Providing guidance for angiographic sequence imaging

Abstract

The subject matter of this disclosure relates to a computer-implemented method for providing guidance for angiographic sequence imaging, the computer-implemented method including receiving a plurality of angiographic images of a region of interest during angiographic sequence imaging, each of the plurality of angiographic images being associated with a viewpoint of a coronary artery structure in the region of interest, identifying one or more segments of the coronary artery structure in each of the plurality of angiographic images, and providing an indication that one or more of the one or more identified coronary artery segments require one or more additional angiographic images associated with a different viewpoint.

Description

[Technical Field]

[0001] The present invention relates to a computer-implemented method for providing guidance for angiographic sequence imaging, a computer-implemented method for providing guidance for percutaneous coronary intervention, a temporary or non-transitory computer-readable storage medium, and a system. [Background technology]

[0002] During angiographic sequence imaging of the 3D structure of arteries, medical professionals must observe each relevant coronary artery segment from at least two different X-ray views. Without acquiring at least two different X-ray views, it is difficult to correctly identify segments of the coronary artery structure that are not discernible from all 2D projections, such as stenoses. This is because some stenoses are highly isotropic, clearly visible in 2D from one perspective but nearly invisible from others. As a result, a lesion may be visible from one orientation of the segment but not from another. In such cases, a single image may provide a false-negative diagnosis of a particular stenosis. As a result, it is difficult to clearly identify coronary artery segments that require interventional endovascular treatment, such as percutaneous coronary intervention. Summary of the Invention [Problem to be solved by the invention]

[0003] The object of the present invention is to improve upon the prior art. [Means for solving the problem]

[0004] According to a first aspect of the present invention, a computer-implemented method for providing guidance for angiographic sequence imaging is provided, the computer-implemented method comprising the steps of receiving a plurality of angiographic images of a region of interest during angiographic sequence imaging, each of the plurality of angiographic images being associated with a viewpoint of a coronary artery structure in the region of interest; identifying a left or right coronary artery tree having a different arrangement of segments and identifying one or more segments of the identified coronary artery tree in each of the plurality of angiographic images; and providing an indication that one or more of the one or more identified coronary artery segments require one or more additional angiographic images associated with different viewpoints, the different viewpoints having clinical relevance to the one or more identified segments that complement the viewpoints in the plurality of angiographic images.

[0005] Thus, if, during identification of a segment of interest in an available angiographic image, the angiographic image is found to have too limited clinical relevance for this particular segment, indications may be provided to guide the clinician in capturing additional angiographic images of different perspectives that complement the clinical relevance of the original image for the identified segment.

[0006] In particular, for segments of interest that contain, for example, a stenosis, a lesion, or an occlusion, this protects the correct identification of such a stenosis, a lesion, or an occlusion when the available angiographic images have too limited clinical relevance for this particular segment of interest. This may, but is not necessarily, due to the identification step revealing that the segment of interest is visible in angiographic images acquired from only a single viewpoint, or from multiple viewpoints that are not sufficiently clear.

[0007] However, in one embodiment, the clinical relevance of a viewpoint is determined by one or more parameters selected from the group of coronary artery segment clutter and coronary artery segment foreshortening. Thus, clutter and foreshortening of segments of interest in available angiographic images are determined in the identification step, and this information is used in determining to provide an indication that additional images need to be acquired from viewpoints having complementary clinical relevance.

[0008] As used herein, "cluttering" refers to the obstruction of a coronary artery segment due to the presence of different structures (in this case, other blood vessels) within the field of view of the detector. "Shortening" refers to the reduction in length of a coronary artery segment that occurs when the segment extends obliquely relative to the imaging plane.

[0009] In one embodiment, the viewpoint includes one or more rotation angles of the x-ray detector when capturing each angiographic image of the plurality of angiographic images, or an angiographic view classification.

[0010] The X-ray detector may be an X-ray detector from a C-arm device.

[0011] In one embodiment, the computer-implemented method may further include determining a viewpoint associated with each of the angiographic images by applying each angiographic image to a neural network model trained to determine a viewpoint for angiographic images, or, if the viewpoint comprises one or more rotational angles of the X-ray detector at the time of capturing each angiographic image of the plurality of angiographic images, determining an angiographic view classification associated with each of the angiographic images using metadata associated with each of the angiographic images, wherein the metadata describes the one or more angles of the X-ray detector.

[0012] In either case, the clinician's workload is reduced. When using neural networks, the viewpoint can be determined even when metadata is not available. When metadata is available, it means that there is no need to train the neural network, which is computationally more efficient.

[0013] In one embodiment, identifying one or more segments of the coronary artery structure in each angiographic image includes retrieving a coronary artery segment associated with the viewpoint from a database that maps viewpoints to coronary artery segments.

[0014] Retrieving the coronary artery segments from a database is a computationally efficient way to obtain the coronary artery segments.

[0015] In one embodiment, identifying one or more segments of coronary artery structure in each angiographic image includes applying each angiographic image to a neural network model trained to identify one or more coronary artery segments from angiographic images, where a well-trained neural network omits the need for reliable identification of the viewing angle.

[0016] In one embodiment, the computer-implemented method may further include determining, for each of the one or more segments, a number of different viewpoints associated with each of the identified one or more segments of the coronary artery structure in the plurality of angiographic images; and comparing the determined number of different viewpoints associated with each of the one or more coronary artery segments with a threshold number of viewpoints, wherein providing an indication that one or more coronary artery segments require one or more additional angiographic images associated with different viewpoints includes identifying one or more coronary artery segments for which the number of different viewpoints is below the threshold number. For example, guidance is provided to ensure that at least two viewpoints are available to the physician for a vessel segment of interest that includes a stenosis. In other words, the threshold number may be two.

[0017] In one embodiment, identifying the left or right coronary artery tree from each angiographic image is performed using a neural network. Once the coronary artery tree is identified, it becomes easier to identify the coronary artery segments of interest, since left and right arterial trees have different arrangements of segments. This is most beneficial when segment visibility is defined by the acquisition angle (as opposed to being assessed by a network).

[0018] Knowing the angle and using guidelines to determine which segments are well visible, the information needed from the image is whether this is a right or left arterial view being imaged. In one embodiment, providing the indication includes displaying a map of the coronary artery structure, the map including one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different perspective than the remaining one or more coronary artery segments.

[0019] In one embodiment, displaying the one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different perspective from the remaining one or more coronary artery segments includes displaying the one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different perspective in a different color than the other coronary artery segments. Color-coding the segments in the map is intuitive for quick and easy interpretation by a clinician.

[0020] In one embodiment, the map of the coronary artery structure includes at least one of a pictorial map and a textual map.

[0021] According to one aspect, a computer-implemented method for providing guidance for percutaneous coronary intervention is provided, the computer-implemented method comprising: guiding angiographic sequence imaging according to the computer-implemented method of any preceding aspect or embodiment; receiving an image of a region of interest during the percutaneous coronary intervention; and displaying, on a display, the image of the region of interest along with an indication of one or more identified coronary artery segments requiring one or more angiographic images associated with different viewpoints.

[0022] In this way, the clinician performing the intervention can understand in real time whether a segment is a good candidate for balloon placement, for example.

[0023] In one embodiment, the computer-implemented method further includes determining a coronary artery segment position of an interventional tool during percutaneous coronary intervention in the received images, wherein providing an indication of one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint comprises providing an indication that the identified coronary artery segment where the interventional tool is to be positioned requires one or more additional angiographic images associated with a different viewpoint in response to determining that the interventional tool is to be positioned in one of the one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint.

[0024] In one embodiment, the computer-implemented method further includes a step of aligning an image of the region of interest received during the percutaneous coronary intervention with one of a plurality of angiographic images of the region of interest received during the angiographic sequence imaging, and determining a coronary artery segment position of an interventional tool during the percutaneous coronary intervention based on the image of the region of interest received during the percutaneous coronary intervention is based on aligning the image of the region of interest received during the percutaneous coronary intervention with one of a plurality of angiographic images of the region of interest received during the angiographic sequence imaging.

[0025] According to one aspect, a transient or non-transitory computer-readable storage medium is provided that stores instructions that, when executed by a processor, cause the processor to perform the computer-implemented method of any preceding claim.

[0026] According to one aspect, a system is provided having a storage device including a non-transitory computer-readable storage medium of the preceding aspect stored thereon, at least one processor configured to execute instructions from the non-transitory computer-readable storage medium, and a display configured to display an indication of one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint and / or display received images of a region of interest along with an indication of identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint.

[0027] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

[0028] Embodiments of the present invention may be best understood with reference to the accompanying drawings. [Brief explanation of the drawings]

[0029] [Figure 1] A schematic cross-sectional view of a C-arm device is shown. [Figure 2] 1 shows a block diagram of a system for performing a method according to an embodiment of the present invention. [Figure 3] 1 shows a flowchart of steps involved in a computer-implemented method for providing guidance for angiographic sequence imaging and a computer-implemented method for providing guidance for percutaneous coronary intervention according to an embodiment of the invention. [Figure 3A] 3 shows the display 310 from FIG. 3 in more detail. [Figure 4] Figure 3 shows a schematic diagram illustrating well-visible and shortened coronary artery segments from angiographic images captured during angiographic sequence imaging. [Figure 5] 1 shows a schematic diagram illustrating the left and right coronary artery trees. [Figure 6] 4 shows a flowchart summarizing a computer-implemented method for providing guidance for angiographic sequence imaging from FIG. 3. [Figure 7] 4 shows a flowchart summarizing the computer-implemented method for providing guidance for percutaneous coronary intervention from FIG. 3. DETAILED DESCRIPTION OF THE INVENTION

[0030] At least some of the example embodiments described herein may be constructed, in part or in whole, using dedicated, special-purpose hardware. As used herein, terms such as "component," "module," or "unit" may include, but are not limited to, hardware devices, such as circuits in the form of discrete or integrated components, field programmable gate arrays (FPGAs), or application-specific integrated circuits (ASICs), that perform particular tasks or provide related functionality. In some embodiments, the described elements may be configured to reside on tangible, persistent, addressable storage media and to execute on one or more processors. These functional elements, in some embodiments, may include, by way of example, components such as software components, object-oriented software components, class and task components, processes, functions, attributes, procedures, subroutines, program code segments, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. While example embodiments are described with reference to components, modules, and units discussed herein, such functional elements may be combined into fewer elements or separated into additional elements. It will be understood that various combinations of optional features are described herein, and that the described features may be combined in any suitable combination. In particular, features of any one example embodiment may be combined with features of any other embodiment, as appropriate, except where such combinations are mutually exclusive. Throughout this specification, the terms "having" or "including" mean the inclusion of the specified components, but do not exclude the presence of others.

[0031] The methods described herein may be computer-implemented methods. In particular, the methods may include at least a computer-implemented method for generating a synthetic video and a computer-implemented method for training one or more machine learning algorithms. A computer implementing the methods may include a storage device, such as a memory, and a processor. The computer may be a hardware computer, and thus the storage device and the processor may be a hardware storage device and a hardware processor, respectively.

[0032] The computer-implemented method may be provided as instructions stored on a transitory or non-transitory computer-readable medium. The computer-readable medium may be stored in a computer storage device. When the instructions are executed by a processor, the instructions cause the processor to perform any of the method steps described herein.

[0033] Angiographic imaging is a diagnostic medical imaging technique that utilizes X-rays to view coronary artery structures, such as the coronary artery tree or arteries. An angiographic image is a 2D X-ray image of the coronary artery structure, including multiple coronary artery segments. During angiographic imaging, a medical professional aims to image the coronary artery structure from several perspectives to obtain an angiogram with a view of each of the coronary artery segments of the coronary artery structure. If only one perspective of the coronary artery structure is acquired, information present in the dimension orthogonal to the imaging plane is not captured. For example, a dangerous narrowing (i.e., stenosis) of a coronary artery segment is not observable in a single perspective of the coronary artery structure if the coronary artery structure extends orthogonal to the imaging plane.

[0034] Once identified, stenoses can be treated using endovascular procedures. For example, percutaneous coronary intervention treatment involves the insertion of stents and / or balloons over a guidewire into a patient's coronary anatomy to widen and support the narrowed coronary artery segment. Contrast injection is typically not used by medical professionals during percutaneous coronary intervention. Instead, alternative imaging techniques are used to obtain images in which the interventional tools are visible.

[0035] To enable angiographic sequence imaging, the patient is first injected with a contrast agent that is opaque to X-rays, and then multiple angiographic images are acquired using an X-ray detecting medical device, such as a C-arm device.

[0036] 1 , a C-arm device 100 typically includes an X-ray source 140 and an X-ray detector 150 disposed at either end of an arc-shaped arm 130. The source 140 and the detector 150 are disposed opposite each other, thereby forming a line of sight between the source 140 and the detector 150, through which a signal emitted by the source 140 can be detected by the detector 150. During imaging, a patient is positioned on a table surface between the source 140 and the detector 150 of the C-arm 130 such that the line of sight impinges on the patient's body. For a given position of the line of sight relative to the patient's body, the C-arm 130 acquires an image having a specific perspective of the patient's body.

[0037] The C-arm 130 is connected to a base structure 110, such as a stand, by a support arm 120. The C-arm 130 is movable with respect to the base structure 110 by at least rotational movement. The C-arm 130 is movable with respect to a table surface and a patient thereon by translational and rotational movement. With respect to translational movement, the C-arm 130 is movable in horizontal and vertical planes. With respect to rotational movement, the C-arm 130 is movable with orbital rotation 160 about the geometric center of the C-arm's arc, i.e., rotation within the plane of the C-arm. Additionally, the C-arm is movable with angular rotation 170 about the axis of the support arm to provide propeller-like rotation of the C-arm.

[0038] The C-arm 130 can thus be repositioned to obtain multiple angiograms of the coronary artery structure, where each angiogram is associated with a different viewpoint relative to the patient. For example, for a particular region of interest, such as a particular translational position of the patient's body, the C-arm 130 can be repositioned by orbit and / or angular rotation. Thus, each angiogram is associated with a viewpoint that includes one or more of the translational position, orbital rotation (i.e., orbital angle) 160, and angular rotation (i.e., angle) 170 of the C-arm 130. The translational component can be used to position the heart at the C-arm isocenter. Therefore, the C-arm angle is more important in identifying which segments are visible.

[0039] The metadata for each angiogram image includes the viewpoint information described above. The angiograms and corresponding metadata may be stored on the C-arm device 100 and / or a remote server communicatively linked to the C-arm device 100.

[0040] Angiographic view classification corresponds to the classification of views with respect to orbital (rotational) and / or angular positions of the C-arm, i.e., achievable by roll and propeller motions, respectively. Coronary artery segments have different levels of visibility in different perspectives. Clinically relevant views for specific segments of interest provide good visibility of these specific coronary artery segments.

[0041] In particular, clinically relevant viewpoints are associated with minimal cluttering and / or shortening of the imaged coronary artery segment of interest, allowing for proper diagnosis of, for example, vessel stenosis.

[0042] Typically, angiography views are classified as either a right anterior (RAO) view, a left anterior (LAO) view, or an anteroposterior (AP) view. These view classifications refer to the angular orientation (propeller motion) of the C-arm 130 relative to the patient. For example, an RAO view corresponds to an angle of the C-arm 130 such that the patient's right side is in the field of view of the detector. Conversely, an LAO view corresponds to an angle of the C-arm 130 such that the patient's left side is in the field of view of the detector. An AP view corresponds to an angle of the C-arm 130 such that the detector is perpendicular to the patient.

[0043] Each of the RAO, LAO, and AP views can be further classified as cranial or caudal. These view classifications refer to the orbital rotation direction (roll motion) of the C-arm 130 relative to the patient. The cranial view classification corresponds to orbital rotation of the C-arm 130 toward the patient's head. Conversely, the caudal view classification corresponds to orbital rotation of the C-arm 130 toward the patient's feet.

[0044] The methods disclosed herein may be performed by a system 200 as shown in Figure 2. The system 200 includes a storage device 210 that includes a non-transitory computer-readable storage medium, a processor 220, and a display 230.

[0045] 2, the system is configured to receive a plurality of angiogram images. For example, the system 200 may be communicatively linked to a server that is communicatively linked to the C-arm device 100. The system 200 may receive the angiograms from the server via a wireless connection, such as Bluetooth or Wi-Fi, or by a physical connection, such as a cable. Alternatively, the system may receive the angiograms from a storage means directly connected to the system.

[0046] With further reference to FIG. 2 , the system display 230 is configured to display an indication of one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different perspective, and / or is configured to display the received image of the region of interest along with an indication of the identified coronary artery segments requiring one or more additional angiographic images associated with a different perspective.

[0047] In particular, the perspective of the additional angiographic images may complement the perspective of the original angiographic image in terms of clinical relevance. For example, the additional angiographic images may visualize the identified coronary artery segments with less foreshortening and / or less clutter. The system display 230 may further be configured to display one or more angiographic images along with the received image and indication of the region of interest.

[0048] Figure 3 shows an overview of two computer-implemented methods. The first computer-implemented method provides guidance for angiographic sequence imaging. The second computer-implemented method provides guidance for percutaneous coronary intervention.

[0049] 3, in a method for providing guidance during a diagnostic procedure, a plurality of angiographic images 300 are captured using a C-arm during an angiographic image sequence. A viewpoint 302 of a coronary artery structure in a region of interest of the angiographic images is determined.

[0050] As mentioned above, the viewpoint includes one or more rotation angles of the X-ray detector when capturing each angiographic image of the plurality of angiographic images, or angiographic view classification.

[0051] If determining viewpoint 302, the method may include determining viewpoint 302 associated with each of the angiographic images 300 by applying each angiographic image to a neural network model trained to determine viewpoints for angiographic images. In other embodiments, if the viewpoint comprises one or more rotational angles of the x-ray detector at the time of capturing each angiographic image of the plurality of angiographic images, the method may include determining an angiogram view classification associated with each of the angiographic images 300 using metadata associated with each of the angiographic images. The metadata describes one or more angles of the x-ray detector at the time the associated angiographic image was captured.

[0052] In other embodiments, the view classification can be determined by applying the angiographic image to a neural network that is trained to determine the view classification from features in the angiographic image, which may be the first neural network.

[0053] The method also includes identifying one or more segments 304 (or visible segments) of the coronary artery structure in each angiographic image of the plurality of angiographic images 300 .

[0054] In some embodiments, this step includes retrieving the coronary artery structure associated with the viewpoint from a database that maps coronary artery structures to viewpoints. The mapping may be based on medical guidelines. Each angiographic image may be subjected to a neural network classifier trained to identify coronary artery structures, e.g., left or right, from angiographic images. The coronary artery structure may be used to more accurately retrieve the specific segment or segments. The neural network classifier may be a third neural network. The neural network classifier may be a convolutional neural network.

[0055] In other embodiments, this step includes applying each angiographic image to a neural network model trained to identify one or more coronary artery segments directly from the angiographic image. This method eliminates the need to retrieve segments from a database. The neural network may be a second neural network, which may be a convolutional neural network. The neural network may be trained using angiographic images labeled with any coronary artery segments visible in the angiographic image.

[0056] By "visible" it is meant that the segments are shown in a way that does not obscure the segments, for example, the segments are not shown with abbreviations.

[0057] Referring briefly to Figure 4, a segment 400 is shown. An orthogonal projection 402 of the depicted segment and a lesion 404 can be seen on that projection. Such an orthogonal projection 402 is considered satisfactory because there is no foreshortening of the segment 400. In other words, in an orthogonal projection, the segment is well visible and may be considered to have high clinical relevance for the segment 400, provided there is no clutter as in this schematic representation.

[0058] A longitudinal projection 406 is also shown in FIG. 4, where the segment is shown in-line in the angiogram. In other words, the detector is positioned along the longitudinal axis of the segment. The longitudinal projection 406 shows a shortened segment when the lesion is not visible. In other words, when shortening is present, the segment is less visible.

[0059] As in this example, an angiographic image showing a segment with significant shortening may be deemed to have insufficient clinical relevance for the clinical problem at hand, e.g., determining the severity of a stenosis. In this case, acquiring additional angiographic images from a viewpoint is required, the additional images complementing the original image in terms of clinical relevance. In this particular example, a viewpoint is selected for the additional angiographic image showing the segment of interest with no or little shortening.

[0060] Coronary artery structure can be best understood with reference to FIG. 5, which shows the right coronary artery structure RCA or coronary artery tree 410 and the left coronary artery structure LCA or coronary artery tree 412. Both the RCA 410 and LCA 412 arterial trees include multiple segment arrangements. The segment arrangements between the arterial trees differ. The right arterial tree 410 includes segments 1 through 4 and 16 through 16c. The left arterial tree 412 includes segments labeled 5 through 14b. As an example, FIG. 5 shows the arrangement of coronary artery segments according to the American Heart Association classification.

[0061] As shown in Figure 5, since the arterial tree extends in three dimensions, several orientations of the detectors mean that the segment can be anywhere from orthogonal to longitudinal to the longitudinal axis of the segment.

[0062] With further reference to FIG. 3 , the method may also include determining, for each of the one or more segments, a number of different viewpoints associated with each of the identified one or more segments of the coronary artery structure in the plurality of angiographic images, and comparing the determined number of different viewpoints associated with each of the one or more coronary artery segments with a threshold number of viewpoints. In some embodiments, the viewpoint threshold may be 2. In other words, the method may check whether two different viewpoints exist for each segment. A clinical recommendation is to ensure that each segment is observed from at least two different viewpoints to ensure that significant stenosis characteristic risks are not missed. The different viewpoints may be substantially different. For example, the different viewpoints may be viewpoints that show the segments substantially orthogonal to each other.

[0063] The method may include providing an indication that one or more of the one or more identified coronary artery segments require one or more additional angiographic images associated with different perspectives having complementary clinical relevance. This may be accomplished by displaying a map on display 310. Display 310 may correspond to display 230 from FIG. 2. FIG. 3A shows an expanded view of display 310 from FIG. 3.

[0064] The map may include a pictorial map showing the coronary artery structure (left and right coronary artery trees 412, 410) on the display along with a textual map 308 of the coronary artery structure. The textual map 308 may be a table. The table may include two columns. The first column includes a segment number, e.g., the segment number from FIG. 5. The second column may include a description, name, or label of the angiogram viewpoint captured and from which the segment is visible. The textual map 308 may be color-coded. In other embodiments, the pictorial map may also be color-coded. The color-coding may indicate the number of views of that segment. The color-coding may be achieved by color-coding the segment number in the first column and the viewpoint label in the second column.

[0065] In one example, the color coding may include red when there are no clinically relevant angiographic images captured showing the segment, yellow, orange, or amber when there is one angiographic image showing the segment but the clinical relevance of the viewpoint is questionable, and green when the number of angiographic images showing the segment in different orientations is equal to or greater than a threshold.

[0066] Thus, a map showing a segment in green means that the available angiographic images have sufficient clinical relevance for the segment of interest and no additional images need to be acquired for the clinical task at hand. On the other hand, for segments shown in yellow or red, additional images with complementary clinical relevance may be required. Optionally, viewpoints with complementary clinical relevance may be suggested by the system, or alternatively, the physician may rely on established clinical guidelines when selecting viewpoints that have a reasonable likelihood of complementing the already available views.

[0067] Also, as mentioned above, displaying one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint differently from the remaining one or more coronary artery segments includes displaying one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint in a different color than the other coronary artery segments.

[0068] In other embodiments, providing an indication may be active in the sense of actively alerting the clinician upon capturing different views of any segment where the number of different clinically relevant views falls below a threshold number.

[0069] In the second method, multiple fluoroscopic images 320 are captured during a percutaneous coronary intervention procedure.

[0070] The method may include determining a coronary artery segment position of an interventional tool during percutaneous coronary intervention in the received images. The interventional tool may be a deflated balloon. This may first include balloon segmentation 322, which may be performed using an object identification model. The object identification model may be, for example, a "You Only Look Once" (YOLO) algorithm or a Retinanet algorithm.

[0071] Unlike diagnostic procedures in which contrast agents are used, fluoroscopic images do not show blood vessels because no contrast agents are used during the procedure. To determine the segment location of the balloon, the fluoroscopic image can be overlaid on an angiographic image with the same perspective. This overlay is achieved using an association algorithm 326. The coronary artery segment location of the balloon is then determined using one or more of the same methods described above for the diagnostic method. In other words, the segment location method 328 is used to determine the segment location of an interventional tool, e.g., a balloon.

[0072] In this manner, determining the coronary artery segment position of an interventional tool during percutaneous coronary intervention is based on registering an image of the region of interest received during percutaneous coronary intervention with one of multiple angiographic images of the region of interest received during angiographic sequence imaging.

[0073] A display 330 may be present during the interventional procedure. The display may be the same display as display 310 in some embodiments and / or the same display as display 230 of FIG. 2 in some embodiments. Display 330 may display a fluoroscopic image and a map (from a diagnostic procedure) side-by-side. In this manner, the method may include displaying an image of the region of interest (fluoroscopic image) on display 330 along with an indication of one or more identified coronary artery segments that require one or more angiographic images associated with different perspectives. This approach of displaying indications is passive in the sense that the clinician can make their own decision regarding the suitability of a segment for balloon deployment.

[0074] In other embodiments, the indication may be actively displayed. In other words, in response to determining that an interventional tool is to be placed in one of one or more identified coronary artery segments that requires one or more additional angiographic images associated with a different perspective, the method may include providing an indication that the identified coronary artery segment in which the interventional tool is placed requires one or more additional angiographic images associated with a different perspective. This may be accomplished by outputting a warning when the balloon reaches a steady-state position within a segment for which too few angiographic images from different perspectives have been captured during the diagnostic phase.

[0075] The above method can be summarized with reference to FIGS.

[0076] 6, in accordance with one or more embodiments, a computer-implemented method for providing guidance for angiographic sequence imaging is provided, comprising steps of: receiving S100 a plurality of angiographic images of a region of interest during angiographic sequence imaging, wherein each angiographic image of the plurality of angiographic images is associated with a viewpoint of a coronary artery structure in the region of interest; identifying S102 a left or right coronary artery tree having a different arrangement of segments and identifying one or more segments of the identified coronary artery tree in each angiographic image of the plurality of angiographic images; and providing S104 an indication that one or more of the one or more identified coronary artery segments require one or more additional angiographic images associated with a different viewpoint, wherein the different viewpoint has clinical relevance to the one or more identified segments that complement the viewpoint in the plurality of angiographic images. 7, a computer-implemented method for providing guidance for percutaneous coronary intervention is provided, comprising the steps of: guiding angiographic sequence imaging S200 according to the embodiment of the computer-implemented method associated with FIG. 6; receiving an image of a region of interest during percutaneous coronary intervention S202; and displaying, on a display S204, the image of the region of interest together with an indication of one or more identified coronary artery segments requiring one or more angiographic images associated with different viewpoints.

[0077] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. The invention is not limited to the disclosed embodiments.

[0078] Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. 1. A computer-implemented method for providing guidance for angiographic sequence imaging, comprising: receiving a plurality of angiographic images of a region of interest during angiographic sequence imaging, each of the plurality of angiographic images being associated with a viewpoint of a coronary artery structure in the region of interest; identifying a left or right coronary artery tree having a different arrangement of segments, and identifying one or more of the segments of the identified coronary artery tree in each angiographic image of the plurality of angiographic images; providing an indication that one or more of the one or more identified coronary artery segments requires one or more additional angiographic images associated with a different perspective, the different perspective having a clinical relevance to the one or more identified segments that complements the perspective in the plurality of angiographic images; 10. A computer-implemented method comprising:

2. The computer-implemented method of claim 1 , wherein the viewpoint comprises one or more rotation angles of an X-ray detector when capturing each angiographic image of the plurality of angiographic images, or an angiogram view classification.

3. determining the viewpoint associated with each of the angiographic images by applying each angiographic image to a neural network model trained to determine the viewpoint of angiographic images; or if the viewpoint includes one or more rotational angles of an x-ray detector when capturing each angiographic image of the plurality of angiographic images, determining an angiographic view classification associated with each of the angiographic images using metadata associated with each of the angiographic images, the metadata describing the one or more angles of the x-ray detector; 3. The computer-implemented method of claim 1 or 2, further comprising:

4. identifying one or more segments of the identified coronary artery tree in each angiographic image, retrieving a coronary artery segment associated with said viewpoint from a database mapping coronary artery segments to viewpoints; 4. The computer-implemented method of claim 1, comprising:

5. The step of identifying one or more segments of the coronary artery tree in each angiographic image comprises: applying each angiographic image to a neural network model trained to identify one or more coronary artery segments from the angiographic image; 4. A computer-implemented method according to any one of claims 1 to 3, comprising:

6. For each of the one or more segments, determining a number of different viewpoints associated with each of the identified one or more segments of the coronary artery structure in the plurality of angiographic images; comparing the determined number of different viewpoints associated with each of the one or more coronary artery segments to a threshold number of viewpoints; and providing an indication that the one or more coronary artery segments require one or more additional angiographic images associated with different viewpoints comprises identifying the one or more coronary artery segments for which the number of different viewpoints is below the threshold number; 6. A computer-implemented method according to any one of claims 1 to 5.

7. The computer-implemented method of claim 1 , wherein the clinical relevance of the viewpoint is determined by one or more parameters selected from the group of coronary artery segment clutter and coronary artery segment shortening.

8. 8. The computer-implemented method of claim 1, wherein the step of providing the indication comprises displaying a map of the coronary artery structure that includes the one or more identified coronary artery segments that require one or more additional angiographic images associated with a different perspective, differently from the remaining one or more coronary artery segments.

9. 9. The computer-implemented method of claim 8, wherein the step of displaying the one or more identified coronary artery segments requiring one or more additional angiographic images associated with a different viewpoint differently from the one or more remaining coronary artery segments comprises displaying the one or more identified coronary artery segments requiring the additional angiographic images associated with a different viewpoint in a different color than other coronary artery segments.

10. 10. The computer-implemented method of claim 7, wherein the map of coronary artery structure comprises at least one of a pictorial map and a textual map.

11. 1. A computer-implemented method for providing guidance for percutaneous coronary intervention, comprising: guiding an angiographic sequence imaging by the computer-implemented method of claims 1 to 10; receiving an image of the region of interest during a percutaneous coronary intervention; displaying, on a display, the image of the region of interest along with the indication of the one or more identified coronary artery segments requiring one or more angiographic images associated with different viewpoints; 10. A computer-implemented method comprising:

12. determining a coronary artery segment position of an interventional tool during the percutaneous coronary intervention in the received images; and providing the indication of one or more identified coronary artery segments requiring one or more additional angiographic images associated with the different perspectives, In response to determining that the interventional tool is positioned on one of the one or more identified coronary artery segments that requires one or more additional angiographic images associated with a different viewpoint, providing an indication that the identified coronary artery segment on which the interventional tool is positioned requires one or more additional angiographic images associated with a different viewpoint; 12. The computer-implemented method of claim 11, comprising:

13. registering the image of the region of interest received during the percutaneous coronary intervention with one of the plurality of angiographic images of the region of interest received during angiographic sequence imaging; and determining a coronary artery segment position of an interventional tool during the percutaneous coronary intervention based on an image of the region of interest received during the percutaneous coronary intervention is based on registering the image of the region of interest received during the percutaneous coronary intervention with one of the plurality of angiographic images of the region of interest received during angiographic sequence imaging; 13. The computer-implemented method of claim 12.

14. A transitory or non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the computer-implemented method of any one of claims 1 to 13.

15. a storage device including the non-transitory computer-readable storage medium of claim 14 stored thereon; at least one processor configured to execute the instructions from the non-transitory computer-readable storage medium; displaying the indication of one or more identified coronary artery segments requiring one or more additional angiographic images associated with the different perspectives; and / or displaying the received image of the region of interest along with the indication of the identified coronary artery segments requiring one or more additional angiographic images associated with the different perspectives. a display configured as follows: A system having:

Images