JP2006025836A - Method for processing electrocardiograph signal having superimposed complex - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a computerized method to execute a heart analysis such as a real-time pace mapping, and the like, objectively and efficiently by the electrocardiograph signal processing to display the generated P wave without superimposing with the T wave which precedes in the premature atrial contraction (PAC). <P>SOLUTION: The method for generating the P wave signal from the PAC pulse to support a person to diagnose the heart includes (a) a step to select a QRS-T segment of a reference ECG signal, (b) a step to enable a user to mark a starting point and an ending point of the selected segment, (c) a step to prescribe a reference template as a wavelike segment between the marked starting point and the ending point, (d) a step to acquire the PAC pulse by a signal processing unit from two or more leads, and (e) a step to process the PAC pulse to generate the P wave signal. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a method for processing an electrocardiogram signal having a composite property (hereinafter referred to as a complex) that has been overwritten (hereinafter referred to as a superimpose). (Methods for Processing Electrocardiac Signals Having Superimposed Complexes) is a continuation-in-part application of US patent application Ser. No. 10 / 005,470. S. C. §119, title of invention filed on November 10, 2000 “Method for Viewing and Comparing ECG Signals Having Superimposed Complexes” US Patent Application No. 60 / 247,269 and the title of the invention filed on June 1, 2001 "An Algorithm to Measurement T-wave Subtraction Quality" No. 60 / 295,217, all of which are incorporated herein by reference in their entirety.

  The present invention relates to a method for processing electrical signals obtained from the heart, and more particularly to being able to track natural, paced and generated beat signals. , A method for processing an electrocardiogram signal having a composite of superimposed sub-components.

Certain cardiac arrhythmias are induced or start from a site of heart tissue, not a sinus node. These arrhythmias are generally classified as “focal” in nature. Focal arrhythmia treatment involves locating and ablating the site of the arrhythmia. One way to locate the focal site locally is to use a diagnostic 12-Lead (hereinafter simply referred to as lead (meaning lead, write, start, etc.)) ECG. The 12 leads can be used with pacing via a roving intracardiac catheter to pace map the heart. The theoretical basis of this method is the paced 12-lead ECG if the cycle length (ie, paced heartbeat) and the paced site match the unpaced heartbeat and the starting focal site. That looks the same as an unpaced ECG.
One problem (in current practice) with this method is the subjectivity involved in visually comparing an unpaced 12-lead ECG with a paced 12-lead ECG.

  The second problem is that the procedure of recording spontaneous ectopic beats and printing on paper is typically a time consuming process. A roving mapping catheter is positioned at a likely ectopic site, pacing is initiated, a record is taken, a printout is made, and the spontaneous and paced beats are printed against each other. Visual comparisons are made by aligning them. This process is repeated iteratively until the physician determines that a good match has been found between spontaneous ectopic and paced beats.

  A third problem occurs when there are multiple arrhythmia focal points and each focal point produces a deformation in a 12-lead ECG. A good distinction between these focal points may be advantageous during pace mapping and during other EP procedures (Reference: Throne RD, Jenkins JM, Winston SA et al., “Recurrent monogenic ventricular frequency. Use of tachycardia templates for recognition of recurrent monomorphic VT. "Comp. Cardology, 1989: 171-174).

The fourth problem involves superimposing the ECG P-wave and T-wave components. An electrocardiogram typically includes an initial impulse emanating from the atria, referred to as the P wave, followed by what is emanating from the ventricle, referred to as a QRS complex, followed by a T wave resulting from ventricular repolarization (FIG. 1). Thus, the heartbeat begins with a P wave, ends with a T wave, and the next heartbeat begins with another P wave.
P-waves can be a valuable tool that clinicians use to diagnose heart conditions. Therefore, clinicians often monitor the heart's electrocardiogram (ECG) to assist in the diagnosis of atrial and ventricular arrhythmias. This can be done in a variety of ways, for example, by observing the bioactivity recorded on the intracardiac electrode performed by the transthoracic catheter and monitoring the 12 lead (surface) ECG.

In some focal system arrhythmias, the atrial heart tissue begins to beat very quickly as the focus origin moves from the sinus node to the ectopic site. This faster heart rate may be maintained for more than two beats and is referred to as tachycardia. Faster heartbeats can be intermittent or as short as one heartbeat. In either case, the first beat of an atrial arrhythmia is usually initiated by what is referred to as an atrial extrasystole (“PAC”), which results in a T wave of the preceding beat. Overlapping continuous pulsating P waves can result (FIG. 2). This is not only a physiological compromise as the heart is inside, but because the P wave is covered by a T wave, the clinician can no longer use the P wave for diagnosis.
Thus, there is a method that allows a clinician to more efficiently pace map, and in addition, to monitor a patient's heart rate P wave even when the P wave overlaps the preceding T wave. Clearly, it is still necessary. The present invention addresses these needs.

  To reveal the ECG P-wave morphology of the PAC by subtracting the QRS-T template from the PAC, T-wave subtraction is a useful method in electrophysiological procedures, but the ECG baseline produced by breathing or body movement. Floating can cause certain changes in the result of T-wave subtraction. Thus, there is a further need in the industry to quantitatively measure the quality of T-wave subtraction results, particularly to monitor respiratory changes in T-wave subtraction.

  The present invention, in certain aspects, provides real-time pace mapping and processing by processing input electrical signals that represent cardiac activity to display the generated P-wave without overlapping the preceding T-wave during PAC. Provide the physician with a computerized way to perform other cardiac analyzes objectively and efficiently, and generate the P wave generated to determine whether the P wave originates from the same focus Enables objective comparison. As a direct result of the heart signal processing of the present invention, otherwise hidden signals and correlations are identified in the heart rate and heart rate segments by calculation in the derivation of the acquired signal and / or new signal. The The physician can be guided by visual aids such as bar graphs and overlapping cardiac signals of signal matching quality. These signal alignments can aid in patient diagnosis and the effectiveness of ongoing treatments such as ablation procedures.

Because of the relationship between timing and amplitude during the heart beat, individual waveforms can be covered or hidden. If a single pure subcomponent is identified and if this subcomponent has similar timing characteristics that allow it to synchronize with the composite waveform, then subtraction processing is performed according to aspects of the present invention. Can be performed, thereby generating other component waveforms (single / multiple). The sub-component waveforms, whether generated or naturally paced, can be quantitatively compared to each other using correlation analysis. This analysis may be performed retroactively or in real time.
More specifically, the present invention provides systems, programmed machines and methods that are capable of superior signal processing over prior art electrophysiological signal processors and accomplish this using a standard 12 lead ECG. can do.

In accordance with one aspect of the invention, a system for tracking ectopic beats comprises a signal sensing unit, a signal processor, and an output device. The signal detection unit is configured to capture the first ECG signal. The signal processor is connected to receive a first ECG signal from the signal sensing unit, and the start and end points of the first ECG signal used when the user defines the waveform segment as a reference template. Allows you to mark, allows you to acquire data from multiple leads, and uses correlation coefficient calculation to identify the best fit between reference template and acquired data Configured to do. The output device presents the best fit identified.
In accordance with another aspect of the present invention, a system for generating a p-wave signal from an atrial premature contraction (“PAC”) beat comprises a signal sensing unit, a signal processor, and an output device. The signal processor is connected to receive an electrocardiogram signal from the signal sensing unit and is configured to process the electrocardiogram signal to generate a P-wave signal from the PAC beat. The output device presents the generated P-wave signal.

  In certain embodiments of the foregoing system, the processor includes: (a) selecting a QRS-T segment of the reference ECG signal; and (b) the user marks the start and end points of the selected ECG signal. (C) defining a reference template as a waveform segment between the marked start and end points of the selected ECG signal; and (d) a plurality of leads (preferably A PAC beat at the signal processing unit (with leads not exceeding 12) and (e) processing the PAC beat to generate a P-wave signal. .

  In accordance with yet another aspect of the present invention, an electrophysiological computer system includes a processor configured to generate a P-wave signal that is hidden within a pre-atrial contraction (“PAC”) beat. The processor (a) selecting a QRS-T segment of the reference ECG signal; and (b) allowing the user to mark the start and end points of the selected segment of the reference ECG signal; (C) defining a reference template as a waveform segment between the marked start and end points of a selected segment of the reference ECG signal; and (d) a PAC beat in a signal processing unit from a plurality of ECG leads. And (e) processing the PAC beat to generate a p-wave signal.

  In certain embodiments of the aforementioned system, the processor performs subtraction of the reference template from a predetermined segment of the PAC beat using a correlation coefficient calculation. In a more specific embodiment, the processor compares P waves generated from a plurality of beats with each other and infers whether they indicate a common focus origin in several generated P waves. Predict the most likely site of focus origin using a focus-initiated p-wave (preferably 12 lead) library and generate paced P-waves for comparison with spontaneous p-waves Determining the integral of the QRS area of the generated P-wave signal, normalizing any integral over the length of the generated P-wave signal, and separately processing the QRS segment of the beat to A further decision is reached and configured to perform the aforementioned combination.

  In accordance with yet another aspect of the invention, an electrophysiological computer system includes a processor configured to perform steps in substantially the same manner as a processor that generates P waves from PAC beats, but more In general, the asynchronous subcomponent is configured to be generated from a first heartbeat signal having a composite waveform that includes a synchronous subcomponent that overlaps the asynchronous subcomponent. The processor selects a second heartbeat signal synchronization subcomponent corresponding to the first heartbeat signal synchronization subcomponent, and the user marks the start and end points of the selected synchronization subcomponent. The step of defining the reference template as a waveform segment between the marked start and end points of the selected synchronization subcomponent, and a first in the signal processing unit from the plurality of leads. Performing a step of obtaining a composite waveform of the heartbeat signal and processing the composite waveform beat to generate asynchronous subcomponents.

  In accordance with a further aspect of the present invention, a method for tracking ectopic beats by template matching is described, comprising: (a) capturing a first ECG signal in a signal processing unit; (b) Allowing the user to mark the start and end points of the captured first ECG signal; and (c) the waveform between the marked start and end points of the first ECG signal Defining a reference template as a segment; (d) obtaining data in a signal processing unit; and (e) using a correlation coefficient calculation on the obtained data to obtain a reference template and the obtained data. Identifying the best fit between.

  In accordance with a further aspect of the invention, a method for generating a P-wave signal from an atrial premature contraction ("PAC") beat is described, which can assist a person diagnosing the heart. The method includes (a) selecting a QRS-T segment of a reference ECG signal, and (b) allowing a user to mark the start and end points of the selected segment of the reference ECG signal. And (c) defining a reference template as a waveform segment between the marked start and end points of a selected segment of the reference ECG signal, and (d) a PAC beat in a signal processing unit from a plurality of leads. Obtaining motion, and (e) processing the PAC beat to generate a p-wave signal.

  In certain embodiments of the foregoing method, the PAC beat is processed to perform a subtraction of the reference template from a predetermined segment of the PAC beat using a correlation coefficient calculation. Further, the above-described method includes a step of comparing P waves generated from a plurality of beats with each other, a step of inferring whether or not a common focus start is generated in several types of generated p waves, and a known focus start. Predict the most likely site of focus origin using a library of initial P waves (preferably 12 leads) and generate paced P waves for comparison with spontaneous P waves Separate the QRS segment of the generated P wave signal, determine the integral value of the QRS area of the generated P wave signal, normalize any integral value over the length of the generated p wave signal, Processing to arrive at further decisions regarding heart rate data and performing a combination of the foregoing steps.

  Further methods according to further aspects of the invention include determining integrals for sections of QRS-T segments and processing those integrals. QRS segment integration can be used as a measure of QRS residual, which is an indicator of alignment or synchronization quality between template QRS and PAC QRS. In addition, baseline float can be monitored as a change in the QRS absolute peak (integral) value percentage between the template and the PAC. These methods are implemented by a suitably configured computer processor.

  A further method according to another aspect of the invention proceeds in substantially the same manner as generating a p-wave from a PAC beat, but more generally selects a synchronization subcomponent of the heartbeat signal And enabling the user to mark the start and end points of the selected synchronization subcomponent and between the marked start and end points of the selected synchronization subcomponent Defining a reference template as a waveform segment, acquiring a composite waveform from a plurality of leads with a signal processing unit, and processing the composite waveform beat to generate asynchronous subcomponents.

  In accordance with the present invention, a comparative display method is provided for displaying sequentially paced signal / template matches, which reduces the time required for the pace mapping procedure. In one embodiment, the comparative display method includes simultaneously displaying the template and the most recently paced signal / template match and the second most recently paced signal / template match. . In another embodiment, the comparative display method includes simultaneously displaying the template, the most recently paced signal / template match, and the best match of the previous paced signal / template. .

  In accordance with yet another aspect of the invention, a template optimization method is disclosed that dynamically uses different templates. QRS beats that precede or follow the PAC can be selected by hand or by operation of a programmed machine that selects and sets a new template to be used for the next calculation. The method is implemented by a suitably configured computer processor.

  Other aspects, features and advantages of the present invention can be more clearly understood from the following detailed description of exemplary embodiments and the accompanying drawings.

  In order to facilitate an understanding of the methods that can be performed according to preferred embodiments of the present invention, several suitable aspects are discussed below under their respective headings.

<Template matching / pace mapping>
Any recorded ECG waveform can be used as a reference and compared to another recorded ECG waveform or a real-time ECG waveform. The comparison is performed in a two-step process, where a reference template is first selected by the user to describe the start and end of an ECG waveform segment to be used as a comparison template. The user then selects an area of data to be used for comparison, either from preselected data or from a real-time data stream. A properly configured computer processor can find the best match against a reference template over a specific area, or, in the case of real-time analysis, the best match is updated over a defined period of time, for example every second. Is done. The criteria for “best match” uses correlation coefficient calculations across all 12 leads of the ECG to find the best alignment. This calculation may be preceded by a correlation assessment taken over a smaller number of leads, eg, one lead, and generally aligns the reference template to a selected region of the data of interest. A visual display showing aligned reference beats (templates) superimposed on the beats being analyzed provides feedback to the user regarding the closeness of the match. The correlation coefficient calculated for each ECG lead provides a quantitative indicator of alignment. A composite average is also calculated and displayed in its own color-enhanced bar graph indicator, which is particularly useful when real-time template matching is being performed. The composite average can be updated as a moving average for a preselected number of beats.

  Template matching may be used to compare two spontaneous beats, or it can be used to pace map, i.e., paced beats are referred to as spontaneous beats. Can be compared. A user can manipulate a region of interest (ROI) indicator to exclude certain portions of the waveform from analysis. This is useful during pace mapping where pacing artifacts in the surface lead can be excluded from the area of analysis. ROI indicators can also be used to specify preferences for T-waves or P-waves because T-waves and P-waves are often very similar in shape.

<T-wave subtraction>
In one embodiment of the present invention, an ECG with overlapping P and T waves is processed to remove T waves, thereby displaying non-overlapping P waves, so that the clinician can diagnose the heart. A method is provided that, when done, can observe P waves.
Referring to FIG. 1, this illustrates a normal ECG for three beats that can distinguish the discriminating P and T waves. FIG. 2 shows the rhythm that the P wave from the third beat (P ′) arrives early and is covered by the T wave from the second beat. This is consequently referred to as a P complex on T and is referred to as QRS-TP ′ in the drawing.

  Generally, according to this method, a beat-less QRS-T segment without a PAC is selected as a template. This template is subtracted from the QRS-T-P 'of the PAC to be studied to produce a P-wave. The QRS-T signal used as a template may be from a single beat or may be generated from the average of multiple beats. The QRS-T signal (or average) used as a template is such that the preceding QRS-QRS interval is equal (or nearly equal) to the QRS-QRS interval immediately preceding the QRS-TP-P 'signal to be studied. Selected. Preferably, the beat immediately preceding the PAC is selected QRS-T template when the cycle length and blood flow state of this beat is closest to that of the next beat containing the PAC and T on P complex. Can be used (see FIGS. 2 and 3).

The QRS complex is used as a means to synchronize and align QRS-T templates and PAC beats for subtraction. Alignment is automated by an algorithm for best matching based on a composite correlation coefficient over a 12 lead ECG. The physician has the option of moving the template alignment to the right or left on a sample-by-sample basis, and the resulting composite correlation coefficient is updated at each new location. The physician also has the option of choosing the previous or next QRS-T segment as a reference template. The software automatically locates the previous or next beat based on the current reference template and uses the corresponding QRS-T segment of that beat as the new reference template in the generated P-wave calculation.
Different display screens showing the generated P wave alone or the P wave superimposed on the original PAC beat or reference template can be used as an aid to the physician.
P waves generated using the T-wave subtraction method can be further signal processed to remove unwanted artifacts caused by respiration or noise.

<3. Template matching of generated P wave>
Once having a generated P-wave identified from tachycardia or early atrial beat (PAC), this generated P-wave can be compared to a previously captured reference template.

3a. More specifically, one or more spontaneous P-waves may be identified using the subtraction method described above and compared to each other using correlation waveform analysis. This can be used to determine whether spontaneous P-waves have the same focal origin. This can be done in real time or by review from recorded data.
3b. In addition, one or more generated spontaneous P-waves may be identified and compared to a known focal origin library to predict the most likely site of origin.
3c. In addition, once the spontaneous P-wave generated is identified using the subtraction method described above, the physician initiates atrial pace mapping, which is also followed by the T-wave template matching / pace mapping method described above. can do. The roving pace mapping catheter is operated in the atrium (or adjacent blood vessels such as pulmonary veins) until the generated paced P-wave is approximately the same as the generated spontaneous P-wave. The comparison of the generated P waves in this way may be performed on pre-recorded data or in real time.

  More generally, two or more waveforms X, Y,... May form a composite waveform that covers or hides individual waveforms due to timing and amplitude relationships. The composite waveform includes a synchronous subcomponent that overlaps the asynchronous subcomponent. If a single pure subcomponent waveform (e.g., X or Y) can be identified and has similar timing features that allow it to be synchronized with the composite waveform (i.e., this identified Sub-components are synchronous sub-components) and can be subtracted from the composite waveform to generate other sub-component waveforms (single / duplex) (ie asynchronous sub-components (single / duplex)). The sub-component waveforms, whether generated, naturally occurring or pace induced, can be quantitatively compared to each other using correlation analysis. This analysis may be performed retroactively or in real time. Those skilled in the art will appreciate that a number of algorithms can be used to compare waveforms, including but not limited to bin area methods and integration. Any of these methods can assist the goal of aligning the composite components of the composite waveform and / or comparing the generated results.

  The method according to this more general teaching proceeds generally as outlined above. Specifically, the method proceeds in substantially the same way as generating a P-wave from a PAC beat, but more generally, selecting a synchronization subcomponent of the heartbeat signal; Allows the user to mark the start and end points of the selected sync subcomponent and as a waveform segment between the marked start and end points of the selected sync subcomponent Defining a reference template, obtaining a composite waveform from a plurality of leads with a signal processing unit, and processing the composite waveform beat to generate asynchronous subcomponents.

  Referring now to the drawings, and in particular to FIG. 4, a system 10 for receiving and processing electrical signals in accordance with one exemplary embodiment of the present invention is shown. In one exemplary embodiment, the system 10 includes a signal sensing unit 12, which may take different forms such as a standard 12 lead ECG, an intracardiac lead, or a combination thereof. The signal detection unit is electrically connected to the signal processing device 14, which receives the detected signal from the unit 12, as will be described in more detail below. A signal processing device (“signal processor” or “processor”) 14 is preferably connected to a suitable display 16, which presents the processed system to a clinician or other party. Information is stored and can be recalled from storage device 18. Preferably, the signal processor 14 and the display 16 are a C.I. of Murray Hill, NJ. R. It is equipped with a trademark EP lab system (CR LabSystem) of Bird Corporation (CR Bard, Inc.). The trademark EP lab system supports various data collection and processing functions that are standard in electrophysiological procedures, for example through software (eg, modules, procedures, functions or purposes) or firmware as described above for subtraction and lead methods. May have its hardware configured to implement (ie, processor 14). The processor 14 communicates with memory or storage 18, which configures the processor to implement the subtraction and read methods described above (and the integration techniques described below).

  In one exemplary embodiment, certain features of the system of the present invention are implemented, in part, by a processor that uses program information stored in the memory of signal processor 14. The processor 14 may access one or more files as necessary to perform the requested function as described in more detail in connection with FIGS.

  Referring now to FIG. 5, the operation of the signal processor 14 of the present invention will be described in connection with the above structural description of the system 10. As illustrated in FIG. 5, the process begins when the clinician wishes to form a reference template, which occurs by capturing a reference ECG signal, as shown in step 502. Preferably, the reference ECG signal is captured using a standard 12 lead device and / or one or more intracardiac leads. As described above in connection with FIG. 2, a beating QRS-T signal component that does not exhibit a P wave on a T wave is selected as a template and captured in step 502 is an ECG signal configuration. This set of elements. Such beats can be captured with sinus rhythms or during local arrhythmias such as tachycardia. Furthermore, the reference template arises from signals captured at the surface, from intracardiac leads that can be placed at various locations in the heart, or from a combination of signals from the surface and intracardiac leads Is contemplated. The QRS-T signal used as a template can be captured from a single heart beat or can be a signal generated from the average of multiple heart beats.

At step 504, the start and end points of the reference template are marked by the clinician using the interface to the signal processing unit 14. The marked points define a segment of the ECG waveform that is used as a comparison template.
At step 506, the clinician selects either recorded data or real-time data to use for template matching analysis. (This step can be performed at any time prior to the waveform matching analysis of step 508, for example, prior to performing steps 502 and 504.) If recorded data is used for template matching analysis In particular, a specific area of pre-recorded data is provided to the signal processing unit for comparison with a reference template. On the other hand, if real-time data is used for template matching analysis, a stream of data from the ECG lead is provided to the signal processing unit 14 for a defined time period for comparison with a reference template.

In step 508, the signal processor 14 finds the “best match”, in other words, the best alignment between the selected region or time period and the reference template.
At step 510, the display 16 is updated to show the results of template matching to the clinician (or other person). The results can be shown quantitatively as superimposed ECG waveform signals, ie as reference beats (templates) superimposed on the beat under analysis, indicating the degree of alignment between them, or It can be quantitatively shown as a correlation coefficient calculated for the ECG lead. Preferably, a composite average is also calculated and displayed. This is illustrated in the computer display shown in FIG.

FIG. 8 (a) shows a comparison in which the second most recently paced signal / template match (Control 1), the template, and the most recently paced signal / template match (Control 2) are displayed simultaneously. A display is illustrated.
At step 512, a test is performed to determine whether the user has selected real-time processing at step 506. If so, the flow loop returns to step 508 to perform a template matching analysis again and update the display accordingly. Otherwise, if the previously recorded segment has been analyzed, the user is given the option to save the analysis (as tested in step 514), as shown in step 516. Correlation analysis is saved. Real-time analysis can be saved if desired.

  Referring to FIG. 6, the operation of the signal processor 14 of the present invention will be described in conjunction with the above structural description of the system 10. As illustrated in FIG. 6, the process begins at step 602, where the clinician wishes to capture a PAC and subtract a QRS-T reference template from the PAC. The QRS-T reference template is marked by the clinician at step 604 (as described above) and the region containing the PAC is selected by the clinician at step 606 for analysis. The QRS portion of the reference template is aligned at step 608 to best match the QRS complex immediately preceding the PAC. When the best match is found, the processor 14 subtracts the QRS-T reference template from the QRS-T-P 'segment of the PAC at step 610.

  The difference is the generated P wave that is output to the display 16 at step 612. This is illustrated in the computer display shown in FIG. 7, where the leftmost window has two vertical lines (a dotted line in front of the 14 second mark at the top (highlighted with an arrow)) and 14 2nd The selected QRS-T reference template between the second solid line immediately after the mark) is displayed. The rightmost window shows the original PAC waveform, and the generated P wave overlaps at the top of the ECG portion that occurs in the first 15 seconds. The overlapped P wave appears as a second graph overlaid on the ECG signal. Visual aids can be provided to automatically align and overlap waveforms for visual comparison on a computer display or printout.

  FIG. 8 illustrates an exemplary display for template matching (no subtraction) that can be displayed to the operator. The leftmost window displays markers that indicate the presence and use of the reference template, which starts with the leftmost vertical line (highlighted with an arrow) and ends with the second vertical line. In this example, the reference template marks the start and end of the P wave. However, any waveform segment can be used if the region of interest is marked for use as a template. The larger display window on the right shows the correlation value for each channel of the 12 lead ECG when compared to the reference template. The right bar graph does not move in this example because the analysis area is taken from recorded data rather than real-time data collected during the medical procedure.

The data can be stored and / or printed as desired in response to a user input to do so, as tested at step 614 and performed at step 616.
From the foregoing, it will be apparent to those skilled in the art that the present invention provides a method for reliably and efficiently recovering P waves from waveforms having overlapping P and T waves. Furthermore, the template matching capability of the present invention provides the added benefit of comparing ECG waveform components immediately and objectively in their natural or generated state. It should also be understood that the subtraction and lead methods described herein apply to data that can be obtained from conventional 12-lead surface ECG signals and intracardiac signals or a combination of both surface and intracardiac signals. I must.

  The two waveforms can have a high correlation with each other, but still match absolutely bad due to amplitude fluctuations and drift caused by the effects of respiration. This can be a problem when the two waveforms are aligned and then one is subtracted from the other. This is why it is usually desirable to have adjacent beats as reference (QRS-T) and PAC (QRS-T-P '). This is not always possible and is not practical when performing real-time pace mapping.

A method for monitoring the quality of T-wave subtraction will now be described with reference to FIG. In step 902, a subtraction process (as illustrated in FIGS. 3 and 6 and described above) is performed to subtract the QRS-T template from the PAC (QRS-TP ′), thereby generating a waveform. The method of FIG. 9 then proceeds by providing the physician with an integral calculation that allows for multiple measurements of interest, measuring the quality of the QRS residual and T-wave subtraction process, and in some cases, baseline floating Including, but not limited to, optimization of the selection of templates used for measurement and subtraction processing.
At step 904, the area of the generated waveform is measured. Step 906 normalizes this value by dividing the integral value by the length of the generated waveform. In addition, at step 908, the normalized integral amplitude is measured and displayed as a voltage at the input of the ECG channel. This voltage value is referred to as the QRS residual.

  As explained above, correlation analysis is used to align the QRS segment of the reference ECG template with the QRS segment of the PAC beat. Thus, further improvements may use a correlation coefficient with the so-called QRS residual of the generated waveform to give an indication of quality when a match between two beats is selected for subtraction. At the same time, it provides an indication of alignment or synchronization quality between the template QRS and the PAC QRS. For perfect alignment and good subtraction results, the generated QRS segments must be flat, exhibit high correlation to the template, and the QRS residual must be very small, exhibiting small differences in absolute amplitude (floating including).

<Comparison display of paced signal / template matching>
Use pace mapping to determine the onset of arrhythmia. Pace mapping is a time consuming procedure because an electrocardiogram signal is derived sequentially by taking a pace with the heart with intracardiac electrodes and then comparing the derived signal with a spontaneous arrhythmia signal. The spontaneous arrhythmia signal acts as a template against which the paced signal is matched. A comparison is made between each paced signal and the arrhythmia signal. A close match between the paced signal and the arrhythmia signal indicates that the onset of the arrhythmia is identified.

  In pace mapping, the user analyzes each iteration and the most recently paced signal is more arrhythmic than the second most recently paced signal or the previous best-matched paced signal. Determine whether it is closer to or farther from the beginning. Analysis traditionally uses either (1) a printout of the second most recently paced signal or (2) the user's memory of the second most recently paced signal. This is done by comparing (or matching) the paced signal to the arrhythmia signal template. Based on this analysis, the user determines where to position the intracardiac electrode for the next paced signal. The user sequentially moves the pacing catheter in the direction of the paced signal / template match with a relatively high correlation and moves it away from the paced signal / template match with a relatively low correlation. Try to “walk” towards the beginning. By this iterative process, the onset of arrhythmia is eventually identified when a more highly correlated match is found between the paced signal and the template.

  The comparative display method of the present invention allows the user to see the most recent “step” of the walk (ie, the current probe location) and the step immediately preceding the walk (ie, the immediately preceding probe location) simultaneously. By reducing the time required to perform pace mapping. Using the comparative display method of the present invention, the user does not need to take the time to print the second most recently paced signal to indicate the next placement or orientation for placing the probe. In another embodiment, the user can use the most recent “step” of the walk (ie, the current probe location) and the best step ahead of the walk (ie, the destination that generated the signal with the best template match). And the location of the probe).

In one embodiment, the comparative display method displays the template on the first panel of the three-way split screen and displays the most recently paced signal / template match on the second panel of the three-way split screen. And displaying the second most recently paced signal / template match on the third panel of the three-way split screen.
By using the pace mapping procedure of this embodiment, multiple steps are displayed simultaneously on the “walk” so that ectopic signals can be located more quickly. A roving catheter is introduced into the heart in a conventional manner. The roving catheter includes a pacing electrode that delivers a signal that depolarizes the heart to a location in contact with the heart wall. One or more such electrodes can be included in the roving catheter.
The roving catheter is brought into contact with the heart wall at a first location and pacing pulses are delivered in a conventional manner. The pacing pulse causes cardiac depolarization, resulting in a heart waveform derived by the pacing pulse.

The pacing electrode of the roving catheter is then subjected to a voltage at a second location on the heart wall, either by applying a voltage to a different electrode while maintaining the catheter in place, or by moving the catheter to a different location. Applied. A second paced signal is derived at a second location in response to the second application.
Each of the paced signals represents a cardiac response to a pacing pulse and comprises at least one cardiac signal or cardiac signal segment. A reference template as described above is used to correlate the cardiac response to the signal beat to be located. More specifically, the reference template is a waveform representing the ectopic signal of interest, and a correlation is performed to find the best match between the reference template and the paced signal. A high correlation (ie, the best fit between the template and the paced signal) indicates a roving catheter placed at the focal point of the ectopic beat.

In order to better guide the operator during this “walk”, the best fit between the reference template and each of the first and second paced signals is displayed simultaneously. As shown in FIG. 8A, the reference template 810 is shown in the central window or frame 820 of the electronic display 800. In another window or frame 830, the reference template 810 is shown with the best fit of the relationship overlaid with the first paced signal 840. Display in window or frame 830 additional information including either a quantitative indicator of the correlation coefficient calculated to arrive at the best fit, or a graph indicator of the degree of matching or percentage indicated in that window Can do. In a further window or frame 850, the reference template 810 is again illustrated this time in an overlapping relationship with the second paced signal 860.
As a result, the operator can easily see progress or retraction seeking the focus of the ectopic signal.

For illustrative simplification, FIG. 8A shows a template constructed from only one lead signal and paced signals from only one lead.
Template matching using triggers and offsets Triggers and offsets can be set to minimize the computer requirements for processing the template on the data signal and to immediately focus the user on the ROI, resulting in less time. It becomes an electrophysiological procedure that does not consume. The trigger can be an electrocardiogram signal (eg, Q wave, R wave), a paced pulse (eg, the last pulse of a paced stimulus train), an EP waveform event, an activation pattern, or an external timing signal (eg, , A stimulator or QRS detector that provides a timing signal such as the last stimulus pulse or QRS start), or any part of any combination of the foregoing. A trigger can be further defined as a characteristic of a specified portion of an electrocardiogram signal (eg, threshold amplitude, peak, slope). For example, a user or program that manages signal processing functions defines the peak of the R wave as a trigger, which in turn triggers at the right point when the R wave reaches the peak. The characteristic of the ECG signal may be either a positive value or a negative value. For example, a positive slope or a negative slope of the Q wave may be defined as the characteristic.

The offset is the time delay following the trigger. The offset is typically set by the user in the millisecond range. The default offset can be stored and used by management software. The template matching process is executed following each offset. The offset is preferably defined by the user so that the region of interest appears in the acquired data immediately after the expiration of the offset. In this way, the computational need is concentrated in the region of interest.
Alternatively, the offset may be a negative value that represents just the desired point before the trigger event that the operator wishes to use. Thus, a trigger occurs after the start of the ROI and the software retrieves the data signal from the buffer or storage and starts template matching at the right time before the trigger event equal to the offset value.
FIG. 10A illustrates a process in which template matching is performed without a trigger or offset. In this mode of operation, the entire incoming stream of data is processed for template matching, and the user must observe the data both inside and outside the ROI.

  FIG. 10 (b) illustrates the advantage of using triggers and offsets. The user plans to use a template matching method for the stream of real-time ECG data. The user's template is an ectopic P wave. The user selects the R wave peak as a trigger and 200 milliseconds as an offset. This offset is selected, for example, based on the expectation that a P-wave (ROI) will appear at a certain time following the trigger. Real-time data is then acquired. Rather than processing a continuous stream of incoming data, rather, the microprocessor identifies the trigger event, delays the offset time period, and then initiates a template matching calculation that is close to the ROI, thereby creating a template Minimize the effort of matching processing calculations. The user can concentrate on template matching performed on the ROI without being disturbed by irrelevant comparisons performed outside the ROI.

Thus, referring to FIG. 10 (c), the trigger is identified by the peak of the R wave as indicated by arrow 1010. The trigger point starts an offset interval 1020, which in this example is 200 milliseconds. During that 20 millisecond interval, the portion of the ECG signal that precedes the region of interest, such as bounces from the R and S waves, is not processed. Thereafter, ROI 1030 occurs and this portion of the cardiac cycle is aligned to the template as described above.
Optionally, the user can select the end of the matching interval 1030. For example, the end of the matching interval 1030 can coincide with a cardiac event such as the detection of a Q wave or other waveform segment, or can coincide with a cardiac parameter such as a threshold amplitude or slope, or after an offset Or a parameter related to correlation calculation such that the correlation coefficient threshold is exceeded. In any such indicated event, there is a time window 1040 before the next trigger 1010 during which no data signal needs to be processed. This cycle again proceeds with the next trigger 1010. As a result, the above-described method in which triggers and offsets are used minimizes template matching processing outside the ROI, and without the delay and disturbance introduced by the user moving through irrelevant template matching outside the ROI. Allows you to focus on template alignment.

<Multiple spontaneous arrhythmia matching and pace mapping>
It can be difficult for the user to tell if one or more arrhythmias are present. For example, two spatially close but separate separate ectopic foci can emit morphologically similar arrhythmia signals. Determining the number of distinct arrhythmias and the location of the ectopic focus of each arrhythmia is diagnostically and therapeutically important for subsequent treatment of the arrhythmia by ablating each arrhythmia with the ectopic focus It is. According to the following method, the user can (1) determine the number of distinct arrhythmias present, and (2) use pacing mapping to locate the ectopic focus of each distinct arrhythmia. .

(1) Determine the number of distinct arrhythmias A first arrhythmia signal is acquired and defined as a template (template 1) by the user. A second arrhythmia signal is acquired and selected to correlate with template 1. The correlation coefficient is calculated to find the best alignment between template 1 and the second arrhythmia signal. The best aligned correlation that falls below a predetermined criterion (eg, the minimum correlation coefficient) causes the first arrhythmia signal and the second arrhythmia signal to be different because distinct ectopic focal points produce distinct arrhythmia signal patterns. Indicates that begins with a different ectopic focus. The best alignment correlation that meets or exceeds a predetermined criterion indicates that the first arrhythmia signal and the second arrhythmia signal are the same signal and therefore start from the same ectopic focus.

  The method described herein for determining the number of distinct arrhythmias can be repeated to allow determination of multiple distinct arrhythmias. For example, the user suspects that there are three separate arrhythmias. Subsequent to the correlation between template 1 and the second arrhythmia signal (described above), the user has the first arrhythmia signal and the second arrhythmia signal because the correlation is below the correlation criteria set by the user. Represents a separate arrhythmia. The user defines the second arrhythmia signal as template 2. A third arrhythmia signal is acquired. The third arrhythmia signal is selected to correlate with template 1 and template 2 in order. If the correlation between the third arrhythmia signal and either template 1 or template 2 meets or exceeds a predetermined criterion, this means that there are a total of two ectopic foci. It shows that. If the sequential correlation between the third arrhythmia signal and templates 1 and 2 is below a predetermined criterion, there are three ectopic foci. The method of determining the number of distinct arrhythmias can be repeated until all arrhythmia signals have been accounted for and the number of distinct arrhythmia signals has been determined. Each separate arrhythmia signal can define a separate template.

(2) Using pacing mapping to locate ectopic focus of each individual arrhythmia The method according to the present invention facilitates pace mapping of multiple ectopic focus and consequently results in less time-consuming mapping procedure become. After deriving the paced signal by the pace mapping catheter, the user can sequentially correlate the paced signal to each of the plurality of templates. Each of the plurality of templates represents a separate arrhythmia signal as defined by the user. If the correlation between the paced signal and one of the templates meets or exceeds a predetermined criterion (eg, the minimum correlation coefficient), this has been stimulated by a pace mapping catheter It shows that the location of the heart (and resulting in a paced signal) is the local focal point of a separate arrhythmia signal that defines the template used for correlation. The user can then ablate that focus. The user operates the pace mapping catheter in the heart and takes pace for subsequent sequential template correlation until all local focal points (as represented by multiple templates) have been identified and / or ablated. Deriving the correct signal.

<Activation pattern matching>
Although the reference template for the above discussion provided a defined time interval of cardiac signals obtained from one or more leads, the present invention is not so limited. As will be appreciated by those skilled in the art, a reference template can then be defined over different times, as discussed in connection with FIG. As shown in FIG. 11, the user can define templates containing electrocardiogram signals from different leads, such signals occurring at different times. Preferably, the reference ECG signal is captured using a standard 12 lead device and / or one or more intracardiac leads. In FIG. 11, the user has selected ECG surface leads I, II and III and intracardiac leads 1, 2 and 3 for the template. The user starts and ends ECB lead I waveform (A) and end point (B), ECG lead II waveform start point (C) and end point (D), ECG lead III waveform start point (E) and end point. (F), start point (G) and end point (H) of intracardiac lead 1, start point (I) and end point (J) of intracardiac lead 2, and start point (K) of intracardiac lead 3 And the end point (L) is marked. Thus, the user has defined a template that is structured from signals that appear on different leads, and some signals have the same starting point (A, C, E) and the same time as described above. It has end points (B, D, F) that occur at time, while start points (eg, E, G, I, K) that occur at different times and end points (eg, F, Some have H, J, L). Template matching then proceeds as described in connection with steps 506, 508, 510, 512, 514 and 516. Such a template can also be used for template subtraction as described in connection with the steps of FIG.

  While preferred embodiments of the present invention have thus been described, the above-described arrangements and systems are merely illustrative of the principles of the present invention and other arrangements and systems are within the spirit and scope of the following claims. It should be understood that it may be devised by those skilled in the art without departing.

It is the schematic of a normal heartbeat. 1 is a schematic diagram of atrial premature contraction (PAC). FIG. It is the schematic of T wave subtraction. FIG. 2 is a block diagram of a system programmed to perform a method according to a preferred embodiment of the present invention. It is a flowchart which shows the process for template matching according to suitable embodiment. 6 is a flowchart illustrating a process for T-wave subtraction according to a preferred embodiment. Fig. 4 represents a computer display interface for T-wave subtraction that can be displayed to an operator. Fig. 4 represents a computer display interface for template matching that can be displayed to an operator. It is the schematic of the multiple signal display of the waveform correlation with respect to a template. 6 illustrates a form for determining the integral of a section of a QRS-T segment after subtraction processing. FIG. 6 is a schematic diagram of real-time template matching without triggers and offsets. FIG. 6 is a schematic diagram of real-time template matching with trigger and offset. FIG. 6 is a schematic diagram of real-time template matching with trigger and offset. FIG. 5 is a schematic diagram of a template constructed from signals selected from multiple leads at various times.

Explanation of symbols

10 system 12 signal detection unit 14 signal processor, processor 16 display 18 storage device, memory, storage 800 electronic display 810 reference template 820 central window or frame 830 another window or frame 840 first paced signal 850 further window Or frame 860 second paced signal 1010 arrow, next trigger 1020 offset interval 1030 ROI, matching interval 1040 window of time

Claims (73)

A method for tracking ectopic beats by template matching,
(A) capturing a first ECG signal in the signal processing unit;
(B) allowing the user to mark the start and end points of the captured first ECG signal;
(C) defining a reference template as a waveform segment between the marked start and end points of the first ECG signal;
(D) obtaining data in the signal processing unit;
(E) identifying a best fit between the reference template and the acquired data using a correlation coefficient calculation on the acquired data;
Including methods.   The method of claim 1, wherein the acquired data is acquired with multiple leads at just the right point and provided from a data storage device or a real-time data stream.   An image of the reference template is displayed on the display within the data acquired across the plurality of leads so as to display the identified best fit degree between the reference template and the acquired data from each of the plurality of leads. The method of claim 2 including the additional step of aligning to the beats.   The method of claim 1 including the additional step of outputting a quantitative indicator of the correlation coefficient calculation.   5. The method of claim 4, wherein the data is obtained from the plurality of leads and the quantitative indicator is a composite average of coefficients calculated from the plurality of leads.   The method of claim 5, wherein the quantitative indicator is displayed as a graph showing a percentage of fit.   The method of claim 1, wherein the reference template is a segment of spontaneous beats and the acquired data is a paced beat.   The reference template is a segment of a first spontaneous beat, and the acquired data is a second spontaneous beat different from the first spontaneous beat. Method.   The method of claim 1, wherein the acquired data is from a real-time data stream and the method includes the additional step of repeating a correlation coefficient calculation on the acquired data at predetermined intervals. A method for generating a p-wave signal from an atrial extrasystole ("PAC") beat to assist a person diagnosing the heart, comprising:
(A) selecting a QRS-T segment of the reference ECG signal;
(B) allowing a user to mark the start and end points of the selected segment of the reference ECG signal;
(C) defining a reference template as a waveform segment between the marked start and end points of the selected segment of the reference ECG signal;
(D) obtaining the PAC beat from a plurality of leads with a signal processing unit;
(E) processing the PAC beat to generate the p-wave signal;
Including methods.   The method of claim 10, wherein the processing step includes subtracting the reference template from a predetermined segment of the PAC beat.   The method of claim 10, wherein the reference ECG signal is a single beat.   The method of claim 10, wherein the reference ECG signal is a signal generated from an average of a plurality of beats.   The method of claim 10, wherein the reference ECG signal is a beat that immediately precedes the PAC beat.   11. The method of claim 10, comprising the additional step of synchronizing the reference template and the PAC beat by aligning respective waveform segments.   The method of claim 15, wherein the alignment identifies a best fit between the respective waveform segments by using a correlation coefficient calculation on the acquired data.   16. The method of claim 15, wherein each of the waveform segments is a QRS complex of the reference template and the PAC beat.   18. The method of claim 17, wherein the alignment identifies a best fit between the QRS complex of the reference template and the QRS complex of the PAC beat by using a correlation coefficient calculation for the PAC beat.   The method of claim 16, comprising the additional step of allowing the alignment to be artificially moved, thereby causing a change in correlation coefficient calculation.   17. The additional step of allowing the reference template to be artificially moved to a waveform segment between corresponding start and end points of different beats, thereby causing a change in correlation coefficient calculation. The method described. An additional step of repeating the acquisition step and the processing step to generate a p-wave from at least two different PAC beats;
An additional step of comparing the generated p-waves to each other;
The method of claim 10 comprising:   The method of claim 21, wherein the comparing step includes performing a cross-correlation waveform analysis.   22. An additional step of selectively indicating the quality of matching to an output device as a function of the comparison step, thereby providing an indicator as to whether the generated p-waves have the same focus origin. The method described.   11. An additional step of comparing the generated p-wave to a known focal-origin p-wave library and an additional step of predicting the most likely site of the origin as a function of the comparison. The method described.   The generated p-wave is a generated spontaneous p-wave and the method paced with an additional step of manipulating a pace mapping catheter in or adjacent to the atrium while pacing the heart an additional step of repeating the acquisition step and the processing step to generate a p-wave; and an additional step of comparing the generated paced p-wave with the generated spontaneous p-wave. Item 11. The method according to Item 10.   The method of claim 10, comprising the additional step of determining an integral value of an area of the generated p-wave signal.   27. The method of claim 26, comprising the additional step of normalizing the integral over the length of the generated p-wave signal.   The marked start and end points define a QRS segment of the reference ECG signal, and the method measures a QRS residual of the generated p-wave signal to measure the QRS segment of the PAC beat and the QRS segment 28. The method of claim 27, comprising the additional step of providing an indicator of alignment quality with the QRS segment of the reference template.   29. The method of claim 28, wherein the processing step includes subtracting the reference template from the QRS segment of the PAC beat, and the QRS residual is an integral value calculated after the processing step. The obtaining step and the processing step are repeated, the method comprising:
An additional step of calculating an integral value of the QRS segment of the reference template and an integral value of the PAC beat for each iteration of the obtaining step and the processing step;
An additional step of determining any change in the absolute peak value percentage of the integral between the reference template and the PAC beat;
Including
11. The method of claim 10, whereby any baseline float is identified. A method for generating an asynchronous subcomponent from a first heartbeat signal having a composite waveform that includes a synchronous subcomponent overlying an asynchronous subcomponent to assist a person diagnosing the heart comprising:
(A) selecting a synchronization subcomponent of a second heartbeat signal corresponding to the synchronization subcomponent of the first heartbeat signal;
(B) allowing the user to mark the start and end points of the selected synchronization subcomponent;
(C) defining a reference template as a waveform segment between the marked start and end points of the selected synchronization subcomponent;
(D) obtaining a composite waveform of the first heartbeat signal from a plurality of leads with a signal processing unit;
(E) processing the composite waveform beat to generate the asynchronous subcomponent.   32. The method of claim 31, wherein the processing step includes subtracting the reference template from a predetermined segment of the composite waveform.   32. The method of claim 31, wherein the selected synchronization subcomponent is from a single beat.   32. The method of claim 31, wherein the selected synchronization subcomponent is a signal generated from an average of a plurality of beats.   32. The method of claim 31, wherein the selected synchronization subcomponent is from a beat that immediately precedes the composite waveform.   32. The method of claim 31, comprising the additional step of synchronizing the reference template and the composite waveform by aligning respective synchronized waveform segments.   The method of claim 36, wherein the alignment identifies a best fit between the respective synchronized waveform segments by using a correlation coefficient calculation on the acquired data.   40. The method of claim 36, wherein each waveform segment is a synchronization subcomponent of the reference template and the composite waveform.   39. The method of claim 38, wherein the alignment identifies a best match between the reference template and the composite waveform synchronization subcomponent by using a correlation coefficient calculation for the composite waveform.   38. The method of claim 37, comprising the additional step of allowing the alignment to be artificially moved, thereby causing a change in correlation coefficient calculation.   38. The method includes the additional step of allowing the reference template to be artificially moved to a waveform segment between corresponding start and end points of different heartbeats, thereby causing a change in correlation coefficient calculation. the method of. An additional step of repeating the acquisition step and the processing step to generate an asynchronous subcomponent from at least two different composite waveforms;
32. The method of claim 31, comprising the additional step of comparing the generated asynchronous subcomponents to each other.   43. The method of claim 42, wherein the comparing step includes performing a cross-correlation waveform analysis.   Including the additional step of selectively indicating quality of matching to an output device as a function of the comparing step, thereby providing an indicator as to whether or not the generated asynchronous sub-components have the same focus origin. Item 43. The method according to Item 42.   An additional step of comparing the generated asynchronous subcomponent to a library of known focus-origin asynchronous subcomponents, and an additional step of predicting the most likely portion of the origin as a function of the comparison. 32. The method of claim 31, comprising.   The generated asynchronous subcomponent is a generated spontaneous asynchronous subcomponent, the method comprising the additional step of manipulating a pace mapping catheter in or adjacent to the atrium while pacing the heart; The acquiring step and the generating step to generate a paced asynchronous subcomponent until such time as the generated paced subcomponent and the spontaneous subcomponent are correlated with each other within a predetermined criterion; 32. The method of claim 31, comprising the additional step of repeating the processing steps.   32. The method of claim 31, comprising the additional step of determining an integral value of the generated asynchronous subcomponent area.   48. The method of claim 47, comprising the additional step of normalizing the integral over the length of the generated asynchronous subcomponent. An additional step of comparing the generated asynchronous subcomponent to a known focal-origin asynchronous subcomponent library, wherein the generated asynchronous subcomponent is a spontaneous asynchronous subcomponent When,
An additional step of predicting the most likely part of the origin as a function of the comparison;
An additional step of manipulating the pace mapping catheter in or adjacent to the atria while pacing the heart in real time;
The acquiring step and the generating step to generate a paced asynchronous subcomponent until such time as the generated paced subcomponent and the spontaneous subcomponent are correlated with each other within a predetermined criterion; An additional step that repeats the processing steps;
32. The method of claim 31 comprising: A method for locating ectopic beats during pace mapping with a roving catheter,
(A) deriving at least first and second paced signals from first and second locations, respectively, of the roving catheter;
(B) using a correlation coefficient calculation on the derived first and second paced signals to best between a reference template and each of the first and second paced signals; Identifying the fit of
(C) simultaneously displaying the best fit for each of the first and second paced signals on a display;
Including methods.   51. The method of claim 50, wherein the reference template comprises a single cardiac signal waveform segment comprising an arrhythmia component.   52. The method of claim 51, further comprising displaying the reference template on the display while displaying the first and second paced signals.   51. The method of claim 50, comprising the additional step of displaying a quantitative indicator for each correlation coefficient calculation on the display.   51. The method of claim 50, wherein data is obtained from a plurality of leads and the quantitative indicator is a composite average of coefficients calculated from the plurality of leads.   51. The method of claim 50, wherein the quantitative indicator is displayed as a graph showing a percentage of fit. A method for tracking ectopic beats by template matching,
(A) establishing a reference template over an interval of the first ECG signal;
(B) monitoring a data signal for a trigger event;
(C) starting an offset period in response to the trigger event;
(D) identifying a best match between the reference template and the data signal over the interval using a correlation coefficient calculation on the data signal after the offset period has passed;
Including methods.   57. The method of claim 56, wherein the trigger event is defined by a user.   57. The method of claim 56, wherein the trigger event comprises one of a waveform characteristic, a pacing pulse, an activation sequence, an external timing signal, and combinations thereof.   The data signal is obtained from a real-time data stream including continuous trigger events, and the method includes the additional step of repeating steps (c) and (d) in response to each continuous trigger event of the real-time data stream. 56. The method according to 56.   57. The method of claim 56, wherein the correlation coefficient calculation ends at a specified event.   61. The method of claim 60, wherein the specified event is a correlation coefficient value threshold.   57. The method of claim 56, comprising the additional step of processing a portion of the data signal corresponding to the identified best fit.   The processing step includes defining a waveform generated by subtracting the reference template from a portion of the data signal corresponding to the identified best fit, the method comprising: 64. The method of claim 62, comprising the additional step of displaying on a display.   64. The method of claim 62, wherein the processing step comprises ablating heart tissue. A method for identifying a plurality of distinct arrhythmias comprising:
(A) obtaining a first arrhythmia signal;
(B) defining the first arrhythmia signal as a first template;
(C) obtaining a second arrhythmia signal;
(D) correlating the first arrhythmia signal and the second arrhythmia signal;
(E) identifying the second arrhythmia signal as a second separate arrhythmia if the correlation does not reach a predetermined criterion;
Including methods. There are more than two arrhythmia signals and comprises a set of templates, the method comprising:
(A) defining the second discrete arrhythmia as a second template;
(B) obtaining an additional arrhythmia signal;
(C) sequentially correlating the additional arrhythmia signal to each template in the set of templates;
(D) identifying the additional arrhythmia signal as a separate arrhythmia signal if the correlation does not reach a predetermined criterion;
(E) defining the additional separate arrhythmia signal as an additional template;
Repeating steps (b)-(e) until there are no additional arrhythmia signals that do not reach the predetermined criteria;
66. The method of claim 65, further comprising: (A) an additional step of obtaining a paced signal produced by a pace mapping catheter in or adjacent to the heart;
(B) an additional step of sequentially correlating the paced signal to each of the templates of the set;
(C) an additional step of identifying the location of the ectopic focus when the correlation of the paced signal to one of the templates meets or exceeds a predetermined criterion;
68. The method of claim 66, comprising:   68. The method of claim 67, further comprising ablating the ectopic focus.   68. The method of claim 67, further comprising repeating steps (a)-(c) until the location of the focal focus of each distinct arrhythmia signal is identified.   The method of claim 1, wherein the defining step includes defining the reference template as a set of waveform segments obtained from a plurality of electrocardiogram leads between a marked start and end point.   71. The method of claim 70, wherein the starting point of the waveform segment occurs in a very different template.   71. The method of claim 70, wherein the end points of the waveform segments occur in exactly the same template.   The first paced signal is the most recently paced signal, and the second paced signal is before the first paced signal having the best fit. The method of claim 70, wherein the signal is a paced signal.

Images