Cardiac conduction system imaging

Abstract

There is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating an indication of the cardiac conduction system of the subject according to computed correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features, and wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

Description

[0001] CARDIAC CONDUCTION SYSTEM IMAGING

[0002] RELATED APPLICATION / S

[0003] This application claims the benefit of priority of U.S. Provisional Patent Application No. 63 / 530,724 filed on August 4, 2023, of U.S. Provisional Patent Application No. 63 / 531,063 filed on August 7, 2023, of U.S. Provisional Patent Application No. 63 / 532,995 filed on August 16, 2023, of U.S. Provisional Patent Application No. 63 / 621,140 filed on January 16, 2024, of U.S. Provisional Patent Application No. 63 / 656,164 filed on June 5, 2024, of U.S. Provisional Patent Application No. 63 / 664,786 filed on June 27, 2024, the contents of which are incorporated herein by reference in their entirety.

[0004] BACKGROUND

[0005] The present invention, in some embodiments thereof, relates to cardiac imaging and, more specifically, but not exclusively, to systems and methods for imaging the cardiac conduction system.

[0006] Different approaches have been developed to obtain information about the cardiac conduction system of the heart. For example, an electrocardiogram (ECG) records the electrical activity of the heart. In another example, electrophysiological studies (EPS) direct map electrical activity in the heart using catheters and electrodes.

[0007] SUMMARY

[0008] According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, computing correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations, and generating an indication of the cardiac conduction system of the subject according to the computed correlations, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0009] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, computing correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations, and generating an indication of the cardiac conduction system of the subject according to the computed correlations, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0010] According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating an indication of the cardiac conduction system of the subject according to computed correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features, and wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0011] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating an indication of the cardiac conduction system of the subject according to computed correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features, and wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0012] Optionally, the correlations are based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations.

[0013] Optionally, further comprising computing the correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features.

[0014] Optionally, further comprising obtaining the correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features.

[0015] Optionally, the correlations are computed between at least one combination of the plurality of structures of the cardiac conduction system and at least one combination of the plurality of gross anatomical features. Optionally, the indication of the cardiac conduction system comprises a 3D image including the plurality of structures and interpolations therebetween.

[0016] Optionally, the plurality of gross anatomical features are selected from: non-coronary cusp (NCC), right coronary cusp (RCC), left coronary cusp (LCC), nadir of one or more of NCC RCC and LCC, virtual annulus plane, membranous septum, rotation angle between aortic valve and membranous septum, membranous septum floor, posterior papillary muscle of the left ventricle, anterior papillary muscle of the left ventricle, apex of the left ventricle, medial papillary muscle of the right ventricle, the moderator band of the right ventricle, the anterior papillary muscle of the right ventricle, intraventricular septum, myocardial ring connecting to aortic valve, myocardial ring connecting to mitral valve, myocardial notch between plane of myocardial rings of the aortic valve and mitral valve, base (i.e., inferior border) of the infra pyramidal space (IPS), apex of the IPS, anterior Border of the IPS, posterior Border of the IPS, lateral borders of the IPS, the roof of the inferoseptal recess, anterior / superior border of the inferoseptal recess, posterior / inferior border of the inferoseptal recess, right side of the inferoseptal recess, left side of the inferoseptal recess, diverging vestibules of the mitral and tricuspid valves, the ventricular boundary of the inferoseptal recess, the apex of the inferoseptal recess, medial papillary muscle (MPM), septal band (SB), moderator band (MB), muscle between septum and free wall (myocardium), MB1 - connected to the interventricular septum, and MB2 - connected to the ventricular myocardium.

[0017] Optionally, the structures of the cardiac conduction system are selected from: atrioventricular (AV) node, bundle of his, penetrating bundle, non-branching bundle, branching point of a left branch bundle (LBB), LBB, right branch bundle (RBB), left anterior fascicle, left posterior fascicle, left septal fascicle, Mahaim fibers, Purkinje fibers, and connection paths between two of the aforementioned along the cardiac conduction system.

[0018] It is noted that fascicles may be further divided into sub-regions, for example, a proximal / superior region that is closer to the LBB, and a distal / inferior region that is closer to the apex, or other divisions and / or great number of sub-regions.

[0019] Optionally, the plurality of structures include an atrioventricular (AV) node estimated as being location within at least one of: (i) fat of the infra-septal recess at the level close to the septal commissure of the mitral annulus, (ii) within the atrioventricular septum, near the opening of the coronary sinus, and part of the triangle of Koch, (iii) located just above the infra septal recess, and (iv) adjacent to an apex of the infra pyramidal space..

[0020] Optionally, the indication of the cardiac conduction system is generated as a visual marking depicted within a 3D representation based on the 3D image. Optionally, further comprising presenting the 3D representation depicting the visual marking on a display and within a graphical user interface configured for rotation of the 3D representation.

[0021] Optionally, further comprising: receiving a 2D image of the heart, registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image, projecting a 3D indication of the cardiac conduction system to the 2D plane for creating a 2D projection of the indication of the cardiac conduction system, and presenting the 2D projection on the 2D image.

[0022] Optionally, the receiving, the registering, the projecting and the presenting, are dynamically iterated during a procedure for varying poses at which the 2D image is captured.

[0023] Optionally, further comprising: receiving a 2D image of the heart, registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image, mapping the plurality of gross anatomical features depicted in the 3D image to the 2D image, generating a 2D indication of the cardiac conduction system of the subject according to the computed correlations and according to the mapping, and presenting the 2D indication of the cardiac conduction system as an overlay over the 2D image.

[0024] Optionally, the structures of the cardiac conduction system are within about 3 centimeters of an aortic valve of the heart.

[0025] Optionally, further comprising feeding the indication of the cardiac conduction system of the subject into an artificial intelligence (Al) process for further analysis.

[0026] Optionally, the Al process is configured for at least one of: estimating risk of damage to the conduction system, and planning a transcatheter procedure for reducing or avoiding damage to the conduction system.

[0027] Optionally, the 3D image excludes visualization of the plurality of structures of the cardiac conduction system.

[0028] Optionally, the indication of the cardiac conduction system is generated by adapting a template of the cardiac conduction system according to the computed correlations.

[0029] Optionally, the plurality of structures include the AV node, estimated to be in at least one of: (i) a middle of an infra-septal recess, (ii) within an intraventricular septum adjacent to a septal commissure between a posterior leaflet and an anterior leaflet of a mitral valve, and / or (iii) the intraventricular septum at the left most proximal part of the posterior leaflet adjacent to the septal commissure between the posterior and the anterior leaflets of the mitral valve.

[0030] Optionally, AV node within the indication of the cardiac conduction system denotes an anatomical region of at least one of: (i) the infra-septal recess, (ii) the intraventricular septum adjacent to the septal commissure between the posterior leaflet and the anterior leaflet of the mitral valve, and / or (iii) the intraventricular septum at the left most proximal part of the posterior leaflet adjacent to the septal commissure between the posterior and the anterior leaflets of the mitral valve.

[0031] Optionally, the plurality of structures include a non bifurcating bundle including the bundle of His, estimated to be between the estimated location of the AV node to the most posterior part of the floor of the membranous septum.

[0032] Optionally, the non bifurcating bundle including the bundle of His within the indication of the cardiac conduction system denotes an anatomical region between the estimated location of the AV node and the most posterior part of the floor of the membranous septum.

[0033] Optionally, a location of the bundle of His may be estimated as at least one of: (i) originating from the AV node and traveling through the floor of the membranous part of the interventricular septum, in relation to the inferoseptal recess, (ii) as passing close to the inferoseptal recess as the inferoseptal recess descends through the septum, and (iii) being closely associated with the infra pyramidal space at its origin from the AV node.

[0034] Optionally, the plurality of structures include the penetrating bundle, estimated to be between the most posterior part of the floor of the membranous septum to the most anterior part of the floor of the membranous septum tracking over the floor of the membranous septum.

[0035] Optionally, the penetrating bundle within the indication of the cardiac conduction system denotes an anatomical region between the most posterior part of the floor of the membranous septum to the most anterior part of the floor of the membranous septum tracking over the floor of the membranous septum.

[0036] Optionally, the plurality of structures included the bifurcating bundle, estimated to be toward the left ventricle next to the most anterior part of the floor of the membranous septum.

[0037] Optionally, the bifurcating bundle within the indication of the cardiac conduction system denotes an anatomical region toward the left ventricle next to the most anterior part of the floor of the membranous septum.

[0038] Optionally, the plurality of structures included the right bundle branch, estimated to be across from the estimated location of the penetrating bundle on the right ventricle wall.

[0039] Optionally, the right bundle branch within the indication of the cardiac conduction system denotes an anatomical region across from the estimated location of the penetrating bundle on the right ventricle wall.

[0040] Optionally, a location of the right branch bundle is estimated as at least one of: (i) originating from the branching bundle during the path of the branching bundle over the membranous septal floor, and (ii) as traveling along the right side of the interventricular septum toward the medial papillary muscle of the right ventricle.

[0041] Optionally, the plurality of structures include the left bundle branch, on the left ventricle toward the apex of the left ventricle.

[0042] Optionally, the left bundle branch within the indication of the cardiac conduction system denotes an anatomical region on the left ventricle toward the apex of the left ventricle.

[0043] Optionally, a location of the left bundle branch is estimated as at least one of: (i) originating from the branching bundle during a path of the branching bundle over the membranous septal floor, and (ii) as traveling along the left side of the interventricular septum.

[0044] Optionally, the plurality of structures include the left anterior fascicle, estimated to be along a region defined by a line from below the location of the branching point of the left bundle branch to a base of the anterior papillary muscle of the left ventricle.

[0045] Optionally, the left anterior fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to a base of the anterior papillary muscle of the left ventricle.

[0046] Optionally, the plurality of structures include the left posterior fascicle, estimated to be along a region defined by a line from below the location of the branching point of the left bundle branch to a base of the posterior papillary muscle of the left ventricle.

[0047] Optionally, the left posterior fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to a base of the posterior papillary muscle of the left ventricle.

[0048] Optionally, the plurality of structures include the left septal fascicle, estimated to be along a region defined by a line from below the location of the branching point of the left bundle branch to the apex of the left ventricle with a path that passes between the anterior and posterior papillary muscles.

[0049] Optionally, the left septal fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to the apex of the left ventricle with a path that passes between the anterior and posterior papillary muscles.

[0050] Optionally, the plurality of gross anatomical features include at least a portion of the membranous septum, wherein analyzing comprises analyzing the 3D image for detecting the membranous septum. Optionally, the membranous septum is identified as a region between a ventricular septum and an inferoseptal recess, wherein analyzing comprises analyzing the 3D image for detecting the ventricular septum and the inferoseptal recess.

[0051] Optionally, the inferoseptal recess is identified according to at least neighboring tissue detected in the 3D image, selected from: (i) a roof of the inferoseptal recess identified as an area of fibrous tissue continuity between leaflets of a mitral valve and a tricuspid valve, and / or identifying the fibrous tissue continuity as tissue at a base of the atrial septum, (ii) anterior and / or superior border of the inferoseptal recess identified as an aortic leaflet of the mitral valve, (iii) posterior and / or inferior border of the inferoseptal recess identified as a septal surface of the left ventricle, (iv) right wall of the inferoseptal recess identified as fibrous tissue through which an identified His bundle penetrates, (v) left side of the inferoseptal recess identified as a border of a left ventricular cavity, (vi) detecting the diverging vestibules to define the atrial boundary of the infero septal recess, (vii) detecting a crest of the muscular ventricular septum to define the ventricular boundary, and (viii) detecting the apex of the inferoseptal recess to identify the inferior margin of the membranous septum.

[0052] Optionally, the plurality of gross anatomical features include an infra pyramidal space (IPS), wherein the plurality of structure include an atrioventricular (AV) node, wherein the correlations include a location of the AV node at an end of a tip of a pyramid-like shape of the infra pyramidal space.

[0053] Optionally, the IPS is identified by detecting one or more of the following: (i) base of the IPS identified by detecting a diaphragmatic surface of the heart, (ii) apex of the IPS identified by detecting a central fibrous body of the atrioventricular septum, close to the noncoronary sinus of the aortic root, (iii) anterior Border of the IPS identified by detecting a crest of the interventricular septum, (iv) posterior Border of the IPS identified as being continuous with a detected diaphragmatic surface of the heart at the crossing of the left and right inferior atrioventricular and interatrial grooves, and (v) lateral Borders of the IPS identified by detecting two septal atrial walls.

[0054] Optionally, further comprising creating a fused image by registering the indication of the cardiac conduction system to an anatomical image, and presenting the fused image depicting the anatomical image with the indication of the cardiac conduction system.

[0055] Optionally, further comprising creating a fused image by registering the indication of the cardiac conduction system to a functional image, and presenting the fused image depicting the functional image with the indication of the cardiac conduction system. Optionally, the functional image depicts at least one of: conduction functionality of nerves of the cardiac conduction system, and viability of the myocardium of the heart innervated by the cardiac conduction system.

[0056] Optionally, the gross anatomical features include gross anatomical landmarks.

[0057] Optionally, the indication of the cardiac conduction system is represented as a plurality of spaced apart visual elements represents a probability distribution indicating likelihood of location of structures of the cardiac conduction system, the probability distribution computed from an analysis of the location of the structures of the cardiac conduction system in a plurality of other subjects.

[0058] Optionally, the indication of the cardiac conduction system is generated excluding data from at least one sensor sensing electrical activity of the cardiac conduction system.

[0059] Optionally, further comprising: iterating the receiving, the analyzing, the computing, and the generating, for each of a plurality of sample 3D images of sample hearts of sample subjects, for creating a training dataset of a plurality of records, wherein a record includes the sample 3D image, and a ground truth of the indication of the cardiac conduction system generated for the sample 3D image, and training a machine learning model on the training dataset.

[0060] Optionally, further comprising identifying a membranous septum, and identifying a membranous septum floor by drawing a highest elevation line of a myocardium and identifying a crossing of the identified membranous septum and the highest elevation line of the myocardium.

[0061] Optionally, the indication comprises an AV node, and the plurality of gross anatomical features comprise an infrapyramidal space and / or inferoseptal recess, and the AV node is identified as being located closest to an apex of the infrapyramidal space, and / or at an intersection of the infrapyramidal space and inferoseptal recess.

[0062] Optionally, the indication comprises a perforating bundle and the plurality of gross anatomical features comprise an inferoseptal recess and membranous septum floor, and the perforating bundle is identified at an intersection of the inferoseptal floor and posterior border of the membranous septum floor.

[0063] Optionally, the indication comprise a non branching bundle of His and an AV node and a perforating bundle, and the plurality of gross anatomical features comprise an inferoseptal recess, and wherein the non branching bundle of His is identified as extending from the AV node to the perforating bundle over a surface of the inferoseptal recess.

[0064] Optionally, the indication comprise a branching point of the left bundle branch, and the plurality of gross anatomical features comprise a membranous septum floor, wherein the branching point of the left bundle branch is identified as being located at about 60-80% (e.g., 70%) of a length of the membranous septum floor.

[0065] Optionally, the indication comprises a right bundle branch, and the plurality of gross anatomical features comprise a membranous septum floor, wherein right bundle branch is identified as being located at about 50-70% (e.g., 60%) of a length of the membranous septum floor.

[0066] Optionally, the indication comprises a left bundle branch, and the plurality of gross anatomical features comprise a membranous septum floor, wherein left bundle branch is identified as extending along an entire length of the membranous septum floor, and covering an endocardium of a left ventricle for a distance of about 1-4 times a length of the membranous septum, and breaking into three fascicles.

[0067] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, estimating locations of one or more structures of the cardiac conduction system using the plurality of gross anatomical features, and marking locations of the one or more structures of the cardiac conduction system on a visual rendition of the heart.

[0068] According to some embodiments of the present invention, there is provided a system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, estimating locations of one or more structures of the cardiac conduction system using the plurality of gross anatomical features, and marking locations of the one or more structures of the cardiac conduction system on a visual rendition of the heart.

[0069] Optionally, the estimating the locations is by applying at least one rule derived from an analysis of a plurality of hearts of different individuals.

[0070] Optionally, further comprising: creating a training dataset of a plurality of records for a plurality of sample 3D images of a plurality of sample individual, wherein a record includes a sample 3D image and a ground truth of the marking of the locations of the one or more structures of the cardiac conduction system, and training a machine learning model on the training dataset, wherein the machine learning model generates the locations of the one or more structures in response to an input of a target 3D image. Optionally, the estimating the locations is by feeding at least a portion of the 3D image depicting at least one of the gross anatomical features into a machine learning model.

[0071] Optionally, the machine learning model is trained on a training dataset of a plurality of records, wherein a record includes at least a sample of a portion of a sample 3D image of a sample individual depicting the at least one of the gross anatomical features, and a ground truth indicating the locations of the one or more structures.

[0072] Optionally, further comprising rendering an interpolation of a non-trivial geometry of the cardiac conduction system between the marked locations of the one or more structures.

[0073] Optionally, the rendering the interpolation is according to at least one rule.

[0074] Optionally, further comprising adapting one or more of the computed correlations between at least one of the gross anatomical features and at least one of the structures of the cardiac conduction system based on a current medical state of the subject.

[0075] Optionally, the current medical state is selected from: genetic conditions and acquired disease conditions.

[0076] Optionally, further comprising: dynamically training a machine learning model on published cases describing variation in deviations of locations of structures of the cardiac conduction system for different medical states of individuals, feeding at least one of the computed correlations and the current medical state of the subject into the machine learning model, and obtaining the adaptation of the computed correlation from the machine learning model.

[0077] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, estimating locations of one or more structures of the cardiac conduction system in the 3D image using the plurality of gross anatomical features, and marking locations of the one or more structures of the cardiac conduction system on a 2D image of the heart.

[0078] A system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, estimating locations of one or more structures of the cardiac conduction system in the 3D image using the plurality of gross anatomical features, and marking locations of the one or more structures of the cardiac conduction system on a 2D image of the heart.

[0079] According to some embodiments of the present invention, there is provided a system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, estimating locations of one or more structures of the cardiac conduction system in the 3D image using the plurality of gross anatomical features, and marking locations of the one or more structures of the cardiac conduction system on a 2D image of the heart.

[0080] A system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, estimating locations of one or more structures of the cardiac conduction system in the 3D image using the plurality of gross anatomical features, and marking locations of the one or more structures of the cardiac conduction system on a 2D image of the heart.

[0081] Optionally, further comprising: receiving the 2D image of the heart, registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image, projecting the 3D estimated location of the one or more structures of the cardiac conduction system to the 2D plane for creating a 2D projection of the one or more structures, and presenting the 2D projection on the 2D image.

[0082] Optionally, the receiving, the registering, the projecting and the presenting, are dynamically iterated during a procedure for varying poses at which the 2D image is captured.

[0083] Optionally, further comprising: receiving the 2D image of the heart, registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image, mapping the plurality of gross anatomical features depicted in the 3D image to the 2D image, generating a 2D indication of the cardiac conduction system of the subject according to the estimated locations and according to the mapping, and presenting the 2D indication of the cardiac conduction system as an overlay over the 2D image.

[0084] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and estimating locations of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0085] According to some embodiments of the present invention, there is provided a system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and estimating locations of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0086] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and providing an indication of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0087] A system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and providing an indication of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0088] According to some embodiments of the present invention, there is provided a system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and providing an indication of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0089] A system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and providing an indication of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0090] Optionally, further comprising: computing a registration for registering the 2D image with the 3D image, wherein the plurality of gross anatomical features and the one or more structures of the cardiac conduction system are located within the 3D image, and associated with CT derived information (CDI) indicating coordinates of locations of corresponding structures of the cardiac conduction system within the 3D image, detecting the plurality of gross anatomical features on the 2D image, and extracting fluoroscopy derived information (FDI) indicating the location of the corresponding structures of the cardiac conduction system on the 2D image and a pose of a sensor that captured the 2D image, estimating the location of the one or more structures of the cardiac conduction system on the 2D image according to the detected locations of the plurality of gross anatomical features on the 2D image, computing a pose of the 3D image that is closest to the pose of the sensor that captured the 2D image, applying the registration for projecting the one or more structures of the cardiac conduction system of the 3D image defined by the CDI to a 2D plane corresponding to the FDI, according to the pose for generating a synthetic view, and providing the synthetic view for presentation.

[0091] Optionally, further comprising: computing a registration for registering the 2D image with the 3D image, wherein the plurality of gross anatomical features are located within the 3D image, and associated with CT derived information (CDI) indicating coordinates of locations thereof within the 3D image, detecting the plurality of gross anatomical features on the 2D image, and extracting fluoroscopy derived information (FDI) indicating the location of the plurality of gross anatomical features on the 2D image and a pose of a sensor that captured the 2D image, computing a pose of the 3D image that is closest to the pose of the sensor that captured the 2D image, applying the registration for projecting the plurality of gross anatomical features of the 3D image defined by the CDI to a 2D plane corresponding to the FDI, according to the pose for generating a synthetic view, wherein the indication of the of the one or more structures of the cardiac conduction system comprises rendering the one or more structures of the cardiac conduction system in the synthetic view according to the locations of the plurality of gross anatomical features.

[0092] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and providing an indication based on the one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0093] According to some embodiments of the present invention, there is provided a system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, and providing an indication based on the one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

[0094] Optionally, the indication based on the one or more structures of the cardiac conduction system comprises rendering an interpolation of a non-trivial geometry of the cardiac conduction system between detected locations of the one or more structures.

[0095] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving an image depicting a heart of a subject, feeding the image into a machine learning model, and obtaining from the machine learning model, a fused image comprising the image and a plurality of structures of the cardiac conduction system.

[0096] According to some embodiments of the present invention, there is provided a system for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving an image depicting a heart of a subject, feeding the image into a machine learning model, and obtaining from the machine learning model, a fused image comprising the image and a plurality of structures of the cardiac conduction system.

[0097] Optionally, the machine learning model is trained on a training dataset of a plurality of records generated from a plurality of sample 3D images of sample hearts of sample subjects, for creating a training dataset of a plurality of records, wherein a record includes the sample 3D image, and a ground truth of an indication of the cardiac conduction system generated for the sample 3D image, and training a machine learning model on the training dataset.

[0098] Optionally, the indication of the cardiac conduction system is generated for the sample 3D image by: analyzing the sample 3D image for detecting a plurality of gross anatomical features, computing correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features based on an analysis of a plurality of other 3D images of a plurality of other subjects representing anatomical variations, and generating the indication of the cardiac conduction system according to the computed correlations.

[0099] According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting at least one gross anatomical feature, identifying a location of a membranous septum according to the at least one gross anatomical feature, and generating an indication of the cardiac conduction system of the subject according to the location of the membranous septum, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0100] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting at least one gross anatomical feature, identifying a location of a membranous septum according to the at least one gross anatomical feature, and generating an indication of the cardiac conduction system of the subject according to the location of the membranous septum, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject. Optionally, the indication includes a visual element denoting the perforating bundle, bundle of His and a branching point of the LBB are located according to the membranous septum, the perforating bundle located at an anterior location of the membranous septum, the branching point of the LBB located at a posterior point location of the membranous septum, and the bundle of His connecting the perforating bundle and the branching point of the LBB.

[0101] Optionally, the membranous septum is identified as a region location at a minimum thickness of a septum between a left ventricle and a right ventricle, wherein the region has a predetermined area corresponding to a predicted area of the membranous septum.

[0102] Optionally, the membranous septum is identified by: identifying a mitral ring of a left ventricle myocardium connected to an mitral valve, identifying a tricuspid ring of a right ventricle myocardium connected to a tricuspid valve, wherein a plane of the mitral ring is at an angle with respect to a plane of the tricuspid ring, identifying the membranous septum at a bump between the mitral ring and the tricuspid ring.

[0103] Optionally, the membranous septum is found by identifying the location of the membranous septum of a plurality of sample images of a plurality of sample individuals with respect to the at least one gross anatomical feature, clustering the locations with respect to the at least one gross anatomical feature, and applying the clustering to the at least one gross anatomical feature of the 3D image.

[0104] Optionally, the membranous septum is found by segmenting a myocardium of a left ventricle, and identifying an aortic notch in the myocardium, wherein the membranous septum is located at the aortic notch.

[0105] According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating an indication of the cardiac conduction system of the subject by adapting a template according to the plurality of gross anatomical features, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0106] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating an indication of the cardiac conduction system of the subject by adapting a template according to the plurality of gross anatomical features, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0107] Optionally, the template comprises an image of at least a left bundle branch, an anterior fascicle, a posterior fascicle, and a septal fascicle.

[0108] Optionally, adapting comprises fitting a plurality of anchor points according to the plurality of gross anatomical features, and adapting connections between the plurality of anchor points, and / or re-sizing the template to fit the plurality of gross anatomical features.

[0109] According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating a 3D indication of the cardiac conduction system of the subject according to the plurality of gross anatomical features, wherein the 3D indication is defined according to the 3D image, receiving a 2D image of the heart, registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image, projecting the 3D indication of the cardiac conduction system to the 2D plane for creating a 2D projection of the indication of the cardiac conduction system, and presenting the 2D projection on the 2D image.

[0110] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating a 3D indication of the cardiac conduction system of the subject according to the plurality of gross anatomical features, wherein the 3D indication is defined according to the 3D image, receiving a 2D image of the heart, registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image, projecting the 3D indication of the cardiac conduction system to the 2D plane for creating a 2D projection of the indication of the cardiac conduction system, and presenting the 2D projection on the 2D image.

[0111] Optionally, the registration is performed using features that are different than the plurality of gross anatomical features used to generate the 3D indication of the cardiac conduction system.

[0112] Optionally, further comprising: computing correlations between a calcification template of the 3D image and the 3D indication of the cardiac conduction system, identifying the calcification template in the 2D image, and registering the 2D image with a 2D plane defining a projection of the 3D image according to the calcification template. According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating a 3D indication of the cardiac conduction system of the subject according to the plurality of gross anatomical features, wherein the 3D indication is defined according to the 3D image, computing correlations between a calcification template of the 3D image and the 3D indication of the cardiac conduction system, receiving a 2D image of the heart, identifying the calcification template in the 2D image, and computing a 2D indication of the cardiac conduction system for the 2D image according to a registration function that registers the calcification template in the 2D image to the calcification template of the 3D image.

[0113] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, analyzing the 3D image for detecting a plurality of gross anatomical features, generating a 3D indication of the cardiac conduction system of the subject according to the plurality of gross anatomical features, wherein the 3D indication is defined according to the 3D image, computing correlations between a calcification template of the 3D image and the 3D indication of the cardiac conduction system, receiving a 2D image of the heart, identifying the calcification template in the 2D image, and computing a 2D indication of the cardiac conduction system for the 2D image according to a registration function that registers the calcification template in the 2D image to the calcification template of the 3D image.

[0114] According to some embodiments of the present invention, there is provided a computer implemented method of computing an area of a conduction system at risk during a transcatheter procedure on a subject, comprising: analyzing a combination of at least one personal parameter of the subject and at least one procedure parameter of the transcatheter procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure, and providing the area of the conduction system at risk.

[0115] According to some embodiments of the present invention, there is provided a system for computing an area of a conduction system at risk during a transcatheter procedure on a subject, comprising: at least one processor executing a code for: analyzing a combination of at least one personal parameter of the subject and at least one procedure parameter of the transcatheter procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure, and providing the area of the conduction system at risk. Optionally, the area of the conduction system at risk is presented as an overlay on 2D images captured during the transcatheter procedure.

[0116] Optionally, the area of the conduction system at risk is determined on a 3D image obtained pre-procedure, and further comprising projecting the area of the conduction system to a 2D plane registered with a 2D image obtained during the procedure, wherein a pose of the 2D plane substantially matches a pose of the 2D image.

[0117] Optionally, the analyzing is performed by feeding the combination of at least one personal parameter of the subject and at least one procedure parameter of the transcatheter procedure into a machine learning model, and obtaining the area of the conduction system at risk as an outcome of the machine learning model, wherein the machine learning model is trained on a training dataset of records, each record including the combination of at least one personal parameter of a sample individual, and at least one procedure parameter of the transcatheter procedure performed on the sample individual, and a ground truth indication of the conduction system at risk.

[0118] Optionally, the at least one personal parameter includes one or more of: subject demographics, mapping data including a location of the conduction system computed using a 3D image obtained pre-procedure, patient history, and ECG conduction analysis.

[0119] Optionally, the at least one procedure parameter includes one or more of: procedure related incidence of conduction disturbances, procedure specific tools related incidence of conduction disturbances, type and size of the tools used in the procedure, indication of stress-strain of the tools being used (e.g., extracted from imaging during procedure), indication of the stress-strain of the tools being used extracted from a strain stress sensor on the tool itself during procedure, and ECG conduction analysis during the procedure.

[0120] Optionally, the area of the conduction system at risk includes one or more of: sinoatrial (SA) node, atrioventricular (AV) node, the non bifurcating bundle including the bundle of His, the penetrating bundle, the bifurcating bundle, the right bundle, the left bundle, and / or the branches of the left bundle.

[0121] Optionally, 2D images captured during the transcatheter procedure are dynamically fed into the machine learning model for monitoring the area of the conduction system at risk.

[0122] Optionally, further comprising determining a risk level for at least one region of the area of the conduction system, and visually coding the at least one region according to the risk level.

[0123] Optionally, further comprising monitoring that a pressure applied by a transcatheter device to the area of the conduction system at risk remains below a threshold.

[0124] Optionally, further comprising monitoring that a distance between a transcatheter device and the area of the conduction system remains below a distance threshold. Optionally, further comprising monitoring that an accessory catheter is not present during certain parts of the procedure that may damage the area of the conduction system at risk and / or that the accessory catheter is not placed in proximity to the area of the conduction system at risk where the accessory catheter can cause damage.

[0125] Optionally, further comprising generating a recommendation for a safe location for placement of the accessory catheter for not damaging or reducing risk of damage to the area of the conduction system at risk, below a risk threshold.

[0126] Optionally, further comprising generating, in response to the monitoring of the accessory location, a first indication of safe expansion of a THV, or a second indication of unsafe expansion of the THV.

[0127] Optionally, further comprising in response to detecting a location of a transcatheter device likely to damage the area of the conduction system at risk, generating instructions for a mini-robot that is attached to the transcatheter device, for blocking further movement of the transcatheter device.

[0128] Optionally, further comprising computing a target range for a device depth for deployment of a THV according to the area of the conduction system at risk.

[0129] Optionally, transcatheter procedure is selected from a group including: TAV repair, Mitral Valve repair, Tricuspid valve repair, TAV replacement, Mitral Valve replacement, and Tricuspid valve replacement.

[0130] According to some embodiments of the present invention, there is provided a computer implemented method of displaying an area of a conduction system at risk during a transcatheter procedure on a subject, comprising: analyzing a 3D image obtained pre-procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure, and presenting the conduction system at risk as an overlay on 2D images captured during the transcatheter procedure.

[0131] According to some embodiments of the present invention, there is provided a system for displaying an area of a conduction system at risk during a transcatheter procedure on a subject, comprising: at least one processor executing a code for: analyzing a 3D image obtained preprocedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure, and presenting the conduction system at risk as an overlay on 2D images captured during the transcatheter procedure.

[0132] Optionally, the area of the conduction system at risk includes one or more of: sinoatrial (SA) node, atrioventricular (AV) node, the non bifurcating bundle including the bundle of His, the penetrating bundle, the bifurcating bundle, the right bundle, the left bundle, and / or the branches of the left bundle.

[0133] According to some embodiments of the present invention, there is provided a computer implemented method of reducing risk of conductance disturbance during a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure and monitoring that a pressure applied by a transcatheter device to the area of the conduction system at risk remains below a threshold.

[0134] According to some embodiments of the present invention, there is provided a system for reducing risk of conductance disturbance during a transcatheter procedure on a subject, comprising: at least one processor executing a code for: computing an area of a conduction system at risk during the transcatheter procedure and monitoring that a pressure applied by a transcatheter device to the area of the conduction system at risk remains below a threshold.

[0135] Optionally, the threshold is lOgr.

[0136] According to some embodiments of the present invention, there is provided a computer implemented method of reducing risk of conductance disturbance during a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure and avoiding applying a pressure by a transcatheter device to the area of the conduction system at risk above a threshold.

[0137] According to some embodiments of the present invention, there is provided a system for reducing risk of conductance disturbance during a transcatheter procedure on a subject, comprising: at least one processor executing a code for: computing an area of a conduction system at risk during the transcatheter procedure and avoiding applying a pressure by a transcatheter device to the area of the conduction system at risk above a threshold.

[0138] Optionally, the threshold is 10 gram (gr).

[0139] According to some embodiments of the present invention, there is provided a computer implemented method of guiding a transcatheter procedure on a subject, comprising: analyzing a 3D image obtained pre-procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure, and presenting the conduction system at risk as an overlay on 2D images captured during the transcatheter procedure.

[0140] According to some embodiments of the present invention, there is provided a system for guiding a transcatheter procedure on a subject, comprising: at least one processor executing a code for: analyzing a 3D image obtained pre-procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure, and presenting the conduction system at risk as an overlay on 2D images captured during the transcatheter procedure. According to some embodiments of the present invention, there is provided a computer implemented method of guiding a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure and computing a target range for a device depth for deployment of a THV according to the area of the conduction system at risk.

[0141] According to some embodiments of the present invention, there is provided a system for guiding a transcatheter procedure on a subject, comprising: at least one processor executing a code for: computing an area of a conduction system at risk during the transcatheter procedure and computing a target range for a device depth for deployment of a THV according to the area of the conduction system at risk.

[0142] Optionally, computing an area of a conduction system at risk comprises analyzing a 3D image obtained pre-procedure.

[0143] According to some embodiments of the present invention, there is provided a computer implemented method of guiding a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure, and monitoring that an accessory catheter is not present during certain parts of the procedure that may damage the area of the conduction system at risk and / or that the accessory catheter is not placed in proximity to the area of the conduction system at risk where the accessory catheter can cause damage.

[0144] According to some embodiments of the present invention, there is provided a system for guiding a transcatheter procedure on a subject, comprising: at least one processor executing a code for: computing an area of a conduction system at risk during the transcatheter procedure, and monitoring that an accessory catheter is not present during certain parts of the procedure that may damage the area of the conduction system at risk and / or that the accessory catheter is not placed in proximity to the area of the conduction system at risk where the accessory catheter can cause damage.

[0145] Optionally, computing an area of a conduction system at risk comprises analyzing a 3D image obtained pre-procedure.

[0146] According to some embodiments of the present invention, there is provided a computer implemented method of guiding a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure, and monitoring that a distance between a transcatheter device and the area of the conduction system remains below a distance threshold.

[0147] According to some embodiments of the present invention, there is provided a system for guiding a transcatheter procedure on a subject, comprising: at least one processor executing a code for: computing an area of a conduction system at risk during the transcatheter procedure, and monitoring that a distance between a transcatheter device and the area of the conduction system remains below a distance threshold.

[0148] Optionally, computing an area of a conduction system at risk comprises analyzing a 3D image obtained pre-procedure.

[0149] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a cardiac conduction system of a subject. The method comprises: accessing anatomical data (e.g., an image, such as CT image) of a heart of a subject, detecting multiple gross anatomical features (e.g., by analyzing the anatomical data), estimating locations of structures of the cardiac conduction system (e.g., based on the detected by gross anatomical features, for example: based on correlations between multiple structures of the cardiac conduction system and the gross anatomical features), generating an indication of the cardiac conduction system of the subject and marking the structures of the cardiac conduction system on one or more anatomical images, optionally a 3D image such as a CT scan captured pre-procedure.

[0150] According to some embodiments of the present invention, there is provided a computer implemented method for visualizing a cardiac conduction system of a subject. The method comprising: obtaining an indication of the structures of the cardiac conduction system, obtaining a 2D image such as a fluoroscopy image, and marking and / or presenting the indication of the structures of the cardiac conduction system over the 2D image. The method may be performed during a medical procedure.

[0151] According to some embodiments of the present invention, there is provided a computer implemented method for treating a cardiac conduction system of a subject. The method comprises: registering one or more conduction system structures detected at a prior time (e.g., pre-procedure) to a 2D image obtained during the procedure, and displaying a treatment tool together with said one or more conduction system structures.

[0152] According to some embodiments of the present invention, there is provided a computer implemented method for imaging a cardiac conduction system of a subject. The method comprises: registering one or more conduction system structures detected from a first imaging modality (e.g., 3D scanner) and displaying with a treatment tool detected by a second imaging modality (e.g., 2D imager) together with said one or more conduction system structures, optionally on a 2D image.

[0153] According to some embodiments of the present invention, there is provided a computer implemented method for indicating a subject’s non-parallax viewing angles of one or more structures of a cardiac conduction system and / or a treatment tool.

[0154] According to some embodiments of the present invention, there is provided a computer implemented method for treating a subject’s cardiac conduction system, comprising: registering a structure of the cardiac conduction system denoting a treatment landmark target, optionally to a 2D image, registering a path indicator for the treatment landmark target to the 2D image denoting the path for treatment of the treatment landmark target by a treatment tool, and visualizing the treatment tool together with the target landmark target and path, optionally on the 2D image, such as an overlay over a fluoroscopy image of a heart.

[0155] According to some embodiments of the present invention, there is provided a computer implemented method for treating a subject’s cardiac conduction system, comprising: calculating dimensions of a treatment tool dimensions from a 2D image, optionally a fluoroscopy image, and visualizing said dimensions, optionally on the 2D image.

[0156] According to some embodiments of the present invention, there is provided a computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 2D image depicting a heart of a subject, obtaining an indication of the cardiac conduction system of the subject, and marking the indication of the cardiac conduction system of the subject on the 2D image, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0157] According to some embodiments of the present invention, there is provided a system for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: at least one processor executing a code for receiving a 2D image depicting a heart of a subject, obtaining an indication of the cardiac conduction system of the subject, and marking the indication of the cardiac conduction system of the subject on the 2D image, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0158] Optionally, the indication of the cardiac conduction system is based on correlations with a plurality of gross anatomical features identified on a 3D image of the subject, the correlations are based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations.

[0159] According to some embodiments of the present invention, there is provided a computer implemented method for identifying a gross anatomical structures to define at least one structure of a cardiac conduction system, comprising: receiving a 3D image depicting a heart of a subject, identifying an infrapyramidal space in the 3D image, identifying an inferoseptal recess in the 3D image, identifying a membranous septum in the 3D image, and generating an indication of at least one structure of the cardiac conduction system of the subject according to the infrapyramidal space, the inferoseptal recess, and the membranous septum. According to some embodiments of the present invention, there is provided a system for identifying a gross anatomical structures to define at least one structure of a cardiac conduction system, comprising: at least one processor executing a code for: receiving a 3D image depicting a heart of a subject, identifying an infrapyramidal space in the 3D image, identifying an inferoseptal recess in the 3D image, identifying a membranous septum in the 3D image, and generating an indication of at least one structure of the cardiac conduction system of the subject according to the infrapyramidal space, the inferoseptal recess, and the membranous septum.

[0160] Optionally, the indication of at least one structure of the cardiac conduction system includes an AV node, a Point B denoting a perforating bundle, and a Point C denoting a branching point of the LBB.

[0161] Optionally, the at least one structure of the cardiac conduction system is identified by applying different HUE levels to the 3D image, the different HUE levels selected to visualize different types of tissues of the heart, wherein the infrapyramidal space and the inferoseptal recess are of a first type of tissue at a first HUE level that is different than a myocardium of the heart of a second type of tissue at a second HUE level.

[0162] Optionally, the indication comprises an AV node, and the AV node is identified as being located within the infrapyramidal space and closest to the inferoseptal recess.

[0163] Optionally, the indication comprises Point B, and Point B is identified as being located within the posterior most position of the membranous septum floor.

[0164] Optionally, the indication comprises Point C, and Point C is identified as being located on the membranous septum floor at a point between Point B and the most anterior point of the membranous septum floor.

[0165] Optionally, Point C is located at about 40-80% of a length of the line between Point B and the most anterior point of the membranous septum floor.

[0166] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of embodiments of the invention, exemplary methods and / or materials are described below. In case of conflict, the patent specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and are not intended to be necessarily limiting. BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

[0167] Some embodiments of the invention are herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of embodiments of the invention. In this regard, the description taken with the drawings makes apparent to those skilled in the art how embodiments of the invention may be practiced.

[0168] In the drawings:

[0169] FIG. 1A is flowchart of a method for visualizing a cardiac conduction system of a subject performed prior to a medical procedure, in accordance with some embodiments of the present invention;

[0170] FIG. IB is flowchart of a method for visualizing a cardiac conduction system of a subject performed during a medical procedure, in accordance with some embodiments of the present invention

[0171] FIG. 2 is a block diagram of components of a system for visualizing a cardiac conduction system of a patient, in accordance with some embodiments of the present invention;

[0172] FIG. 3 is a schematic of a template that may be adapted for creating an indication of the cardiac conduction system of a subject, in accordance with some embodiments of the present invention;

[0173] FIG. 4 is an image that includes a heart of a subject and an indication of the cardiac conduction system, in accordance with some embodiments of the present invention;

[0174] FIG. 5 is an image representing another view of the image of FIG. 4, in accordance with some embodiments of the present invention;

[0175] FIG. 6 is an image of a heart depicting a right bundle branch of the cardiac conduction system, in accordance with some embodiments of the present invention;

[0176] FIG. 7 is a schematic of a 2D image of a heart of a subject enhanced with an indication of a cardiac conduction system, in accordance with some embodiments of the present invention;

[0177] FIG. 8 is a schematic of a 2D image representing a different view of the image of FIG. 7, in accordance with some embodiments of the present invention;

[0178] FIG. 9 is a schematic depicting a fluoroscopy image that includes one or more structures of the cardiac conduction system, in accordance with some embodiments of the present invention;

[0179] FIG. 10 is an image depicting an indication of a cardiac conduction system depicted as dots representing predicted locations of structures, in accordance with some embodiments of the present invention; FIGs. 11A-11B is a schematic depicting a process for computing the anatomical features of a membranous septum depth for generating the indication of the cardiac conduction system, in accordance with some embodiments of the present invention;

[0180] FIG. 12 is a schematic depicting an example of a segmentation of a membranous septum from a myocardium of a left ventricle in a 3D image, in accordance with some embodiments of the present invention;

[0181] FIG. 13 is a schematic depicting another exemplary approach for segmentation of a membranous septum from a 3D image, in accordance with some embodiments of the present invention;

[0182] FIG. 14 is a schematic depicting examples of anatomical features used for generating the indication of the cardiac conduction system which may be identified in a 3D image of a heart, in accordance with some embodiments of the present invention;

[0183] FIG. 15 includes schematics depicting an example of a 3D model of a heart and LBB with three proximal fascicles, in accordance with some embodiments of the present invention; and

[0184] FIG. 16 is a pictorial illustration of a catheterization laboratory equipped with a system for visualizing a cardiac conduction system of a patient, in accordance with some embodiments of the present invention.

[0185] DETAILED DESCRIPTION

[0186] The present invention, in some embodiments thereof, relates to cardiac imaging and, more specifically, but not exclusively, to systems and methods for imaging the cardiac conduction system.

[0187] As used herein, the term “image of the cardiac conduction system” or “imaging the cardiac conduction system” may refer to one or more of: locating structures of the cardiac conduction system, identifying structures of the cardiac conduction system, and displaying the structures of the cardiac conduction system on an image,. As used herein, "imaging" refers to the process, method, or system for capturing, processing, generating, and / or rendering information, data, or representations in any dimensional format. This includes but is not limited to one-dimensional (ID), two-dimensional (2D) and three-dimensional (3D). The term is intended to cover both digital and analog formats and is not limited to visual representations displayed on screens but includes any form of data representation.

[0188] As used herein, the structures of the cardiac conduction system may refer to a conduction system axis. In some embodiments, systems and methods for imaging the cardiac conduction system may be used pre-medical procedure (e.g., for planning the procedure) and / or during a medical procedure.

[0189] Medical procedures may include, for example: transcatheter aortic valve implantation (TAVI), percutaneous coronary intervention (PCI), transcatheter mitral valve repair (TMVR), left atrial appendage closure (LAAC), septal defect closure, patent foramen ovale (PFO) closure, and the like.

[0190] As used herein, the term anatomical data and image may sometimes be interchanged.

[0191] As used herein, the term “cardiac conduction system” refers to one or more components of the cardiac conduction system. Examples of components of the cardiac conduction system are provided herein.

[0192] As used herein, the terms anatomical feature and anatomical structures may be interchanged. The terms anatomical features and / or anatomical structures may include and / or may refer to anatomical landmarks.

[0193] As used herein, the terms virtual annulus plane, virtual aortic valve plane, virtual aortic annulus plane, and annulus plane, may be interchanged.

[0194] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject. The personalized surrogate may represent an estimated and / or predicted location of the cardiac conduction system for the particular subject, based on an analysis of determined locations of the cardiac conduction system in multiple other individuals representing anatomical variations. The personalized surrogate may be in contrast to other approaches that directly sense the location of the cardiac conduction system in an individual, for example, very high-resolution imaging devices that directly image the cardiac conduction system, and / or functional approaches that determine the location of the cardiac conduction system by measuring electrical activity generated by the cardiac conduction system.

[0195] Cardiac conduction system may include one of more structures of the cardiac conduction system. Examples of structures of the cardiac conduction system include: Sinoatrial (SA) node, atrioventricular (AV) node, bundle of His, penetrating bundle, branching point of a left branch bundle (LBB), right branch bundle (RBB), left anterior fascicle, left posterior fascicle, left septal fascicle, Mahaim fibers, Purkinje fibers, and connection paths between two of the aforementioned along the cardiac conduction system. A 3D image depicting a heart of a subject is received, for example, a CT (e.g., cardiac CT), MRI, US, and the like. It may be difficult or impossible to visualize the cardiac conduction system of the subject directly in the 3D image. The 3D image is analyzed for detecting multiple gross anatomical features, for example, non-coronary cusp (NCC), right coronary cusp (RCC), left coronary cusp (LCC), nadir of one or more of NCC RCC and LCC, virtual annulus plane, membranous septum, rotation angle of the aorta (e.g., cusp and / or commissure) relative to the membranous septum, ventricular septum, inferoseptal recess, posterior papillary muscle of the left ventricle, anterior papillary muscle of the left ventricle, apex of the left ventricle, medial papillary muscle of the right ventricle, the moderator band of the right ventricle, the anterior papillary muscle of the right ventricle and the Septal Band.

[0196] Correlations between multiple structures of the cardiac conduction system, optionally combinations of structures, and the gross anatomical features, optionally combinations of the gross anatomical features, may be used to identify and / or image the cardiac conduction system.

[0197] Correlations between multiple structures of the cardiac conduction system, optionally combinations of structures, and the gross anatomical features, optionally combinations of the gross anatomical features, are computed based on an analysis of sample 3D images of sample subjects representing anatomical variations. The correlations may be, for example, a vector (i.e., distance and location direction) indicating expected location of one or more structures of the cardiac conduction system relative to one or more gross anatomical features. An indication personalized surrogate of the cardiac conduction system of the subject is according to the computed correlations. The indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject. The indication of the cardiac conduction system may be, for example, a second 3D image fused with the 3D image, a standalone second 3D image (e.g., may be fused with another 3D image and / or another 2D image), a marking of the 3D image, a marking of a 3D model of the heart based on the 3D image, and an overlay for being placed over a 3D and / or 2D image.

[0198] As used herein, the locations of the structures of the cardiac conduction system may be defined by coordinates within a 3D space and / or a 2D plane, for example, x,y and / or z coordinates along the x-axis, y-axis, and / or z-axis. The coordinates of the structures of the cardiac conduction system may be defined within the 3D space defined by a 3D image of the heart, for example, a pre-procedure CT scan. The coordinates may be registered to a 2D image, such as a fluoroscopy image of the heart. The 2D image may be defined within the 3D space of the 3D image, and / or may be mapped to the 3D image, and / or may be defined by a 2D plane that is projected from the 3D space of the 3D image and / or located within the 3D space of the 3D image, and / or mapped from the 3D space of the 3D image. The coordinates within the common space and / or projected between spaces and / or planes and / or mapped between spaces and / or planes enable registering the location of the structures of the cardiac conduction system in the 3D image to the 2D image.

[0199] Alternatively or additionally, the locations of the structures of the cardiac conduction system may be defined relative to anatomical landmarks of the 3D image, for example, relative to the gross anatomical features of the 3D image. For example, an end of a vector having an origin at a certain gross anatomical feature may indicate the location of a certain structure of the cardiac conduction system relative to the certain gross anatomical feature.

[0200] Inventor discovered that the locations of structures of the cardiac conduction system vary between individuals. The indication (i.e., personalized surrogate) of the structures of the cardiac conduction system is designed as a way to visualize where the structures of the cardiac conduction system are expected to be located for each subject.

[0201] Although the locations of individual structures of the cardiac conduction system vary between individuals, since the individual structures make up the cardiac conduction system, the location of the individual structures may vary as a combination of multiple structures. For example, the correlation of a combination of multiple structures of the cardiac conduction system relative to a combination of multiple gross anatomical structures may be more accurate than considering individual structures of the cardiac conduction system.

[0202] Optionally, one or more of the computed correlations between at least one of the gross anatomical features and at least one of the structures of the cardiac conduction system are adapted based on a current medical state of the subject. Examples of current medical states include: a genetic condition, and an acquired disease condition. The current medical state may include an indication of presence or absence of one or more of known genetic and / or acquired disease conditions. For example, in patients with congestive heart failure and dilated cardiomyopathy, the predictive correlation of one or more of the structures of the cardiac conduction system may be adapted. For example, the correlation between the location of the posterior left ventricle papillary muscle and the direction of the left posterior fascicle is lost compared to normal hearts. In another example, in the case of transposition of heart vessels (congenital heart condition), the correlation of the membranous septum to the location of the bundle of His is attenuated.

[0203] Optionally, a machine learning model is dynamically trained on collected published and / or observed cases. The cases may be obtained, for example, from feeds from medical journals, accessing websites publishing new cases online, and the like. The cases describe variations in deviations of locations of structures of the cardiac conduction system for different medical states of individuals. A record may be dynamically created by analyzing each case. The record may include the current medical state of the individual in the case, and a ground truth of the deviation of the location of the structure(s) of the cardiac conductions system, such as relative to anatomy of normal hearts. The computed correlations and / or the current medical state of the subject may be fed into the trained machine learning model. The adaptation of the computed correlation may be obtained from the machine learning model. The adapted correlation may be used for generating the visualization of the structures of the cardiac conduction system, as described herein.

[0204] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting multiple gross anatomical features. Locations of one or more structures of the cardiac conduction system are estimated using the gross anatomical features. The location may be estimated by applying one or more rules derived from an analysis of hearts of different individuals. For example, the location of a certain structure may be located at an intersection between a first vector from a first gross anatomical feature and a second vector from a second gross anatomical features, where the two vectors are computed as an average of from multiple hearts of different individuals. Locations of the structures of the cardiac conduction system are presented on a visual rendition of the heart, for example, a 3D model, 3D image, 2D image, and the like. An indication of one or more structures of the cardiac conduction system in a 3D image of the heart is provided using the gross anatomical features. For example, the structures are marked using an arrow, an overlay representing the structures is overlaid over the 3D image, and a text description of the structures is generated and presented.

[0205] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting multiple gross anatomical features. Locations of one or more structures of the cardiac conduction system are estimated in the 3D image using the gross anatomical features. Locations of the structures of the cardiac conduction system may be marked on a 2D image of the heart. The locations may be marked on the 2D image which may be received in real time or near real time during a medical procedure, by registering the 2D image to the 3D image, mapping the locations from the 3D image to the 2D image, and updating a display presenting the 2D image with mapped location in real time or near real time during the medical procedure. In another example, the 2D image may be registered to a 2D projection of the 3D image at a plane having a pose corresponding to the pose of the 2D image (e.g., pose of sensor that captured the 2D image). The locations on the 3D image may be projected to the plane, and mapped to the 2D image registered to the plane.

[0206] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting multiple gross anatomical features. Locations of one or more structures of the cardiac conduction system are estimated in a 2D image of the heart using the gross anatomical features.

[0207] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting multiple gross anatomical features. An indication of one or more structures of the cardiac conduction system in a 2D image of the heart is provided using the gross anatomical features. For example, the structures are marked using an arrow, an overlay representing the structures is overlaid over the 2D image, and a text description of the structures is generated and presented.

[0208] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting multiple gross anatomical features. An indication based on the structures of the cardiac conduction system is provided in a 2D image of the heart using the gross anatomical features.

[0209] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. An image depicting a heart of a subject is received. The image may be a 3D image and / or a 2D image. The image is fed into a machine learning model. A fused image including the image and structures of the cardiac conduction system is obtained from the ML model. The ML model may be trained on records generated from sample images of sample hearts of sample subjects, for creating a training dataset of records. A record includes the sample image, and a ground truth of an indication of the cardiac conduction system generated for the sample image, generated using one or more embodiments described herein.

[0210] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting one or more gross anatomical features. A location of a membranous septum is identified (e.g., detected, segmented) according to the at least one gross anatomical feature. Exemplary approaches for identifying the location of the membranous septum are described herein. An indication of the cardiac conduction system of the subject is generated according to the location of the membranous septum. For example, a visual indication of the bundle of His and / or the branching point of the left bundle branch into the fascicles (anterior, posterior, septal) is generated corresponding to the location of the membranous septum.

[0211] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for visualizing a personalized surrogate of a cardiac conduction system of a subject. A 3D image depicting a heart of a subject is received. The 3D image is analyzed for detecting multiple gross anatomical features. An indication of the cardiac conduction system of the subject is generated by adapting a template according to the gross anatomical features. The template may include an image of at least a left bundle branch, an anterior fascicle, a posterior fascicle, and a septal fascicle. The template may be adapted by fitting anchor points according to the gross anatomical features, and / or adapting connections between the anchor points, and / or re-sizing the template to fit the gross anatomical features.

[0212] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for creating a fused image (e.g., 3D image) that includes a 3D image depicting structures of the cardiac conduction system of a subject registered with functional data (e.g., functional image) of the heart. The 3D image depicting the structures of the cardiac conduction system is created according to one or more embodiments described herein. The functional data may indicate functional status of the heart, for example, scar tissue location, viability of tissues of the heart, tissues suffering from ischemia, blood perfusion to tissues of the heart, and the like. The functional data of the heart may be acquired by an imaging modality (e.g., CT scanner, MRI) used to generate the image data that includes the gross anatomical features used to identify locations of the structures of the cardiac conduction system. Alternatively or additionally, the heart functional data may be generated using a second imaging technology, for example, perfusion scan, ultrasound images, and the like. Alternatively or additionally, the functional data of the heart may be derived, for example, from body surface electrocardiogram (ECG) electrodes measuring ECG signals, vectorcardiography (VCG) analysis, invasive electrophysiology (EP) testing, electrical system imaging studies (e.g., body surface electrical mapping), and the like. The functional data of the heart may indicate status of the conduction path of the cardiac conduction system. For example, presence of: a left anterior conduction block, left posterior conduction block, left bundle branch block, left septal branch block, and right bundle branch block.

[0213] An aspect of some embodiments of the present invention relates to systems, methods, computing device, and code instructions (stored on a memory and executable by one or more processors) for identifying a gross anatomical structures to define at least one structure of a cardiac conduction system, comprising: receiving a 3D image depicting a heart of a subject, identifying an infrapyramidal space in the 3D image, identifying an inferoseptal recess in the 3D image, identifying a membranous septum in the 3D image, and generating an indication of at least one structure of the cardiac conduction system of the subject according to the infrapyramidal space, the inferoseptal recess, and the membranous septum. Optionally, the indication of at least one structure of the cardiac conduction system includes an AV node, a Point B denoting a perforating bundle, and a Point C denoting a branching point of the LBB. The AV node, Point B, and Point C may be identified based on the membranous septum floor, the infrapyramidal space, and the inferoseptal recess. Optionally, the at least one structure of the cardiac conduction system is identified by applying different HUE levels to the 3D image. The different HUE levels are selected to visualize different types of tissues of the heart. The infrapyramidal space and the inferoseptal recess are of a first type of tissue at a first HUE level that is different than a myocardium of the heart of a second type of tissue at a second HUE level. Optionally, the indication comprises an AV node, and the AV node is identified as being located within the infrapyramidal space and closest to the inferoseptal recess. Optionally, the indication comprises Point B, and Point B is identified as being located within the posterior most position of the membranous septum floor. Optionally, the indication comprises Point C, and Point C is identified as being located on the membranous septum floor at a point between Point B and the most anterior point of the membranous septum floor. Optionally, Point C is located at about 40-80%, or about 60-80%, or about 50-70%, or other values, of a length of the line between Point B and the most anterior point of the membranous septum floor.

[0214] Optionally, the indication of the cardiac conduction system is generated excluding (i.e., without using) data from at least one sensor sensing electrical activity of the cardiac conduction system. Examples of sensors sensing electrical activity of the cardiac conduction system include ECG sensors, and / or electrophysiology (EPS) sensors. Data from ECG and / or EPS sensors is not required to visualize structures of the cardiac conduction system according to at least one embodiment described herein. At least some embodiments described herein address the technical problem of improving imaging of the cardiac conduction system of the heart. The cardiac conduction system is difficult or cannot be directly visualized on images. Approaches such as ECG and EPS sense electrical activity of the cardiac conduction system but do provide direct images of the cardiac conduction system, in particular when functionality is impaired. At least some embodiments described herein improve the technology of imaging of the cardiac conduction system of the heart. At least some embodiments described herein improve upon prior approaches of imaging the cardiac conduction system of the heart. At least some embodiments described herein improve the technology of imaging of the cardiac conduction system of the heart pre-procedure, e.g., for planning the procedure for the specific patient. At least some embodiments described herein improve the technology of imaging of the cardiac conduction system of the heart during the medical procedure, e.g., for imaging the conduction system of the patient in real-time.

[0215] At least some embodiments described herein provide solutions for the aforementioned technical problem, and / or improve the aforementioned technical field, and / or improve upon aforementioned prior approaches, by accessing a 3D image of a heart of a subject, for example, a CT (e.g., cardiac CT) and / or MRI. The 3D image may be obtained pre-procedure. The 3D image is analyzed to identify multiple gross anatomical features. The location of one or more structures of the cardiac conduction system may be inferred according to relationships between the gross anatomical features and known locations of the structures of the cardiac conduction system in patients, for example, by clustering measurements made on images of multiple patients to obtain a statistical relationship between structures of the cardiac conduction system and the anatomical features. A visual indication is generated at the location(s) corresponding to the predicted / expected location(s) of the structures of the cardiac conduction system. The visual indication may be on the 3D image. The 3D image marked with locations of the structures of the cardiac conduction system may be projected to a 2D plane corresponding to view of a 2D image (e.g., fluoroscopy image) captured during the procedure. A realtime 2D image (e.g., fluoroscopy image) marked-up with a visualization of the location(s) of structure(s) of the cardiac conduction system may be created.

[0216] Potential advantages of at least one embodiment described herein that generates a visualization of the cardiac conduction system includes:

[0217] • Enables pre-procedure planning according to locations of one or more structures of the cardiac conduction system, for example, where a distal end of an aortic valve prosthesis device may be safely positioned during a TAVI procedure for reducing or preventing damage to the structures of the cardiac conduction system. • The visualization of the cardiac conduction system is done non-invasively, for example, in contrast to other approaches such as EPS that rely on in-procedure mapping using electrodes on catheters.

[0218] • The visualization of the cardiac conduction system may be of the structure of the cardiac conduction system, for example, in contrast to approaches that image the cardiac conduction system based on its function. For example, the conduction path from the SA node and / or AV node and / or bundle of His to the end of the anterior, posterior, and / or septal fascicles (e.g., start of the Purkinje fibers) may be visualized, even when regions of the conduction path have been damaged and do not conduct or have non-significant conduction, such as due to a myocardial infarction that causes the conduction nerves to die in certain regions.

[0219] • Adding (e.g., fusing) functional data of the heart to the visualization (e.g., image) of the structures of the cardiac conduction system enables presenting a combination of anatomical and functional imaging data on a display (e.g., to an operator), for example, for planning of optimal therapy for a patient, and / or for aiding in diagnosis.

[0220] • Real-time overlays of the location of the conduction system on 2D images of the heart may be generated. For example, fluoroscopy image of the heart captured during a transcatheter procedure may be enhanced to visualize the location of the structures of the cardiac conduction system, taking into account the view at which the fluoroscopy image is captured.

[0221] Before explaining at least one embodiment of the invention in detail, it is to be understood that the invention is not necessarily limited in its application to the details of construction and the arrangement of the components and / or methods set forth in the following description and / or illustrated in the drawings and / or the Examples. The invention is capable of other embodiments or of being practiced or carried out in various ways.

[0222] The present invention may be a system, a method, and / or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.

[0223] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0224] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0225] Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.

[0226] Aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer readable program instructions.

[0227] These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function / act specified in the flowchart and / or block diagram block or blocks.

[0228] The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0229] The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustration, and combinations of blocks in the block diagrams and / or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0230] Reference is now made to FIG. 1 A, which is flowchart of a method for visualizing a cardiac conduction system of a subject (also referred to herein as a personalized surrogate of a cardiac conduction system of a subject) performed prior to a medical procedure (also referred to herein as a method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject), in accordance with some embodiments of the present invention. Reference is also made to FIG. IB, which is flowchart of a method for visualizing a cardiac conduction system of a subject (also referred to herein as a personalized surrogate of a cardiac conduction system of a subject) performed during a medical procedure (also referred to herein as method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject), in accordance with some embodiments of the present invention. Reference is also made to FIG. 2, which is a block diagram of components of a system 100 for visualizing a cardiac conduction system of a patient, in accordance with some embodiments of the present invention. Reference is also made to FIG. 3, which is a schematic of a template 402 that may be adapted for creating an indication of the cardiac conduction system of a subject, in accordance with some embodiments of the present invention. Reference is also made to FIG. 4, which is an image that includes a heart of a subject and an indication of the cardiac conduction system, in accordance with some embodiments of the present invention. Reference is also made to FIG. 5, which is an image 602 representing another view of image 502 of FIG. 4, in accordance with some embodiments of the present invention. Reference is also made to FIG. 6, which is an image 650 of a heart depicting a right bundle branch 604 of the cardiac conduction system, in accordance with some embodiments of the present invention. Reference is also made to FIG. 7, which is a schematic of a 2D image 702 of a heart 704 of a subject enhanced with an indication of a cardiac conduction system 706, in accordance with some embodiments of the present invention. Reference is also made to FIG. 8, which is a schematic of a 2D image 802 representing a different view of image 702 of FIG. 7, in accordance with some embodiments of the present invention. FIG. 9 is a schematic depicting a fluoroscopy image that includes one or more structures of the cardiac conduction system, in accordance with some embodiments of the present invention. Reference is also made to FIG. 10, which is an image 1002 depicting an indication of a cardiac conduction system depicted as dots representing predicted locations of structures , in accordance with some embodiments of the present invention. Reference is also made to FIGs. 11A-11B, which is a schematic depicting a process for computing the anatomical features of a membranous septum depth for generating the indication of the cardiac conduction system, in accordance with some embodiments of the present invention. Reference is also made to FIG. 12, which is a schematic depicting an example of a segmentation of a membranous septum from a myocardium of a left ventricle 1304 in a 3D image, in accordance with some embodiments of the present invention. Reference is also made to FIG. 13, which is a schematic 1402 depicting another exemplary approach for segmentation of a membranous septum from a 3D image, in accordance with some embodiments of the present invention. Reference is also made to FIG. 14, which is a schematic depicting examples of anatomical features used for generating the indication of the cardiac conduction system which may be identified in a 3D image of a heart, in accordance with some embodiments of the present invention. Reference is also made to FIG. 15 which includes schematics depicting an example of a 3D model of a heart and LBB with three proximal fascicles, in accordance with some embodiments of the present invention. Reference is also made to FIG. 16, which is a pictorial illustration of a catheterization laboratory equipped with a system for visualizing a cardiac conduction system of a patient, in accordance with some embodiments of the present invention.

[0231] Referring now back to FIG. 1A, at 302, anatomical data of a heart of a subject is accessed. The heart is beating inside a living subject. This is in contrast to approaches using hearts of cadavers which are not beating, and may be dissected.

[0232] The anatomical data may be, for example, anatomical images such as a 3D image, a model (e.g., 3D, 2D) which may be created based on the image, segmentations of the image, and the like. The anatomical image may be captured, for example, by a CT machine (e.g., standard, dual energy) using one or more protocols, for example, cardiac CT, delayed enhancement, and the like. Other examples of anatomical images include images captured by an MRI device, an x-ray device, and an ultrasound device.

[0233] The anatomical data may exclude visualization of the structures of the cardiac conduction system. The structures of the cardiac conduction system cannot be directly detected and / or segmented from the anatomical data.

[0234] As used herein, the term image is an exemplary implementation of the anatomical data, and is not meant to be necessarily limiting. Other implementations of the anatomical data may be substituted for the term image.

[0235] As used herein, the term visual rendition may refer to, and / or be interchanged with, the term: image, model, anatomical data, and others described herein.

[0236] The image may be based on the non-visual electromagnetic spectrum, i.e., different than images captured in the visible spectrum. The 3D image may be processed, for example, using image processing approaches. The image processing approaches may include performing HUE filtering of the 3D image (e.g., cardiac CT) and generating 3D image (e.g., CT images) of values of HUE that corresponds with different tissues, for example, fat, fibrous, and connective tissue. Some different gross anatomical features may be differentiate due to being different types of tissues depicted by applying filtering at different HUE values. Optionally, the infrapyramidal space and / or the inferoseptal recess are of a different tissue type than the rest of the heart, such as the heart myocardium. The infrapyramidal space and the inferoseptal recess are of a first type of tissue that may be better visualized at a first HUE level, that is different than a myocardium of the heart of a second type of tissue that is better visualized at a second HUE level. For example, selecting a window of HUE of values below 30 HUE or below 0 HUE. In another example, multiple images of different HUE windowing may be generated. In yet another example, a dual beam CT image may be obtained, and different HUE may be selected for each beam depending on the beam energy. In yet another example, the dual beam CT image may be generated using a ratio between the two energy values of the corresponding voxels. In yet another example, a conventional cardiac CT is processed into multiple HUE windows, and a derived ratio formula that enhances the difference between the aforementioned mentioned tissues and the rest of the myocardium and the blood within it may be used.

[0237] At 304, image is analyzed for detecting multiple gross anatomical features.

[0238] The gross anatomical features may include anatomical features.

[0239] The gross anatomical features may be visualized in the image. The gross anatomical features may be directly detected and / or segmented from the image.

[0240] The gross anatomical features may be detected and / or segmented using machine learning methods (e.g., detector trained on images labelled with a ground truth indicating the gross anatomical feature) and / or using image processing approaches (e.g., based on contour, visual features, edges, patterns, and the like).

[0241] Examples of gross anatomical features that may be detected include: non-coronary cusp (NCC), right coronary cusp (RCC), left coronary cusp (LCC), nadir of one or more of NCC RCC and LCC, virtual annulus plane (also referred to herein as virtual aortic valve plane, or virtual aortic annulus plane), membranous septum, membranous septum floor (i.e., the bottom part of the membranous septum), membranous septum depth, angle of rotation (e.g., between a location on the aortic valve and the septum such as the membranous septum), ventricular septum, inferoseptal recess, posterior papillary muscle of the left ventricle, anterior papillary muscle of the left ventricle, apex of the left ventricle, medial papillary muscle of the right ventricle, the moderator band of the right ventricle, the anterior papillary muscle of the right ventricle, septal band, intraventricular septum, myocardial ring connecting to aortic valve, myocardial ring connecting to mitral valve, myocardial notch between plane of myocardial rings of the aortic valve and mitral valve, base (i.e., inferior border) of the infra pyramidal space (IPS), apex of the IPS, anterior Border of the IPS, posterior Border of the IPS, lateral borders of the IPS, the roof of the inferoseptal recess, anterior / superior border of the inferoseptal recess, posterior / inferior border of the inferoseptal recess, right side of the inferoseptal recess, left side of the inferoseptal recess, diverging vestibules of the mitral and tricuspid valves, the ventricular boundary of the inferoseptal recess, the apex of the inferoseptal recess, medial papillary muscle (MPM), septal band (SB), moderator band (MB), muscle between septum and free wall (myocardium), MB1 - connected to the interventricular septum, and MB2 - connected to the ventricular myocardium.

[0242] The membranous septum and / or the membranous septal floor may be identified (e.g., segmented) using one or more approaches described herein. Optionally, the membranous septum is identified by searching the most proximal endocardium of right ventricle and left ventricle area, as described herein in detail.

[0243] The membranous septal floor may be identified, for example, by drawing a highest elevation line of the myocardium and identifying a crossing of the identified membranous septum and the highest elevation line of the myocardium.

[0244] Optionally, the infrapyramidal space (IPS) and / or the inferoseptal recess (ISR) may be identified and / or segmented after the membranous septum and / or membranous septum floor is identified. Exemplary approaches for identifying and / or segmenting the IPS and / or ISR are described herein.

[0245] Exemplary approaches for computing the anatomical features of membranous septum depth, which may be used for generating the indication of the cardiac conduction system (e.g., using correlations) are now described.

[0246] Referring now back to FIGs. 11A-11B, schematics 1202 and 1212 indicate a computed membranous septum depth, computed as a distance between a membranous septum to a virtual annulus plane, which may be used for generating the indication of the cardiac conduction system. Schematic 1202 depicts a membranous septum depth 1204 of 7.06 millimeters (mm). Membranous septum depth 1204 is defined between a virtual annulus plane (shown as a line) 1206 and membranous septum 1208. Annulus line 1206 may be defined as a plane / line that inserts three nadirs of the leaflets of the aortic valve. Membranous septum depth 1208 may be defined as a line between an anterior position 1250 on the membranous septum and a posterior position 1252 on the membranous septum, representing the line between the perforating bundle 1250 (also labelled point B) and the branching point of the LBB 1252 (also labelled point C). It is noted that perforating bundle 1250 and branching point of the LBB 1252 may be included in the structures of the cardiac conduction system which are being visualized using at least one embodiment described herein. Schematic 1212 depicts a membranous septum depth 1214 of -0.92 mm. It is noted that in schematic 1212, annulus line 1216 is below membranous septum 1218.

[0247] An exemplary approach for identifying the infra pyramidal space (IPS) is now described. The inferior pyramidal space is a pyramid- shaped fibrofatty structure located on the diaphragmatic (inferior) surface of the heart. It wedges between the four cardiac chambers from the diaphragmatic surface. The IPS may be identified by identifying differentiated tissue content on the 3D image, i.e., fat and fibers opposite to surrounding muscle. In an example, by identifying low HUE epicardial invagination volume of the 3D image implemented as a cardiac CT. In another example, by identifying high signal intensity on T1 weighted images of a cardiac MRI.

[0248] The IPS may be identified by detecting one or more of the following neighboring anatomical landmarks (which may also referred to as gross anatomical features):

[0249] • Base (i.e., inferior border) of the IPS - The base of the IPS may be identified by detecting a diaphragmatic surface of the heart. This is the bottom part of the pyramidshaped space.

[0250] • Apex of the IPS - The apex of the IPS may be identified by detecting a central fibrous body of the atrioventricular septum, close to the noncoronary sinus of the aortic root. The apex is in proximity to the central fibrous body.

[0251] • Anterior Border of the IPS - The anterior border of the IPS may be identified by detecting a crest of the interventricular septum. The anterior border is formed by the crest.

[0252] • Posterior Border of the IPS - The posterior border of the IPS may be identified as being continuous with a detected diaphragmatic surface of the heart at the crossing of the left and right inferior atrioventricular and interatrial grooves. The IPS may be identified as being continuous with the left and right inferior atrioventricular and interatrial grooves at their intersection on the diaphragmatic surface.

[0253] • Lateral Borders of the IPS - The lateral borders of the IPS may be identified by detecting the two septal atrial walls. The lateral borders are defined as the two septal atrial walls.

[0254] Exemplary approaches for identifying (e.g., segmenting, detecting) the membranous septum, are now described. Additional exemplary approaches for identifying the membranous septum are described, for example, with reference to U.S. Provisional Patent Application No. 63 / 656,164, filed on June 5, 2024, “PERSONALIZED SAFE ZONE FOR A TA VI PROCEDURE”, and / or U.S. Provisional Patent Application No. 63 / 531,063, filed on August 7, 2023, “SYSTEM AND METHODS FOR SUPPORTING A TA VI PROCEDURE”, incorporated herein by reference in their entirety.

[0255] The membranous septum described herein may refer to the portion of the membranous septum between the ventricular septum and the inferoseptal recess. The membranous septum may be identified by locating the ventricular septum and the inferoseptal recess.

[0256] The inferoseptal recess may be identified by differentiating tissue content within the inferoseptal recess from the rest of the adjacent tissue, i.e., fibrous tissue within the inferoseptal recess versus adjacent muscle and / or blood, as depicted in the 3D image.

[0257] Alternatively or additionally, the inferoseptal recess may be identified based on identifying neighboring anatomical landmarks (which may also referred to as gross anatomical features), such as one or more of:

[0258] • The roof of the inferoseptal recess. The roof may be identified as an area of fibrous continuity between the leaflets of the mitral and tricuspid valves. This fibrous tissue may also be identified as the tissue at (i.e., for supporting) the base of the atrial septum.

[0259] • Anterior / Superior border of the inferoseptal recess: The anterior / superior border may be identified as the aortic leaflet of the mitral valve.

[0260] • Posterior / Inferior border of the inferoseptal recess: The posterior / inferior border may be identified as the septal surface of the left ventricle.

[0261] • Right side of the inferoseptal recess: The right wall of the inferoseptal recess may be identified as fibrous tissue through which the identified conduction axis (i.e., His bundle) penetrates.

[0262] • Left side of the inferoseptal recess: The left side of the inferoseptal recess may be identified as a border of the left ventricular cavity.

[0263] • Diverging vestibules of the mitral and tricuspid valves. The inferoseptal recess is bounded on the atrial side by the diverging vestibules of the mitral and tricuspid valves. Detecting the diverging vestibules may be used to define the atrial boundary of the infero septal recess. These vestibules are part of the atrial anatomy and play a crucial role in the function of the atrioventricular valves.

[0264] • The ventricular boundary of the inferoseptal recess is formed by the crest of the muscular ventricular septum. Detecting the crest of the muscular ventricular septum may be used to define the ventricular boundary. This septum is the thick, muscular wall that separates the left and right ventricles of the heart. * The apex of the inferoseptal recess may be detected and may be used to identify the inferior margin of the membranous septum. Alternatively or additionally, the inferior margin of the membranous septum may be detected and used to identify the apex of the inferoseptal recess. At its apex, the inferoseptal recess is marked by the inferior margin of the membranous septum. This region of the membranous septum typically acts as an atrioventricular partition at this level and is often overlaid by a layer of atrial vestibular myocardium. This layer separates the apex of the pyramidal space from the inferoseptal recess within the left ventricle

[0265] The membranous septum may be segmented from the 3D image. Segmenting / detecting the membranous septum is technically challenging due to the small size of the membranous septum and / or difficulty in visualizing the membranous septum in medical images.

[0266] The septum is a structure within the heart that separates the two lower chambers, or ventricles: the right ventricle and the left ventricle. The heart is divided into four chambers — two atria (upper chambers) and two ventricles (lower chambers) — and the septum serves to keep the oxygenated blood on the left side of the heart from mixing with the deoxygenated blood on the right side. There are two main structures of the septum: the membranous septum and the muscular septum. The membranous septum is a thin, fibrous partition that lies between the two ventricles. Structures of the cardiac conduction system includes SA node, AV node, perforating bundle, and a branching point of the LBB. A structure of the cardiac conduction region may be indicated as a line between the perforating bundle and the branching point. The structure of the cardiac conduction region may run along the floor of membranous septum. The location of the structure of the cardiac conduction region relative to membranous septum may be determined, for example, using one or more anatomical features described herein.

[0267] Exemplary approaches of identifying the membranous septum include:

[0268] * As a region location at a minimum thickness of a septum between a left ventricle and a right ventricle. The region may have a predetermined area and / or shape corresponding to a predicted area of the membranous septum, for example, average area and / or shape computed from areas and / or shapes of multiple membranous septum of multiple individuals.

[0269] * Identifying a mitral ring of a left ventricle myocardium connected to a mitral valve. Identifying a tricuspid ring of a right ventricle myocardium connected to a tricuspid valve. A plane of the mitral ring is at an angle with respect to a plane of the tricuspid ring. Identifying the membranous septum at a bump between the mitral ring and the tricuspid ring. * The membranous septum may be found by identifying the location of the membranous septum of sample images of sample individuals with respect to the one or more gross anatomical features, and clustering the locations with respect to the gross anatomical feature(s). The cluster may be applied to the gross anatomical feature(s) of the 3D image to identify the location of the membranous septum.

[0270] * Segmenting a myocardium of a left ventricle, and identifying an aortic notch in the myocardium. The membranous septum is located at the aortic notch.

[0271] The location of the membranous septum may be detected, for example, by obtaining orientation of a NRL and / or depth of the aorta. Using this data, the membranous septum may be detected, for example, according to the calculations described with reference to Figs. 7A-7C of the paper "Vulnerability of the ventricular conduction axis during transcatheter aortic valvar implantation: A translational pathologic study” by Tretter et al. dated Jan 2023, incorporated herein in its entirety. The location of the membranous septum may be computed, for example, according to the publication “The atrioventricular conduction axis and the aortic root — Inferences for transcatheter replacement of the aortic valve”, by Yolanda Macias et at., Clinical Anatomy. 2022;35: 143-154, incorporated herein by reference in its entirety.

[0272] The location of one or more gross anatomical features used for determining the location of the cardiac conduction system may be computed, for example, according to the following exemplary process, using the 3D image (e.g., CT) which may be obtained pre-procedure:

[0273] 1. Identification of the nadir of each of the aortic cusps (the right coronary cusp, the left coronary cusp, and the non-coronary cusp) - which may be used for defining the Virtual Aortic Annulus plane. The virtual aortic annulus plane may be computed as a plane intersecting the three nadirs.

[0274] 2. Identifying the location of the right and the left fibrous trigones, which are fibrous connective tissue regions near the aortic cusps.

[0275] 3. Identifying the triangular attachment of the atrial septum. The triangular attachment of the atrial septum may serve as an anatomical reference point.

[0276] 4. Drawing a first line in a posterior anterior direction from step #3 (e.g., across the Aortic Root).

[0277] 5. Drawing a second line connecting the center of the non-coronary cusp to the area of apposition of apposition (e.g., where the left and right coronary cusps meet).

[0278] 6. Identifying the ventricular septum and the infero-septal recess

[0279] 7. Identifying the width of the roof of the infero-spetal recess at the level of the virtual aortic valve plane. 8. Identifying the location of the apex of the infero-septal recess in section of the preprocedure CT that are parallel and lower (toward the apex of the left ventricle) than the apex.

[0280] 9. Measuring the virtual aortic valve plane.

[0281] 10. Identifying points A denoting the AV node, B denoting the perforating bundle, and C denoting the branching point of the LBB. Exemplary approaches for identifying point A, B, and / or C are described, for example, with reference to U.S. Provisional Patent Application Nos. 63 / 664,786, 63 / 656,164, and 63 / 621,140, incorporated herein by reference in their entirety.

[0282] One or more of the following gross anatomical features, which may be used for identifying the indication of the cardiac conduction system, may be computed:

[0283] • Amount of counter clockwise rotation of the aortic root.

[0284] • Amount of infero septal recess roof.

[0285] • The location and the area of the supra ventricular membranous above the virtual aortic plane.

[0286] • The depth of the of the infero septal recess.

[0287] • Anterior papillary muscles (AP), which may be used for imaging the LBB, as described herein.

[0288] • Posterior papillary muscles (PP) , which may be used for imaging the LBB, as described herein.

[0289] • Apex of the LV, which may be used for imaging the LBB, as described herein.

[0290] • Medial papillary muscle (MPM), which may be used for imaging the RBB, as described herein.

[0291] • Septal band (SB), which may be used for imaging the RBB, as described herein.

[0292] • Moderator band (MB), muscle between septum and free wall (myocardium), which may be used for imaging the RBB, as described herein.

[0293] • MB 1 - connected to interventricular septum, which may be used for imaging the RBB, as described herein.

[0294] • MB2 - connected to ventricular myocardium, which may be used for imaging the RBB, as described herein.

[0295] Referring now back to FIG. 14, schematic 1602 may be computed from a 3D image, such as a pre-procedure CT scan. Schematic 1602 depicts a LV myocardium 1604, a non-coronary cusp (NCC) 1606 (of the aortic valve), and a right coronary cusp (RCC) 1608 of the aortic valve. The following anatomical features may be automatically identified on the 3D image: NCC-RCC commissure 1610, predicted location of SA node (not shown), predicted location of AV node 1612 (Point A), membranous septum 1614 and optionally the membranous septum floor, anterior position on membranous septum floor 1616 (Point C), and posterior position on membranous septum floor 1618 (Point B). The membranous septum floor lies on the myocardium. The location of the structures of the cardiac conduction system may be computed based on the detected anatomical features. Since the conduction system is not shown in a CT scan, the conduction system may be computed, identified and / or imaged using known correlation between gross anatomical structures (detected from CT scan) and the conduction system. The conduction system may be computed, identified and / or imaged by a trained machine learning model and / or heuristic models, as described herein.

[0296] One exemplary approach for automatic segmentation of the membranous septum from a 3D image is now described. The exemplary approach is based on segmenting the posterior border of the membranous septum and the anterior border (e.g., point) of the membranous septum. The SA node and / or AV node may not necessarily be segmented since it cannot usually be visualized.

[0297] The membranous septum sits between the tip of the septal muscle and connects to the aortic root. The floor of the membranous septum includes part of the conduction axis (His Bundle and / or perforating bundle).

[0298] A surrogate location of the AV node may be segmented using the following exemplary approach: Identification of both hinges of the anterior and posterior leaflets of the mitral valve may be determined, in a two chamber and three chamber view using multiplanar reformatting (MPR), creating a short axis plane of the mitral valve annulus. In the three chamber view the hinge line of the anterior leaflet is distinguished from the adjacent fibrous tissue between the aortic and mitral valves. From the short axis plane of the mitral valve annulus the medial commissure of the mitral valve is identified. A horizontal line from the medial commissure to the right atrial endocardium is drawn, and the estimated location of the AV node is annotated in close proximity to the right atrial wall (Point A). If the apex of the inferior pyramidal space is clearly visible the AV node can be further confirmed by this estimation.

[0299] Alternatively or additionally, the AV node (point A) is identified the AV node is identified as being located within the infrapyramidal space and closest to the inferoseptal recess. Alternatively or additionally, the AV node is identified as being located closest to an apex of the IPS, and / or at an intersection of the IPS and ISR. Point B may be identified as being located within the posterior most position of the membranous septum floor. Alternatively or additionally, Point B may be identified at an intersection of the ISR and posterior border of the membranous septum floor.

[0300] Point C may be identified as being located on the membranous septum floor at a point between Point B and the most anterior point of the membranous septum floor.

[0301] A non branching bundle of His is identified as extending from point A to Point B over a surface of the ISR.

[0302] Optionally, Point C is located at about 40-80%, or 60-80%, or 50-70%, or other values, of a length of the line between Point B and the most anterior point of the membranous septum floor. Alternatively or additionally, Point C may be identified as being located at about 70%, or about 60-80%, or about 40-80%, or about 50-70%, or other values, of a length of the membranous septum floor.

[0303] The inferior border of the membranous septum may be segmented using the following exemplary approach: The aortic annulus plane is identified by the plane connecting the nadirs, or lower attachment points, of the three leaflets of the aortic valve. Using MPR from the short axis plane of the aortic valve, an orthogonal long axis plane is obtained across the membranous septum. The inferior margin of the membranous septum is identified as the junction between the fibrous membranous septum and underlying muscular interventricular septum.

[0304] The posterior and anterior points of the inferior margin of the membranous septum may be segmented using the following exemplary approach: From the short axis plane of the aortic valve annulus using MPR, the orthogonal long axis plane is rotated from the most posterior to the most anterior component of the membranous septum (Point B and Point C), respectively.

[0305] The posterior border of the floor of the membranous septum may be segmented using the following exemplary approach: Identification of the aortic annulus plane, such as by intersecting nadirs of the three cusps of the aortic valve (i.e., NCC, LCC, RCC). Identification of the membranous septum posterior border by identification of the roof of the inferoseptal recess in the short axis plane. Identification and optional annotation of an inferior area of the membranous septum adjacent to the roof of the inferoseptal recess, optionally in a long axis view.

[0306] The anterior point of the floor of the membranous septum may be segmented using the following exemplary approach: Performing clockwise rotation of each slice plane of the 3D image, optionally in a short axis, until the anterior edge of the membranous septum is identified, optionally in a long axis view. Annotation of a point between the inferior area of the membranous septum and the crest of the septum, optionally in the long axis view. Another exemplary approach for automatic segmentation of the membranous septum from a 3D image is now described. The 3D image may be a contrast enhanced cardiac CT scan. The 3D image may be segmented to identify the left and the right ventricles and the muscular ventricular septum between the ventricles. A tip of the muscular ventricular septum is identified and optionally marked for each slice of multiple slices of the 3D image, such as CT scan. A line connecting the superior point of the ventricular septal muscle through all the slices may be defined as the membranous septum floor. A sub-region within the line that includes the His bundle is identified within the membranous septum floor. The sub-segment may be defined between a first inferior about 10% or 20% or other percentage of the membranous septum floor to an end of the conduction path located at about 70%, or 80%, or 85%, or 90% of the length of the membranous septum floor.

[0307] Referring now back to FIG. 12, schematic 1302 depicts exemplary segmentations of a 3D image used for identifying the conduction system. Segmentation 1304 represents the endocardium of the left ventricle. Segmentation 1306 represents the right ventricle, in particular, the endocardium. Segmentation 1308 represents the myocardium of the left ventricle. Surrogate locations of one or more of: SA node (not shown), AV node 1310 (also referred to herein as point A), penetrating bundle (also referred to herein as point B) 1312, and branching point of the LBB 1314 (also referred to herein as point C), of the cardiac conduction system, may be identified based on anatomical features. The membranous septum may be used to define structures of the cardiac conduction system. The membranous septum is a region 1316 that may be defined as between penetrating bundle 1312 representing a posterior part of the membranous septum and branching point of the LBB 1314 representing an anterior part of the membranous septum, where segmentation 1308 is minimal between segmentation 1304 and segmentation 1306, and / or is in the posterior slope of the LV myocardium notch shown with reference to FIG. 13. The identified region may have an area corresponding to a predicted area of the membranous septum, for example, an average area computed from areas of membranous septum obtained from multiple sample subjects. 1316 may represent an area of apposition of the right ventricle (RV) and the left ventricle (LV).

[0308] The membranous septum may be identified between the ventricular septum and the inferoseptal recess as viewed from the aorta.

[0309] The membranous septum may be segmented, for example, using the following exemplary approach: identifying a region between the RV myocardium segmentation and the LV myocardium segmentation that includes a minimum spatially consistent distance between the RV myocardium and the LV myocardium. A length of a base of the region may be, for example, greater than about 0.8 mm, or 1mm, or 1.2 mm, and / or shorter than about 10 mm, or 12mm, or 14 mm, or other values. The thickness of membranous septum may be on the lower (possibly the lowest) percentile of distance between the segmentations of the LV and RV. Identifying an area defined as the edge of the LV myocardial segmentation that corresponds to the region of the left and right ventricles segmentation. Verifying that the identified area substantially matches a posterior down slope, optionally the notch, of the LV myocardium segmentation. Marking the path of the membranous septum floor that is included in the region of the LV myocardium segmentation apposition of the right ventricle to the left ventricle. Marking on the identified area the posterior down slope of left ventricle myocardium segmentation.

[0310] Referring now back to FIG. 13, the segmentation of the myocardium of left ventricle 1404 from a 3D image may is presented, optionally on a display. A notch in the left ventricle wall is indicated by dashed line 1406. The notch may be on the left ventricle myocardium segmentation, below the aortic root. The posterior down slope of the notch may be marked, for example, using a straight line with thickness and / or using a thick line (e.g., thickness is a function of the CT resolution and quality). Predicted locations of the SA node (not shown), AV node 1410, penetrating bundle 1412, and branching bundle 1414, may be determined according to notch 1406.

[0311] Optionally, a location of an annulus plane of the aortic valve may be detected on the reference image. The annulus plane may be detected by identifying a nadir of each respective cusp of the aortic valve. I.e., in an aortic valve with three cusps, three nadirs are identified, for the right coronary cusp, the left coronary cusp, and the non-coronary cusp. The aortic valve and / or cusps and / or nadirs may be detected, for example, by a machine learning model trained on a training dataset of images labelled with ground truths indicating the aortic valve and / or cusps and / or nadirs. Alternatively, image processing approaches may be applied. For example, the lower boundary of the aortic valve may be determined. In a contrast image where contrast is injected into the aortic root, the lower boundary is the boundary between the lower contrast enhanced aortic root and the left ventricle without contrast. The cusps may be detected as local curvatures of the boundary. For each cusp, the nadir may be computed as a local minimum, for example, using image processing approaches. The annulus plane may be detected by drawing a circle or other planar shape, intersecting the three nadirs.

[0312] At 306, locations of structures of the cardiac conduction system are estimated. The locations may be surrogate locations, representing a prediction of an expected location, rather than an actual measurement of the conduction system made at that location (e.g., direct visualization and / or electrical activity of the cardiac conduction system). The locations of one or more structures of the cardiac conduction system may be estimated using the plurality of gross anatomical features. The structures of the cardiac conduction system may be are within about 2 centimeters (cm), or about 3 cm, or about 5 cm of the aortic valve of the heart.

[0313] Examples of structures of the cardiac conduction system which may be estimated include: atrioventricular (AV) node (point A), bundle of His, penetrating bundle (point B), non-branching bundle, branching point of the LBB (point C), LBB, right branch bundle (RBB), left anterior fascicle, left posterior fascicle, left septal fascicle, Mahaim fibers, Purkinje fibers, and connection paths between two of the aforementioned along the cardiac conduction system.

[0314] The location of the structures of the cardiac conduction system may be based on the identified anatomical features, for example, based on predefined correlations such as according to an analysis of anatomical features and structures of the cardiac conduction system identified in multiple other subjects, using a trained machine learning model, using rules, by fitting a predefined template, and the like.

[0315] Inventor discovered that the cardiac conduction region is located in different places in different patients, with different correlations to anatomical structures, for example, a distance below or above an annulus plane and / or at different distances from the annulus plane.

[0316] Inventors analyzed data of multiple patents and discovered that:

[0317] The depth of the structure of the cardiac conduction system which runs within the membranous septum (i.e., between the perforating bundle and the LBB), below the annulus plane, is variable, as presented in Table 1 below:

[0318] The location of the structure of the cardiac conduction region which runs within the membranous septum, is variable, as presented in Table 2 below (RCC=right coronary cusp, NNC = non-coronary cusp, LCC=left coronary cusp):

[0319] The location of different structures of the cardiac conduction system is variable, as presented in Table 3 below:

[0320] The length between the penetrating bundle and the LBBB branching point of the cardiac conduction system (referred to as BC length) is variable, as presented in Table 4 below:

[0321] The variability of the location of the cardiac conduction system in different patients makes it difficult to predict its location using images in which the cardiac conduction system itself cannot be directly visualized, and / or without using direct electrical measurements of the nerves of the conduction system (e.g., ECG, EPS). At least one embodiment descried herein predicts surrogate locations for structures of the cardiac conduction system according to anatomical features identified in the image (e.g., 3D image), without using direct electrical measurements.

[0322] Optionally, correlations are computed between structures of the cardiac conduction system and the gross anatomical features. The correlations may be, for example, correlations between visual features extracted from the gross anatomical features and locations of the structures, correlations between distances and / or vectors (distance and direction), correlation between combinations of gross anatomical features and structures, and the like.

[0323] Locations of one or more structures of the cardiac conduction system, optionally a combination, may be predicted using the correlation with the gross anatomical features. The correlations may be, for example, a vector indicating distance and / or direction. The correlations may be computed based on an analysis of sample 3D images of sample subjects representing anatomical variations. For example, an average vector is computed, such as from a certain gross anatomical feature, by applying the vector, the predicted location of a certain structure may be obtained. The correlations may be computed between one or more combinations of structures of the cardiac conduction system and one or more combinations of gross anatomical features. Since the structures of the cardiac conduction system do not work in isolation, but in a path for conducting signals, location of one structure may impact the location of other structures, in particular neighboring structures. For example, a set of vectors pointing to two neighboring structures are computed from a set of multiple gross anatomical features. The set of vectors predict the locations of the neighboring structures. Alternatively or additionally, the locations of the structures of the cardiac conduction system are estimated by applying at least one rule derived from an analysis of hearts of different individuals. For example, when a distance between a first and a second gross anatomical feature is greater than a first threshold, and a curvature of a third anatomical structure is less than a second threshold, a certain structure is located midpoint between the first and third anatomical structures. When the distance is less than the first threshold and the curvature is greater than the second threshold, the certain structure is located midpoint between the first and second anatomical structures.

[0324] Alternatively or additionally, the locations are estimated is by feeding at least a portion of the 3D image depicting at least one of the gross anatomical features into a machine learning model (e.g., neural network). The machine learning model may be trained on a training dataset of records. A record may include at least a sample of a portion of a sample 3D image of a sample individual depicting the gross anatomical features(s), and a ground truth indicating the location(s) of the structure(s).

[0325] Exemplary architectures of machine learning models described herein include a detector architecture, a classifier architecture, and / or a pipeline combination, statistical classifiers and / or other statistical models, neural networks of various architectures (e.g., convolutional, fully connected, deep, encoder-decoder, recurrent, transformer, graph), support vector machines (SVM), logistic regression, k-nearest neighbor, decision trees, boosting, random forest, a regressor, and / or any other commercial or open source package allowing regression, classification, dimensional reduction, supervised, unsupervised, semi-supervised, and / or reinforcement learning. Machine learning models may be trained using supervised approaches and / or unsupervised approaches.

[0326] Optionally, the location of the membranous septum is determined, as described herein. The location of the membranous septum may be determined according to the one or more gross anatomical features (for example: the ventricular septum and the infero-septal recess), since the membranous septum is difficult to directly visualize. Structures of the cardiac conduction system may be determined relative to the membranous septum, for example, based on correlations and / or rules. For example, visual elements (e.g., image, such as dots, nerve-like icon) denoting the perforating bundle, bundle of His and a branching point of the LBB are located according to the membranous septum. The perforating bundle may be located at an anterior location of the membranous septum. The branching point of the LBB may be located at a posterior point location of the membranous septum. The bundle of His may be located at a connection between the perforating bundle and the branching point of the LBB and / or RBB. Exemplary approaches for determining structures of the cardiac conduction system based on one or more gross anatomical features are now described. It is noted that gross anatomical features used for determining the structures of the cardiac conduction system may be identified on the image, as described herein.

[0327] The Infra Pyramidal Space is closely related to the AV node and the initial part of the Bundle of His.

[0328] The location of the AV node may be estimated by locating the infra pyramidal space. The infra pyramidal space may be identified as a fat space between the left ventricle and right ventricle until the right atrium. The AV node may be identified at the end of the tip of the pyramid-like shape of the infra pyramidal space.

[0329] The location of the AV Node may be identified as being situated within the atrioventricular septum, near the opening of the coronary sinus, and is part of the triangle of Koch. The triangle of Koch is bounded by the tendon of Todaro, the tricuspid valve annulus, and the coronary sinus ostium. The AV node may be identified as being located just above the infra septal recess. The infra septal recess may be used as a landmark for accessing the AV node during transcatheter procedures. The AV node may be detected as being adjacent to an apex of the infra pyramidal space. The apex may provide a pathway for the AV node to transition into the Bundle of His.

[0330] The location of the SA node, may be estimated to be at the juncture of the superior vena cava with the upper portion of the crista terminalis of the right atrium. Alternatively or additionally, the location of AV node is estimated to be the middle of the infra-septal recess. Alternatively or additionally, the location of AV node is estimated to be within the intraventricular septum adjacent to the septal commissure between the posterior and the anterior leaflets of the mitral valve. Alternatively or additionally, the location of AV node is estimated to be within the intraventricular septum at the left most proximal part of the posterior leaflet adjacent to the septal commissure between the posterior and the anterior leaflets of the mitral valve. Alternatively or additionally, the location of AV node is estimated to be within fat of the infra-septal recess at the level close to the septal commissure of the mitral annulus.

[0331] The location of the non bifurcating bundle (including the bundle of His), may be estimated to be between the estimated location of the AV node to the most posterior part of the floor of the membranous septum.

[0332] The location of the bundle of His may be within the membranous septum floor, which may be defined between Point B and Point C described herein. Point C may be the most anterior location without touching the ventricular septum. The Bundle of His may be identified as originating from the AV node and traveling through the floor of the membranous part of the interventricular septum, in relation to the inferoseptal recess. The Bundle of His may be detected as passing close to the inferoseptal recess as the inferoseptal recess descends through the septum. The Bundle of His may be detected as being closely associated with the infra pyramidal space, particularly at its origin from the AV node. During the path of the bundle of His over the membranous septal floor it give rise to the right and left bundle branches.

[0333] The inferoseptal recess may be identified as described herein. The inferoseptal recess is significant because it is close to the atrioventricular (AV) node and the bundle of His. The located conduction system may be estimated by excluding the location of the conduction system within the inferoseptal recess.

[0334] The location of the penetrating bundle may be estimated to be between the most posterior part of the floor of the membranous septum to the most anterior part of the floor of the membranous septum tracking over the floor of the membranous septum.

[0335] The location of the bifurcating bundle may be estimated to be toward the left ventricle next to the most anterior part of the floor of the membranous septum.

[0336] The location of the right bundle (i.e., right bundle branch (RBB)) may be estimated to be across from the estimated location of the penetrating bundle on the right ventricle wall. The location of the RBB may be estimated as originating from the branching bundle during the path of the branching bundle over the membranous septal floor. The location of the RBB may be estimated as traveling along the right side of the interventricular septum toward the medial papillary muscle of the right ventricle.

[0337] The location of the RBB may be estimated to be at about 60%, or about 50%-70% of a length of the membranous septum floor. The RBB may emerge as a narrow bundle.

[0338] The location of the apex of the left ventricle may be found as the lowest point of the left ventricle.

[0339] The location of the left bundle (i.e., left bundle branch (LBB)) may be estimated to be on the left ventricle toward the apex of the left ventricle. The location of the left bundle may be estimating as originating from the branching bundle, during a path of the branching bundle over the membranous septal floor. The location of the LBB may be estimated as traveling along the left side of the interventricular septum.

[0340] The location of the LBB may be identified as extending along the length (e.g., entire length, 100% of the length) of the membranous septum floor. The LBB may cover the endocardium of the left ventricle. At a distance of about 1-4 times the length of the membranous septum floor, after the membranous septum, the LBB breaks into three fascicles. Each respective fascicle extends until its corresponding anterior papillary muscle, posterior papillary muscle, or apex of the left ventricle, as described herein.

[0341] The anterior and posterior papillary muscles (also referred herein as AP and PP) may be found by searching the left ventricle in the image using an axial view from a location corresponding to a plane of the aortic annulus. The axial view may enable better visualization of the entire (or larger portion of) the papillary muscle(s). The anterior papillary muscle may be found at a middle point (approximately) of the base of the left ventricle. The posterior papillary muscle may be found by searching posteriorly from the anterior papillary muscle. The location of the left anterior bundle (i.e., anterior fascicle) may be estimated to be below the location of the left bundle oriented toward the anterior papillary muscle. For example, by tracing a curve within the wall of the left ventricle from the left bundle to the anterior papillary muscle.

[0342] The location of the left posterior bundle (i.e., posterior fascicle) may be estimated to be below the location of the left bundle toward the posterior papillary muscle. For example, by tracing a curve within the wall of the left ventricle from the left bundle to the posterior papillary muscle.

[0343] The location of the left central bundle (i.e., septal fascicle) may be estimated to be below the location of the bifurcating bundle oriented toward the apex of the left ventricle with a path that passes between the two papillary muscles. For example, by tracing a curve within the wall of the left ventricle from the left bundle to a midpoint between the anterior and posterior papillary muscles.

[0344] The location of the RBB may be estimated to run from the bundle of His, where the EBB emerges, splitting off before the EBB to the location of the medical papillary muscle of the right ventricle, to the septal band, to MB1, and terminating at MB2.

[0345] The location of the medial papillary muscle may be estimated by viewing the right ventricle at an axial view from the plane of the aortic annulus. The papillary muscle may be identified at about 12 o’clock by searching for a triangle-like shape within the right ventricle. The conduction system may be identified at the lower right point of the triangle. It is noted that the papillary muscle may be very small in most people, making it difficult to directly visualize. Moreover, there are anatomical variations, such as some people not having a medial papillary muscle at all, and other people having 4 papillary muscles where it is unclear which one is the medical papillary muscle.

[0346] The septal band may be found as a region of muscle below the medial papillary muscle.

[0347] The moderator band (MB) may be found as a region of muscle between the septum and the free wall of the heart (myocardium). The MB may be identified across multiple slices of the 3D image. MB1 and MB2 are the two ends of the MB. MB1 may be found in close proximity to an identified the anterior papillary muscle. MB1 may be found as muscle connected to the interventricular septum. MB2 may be found at the extension of the muscle connected to MB 1 that connects to the free wall, i.e., the ventricular myocardium.

[0348] Optionally, other non-imaging parameters may be used in combination with the anatomical features for estimating the location of the structures of the cardiac conduction system, for example, being fed into the machine learning model in combination with image data, and / or used in the set of rules, and the like. Examples of non-imaging parameters include: patient demographics (e.g., age, biological gender), patient history (e.g., smoking history, previous procedures, cardiovascular history, family history, medications), and ECG conduction analysis (e.g., manual and / or automated, such a detection of arrhythmia and conduction disturbances, risk of arrhythmia, other ECG abnormalities). It is noted that at least one embodiment described herein is designed to be performed without requiring ECG data, however, if ECG data is available, other embodiments may include features for using it.

[0349] At 308, an indication of the cardiac conduction system of the subject (also referred to as just “indication”) is generated. As used herein, the indication of the cardiac conduction system may include one or more structures of the cardiac conduction system (also referred to as just “structures”). The indication of the cardiac conduction system of the subject may be generated according to the computed correlations, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

[0350] The indication may be registered to the 3D image, such as defined within the same coordinate system as the 3D image, and / or fused with the 3D image.

[0351] The indication may be designed to be visually presented.

[0352] The indication may be, for example, a second image (e.g., 3D image) fused with the image (e.g., 3D image) of the heart, an overlay over the image of the heart, visual indicators located within the image (e.g., arrows, dots, icons visually resembling nerve tissue) and / or wihin a 3D representation based on the image, coordinates within the image.

[0353] The indication may include interpolations between the structures.

[0354] The indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject. For example, rather than identifying a structure in the same location in all subjects, at least one embodiment described herein estimates a more accurate location of the structure according to the anatomy of the particular subject.

[0355] Optionally, an interpolation of a non-trivial geometry of the cardiac conduction system is rendered between the locations of the one or more structures. The non-trivial geometry may be based on an aggregation of visual representation of multiple conduction systems, for example, depicting a complex branching of multiple nerves of the fascicles at various levels for innervating different parts of the myocardium. The rendering of the interpolation of the non-trivial geometry may be according to at least one rule. An example of the interpolation of the non-trivial geometry is shown with respect to template 402 of FIG. 3.

[0356] The structures included in the indication of the cardiac conduction system may indicate an anatomical fact. Examples of structures included in the indication of the cardiac conduction system:

[0357] * The non bifurcating bundle including the bundle of His within the indication of the cardiac conduction system denotes an anatomical region between the estimated location of the AV node and the most posterior part of the floor of the membranous septum.

[0358] * The penetrating bundle within the indication of the cardiac conduction system denotes an anatomical region between the most posterior part of the floor of the membranous septum to the most anterior part of the floor of the membranous septum tracking over the floor of the membranous septum.

[0359] * The bifurcating bundle within the indication of the cardiac conduction system denotes an anatomical region toward the left ventricle next to the most anterior part of the floor of the membranous septum.

[0360] * The right bundle branch within the indication of the cardiac conduction system denotes an anatomical region across from the estimated location of the penetrating bundle on the right ventricle wall.

[0361] * The left bundle branch within the indication of the cardiac conduction system denotes an anatomical region on the left ventricle toward the apex of the left ventricle.

[0362] * The left anterior fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to a base of the anterior papillary muscle of the left ventricle.

[0363] * The left posterior fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to a base of the posterior papillary muscle of the left ventricle.

[0364] * The left septal fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to the apex of the left ventricle with a path that passes between the anterior and posterior papillary muscles.

[0365] At 310, the structures of the cardiac conduction system may be marked on one or more anatomical images, optionally a 3D image such as a CT scan captured pre-procedure. Optionally, the structures of the cardiac conduction system may be marked based on certainty of their locations. The different structures may be marked differently according to certainty of their estimated locations, for example, different line and / or transparency and / or HUE character and / or other visualizations. For example, the brighter and / or the less transparent the structure, the greater the certainty of its estimated location. The lower the brightness and / or the more transparent the structure, the less the certainty of its estimated location. The following is an example of structures of the cardiac conduction system in descending order of certainty: Point B, Point A, Point C, LBB, RBB, proximal septal fascicle, proximal anterior fascicle, proximal posterior fascicle, distal septal fascicle, distal posterior fascicle, distal anterior fascicle.

[0366] Optionally, the indication of the cardiac conduction system of the subject is created by adapting a template according to the gross anatomical features and / or according to the computed correlations. The template may be an image, a set of connected points, a visual model, and the like. The template may include at least the left bundle branch, the anterior fascicle, the posterior fascicle, and the septal fascicle. The template may include other structures of the cardiac conduction system. The template may be fitted to the image by adapting and / or fitting anchor points according to the gross anatomical features, and / or adapting connections between the anchor points, and / or by re-sizing the template to fit the gross anatomical features. For example, the anchor point at the left bundle branch of the template is placed at the posterior part of the membranous septum, and anchor pointes at ends of the fascicles (anterior, posterior, septal) are positioned closest towards corresponding papillary muscles. The template may be re-sized accordingly.

[0367] Referring now back to FIG. 3, template 402 may be adapted for creating the indication of the cardiac conduction system. There may be one or more templates. Template 402 is meant to serve as a not necessarily limiting example. Template 402 includes a branch point of a left bundle branch 404, a left anterior fascicle 406, a left septal fascicle 408, and a left posterior fascicle 410. It is noted that proximal and / or superior portions of fascicles 406, 408, and 410 are depicted. Distal and / or inferior portions of the fascicles may not necessarily be depicted, for example, due to their thin structures. Dimension of template 402 may be adapted, for customization for the subject, such as based on the gross anatomical features. For example, a first dimension 412 indicating total length just before branching into the fascicles may be adapted. In another example, a second dimension 414 indicating total length from an end of the left anterior fascicle to an end of the left posterior fascicle may be adapted accordingly. The adapted template may be integrated within the 3D image, within the 3D model, projected and overlaid on the 2D image, and the like as described herein. Referring now back to FIG. 4, image 502 includes heart 504 and indication of the cardiac conduction system 506. Indication of the cardiac conduction system 506 is generated as described herein. Indication of the cardiac conduction system 506 is shown face-on (i.e., anfas view) as seen from the left ventricle. The face-on view may be determined, for example, as the view from which an area of the indication of the cardiac conduction system is maximized. Heart 504 may be a 3D image and / or a 3D rendition and / or 3D model generated from the 3D image. Heart 504 includes a left ventricle 508, a left atrium, 510, and an aortic root 512. Indication of the cardiac conduction system 506 may include SA node, AV node 514, bundle of His 516, penetrating bundle 518, branching point of the LBB 520, left anterior fascicle 522, left posterior fascicle 524, and left septal fascicle 526. It is noted that proximal and / or superior portions of fascicles 522, 524, and 526 are depicted. Distal and / or inferior portions of the fascicles may not necessarily be depicted, for example, due to their thin structures.

[0368] Referring now back to FIG. 5, image 602 is a face-on view of indication of the cardiac conduction 506 and image 502 as shown in FIG. 4, as seen from the right ventricle.

[0369] Referring no back to FIG. 6, image 650 depicts RBB 604, which is created as described herein based on gross anatomical features. RBB passes within a septum 606 and a free wall (myocardium) of the right ventricle 608. RBB 604 may be visualized by drawing a curve along the surface and / or below the surface of septum 606 and free wall 608. The curve may connect the following gross anatomical features, in the following order: medial papillary muscle (MPM) 610 septal band 612 MB1 614 of a moderator band (MB) MB2 616 of the moderator band. MPM 610, SB 612, and MB1 614 lie along septum 606. MB1 614 and MB2 616 are part of the moderator band (not shown), optionally opposite ends thereof, that connects septum 606 to free wall 608. MB2 lies on free wall 608.

[0370] Referring now back to FIG. 15, schematic 1502 and 1504 may depict a 3D model of a heart 1506A and 1506B, marked with structures of a cardiac conduction system 1508 A and 1508B, generated as described herein. Structures of the cardiac conduction system 1508A-B may include a LBB and proximal portion of the three fascicles, as described herein. Schematics 1502 and 1504 may be generated as representing non-parallax views, for example, for selecting a LAO and / or CRA angle of a fluoroscopy machine corresponding to a selected S curve, during the procedure for obtaining fluoroscopy images on which the structures of the cardiac conduction system 1508A- B may be overlaid, as described herein. Schematic 1502 depicts an enface view of structures 1508 A. Schematic 1508B may depict a portrait view of structures 1508 A. Structures of the cardiac conduction system 1508A-B may be overlaid on fluoroscopy image captured during the procedure in real time, as described herein. Referring now back to FIG. 10, image 1002 depicts the indication of the cardiac conduction system as dots (or other spaced apart visual indications). One or more dots may represent a predicted location of a structure of the cardiac conduction system. For example, dots 1004 may denote predicted location of an anterior fascicle, dots 1006 may denote predicted location of a posterior fascicle, and dots 1008 may denote predicted location of a septal fascicle. A left bundle branch 1010 from which fascicles 1004, 1006, and 1008 originate may be presented. The dots are positioned at corresponding location on a representation of a heart 1012, such as a 3D image, 2D image, model, and the like.

[0371] Exemplary markings and / or uses of the indication of the cardiac conduction system are now described.

[0372] The indication of the cardiac conduction system may be presented in conjunction with the 3D representation (e.g., 3D image) of the heart, on a display, optionally within a graphical user interface configured for rotation, displacement, and / or zoom in / out of the 3D representation.

[0373] A fused image may be created by registering the indication of the cardiac conduction system to an anatomical image (e.g., the 3D image, a visual model based on the 3D image). The fused image may be presented on a display, within the GUI.

[0374] Alternatively or additionally, the indication of the cardiac conduction system may be registered to a functional image, for creating a fused image. Alternatively or additionally, an anatomical image with the indication of the cardiac conduction system may be registered to the functional image for creating the fused image. The fused image depicting the functional image with the indication of the cardiac conduction system may be presented on a display. Functional images may depict for example at least one of: conduction functionality of nerves of the cardiac conduction system, and viability of the myocardium of the heart innervated by the cardiac conduction system. The fused image may indicate which structures of the cardiac conduction system innervate non-viable myocardium, and / or which structures of the cardiac conduction system do not conduct.

[0375] Alternatively or additionally, the indication of the cardiac conduction system of the subject may be fed into an artificial intelligence (Al) process for further analysis. Examples of Al processes include: estimating risk of damage to the conduction system (e.g., as described with reference to Provisional Patent Application No. 63 / 656,164), and planning a transcatheter procedure for reducing or avoiding damage to the conduction system (e.g., as described with reference to Provisional Patent Application No. 63 / 532,995).

[0376] Alternatively or additionally, the indication of the cardiac conduction system of the subject may be generated for multiple different subjects using respective sample images. Multiple records of a training dataset may be created, where a record includes the sample 3D image, and a ground truth of the indication of the cardiac conduction system generated for the sample 3D image. A machine learning model may be trained on the training dataset. The machine learning model may be more robust and / or may be able to handle more anatomical variations than creating the indication using non-machine learning approaches.

[0377] Optionally, the fused 3D image may be used for computing an optimal 2D projection for mapping the 3D indication of the cardiac conduction system to a 2D image of the heart according to a single and / or double S curve. The single and / or double S -curve is a view in which a target anatomical structure, such as the membranous septum, other region of the septum, myocardium of the left ventricle, and / or other structure (e.g., aortic valve annulus and / or delivery catheter) planes are displayed perpendicularly on fluoroscopy without foreshortening. Using the fused 3D image obtained pre-procedure, comparable projections to that of the single and / or double S-curve model may preclude the need for intraprocedural image processing. The S-curve of the target anatomical structure, representing the perpendicular views of the target anatomical structure, may be generated using pre-procedural CT analysis. During the procedure, the operator may be informed of the optimal pose of the fluoroscopic device to obtain the optimal 2D projection for viewing the structure of the cardiac conduction system at the target anatomical structure. In another example, during the projection, the operator may use any pose, which is then registered to the 3D image and compared to the optimal 2D projection. An indication of correctness may be generated, as described herein.

[0378] Calcification deposits may be detected on the 3D image. A pattern of calcification deposits may be referred to herein as a calcium signature and / or calcification template. The calcium may be deposited in multiple clumps, which may be “randomly” distributed, in the sense that each patient is expected to have their own unique pattern of distribution of calcium deposits. The calcification deposits may be detected, for example, according to a voxels having Hounsfield values within a range that defines calcium.

[0379] The pre-procedure image may be used for generating a template which may be registered (e.g., substantially matched and / or correlated) with a signature of a 2D image (e.g., fluoroscopic image) obtained during the procedure. The registration between the 2D image obtained during the procedure (e.g., in real time or near real time) and the 3D image (e.g., obtained pre-procedure) enables mapping the structures of the cardiac conduction system whose location are estimated in the 3D image to the 2D image, even when the structures of the cardiac conduction system are not visible or not sufficiently visible on the 2D image. Optionally, 2D projections of the 3D image are computed. The 2D projections may be used as templates, which are registered with the 2D fluoroscopic image(s), which may be used to map structures of the cardiac conduction system of the 3D image to the 2D fluoroscopic image(s).

[0380] The fused 3D image may be used for performing a registration of the 2D fluoroscopy image(s) obtained during the procedure to a 2D projection of the 3D image. This registration process may be used to correlate the structures of the cardiac conduction system defined on the 3D image with the actual patient anatomy visible on the 2D fluoroscopy images during the transcatheter procedure.

[0381] Optionally, the 3D image may be analyzed to identify a parallax view for the structures of the cardiac conduction system and / or anatomical structure in proximity thereof. Valve parallax occurs since the structures of the cardiac conduction system are depicted differently in different 2D images captured at different poses of the image sensor. An optimal pose of the image sensor may be computed for optimal viewing of the structures of the cardiac conduction system.

[0382] Optionally, a parallax free presentation of one or more structures of the cardiac conduction system and / or anatomical features in proximity thereof depicted in a 2D fluoroscopy image A registration (e.g., function) for registering a 2D fluoroscopy image with a 3D image (e.g., preprocedure cardiac CT) of a subject is computed. One or more features are located within the 3D image, for example, the aortic annulus plane, membranous septum floor, and / or a simulation of a location of the aortic valve prosthesis device. The features are associated with CT derived information (CDI) indicating coordinates of the location within the 3D image. The features are detected on the 2D fluoroscopy image. Fluoroscopy derived information (FDI) is extracted. The FDI includes the location of the feature(s) on the 2D fluoroscopy image and a pose of a sensor that captured the 2D fluoroscopy image. A parallax free pose of the structures of the cardiac conduction system and / or anatomical features in proximity thereof depicted in the 3D image is computed. The parallax free pose may be the pose that is closest to the pose of the sensor that captured the 2D fluoroscopy image. The registration is applied for projecting the feature(s) of the 3D image defined by the CDI to a 2D plane corresponding to the FDI, according to the parallax free pose for generating a synthetic parallax free view (SPFV) of the feature(s). Alternatively, the whole 3D image is projected to the 2D plane to generate a synthetic fluoroscopy image. The SPFV of the feature(s) is provided for presentation, for example, as an overlay over the fluoroscopy image and / or as a separate presentation.

[0383] Optionally, the indication of the cardiac conduction system is validated using ECG measurements of the subject. The ECG measurements, collected using multiple sensors located at different positions on the chest of the subject, may be correlated with the locations of the structures of the generated indication of the cardiac conduction system. The validation may be done as a rough sanity check, to validate generate correctness of the indication of the cardiac conduction system. For example, if the ECG measurements made by one or more sensors indicate activity of the anterior fascicle as expected, but the anterior fascicle is positioned at a different location (and hence would look different on the ECG measurements), the indication of the cardiac conduction system may be determined to be invalid, for example, prompting an alert for an operator to investigate, and / or prompting a re-computation of the indication.

[0384] Referring now back to FIG. IB, at 320, the indication of the structures of the cardiac conduction system are obtained. The indication of the structures of the cardiac conduction system may be generated, for example, as described with reference to 308 of FIG. 1A. Alternatively or additionally, the markings of the cardiac conduction system on the anatomical image(s) and / or the marked anatomical image(s) are obtained. The markings are generated, for example, as described with reference to 310 of FIG. 1 A.

[0385] At 322, a 2D image is obtained.

[0386] The 2D image may be a fluoroscopy image, which may be obtained in real time during a transcatheter procedure.

[0387] At 324, an indication of the structures of the cardiac conduction system may be marked and / or presented over the 2D image.

[0388] Referring now back to FIG. 9, image 902 depicts a fluoroscopy image 904 on which structures 906 of the cardiac conduction system are overlaid and / or marked, as described herein. Structures 906 may include the LBB, the anterior fascicle, posterior fascicle, and septal fascicle, as described herein.

[0389] Referring now back to 324 of FIG. IB, the indication may be based on the gross anatomical features identified in the 3D image and / or in the 2D image.

[0390] Locations of the structures of the cardiac conduction system may be marked on a visual rendition of the heart generated based on the 2D image and / or the 3D image which is mapped to the 2D image.

[0391] Exemplary approaches for presenting the structures of the cardiac conduction system on the 2D image are now described:

[0392] • The 2D image may be registered with a 2D plane defining a projection of the 3D image. A pose of the 2D plane substantially matches a pose of the 2D image. For example, a pose of the fluoroscopy machine is extracted via optical character recognition from the 2D image, and used to compute the corresponding pose of the 2D plane. The registration may be between features visible in the 2D image and in the 3D image, for example, a calcification signature as described herein. The indication of the cardiac conduction system is registered to the 3D image. The indication (e.g., 3D) of the cardiac conduction system is registered to the 2D plane for creating a 2D projection of the indication of the cardiac conduction system. The 2D projection is presented on the 2D image, for example, as an overlay.

[0393] • The 2D image is registered with a 2D plane defining a projection of the 3D image. A pose of the 2D plane substantially matches a pose of the 2D image. The gross anatomical features depicted in the 3D image are mapped to the 2D image, for example, by projecting to the 2D plane, and mapping from the 2D plane to the 2D image. A 2D indication of the cardiac conduction system of the subject is generated according to the computed correlations and according to the mapping. The 2D indication of the cardiac conduction system is presented as an overlay over the 2D image.

[0394] • Locations of one or more structures of the cardiac conduction system are estimated in the 2D image of the heart using the gross anatomical features identified in the 3D image. Gross anatomical features in the 2D image corresponding to the gross anatomical features in the 3D image may be identified. The gross anatomical features in the 2D image may be used to location the structures of the cardiac conduction system in the 2D image, by considering the pose of the 2D image (e.g., view of the fluoroscopy machine) when generating the indication of the structures on the 2D image. The indication of one or more structures of the cardiac conduction system in the 2D image of the heart generated using the gross anatomical features may be presented, for example, as an overlay.

[0395] An exemplary approach for generating a parallax free presentation of one or more structures of the cardiac conduction system and / or gross anatomical features depicted in a 2D fluoroscopy image (e.g., as an overlay, as markings). The gross anatomical features may be located in proximity to the structures, for example, membranous septum, intraventricular septum, aortic valve, and the like is described. A registration (e.g., function) for registering a 2D fluoroscopy image with a 3D image (e.g., pre-procedure cardiac CT) of a subject is computed. One or more gross anatomical features and / or the structures of the cardiac conduction system are located within the 3D image. The gross anatomical features are associated with CT derived information (CDI) indicating coordinates of the location of corresponding structures of the cardiac conduction system within the 3D image. The gross anatomical features are detected on the 2D fluoroscopy image. Fluoroscopy derived information (FDI) is extracted. The FDI includes the location of the corresponding structures of the cardiac conduction system on the 2D fluoroscopy image and a pose of a sensor that captured the 2D fluoroscopy image. The location of one or more structures of the cardiac conduction system on the 2D image may be estimated according to the detected locations of the gross anatomical features on the 2D image. A parallax free pose of the gross anatomical feature(s) and / or structures of the cardiac conduction system depicted in the 3D image is computed. The parallax free pose may be the pose that is closest to the pose of the sensor that captured the 2D fluoroscopy image. The registration is applied for projecting the structures of the cardiac conduction system and / or the gross anatomical features of the 3D image defined by the CDI to a 2D plane corresponding to the FDI, according to the parallax free pose for generating a synthetic parallax free view (SPFV) of the structures of the cardiac conduction system and / or of the gross anatomical features. Alternatively, the whole 3D image is projected to the 2D plane to generate a synthetic fluoroscopy image. The SPFV of the structures of the cardiac conduction system are provided for presentation, for example, as an overlay over the fluoroscopy image and / or as a separate presentation.

[0396] Optionally, a calcification template (also referred to herein as calcification signature) identified in the 3D image and / or the 2D may be used for generating the indication of the cardiac conduction system on the 3D image and / or the 2D image, correlations between a calcification template of the 3D image and the 3D indication of the cardiac conduction system may be computed. The calcification template may be identified in the 2D image. The 2D image may be registered with a 2D plane defining a projection of the 3D image according to the calcification template. For example, the calcification template in the 2D image is captured at a certain pose. Different projections of the calcification template in the 3D image to different 2D planes at different poses may be computed. The 2D plane most correlated with the calcification template in the 2D image may be identified, for example, highest correlation value, and / or by performing a search, and / or by reading the pose from the 2D image. Optionally, the 3D indication of the cardiac conduction system may be projected onto the 2D image according to the pose of the selected 2D plane. Alternatively, the 2D plane with projected 3D indication is mapped to the 2D image. Alternatively, a 2D indication of the cardiac conduction system for the 2D image is generated (e.g., overlay) according to a registration function that registers the calcification template in the 2D image to the calcification template of the 3D image

[0397] Referring now back to FIG. 7, 2D image 702 of heart 704 may be, a fluoroscopy image obtained during a procedure. In another example, image 702 may be created from a 3D image, such as by projecting the 3D image to a 2D plane, which may correspond to a view of a fluoroscopy image obtained during a procedure. Indication of the cardiac conduction system 706 may be created as described herein. For example, indication of the cardiac conduction system 706 is overlaid on 2D image 702, a 3D representation of indication of the cardiac conduction system 706 may be projected to a 2D plane corresponding to a view of 2D image 702 and overlaid on 2D image 702, indication of the cardiac conduction system 706 may be marked on 2D image 702 according to anatomical features detected on 2D image 702, and the like. Indication of the cardiac conduction system 706 may include a LBB 708, a posterior fascicle 710, an anterior fascicle 712, and a septal fascicle 714.

[0398] Referring now back to FIG. 8, 2D image 802 depicts image 702 of FIG. 7 captured at a different view, showing a face on view of a septum 804 of the heart. Views of the indication of the cardiac conduction system 706 with respect to the view of the heart may be dynamically generated and presented, such as in response to real time changes of a C-arm of the fluoroscopy device capturing the fluoroscopy images, as described herein.

[0399] Referring now back to 324 of FIG. IB, an exemplary approach for registering an anatomical structure (e.g., ascending aorta) depicted in a 3D image (e.g., pre-procedure CT scan) to a 2D fluoroscopy image (obtained during the projection). The registration enables mapping structures of the cardiac conduction system detected in the 3D image to the fluoroscopy image, optionally in real time during the procedure. The registration may be performed by identifying the anatomical structure (e.g., ascending aorta) and the anatomical feature(s) in the 3D image. The contour of the anatomical structure (e.g., ascending aorta) is detected in the fluoroscopy image, which may depict contrast injected into the anatomical structure (e.g., ascending aorta). A pose of the sensor that captured the fluoroscopy image (e.g., C-arm of an x-ray machine) is obtained. The contour of the anatomical structure (e.g., ascending aorta) detected in the 3D image is projected to a 2D plane according to the pose, representing what the anatomical structure (e.g., ascending aorta) looks like as viewed from the sensor at the pose. The contour projected from the 3D image is registered to the contour detected in the fluoroscopy image. A registration (e.g., function) is computed. The registration may applied for registering (e.g., mapping) the structures of the cardiac conduction system detected in the 3D image to the fluoroscopy image, e.g., during the medical procedure. The anatomical features are presented on the fluoroscopy image, such as using an overlay, for example, for depicting the structures of the cardiac conduction system in real time.

[0400] The structures of the cardiac conduction system may be presented on a display, optionally as an overlay over 2D images captured during the procedure, for example, fluoroscopic images. The location of the structures of the cardiac conduction system may be determined on the 3D image (e.g., CT obtained pre-procedure). The location of the structures of the cardiac conduction system on the 3D image may be projected to a 2D plane registered with a 2D image obtained during the procedure. The 2D plane may be obtained, for example, based on an indication of a pose of the image sensor (e.g., C-arm), and / or calcification signature, and / or marker(s) on a transcatheter device (e.g., the THV). The projection on the 2D plane registered with the 2D image (e.g., fluoroscopic image) may be used as the overlay.

[0401] The tracking of the structures of the cardiac conduction system may be done, for example, by identifying a 2D projection of a 3D image of the heart (e.g., obtained by a pre-procedure CT scan) where the structures of the cardiac conduction system are indicated, registering the 2D fluoroscopic image(s) (obtained during the procedure) with the 2D projection, and mapping the location of the structures of the cardiac conduction system on the 3D image, which are projected to the 2D projection, to the 2D fluoroscopic image(s). The location of the structures of the cardiac conduction system may be visually indicated, for example, as an overlay, such as a line to indicate the annulus line.

[0402] At 326, features described with reference 322 and 324 may be iterated in real time, for dynamically updating the presentation of the indication of the structures of the cardiac conduction system on real time 2D images, such as according to varying poses of the fluoroscopy machine.

[0403] Optionally, a change in settings of the 2D image and / or the sensor that captured the 2D image (e.g., fluoroscopy machine) may be detected. Examples of changes in settings include: zoom, subject orientation, and the pose of the imaging device.

[0404] The change in settings may be detected, for example, by monitoring input of the operator (e.g., user enters instructions to change the settings), based on output of a sensor measuring pose (e.g., position sensor, angulation sensor, accelerometer), optical character recognition of the pose added to the 2D image(s), and / or based on detecting a change of 2D projection (e.g., template) of the 3D image that is registered to the 2D fluoroscopic image.

[0405] The detected change may trigger a computation for correcting the mapping of the 2D fluoroscopic image to the 3D image, such as to a 2D projection of the 3D image. The computation may be for finding a new 2D projection (e.g., template) of the 3D image that is registered with the 2D fluoroscopic image.

[0406] Alternatively or additionally, the detected change triggers a new search for the calcification signature on the 2D image fluoroscopic image, which is then matched to a projection of the 3D image.

[0407] Optionally, one or more structures of the cardiac conduction system depicted in the 2D fluoroscopic images, may be tracked. The structures may be dynamically tracked on subsequent non-contrast images, which reduces the amount of contrast used. The structures may be dynamically tracked on subsequent non-contrast images, even as the pose of the image sensor is changed, which gives the operator freedom to adapt the pose as needed. The tracked structure(s) may be displayed, for example, as an overlay on the fluoroscopic image.

[0408] One or more different approaches may be used for tracking the structure(s), for example, described as follows:

[0409] The structure(s) may be detected by using the calcification template as the gross anatomical features correlated with the structures of the cardiac conduction system. The calcification template may be computed by searching the calcification signature for a 2D calcification template of the 3D image according to a correlation requirement. Optionally, multiple 2D projections of the 3D image are pre-computed according to predicted poses of the image sensor. The search may be done using the pre-computed 2D projections, to find the closest matching projection. For example, the 2D projection with highest correlation to the calcification signature of the 2D target image. Structure of the cardiac conduction system marked on the 3D image are projected to a 2D plane of the calcification template. The 2D image may be registered to the calcification template. Registration may be performed, for example, by matching features, and / or other approaches for image registration as described herein. The indication of the structures of the cardiac conduction system on the calcification template may be mapped to the registered 2D image. The indication of the structures of the cardiac conduction system is provided, optionally presented on the 2D target image such as an overlay.

[0410] Alternatively, or additionally, the indication of the structures of the cardiac conduction system may be automatically detected by processing the 2D image by cross correlation of the 2D image with the 3D image, and mapping the structure(s) found in the 3D image to the 2D image. A projection of the 3D image to a 2D plane is computed, where the 2D projection of the 3D image is registered with the 2D image. The projection may be computed, for example, using the pose of the image sensor that captured the 2D image. The pose of the image sensor may be found, for example, using a pose sensor, and / or by optical character recognition to extract text of the target image indicating the pose of the image sensor. The structure(s) may be found on the 2D projection, and mapped to the 2D image using the registration.

[0411] Other exemplary embodiments are now described.

[0412] At least some embodiments described herein address the technical problem of reducing risk of damage to a conduction system of a heart of a subject, by imaging the conduction system, and presenting the area as an overlay on fluoroscopic images, which may enable an operator to avoid the area and / or be more careful when moving devices near the area.

[0413] The conduction system of the heart is responsible for generating and coordinating the electrical impulses that regulate the heart's rhythm and ensure its proper functioning. During a transcatheter procedure, which is a minimally invasive heart procedure, there is a potential risk of damage to the heart's conduction system.

[0414] An exemplary application of at least one embodiment for visualizing structures of the conduction system described herein is computing the area of conduction system at risk. The area of conduction system at risk may be computed by feeding in a combination of at least one indication of a structure of the conduction system of the subject and optionally at least one procedure parameter of the transcatheter procedure into a machine learning model. The area of the conduction system at risk may be obtained as an outcome of the machine learning model. The machine learning model may be trained on a training dataset of records. Each record includes the combination of one or more indications of structures of the conduction system of a sample individual, and optionally one or more procedure parameters of the transcatheter procedure performed on the sample individual, and a ground truth indication of the conduction system at risk.

[0415] Examples of additional personal parameters that may be fed into the machine learning model (and / or used for training the machine learning model) include one or more of the following: patient demographics (e.g., age, biological gender), mapping data (e.g., location of conduction system computed using the 3D image obtained pre-procedure), patient history (e.g., smoking history, previous procedures, cardiovascular history, family history, medications), and ECG conduction analysis (e.g., manual and / or automated, such a detection of arrhythmia and conduction disturbances, risk of arrhythmia, other ECG abnormalities).

[0416] Examples of procedure parameter(s) include one or more of the following: type of procedure (e.g., TAVI, PCI), and type and size of the tools used in the procedure.

[0417] The area of the conduction system at risk may include one or more structures of the conduction system which are at risk of being damaged during the transcatheter procedure, for example, due to application of high pressure by a tool used during the procedure, and / or due to puncture by a tool during the procedure.

[0418] According to some embodiments of the present invention, there is provided a computer implemented method of and / or system for segmenting a membranous septum of a 3D image, comprising: identifying a left ventricle myocardium base rim, identifying a first point and a second point defining an anterior myocardial notch passing therebetween, identifying a first location spaced apart from the first point as a start of a cardiac conduction region between a perforating bundle and a branching point of the left branch bundle, passing through the membranous septum, defining an box having a predefined width, a predefined height and a length computed as a predicted length of the cardiac conduction region with margin of error, positioning the box parallel to and a preselected distance above the left ventricle myocardial base rim, wherein the box defines the location of the membranous septum.

[0419] Optionally, the predefined width is about 1 millimeter (mm), the predefined height is about 1 mm, the predefined length is obtained as an average of a length measured for a plurality of subjects and one standard deviation, and the preselected distance is about 0.5 mm.

[0420] According to some embodiments of the present invention, there is provided a computer implemented method of and / or system for locating a membranous septum on a 3D image, comprising: for each slice of a plurality of slices of a 3D image, segmenting a left ventricle, a right ventricle, and a muscular intraventricular septum, and identifying a tip of the muscular intraventricular septum, identifying a path denoted by a line connecting a superior point of the muscular intraventricular septum through the plurality of slices, and detecting a sub-segment that includes a bundle of His within the path, wherein the membranous septum floor is defined at the location of the sub-segment.

[0421] Optionally, the sub-segment is defined between a first inferior about 10% of the membranous septum floor to an end of the conduction path located at about 90% of the length of the membranous septum floor.

[0422] Optionally, sub-segment of the path is defined between about 20% and about 70%, 80% or 85% of a total length of the membranous septum floor.

[0423] Optionally, the 3D image comprises a contrast enhanced cardiac CT.

[0424] According to some embodiments of the present invention, there is provided a computer implemented method of and / or system for locating an atrioventricular (AV) node on a 3D image, comprising: identification of a first set of hinges of an anterior cusp of the aortic valve, and a second set of hinges of a posterior cusp of the aortic valve, adjusting a plurality of parallel slice plane of the 3D image to the superior edge of the first set of hinges and the second set of hinges, localizing an area between a medial commissure of the mitral valve and the right atrium, and detecting a location of the AV node in close proximity to an atrial wall or an apex of the inferior pyramidal space.

[0425] Optionally, the first and second set of hinges are identified in a two chamber and / or a three chamber view in NPR.

[0426] Optionally, the localizing of the area is performed in a short axis view.

[0427] According to some embodiments of the present invention, there is provided a computer implemented method of and / or system for segmenting a floor of a membranous septum on a 3D image, comprising: detecting a posterior border of the floor of the membranous septum by: identifying an aortic annulus plane as a plane intersecting three nadirs of three cusps of the aortic valve, identifying of a roof of an inferoseptal recess in a short axis plane, wherein the roof denotes the posterior border of the membranous septum, identifying and annotating an inferior area of the membranous septum adjacent to the roof of the inferoseptal recess in long axis view, and detecting an anterior border of the floor of the membranous septum by: clockwise rotating each slice of a plurality of slices of a 3D image in a short axis until the anterior edge of the membranous septum is identified in the long axis view, and annotating the point between the inferior area of the membranous septum and the crest of the septum in long axis view.

[0428] According to some embodiments of the present invention, there is provided a computer implemented method of and / or system for segmenting (optionally a floor) of a membranous septum on a 3D image, comprising: segmenting a right ventricle endocardium (RV), a RV myocardium, a left ventricle (LV) endocardium, and a LV myocardium, identifying a region between the RV myocardium and the LV myocardium includes a minimum spatially consistent distance between the RV myocardium and the LV myocardium, wherein the membranous septum is segmented according to the region.

[0429] Optionally, further comprising: detecting a posterior down slope of a notch on the LV myocardium below an aortic root, and verifying that the region corresponds to the posterior down slope of the notch of the LV myocardium.

[0430] Optionally, further comprising: detecting a cardiac conduction region defined between a perforating bundle of a conduction system of the heart and a branching point of a left bundle branch (LBB) of the conduction system with the region, wherein the membranous septum is segmented according to the cardiac conduction region.

[0431] Referring now back to FIG. 16, system 300 is depicted as being used for visualizing structures of a cardiac conduction system of a patient (also referred to herein as system used for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject),. System 300 may be used for generating images of structures of the cardiac conduction system pre-procedure, for example, for helping to plan the procedure, for example, when features described with reference to FIG. 1A are performed pre-procedure. Alternatively or additionally, system 300 may be used during a procedure, for example, when features described with reference to FIG. IB are performed during the procedure. For example, for generating overlays over 2D images (e.g., fluoroscopy images) of the heart captured during a transcatheter procedure and / or other features as described herein. An exemplary environment for system 300 is now described. A catheter 15 may be percutaneously inserted into a living body 17 of a patient lying on a gurney 19. Catheter 15 may be controlled and manipulated by operator 70 (e.g., a surgeon). An imaging device 32 captures images of a heart of a patient in real time during a procedure, for example, a fluoroscopy device captures 2D x-rays of the heart. A computing environment 30 may be used to dynamically generate images 18 depicting the 2D image with a visualization of locations of the structures of the cardiac conduction system, such as by projection from a 3D image marked with the location of the cardiac conduction system, as described herein. An image of a heart of the patient is obtained in real time during the procedure. Generated and / or enhanced image 18 may be displayed to an operator 70 on an output display 50. A window 60 may present a zoom-in of structures of the cardiac conduction system presented in image 18.

[0432] Referring now back to FIG. 2, system 100 may implement the acts of the methods described herein optionally by one or more processor(s) 102 of a computing environment 104 executing code instructions stored in a memory 106.

[0433] Computing environment 104 may be implemented as, for example, a client terminal, a server, a virtual server, a radiology workstation, a catheterization laboratory workstation, a virtual machine, a computing cloud, a mobile device, a desktop computer, a thin client, a Smartphone, a Tablet computer, a laptop computer, a wearable computer, glasses computer, and a watch computer. Computing environment 104 may include an advanced visualization process that sometimes is add-on to a catheterization laboratory workstation and / or other devices for generating visualization of the cardiac conduction system (e.g., 3D, 2D), as described herein. Computing environment 104 may include locally stored software that performs one or more of the acts of the method described herein and / or may act as one or more servers (e.g., network server, web server, a computing cloud, virtual server) that provides services (e.g., one or more of the acts of the method described herein) to one or more client terminals 108 (e.g., remotely located catheterization laboratory workstation) over a network 110, for example, providing software as a service (SaaS) to the client terminal(s) 108, providing an application for local download to the client terminal(s) 108, as an add-on to a web browser and / or a medical imaging viewer application, and / or providing functions using a remote access session to the client terminals 108, such as through a web browser.

[0434] Different architectures based on system 100 may be implemented. In one example, computing environment 104 provides centralized services. Computing environment 104 may obtain images from an image repository 122A (e.g., fluoroscopy images captured during a procedure and / or a 3D pre-procedure image such as a CT scan) storing images from one or more different imaging devices 112 (e.g., fluoroscopy machines and / or CT scanners). The images 122A may be provided to computing environment 104 for centralized analysis, for example, for generating a visualization of structures of the cardiac conduction system on a 3D image of the heart, and / or for generating visualization of structures of the cardiac conduction system on a 2D image of the heart obtained during a catheterization procedure, as described herein. The images (e.g. pre-procedure and / or intra-procedure) may be provided to computing environment 104, for example, via an application programming interface (API), software development kit (SDK), and / or other local application, over a network 110, and / or transmitted using a suitable transmission protocol. The generated visualization(s), for example, to client terminal(s) 108 for presentation on a display such as within a GUI and / or other indications (e.g., audio and the like) and / or local storage, stored in an electronic medical record (e.g., hosted by server 118), and / or stored by computing environment 104.

[0435] In another architecture, computing environment 104 provides localized services, by locally analyzing the images (e.g., pre-procedure and / or intra-procedure) and / or locally generating the visualization of structures of the cardiac conduction system. For example, computing environment 104 is implemented as code executed by a processor of a catheterization laboratory workstation that analyses pre-procedure CT scans for generating visualization of the structures of the cardiac conduction system, and / or generates an overlay of the structures of the cardiac conduction system for presentation over fluoroscopy image captured by a fluoroscopy machine of the catheterization laboratory. Computing environment 30 described with reference to FIG. 16. may include and / or be implemented as computing environment 104 described with reference to FIG. 2.

[0436] System 100 and / or computing environment 104 described with reference to FIG. 2. may include and / or be implemented as system 300 described with reference to FIG. 16.

[0437] Imaging device 112 provides the medical images, optionally the pre-procedure image used to generate the visualization of the structures of the cardiac conduction system and / or intraprocedure image on which the structures of the cardiac conduction system may be presented. Imaging device(s) 112 may include a pre-procedure imaging device such as a 3D scanner, for example, a CT and / or MRI machine. Imaging device(s) 112 may include an intra-procedure imaging device for capturing image during the procedure, for example, a 2D fluoroscopy and / or x-ray machine, as commonly used in catheterization labs (e.g., cardiac).

[0438] Imaging device(s) 112 described with reference to FIG. 2 may include and / or be implemented as imaging device 32 described with reference to FIG. 16.

[0439] The pre-procedure images and / or intra-procedures images may be stored in image repository 122A. Image repository 122A may be stored in a data repository 114 and / or data storage device 122, for example, a storage server, a computing cloud, virtual memory, and a hard disk. It is noted that the pre-procedure images and / or intra-procedure images may be stored by a server 118, accessible by computing environment 104 over network 110.

[0440] Computing environment 104 may receive the pre-procedure images and / or intra-procedure images using one or more data interfaces 120, for example, a wire connection (e.g., physical port), a wireless connection (e.g., antenna), a local bus, a port for connection of a data storage device, a network interface card, other physical interface implementations, and / or virtual interfaces (e.g., software interface, virtual private network (VPN) connection, application programming interface (API), software development kit (SDK)), and / or a network interface 124.

[0441] Processor(s) 102 may be implemented, for example, as a central processing unit(s) (CPU), a graphics processing unit(s) (GPU), field programmable gate array(s) (FPGA), digital signal processor(s) (DSP), and application specific integrated circuit(s) (ASIC). Processor(s) 102 may include one or more processors (homogenous or heterogeneous), which may be arranged for parallel processing, as clusters and / or as one or more multi core processing units.

[0442] Memory 106 (also referred to herein as a program store, and / or data storage device) may store code instruction for execution by hardware processor(s) 102, for example, a random access memory (RAM), read-only memory (ROM), and / or a storage device, for example, non-volatile memory, magnetic media, semiconductor memory devices, hard drive, removable storage, and optical media (e.g., DVD, CD-ROM). For example, memory 106 may store code 106A that implement one or more acts and / or features of the methods described herein.

[0443] Computing environment 104 may include a data storage device 122 for storing data, for example, image repository 122A set for storing pre -procedure and / or intra-procedure images (e.g., 2D, 3D), a repository of models and / or rules 122B set for generating the visualization based on an analysis of the 3D image, and a repository of visual representations with indications of structures of the cardiac conduction system 122C set for storing for example, 3D images of different patients labeled with locations of structures of the cardiac conduction system for generating real-time overlays over 2D fluoroscopy images captured during a procedure for indicating the locations of structures of the cardiac conduction system, as described herein. Data storage device 122 may be implemented as, for example, a memory, a local hard-drive, a removable storage device, an optical disk, a storage device, and / or as a remote server and / or computing cloud (e.g., accessed over network 110).

[0444] Computing environment 104 may include network interface 124 for connecting to network 110, for example, one or more of, a network interface card, a wireless interface to connect to a wireless network, a physical interface for connecting to a cable for network connectivity, a virtual interface implemented in software, network communication software providing higher layers of network connectivity, and / or other implementations. Computing environment 104 may access one or more remote servers 118 using network 110.

[0445] It is noted that data interface 120 and network interface 124 may exist as two independent interfaces (e.g., two network ports), as two virtual interfaces on a common physical interface (e.g., virtual networks on a common network port), and / or integrated into a single interface (e.g., network interface). Computing environment 104 may communicate using network 110 (or another communication channel, such as through a direct link (e.g., cable, wireless) and / or indirect link (e.g., via an intermediary computing device such as a server, and / or via a storage device) with one or more of:

[0446] * Client terminal(s) 108, for example, when computing environment 104 acts as a server providing services (e.g., SaaS) to remote catheterization laboratory workstations, for generative visualizations of structures of the cardiac conduction system for remotely obtained images 122 A (e.g., CT scans, fluoroscopy images) and / or for generating an overlay of the structures of the cardiac conduction system.

[0447] * Server 118, for example, implemented in association with a PACS, which may store captured pre-procedure images used to generate the visualization of the structures of the cardiac conduction system. * Imaging device 112 and / or data repository 114 that store images acquired by imaging device 112.

[0448] Computing environment 104 and / or client terminal(s) 108 and / or server(s) 118 may include and / or may be in communication with a user interface(s) 126 that includes a mechanism designed for a user to enter data and / or to view data. Exemplary user interfaces 126 include, for example, one or more of, a touchscreen, a display, a keyboard, a mouse, and voice activated software using speakers and microphone.

[0449] The descriptions of the various embodiments of the present invention have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

[0450] It is expected that during the life of a patent maturing from this application many relevant images will be developed and the scope of the term image is intended to include all such new technologies a priori.

[0451] As used herein the term “about” refers to ± 10 %.

[0452] The terms "comprises", "comprising", "includes", "including", “having” and their conjugates mean "including but not limited to". This term encompasses the terms "consisting of" and "consisting essentially of".

[0453] The phrase "consisting essentially of" means that the composition or method may include additional ingredients and / or steps, but only if the additional ingredients and / or steps do not materially alter the basic and novel characteristics of the claimed composition or method.

[0454] As used herein, the singular form "a", "an" and "the" include plural references unless the context clearly dictates otherwise. For example, the term "a compound" or "at least one compound" may include a plurality of compounds, including mixtures thereof.

[0455] The word “exemplary” is used herein to mean “serving as an example, instance or illustration”. Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and / or to exclude the incorporation of features from other embodiments.

[0456] The word “optionally” is used herein to mean “is provided in some embodiments and not provided in other embodiments”. Any particular embodiment of the invention may include a plurality of “optional” features unless such features conflict. Throughout this application, various embodiments of this invention may be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

[0457] Whenever a numerical range is indicated herein, it is meant to include any cited numeral (fractional or integral) within the indicated range. The phrases “ranging / ranges between” a first indicate number and a second indicate number and “ranging / ranges from” a first indicate number “to” a second indicate number are used herein interchangeably and are meant to include the first and second indicated numbers and all the fractional and integral numerals therebetween.

[0458] It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable subcombination or as suitable in any other described embodiment of the invention. Certain features described in the context of various embodiments are not to be considered essential features of those embodiments, unless the embodiment is inoperative without those elements.

[0459] Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims.

[0460] It is the intent of the applicant(s) that all publications, patents and patent applications referred to in this specification are to be incorporated in their entirety by reference into the specification, as if each individual publication, patent or patent application was specifically and individually noted when referenced that it is to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the present invention. To the extent that section headings are used, they should not be construed as necessarily limiting. In addition, any priority document(s) of this application is / are hereby incorporated herein by reference in its / their entirety.

Claims

WHAT IS CLAIMED IS:

1. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; generating an indication of the cardiac conduction system of the subject according to computed correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features; and wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

2. The computer implemented method of claim 1, wherein the correlations are based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations.

3. The computer implemented method of claim 1 or 2, further comprises computing the correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features.

4. The computer implemented method of claim 1 or 2, further comprises obtaining the correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features.

5. The computer implemented method of any of claims 1-4, wherein the correlations are computed between at least one combination of the plurality of structures of the cardiac conduction system and at least one combination of the plurality of gross anatomical features.

6. The computer implemented method of any of claims 1-5, wherein the indication of the cardiac conduction system comprises a 3D image including the plurality of structures and interpolations therebetween.

7. The computer implemented method of any of claims 1-6, wherein the plurality of gross anatomical features are selected from: non-coronary cusp (NCC), right coronary cusp(RCC), left coronary cusp (LCC), nadir of one or more of NCC RCC and LCC, virtual annulus plane, membranous septum, membranous septum floor, rotation angle between aortic valve and membranous septum, posterior papillary muscle of the left ventricle, anterior papillary muscle of the left ventricle, apex of the left ventricle, medial papillary muscle of the right ventricle, the moderator band of the right ventricle, the anterior papillary muscle of the right ventricle, intraventricular septum, myocardial ring connecting to aortic valve, myocardial ring connecting to mitral valve, myocardial notch between plane of myocardial rings of the aortic valve and mitral valve, base (i.e., inferior border) of the infra pyramidal space (IPS), apex of the IPS, anterior Border of the IPS, posterior Border of the IPS, lateral borders of the IPS, the roof of the inferoseptal recess, anterior / superior border of the inferoseptal recess, posterior / inferior border of the inferoseptal recess, right side of the inferoseptal recess, left side of the inferoseptal recess, diverging vestibules of the mitral and tricuspid valves, the ventricular boundary of the inferoseptal recess, the apex of the inferoseptal recess, medial papillary muscle (MPM), septal band (SB), moderator band (MB), muscle between septum and free wall (myocardium), MB1 - connected to the interventricular septum, and MB2 - connected to the ventricular myocardium.

8. The computer implemented method of any of claims 1-7, wherein the structures of the cardiac conduction system are selected from: atrioventricular (AV) node, bundle of his, penetrating bundle, non-branching bundle, branching point of a left branch bundle (LBB), LBB, right branch bundle (RBB), left anterior fascicle, left posterior fascicle, left septal fascicle, Mahaim fibers, Purkinje fibers, and connection paths between two of the aforementioned along the cardiac conduction system.

9. The computer implemented method of any of claims 1-8, wherein the plurality of structures include an atrioventricular (AV) node estimated as being location within at least one of: (i) fat of the infra-septal recess at the level close to the septal commissure of the mitral annulus, (ii) within the atrioventricular septum, near the opening of the coronary sinus, and part of the triangle of Koch, (iii) located just above the infra septal recess, and (iv) adjacent to an apex of the infra pyramidal space.

10. The computer implemented method of any of claims 1-9, wherein the indication of the cardiac conduction system is generated as a visual marking depicted within a 3D representation based on the 3D image.

11. The computer implemented method of claim 10, further comprising presenting the 3D representation depicting the visual marking on a display and within a graphical user interface configured for rotation of the 3D representation.

12. The computer implemented method of any of claims 1-11, further comprising: receiving a 2D image of the heart; registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image; projecting a 3D indication of the cardiac conduction system to the 2D plane for creating a 2D projection of the indication of the cardiac conduction system; and presenting the 2D projection on the 2D image.

13. The computer implemented method of claim 12, wherein the receiving, the registering, the projecting and the presenting, are dynamically iterated during a procedure for varying poses at which the 2D image is captured.

14. The computer implemented method of any of claims 1-13, further comprising: receiving a 2D image of the heart; registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image; mapping the plurality of gross anatomical features depicted in the 3D image to the 2D image; generating a 2D indication of the cardiac conduction system of the subject according to the computed correlations and according to the mapping; and presenting the 2D indication of the cardiac conduction system as an overlay over the 2D image.

15. The computer implemented method of any of claims 1-14, wherein the structures of the cardiac conduction system are within about 3 centimeters of an aortic valve of the heart.

16. The computer implemented method of any of claims 1-15, further comprising feeding the indication of the cardiac conduction system of the subject into an artificial intelligence (Al) process for further analysis.

17. The computer implemented method of claim 16, wherein the Al process is configured for at least one of: estimating risk of damage to the conduction system, and planning a transcatheter procedure for reducing or avoiding damage to the conduction system.

18. The computer implemented method of any of claims 1-17, wherein the 3D image excludes visualization of the plurality of structures of the cardiac conduction system.

19. The computer implemented method of any of claims 1-18, wherein the indication of the cardiac conduction system is generated by adapting a template of the cardiac conduction system according to the computed correlations.

20. The computer implemented method of any of claims 1-19, wherein the plurality of structures include the AV node, estimated to be in at least one of: (i) a middle of an infra- septal recess, (ii) within an intraventricular septum adjacent to a septal commissure between a posterior leaflet and an anterior leaflet of a mitral valve, and / or (iii) the intraventricular septum at the left most proximal part of the posterior leaflet adjacent to the septal commissure between the posterior and the anterior leaflets of the mitral valve.

21. The computer implemented method of any of claims 1-20, wherein the AV node within the indication of the cardiac conduction system denotes an anatomical region of at least one of: (i) the infra-septal recess, (ii) the intraventricular septum adjacent to the septal commissure between the posterior leaflet and the anterior leaflet of the mitral valve, and / or (iii) the intraventricular septum at the left most proximal part of the posterior leaflet adjacent to the septal commissure between the posterior and the anterior leaflets of the mitral valve.

22. The computer implemented method of any of claims 1-21, wherein the plurality of structures include a non bifurcating bundle including the bundle of His, estimated to be between the estimated location of the AV node to the most posterior part of the floor of the membranous septum.

23. The computer implemented method of claim 22, wherein the non bifurcating bundle including the bundle of His within the indication of the cardiac conduction system denotes an anatomical region between the estimated location of the AV node and the most posterior part of the floor of the membranous septum.

24. The computer implemented method of claim 22, wherein a location of the bundle of His may be estimated as at least one of: (i) originating from the AV node and traveling through the floor of the membranous part of the interventricular septum, in relation to the inferoseptal recess, (ii) as passing close to the inferoseptal recess as the inferoseptal recess descends through the septum, and (iii) being closely associated with the infra pyramidal space at its origin from the AV node.

25. The computer implemented method of any of claims 1-24, wherein the plurality of structures include the penetrating bundle, estimated to be between the most posterior part of the floor of the membranous septum to the most anterior part of the floor of the membranous septum tracking over the floor of the membranous septum.

26. The computer implemented method of any of claims 1-25, wherein the penetrating bundle within the indication of the cardiac conduction system denotes an anatomical region between the most posterior part of the floor of the membranous septum to the most anterior part of the floor of the membranous septum tracking over the floor of the membranous septum.

27. The computer implemented method of any of claims 1-26, wherein the plurality of structures included the bifurcating bundle, estimated to be toward the left ventricle next to the most anterior part of the floor of the membranous septum.

28. The computer implemented method of any of claims 1-27, wherein the bifurcating bundle within the indication of the cardiac conduction system denotes an anatomical region toward the left ventricle next to the most anterior part of the floor of the membranous septum.

29. The computer implemented method of any of claims 1-28, wherein the plurality of structures included the right bundle branch, estimated to be across from the estimated location of the penetrating bundle on the right ventricle wall.

30. The computer implemented method of claim 29, wherein the right bundle branch within the indication of the cardiac conduction system denotes an anatomical region across from the estimated location of the penetrating bundle on the right ventricle wall.

31. The computer implemented method of claim 29, wherein a location of the right branch bundle is estimated as at least one of: (i) originating from the branching bundle during the path of the branching bundle over the membranous septal floor, and (ii) as traveling along the right side of the interventricular septum toward the medial papillary muscle of the right ventricle.

32. The computer implemented method of any of claims 1-31, wherein the plurality of structures include the left bundle branch, on the left ventricle toward the apex of the left ventricle.

33. The computer implemented method of claim 32, wherein the left bundle branch within the indication of the cardiac conduction system denotes an anatomical region on the left ventricle toward the apex of the left ventricle.

34. The computer implemented method of claim 32, wherein a location of the left bundle branch is estimated as at least one of: (i) originating from the branching bundle during a path of the branching bundle over the membranous septal floor, and (ii) as traveling along the left side of the interventricular septum.

35. The computer implemented method of any of claims 1-34, wherein the plurality of structures include the left anterior fascicle, estimated to be along a region defined by a line from below the location of the branching point of the left bundle branch to a base of the anterior papillary muscle of the left ventricle.

36. The computer implemented method of any of claims 1-35, wherein the left anterior fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to a base of the anterior papillary muscle of the left ventricle.

37. The computer implemented method of any of claims 1-36, wherein the plurality of structures include the left posterior fascicle, estimated to be along a region defined by a line from below the location of the branching point of the left bundle branch to a base of the posterior papillary muscle of the left ventricle.

38. The computer implemented method of any of claims 1-37, wherein the left posterior fascicle within the indication of the cardiac conduction system denotes an anatomical regiondefined by a line from below the location of the branching point of the left bundle branch to a base of the posterior papillary muscle of the left ventricle.

39. The computer implemented method of any of claims 1-38, wherein the plurality of structures include the left septal fascicle, estimated to be along a region defined by a line from below the location of the branching point of the left bundle branch to the apex of the left ventricle with a path that passes between the anterior and posterior papillary muscles.

40. The computer implemented method of any of claims 1-39, wherein the left septal fascicle within the indication of the cardiac conduction system denotes an anatomical region defined by a line from below the location of the branching point of the left bundle branch to the apex of the left ventricle with a path that passes between the anterior and posterior papillary muscles.

41. The computer implemented method of any of claims 1-40, wherein the plurality of gross anatomical features include at least a portion of the membranous septum, wherein analyzing comprises analyzing the 3D image for detecting the membranous septum.

42. The computer implemented method of claim 41, wherein the membranous septum is identified as a region between a ventricular septum and an inferoseptal recess, wherein analyzing comprises analyzing the 3D image for detecting the ventricular septum and the inferoseptal recess.

43. The computer implemented method of claim 42, wherein the inferoseptal recess is identified according to at least neighboring tissue detected in the 3D image, selected from:(i) a roof of the inferoseptal recess identified as an area of fibrous tissue continuity between leaflets of a mitral valve and a tricuspid valve, and / or identifying the fibrous tissue continuity as tissue at a base of the atrial septum,(ii) anterior and / or superior border of the inferoseptal recess identified as an aortic leaflet of the mitral valve,(iii) posterior and / or inferior border of the inferoseptal recess identified as a septal surface of the left ventricle,(iv) right wall of the inferoseptal recess identified as fibrous tissue through which an identified His bundle penetrates,(v) left side of the inferoseptal recess identified as a border of a left ventricular cavity,(vi) detecting the diverging vestibules to define the atrial boundary of the infero septal recess,(vii) detecting a crest of the muscular ventricular septum to define the ventricular boundary, and(viii) detecting the apex of the inferoseptal recess to identify the inferior margin of the membranous septum.

44. The computer implemented method of any of claims 1-43, wherein the plurality of gross anatomical features include an infra pyramidal space (IPS), wherein the plurality of structure include an atrioventricular (AV) node, wherein the correlations include a location of the AV node at an end of a tip of a pyramid-like shape of the infra pyramidal space.

45. The computer implemented method of claim 44, wherein the IPS is identified by detecting one or more of the following:(i) base of the IPS identified by detecting a diaphragmatic surface of the heart,(ii) apex of the IPS identified by detecting a central fibrous body of the atrioventricular septum, close to the noncoronary sinus of the aortic root,(iii) anterior Border of the IPS identified by detecting a crest of the interventricular septum, (iv) posterior Border of the IPS identified as being continuous with a detected diaphragmatic surface of the heart at the crossing of the left and right inferior atrioventricular and interatrial grooves, and(v) lateral Borders of the IPS identified by detecting two septal atrial walls.

46. The computer implemented method of any of claims 1-45, further comprising creating a fused image by registering the indication of the cardiac conduction system to an anatomical image, and presenting the fused image depicting the anatomical image with the indication of the cardiac conduction system.

47. The computer implemented method of any of claims 1-46, further comprising creating a fused image by registering the indication of the cardiac conduction system to a functional image, and presenting the fused image depicting the functional image with the indication of the cardiac conduction system.

48. The computer implemented method of claim 47, wherein the functional image depicts at least one of: conduction functionality of nerves of the cardiac conduction system, and viability of the myocardium of the heart innervated by the cardiac conduction system.

49. The computer implemented method of any of claims 1-48, wherein the gross anatomical features include gross anatomical landmarks.

50. The computer implemented method of any of claims 1-49, wherein the indication of the cardiac conduction system is represented as a plurality of spaced apart visual elements represents a probability distribution indicating likelihood of location of structures of the cardiac conduction system, the probability distribution computed from an analysis of the location of the structures of the cardiac conduction system in a plurality of other subjects.

51. The computer implemented method of any of claims 1-50, wherein the indication of the cardiac conduction system is generated excluding data from at least one sensor sensing electrical activity of the cardiac conduction system.

52. The computer implemented method of any of claims 1-51, further comprising: iterating the receiving, the analyzing, the computing, and the generating, for each of a plurality of sample 3D images of sample hearts of sample subjects, for creating a training dataset of a plurality of records, wherein a record includes the sample 3D image, and a ground truth of the indication of the cardiac conduction system generated for the sample 3D image; and training a machine learning model on the training dataset.

53. The computer implemented method of any of claims 1-52, further comprising: identifying a membranous septum, and identifying a membranous septum floor by drawing a highest elevation line of a myocardium and identifying a crossing of the identified membranous septum and the highest elevation line of the myocardium.

54. The computer implemented method of any of claims 1-53, wherein the indication comprises an AV node, and the plurality of gross anatomical features comprise an infrapyramidal space and / or inferoseptal recess, and the AV node is identified as being located closest to an apex of the infrapyramidal space, and / or at an intersection of the infrapyramidal space and inferoseptal recess.

55. The computer implemented method of any of claims 1-54, wherein the indication comprises a perforating bundle and the plurality of gross anatomical features comprise an inferoseptal recess and membranous septum floor, and the perforating bundle is identified at an intersection of the inferoseptal floor and posterior border of the membranous septum floor.

56. The computer implemented method of any of claims 1-55, wherein the indication comprises a non branching bundle of His and an AV node and a perforating bundle, and the plurality of gross anatomical features comprise an inferoseptal recess, and wherein the non branching bundle of His is identified as extending from the AV node to the perforating bundle over a surface of the inferoseptal recess.

57. The computer implemented method of any of claims 1-56, wherein the indication comprises a branching point of the left bundle branch, and the plurality of gross anatomical features comprise a membranous septum floor, wherein the branching point of the left bundle branch is identified as being located at about 60-80% of a length of the membranous septum floor.

58. The computer implemented method of any of claims 1-57, wherein the indication comprises a right bundle branch, and the plurality of gross anatomical features comprise a membranous septum floor, wherein right bundle branch is identified as being located at about 50- 70% of a length of the membranous septum floor.

59. The computer implemented method of any of claims 1-58, wherein the indication comprises a left bundle branch, and the plurality of gross anatomical features comprise a membranous septum floor, wherein left bundle branch is identified as extending along an entire length of the membranous septum floor, and covering an endocardium of a left ventricle for a distance of about 1-4 times a length of the membranous septum, and breaking into three fascicles.

60. A computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; estimating locations of one or more structures of the cardiac conduction system using the plurality of gross anatomical features; andmarking locations of the one or more structures of the cardiac conduction system on a visual rendition of the heart.

61. The computer implemented method of claim 60, wherein the estimating the locations is by applying at least one rule derived from an analysis of a plurality of hearts of different individuals.

62. The computer implemented method of any of claims 60-61, further comprising: creating a training dataset of a plurality of records for a plurality of sample 3D images of a plurality of sample individual, wherein a record includes a sample 3D image and a ground truth of the marking of the locations of the one or more structures of the cardiac conduction system; and training a machine learning model on the training dataset, wherein the machine learning model generates the locations of the one or more structures in response to an input of a target 3D image.

63. The computer implemented method of any of claims 60-62, wherein the estimating the locations is by feeding at least a portion of the 3D image depicting at least one of the gross anatomical features into a machine learning model.

64. The computer implemented method of claim 63, wherein the machine learning model is trained on a training dataset of a plurality of records, wherein a record includes at least a sample of a portion of a sample 3D image of a sample individual depicting the at least one of the gross anatomical features, and a ground truth indicating the locations of the one or more structures.

65. The computer implemented method of any of claims 60-64, further comprising rendering an interpolation of a non-trivial geometry of the cardiac conduction system between the marked locations of the one or more structures.

66. The computer implemented method of claim 65, wherein the rendering the interpolation is according to at least one rule.

67. The computer implemented method of any of claims 60-66, further comprising adapting one or more of the computed correlations between at least one of the gross anatomicalfeatures and at least one of the structures of the cardiac conduction system based on a current medical state of the subject.

68. The computer implemented method of any of claims 60-67, wherein the current medical state is selected from: genetic conditions and acquired disease conditions.

69. The computer implemented method of claim 68, further comprising: dynamically training a machine learning model on published cases describing variation in deviations of locations of structures of the cardiac conduction system for different medical states of individuals; feeding at least one of the computed correlations and the current medical state of the subject into the machine learning model; and obtaining the adaptation of the computed correlation from the machine learning model.

70. A computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; estimating locations of one or more structures of the cardiac conduction system in the 3D image using the plurality of gross anatomical features; and marking locations of the one or more structures of the cardiac conduction system on a 2D image of the heart.

71. The computer implemented method of claim 70, further comprising: receiving the 2D image of the heart; registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image; projecting the 3D estimated location of the one or more structures of the cardiac conduction system to the 2D plane for creating a 2D projection of the one or more structures; and presenting the 2D projection on the 2D image.

72. The computer implemented method of claim 71, wherein the receiving, the registering, the projecting and the presenting, are dynamically iterated during a procedure for varying poses at which the 2D image is captured.

73. The computer implemented method of any of claims 70-72, further comprising: receiving the 2D image of the heart; registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image; mapping the plurality of gross anatomical features depicted in the 3D image to the 2D image; generating a 2D indication of the cardiac conduction system of the subject according to the estimated locations and according to the mapping; and presenting the 2D indication of the cardiac conduction system as an overlay over the 2D image.

74. A computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; and estimating locations of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

75. A computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; and providing an indication of one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

76. The computer implemented method of claim 75, further comprising: computing a registration for registering the 2D image with the 3D image; wherein the plurality of gross anatomical features and the one or more structures of the cardiac conduction system are located within the 3D image, and associated with CT derived information (CDI) indicating coordinates of locations of corresponding structures of the cardiac conduction system within the 3D image; detecting the plurality of gross anatomical features on the 2D image, and extracting fluoroscopy derived information (FDI) indicating the location of the corresponding structures of the cardiac conduction system on the 2D image and a pose of a sensor that captured the 2D image;estimating the location of the one or more structures of the cardiac conduction system on the 2D image according to the detected locations of the plurality of gross anatomical features on the 2D image; computing a pose of the 3D image that is closest to the pose of the sensor that captured the 2D image; applying the registration for projecting the one or more structures of the cardiac conduction system of the 3D image defined by the CDI to a 2D plane corresponding to the FDI, according to the pose for generating a synthetic view; and providing the synthetic view for presentation.

77. The computer implemented method of any of claims 75-76, further comprising: computing a registration for registering the 2D image with the 3D image; wherein the plurality of gross anatomical features are located within the 3D image, and associated with CT derived information (CDI) indicating coordinates of locations thereof within the 3D image; detecting the plurality of gross anatomical features on the 2D image, and extracting fluoroscopy derived information (FDI) indicating the location of the plurality of gross anatomical features on the 2D image and a pose of a sensor that captured the 2D image; computing a pose of the 3D image that is closest to the pose of the sensor that captured the 2D image; applying the registration for projecting the plurality of gross anatomical features of the 3D image defined by the CDI to a 2D plane corresponding to the FDI, according to the pose for generating a synthetic view; wherein the indication of the of the one or more structures of the cardiac conduction system comprises rendering the one or more structures of the cardiac conduction system in the synthetic view according to the locations of the plurality of gross anatomical features.

78. A computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; and providing an indication based on the one or more structures of the cardiac conduction system in a 2D image of the heart using the plurality of gross anatomical features.

79. The computer implemented method of claim 78, wherein the indication based on the one or more structures of the cardiac conduction system comprises rendering an interpolation of a non-trivial geometry of the cardiac conduction system between detected locations of the one or more structures.

80. A computer implemented method for visualizing a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving an image depicting a heart of a subject; feeding the image into a machine learning model; and obtaining from the machine learning model, a fused image comprising the image and a plurality of structures of the cardiac conduction system.

81. The computer implemented method of claim 80, wherein the machine learning model is trained on a training dataset of a plurality of records generated from a plurality of sample 3D images of sample hearts of sample subjects, for creating a training dataset of a plurality of records, wherein a record includes the sample 3D image, and a ground truth of an indication of the cardiac conduction system generated for the sample 3D image; and training a machine learning model on the training dataset.

82. The computer implemented method of claim 81, wherein the indication of the cardiac conduction system is generated for the sample 3D image by: analyzing the sample 3D image for detecting a plurality of gross anatomical features; computing correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features based on an analysis of a plurality of other 3D images of a plurality of other subjects representing anatomical variations; and generating the indication of the cardiac conduction system according to the computed correlations.

83. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting at least one gross anatomical feature; identifying a location of a membranous septum according to the at least one gross anatomical feature; andgenerating an indication of the cardiac conduction system of the subject according to the location of the membranous septum, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

84. The computer implemented method of claim 83, wherein the indication includes a visual element denoting the perforating bundle, bundle of His and a branching point of the LBB are located according to the membranous septum, the perforating bundle located at an anterior location of the membranous septum, the branching point of the LBB located at a posterior point location of the membranous septum, and the bundle of His connecting the perforating bundle and the branching point of the LBB.

85. The computer implemented method of any of claims 83-84, wherein the membranous septum is identified as a region location at a minimum thickness of a septum between a left ventricle and a right ventricle, wherein the region has a predetermined area corresponding to a predicted area of the membranous septum.

86. The computer implemented method of any of claims 83-85, wherein the membranous septum is identified by: identifying a mitral ring of a left ventricle myocardium connected to an mitral valve; identifying a tricuspid ring of a right ventricle myocardium connected to a tricuspid valve; wherein a plane of the mitral ring is at an angle with respect to a plane of the tricuspid ring; identifying the membranous septum at a bump between the mitral ring and the tricuspid ring.

87. The computer implemented method of any of claims 83-86, wherein the membranous septum is found by identifying the location of the membranous septum of a plurality of sample images of a plurality of sample individuals with respect to the at least one gross anatomical feature, clustering the locations with respect to the at least one gross anatomical feature, and applying the clustering to the at least one gross anatomical feature of the 3D image.

88. The computer implemented method of any of claims 83-87, wherein the membranous septum is found by segmenting a myocardium of a left ventricle, and identifying an aortic notch in the myocardium, wherein the membranous septum is located at the aortic notch.

89. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; generating an indication of the cardiac conduction system of the subject by adapting a template according to the plurality of gross anatomical features, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

90. The computer implemented method of claim 89, wherein the template comprises an image of at least a left bundle branch, an anterior fascicle, a posterior fascicle, and a septal fascicle.

91. The computer implemented method of any of claims 89-90, wherein adapting comprises fitting a plurality of anchor points according to the plurality of gross anatomical features, and adapting connections between the plurality of anchor points, and / or re-sizing the template to fit the plurality of gross anatomical features.

92. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; generating a 3D indication of the cardiac conduction system of the subject according to the plurality of gross anatomical features, wherein the 3D indication is defined according to the 3D image; receiving a 2D image of the heart; registering the 2D image with a 2D plane defining a projection of the 3D image, wherein a pose of the 2D plane substantially matches a pose of the 2D image; projecting the 3D indication of the cardiac conduction system to the 2D plane for creating a 2D projection of the indication of the cardiac conduction system; andpresenting the 2D projection on the 2D image.

93. The computer implemented method of claim 92, wherein the registration is performed using features that are different than the plurality of gross anatomical features used to generate the 3D indication of the cardiac conduction system.

94. The computer implemented method of any of claims 92-93, further comprising: computing correlations between a calcification template of the 3D image and the 3D indication of the cardiac conduction system; identifying the calcification template in the 2D image; and registering the 2D image with a 2D plane defining a projection of the 3D image according to the calcification template.

95. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; generating a 3D indication of the cardiac conduction system of the subject according to the plurality of gross anatomical features, wherein the 3D indication is defined according to the 3D image; computing correlations between a calcification template of the 3D image and the 3D indication of the cardiac conduction system; receiving a 2D image of the heart; identifying the calcification template in the 2D image; and computing a 2D indication of the cardiac conduction system for the 2D image according to a registration function that registers the calcification template in the 2D image to the calcification template of the 3D image.

96. A computer implemented method of computing an area of a conduction system at risk during a transcatheter procedure on a subject, comprising: analyzing a combination of at least one personal parameter of the subject and at least one procedure parameter of the transcatheter procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure; and providing the area of the conduction system at risk.

97. The computer implemented of claim 96, wherein the area of the conduction system at risk is presented as an overlay on 2D images captured during the transcatheter procedure.

98. The computer implemented method of claim 97, wherein the area of the conduction system at risk is determined on a 3D image obtained pre-procedure, and further comprising projecting the area of the conduction system to a 2D plane registered with a 2D image obtained during the procedure, wherein a pose of the 2D plane substantially matches a pose of the 2D image.

99. The computer implemented method of any of claims 96-98, wherein the analyzing is performed by feeding the combination of at least one personal parameter of the subject and at least one procedure parameter of the transcatheter procedure into a machine learning model, and obtaining the area of the conduction system at risk as an outcome of the machine learning model, wherein the machine learning model is trained on a training dataset of records, each record including the combination of at least one personal parameter of a sample individual, and at least one procedure parameter of the transcatheter procedure performed on the sample individual, and a ground truth indication of the conduction system at risk.

100. The computer implemented method any of claims 96-99, wherein the at least one personal parameter includes one or more of: subject demographics, mapping data including a location of the conduction system computed using a 3D image obtained pre-procedure, patient history, and ECG conduction analysis.

101. The computer implemented method any of claims 96-100, wherein the at least one procedure parameter includes one or more of: procedure related incidence of conduction disturbances, procedure specific tools related incidence of conduction disturbances, type and size of the tools used in the procedure, indication of stress-strain of the tools being used (e.g., extracted from imaging during procedure), indication of the stress-strain of the tools being used extracted from a strain stress sensor on the tool itself during procedure, and ECG conduction analysis during the procedure.

102. The computer implemented method of any of claims 96-101, wherein the area of the conduction system at risk includes one or more of: sinoatrial (SA) node, atrioventricular (AV) node, the non bifurcating bundle including the bundle of His, the penetrating bundle, the bifurcating bundle, the right bundle, the left bundle, and / or the branches of the left bundle.

103. The computer implemented method of any of claims 96-102, wherein 2D images captured during the transcatheter procedure are dynamically fed into the machine learning model for monitoring the area of the conduction system at risk.

104. The computer implemented method of any of claims 96-103, further comprising determining a risk level for at least one region of the area of the conduction system, and visually coding the at least one region according to the risk level.

105. The computer implemented method of any of claims 96-104, further comprising monitoring that a pressure applied by a transcatheter device to the area of the conduction system at risk remains below a threshold.

106. The computer implemented method of any of claims 96-105, further comprising monitoring that a distance between a transcatheter device and the area of the conduction system remains below a distance threshold.

107. The computer implemented method of any of claims 96-106, further comprising monitoring that an accessory catheter is not present during certain parts of the procedure that may damage the area of the conduction system at risk and / or that the accessory catheter is not placed in proximity to the area of the conduction system at risk where the accessory catheter can cause damage.

108. The computer implemented method of claim 107, further comprising generating a recommendation for a safe location for placement of the accessory catheter for not damaging or reducing risk of damage to the area of the conduction system at risk, below a risk threshold.

109. The computer implemented method of any of claims 107-108, further comprising generating, in response to the monitoring of the accessory location, a first indication of safe expansion of a THV, or a second indication of unsafe expansion of the THV.

110. The computer implemented method of any of claims 96-109, further comprising in response to detecting a location of a transcatheter device likely to damage the area of theconduction system at risk, generating instructions for a mini-robot that is attached to the transcatheter device, for blocking further movement of the transcatheter device.

111. The computer implemented method of any of claims 96-110, further comprising computing a target range for a device depth for deployment of a THV according to the area of the conduction system at risk.

112. The computer implemented method of any of claims 96- 111, wherein transcatheter procedure is selected from a group including: TAV repair, Mitral Valve repair, Tricuspid valve repair, TAV replacement, Mitral Valve replacement, and Tricuspid valve replacement.

113. A computer implemented method of displaying an area of a conduction system at risk during a transcatheter procedure on a subject, comprising: analyzing a 3D image obtained pre-procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure; and presenting the conduction system at risk as an overlay on 2D images captured during the transcatheter procedure.

114. The computer implemented method of claim 113, wherein the area of the conduction system at risk includes one or more of: sinoatrial (SA) node, atrioventricular (AV) node, the non bifurcating bundle including the bundle of His, the penetrating bundle, the bifurcating bundle, the right bundle, the left bundle, and / or the branches of the left bundle.

115. A computer implemented method of reducing risk of conductance disturbance during a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure and monitoring that a pressure applied by a transcatheter device to the area of the conduction system at risk remains below a threshold.

116. The computer implemented method of claim 115, wherein the threshold is lOgr.

117. A computer implemented method of reducing risk of conductance disturbance during a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure andavoiding applying a pressure by a transcatheter device to the area of the conduction system at risk above a threshold.

118. The computer implemented method of claim 117, wherein the threshold is 10 gram (gr).

119. A computer implemented method of guiding a transcatheter procedure on a subject, comprising: analyzing a 3D image obtained pre-procedure for computing an area of a conduction system at risk for being damaged during the transcatheter procedure; and presenting the conduction system at risk as an overlay on 2D images captured during the transcatheter procedure.

120. A computer implemented method of guiding a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure and computing a target range for a device depth for deployment of a THV according to the area of the conduction system at risk.

121. The computer implemented method of claim 120, wherein computing an area of a conduction system at risk comprises analyzing a 3D image obtained pre-procedure.

122. A computer implemented method of guiding a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure, and monitoring that an accessory catheter is not present during certain parts of the procedure that may damage the area of the conduction system at risk and / or that the accessory catheter is not placed in proximity to the area of the conduction system at risk where the accessory catheter can cause damage.

123. The computer implemented method of claim 122, wherein computing an area of a conduction system at risk comprises analyzing a 3D image obtained pre-procedure.

124. A computer implemented method of guiding a transcatheter procedure on a subject, comprising: computing an area of a conduction system at risk during the transcatheter procedure, and monitoring that a distance between a transcatheter device and the area of the conduction system remains below a distance threshold.

125. The computer implemented method of claim 124, wherein computing an area of a conduction system at risk comprises analyzing a 3D image obtained pre-procedure.

126. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 2D image depicting a heart of a subject; obtaining an indication of the cardiac conduction system of the subject; and marking the indication of the cardiac conduction system of the subject on the 2D image; wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

127. The computer implemented method of claim 126, wherein the indication of the cardiac conduction system is based on correlations with a plurality of gross anatomical features identified on a 3D image of the subject, the correlations are based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations.

128. A computer implemented method for generating an image depicting a personalized surrogate of a cardiac conduction system of a subject, comprising: receiving a 3D image depicting a heart of a subject; analyzing the 3D image for detecting a plurality of gross anatomical features; computing correlations between a plurality of structures of the cardiac conduction system and the plurality of gross anatomical features based on an analysis of a plurality of sample 3D images of a plurality of sample subjects representing anatomical variations; and generating an indication of the cardiac conduction system of the subject according to the computed correlations, wherein the indication of the cardiac conduction system denotes a surrogate of the cardiac conduction system personalized for the subject.

129. A computer implemented method for identifying a gross anatomical structures to define at least one structure of a cardiac conduction system, comprising: receiving a 3D image depicting a heart of a subject; identifying an infrapyramidal space in the 3D image; identifying an inferoseptal recess in the 3D image; identifying a membranous septum in the 3D image; and generating an indication of at least one structure of the cardiac conduction system of the subject according to the infrapyramidal space, the inferoseptal recess, and the membranous septum.

130. The computer implemented method of claim 129, wherein the indication of at least one structure of the cardiac conduction system includes an AV node, a Point B denoting a perforating bundle, and a Point C denoting a branching point of the LBB.

131. The computer implemented method of any of claims 129-130, wherein the indication of at least one structure of the cardiac conduction system are identified by applying different HUE levels to the 3D image, the different HUE levels selected to visualize different types of tissues of the heart, wherein the infrapyramidal space and the inferoseptal recess are of a first type of tissue at a first HUE level that is different than a myocardium of the heart of a second type of tissue at a second HUE level.

132. The computer implemented method of any of claims 129-131, wherein the indication comprises an AV node, and the AV node is identified as being located within the infrapyramidal space and closest to the inferoseptal recess.

133. The computer implemented method of any of claims 129-132, wherein the indication comprises Point B, and Point B is identified as being located within the posterior most position of the membranous septum floor.

134. The computer implemented method of any of claims 129-133, wherein the indication comprises Point C, and Point C is identified as being located on the membranous septum floor at a point between Point B and the most anterior point of the membranous septum floor.

135. The computer implemented method of claim 134, wherein Point C is located at about 40-80% of a length of the line between Point B and the most anterior point of the membranous septum floor.

136. An apparatus comprising at least one processor configured to execute a method according to any one of claims 1-135.

Images