Classification of ablation points according to catheter data
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- BIOSENSE WEBSTER (ISRAEL) LTD
- Filing Date
- 2025-11-18
- Publication Date
- 2026-05-29
Smart Images
Figure 2026089039000001_ABST
Abstract
Description
Technical Field
[0001] The present disclosure generally relates to cardiac ablation, and more specifically to systems and methods for generating ablation maps.
Background Art
[0002] An ablation map is typically a 3D rendering of at least a portion of an organ that can be overlaid with ablation tags and is used by physicians when performing ablation therapy and verifying its effectiveness. The 3D rendering may be an electroanatomical map of an anatomical surface such as the inner wall of a cardiac chamber. The ablation display on the map can assist a physician when evaluating the status of an ablation therapy or when identifying characteristics associated with an ablation site. For example, a physician can use the ablation display to determine whether a sufficient number of ablations have been performed in a desired area or to identify the location of gaps within an ablated tissue pathway that need to be closed with further ablations to eliminate arrhythmias. An appropriate display of the area where treatment has been applied is beneficial to enable easy identification of ablation characteristics and to generate the information required for treatment reports and / or research purposes.
[0003] A more complete understanding of the present disclosure will be obtained by reading the following detailed description of the embodiments of the disclosure in conjunction with the drawings.
Brief Description of the Drawings
[0004] [Figure 1] FIG. is a schematic illustrative diagram of a catheter-based electroanatomical (EA) mapping and ablation system according to an embodiment of the present disclosure. [Figure 2] FIG. is a schematic illustrative diagram of the classification of ablation points based on catheter data according to an embodiment of the present disclosure. [Figure 3]This is an ablation map of cardiac chambers, induced using one of the disclosed algorithms for classifying ablation points, according to an embodiment of the present disclosure. [Figure 4] This flowchart schematically illustrates a method for generating an ablation map of cardiac chambers by classifying ablation points using catheter data, according to an embodiment of the present disclosure. [Modes for carrying out the invention]
[0005] overview In cardiac ablation procedures such as pulmonary vein isolation (PVI) to treat atrial fibrillation, physicians can ablate tissue within a specific anatomical region (e.g., around the entire circumference of the PV orifice). Ablations using pulsed-field ablation (PFA) or radiofrequency (RF) techniques may require multiple repetitions in the region to completely cover the entire circumference of the orifice. Each repetition may require moving the multi-electrode ablation catheter to either an area of the region that the physician deems inadequately ablated or an area that has not been ablated at all.
[0006] Physicians may be assisted by an ablation map that includes an anatomical 3D map of the region, which also displays the classification of one or more ablation tags to indicate the locations where ablation has already been performed.
[0007] However, this classification can be difficult for some anatomical structures. For example, in the left atrium (LA), classification may be difficult in the following cases: 1. Placing a catheter in an adjacent superior or inferior vein creates adjacent ablation points that are close together but spaced apart, making it difficult to distinguish and classify each location, especially if the catheter positioning can cause movement of the LA wall (i.e., tenting effect). 2. Identify whether the goal of ablation therapy is to isolate the pulmonary vein (PV) or to create an ablation isolation line near the PV (e.g., on the posterior wall of the LA or on the ceiling wall of the LA).
[0008] Each ablation sequence is performed on a specific anatomical region, but since some electrodes may be located outside that region, the exemplary challenges described above become more significant in PFA multi-electrode ablation. Therefore, each set of ablation points created by a multi-electrode catheter at a given time must be treated as a whole set where points can be mixed with other ablation points, rather than as individual points, so connecting resulting sets of ablation points (e.g., from several iterations of ablation) does not make much sense.
[0009] The aforementioned issues, along with other problems, can cause ablation tags to fail to accurately display the correct locations of ablations. This can further lead to incorrectly indicating that certain locations have been ablated when they have not actually been ablated.
[0010] For example, one or more of the reasons mentioned above may cause an ablation tag created during ablation of the superior PV orifice tissue location to incorrectly mark the inferior PV orifice location. Such errors on the ablation map may mislead or make it difficult for the physician to determine whether further ablation is necessary.
[0011] The embodiments of the present disclosure described herein provide an ablation point classification technique that can eliminate the inaccuracies in tagging caused by the aforementioned problems. As described herein, a set of ablation points may be one of the following: (i) a set of catheter electrode positions in an instance of ablation, with any additional information assigned to the electrodes therein, such as contact force, ablation power, and ablation duration; and (ii) a set of planned ablation points generated by a multi-electrode catheter in an instance of ablation.
[0012] The algorithm described in more detail below can classify a set of ablation points from instances of ablation performed by a multi-electrode catheter, taking into account the specific location of the ablated region and the shape of the catheter (e.g., a circular catheter). The disclosed method can consider the ablation points generated in any given instance as a group of points rather than a single, unrelated point, and can help to limit the tissue location of the ablation points to the correct anatomical region or location (e.g., the correct mouth) on an anatomical map with high confidence.
[0013] Additional catheter-related data, such as electrode contact force with cardiac wall tissue or tissue proximity index (TPI), may also be used for classification.
[0014] In a first embodiment of the disclosed technology, the processor can perform the following steps of a majority voting algorithm: 1. The step of assigning an ablation instance to an anatomical region (e.g., a given mouth) where the majority of the catheter electrodes are considered to be in contact during the ablation, and 2. A step of using the position and shape of the catheter in space to identify the tissue location of the electrode in contact with the tissue in that region and generate the correct ablation tag.
[0015] In the second embodiment, the processor can perform the following centroid mapping algorithm steps. 1. A step of calculating the geometric center (centroid) of all electrodes of the catheter that are in contact with the tissue during the ablation sequence, and 2. A step of assigning an ablation point to the region where the center of gravity is located, by limiting the tissue position of the electrode in contact with the catheter according to a catheter model of the electrode position, for example, on the circular curve of the loop catheter around a center of gravity located inside a given mouth, taking into consideration the shape of the catheter in space.
[0016] In the third embodiment, the processor can perform the following weighted voting algorithm steps. 1. A step of assigning weights to each of the electrodes of the catheter based on factors such as contact quality and / or signal intensity, and 2. Using the position of the catheter in space and the shape of the catheter, a step of identifying the tissue location corresponding to the electrode of the catheter having the highest total weight (e.g., the sum of the weights assigned to a particular electrode or group of electrodes).
[0017] In the fourth embodiment, the processor can perform the following probabilistic allocation algorithm steps. 1. A process of assigning the probability of belonging to each of multiple regions to each ablation point based on the distance from each region. 2. A step of aggregating probabilities to determine the most likely region associated with the ablation point, and 3. Using the position and shape of the catheter in space, the process of constraining the tissue location of the ablation point to the most likely area.
[0018] In a fifth embodiment, the processor can execute a machine learning (ML) classification algorithm that can be trained on existing labeled ablation data to predict the anatomical region of each ablation sequence (e.g., a group of ablation points) based on the electrode configuration and other aforementioned data (e.g., contact force and / or catheter shape).
[0019] In a sixth embodiment, the processor can use the output from any one or more of the first to fifth algorithms to train another ML-based algorithm, such as a random forest, to achieve the best overall results.
[0020] Description of the System The following detailed description should be read with reference to the drawings, in which like elements in different drawings are numbered the same. The drawings are not necessarily to scale and are presented to show selected embodiments and are not intended to limit the scope of the disclosure. The detailed description is not intended to limit the principles of the disclosure but is presented by way of example only. This description explains some embodiments, adaptations, variations, alternatives, and uses of the invention, including what is currently considered to be the best mode for practicing the invention, which will enable those skilled in the art to make and use the invention.
[0021] FIG. 1 is a schematic illustrative diagram of a catheter-based anatomical / electrical anatomical (EA) mapping and ablation system 10 according to an embodiment of the present disclosure.
[0022] System 10 may include a plurality of catheters that can be percutaneously inserted by physician 24 into a chamber of heart 12 or a vascular structure (seen in the insert figure 45) through the patient's vasculature. In some embodiments, the delivery sheath catheter can be inserted into a heart chamber such as the left atrium or the right atrium near the desired location within heart 12. Thereafter, a plurality of catheters can be inserted into the delivery sheath catheter to reach the desired location. The plurality of catheters can include a pacing-only catheter, a catheter for sensing intracardiac electrogram signals, an ablation-only catheter, and / or a catheter for both EA mapping and ablation. An exemplary loop ablation catheter 14 is shown in FIG. 1. In one aspect, loop catheter 14 can be configured to sense bipolar electrograms and apply PFA. Physician 24 can contact the distal tip assembly 28 of catheter 14 with the heart wall to ablate the target site of heart 12.
[0023] As shown in the insert figure 65, the distal end assembly 28 of catheter 14 can include a plurality of electrodes 26 dispersed on a curved spline 22 in some embodiments. Catheter 14 can include a position sensor 29 embedded within or near distal end assembly 28 on shaft 46 of catheter 14 to track the position and orientation of distal end assembly 28. Optionally, preferably, position sensor 29 can be a magnetic-based position sensor having three magnetic coils for sensing three-dimensional (3D) position and orientation.
[0024] The magnetic-based position sensor 29 may operate in conjunction with a position pad 25 which may include a plurality of magnetic coils 32 configured to generate a magnetic field within a predefined working volume. The real-time position of the distal end assembly 28 of the catheter 14 may be tracked based on the magnetic field generated by the position pad 25 and sensed by the magnetic-based position sensor 29. Details of the magnetic-based position sensing technology are described in U.S. Patents No. 5,391,199, No. 5,443,489, No. 5,558,091, No. 6,172,499, No. 6,239,724, No. 6,332,089, No. 6,484,118, No. 6,618,612, No. 6,690,963, No. 6,788,967, and No. 6,892,091.
[0025] In another embodiment, the system 10 may include one or more electrode patches 38, which are positioned on the patient 23 in contact with the skin and establish a position reference for a position pad 25 and for impedance-based tracking of the electrodes 26. For impedance-based tracking, a current may be directed to the electrodes 26 and sensed by the electrode skin patches 38, thereby allowing the position of each electrode to be triangulated through the electrode patches 38. Details of impedance-based position tracking techniques are described in U.S. Patents 7,536,218, 7,756,576, 7,848,787, 7,869,865, and 8,456,182.
[0026] In some embodiments, the recorder 11 may display cardiac signals 21 acquired using surface ECG electrodes 18 (e.g., electrophoresis acquired at each tracked cardiac tissue location) and intracardiac electrophoresis acquired using electrodes 26 of the catheter 14. The recorder 11 may include pacing capabilities for pacing the rhythm of the heart and / or may be electrically connected to a standalone pacer.
[0027] The system 10 may further include an ablation energy generator 50 adapted to transfer ablation energy to one or more electrodes 26 configured for ablation. The energy generated by the ablation energy generator 50 may include, but is not limited to, radiofrequency (RF) energy or pulsed magnetic field (PF) energy, including unipolar or bipolar high-voltage DC pulses (i.e., PFAs).
[0028] The patient interface unit (PIU) 30 is an interface configured to establish electrical communication between the catheter, the electrophysiological equipment, the power supply, and the workstation 55 to control the operation of the system 10 and to receive EA signals from the catheter. The electrophysiological equipment of the system 10 may include, for example, multiple catheters, a position pad 25, a body surface ECG electrode 18, an electrode patch 38, an ablation energy generator 50, and a recorder 11. Optionally, and preferably, the PIU 30 may additionally include processing capabilities for performing real-time calculations of catheter position and performing ECG calculations.
[0029] The workstation 55 may include memory 57, a processor unit 56 having memory or storage device in which appropriate operating software is loaded, and user interface capabilities. In one embodiment, the workstation 55 may optionally provide several functions, including (i) modeling the endocardial anatomical structure in three dimensions (3D) and rendering the model or anatomical map 20 to be displayed on a display device 27, (ii) displaying an activation sequence (or other data) compiled from recorded cardiac signals 21 on the display device 27 as a representative visual representation or image superimposed on the rendered anatomical map 20, (iii) displaying the real-time position and orientation of multiple catheters in the cardiac chambers, and (iv) displaying sites of interest on the display device 27, such as the location where ablation energy is applied. One commercially available product embodying the elements of system 10 is available as the CARTO™3 system, commercially available from Biosense Webster, Inc.
[0030] In the disclosed embodiments, the processor 56 can execute an algorithm that groups together multiple ablation points, such as points associated with each electrode throughout the distal end assembly 28 of the catheter loop, taking into account the specific location and spatial configuration (i.e., shape) of the distal end assembly 28, and constrains the tissue location to conform to the catheter loop shape, as shown in Figure 2.
[0031] In some embodiments, the processor 56 may comprise a general-purpose computer programmed with software to perform the functions described herein. The software may be downloaded to the computer in electronic form, for example, via a network, or alternatively or additionally, provided and / or stored on a non-temporary tangible medium such as magnetic memory, optical memory, or electronic memory.
[0032] This configuration of System 10 is shown as an example to illustrate a specific problem addressed by the embodiments of this disclosure and to demonstrate the application of these embodiments in improving the performance of such systems. However, the embodiments of this disclosure are not limited to this particular exemplary system, and the principles described herein may be similarly applied to other medical systems. For example, the systems, methods, and techniques described herein can be used in conjunction with other multi-electrode ablation catheter types, such as basket catheters, balloon catheters, or several other suitable shapes of multi-electrode ablation catheters.
[0033] Classification of ablation points according to catheter data Figure 2 is a schematic diagram illustrating the classification of ablation points 220 based at least partially on data received from catheter 14 according to an embodiment of the present disclosure.
[0034] As shown in Figure 2, flexible tissues in anatomical regions such as orifices 210A and 212A can respond to contact with the catheter (e.g., tenting or deformation), which may cause the tissue location of the ablation to be misrepresented (e.g., tagged) in the mapped (200) representations 210B and 212B of the orifice.
[0035] In the illustrated embodiment, the ablation point 220 may occur at the pushed wall of the tissue 202 at the opening 210A. Based solely on the coordinates of the map 200, the algorithm may incorrectly assign the ablation point 220 using the tag 222B located at the lower opening 212B on the map 200.
[0036] By applying one or more of the algorithms considered above, which can take into account the shape of the distal end assembly 28, the disclosed technique can correctly assign ablation points 220 using ablation tags 222A located at the mapped mouth 210B, instead of incorrectly assigning ablation tags 222B to mouth 212B.
[0037] In particular, in the illustrated embodiment, the disclosed technique can accurately assign the correct ablation tag 222A by taking into account the locations of other ablation points 230 performed by the loop catheter during the same ablation instance. In one embodiment, using one of the classification algorithms described above may involve determining the correct anatomical region to associate with an ablation point by assuming that the centroid position 250 of the ablation point (220, 230) lies within that region.
[0038] The disclosed technique can also ensure that ablation tags for other ablation lines, such as those fabricated on the posterior wall of the left atrium, are precisely positioned. In this embodiment, the tags are precisely positioned despite changes in the flexible posterior wall tissue coordinates between those recorded during ablation and those represented by the map (i.e., representing the tissue in the absence of a catheter and before it experiences deflection or tenting).
[0039] An ablation session may include multiple instances of the type shown in Figure 2. An ablation map based on the above analysis applied to multiple ablation instances is shown in Figure 3.
[0040] Ablation map based on ablation points classified according to catheter data Figure 3 shows an ablation map 300 of cardiac chambers obtained using one of the disclosed algorithms for classifying ablation points, according to an embodiment of the present disclosure. The ablation map 300 may include an anatomical map 302 presented (e.g., superimposed) together with ablation tags such as tags 304, 306, and 310.
[0041] The illustrated cavity is the left atrium, and can be graphically coded by a segmentation algorithm to visualize, among other things: Left atrial wall and four PVs, Upper left PV (LSPV), Lower left PV (LIPV), The upper right PV (RSPV), and / or, Lower right PV (RIPV)
[0042] Map 300 can further be based on one or more of the disclosed classification algorithms that, based on catheter data, assign each set of ablation points of the loop catheter 14 to the correct region. As illustrated, each ablation tag 304, 306, and 310 may appear as a set, each set distributed across the circular contour (322) of the distal end assembly 28. Note that once the tissue returns to its normal shape (i.e., once the catheter is repositioned after the application of ablation energy so as not to deform or tentate the tissue), the arc of the contour may appear as a straight line rather than a loop, as in the case between the ablation tags 310 of the ablation line on the left atrial wall.
[0043] Map 300 can be generated offline using stored data, or it can be generated during the clinical ablation procedure and updated in real time as the ablation procedure progresses. Depending on the shape of the catheter, Map 300 may have a slightly different appearance than the one exemplified.
[0044] How to generate an ablation map using ablation point classification Figure 4 is a schematic flowchart illustrating a method for generating an ablation map of cardiac chambers using the above-described technique for classifying ablation points, according to an embodiment of the present disclosure. One or more of the algorithms described in the presented embodiment can perform a process in which, in the map receiving step 402, the processor 56 receives the anatomical map 302.
[0045] In step 404, the processor may further receive a set of ablation points from instances of ablation by the catheter 14 in the cardiac chambers. Step 404 may optionally include the application of ablation energy to the tissue by the catheter 14 before receiving the set of ablation points.
[0046] Next, in the classification step 406, the processor may apply one or more of the disclosed classification algorithms that take into account the shape of the catheter before assigning the ablation points to the correct region of the map.
[0047] In the ablation tag generation step 408, the processor can generate each set of ablation tags. In the ablation tag overlay step 410, the processor can overlay the set of ablation tags 306 onto the map 302 in order to generate the ablation map 300.
[0048] In the graphical coding step 412, the processor can graphically code the ablation tag 306 to display ablation characteristics such as PV port identification information and the level of ablation. Graphical coding of the tag ablation map may include coloring ablation tags in one area with one color and ablation tags in another area with a different color.
[0049] Finally, in the ablation map presentation step 414, the processor can present the graphically coded ablation map 300 to the user on the display device 27.
[0050] The flowchart in Figure 4 is simplified to illustrate a single example of ablation. However, in actual procedures, multiple ablation instances may occur (for example, multiple ablations being performed iteratively in step 404), and the processor may update the ablation map 300 multiple times during the procedure. The final map 300 may reflect the complete treatment if the ablation session is fully completed.
[0051] Map 300 may be generated using offline data, or it may be generated during the clinical ablation procedure and updated in real time as the ablation procedure progresses. [Examples]
[0052] In one embodiment, the method according to this description may include receiving an anatomical map (200) relating to at least a portion of the wall tissue (202) of the cardiac chambers, the map comprising different anatomical regions (210, 212). The method may further include receiving a set of ablation points (220, 230) generated by a multi-electrode catheter (14), and one or more sets of planned ablation points to be generated by the multi-electrode catheter (14) in ablation instances in one of the anatomical regions. Using a classification algorithm that takes into account the shape and position data of the catheter (14), one of the anatomical regions (210, 212) (210) can be identified, and all ablation points (220, 230) can be classified as belonging to the identified anatomical region (210). Each ablation tag (222A) may be presented on the anatomical map (200) according to the classified ablation points (220, 230) and the shape and position data of the catheter.
[0053] In a further embodiment, the classification algorithm may further include determining the correct anatomical region to which the ablation points are associated by assuming that the majority of the ablation points (220, 230) lie within that region.
[0054] In other embodiments, the classification algorithm may include determining the correct anatomical region (220) to which the ablation point is associated by assuming that the centroid (250) position of electrodes distributed around the shape of the ablation point and / or the distal end of the catheter lies within that region.
[0055] Alternatively or additionally, the classification algorithm may include assigning weights to each ablation point associated with the shape data of the catheter (14) and identifying the correct anatomical region (210) to which the ablation points should be associated by determining which region the subset of ablation points (220, 230) that yield the highest total weights are located.
[0056] In other embodiments, the classification algorithm may include determining a suitable anatomical region (210) to associate an ablation point with by calculating the distance between each of the multiple ablation points (220, 230) from each of the multiple regions, and determining the correct region (210) on the map (200) to associate the ablation point with based on probabilities derived from the distances.
[0057] In yet another embodiment, the classification algorithm may include training a machine learning (ML)-based algorithm based on a dataset of labeled ablation points, and using the trained ML-based algorithm to infer the correct anatomical region (210) based on the received set of ablation points (220, 230).
[0058] Training an ML model may involve training the model on data labeled using one or more of the following classification algorithms: majority voting classification algorithms, centroid location classification algorithms, distance-based probabilistic classification algorithms, and weighted ablation point-based classification algorithms (each described herein).
[0059] In another embodiment, the system (10) according to this description may include a memory (57) and a processor (56). The memory may be configured to store a classification algorithm. The processor may be further configured to (i) receive an anatomical map (200) of at least some wall tissue (202) of the cardiac chambers, the map including different anatomical regions (210, 212), and (ii) receive one or more of a set of ablation points (220, 230) generated by a multi-electrode catheter (14) and a set of planned ablation points to be generated by the multi-electrode catheter (14) in ablation instances in one of the anatomical regions. In one embodiment, the processor (56) may be further configured to (a) identify one of the anatomical regions (210) and classify all ablation points (220, 230) as belonging to the identified anatomical region, and (b) present each ablation tag (222A) on the anatomical map (200) according to the classified ablation points (220, 230) and the shape and position data of the catheter.
[0060] The embodiments described above are illustrative examples, and it should be understood that this disclosure is not limited to those illustrated and described above. Rather, the scope of this disclosure includes both combinations and partial combinations of the various functions described above, as well as variations and modifications thereof that a person skilled in the art would conceive of from reading the foregoing description and that are not disclosed in the prior art.
[0061] For example, any features, materials, properties, or groups described in relation to a particular aspect, configuration, or example should be understood to be applicable to any other aspect, configuration, or example described in this section or elsewhere in this specification, provided that they do not conflict. All features disclosed herein (including the appended claims, abstract, and drawings) and / or all steps of any method or process disclosed herein may be combined in any combination, except for any combination in which at least some of such features and / or steps are mutually exclusive.
[0062] Operations may be shown in the drawings or described in the specification in a specific order, but such operations do not need to be performed in the specific order or sequence shown, or not all operations need to be performed, in order to achieve the desired result. Other operations not described or described may be incorporated into exemplary methods and processes. For example, one or more additional operations may be performed before, after, simultaneously with, or between any of the operations described. Furthermore, in other implementations, operations may be reconfigured or rearranged. Those skilled in the art will understand that in some configurations, the actual steps performed in the illustrated and / or disclosed processes may differ from the steps shown in the drawings. Depending on the configuration, some of the steps described above may be omitted and others may be added. Furthermore, the features and attributes of the specific configurations disclosed above may be combined in different ways to form additional configurations, all of which fall within the scope of this disclosure.
[0063] The scope of this disclosure is defined solely by reference to the following claims.
[0064] [Implementation Method] (1) A method for generating an ablation map of cardiac chambers, Receiving an anatomical map of at least a portion of the wall tissue of the cardiac chambers, wherein the map includes different anatomical regions. Receiving multiple ablation points generated by a multi-electrode catheter, Using a classification algorithm that takes catheter shape data into account, an anatomical region is identified, and all of the multiple ablation points are classified as belonging to the identified anatomical region. A method comprising presenting each ablation tag on the anatomical map according to the classified ablation points and catheter shape data. (2) The method according to Embodiment 1, wherein using the classification algorithm includes identifying the anatomical region by determining that the majority of the ablation points fall within the anatomical region. (3) The method according to Embodiment 1, wherein using the classification algorithm includes identifying the anatomical region by considering that the centroid of the ablation point falls within the anatomical region. (4) Using the classification algorithm described above means Assigning weights to each ablation point based at least partially on the catheter shape data, The method according to Embodiment 1, comprising identifying the anatomical region at least partially based on a subset of ablation points that yield the best weighted sum. (5) The method according to Embodiment 1, wherein using the classification algorithm includes identifying the anatomical region by determining the probability that the ablation point belongs to the anatomical region, the probability being calculated based on the distance of each of the multiple ablation points from the anatomical region on the map.
[0065] (6) The method according to Embodiment 1, wherein using the classification algorithm comprises training a machine learning (ML) based algorithm on a dataset of labeled ablation points, and using the trained ML-based algorithm to infer the anatomical regions of the received ablation points. (7) The method according to Embodiment 6, wherein training the machine learning (ML) model is trained on data labeled using one or more of the following classification algorithms: majority voting classification algorithms, centroid position classification algorithms, distance-based probabilistic classification algorithms, and weighted ablation point-based classification algorithms. (8) The method according to Embodiment 1, wherein the anatomical map is an electroanatomical (EA) map. (9) A system for generating an ablation map of cardiac chambers, Memory configured to store classification algorithms, It is a processor, Receiving an anatomical map of at least a portion of the wall tissue of the cardiac chambers, wherein the map includes different anatomical regions. Receiving multiple ablation points generated by a multi-electrode catheter, Using the classification algorithm that takes catheter shape data into consideration, an anatomical region is identified, and all of the multiple ablation points are classified as belonging to the identified anatomical region. A system including a processor configured to display ablation tags on the anatomical map according to the classified ablation points and catheter shape data. (10) The system according to embodiment 9, wherein the processor is configured to use the classification algorithm by identifying the anatomical region by determining that the majority of the ablation points fall within the anatomical region.
[0066] (11) The system according to embodiment 9, wherein the processor is configured to use the classification algorithm by identifying the anatomical region by considering that the centroid position of the ablation point falls within the anatomical region. (12) The processor Assigning weights to each ablation point based at least partially on the catheter shape data, The system according to embodiment 9, configured to use the classification algorithm by identifying the anatomical region on at least a portion of a subset of ablation points that yield the best weight sum. (13) The system according to Embodiment 9, wherein the processor is configured to use the classification algorithm by identifying the anatomical region by determining the probability that the ablation point belongs to the anatomical region, the probability being calculated based on the distance of each of the multiple ablation points from the anatomical region on the map. (14) The system according to Embodiment 9, wherein the processor is configured to use the classification algorithm by training a machine learning (ML) based algorithm on a dataset of labeled ablation points and using the trained ML-based algorithm to infer the anatomical regions of the received ablation points. (15) The system according to Embodiment 14, wherein the processor is configured to train the machine learning (ML) model by training the model on data labeled using one or more of the following classification algorithms: a majority voting classification algorithm, a centroid position classification algorithm, a distance-based probabilistic classification algorithm, and a weighted ablation point-based classification algorithm.
[0067] (16) The system according to embodiment 9, wherein the anatomical map is an electroanatomical (EA) map.
Claims
1. A system for generating ablation maps of cardiac chambers, Memory configured to store classification algorithms, It is a processor, Receiving an anatomical map of at least a portion of the wall tissue of the cardiac chambers, wherein the map includes different anatomical regions. Receiving multiple ablation points generated by a multi-electrode catheter, Using the classification algorithm that takes catheter shape data into consideration, an anatomical region is identified, and all of the multiple ablation points are classified as belonging to the identified anatomical region. A system including a processor configured to display ablation tags on the anatomical map according to the classified ablation points and catheter shape data.
2. The system according to claim 1, wherein the processor is configured to use the classification algorithm by identifying the anatomical region by determining that the majority of the ablation points fall within the anatomical region.
3. The system according to claim 1, wherein the processor is configured to use the classification algorithm by identifying the anatomical region by considering that the centroid position of the ablation point falls within the anatomical region.
4. The aforementioned processor, Assigning weights to each ablation point based at least partially on the catheter shape data, The system according to claim 1, configured to use the classification algorithm by identifying the anatomical region on at least a portion of a subset of ablation points that yield the best weight sum.
5. The system according to claim 1, wherein the processor is configured to use the classification algorithm by identifying the anatomical region by determining the probability that the ablation points belong to the anatomical region, the probability being calculated based on the distance of each of the plurality of ablation points from the anatomical region on the map.
6. The system according to claim 1, wherein the processor is configured to use the classification algorithm by training a machine learning (ML) based algorithm on a dataset of labeled ablation points, and by using the trained ML-based algorithm to infer the anatomical regions of the received a plurality of ablation points.
7. The system according to claim 6, wherein the processor is configured to train the machine learning (ML) model by training the model with data labeled using one or more of the following classification algorithms: a majority-vote classification algorithm, a centroid-position classification algorithm, a distance-based probabilistic classification algorithm, and a weighted ablation point-based classification algorithm.
8. The system according to claim 1, wherein the anatomical map is an electroanatomical (EA) map.
9. A method for generating an ablation map of cardiac chambers, Receiving an anatomical map of at least a portion of the wall tissue of the cardiac chambers, wherein the map includes different anatomical regions. Receiving multiple ablation points generated by a multi-electrode catheter, Using a classification algorithm that takes catheter shape data into account, an anatomical region is identified, and all of the multiple ablation points are classified as belonging to the identified anatomical region. A method comprising presenting each ablation tag on the anatomical map according to the classified ablation points and catheter shape data.
10. The method of claim 9, wherein using the classification algorithm includes identifying the anatomical region by determining that the majority of the ablation points fall within the anatomical region.
11. The method according to claim 9, wherein using the classification algorithm includes identifying the anatomical region by considering that the centroid of the ablation point falls within the anatomical region.
12. Using the aforementioned classification algorithm means Assigning weights to each ablation point based at least partially on the catheter shape data, The method according to claim 9, comprising identifying the anatomical region on at least a portion of a subset of ablation points that yield the best weight sum.
13. The method according to claim 9, wherein using the classification algorithm includes identifying the anatomical region by determining the probability that the ablation point belongs to the anatomical region, the probability being calculated based on the distance of each of the plurality of ablation points from the anatomical region on the map.
14. The method according to claim 9, wherein using the classification algorithm comprises training a machine learning (ML) based algorithm on a dataset of labeled ablation points, and using the trained ML-based algorithm to infer the anatomical regions of the received ablation points.
15. The method according to claim 14, wherein training the machine learning (ML) model involves training the model with data labeled using one or more of the following classification algorithms: majority voting classification algorithms, centroid location classification algorithms, distance-based probabilistic classification algorithms, and weighted ablation point-based classification algorithms.
16. The method according to claim 9, wherein the anatomical map is an electroanatomical (EA) map.