WO2008035070A2 - Atrial fibrillation analysis - Google Patents

Atrial fibrillation analysis Download PDF

Info

Publication number
WO2008035070A2
WO2008035070A2 PCT/GB2007/003558 GB2007003558W WO2008035070A2 WO 2008035070 A2 WO2008035070 A2 WO 2008035070A2 GB 2007003558 W GB2007003558 W GB 2007003558W WO 2008035070 A2 WO2008035070 A2 WO 2008035070A2
Authority
WO
WIPO (PCT)
Prior art keywords
step
electrogram
method according
patient
dominant frequency
Prior art date
Application number
PCT/GB2007/003558
Other languages
French (fr)
Other versions
WO2008035070A3 (en
Inventor
Julian William Ernest Jarman
Darrel Parthipan Francis
Nicholas Simon Peters
Justin Edgar Rees Davies
Original Assignee
Imperial Innovations Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to GB0618522.7 priority Critical
Priority to GBGB0618522.7A priority patent/GB0618522D0/en
Application filed by Imperial Innovations Limited filed Critical Imperial Innovations Limited
Publication of WO2008035070A2 publication Critical patent/WO2008035070A2/en
Publication of WO2008035070A3 publication Critical patent/WO2008035070A3/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/04Measuring bioelectric signals of the body or parts thereof
    • A61B5/0402Electrocardiography, i.e. ECG
    • A61B5/0452Detecting specific parameters of the electrocardiograph cycle
    • A61B5/046Detecting fibrillation
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7253Details of waveform analysis characterised by using transforms
    • A61B5/7257Details of waveform analysis characterised by using transforms using Fourier transforms
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B18/00Surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body
    • A61B18/04Surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body by heating
    • A61B18/12Surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body by heating by passing a current through the tissue to be heated, e.g. high-frequency current
    • A61B18/14Probes or electrodes therefor
    • A61B18/1492Probes or electrodes therefor having a flexible, catheter-like structure, e.g. for heart ablation
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B18/00Surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body
    • A61B2018/00315Surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body for treatment of particular body parts
    • A61B2018/00345Vascular system
    • A61B2018/00351Heart
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B18/00Surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body
    • A61B2018/00636Sensing and controlling the application of energy
    • A61B2018/00773Sensed parameters
    • A61B2018/00839Bioelectrical parameters, e.g. ECG, EEG
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10Computer-aided planning, simulation or modelling of surgical operations
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61NELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N1/00Electrotherapy; Circuits therefor
    • A61N1/18Applying electric currents by contact electrodes
    • A61N1/32Applying electric currents by contact electrodes alternating or intermittent currents
    • A61N1/36Applying electric currents by contact electrodes alternating or intermittent currents for stimulation
    • A61N1/362Heart stimulators
    • A61N1/3621Heart stimulators for treating or preventing abnormally high heart rate
    • A61N1/3624Heart stimulators for treating or preventing abnormally high heart rate occurring in the atrium, i.e. atrial tachycardia

Abstract

A method of generating a dominant frequency map of a cardiac area in a patient. The method comprises the steps of: (i) processing a plurality of electrograms from the patient by Fourier transformation to generate a power frequency spectrum. Each electrogram corresponds to a portion of the cardiac area measured substantially simultaneously. Step (ii) comprises selecting the highest value from each power frequency spectrum as the dominant frequency at the corresponding portion of the cardiac area. Each dominant frequency is determined in respect of a period of at least 7 seconds.

Description

ATRIAL FIBRILLATION ANALYSIS

TECHNICAL FIELD

The present invention relates to a method of generating a dominant frequency map of a cardiac area of a patient. The present invention also relates to a method of determining the dominant frequency of an electrogram from a patient and to a method of mapping a cardiac area. The present invention also relates to a method of elucidating the atrial activity from an electrocardiographic recording that contains ventricular signals that are unwanted. The present invention further relates to a method for assessing the requirement for ablation therapy on a patient suffering from atrial fibrillation and to a method of ablating portions of a cardiac area of such a patient. The present invention also relates to a method of recording an electrogram of a patient.

BACKGROUND ART

Atrial fibrillation (AF) is a chaotic heart rhythm which causes strokes and increased mortality as well as reduced quality of life. Despite decades of intense speculation and investigation, the mechanisms underlying atrial fibrillation (AF), paramount to further development of effective treatment strategies, are not fully understood. This is in part related to the difficulties in interpreting complex and constantly changing electrical activation, for which high resolution simultaneous global atrial mapping technology is required (or sequential mapping with later registration).

Cardiac electrophysiology (EP) is a subspecialty of cardiology which is growing rapidly. The curative treatment of AF by catheter ablation (burning inside the heart using a minimally invasive approach) is the central focus of EP research and the fastest growing area in EP. Current techniques for EP are inadequate with disappointing success rates, particularly in chronic AF. One area of development of AF catheter ablation is in the use of spectral analysis techniques to identify critical areas within the atria to target with ablation. Spectral analysis techniques use Fourier transformation to identify high dominant frequencies within the chaotic AF signals recorded within the heart - areas where high dominant frequencies are seen have been shown to be promising areas for targeting with ablation. Identification of critical areas is advantageous as targeting them can increase procedural success rates and also reduce complications by reducing the amount of radiofrequency energy it is necessary to deliver to achieve success.

However, it has now been found that there are serious flaws in the prior art techniques of spectral analysis.

In existing procedures of spectral analysis, it is typical for investigations to be based on data recordings of unipolar electrograms of less than 7 seconds in length. Such recordings are analysed using Fourier transformation in order to identify portions of the cardiac area with high dominant frequencies. This practice stems from the belief that dominant frequency values are stable over time (Sanders P, Berenfeld O, Hocini M et al. Spectral Analysis Identifies Sites of High-Frequency Activity Maintaining Atrial Fibrillation in Humans. Circulation 2005 August 9;112(6):789-97). However, the present inventors have found that dominant frequency values are not generally stable over time and that investigations using data recordings of longer than 7 seconds have significantly improved accuracy.

As has been explained, the analysis of atrial signals is carried out using

Fourier transformation in order to determine the dominant frequency. However, the present inventors have found these results to be inaccurate when using a unipolar electrogram unless the ventricular far field component of the unipolar electrogram signal is first removed. Furthermore, other inaccuracies can occur due to the presence of AC "hum" in the signal when attempting to subtract the far field ventricular component.

It has previously been proposed to identify critical areas for ablation in patients suffering from atrial fibrillation by referring to dominant frequency maps of the cardiac area of the patient. The present inventors have now determined that it is beneficial to target portions with highly organised AF and which are temporospatially stable as well as having a relatively high dominant frequency.

One problem with cardiac ablation therapy is that it requires the application of radiofrequency energy via catheter and this is associated with certain procedural complications and may even cause long term damage to the atria. It would therefore be desirable to be able to predict the long term outcome of catheter ablation so that the length and aggression of the procedure can be adapted to the clinical requirements of the patient. This would be useful both prior to ablation therapy and after a first procedure when the need for a subsequent procedure could be assessed.

The ablating procedure typically involves ablating individual sites of high dominant frequency following standard lines. However, the present inventors have now found that it is advantageous to carry out a linear ablation by modifying the standard lines in order to pass through all sites of high dominant frequency.

The present invention seeks to improve on the prior art methods. SUMMARY OF THE INVENTION

According to one aspect of the present invention, there is provided a method of generating a dominant frequency map of a cardiac area in a patient comprising the steps of:

(i) processing at least one electrogram from the patient by Fourier transformation to generate a power frequency spectrum, the at least one electrogram corresponding to a portion of the cardiac area; and

(ii) selecting the highest value from the or each power frequency spectrum as the dominant frequency at the corresponding portion of the cardiac area, wherein the or each dominant frequency is determined in respect of a period of at least 7 seconds.

A "power frequency spectrum" shows the power of each frequency within a signal (in this case an electrogram). Typically, the power is shown in the y- axis and the frequency in the x-axis.

By determining the dominant frequency over a period of this length (in particular, longer than 7 seconds) the variations in dominant frequency which occur over time are avoided and the accuracy of the determination improved.

It is preferred that fast Fourier transformation is carried out in the methods of the present invention although other forms of Fourier transformation, or other frequency transforms, such as any of a family of wavelet transforms, may be usable in a similar manner.

Generally, the highest output in the 3 to 30 Hz range, or thereabouts, is select in step (ii).

Conveniently, the period is at least 10, 15, 20, 25 or 30 seconds. Preferably, the method further comprises the step of, prior to step (i), linking together a plurality of electrograms for the or each portion, each electrogram being shorter than the period in order to produce an electrogram of at least the length of the period for the or each portion. This is necessary if recordings are generated for time periods of insufficient length (such as the current EnSite system)

Alternatively, step (i) comprises, for the or each portion, processing a plurality of electrograms, each being in relation to a period of less than the period and then averaging the processed signals so as to relate to at least the length of the period. This approach also permits analysis over a period of sufficient length to be achieved when recordings are for less than the required time.

Advantageously, the cardiac area is at least the left atrium of the patient.

Conveniently, the method further comprises the step of, prior to step (i), recording the or each electrogram.

Preferably, for unipolar electrograms the step of recording the or each electrogram is carried out using a contactless system. This is preferred because contact recording systems generally require each portion of the cardiac area to be measured sequentially (rather than simultaneously) so contactless systems are much quicker at recording from multiple sites.

Advantageously, step (i) comprises processing a plurality of electrograms, each electrogram corresponding to a portion of the cardiac area measured substantially simultaneously. It is to be understood that in preferred embodiments, the map is used in order to guide ablation (e.g. catheter ablation) procedures including the methods of ablating of the present invention.

According to another aspect of the present invention, there is provided a method of determining the dominant frequency of an electrogram from a patient comprising the steps of:

(i) selecting a timing reference signal;

(ii) detecting a repeating physiological signal in the electrogram; (iii) subtracting a representation of the repeating physiological signal from the electrogram using the timing reference signal to calculate the timing of each physiological signal;

(iv) processing the resultant signal using Fourier transformation to generate a power frequency spectrum; and (v) selecting the highest value (or highest region) from the power frequency spectrum as the dominant frequency.

In some embodiments, steps (iv) and/or (v) are omitted.

The physiological signal need not be perfectly identical at each repeat but may vary mildly in morphology. The "representation" of the repeating physiological signal is a synthesis, average or the like of the signal or may be the signal itself.

The subtraction of the physiological signal from the electrogram prior to processing in step (iv) results in a much more accurate power frequency spectrum for atrial activity which results in the selecting of the dominant frequency of atrial activity in step (v) being more accurate. It is to be understood that this method may be used in the methods of generating cardiac maps of the invention and may rely on information provided by the method of recording an electrogram of the invention.

It is preferred that the selection in step (i) is carried out on the basis of a skew of the distribution of voltages of the electrogram. That is to say, those voltages of the electrogram which have a large deviation from the mean voltages are selected and used to generate the timing reference signal.

Conveniently, the physiological signal is a ventricular depolarization and repolarization wave.

Alternatively, the physiological signal is a respiratory wave.

Preferably, the method further comprises the step of filtering the electrogram for signals corresponding to the frequency of mains electricity. In Europe and some parts of Japan this is generally 50Hz. In North America and other parts of Japan this is generally 60Hz.

Advantageously, the method further comprises the step of, prior to step (i), filtering the electrogram through a bandwidth filter. The bandwidth that is selected for filtering may be varied depending on circumstances. A range of 1 to 150Hz is preferred but a range of, for example 2 to 300 Hz is also suitable.

Conveniently, step (iii) comprises the step of averaging each physiological signal to generate a template and subtracting the template from the electrogram at each incidence of the physiological signal.

Preferably, the step of averaging the physiological signal comprises overlaying each signal from approximately 100ms before to approximately 800ms after the timing reference signal. This ensures that the whole ventricular depolorization and repolorization wave is encompassed.

Advantageously, the step of averaging each physiological signal comprises carrying out a weighted average.

Conveniently, the physiological signal is a ventricular depolarization wave and the electrogram has been recorded by a monitoring device and wherein the weighted average is calculated with reference to: the phase of the respiratory cycle of the patient; the interval between a current ventricular complex and a previous ventricular complex; the interval between two preceding ventricular complexes; the location and orientation of the monitoring device; the drift in position of the monitoring device; and combinations thereof.

Preferably, the method is carried out during recording of the electrogram.

Conveniently, at least step (iii) is carried out continually.

Step (v) is preferably carried out in approximately the 3 to 30Hz range.

Advantageously, the method is carried out on a plurality of electrograms each in relation to different portions of a cardiac area of the patient, preferably wherein each is obtained substantially simultaneously. This approach is used, for example, in order to generate a map of a cardiac area of a patient showing the dominant frequencies in various portions (i.e. locations).

Conveniently, step (i) comprises the step of, generating a series of timing reference points from the timing reference signal (that is to say, the timing reference signal for one particular time during the recording) by selecting one electrogram to provide each timing reference point, said timing reference point being applied to every electrogram in step (iii). By adopting this procedure, the output from one particular portion is used to generate the timing reference signal which is then used to calculate the position of a timing reference point or fiducial point for the electrogram at each portion in order to carry out the subtracting step correctly. The output from the particular portion providing the clearest signal at that section of the recording is generally selected.

According to further aspect of the present invention, there is provided a method of generating a map of dominant frequencies of a cardiac area of a patient comprising carrying out the method of the invention of determining the dominant frequency of an electrogram from the patient and plotting the frequencies at each portion where the dominant frequency has been determined.

According to another aspect of the present invention, there is provided a method of generating a map of a cardiac area of a patient indicating portions of the cardiac area responsible for atrial fibrillation comprising the steps of:

(i) processing at least one electrogram from the patient by Fourier transformation to generate a power frequency spectrum, the or each electrogram corresponding to a portion of the cardiac area;

(ii) selecting the highest value from the or each power frequency spectrum as the dominant frequency of the corresponding portion of the cardiac area, the dominant frequency thereby having a corresponding dominant frequency peak in the power frequency spectrum;

(iii) determining an organisational index at the or each portion by calculating the area beneath the dominant frequency peak and its harmonic peaks as a fraction of the total area under the power frequency spectrum; and

(iv) determining a temporal stability index at the or each portion by determining the dominant frequency at the or each portion over a plurality of successive time periods and calculating the standard deviation or average change of the dominant frequency over the successive time periods.

A "combined" map generated in this way identifies portions of a cardiac area with highly organized atrial fibrillation which are temporospatially stable and have a relatively high dominant frequency. Such maps thereby provide a more accurate prediction of targets for ablation.

In some embodiments, step (iii) or (iv) may be omitted.

It is to be understood that the method may be used in conjunction with the method of recording an electrogram of the invention, the method of determining the dominant frequency of an electrogram of the invention and the method of generating a dominant frequency map of the present invention.

Conveniently, in step (iii), the harmonics of the dominant frequency peak for approximately 0.25 to 0.5H2 on either side of each peak are used.

Preferably, in step (iv), each time period is between three and sixty seconds long, most preferably 6.82 seconds long with the data sampling frequency being 1.2kHz.

Advantageously, the method further comprises the step of plotting the value from step (ii), the organisational index from step (iii) and the temporal stability index from step (iv) at each respective location of the cardiac area. In this way a map of the cardiac area with this data applied to it is formed.

Conveniently, a portion of the cardiac area responsible for atrial fibrillation is indicated by a higher relative value of dominant frequency, organisational index and/or temporal stability index. Preferably, step (i) comprises processing a plurality of electrograms from the patient.

According to yet another aspect of the present invention there is provided a method of assessing the requirement for ablation therapy on a patient suffering from atrial fibrillation comprising the steps of:

(i) determining the dominant frequency at at least one portion of the cardiac area of the patient; and (ii) comparing the dominant frequency with a reference value.

The dominant frequency may be determined from an electrogram of the patient.

Step (i) may be achieved by determining an average value for all sites of a unipolar electrogram (e.g. the mean value for a whole chamber of 256 sites).

By using such a method, the need for an ablation procedure in a patient can be determined and, if so, the intensity of ablation required for the procedure can be assessed. This can be useful at the beginning of an ablation procedure to determine if a more aggressive procedure would be appropriate being carried out and/or during an ablation procedure whether further ablation is required in addition to lesions already created.

It is to be understood that this method may be use in conjunction with the method of recording an electrogram of the invention, the method of determining the dominant frequency of the invention, and the methods of generating maps of the invention. Conveniently, step (i) comprises determining the dominant frequency at a plurality of portions of the cardiac area of the patient.

Advantageously, the dominant frequency is determined for the left atrium of the patient.

Preferably, step (i) further comprises a step of determining the mean dominant frequency of all portions and step (ii) comprises comparing the mean dominant frequency with a reference value.

Advantageously, the reference value in step (ii) is approximtely 5Hz, a mean dominant frequency of greater than 5Hz being indicative of the need for ablation therapy.

Conveniently, step (i) comprises determining the dominant frequency at a portion adjacent to the pulmonary vein; adjacent to the septum; adjacent to the left atrial appendage; within the coronary sinus or combinations thereof.

Preferably step (i) comprises determining the dominant frequency in the left atrium.

Alternatively, step (i) comprises determining the dominant frequency of the surface electrocardiogram of the patient. Such an approach allows an overall dominant frequency value to be determined without invasive assessment.

Advantageously the surface ECG signal is obtained from the V1 lead of the ECG equipment.

Conveniently, an ablation procedure is carried following completion of the assessment method, the procedure being adapted to follow the results of the assessment method. Alternatively, the assessment method is carried out during an ablation procedure in order to determine whether further ablation is required.

According to a further aspect of the present invention, there is provided a method of ablating portions of a cardiac area of a patient suffering from atrial fibrillation comprising the steps of:

(i) identifying a plurality of portions of the cardiac area of the patient having a dominant frequency above a threshold value in an electrogram; and

(ii) carrying out a linear ablation in the cardiac area along a line joining the portions identified in step (i).

This aspect of the invention permits existing techniques of catheter ablation to be adapted to incorporate information from spectral analysis. Such techniques include but are not limited to: pulmonary vein isolation with additional linear lesions; pulmonary vein isolation with additional targeting of critical areas; encircling linear ablation; nonencircling linear ablation; and targeting of "critical structures" without linear ablation or pulmonary vein isolation.

Thus encircling linear lesions still encircle the pulmonary veins but vary in distance from the pulmonary vein ostia to allow them to pass over critical areas on the maps. Non-encircling linear lesions are placed through critical areas of the maps. Additional linear lesions (for instance at roof, at posterior wall, or at mitral valve annulus) continue to transect the roof area, posterior wall area or area superoposterior to the mitral valve annulus, but are varied in exact position to pass over critical areas on the maps (also the order of preference of the use of these lines is dictated by the maps). Targeting of critical areas is performed purely on the basis of the maps, though favouring the base of the left atrial appendage and proximal coronary sinus area also. Alternatively targeting of critical areas is performed, on the basis of the maps, as described above, but without any linear ablation or pulmonary vein isolation.

It is to be understood that this method may rely on the other methods of the invention in particular the method of recording an electrogram; the method of determining the dominant frequency and the method of generating a dominant frequency map. It is particularly to be noted that the method may rely on the provision of a "combined map" comprising information on the organizational index and the temporal stability index, as well as dominant frequency values, as described above. Thus the route of the ablation is carried out taking these indices into account as well as the dominant frequency values.

By modifying standard ablation lines to go through portions of high dominant frequency, while still continuing to their usual destinations, the amount of radiofrequency energy that needs to be delivered to the patient is limited.

According to another aspect of the present invention there is provided a method of recording an electrogram of a patient comprising the steps of: (i) determining the intrinsic ventricular rate of the patient;

(ii) pacing the ventricle of the patient at a rate higher than the intrinsic ventricular rate of the patient; and recording the electrogram of the patient.

By using this method the variability in the morphology of the ventricular component of an electrogram from beat to beat is reduced or eliminated allowing the ventricular component to be subtracted more effectively. This is because the pacing of the ventricle results in the resultant ventricular signal having consistent morphology from beat to beat and being regular in timing. It is preferred that the recordings are discarded in which capture or fusion beats occur.

It is to be understood that this method may be used in conjunction with the other methods of the present invention. In particular the method may be combined with the other method of recording an electrogram of the invention, the methods of the invention for determining a dominant frequency and the method of generating a dominant frequency map.

Conveniently, the method further comprises the step of:

(iv) subtracting at least a part of a physiological component of the patient from the electrogram, preferably the ventricular component and preferably as described above.

Preferably, the method further comprises the step of:

(v) confirming that the ventricular component has been subtracted from the electrogram of the patient up to a predetermined threshold. This step is possible because any remaining ventricular signal has a very distinctive appearance in the power frequency spectrum following Fourier transformation.

It is to be emphasised that the methods of the present invention may be carried out in relation to any type of electrogram including unipolar electrograms (e.g. those produced by the ESI system), and bipolar electrograms e.g. produced by single point, sequential systems such as Carto).

In preferred embodiments, the methods of the present invention are implemented on a computer. Thus in other aspects of the present invention there is provided a processor programmed to carry out the methods described above.

FIGURES

Figure 1 shows a series of traces demonstrating the subtraction of ventricular components from a unipolar electrogram. (A) is a trace of a recorded unipolar electrogram from a patient. (B) is a trace of all ventricular components in the segment from (A) overlaid. (C) and (D) show traces of the mean of the overlaid trace from (B). (E) shows the trace from (A) with the mean ventricular component from (D) overlaid. (F) shows the trace of the unipolar electrogram of (A) with the mean ventricular component from (D) subtracted. (G) shows the results of Fourier transformation of the unipolar electrogram of (A) without subtraction of the ventricular component. (H) shows the results of Fourier transformation of the unipolar electrogram of (A) after subtraction of the ventricular component.

Figure 2 is an illustration of the determination of the dominant frequency at two portions of the cardiac area of a patient comparing contact and contactless electrograms. The left atrial geometry is displayed in a right anterior oblique projection. The position of the Ensite balloon is visible within the geometry, and points on the visible surfaces where validation data was gathered are marked with lighter grey dots. Raw contact and virtual electrograms acquired at a single left atrial location are displayed together (A), the coefficient of correlation between the two signals is 0.93. The power frequency spectra following Fourier transformation for the two signals are also displayed over a 3 to 15Hz range of frequencies on the X axis (B), demonstrating that the highest peak (dominant frequency) occurs at the same frequency in both cases. The same data for an alternative left atrial location is also displayed (C, D). LAA indicates left atrial appendage; MVA, mitral valve annulus; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein.

Figure 3 shows dominant frequency mapping of two postero-anterior views of the left atrium of patients with (A) paroxysmal and (B) persistent AF. The dominant frequencies from 256 evenly distributed sites are displayed in greyscale and the frequency greyscale spectrum is illustrated in the column on the left. On the left map from a patient with paroxysmal AF (A) a discrete area of high dominant frequency (DADF) is visible on the posterior wall near the left inferior pulmonary vein. On the right map from a patient with chronic

AF (B) the frequencies are relatively homogenous without any discrete high frequency areas.

Figure 4A is a pie chart of the location of the focal DFmaχ in a study of 24 patients. Ant. wall = anterior wall; LAA = left atrial appendage; Lat. wall = lateral wall; Post, wall = posterior wall; PVs = pulmonary veins.

Figure 4B is a graph of the magnitude of focal DFmax by location.

Figure 5 shows four consecutive segments, (A) to (D), of greyscale dominant frequency maps during a single episode. They show spatial and temporal stability of the focal area of high dominant frequency in a patient who had paroxysmal AF.

Figure 6 is a series of traces resulting from a unipolar electrogram signal from a patient, (a) shows the raw signal, (b) shows the raw signal following Fourier transformation, (c) shows the raw signal following high quality subtraction of the ventricular components, (d) shows the signal of (c) following Fourier transformation, (e) shows the raw signal following poor quality subtraction of the ventricular components, (f) shows the signal of (e) following Fourier transformation.

Figure 7a is a left atrial dominant frequency map from a patient calculated over a 7 second period. Figure 7b is a left atrial dominant frequency map from the patient calculated over a 7 second period immediately following the 7 second period used to generate the map of Figure 7a.

Figure 8 is an organisational index map of a patient showing the ventricular- subtracted electrogram and corresponding power frequency spectrum at two locations.

In one embodiment, the following steps are carried out on a unipolar electrogram of a patient in order to determine the dominant frequency. The steps are carried out on a recording of a patient monitored with the EnSite system in which 256 leads (i.e. electrodes) in the cardiac area independently monitor different locations.

(a) The signals are filtered with a 1 to 150Hz bandwidth filter prior to export from EnSite system. (b) The best signal for use as the timing reference signal is automatically detected in order to select a lead with a clear ventricular depolarization wave (VDW) on the basis of skew.

(c) The timing of peaks of VDWs are detected on the timing reference signal on the basis of peak voltage. These fiducial points are referenced from the timing reference signal when processing the signals at other leads. (d) A 50Hz filter is applied to eliminate mains hum and thus avoid misalignment of signals when overlaying.

(e) The signals are overlaid around each VDW from 100ms before, and 800ms after the fiducial point. They are aligned on the fiducial point. Overlaid signals are averaged and an averaged signal from the timing reference electrode is presented to the operator, with standard deviation of signals marked on it.

(f) The timing of the onset of VDW and end of ventricular repolarization wave are identified on the average signal. The resultant average signal between these time points is used as a template for subtraction from the raw signal. (g) The template is subtracted from the raw signal around each VDW, aligning itself on the fiducial point of the raw signal. (h) This process is repeated on the signal at each lead, always referencing time points from the timing reference signal, (i) The output of the process is then subjected to Fourier transformation.

The highest peak in the 3 to 30 Hz range is selected as the dominant frequency of the resultant power frequency spectrum to create a dominant frequency map.

In some variants to this embodiment, the averaging carried out in step (e) is carried out in a more sophisticated manner. In particular, the algorithm takes into account physiological properties that can materially affect the shape of the far field ventricular signal, including (but not limited to): the phase of the respiratory cycle the interval between the current ventricular complex and the previous ventricular complex, hereafter the "RR interval" optionally, the immediately preceding few RR intervals - in embodiments in which an ESI balloon (or similar device) is provided for data acquisition, and data is available on the average location and orientation of the balloon device during the current cardiac cycle, the location of the ESI balloon. the gradual drift of position of the balloon with respect to the patient with time over a period of minutes To do this, the "averaged ventricular depolarization and repolarization wave" or "ventricular complex" which the algorithm subtracts at each electrode position at each time point, is not merely the arithmetic mean of the ventricular complexes at that electrode position during the recording window. Instead, it is a weighted mean of the ventricular complexes at that electrode position during the recording window. The weighting factor used is of the form:

The ventricular complex number = i, whose values run from 1 to nqrs

The ventricular complex currently being considered (the one which the algorithm is trying to remove) = k

The weights can be amalgamated by simple averaging, or by geometric averaging, an example for four weighting factors is:

W(i) = ( W_RR(i) * W_Resp(i) * W_balloon(i) * W_drift(i) )Λ(1/4)

A series of weighting factors may be used. Continuing the above example, the following weighting factors are included:

W_RR(i) is the weighting factor that favours ventricular complexes with similar RR intervals to the current.

W_RR(i) = K_RR * Exp(- ((RR(i)-RR(k))/Stdev(RR(1..nqrs)) )Λ2 )

wherein

K_RR = constant indicating how relatively important RR interval is as a predictor of ventricular complex shape (e.g. 1.0) Stdev means standard deviation

W_Resp(i) is the weighting factor that favours ventricular complexes at a similar phase of respiration to the current.

W_Resp(i) = K_Resρ * (pi - abs(mod(Resp(i)-Resp(k)),2*pi) ) )

wherein

K_Resp = constant indicating how relatively important respiratory phase is as a predictor of ventricular complex shape (e.g. 1.0)

Resp(i) is a variable describing the phase of respiration at the i'th ventricular complex, in the form of radians (0 - 2 pi). The form of

"( pi -abs(mod(Resp(i)-Resp(k)),2*pi) ) " is designed to be maximal when the phase of respiration is identical in the two beats being compared, and zero when the phase is most different (180 degrees out of phase). Wherein "mod(x,y)" represents the modulo operator, which has the effect of successively adding or subtracting (in this case) y to the operand x, until the result is within the range of 0 to y.

Details of how to calculate the Resp(i) signal in real time are provided in UK Patent Application 0607939.6 which is hereby incorporated by reference. In the present invention the algorithm disclosed in GB0607939.6 is applied not for the period of the cycle of periodic breathing (typically 1 minute) but to the period of a single breath (typically 3-5 seconds). Thus the input to the Resp(i) signal is, for example, chest wall position, rather than minute ventilation. In summary, in the present invention, the average respiratory rate is estimated using a Fourier transform over approximately the preceding minute. This rate is used to determine the duration of (advantageously) one or two complete cycles of respiration. The segment of chest wall position data from that duration of time is then selected. A new Fourier transform is then applied to a low-pass filtered version of this segment of respiration data. From this, the Fourier component whose frequency corresponds to the patient's current respiratory rate is read, and its phase is examined: this is the current phase of respiration. In one embodiment, the segment of data being examined is a single respiratory cycle, and therefore the Fourier component whose frequency is of interest is the lowest oscillatory frequency. It is only necessary to perform this process of determination of current phase of respiration once for every ventricular complex, since the value is common for application to all the electrogram signals at that instant in time.

W_balloon(i) is the weighting factor relating to the position of the balloon. It can be derived from a 3-dimensional generalisation of W_Resp(i).

W_drift(i) is the weighting factor relating to the drift of the balloon relative to the patient over time.

W_drift(i) = K_drift * exp(- (t(i) - t(k))/T1 )

wherein

K_drift and T1 are constants indicating how important the passage of time is (i.e. how much we de-emphasise ventricular complexes several minutes ago, for example)

t(i) is the time of the i'th ventricular complex In some embodiments, the algorithm and control of the unipolar electrogram are run on a computer. In certain embodiments, the computer is programmed to cumulate ventricular complex information on a beat-by-beat basis during the ECG or electrogram recording.

In the case of a clinical procedure in a cardiac catheter laboratory, at intervals (for example as soon as each new ventricular complex occurs and is recognised) a revised "averaged ventricular complex" is subtracted from the displayed signal. In such embodiments, for the first few heartbeats of a procedure, the quality of the subtraction is poor, but with every passing beat, the signal quality improves.

In some particular embodiments, this feature is retroactive, in that later ventricular complexes are involved in calculation of the average ventricular complex, that can in turn improve the quality of subtraction of the earlier ventricular complexes (if the user scrolls back to earlier time points). This is relevant because the overall duration of the ESI (or other) recording may be very long (a large number of minutes or some hours) and so in the later part of the study the subtraction may be extraordinarily effective.

While the above described embodiments have been described in relation to an ESI balloon system of recordal it is to be understood that the present invention is not limited thereto. In other embodiments, the ventricular complex subtraction methods of the invention are applied to other forms of ECG analysis and, in particular, can be applied to all cardiac electrical mapping systems

In one embodiment a standard 12 lead ECG machine is modified to generate a "de-QRST'd" ECG for a patient. Such an ECG allows a supervising electrophyiologist to comment in greater depth on the nature of atrial (and other non ventricular) activity of the patient.

In further embodiments, the methods of the present invention are applied to remove other recurrent noise which comprise stereotyped, recurrent added voltage phenomena, even if they were not from the ventricle, and even if they were not regular in the interval between the episodes. Examples of such noise phenomena include some forms of tremor, including shivering, and some forms of noise from external machinery.

In another embodiment, "Body surf mapping", which is a form of ECG recording where a large number of electrodes (of the order of 100 or so) are placed in a large grid on the chest of a patient is modified with the ventricular complex subtraction methodology of the present invention.

The methods of the present invention are applicable to any and all cardiac mapping systems and techniques, including electroanatomical/electrospatial mapping systems. The latter provide a log in 3-dimensional space of each electrogram recording, and therefore allow an anatomical reconstruction of the part of the heart being mapped and the electrical signals at each point thereon. Most of these other mapping systems acquire data on a point-by- point basis, rendering them of limited use for mapping some aspects of the chaos of atrial fibrillation. Some such systems acquire bipolar signals which may not require or benefit less from subtraction of ventricular components Nevertheless, embodiments in which such other mapping systems are used are capable of generating "dominant frequency maps" on a point-by-point basis. The most widely known and used mapping systems are the Carto system (Biosense Webster), also Navex (ESI), Realtime Position Management (RPM). In one embodiment, there is a method of recording a unipolar electrogram of a patient comprising the steps of:

(i) determining the intrinsic ventricular rate of the patient; (ii) pacing the ventricle of the patient at a rate higher than the intrinsic ventricular rate of the patient;

(iii) recording the unipolar electrogram of the patient advantageously discarding any recordings which include capture or fusion beats; (iv) subtracting at least a part of the ventricular component of the patient from the unipolar electrogram; and (v) performing a Fourier transformation on the resultant signal in order to determine the dominant frequency.

This embodiment is illustrated in Figure 6. Referring to Figure 6a, the raw unipolar signal is shown during pacing in the right ventricle of the patient at 60 beats per minute. In Figure 6b, the power frequency spectrum following fast Fourier transformation of the raw signal (i.e. without subtraction of the ventricular component step (iv)) is shown. A spike is present at 1 Hz (which equates to 60 beats per minute) and at the harmonics of 1 Hz e.g. 2Hz, 3Hz, 4Hz etc. The dominant frequency is 2Hz.

The signal following subtraction of ventricular components from the raw signal is shown in Figure 6c. Since high quality subtraction has been carried out, there is no sign of ventricular components remaining, apart from the pacing spikes which are of such low duration that they have no effect on the power frequency spectrum.

The power frequency spectrum following fast Fourier transformation of the resultant signal (i.e. following high quality subtraction of ventricular components) is shown in Figure 6d. The spikes at 1 Hz and its harmonics are attenuated and now 5.4Hz is identified as the dominant frequency. By way of comparison, Figure 6e shows the signal following poor subtraction of the ventricular components from the raw signal shown in Figure 6a. Approximately 50ms at the end of the ventricular repolarization wave has not been subtracted as the template's duration is too short. This has left a trace of the ventricular components in the resultant signal although this is not obvious to the naked eye. Figure 6f the power frequency spectrum following fast Fourier transformation of the signal shown in Figure 6e. Despite the presence of some remaining ventricular components not being obvious to the naked eye, the power frequency spectrum clearly has a series of spikes at 1 Hz and its harmonics. The result is that the dominant frequency is identified as 1 Hz (which would clearly not be the correct value for the pure atrial signal). This demonstrates the value of effective subtraction of the ventricular components from the raw signal.

Referring to Figure 7a, there is shown a the results of determining the dominant frequency of a plurality of portions of a cardiac area (in this instance the left atrium) in order to generate a left atrial dominant frequency map. In Figure 7a, this is viewed from a superior aspect with left atrial appendage seen to the right of the figure. The highest frequency portions are at the base of the posterior wall and near right superior pulmonary vein.

Referring to Figure 7b, a left atrial dominant frequency map is shown which was created using the 7 seconds of atrial fibrillation data immediately following the 7 seconds used to create the map shown in Figure 7a. An identical view is displayed. The highest frequency area is now located on the left anterior wall close to the left atrial appendage. This demonstrates the importance of calculating the dominant frequencies of a cardiac area in relation to periods of time of greater than 7 seconds. EXAMPLES

EXAMPLE 1

INTRODUCTION

Spectral analysis is a powerful signal processing algorithm which can identify regular contributions from complex electrical, irregular signals. These contributions are displayed as dominant frequencies and the frequency with the highest power has been shown to correspond to the oscillating cycle lengths of underlying rotors (Mansour M, Mandapati R, Berenfeld O, Chen J, Sarnie FH, Jalife J. Left-to-right gradient of atrial frequencies during acute atrial fibrillation in the isolated sheep heart. Circulation 2001 May 29;103(21 ):2631-6). The use of the Ensite (Endocardial Solutions Inc, USA) has previously described as a unique tool for simultaneous, high-resolution global activation mapping (Schilling RJ, Peters NS, Davies DW. Feasibility of a noncontact catheter for endocardial mapping of human ventricular tachycardia. Circulation 1999 May 18;99(19):2543-52 and Schilling RJ, Kadish AH, Peters NS, Goldberger J, Davies DW. Endocardial mapping of atrial fibrillation in the human right atrium using a non-contact catheter. Eur Heart J 2000 April;21(7):550-64.). The current study combines the respective aptitudes of these two techniques to demonstrate and to characterise left atrial dominant frequency distribution in patients with paroxysmal and persistent AF.

METHODS Patients

Twenty-four consecutive patients who had a history of highly symptomatic AF and episodes of spontaneous AF recorded by non-contact mapping balloon catheter, deployed in their left atria, were included in this study. The mean age of the patients was 53 ± 4 years and 11 patients were male. Eleven patients had paroxysmal AF and 13 patients had persistent AF. Arrhythmia related symptoms had failed to be controlled by at least 1 antiaarhythmic drug in all patients.

All antiarrhythmic drugs were stopped at least 5 half-lives prior to the procedure and none of the patients was on amiodarone. The presence of left atrial appendage endocardial thrombus was excluded by trans-oesophageal echocardiography performed immediately prior to the planned procedure. The study protocol was approved by the local ethics committee and all procedures were carried out after obtaining written informed consent.

Non-contact mapping A non-contact multi-electrode array (EnSite 3000; Endocardial Solutions Inc) and a conventional mapping catheter (Biosense-Webster, Diamond Bar, CA) were deployed trans-septally into the LA. The details of the non-contact system have been described previously (Chow AWC, Schilling RJ, Davies DW, Peters NS. Characteristics of Wavefront Propagation in Reentrant Circuits Causing Human Ventricular Tachycardia. Circulation 2002 May 7;105(18):2172-8; Chow AWC, SEGAL OR, Davies DW, Peters NS. Mechanism of Pacing-Induced Ventricular Fibrillation in the lnfarcted Human Heart. Circulation 2004 September 28;110(13):1725-30; Markides V, Schilling RJ, Ho SY, Chow AW, Davies DW, Peters NS. Characterization of left atrial activation in the intact human heart. Circulation 2003 February 11 ;107(5):733- 9; and Schilling RJ, Peters NS, Davies DW. Simultaneous Endocardial Mapping in the Human Left Ventricle Using a Noncontact Catheter : Comparison of Contact and Reconstructed Electrograms During Sinus Rhythm. Circulation 1998 September 1 ;98(9):887-98). Afterwards patients were anticoagulated with heparin to achieve and to maintain the activated clotting time above 300 seconds. A detailed LA geometry was acquired using the roving mapping catheter and the anatomical landmarks including mitral valve annulus, pulmonary vein ostia, left atrial appendage, and roof, septal, anterior and posterior left atrial wall, were identified and labeled. All spontaneous paroxysmal AF episodes and periods of persistent AF were recorded using the memory buffer of the non-contact system and were subsequently analysed. The filter setting of the non-contact electrograms was 1 to 150 Hz.

Signal processing and spectral analysis of non-contact electrograms

Non-contact electrograms were recorded from the 3360 points of the LA at 1.2kHz. Amongst which electrograms were sampled from 256 evenly distributed LA points and were analysed in 6.8-second-long segments. Offline analysis was performed using a customised software programmed in the Matlab (Mathworks, Natick, MA) environment. The software was specifically designed to process the non-contact LA electrograms thorough three main steps 1 ) to subject the unipolar raw electrograms to a ventricular signal subtraction algorithm, and 2) to filter the subtracted electrograms using a Hanning window, 3) to analyse the processed electrograms using fast Fourier transform algorithm. The details of each step are described below.

Subtraction of the ventricular components

The far field ventricular components of the unipolar non-contact electrograms from each of the 256 sites were subtracted using a four-stage stepwise semi- automated subtraction algorithm as illustrated in Figure 1. First, the timing of the peak of far field ventricular depolarisation was identified automatically from a reference non-contact electrogam (the electrogram from the LA site with the clearest ventricular depolarisation defined by the largest skew) as shown in Figure 1A. Second, pivoted upon the peak ventricular depolarisation, all the ventricular components from the segment at that site were over-laid (Figure 1B) and a mean was obtained (Figure 1C &1D). Third, the mean ventricular complex was used as a template for the individual site for the subtraction of the ventricular component of the electrogram at that site (Figure 1 E &1 F). Fourth, this process was repeated automatically on the electrograms from all 256 individual LA sites with the timing of the ventricular component referenced from the reference electrogram in each case. The disparity in DF value before and after ventricular subtraction is illustrated Figure 1G.

Filtering and spectral analysis

A Hanning window was applied to the ventricular subtracted atrial fibrillatory electrograms to minimise spectral leakage effects and to improve the sharpness of the spectral peak. 8192 filtered data points (6.8s at 1.2k Hz) from each of the 256 left atrial sites were subjected to fast Fourier transform algorithm. The power frequency spectra after spectral analysis between 3- 30Hz (physiologically relevant to the fibrillatory activity in the human atrium) were analysed. The frequency with the greatest power was taken as the dominant frequency (DF). The dominant frequency from each site was then displayed in a colour map on a three-dimensional left atrial rendition in the Matlab environment.

Validation of the Non-Contact Electrograms and the Frequency Spectra

Non-contact mapping of the right atrium during AF using the Ensite system has previously been validated in the human (Lin YJ, Tai CT, Huang JL et al.

Characterization of Right Atrial Substrate in Patients with Supraventricular

Tachyarrhythmias. Journal of Cardiovascular Electrophysiology

2005;16(2):173-80; and Lin YJ, Tai CT, Kao T et al. Electrophysiological

Characteristics and Catheter Ablation in Patients With Paroxysmal Right Atrial Fibrillation. Circulation 2005 September 20;112(12):1692-700.). Further validation of non-contact electrograms during AF in the left atrium was performed by comparing to the contact electrograms from 62 evenly distributed left atrial locations in 4 patients. Unipolar signals were recorded by a 4mm-tip ablation catheter (Biosense and Webster, Diamond Bar, CA, USA) and were directly compared to the simultaneously acquired non-contact electrograms. Additionally, the dominant frequencies from each site from the contact and non-contact electrograms were also compared. This demonstrates the efficacy of the methodology in contract-type data collection and analysis as well.

Definitions

Atrial fibrillation was defined as an irregular tachycardia with beat-to-beat change in contact intracardiac atrial electrogram timing and morphology (right atrium and coronary sinus) and with an irregular ventricular response.The dominant frequency (DF) of spectral analysis from each site was defined as the frequency with the highest power. A focal area of high DF (DFmaχ) was defined as an area of the highest DF in a segment that was at least 20% larger than the neighbouring points (Sanders P, Berenfeld O, Hocini M et al. Spectral analysis identifies sites of high-frequency activity maintaining atrial fibrillation in humans. Circulation 2005 August 9;112(6):789-97.). This level of 20 % was arbitrarily set to define the presence of a dominant frequency gradient in keeping with previous investigations.

Statistical Analysis Continuous data are expressed as mean ± standard deviation when the data are normally distributed or as range and median values otherwise. Comparison between groups was performed with the unpaired Student's t- test. Categorical variables were compared using the Fisher's exact test. Correlation between contact and non-contact dominant frequency values at different sites, as well as morphology correlation between contact and non- contact electrograms, was determined using Pearson's correlation coefficients. Two-sided P values <0.05 were considered significant.

RESULTS

From the 24 patients who entered the study, the clinical characteristics of the two subgroups (11 patients who had paroxysmal and 13 persistent AF) are summarised in table 1. The LA dimension was larger in the persistent compared to paroxysmal AF subgroup. Two of the thirteen patients who had persistent AF also had a concomitant cardiac disease (left ventricular hypertrophy and impairment of left ventricular function respectively).

A total of 83 episodes (229 segments, 2.9 ± 3.0 segments per episode) of spontaneous onset of AF recorded by the non-contact system (range 2-15 and median 6 episodes per patient) were analysed in patients with paroxysmal AF. This was compared to 193 randomly selected segments (range 5-43 and median 9 segments per patient) from 13 patients who had persistent AF.

Table 1. Clinical characteristics

Figure imgf000034_0001

Validation of non-contact data

The mean correlation coefficient between the contact and non-contact electrograms over 6.8 seconds for all 62 sites was 0.85 +0.11 (range 0.51 to 0.99). There was no difference in mean dominant frequency from all 62 sites between the non-contact (and contact electrograms (6.9 ±1.5 Hz versus 7.0 +1.7 Hz). The DF at the contact and non-contact sites were significantly correlated (r=0.88, p<0.0001). Validation of non-contact signals is illustrated in figure 2.

Focal area of high dominant frequency in paroxysmal AF

Focal DFmax were frequently observed in the segments from patients with paroxysmal AF, a typical example of which is shown in Figure 3A. From the location in x, y and z axis of the 256 sites of the LA, a three-dimensional rendition of the LA was constructed and after the subtraction of ventricular component and fast Fourier transform, the DF from each point was displayed on the LA geometry in greyscale. The greyscale spectrum is illustrated on the right of the Figure 3, from 4.8 Hz to >9 Hz. In this map there was a discrete DFmax located on the posterior wall of the LA. The non-contact electrograms from focal DFmax typically displayed more rapid and fibrillatory activation compared to that from areas with lower DF.

Focal DFmax (11.6 ± 2.9 Hz) were observed in 149 /229 (65%) segments of the 64/83 (77%) episodes in all 11 patients who had spontaneous onset of paroxysmal AF.

Location of focal high dominant frequency area in paroxysmal AF

In total 192 focal DFmax were found in 149 segments (1.3 ± 0.6 foci per segment). The location of the focal DFmax was not confined to the pulmonary veins and posterior LA wall. They were found to be near the pulmonary veins in 33%, other parts of the LA posterior wall in 10%, anterior LA in 19%, left atrial appendage in 11%, lateral LA wall in 9%, septum in 9% and roof of LA in 8% (Figure 4A). The magnitude of the focal DFmax did not differ between locations (Figure 4B).

Temporal and spatial stability of focal high DF area in paroxysmal AF

Figure 5 is a typical example of a dominant frequency greyscale map showing both temporal and spatial stability of the DFmax in both magnitude and location through-out 4 consecutive segments in a patient who had paroxysmal AF. In this example the DFmax (12.5 Hz) was located principally at the anterior wall of the LA, although there was some variation in the area of DFmax.

In those episodes where there was more than one segment containing focal DFmax, in 88% (15/17 episodes) the location of the focal DFmax was at the same location although the focal DFmax may not be present in all segments from an episode. However the location of the DFmax varies between different episodes of the same patient.

Distribution of dominant frequency in persistent/permanent AF

Figure 3B demonstrates a typical DF greyscale map of a persistent AF segment. There is an absence of focal DFmaχ compared to the map from the paroxysmal AF segment.

Focal DFmax was rarely seen in persistent AF compared to during paroxysmal episodes of AF (9/193 [5%] versus 149/229 [65%] segment, p<0.0001 ) Figure 7. A focal DFmax was observed in only 4/13 [31 %] of patient who had persistent AF compared to 11/11 [100%] who had paroxysmal AF, p<0.001. Also, although in persistent AF segments the maximum dominant frequency was lower than that from paroxysmal AF (7.2 ± 1.4 versus 9.6 ± 3.6 Hz, p<0.0001 ), the overall mean dominant frequency of the LA did not differ between persistent and paroxysmal AF (5.7 ± 0.9 versus 5.7 ± 0.9 Hz, p=0.77). This reflects the fact that the magnitude of dominant frequencies of the LA during persistent AF are more homogenous compared to those during paroxysmal AF.

Conclusions The current study demonstrates focal areas of high dominant frequency (DFmax ) in the left atria during ongoing fibrillation, which is in keeping with focal driver activity. These focal drivers were not exclusively confined to the posterior wall and were far less frequently detected during persistent AF. Uniquely, the techniques described in this study combine the advantages afforded by Fourier analysis and global simultaneous high density mapping in the conscious patients to analyse spontaneously occurring AF.

Localised areas with high frequency activity can be identified by non-contact spectral analysis of the left atria. These areas, consistent with rotors described in animal models, are not exclusively located in the posterior left atria and are readily detected during paroxysmal AF. In contrast, they were rarely observed during persistent AF. These findings underline significant mechanistic differences between paroxysmal and persistent AF which, in turn have important implications in choosing the most appropriate ablation strategies for patients with different clinical patterns of AF.

EXAMPLE 2

Introduction

Long-term procedural success of wide area circumferential ablation (WACA) cannot be reliably predicted. Observations from spectral analysis of noncontact mapping signals during atrial fibrillation (AF) have not previously been correlated with clinical outcomes. The relationship between global left atrial (LA) dominant frequencies (DFs) in chronic AF and clinical outcomes following WACA were investigated.

Methods

Patients with chronic AF underwent WACA whereby pairs of unilateral pulmonary veins were encircled and mitral valve and roof lines added. A noncontact multielectrode array was deployed in the LA and used to record AF before and after WACA. Non-contact electrograms from 256 evenly distributed LA sites were exported for analysis. Far field ventricular components were subtracted from the raw signals using a novel technique and the outputs subjected to fast Fourier transform. The highest power frequency in the 3 to 30Hz range was selected as the DF at each site. Twenty eight seconds of continuous AF was analysed before and after WACA in each patient. Successful clinical outcome was defined as freedom from arrhythmic symptoms and sinus rhythm on 24-hour ECG monitoring at 3 months after the procedure.

Results Ten patients completed the protocol, of whom 4 had a successful clinical outcome after a single ablative procedure. LA mean DF was significantly lower in the successful group compared to the unsuccessful group, as measured both before (5.6±0.1 vs 6.2±0.2 Hz (mean+SE); P<0.05) and after (4.8±0.2 vs. 5.6±0.2 Hz; P<0.05) WACA. The reduction of LA mean DF from WACA did not predict outcome (P=O.65). Other clinical and anatomical parameters including patient's age, duration of AF history, LA dimensions and volume, left ventricular ejection fraction, and presence of structural heart disease or associated cardiovascular disease did not predict outcome.

Conclusions

Spectral analysis of left atrial noncontact mapping signals during chronic AF predicted the clinical outcome of WACA. Left atrial mean DF appears to be a more accurate predictor of procedural success than previously established predictors in this preliminary series.

EXAMPLE 3

This study demonstrates a link between organisational index and temporal stability. The study is relevant to the methods of the invention of generating a map of a cardiac area of a patient indicating portions of the cardiac area responsible for atrial fibrillation. The combined maps were prepared which were a combination of dominant frequency, organisational index and temporal stability. Such maps permitted the targeting of areas with high organisational index and also high temporal stability. This work showed that these two characteristics are linked in the same areas. On this basis, subsequent live ablation cases have been carried out using maps of organisational index for targeting rather than combined maps as it is now known from this work that organisational index is also a proxy for temporal stability. When doing this, areas of high dominant frequency are preferentially targeted when choosing between different areas of high organisational index. In effect, this allows ablation on a combined map. An exemplary organisational index map is shown in Figure 8 together with the electrogram (following subtraction of its far field ventricular components) and corresponding power frequency spectrum at two locations.

This work also shows the use of the techniques described herein whereby the far field ventricular signal is subtracted from the unipolar signal obtained from the noncontact mapping system.

Purpose: Localised drivers have been implicated in the mechanism of chronic atrial fibrillation (AF). By spectral analysis, regions of highest dominant frequency (DF) in the left atrium (LA) have been said to identify the locations of such drivers, however high DF may be expected at bystander sites where wavefronts collide. We hypothesised that if drivers of AF were present their principal spectral characteristics, in keeping with other focal and re-entrant tachycardias, would instead be (a) organisation (I.e. a narrow frequency range), and (b) temporal stability of the DF (i.e. DF magnitude remaining relatively constant with time). Methods: In patients undergoing ablation for chronic AF, a noncontact mapping array was deployed in the LA to record AF before ablation. Noncontact electrograms at 256 evenly distributed LA sites were subjected to fast Fourier transform following subtraction of far field ventricular components. The frequency spectra of 5 sequential 7-second segments of AF were analysed. The highest power frequency in the 3 to 15Hz range was selected as the DF at each site. Mean absolute difference in DF between successive segments was defined as the DF variability (DFV) at each site. The ratio of the DF and its harmonics to the total power of the spectrum was calculated for each segment and the mean value defined as the organisational index (Ol) at that site.

Results:

Ten patients completed the protocol. Organised areas were defined as having Ol more than one standard deviation above the mean. Mean Ol for all sites in all patients was 0.41 ±0.02 (mean±SE) and in organised areas was 0.51 ±0.02. Mean DFV was significantly lower in organised areas than in all sites (0.34±0.04 vs 0.46±0.04 Hz; P<0.001 ). Mean DF was only modestly higher in organised areas than in all sites (6.31 ±0.18 vs 6.21 ±0.17 Hz; P<0.01 ). Organised areas were most commonly located at the pulmonary vein orifices (12 of 27 sites; 44%).

Conclusions:

Simultaneous spectral mapping throughout the LA during chronic AF led to a unique observation. At sites where activation pattern was more organised during all time segments, frequency of activation was significantly more stable over time (between time segments), a finding that would not be expected with a purely random activation pattern. This observation is consistent with our hypothesis regarding the expected spectral characteristics of "driver" regions, and inconsistent with a purely random activation pattern.

EXAMPLE 4

This study relates to determining the extent of ablation required when treating atrial fibrillation.

The study is of particular relevance to the methods of the present invention of assessing the requirement for ablation therapy on a patient suffering from atrial fibrillation.

Introduction

Many complications of left atrial (LA) ablation of chronic atrial fibrillation (AF) are related to radiofrequency energy delivery. However, there are no end- points to guide the minimal extent of ablation required to achieve long-term success. We investigated the relationship between mean dominant frequency (DF) in the LA and clinical outcomes following wide area circumferential ablation (WACA).

Methods

Patients with chronic AF underwent WACA, anatomically-guided by LA deployment of the Ensite noncontact system, lpsilateral pulmonary vein pairs were encircled and mitral isthmus and roof lines added. Noncontact electrograms at 256 LA sites were subjected to fast Fourier transform following subtraction of their far field ventricular components. DF was identified at each site from 28 seconds of continuous AF recorded before and after WACA in each patient. Successful outcome was defined as freedom from arrhythmic symptoms and sinus rhythm on Holter ECG monitoring 3 months post-ablation. Results

AF persisted in all 10 patients on completion of WACA, and 4 had a successful outcome after a single procedure. Mean DF of all sites was significantly lower in the successful group compared to the unsuccessful group, as measured both before (5.6+0.1 vs 6.2+0.2 Hz (mean±SE); P<0.05) and after (4.8±0.2 vs. 5.6±0.2 Hz; P<0.05) WACA. On univariate analysis, other parameters including patient's age, duration of AF history, LA dimensions, left ventricular ejection fraction, and presence of structural heart disease or other cardiovascular disease did not predict outcome.

Conclusions

LA mean DF was the only parameter able to predict the outcome of WACA in this small series. The ability to predict a successful outcome from linear ablation in an individual patient, without the need for further supplementary ablation such as targeting high frequency activity, provides a procedural end- point and limits unnecessary additional ablation and its risks of complications.

EXAMPLE 5

This study relates to the use of the methods of the present invention for generating a map of a cardiac area in order to target ablation therapy

We used the techniques described herein to target ablation therapy in a patient with long-lasting persistent atrial fibrillation. The patient had been in persistent atrial fibrillation for 7 years and failed therapy with 3 antiarrhythmic drugs, including amiodarone. He also had a dilated left atrium and moderately severe left ventricular systolic impairment. For all of these reasons he would be considered a difficult patient to treat with ablation therapy and the likelihood of success with a single procedure would be low. He had not had any previous procedures.

We deployed a noncontact mapping system and then recorded atrial fibrillation and exported these recordings in overlapping sequential 7 second segments at 256 points on the surface of the left atrium and joined them into single 30 second segments (or "periods") in accordance with the method of the present invention of generating a dominant frequency map of a cardiac area of a patient. We then removed the far field ventricular component of these unipolar noncontact mapping segments by (i) selecting a timing reference signal; (ii) detecting a repeating physiological signal in the electrogram; and (iii) subtracting a representation of the repeating physiological signal from the electrogram using the timing reference signal to calculate the timing of each physiological signal. We also filtered the electrogram for signals corresponding to the frequency of mains electricity in order to remove "mains hum". In addition, we applied the process automatically to the signals from all 256 points as described herein. We then generated organisational index maps by using the methods of the invention of generating a map of a cardiac area of a patient indicating portions of the cardiac area responsible for atrial fibrillation. These maps were then used to target ablation therapy without the use of any conventional techniques.

As a result of these measures, the patient was found to be in sinus rhythm, without evidence of atrial fibrillation, when examined approximately 4 months following his procedure. This indicated that the procedure had successfully abolished his atrial fibrillation despite it having been resistant to any previous therapy in the past. We find these results very encouraging and suggestive that the techniques developed are effective as treatments for atrial fibrillation.

Claims

1. A method of generating a dominant frequency map of a cardiac area in a patient comprising the steps of: (i) processing at least one electrogram from the patient by Fourier transformation to generate a power frequency spectrum, the at least one electrogram corresponding to a portion of the cardiac area; and
(ii) selecting the highest value from the or each power frequency spectrum as the dominant frequency at the corresponding portion of the cardiac area, wherein the or each dominant frequency is determined in respect of a period of at least 7 seconds.
2. A method according to claim 1 wherein the period is at least 10, 15, 20, 25 or 30 seconds.
3. A method according to claim 1 or 2 further comprising the step of, prior to step (i), linking together a plurality of electrograms for the or each portion, each electrogram being shorter than the period in order to produce an electrogram of at least the length of the period for the or each portion.
4. A method according to claim 1 or 2 wherein step (i) comprises, for the or each portion, processing a plurality of electrograms, each being in relation to a period of less than the period and then averaging the processed signals so as to relate to at least the length of the period.
5. A method according to any one of the preceding claims wherein the cardiac area is at least the left atrium of the patient.
6. A method according to any one of the preceding claims further comprising the step of, prior to step (i), recording the or each electrogram.
7. A method according to claim 6 wherein the step of recording the or each electrogram is carried out using a contactless system.
8. A method according to any one of the preceding claims wherein step (i) comprises processing a plurality of electrograms, each electrogram corresponding to a portion of the cardiac area measured substantially simultaneously.
9. A method of determining the dominant frequency of an electrogram from a patient comprising the steps of:
(i) selecting a timing reference signal on the basis of a skew of the distribution of voltages of the electrogram;
(ii) detecting a repeating physiological signal in the electrogram; (iii) subtracting a representation of the repeating physiological signal from the electrogram using the timing reference signal to calculate the timing of each physiological signal;
(iv) processing the resultant signal using Fourier transformation to generate a power frequency spectrum; and (v) selecting the highest value from the power frequency spectrum as the dominant frequency.
10. A method according to claim 9 wherein the physiological signal is a ventricular depolarization and repolarization wave.
11. A method according to claim 9 wherein the physiological signal is a respiratory wave.
12. A method according to any one of claims 9 to 11 further comprising the step of filtering the electrogram for signals corresponding to the frequency of mains electricity, preferably at 50Hz or 60Hz.
13. A method according to any one of claims 9 to 12 further comprising the step of, prior to step (i), filtering the electrogram through a bandwidth filter, preferably in the range of 1 to 150Hz.
14. A method according to any one of claims 9 to 13 wherein step (iii) comprises the step of averaging each physiological signal to generate a template and subtracting the template from the electrogram at each incidence of the physiological signal.
15. A method according to claim 14 wherein the step of averaging each physiological signal comprises overlaying each signal from 100ms before to
800ms after the timing reference signal.
16. A method according to claim 14 or 15 wherein the step of averaging each physiological signal comprises carrying out a weighted average.
17. A method according to claim 16 wherein the physiological signal is a ventricular depolarization and repolarization wave and the electrogram has been recorded by a monitoring device and wherein the weighted average is calculated with reference to: the phase of the respiratory cycle of the patient; the interval between a current ventricular complex and a previous ventricular complex; the interval between two preceding ventricular complexes; the location and orientation of the monitoring device; the drift in position of the monitoring device; and combinations thereof.
18. A method according to any one of claims 9 to 17 wherein the method is carried out during recording of the electrogram.
19. A method according to claim 18 wherein at least step (iii) is carried out continually.
20. A method according to any one of claims 9 to 19 wherein the method is carried out on a plurality of electrograms in relation to different portions of a cardiac area of the patient, preferably wherein each electrogram is obtained substantially simultaneously.
21. A method according to claim 20 wherein step (i) comprises the step of, generating a series of timing reference points from the timing reference signal, by selecting one electrogram to provide each timing reference point, said timing reference point being applied to every electrogram in step (iii).
22. A method of generating a map of dominant frequencies of a cardiac area of a patient comprising carrying out the method of any one of claims 9 to 21 and plotting the frequencies at each portion where the dominant frequency has been determined.
23. A method of generating a map of a cardiac area of a patient indicating portions of the cardiac area responsible for atrial fibrillation comprising the steps of:
(i) processing at least one electrogram from the patient by Fourier transformation to generate a power frequency spectrum, the or each electrogram corresponding to a portion of the cardiac area;
(ii) selecting the highest value from the or each power frequency spectrum as the dominant frequency of the corresponding portion of the cardiac area, the dominant frequency thereby having a corresponding dominant frequency peak in the power frequency spectrum;
(iii) determining an organisational index at the or each portion by calculating the area beneath the dominant frequency peak and its harmonic peaks as a fraction of the total area under the power frequency spectrum; and
(iv) determining a temporal stability index at the or each portion by determining the dominant frequency at the or each portion over a plurality of successive time periods and calculating the standard deviation or average change of the dominant frequency over the successive time periods.
24. A method according to claim 23 wherein, in step (iv), each time period is between three and sixty seconds long, preferably 6.82 seconds long.
25. A method according to claim 23 or 24 further comprising the step of plotting the value from step (H), the organisational index from step (iii) and the temporal stability index from step (iv) at each respective location of the cardiac area.
26. A method according to any one of claims 23 to 25 wherein a portion of the cardiac area responsible for atrial fibrillation is indicated by a higher relative value of dominant frequency, organisational index and/or temporal stability index.
27. A method according to any one of claims 22 to 26 wherein step (i) comprises processing a plurality of electrograms from the patient.
28. A method of assessing the requirement for ablation therapy on a patient suffering from atrial fibrillation comprising the steps of:
(i) determining the dominant frequency at at least one portion of the cardiac area of the patient; and (ii) comparing the dominant frequency with a reference value.
29. A method according to claim 28 wherein step (i) comprises determining the dominant frequency at a plurality of portions of the cardiac area of the patient.
30. A method according to claim 29 wherein step (i) further comprises a step of determining the mean dominant frequency of all portions and step (ii) comprises comparing the mean dominant frequency with a reference value.
31. A method according to claim 30 wherein the reference value in step (ii) is 5Hz, a mean dominant frequency of greater than 5Hz being indicative of the need for ablation therapy.
32. A method according to claim 28 or 29 wherein step (i) comprises determining the dominant frequency at a portion adjacent to the pulmonary vein; adjacent to the septum; adjacent to the left atrial appendage; within the coronary sinus or combinations thereof.
33. A method according to any one of claims 28 to 32 wherein step (i) comprises determining the dominant frequency in the left atrium.
34. A method according to claim 29 wherein step (i) comprises determining the dominant frequency of the surface electrocardiogram of the patient.
35. A method according to claim 34 wherein the surface ECG signal is obtained from the V1 lead.
36. A method of ablating portions of a cardiac area of a patient suffering from atrial fibrillation comprising the steps of: (i) identifying a plurality of portions of the cardiac area of the patient having a dominant frequency above a threshold value in a electrogram; and
(H) carrying out a linear ablation in the cardiac area along a line joining the portions identified in step (i).
37. A method according to claim 36 wherein step (ii) further comprises a pulmonary vein isolation.
38. A method according to claim 36 wherein step (ii) further comprises an encircling linear ablation.
39. A method according to claim 36 wherein step (ii) further comprises a non-encircling linear ablation.
40. A method according to any one of claims 36 to 39 wherein ablation is carried on in the left atrial appendage and/or coronary sinus area of the patient.
41. A method of recording a electrogram of a patient comprising the steps of:
(i) determining the intrinsic ventricular rate of the patient; (ii) pacing the ventricle of the patient at a rate higher than the intrinsic ventricular rate of the patient; and (iii) recording the electrogram of the patient.
42. A method according to claim 41 further comprising the step of:
(iv) subtracting at least a part of a physiological signal of the patient from the electrogram, preferably at least a part of the ventricular component.
43. A method according to claim 42 and further comprising the step of:
(v) confirming that the ventricular component has been subtracted from the electrogram of the patient up to a predetermined threshold.
44. A method according to any one of the preceding claims wherein the electrogram is a unipolar electrogram or a bipolar electrogram.
PCT/GB2007/003558 2006-09-20 2007-09-20 Atrial fibrillation analysis WO2008035070A2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
GB0618522.7 2006-09-20
GBGB0618522.7A GB0618522D0 (en) 2006-09-20 2006-09-20 Atrial fibrillation analysis

Publications (2)

Publication Number Publication Date
WO2008035070A2 true WO2008035070A2 (en) 2008-03-27
WO2008035070A3 WO2008035070A3 (en) 2008-09-18

Family

ID=37421310

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2007/003558 WO2008035070A2 (en) 2006-09-20 2007-09-20 Atrial fibrillation analysis

Country Status (2)

Country Link
GB (1) GB0618522D0 (en)
WO (1) WO2008035070A2 (en)

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012172027A3 (en) * 2011-06-15 2013-04-25 MAX-PLANCK-Gesellschaft zur Förderung der Wissenschaften e.V. Apparatus for and method of terminating a high frequency arrhythmic electric state of a biological tissue
EP2705464A1 (en) * 2011-05-02 2014-03-12 Topera, Inc. System and method for targeting heart rhythm disorders using shaped ablation
WO2014121110A1 (en) * 2013-02-01 2014-08-07 Northwestern University Contribution of oxidative stress to af electrograms
EP2826418A1 (en) * 2013-07-19 2015-01-21 Biosense Webster (Israel), Ltd. Cardiac activity visualization with frequency discrimination
US8989860B2 (en) 2007-03-03 2015-03-24 Max-Planck-Gesellschaft Zur Foerderung Der Wissenschaften E.V. Multisite heart pacing with adjustable number of pacing sites for terminating high frequency cardiac arrhythmias
EP2856935A1 (en) * 2013-09-26 2015-04-08 Biotronik AG Medical device for evaluating a temperature signal
WO2015149153A1 (en) * 2014-03-31 2015-10-08 University Health Network System and method for focal source identification
WO2015171742A1 (en) * 2014-05-09 2015-11-12 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
US9375156B2 (en) 2008-10-09 2016-06-28 The Regents Of The University Of California System for analysis of complex rhythm disorders
US9408544B2 (en) 2014-05-09 2016-08-09 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
US9468387B2 (en) 2011-05-02 2016-10-18 The Regents Of The University Of California System and method for reconstructing cardiac activation information
CN106231998A (en) * 2014-04-01 2016-12-14 加利福尼亚大学董事会 System and method of identifying sources associated with biological rhythm disorders
US9649040B2 (en) 2014-10-03 2017-05-16 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
US9668666B2 (en) 2011-05-02 2017-06-06 The Regents Of The University Of California System and method for reconstructing cardiac activation information
US9717436B2 (en) 2010-04-08 2017-08-01 The Regents Of The University Of California Method and system for detection of biological rhythm disorders
US9724009B2 (en) 2011-12-09 2017-08-08 The Regents Of The University Of California System and method of identifying sources for biological rhythms
US10085659B2 (en) 2014-10-03 2018-10-02 Boston Scientific Scimed, Inc. Medical system for mapping cardiac tissue
US10085655B2 (en) 2013-03-15 2018-10-02 The Regents Of The University Of California System and method to define drivers of sources associated with biological rhythm disorders
US10136860B2 (en) 2008-05-13 2018-11-27 The Regents Of The University Of California System for detecting and treating heart instability
US10271786B2 (en) 2011-05-02 2019-04-30 The Regents Of The University Of California System and method for reconstructing cardiac activation information

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5772604A (en) * 1997-03-14 1998-06-30 Emory University Method, system and apparatus for determining prognosis in atrial fibrillation
WO1999023943A1 (en) * 1997-11-11 1999-05-20 Fachhochschule Offenburg Device for detecting and storing electrocardiogram signals
US20050240112A1 (en) * 2003-08-28 2005-10-27 Fang Dan O System and method for detecting and locating heart disease
WO2005115232A1 (en) * 2004-05-17 2005-12-08 C.R. Bard, Inc. High density atrial fibrillation cycle length (afcl) detection and mapping system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5772604A (en) * 1997-03-14 1998-06-30 Emory University Method, system and apparatus for determining prognosis in atrial fibrillation
WO1999023943A1 (en) * 1997-11-11 1999-05-20 Fachhochschule Offenburg Device for detecting and storing electrocardiogram signals
US20050240112A1 (en) * 2003-08-28 2005-10-27 Fang Dan O System and method for detecting and locating heart disease
WO2005115232A1 (en) * 2004-05-17 2005-12-08 C.R. Bard, Inc. High density atrial fibrillation cycle length (afcl) detection and mapping system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LANGLEY P ET AL: "Frequency analysis of atrial fibrillation" COMPUTERS IN CARDIOLOGY 2000 CAMBRIDGE, MA, USA 24-27 SEPT. 2000, PISCATAWAY, NJ, USA,IEEE, US, 24 September 2000 (2000-09-24), pages 65-68, XP010528497 ISBN: 0-7803-6557-7 *
PRASHANTHAN SANDERS, OMER BERENFELD, MÉLÈZE HOCINI, PIERRE JAÏS, RAVI VAIDYANATHAN, LI-FERN HSU, STÉPHANE GARRIGUE: "Spectral Analysis Identifies Sites of High-Frequency Activity Maintaining Atrial Fibrillation in Humans Spectral Analysis Identifies Sites of High-Frequency Activity Maintaining Atrial Fibrillation in Humans" CIRCULATION, vol. 112, 2005, pages 789-797, XP002477672 Retrieved from the Internet: URL:http://circ.ahajournals.org/cgi/reprint/112/6/789> [retrieved on 2008-04-22] *

Cited By (47)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9855439B2 (en) 2007-03-03 2018-01-02 Max-Planck-Gesellschaft Zur Foerderung Der Wissenschaften E.V. Multisite heart pacing with adjustable number of pacing sites for terminating high frequency cardiac arrhythmias
US8989860B2 (en) 2007-03-03 2015-03-24 Max-Planck-Gesellschaft Zur Foerderung Der Wissenschaften E.V. Multisite heart pacing with adjustable number of pacing sites for terminating high frequency cardiac arrhythmias
US10136860B2 (en) 2008-05-13 2018-11-27 The Regents Of The University Of California System for detecting and treating heart instability
US9375156B2 (en) 2008-10-09 2016-06-28 The Regents Of The University Of California System for analysis of complex rhythm disorders
US9380950B2 (en) 2008-10-09 2016-07-05 The Regents Of The University Of California Methods for detecting biological rhythm disorders
US10092196B2 (en) 2008-10-09 2018-10-09 The Regents Of The University Of California Method for analysis of complex rhythm disorders
US9955879B2 (en) 2008-10-09 2018-05-01 The Regents Of The University Of California System for analysis of complex rhythm disorders
US9717436B2 (en) 2010-04-08 2017-08-01 The Regents Of The University Of California Method and system for detection of biological rhythm disorders
US9468387B2 (en) 2011-05-02 2016-10-18 The Regents Of The University Of California System and method for reconstructing cardiac activation information
EP2705464A4 (en) * 2011-05-02 2014-12-17 Topera Inc System and method for targeting heart rhythm disorders using shaped ablation
US10271786B2 (en) 2011-05-02 2019-04-30 The Regents Of The University Of California System and method for reconstructing cardiac activation information
US9913615B2 (en) 2011-05-02 2018-03-13 The Regents Of The University Of California System and method for reconstructing cardiac activation information
EP3330894A1 (en) * 2011-05-02 2018-06-06 Topera, Inc. System and method for targeting heart rhythm disorders using shaped ablation
EP2705464A1 (en) * 2011-05-02 2014-03-12 Topera, Inc. System and method for targeting heart rhythm disorders using shaped ablation
US10149622B2 (en) 2011-05-02 2018-12-11 The Regents Of The University Of California System and method for reconstructing cardiac activation information
US9655535B2 (en) 2011-05-02 2017-05-23 The Regents Of The University Of California System and method for targeting heart rhythm disorders using shaped ablation
US9668666B2 (en) 2011-05-02 2017-06-06 The Regents Of The University Of California System and method for reconstructing cardiac activation information
US9233248B2 (en) 2011-06-15 2016-01-12 Max-Planck-Gesellschaft Zur Foerderung Der Wissenschaften E.V.) Apparatus for terminating a high frequency arrhythmic electric state of a heart
CN104623807A (en) * 2011-06-15 2015-05-20 马克思-普朗克科学促进协会 Apparatus and method for terminating a high frequency arrhythmic electric state of a biological tissue
EP2823857A1 (en) * 2011-06-15 2015-01-14 Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.V. Apparatus for terminating an atrial fibrillation of an atrium of a heart
US8886309B2 (en) 2011-06-15 2014-11-11 Max-Planck-Gesellschaft Zur Foerderung Der Wissenschaften E.V. Apparatus for and method of terminating a high frequency arrhythmic electric state of a biological tissue
CN103702715A (en) * 2011-06-15 2014-04-02 马克思-普朗克科学促进协会 Apparatus for and method of terminating a high frequency arrhythmic electric state of a biological tissue
WO2012172027A3 (en) * 2011-06-15 2013-04-25 MAX-PLANCK-Gesellschaft zur Förderung der Wissenschaften e.V. Apparatus for and method of terminating a high frequency arrhythmic electric state of a biological tissue
US10058262B2 (en) 2011-12-09 2018-08-28 The Regents Of The University Of California System and method of identifying sources for biological rhythms
US9724009B2 (en) 2011-12-09 2017-08-08 The Regents Of The University Of California System and method of identifying sources for biological rhythms
WO2014121110A1 (en) * 2013-02-01 2014-08-07 Northwestern University Contribution of oxidative stress to af electrograms
US9615758B2 (en) 2013-02-01 2017-04-11 Northwestern University Contribution of oxidative stress to AF electrograms
US9907479B2 (en) 2013-02-01 2018-03-06 Northwestern University Contribution of oxidative stress to AF electrograms
US10098560B2 (en) 2013-03-15 2018-10-16 The Regents Of The University Of California System and method to identify sources associated with biological rhythm disorders
US10271744B2 (en) 2013-03-15 2019-04-30 The Regents Of The University Of California System and method to identify sources associated with biological rhythm disorders
US10085655B2 (en) 2013-03-15 2018-10-02 The Regents Of The University Of California System and method to define drivers of sources associated with biological rhythm disorders
US9351657B2 (en) 2013-07-19 2016-05-31 Biosense Webster (Israel) Ltd. Cardiac activity visualization with frequency discrimination
EP2826418A1 (en) * 2013-07-19 2015-01-21 Biosense Webster (Israel), Ltd. Cardiac activity visualization with frequency discrimination
EP2856935A1 (en) * 2013-09-26 2015-04-08 Biotronik AG Medical device for evaluating a temperature signal
US9770292B2 (en) 2013-09-26 2017-09-26 Biotronik Ag Medical device for evaluating a temperature signal
WO2015149153A1 (en) * 2014-03-31 2015-10-08 University Health Network System and method for focal source identification
US10111598B2 (en) 2014-03-31 2018-10-30 University Health Network System and method for focal source identification
EP3125756A4 (en) * 2014-04-01 2017-12-13 The Regents of the University of California System and method of identifying sources associated with biological rhythm disorders
CN106231998A (en) * 2014-04-01 2016-12-14 加利福尼亚大学董事会 System and method of identifying sources associated with biological rhythm disorders
WO2015171742A1 (en) * 2014-05-09 2015-11-12 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
US9408544B2 (en) 2014-05-09 2016-08-09 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
US9795314B2 (en) 2014-05-09 2017-10-24 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
US9763625B2 (en) 2014-05-09 2017-09-19 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue
JP2017520286A (en) * 2014-05-09 2017-07-27 ボストン サイエンティフィック サイムド,インコーポレイテッドBoston Scientific Scimed,Inc. Medical devices to map the heart tissue
US10264985B2 (en) 2014-10-03 2019-04-23 Boston Scientific Scimed, Inc. Medical system for mapping cardiac tissue
US10085659B2 (en) 2014-10-03 2018-10-02 Boston Scientific Scimed, Inc. Medical system for mapping cardiac tissue
US9649040B2 (en) 2014-10-03 2017-05-16 Boston Scientific Scimed, Inc. Medical devices for mapping cardiac tissue

Also Published As

Publication number Publication date
GB0618522D0 (en) 2006-11-01
WO2008035070A3 (en) 2008-09-18

Similar Documents

Publication Publication Date Title
SAOUDI et al. Classification of atrial flutter and regular atrial tachycardia according to electrophysiologic mechanism and anatomic bases: a statement from a joint expert group from the Working Group of Arrhythmias of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology
EP2298165B1 (en) System for three-dimensional mapping of electrophysiology information
US5109862A (en) Method and apparatus for spectral analysis of electrocardiographic signals
US10335051B2 (en) Beat alignment and selection for cardiac mapping
US8229545B2 (en) System and method for mapping complex fractionated electrogram information
CA2617754C (en) Automated pace-mapping for identification of cardiac arrhythmic conductive pathways and foci
CN102917637B (en) For the detection, diagnosis, and treatment of circadian rhythm disorders methods, systems and devices
D O’Neill et al. The stepwise ablation approach for chronic atrial fibrillation—evidence for a cumulative effect
Jadidi et al. Ablation of persistent atrial fibrillation targeting low-voltage areas with selective activation characteristics
US6370412B1 (en) Method and apparatus for guiding ablative therapy of abnormal biological electrical excitation
Anter et al. High-resolution mapping of scar-related atrial arrhythmias using smaller electrodes with closer interelectrode spacing
US6236883B1 (en) Methods and systems for localizing reentrant circuits from electrogram features
Desjardins et al. Infarct architecture and characteristics on delayed enhanced magnetic resonance imaging and electroanatomic mapping in patients with postinfarction ventricular arrhythmia
US5311873A (en) Comparative analysis of body surface potential distribution during cardiac pacing
Delacretaz et al. Multiple atrial macro–re-entry circuits in adults with repaired congenital heart disease: entrainment mapping combined with three-dimensional electroanatomic mapping
Holm et al. Non-invasive assessment of the atrial cycle length during atrial fibrillation in man: introducing, validating and illustrating a new ECG method
CN103718191B (en) Systems and methods for targeted ablation shaped heart rhythm disorder
US6064906A (en) Method, system and apparatus for determining prognosis in atrial fibrillation
US6760620B2 (en) Non-invasive localization and treatment of focal atrial fibrillation
US9439573B2 (en) Method for analysis of complex rhythm disorders
Rodrigo et al. Body surface localization of left and right atrial high-frequency rotors in atrial fibrillation patients: a clinical-computational study
Lin et al. Spatiotemporal organization of the left atrial substrate after circumferential pulmonary vein isolation of atrial fibrillation
Guillem et al. Noninvasive localization of maximal frequency sites of atrial fibrillation by body surface potential mapping
US7123954B2 (en) Method for classifying and localizing heart arrhythmias
Benharash et al. Quantitative analysis of localized sources identified by focal impulse and rotor modulation mapping in atrial fibrillation

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 07804341

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07804341

Country of ref document: EP

Kind code of ref document: A2