOI (eg, a WOI spanning 62.5 seconds) may be stored in memory accessible by a processor (eg, processor 40).

[0027] Note that the IC EGM signal captured by the electrodes 36 may capture both atrial and ventricular activity. When generating a template in step 220 of process 200, the last activation in the identified WOI may be an overlapping atrial and ventricular pattern. However, such an overlapping pattern may not be optimized for mapping. As a reference for mapping, advanced reference annotations may be applied, and such overlapping activations may be identified so that patterns within the WOI are set around the last reference annotation, which represents only atrial activation. Advanced reference annotation-based algorithms can use one or more center of gravity energy approaches for the reference calculation, which can provide greater stability, especially for different beat morphologies. Methods and algorithms for analyzing multi-channel electrocardiogram signals generated during a medical procedure and determining reference annotation time are disclosed in U.S. Patent Application No. 9,259,165, entitled "Determination of Reference Annotation Time from Multi-Channel Electrocardiogram Signals (Rubenstein, et al.)," the contents of which are incorporated herein by reference in their entirety. Additionally, such algorithms can distinguish between atrial and chamber beats, both of which appear in IC signals (by viewing the body surface signals in parallel). Such distinctions can provide better input for the algorithms disclosed herein, as patterns can be aligned with incoming beats. Furthermore, by applying advanced reference annotation, when determining templates, the system can automatically select beats that are not fused activations of the atria and chambers, resulting in superior quality templates.

[0028] When advanced reference annotations are used, the time difference of the WOI is calculated between the last reference annotation in the saved window and its nearest heart chamber annotation. If the time difference is less than or equal to a threshold time (e.g., 100 ms), the pattern within the WOI can be defined as an overlapping activation pattern. In such a case, the POI is calculated around the previous, penultimate reference annotation within the saved WOI.

[0029] According to an exemplary embodiment of the present invention, if advanced reference annotations are not used, the POI is determined based on the last annotation in the saved WOI.

[0030] According to an exemplary embodiment of the present invention, a POI can be automatically determined based on the pattern exhibited by the signal within the WOI. The POI can be set around the last reference annotation within a saved template window or can be edited by the user. The POI can be calculated based on an activity segment defined around a reference annotation provided by the medical professional 22, which defines the WOI. The activity segment can be defined as set forth in Equation 1 below.

[0031]

number

[0032]

number

[0033] In the above definition of an activity segment, TS represents a timestamp. Note that the automated POI selection techniques disclosed herein can be applied to IC EGM signals disclosed herein and body surface patterns that align with signals provided by body surface electrodes.

[0034] The activity signal may be calculated for the signal defined by the activity segment by applying a median filter with a window of a given time (e.g., 15 ms). The activity signal may be calculated as shown in Equation 2 below.

[0035]

number

[0036] The activity threshold may be calculated based on the activity signal by applying Equation 3 below.

[0037]

number

[0038] The first and last intersections of the activity signal with the activity threshold may be set as the first and last intersections such that the POI is determined based on Equation 4, as described below.

[0039]

number

[0040] As disclosed herein, the template POI may be based on a WOI as identified by a medical professional 22 or as determined automatically.

[0041] In step 230, the multiple channels corresponding to the IC EGM signals provided by each of the multiple electrodes 36 of the catheter 29 may be weighted based on the signal derivation of the template so that significant activation within the multiple-channel IC EGM signals sensed by the electrodes 36 is emphasized and other activation is suppressed. In other words, correlations are first calculated for every channel among the multiple channels. A final correlation is then calculated for all channels. Each channel contributes differently, with the sharpest channels having the most influence on the final correlation. Flat channels have little influence on the final correlation. The subsequent electrical activity may be a unipolar IC EGM signal captured by the electrodes of the catheter 29.

[0042] 2, a correlation function can utilize a weighting mechanism to compare each annotated incoming beat of subsequent electrical activity against predefined POIs of the template. The correlation function can be defined as set forth in Equation 5 below.

[0043]

number

[0044]

number

[0045] Weights may be applied to each channel corresponding to each unipolar IC EGM signal received from each catheter electrode. The channel weights are calculated based on the maximum gradient of the template signal and indicate the dominant channel that should be more effective in the final correlation result. In particular, the maximum gradient of the template signal is the derivative of the template signal, and as a result, the channel weights are the derivative of the template gradient. By using a derivative-based function, sharp activations may be distinguishable from shallow activations, which may provide better template matching compared to amplitude-based functions, which may provide more deterministic results for a particular activation. For example, the channel weights for each channel may be calculated based on Equation 6, described below.

[0046]

number

[0047]

number

[0048] maxThreshold may be defined as a default value (eg, 2).

[0049] In step 240 of process 200 of Figure 2, each annotated beat in the subsequent electrical activity can be compared to a template pattern. Step 240 includes at least one or more of steps 240a-240c, which are further disclosed herein. Annotations are markings on the ECG / EGM signal where activation occurs. The activation can then be correlated with the selected pattern.

[0050] In step 240a, a pre-processing step may be applied such that the input unipolar IC EGM signal is passed through an FIR low pass filter ("LPF") filter (e.g., a 250 Hz FIR LPF filter) before being sampled at a sample rate (e.g., 1 Khz). This additional filtering may be applied to remove baseline fluctuations caused by patient movement and breathing.

[0051] During the pre-processing step 240a, a median filter may be applied to the input IC EGM signal with a given size (e.g., a size of + / - 20 milliseconds). To smooth the signal and remove artifacts, an additional FIR filter with a number of zero samples (e.g., 20 zero samples) may be applied to the median-filtered signal. Figure 3 shows an example of an FIR filter that indicates the desired frequency of interest and removes and discards the frequency of interest.

[0052] The FIR filter coefficients may be calculated according to Equation 7 set forth below.

[0053]

number

[0054] As part of the preprocessing step, the median-filtered signal may be subtracted from the original signal to remove baseline fluctuations while preserving the input IC EGM signal morphology. The subtraction can be expressed by the following Equation 8, described below:

[0055]

number

[0056] In step 240b, a correlation may be calculated between the template and the current activation of each channel such that an overall correlation value is determined based on the weight of each channel.

[0057] A single correlation value for all channels can be determined. The integrated correlation value can be calculated by the following Equation 9, set forth below:

[0058]

number

[0059] In particular, signal correlations for all channels are considered when determining the correlation between the template and subsequent electrical activity.

[0060] A correlation value can be determined for each beat of subsequent electrical activity, whereby for each incoming annotated beat, a correlation value is determined around each sampling point of the heart, applied over a window of + / - T (e.g., 40 ms) around the beat's reference annotation. The maximum correlation value in a given segment is selected as the beat correlation value with the defined template.

[0061] According to an exemplary embodiment of the present invention, a correlation value may be determined for only a subset of channels such that not all available IC EGM channels contribute to the correlation value. For example, if the left and right atria are dissociated, i.e., not synchronized in their contractions, or if one or more channels exhibit noise exceeding a threshold amount, such channels may be excluded from determining the correlation value. The exclusion of such channels may be automatic (e.g., channels exhibiting noise exceeding a threshold amount may be automatically excluded) or may be user-defined.

[0062] In step 240c, a moving window defined by the template POI can be used to determine a correlation value for each shift of the moving window. In other words, the template is placed over the IC ECM signal of the subsequent electrical activity, and the template is shifted from side to side over the subsequent electrical activity to maximize the correlation between the two. The final score for each shift of the moving window can be selected as the maximum correlation value determined for the subsequent electrical activity beat.

[0063] During phase shift implementation, processor 40 may iteratively vary the phase of the IC EGM signal for a given beat relative to the phase of the template POI. Processor 40 may use the value of the overall correlation from Equation 9 and a correlation threshold (as provided by the user, which may be, for example, 0.9) as inputs for phase shift implementation.

[0064] 4 illustrates the iterative process performed by the single channel correlation block 120', the shifted overall correlation block 124', and the phase shift block 142 in a repeating set of blocks. As shown in FIG. 4, in each iteration, the processor repeats the first two steps described above in the "shifted single channel correlation" block 120' and the "shifted overall correlation" block 124'. The process begins by correlating one channel at a time, then calculating the total correlation with the summed weights. During the iterations, the processor 40 (FIG. 1) determines the maximum overall correlation value, as described herein.

[0065] 4, phase shift iterations are performed every 1 ms, with a time window of ±40 ms measured from the annotation of the beat being analyzed. The index k defines the phase shift being evaluated, k={−40,...0,...+40}. After each iteration, the shifted overall correlation value is compared to the maximum correlation so far in comparison block 130, and if comparison block 130 returns a positive value, the overall correlation is updated in update block 134.

[0066] After a return (e.g., "yes" (positive) or "no" (negative)) is made, control continues to comparison block 138 to check whether there are more values ​​of index k to repeat. If so, k is incremented in increment block 142, the new value of k is applied to the ECG signal in signal block 146, and the flowchart returns to block 120'.

[0067] When the iterations are complete, control continues to a final comparison block shown at 152 in FIG. 4. In block 152, the updated overall correlation value obtained from block 134, which is the maximum correlation value for subsequent electrical activity obtained by the iterative process described above, is compared to a predetermined threshold correlation value. Typically, the correlation threshold may be set to a high threshold, e.g., 0.90. Alternatively, the correlation threshold may be set somewhat lower, e.g., 0.80.

[0068] As shown in block 152 of FIG. 4, if the maximum correlation value of the subsequent electrical activity is greater than a threshold, the subsequent electrical activity is assumed to represent an arrhythmia with the same morphological pattern as the POI template pattern. In this case, processor 40 may add this subsequent electrical activity to the cardiac map (82 of FIG. 1), as indicated at 160. Subsequent electrical signals received by the cardiac map (82 of FIG. 1) are collected and displayed on the map of heart 82 (FIG. 1) on display 80 (FIG. 1) as the electrical activity of interest. Alternatively, as noted in block 152, if the maximum correlation value of the subsequent electrical activity is less than a threshold, then the subsequent electrical activity signal is assumed to represent an arrhythmia different from the morphological pattern identified in the POI template pattern, or is assumed to represent sinus rhythm, an ectopic beat, or a mechanically induced beat, and the subsequent electrical activity signal is removed from the map, as indicated at 164.

[0069] According to one exemplary embodiment, subsequent electrical activity can be compared to a template by creating a sequence of activation comparisons by annotating the activation of each channel and measuring the time difference between all channels as a word of the compared values ​​between each beat, and RMS (root mean square) differences can indicate changes in the sequence. The root mean square is used in error calculations in a known manner.

[0070] Alternatively, for each channel, the POI template pattern is placed at an initial position on the subsequent electrical activity, i.e., subsequent heart beat, and a correlation value is calculated. Next, weighting for all channels at this initial position is performed, and final correlation values ​​for all channels are calculated. Next, the POI template pattern is phase-shifted or moved from the initial position to a second position for the same subsequent beat. For example, the phase shift may be 1 millisecond to the left from the initial position. Then, correlation values ​​for each channel for this new position of the POI template pattern over subsequent beats are calculated. Weighting for all channels is performed, and final correlations for all channels with respect to this second position are calculated. Next, the POI template pattern is phase-shifted over subsequent beats from the second position to a third position, and the process is repeated until the POI template pattern has been placed over subsequent beats at all positions within the position range, and final correlations for all channels are obtained for each position within the position range. For example, the POI template pattern may be shifted to different positions over subsequent beats, e.g., within a time window of ±40 milliseconds and in 1 millisecond increments for each different position. The position with the highest correlation is the one selected.

[0071] 2, a given beat of subsequent electrical activity may be determined to be correlated to the template if the maximum correlation value for the given beat of subsequent electrical activity is greater than a threshold correlation value (e.g., 0.8 or 0.9). According to one embodiment, the threshold correlation may be provided by a user (e.g., a medical professional) and / or may have a default value (e.g., 0.8 or 0.9) that may be applied without change of the threshold correlation.

[0072] The overall correlation coefficient calculated by Equation 9 depends on the phase of the ECG signal being tested relative to the phase of the morphological pattern.

[0073] Any of the functions and methods described herein may be implemented in a general-purpose computer, processor, or processor core. Suitable processors include, by way of example, a general-purpose processor, a special-purpose processor, a conventional processor, a digital signal processor (DSP), multiple microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) circuit, any other type of integrated circuit (IC), and / or a state machine. Such a processor may be manufactured by configuring a manufacturing process with the results of processed hardware description language (HDL) instructions and other intermediate data, such as a netlist (such instructions may be stored on a computer-readable medium). The result of such processing may be a mask work, which is then used in a semiconductor manufacturing process to produce a processor implementing features of the present disclosure.

[0074] Any of the functions and methods described herein can be implemented in a computer program, software, or firmware embodied in a non-transitory computer-readable storage medium and executed by a general-purpose computer or processor. Examples of non-transitory computer-readable storage media may include read-only memory (ROM), random access memory (RAM), registers, cache memory, semiconductor memory devices, magnetic media such as internal hard disks and removable disks, magneto-optical media, and optical media such as CD-ROM disks and digital versatile disks (DVDs).

[0075] It should be understood that many variations are possible based on the disclosure herein, and although features and elements are described above in particular combinations, each feature or element may be used alone without other features and elements, or in various combinations with other features and elements, with or without other features and elements.

[0076] [Embodiment] (1) A method for intracardiac pattern matching, comprising: receiving intracardiac (IC) electromyogram (EGM) signals in a unipolar pattern for a region of the heart from a plurality of activity channels corresponding to a plurality of electrodes of the catheter; selecting a window of interest (WOI) of the IC EGM signal, the WOI representing an entire cycle length of a single cardiac beat; selecting a pattern of interest (POI), the POI encompassing at least a portion of the WOI and corresponding to a particular portion of the WOI where arrhythmia activation is present; generating a template POI representative of the arrhythmia activation, the template POI being formed from a plurality of template channels and based on the IC EGM signal within the WOI; receiving subsequent electrical activity comprising an IC EGM signal from the plurality of activity channels and applying weights to the plurality of activity channels to generate weighted subsequent electrical activity, the weights being based on a derivative of a pattern in each template channel; generating a correlation score by comparing the weighted subsequent electrical activity with the template POI; determining that the correlation score exceeds a threshold correlation score. (2) The method of embodiment 1, further comprising providing a visual indication corresponding to the region of the heart based on determining that the correlation score exceeds the threshold correlation score. (3) The method of claim 2, wherein the visual indicator is at least one of a color, a pattern, and a marking. (4) The method of embodiment 1, wherein the POI is received based on user-provided reference annotations. (5) The method of embodiment 1, further comprising filtering the subsequent electrical activity using at least one of a median filter and a finite impulse response (FIR) filter.

[0077] (6) The method of embodiment 5, wherein filtering the subsequent electrical activity removes baseline fluctuations. (7) The method of embodiment 1, wherein the correlation score is a cumulative value of the weighted subsequent electrical activity. (8) The method of embodiment 1, further comprising defining a moving window based on the POI. (9) The method of embodiment 8, further comprising determining a plurality of correlation scores based on each shift of the moving window. (10) The method of embodiment 9, wherein the correlation score is based on a maximum correlation score.

[0078] (11) The method of embodiment 1, wherein the correlation score is determined based on a subset of the subsequent electrical activity. (12) The method of embodiment 11, wherein the subset of subsequent electrical activity is determined based on at least one of atrial dissociation and a noise threshold. (13) An apparatus for identifying cardiac arrhythmia activation by matching subsequent electrical activity to a template created based on identified cardiac arrhythmia activation, the apparatus comprising: one or more catheters having one or more respective electrodes arranged for insertion into a region of a human patient's heart to acquire unipolar patterned intracardiac (IC) electrogram (EGM) signals from the region of the heart; a processor arranged to receive the IC EGM signals from the one or more catheters, the processor comprising: receiving a pattern of interest (POI) in the IC EGM signal, the POI including a portion of the IC EGM signal corresponding to the cardiac arrhythmia activation; generating a template pattern of interest (POI) representative of the arrhythmia activation, the template POI including a plurality of template channels based on the IC EGM signal within the WOI; receiving subsequent electrical activity comprising an IC EGM signal from the plurality of activity channels and applying weights to the plurality of activity channels to generate weighted subsequent electrical activity, the weights being based on a derivative of a pattern in each template channel; generating a correlation score by comparing the weighted subsequent electrical activity with the template POI; determining that the correlation score exceeds a threshold correlation score. (14) The device of embodiment 13, further comprising at least one filter for filtering out respiratory or other motion. (15) The device described in embodiment 14, further comprising a display screen configured to display subsequent electrical activity having a correlation score exceeding a threshold amount to indicate arrhythmia activation.

[0079] (16) The device described in embodiment 15, wherein the subsequent electrical activity having a correlation score exceeding a threshold amount is incorporated into a map of the heart that appears on the display screen, highlighting areas of the heart that exhibit the arrhythmia activation. (17) The device of embodiment 13, wherein the catheter comprises a probe disposed within a sheath. (18) The device of embodiment 17, wherein the catheter comprises at least one of a lasso catheter and a shaft catheter. (19) The device described in embodiment 13, wherein the processor further comprises at least one filter arranged to be applied to the IC EGM signal corresponding to a cardiac condition prior to generation of the template POI. (20) The apparatus of embodiment 13, wherein the processor further comprises at least one of a median filter and a finite impulse response (FIR) filter for filtering the subsequent electrical activity.

[0080] (21) The apparatus of embodiment 13, further comprising a memory accessible by the processor, wherein the template POI can be stored in the memory. (22) The apparatus of embodiment 13, wherein the processor is located within an operation console. (23) The apparatus of claim 13, wherein the processor further comprises a central processing unit including a noise reduction circuit. (24) The device of embodiment 13, further comprising a magnetic tracking system for tracking the one or more catheters when inserted into the heart of a human patient.

Claims

1. 1. A method of operating an apparatus for intracardiac pattern matching, comprising: a processor; and a catheter configured to receive intracardiac (IC) electromyogram (EGM) signals in a unipolar pattern for a region of the heart from a plurality of activity channels corresponding to a plurality of electrodes of the catheter, the method comprising: the processor selecting a window of interest (WOI) of the IC EGM signal, the WOI representing an entire cycle length of a single cardiac beat; the processor selecting a pattern of interest (POI), the POI encompassing at least a portion of the WOI and corresponding to a particular portion of the WOI where arrhythmia activation is present; the processor generating a template POI representing the arrhythmia activation, the template POI being formed from the plurality of activity channels and based on the IC EGM signal within the WOI; the processor receiving subsequent electrical activity comprising an IC EGM signal from the plurality of activity channels and applying weights to the plurality of activity channels to generate weighted subsequent electrical activity, the weights being based on derivatives of patterns within each of the activity channels; generating a correlation score by comparing the weighted subsequent electrical activity with the template POI; the processor determining that the correlation score exceeds a threshold correlation score; A method wherein the weight in a desired activity channel is the maximum gradient of the pattern in the desired activity channel divided by the sum of the maximum gradients of the patterns in all activity channels.

2. and further comprising the processor providing a visual indicator corresponding to the region of the heart based on determining that the correlation score exceeds the threshold correlation score. The method of claim 1.

3. The method of claim 2 , wherein the visual indicator is at least one of a color, a pattern, and a marking.

4. The method of claim 1 , wherein the POIs are selected based on user-provided reference annotations.

5. 10. The method of claim 1, further comprising the processor filtering the subsequent electrical activity using at least one of a median filter and a finite impulse response (FIR) filter.

6. 6. The method of claim 5, wherein filtering the subsequent electrical activity removes baseline drift.

7. The method of claim 1 , wherein the correlation score is a cumulative value of the weighted subsequent electrical activity.

8. The method of claim 1 , further comprising the processor defining a moving window based on the POI.

9. The method of claim 8 , further comprising the processor determining a plurality of correlation scores based on each shift of the moving window.

10. The method of claim 9 , wherein the correlation score is based on a maximum correlation score.

11. The method of claim 1 , wherein the correlation score is determined based on a subset of the subsequent electrical activity.

12. 12. The method of claim 11, wherein the subset of the subsequent electrical activity is determined based on at least one of atrial dissociation and a noise threshold.

13. 1. An apparatus for identifying cardiac arrhythmia activation by matching subsequent electrical activity to a template created based on identified cardiac arrhythmia activation, the apparatus comprising: one or more catheters having one or more respective electrodes arranged for insertion into a region of a human patient's heart to acquire intracardiac (IC) electrogram (EGM) signals in a unipolar pattern from the region of the heart; a processor arranged to receive the IC EGM signals from the one or more catheters, the processor comprising: receiving a pattern of interest (POI) of the IC EGM signal, the POI including a portion of the IC EGM signal corresponding to the cardiac arrhythmia activation; generating a template pattern of interest (POI) representative of the arrhythmia activation, the template POI being formed from the plurality of activity channels based on the IC EGM signal within the WOI; receiving subsequent electrical activity comprising an IC EGM signal from the plurality of activity channels and applying weights to the plurality of activity channels to generate weighted subsequent electrical activity, the weights being based on a derivative of a pattern within each of the activity channels; generating a correlation score by comparing the weighted subsequent electrical activity with the template POI; determining that the correlation score exceeds a threshold correlation score; The apparatus wherein the weight in a desired activity channel is the maximum gradient of the pattern in the desired activity channel divided by the sum of the maximum gradients of the patterns in all activity channels.

14. 14. The device of claim 13, further comprising at least one filter for filtering out respiratory or other motion.

15. 15. The apparatus of claim 14, further comprising a display screen arranged to display subsequent electrical activity having a correlation score exceeding a threshold amount to indicate the arrhythmia activation.

16. 16. The device of claim 15, wherein the subsequent electrical activity having a correlation score above a threshold amount is incorporated into a map of the heart that appears on the display screen, highlighting regions of the heart exhibiting the arrhythmia activation.

17. The apparatus of claim 13 , wherein the catheter comprises a probe disposed within a sheath.

18. 18. The apparatus of claim 17, wherein the catheter comprises at least one of a lasso catheter and a shaft catheter.

19. 14. The apparatus of claim 13, wherein the processor further comprises at least one filter arranged to be applied to the IC EGM signal corresponding to a cardiac condition prior to generation of the template POI.

20. 14. The apparatus of claim 13, wherein the processor further comprises at least one of a median filter and a finite impulse response (FIR) filter for filtering the subsequent electrical activity.

21. The apparatus of claim 13 , further comprising a memory accessible by the processor, wherein the template POI can be stored in the memory.

22. The apparatus of claim 13 , wherein the processor is located in an operator console.

23. The apparatus of claim 13 , wherein the processor further comprises a central processing unit including noise reduction circuitry.

24. 14. The apparatus of claim 13, further comprising a magnetic tracking system for tracking the one or more catheters when inserted into the heart of a human patient.

Images