Occluded blood vessel visualization

Abstract

A system for providing an image representation of an occluded blood vessel is provided. The system includes one or more processors configured to: detect a portion (140p, 140d) of the blood vessel that adjoins the blockage (130) in a sequence (120) of angiographic images; extracting a motion vector (150) from the sequence (120) of angiographic images based on the detected portion of the blood vessel, the motion vector representing motion of the portion (140p, 140d) of the blood vessel; registering the sequence (120) of angiographic images to each other based on the extracted motion vectors (150); processing image intensities in a sequence of mutually registered angiographic images in a region of interest adjacent to the portion (140p, 140d) of the blood vessel to provide an image representation of the occluded blood vessel; and outputting an image representation of the occluded blood vessel.

Description

Technical Field

[0001] This disclosure relates to providing an image representation of an occluded blood vessel. A system, a computer-implemented method, and a computer program product are disclosed. Background Technology

[0002] Over time, blood vessels can become blocked, or in other words, clogged. Blockages can occur for a variety of reasons, including due to the buildup of plaque or the development of clots in the blood vessels. Blockages can occur in blood vessels in various parts of the anatomy, including the heart, brain, lungs, legs, and more.

[0003] In some cases, the blockage is partial, and blood flow in the vessel is restricted but not blocked. For example, in the heart, a partial blockage in a coronary artery can cause the heart to have to work harder to maintain blood flow. However, in more severe cases, blood flow in the vessel may be completely blocked. Chronic total occlusion, or "CTO," is defined as a complete blockage of a vessel for a duration of 3 months or more.

[0004] If a blood vessel becomes blocked, a procedure may be necessary. An example of such a procedure is balloon angioplasty. Balloon angioplasty opens the lumen of a blood vessel, restoring blood flow. Various interventional devices can be used in this procedure, including guidewires, (balloon) catheters, etc. Balloon angioplasty typically involves placing a stent in the blood vessel to keep the lumen open.

[0005] Angiographic images are commonly used in medical research related to occluded blood vessels. X-ray angiography, which involves taking X-ray images after a contrast agent has been injected into the vascular system, is typically used for this purpose. The contrast agent introduces a difference in X-ray attenuation between the surrounding tissue and the blood, thereby improving the contrast between the vessel lumen and the surrounding tissue in X-ray angiography. However, limited blood volume, or, in the case of CTO, a lack of blood and therefore a lack of contrast agent in the occlusion itself, results in poor visibility of the occlusion in X-ray angiography. Thus, an occluded vessel may appear only as blurred image features in the angiography caused by the vessel wall, calcification, and attenuation of any contrast agent attempting to pass along the lumen. The limited amount or absence of blood in the occlusion also results in poor visibility of the occlusion in angiography generated using other types of imaging modalities. This presents challenges in researching, planning, and implementing occlusion-related treatment procedures.

[0006] Therefore, there is a need to improve the visualization of blocked blood vessels. Summary of the Invention

[0007] According to one aspect of this disclosure, a system for providing an image representation of a blocked blood vessel is provided. The system includes one or more processors configured to:

[0008] Receive image data representing a sequence of angiographic images of blockages in blood vessels;

[0009] Detect the portion of the blood vessel adjacent to the blockage in the sequence of angiographic images;

[0010] Motion vectors are extracted from the sequence of angiographic images based on the detected portion of the blood vessel, the motion vectors representing the motion of the detected portion of the blood vessel;

[0011] The sequences of angiography images are mutually registered based on the extracted motion vectors;

[0012] The image intensity in the sequence of mutually registered angiographic images is processed in the region of interest adjacent to the portion of the blood vessel to provide the image representation of the occluded blood vessel; and

[0013] Output an image representation of the blocked blood vessel.

[0014] Compared to non-occluded vessels, occluded vessels are typically only poorly visible in angiographic images. In the system described above, angiographic images are cross-registered using motion vectors determined based on the motion of detected portions of adjacent occluded vessels.

[0015] Registering images to each other using adjacent occluded portions of the blood vessel ensures accurate registration of images near the occlusion. In other words, local motion compensation and registration are performed between sequences of images, focusing on the detected portion of the blood vessel directly near the occlusion. This portion of the blood vessel is also visible in the angiographic images because blood reaches that portion of the vessel. The image intensity in the sequence of mutually registered angiographic images in the region of interest adjacent to this portion of the blood vessel is then processed to provide an image representation of the occluded blood vessel. In particular, the image representation can be an enhanced image representation of the occluded blood vessel, wherein the visibility of the occluded blood vessel is enhanced compared to the sequence of angiographic images. For this purpose, techniques such as time integration can be used, for example, to process the image intensity.

[0016] Because image processing operates on images that are accurately co-registered near the occlusion, the blurred image features of the occluded vessel caused by the attenuation of the vessel wall, calcification, and any contrast agent attempting to pass along the vessel lumen are well aligned in the regions of interest within the mutually registered images. This facilitates their processing to provide an accurate enhanced image representation of the occluded vessel. Therefore, image enhancement techniques, which operate on locally registered images that are well aligned in the direct vicinity of the occluded vessel, make visible what might only be faintly present in the original image data and indistinguishable within it.

[0017] Other aspects, features, and advantages of this disclosure will become apparent from the following description of examples taken with reference to the accompanying drawings. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating an example of a system 100 for providing an image representation of a blocked blood vessel, according to some aspects of this disclosure.

[0019] Figure 2 This is a flowchart illustrating an example of a computer-implemented method for providing an image representation of a blocked blood vessel according to some aspects of this disclosure.

[0020] Figure 3 An example of a sequence 120 of angiographic images of a blockage 130 in a blood vessel 140, according to some aspects of this disclosure, is illustrated.

[0021] Figure 4 The illustration shows a portion 140 of the detection of adjacent blockage 130 of a blood vessel according to some aspects of this disclosure. p and 140 d Examples of angiographic images.

[0022] Figure 5 It is a portion 140 representing an adjacent blockage of a blood vessel 130 according to some aspects of this disclosure. p and 140 d A schematic illustration of an example of extracting the motion vector 150 of the motion.

[0023] Figure 6 This is a first example of an image representation 160 of a blocked blood vessel according to some aspects of this disclosure, the image representing an extension of the blood vessel adjacent to the blockage toward the blockage.

[0024] Figure 7 This is a second example of an image representation 160 of a blocked blood vessel according to some aspects of this disclosure, the image representing an extension of the blood vessel adjacent to the blockage toward the blockage.

[0025] Figure 8This is a third example of an image representation 160 of a blocked blood vessel according to some aspects of this disclosure, the image representing an extension of the blood vessel adjacent to the blockage toward the blockage. Detailed Implementation

[0026] Examples of this disclosure are provided with reference to the following description and accompanying drawings. In this specification, for purposes of explanation, numerous specific details of certain examples are set forth. References to “example,” “implementation,” or similar language in the specification mean that a feature, structure, or characteristic described in connection with that example is included in at least one example. It should also be understood that a feature described with respect to one example may also be used in another example, and for the sake of brevity, not all features are necessarily repeated in every example. For example, features described with respect to a system may be implemented in a corresponding manner in a computer-implemented method and a computer program product.

[0027] In the following description, reference is made to examples of systems used to provide image representations of blocked blood vessels. In some examples, the blocked blood vessel is a coronary artery, i.e., a blood vessel located in the heart. However, it should be understood that coronary arteries and the heart are used only as examples of the location and type of blocked blood vessels. Generally, a blocked blood vessel can be any type of blood vessel, including, for example, arteries or veins. Generally, a blocked blood vessel can be in any location in the anatomical structure, including in the heart, brain, lungs, legs, etc. This document also refers to an example where the blockage is a CTO. However, it should be understood that a blockage can generally be any type of blockage. For example, a blockage can alternatively be a partial blockage.

[0028] Note that the computer-implemented methods disclosed herein can be provided as a non-transient computer-readable storage medium including computer-readable instructions stored thereon, which, when executed by at least one processor, cause at least one processor to perform the method. In other words, the computer-implemented methods can be implemented in a computer program product. The computer program product can be provided by dedicated hardware or hardware capable of running software in association with appropriate software. When provided by a processor, the functionality of the method features can be provided by a single dedicated processor, or by a single shared processor, or by multiple individual processors, some of which may be shared. The functionality of one or more of the method features can be provided, for example, by a processor shared within a networked processing architecture such as a client / server architecture, a peer-to-peer architecture, the Internet, or the cloud.

[0029] The explicit use of the terms "processor" or "controller" should not be construed as specifically referring to hardware capable of running software, but may implicitly include, but is not limited to, digital signal processor (DSP) hardware, read-only memory (ROM) for storing software, random access memory (RAM), non-volatile storage devices, etc. Furthermore, examples of this disclosure may take the form of a computer program product accessible from a computer-usable storage medium or a computer-readable storage medium, which provides program code for use by or in connection with a computer or any instruction execution system. For the purposes of this specification, a computer-usable storage medium or a computer-readable storage medium can be any means that may include, store, transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. The medium may be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system or device or a propagation medium. Examples of computer-readable media include semiconductor or solid-state memory, magnetic tape, removable computer disks, random access memory (RAM), read-only memory (ROM), rigid disks, and optical disks. Current examples of optical discs include CD-ROM (CD-Read-Only Memory), CD-R / W (CD-Read / Write), and Blu-ray. TM and DVD.

[0030] It should also be noted that some operations described as being performed by one or more processors of the system disclosed herein can be implemented using artificial intelligence techniques. Suitable techniques may include machine learning techniques as well as deep learning techniques such as neural networks. For example, one or more neural networks can be trained in a supervised or, in some cases, unsupervised manner to implement operations performed by one or more processors.

[0031] As mentioned above, there is a need to improve the visualization of blocked blood vessels.

[0032] Figure 1 This is a schematic diagram illustrating an example of a system 100 for providing an image representation of a blocked blood vessel, according to some aspects of this disclosure. Figure 2 This is a flowchart illustrating an example of a computer-implemented method for providing an image representation of a blocked blood vessel according to some aspects of this disclosure. It should be noted that the method described is... Figure 1 The operations performed by one or more processors 110 of the system 100 shown can also be performed in Figure 2 Execute in the method shown. Similarly, regarding the reference... Figure 2 The methods described can also be described by Figure 1 One or more processors 110 of the system 100 shown execute.

[0033] refer to Figure 1 and Figure 2The system 100 for providing an image representation of a blocked blood vessel includes one or more processors 110 configured to:

[0034] Receive image data sequence 120 representing an angiographic image of a blockage 130 included in blood vessel 140, S110;

[0035] Detecting a portion of the adjacent occlusion 130 of vessel S120 in sequence 120 of the angiographic image 140. p 140 d ;

[0036] Based on the detected blood vessels, S130, representing the blood vessel portion 140, is extracted from sequence 120 of the angiography image. p 140 d The motion vector of the motion is 150;

[0037] Based on the extracted motion vector 150, the sequences 120 of the angiography images are mutually registered S140;

[0038] Processing S150 in the portion 140 with blood vessels p 140 d The image intensity in adjacent regions of interest within a sequence of mutually registered angiographic images is used to provide an image representation of the occluded vessel 160; and

[0039] The output image of the blocked blood vessel is represented as S160.

[0040] As mentioned above, blocked vessels are typically poorly visible in angiographic images compared to non-blocked vessels. In the system described above, angiographic images are mutually registered using motion vectors determined based on the motion of detected portions of the vessel adjacent to the blocked portion. Mutual registration using portions of the vessel adjacent to the blocked portion ensures accurate registration of the images near the blockage. This portion of the vessel is also visible in the angiographic image because blood reaches that portion. The image intensity in the region of interest adjacent to this portion of the vessel in the sequence of mutually registered angiographic images is then processed to provide an image representation of the blocked vessel. For example, techniques such as time integration can be used to process the image intensity. This image representation can be referred to as an enhanced image representation of the blocked vessel because the visibility of the blocked vessel is enhanced compared to the sequence of angiographic images. Since this processing is based on accurately co-registered images, the blurred image features of the blocked vessel caused by the vessel wall, calcification, and attenuation of any contrast agent attempting to pass along the vessel lumen are well aligned in the region of interest of the mutually registered images. This facilitates their processing to provide an accurate image representation of the blocked vessel.

[0041] The operations performed by system 100 are described in more detail below.

[0042] In operation S110, image data is received. The image data represents a sequence 120 of angiographic images of the blockage 130 in the blood vessel 140.

[0043] The sequence 120 of angiographic images received in operation S110 can be generated by various types of imaging systems. These include two-dimensional imaging systems such as projection X-ray imaging systems, and volumetric imaging systems such as computed tomography (CT) imaging systems and magnetic resonance imaging (MRI) systems. In the example where the sequence 120 of angiographic images is generated by a projection X-ray imaging system or a CT imaging system, the imaging system can be a spectral X-ray projection imaging system or a spectral CT imaging system, respectively. Such an imaging system generates X-ray attenuation data representing X-ray attenuation within multiple different energy intervals. The X-ray attenuation data generated by such an imaging system can be processed (e.g., using various material decomposition algorithms) to distinguish media that have similar X-ray attenuation values ​​when measured within a single energy interval and are indistinguishable in X-ray attenuation data obtained using a conventional projection X-ray imaging system or a conventional CT imaging system. Therefore, X-ray attenuation data generated by a spectral X-ray projection imaging system or a spectral CT imaging system can be processed to provide angiographic images with improved specificity for materials such as contrast agents, tissues, and bone.

[0044] Continuing with operation S110, in some examples, a sequence 120 of angiographic images is generated after a contrast agent is injected into the subject's vascular system. The contrast agent is used to improve the contrast between the blood and surrounding tissue in the angiographic image 120. In these examples, the angiographic image 120 may be referred to as a contrast-enhanced angiographic image. The sequence of contrast-enhanced angiographic images may be generated by a projection X-ray imaging system, a CT imaging system, or an MRI system. MRI angiographic images may alternatively be generated without a contrast agent and are generally referred to as non-contrast-enhanced MR angiography "NC-MRA" images.

[0045] Continuing with operation S110, in one example, the images in the sequence 120 of angiographic images are generated immediately before being received by one or more processors 110. In other words, the sequence 120 of angiographic images can be a live sequence 120 of angiographic images. Alternatively, the sequence 120 of angiographic images may have been generated over a period of seconds, minutes, hours, or even days or longer before being received by one or more processors 110. In the latter case, the sequence 120 of angiographic images may have been generated before performing medical procedures on the anatomical region. In this case, the sequence 120 of angiographic images may be referred to as the preoperative sequence 120 of angiographic images.

[0046] Referring again to operation S110, the sequence 120 of angiographic images can be received from various sources during operation S110, including from a medical imaging system such as one of the imaging systems described above. Alternatively, the sequence 120 of angiographic images can be received from another source, such as from a computer-readable storage medium, the Internet, or the cloud. Typically, the sequence 120 of angiographic images can be received via any form of data communication, including wired and wireless communication. By way of examples, when using wired communication, communication can occur via cable or fiber optic cable, and when using wireless communication, communication can occur, for example, via RF or infrared signals.

[0047] Figure 3 An example of a sequence 120 of angiographic images of a blockage 130 in a blood vessel 140, according to some aspects of this disclosure, is illustrated. Figure 3 The angiography image 120 shown represents the vascular system of the heart. Figure 3 The angiography image 120 shown is a contrast-enhanced X-ray angiography image generated after the contrast agent has been injected into the object. A projection X-ray imaging system (such as...) can be used. Figure 1 The C-arm-based projection X-ray imaging system 210 shown is used to generate... Figure 3 The image shown is a contrast-enhanced X-ray angiography image. Figure 3 In the image 120, vessel 140 includes occlusion 130. Occlusion 130 extends between points marked A and A'. In this example, the occlusion completely obstructs blood flow, and occlusion 130 is CTO. Because there is no blood and therefore no contrast agent along the length of occlusion 130, occlusion 130 is poorly visible in the angiographic image 120.

[0048] Now refer to the above reference. Figure 1 and Figure 2 The aforementioned operation S120; in this operation, a portion 140 of an adjacent occlusion 130 of a blood vessel is detected in a sequence 120 of angiographic images. p 140d As referenced above. Figure 3 Although the blockage between points A and A' in vessel 140 is generally poorly visible due to the absence of contrast agent along the blocked portion of the vessel, the adjacent blocked portion of the vessel is visible in the angiographic image because blood reaches that portion of the vessel. This portion of vessel 140 can be referred to as the "remnant" of the vessel because, apart from this portion, the vessel appears truncated in the angiographic image 120.

[0049] exist Figure 3 In the example shown, two portions of the adjacent blockage 130 of the blood vessel are visible: the proximal portion terminating at point A and the distal portion beginning at point A'. In this context, the terms "proximal" and "distal" refer to their relative positions with respect to the direction of normal blood flow in the vessel. Typically, in operation S120, a single portion of the adjacent blockage 130 of the blood vessel 140, i.e., the proximal portion or the distal portion, or alternatively both, i.e., the proximal and distal portions, can be detected. The portion of the blood vessel detected in operation S120 can be a point, such as the distal or proximal end of the blood vessel adjacent to the blockage 130. Alternatively, it can be a longer portion of the blood vessel, such as a segment of the blood vessel.

[0050] Figure 4 The illustration shows a portion 140 of the detection of an adjacent blood vessel blockage 130 according to some aspects of this disclosure. p and 140 d Example of angiographic images. In this example, two portions of the blood vessel are examined, namely the proximal portion 140. p and distal portion 140 d In this example, part 140 was detected. p and 140 d This indicates the segment representing the adjacent blockage of the blood vessel (130).

[0051] In operation S120, a portion 140 is intended for detecting adjacent occlusion of a blood vessel. p 140 d Various techniques are available. These include segmentation techniques as well as feature detection techniques. An example of a segmentation technique that can be used for this purpose is described in the document “Automated Design of Deep Learning Methods for Biomedical Image Segmentation” by Isensee, F. et al. (pp. 1–55, https: / / arxiv.org / pdf / 1904.08128.pdf).

[0052] Now refer to the above reference. Figure 1 and Figure 2The aforementioned operation S130; in this operation, based on the detected portion of the blood vessel, the portion representing the blood vessel 140 is extracted from the sequence 120 of the angiography image. p 140 d The motion vector is 150. (Reference) Figure 5 Describe this operation. Figure 5 It is a portion 140 representing an adjacent blockage of a blood vessel 130 according to some aspects of this disclosure. p and 140 d A schematic illustration of an example of extracting the motion vector 150 of the motion. Figure 5 a and Figure 5 Figure b illustrates angiographic images 1200 and 1201, respectively, of the blockage 130 included in vessel 140 at times t=t0 and t=t1. Figure 5 a and Figure 5 Each image in b illustrates the detected portion of occlusion of adjacent blood vessel 140, which was identified in the previous operation S120. p and 140 d Due to heart movement, the portion of blood vessel 140 detected 140. p and 140 d Its position changes between times t=t0 and t=t1. This is in Figure 5 c) is illustrated in the diagram. Figure 5 c) The illustration shows the portion of blood vessel 140 that was detected. p and 140 d Position relative to a fixed reference position.

[0053] In operation S130, by determining the positional changes of the detected portions 140p and 140d of the blood vessels in the angiography image, motion vectors 150 representing the motion of the portions 140p and 140d of the blood vessels are extracted from the sequence 120 of the angiography image. Motion vector 150 may, for example, represent the portion 140p of the blood vessel. p 140 d Motion relative to a reference image in the sequence. For example, the reference image could be at time t=t0. Figure 5 The image shown in section a.

[0054] In one example, section 140 describing blood vessels is used. p 140 d The motion vector is expressed using a parametric model of motion over time. The parametric model can represent a portion of the blood vessel (140). p 140 d Two-dimensional motion over time. For example, a parametric model can express the motion as a portion of a blood vessel (140°). p 140 dA combination of translation, rotation, and optional stretching along the axis of the blood vessel over time.

[0055] Now refer to the above reference. Figure 1 and Figure 2 The mentioned operation S140; in this operation, the angiographic images in sequence 120 are mutually registered based on the extracted motion vectors 150. (See reference...) Figure 5 In this operation, motion vector 150 is used to register the image to a common reference image. For example, the image can be registered with a reference image at time t=t0. Figure 5 Image registration is shown in Figure a. This results in the images in the time series being based on the detected blood vessels in portion 140. p 140 d The images are aligned with each other. The registration performed in this operation can be rigid, or alternatively, it can be non-rigid, i.e., elastic. Rigid registration can be used when the vessel 140 undergoes relatively minor deformation between images in the sequence. However, if the vessel 140 undergoes significant deformation, elastic registration can be used. If the vessel 140 undergoes significant deformation, using elastic registration in subsequent operation S150 can improve the accuracy of the image representation of the occluded vessel.

[0056] Typically, two or more images can be registered in operation S140. The angiographic images registered in operation S140 correspond to a time period. This time period can typically be any specified time interval. By way of examples, the images can correspond to a complete cardiac cycle, multiple cardiac cycles, or portions of a cardiac cycle. In some examples, all images in sequence 120 within a specified time period can be registered in operation S140. However, in other examples, a selection of images from the specified time period, i.e., an appropriate subset, can be registered. Some of the motion vectors 150 extracted in operation S130 may be anomalous because they indicate excessive motion or because they indicate portions of vessel 140. p 140 d Unexpected locations. This could be due to poor accuracy in detecting portions of the blood vessel in operation S120, or poor accuracy in extracting motion vector 150 in operation S130. In such cases, operation S140 can be performed for selection of images from the time series, for example, fitting the extracted motion vector 150 to a proper subset of images within the envelope of the expected motion. Omitting such anomalous motion vectors yields a more accurate representation of the blocked blood vessel in subsequent operation S150.

[0057] Now refer to the above reference. Figure 1 and Figure 2 The mentioned operation S150; in this operation, the portion 140 with the blood vessel is processed. p 140d Image intensity in adjacent regions of interest within a sequence of mutually registered angiographic images to provide an image representation of the occluded vessel 160.

[0058] In operation S150, the portion 140 connected to the blood vessel is processed. p 140 d Image intensity in adjacent regions of interest within a sequence of mutually registered angiographic images. Typically, this operation may include processing image intensity only within the region of interest, or it may include processing image intensity holistically across each angiographic image in the sequence, and thus processing across portions inherently including blood vessels 140. p 140 d The image intensity of a region adjacent to the region of interest. The region of interest can be defined in various ways. For example, the region of interest can be defined by one or more portions of a blood vessel. p 140 d The region of interest is defined by a pre-defined shape, such as a rectangle or ellipse. Alternatively, the region of interest can be defined as a portion of the angiographic image that covers the expected path of the blocked vessel. For example, the region of interest can have a pre-defined length along the expected path and / or a pre-defined width on either side of the expected path of the blocked vessel.

[0059] Various techniques can be used to perform the processing performed in operation S150. In one example, one or more processors 110 are configured to process the image intensity in a sequence of mutually registered angiographic images in S150 by performing temporal integration of the image intensity in a region of interest over multiple mutually registered images. Temporal integration is used to enhance the intensity of weak vascular features in the angiographic images, such as vessel walls, calcifications, and any contrast agents that may be present along the occlusion 130. Thus, the result of temporal integration is to provide an image representation of the occluded vessel, wherein the visibility of the occlusion is enhanced compared to its visibility in the sequence of angiographic images 120. Another result of temporal integration is to provide a portion 140 of the vessel adjacent to the occlusion 130 in the image representation 160 of the occluded vessel. p 140 d The extension toward the blockage 130. This extension is a result of increased visibility of the blockage. Temporal integration can also have the beneficial effect of blurring background image features around the blocked vessel. This is because, due to the way the motion vector is calculated, corresponding to the movement of the blocked vessel, the blocked vessel may move differently from the background image features, as in the case of the heart. Registration relative to the blocked vessel then has the effect of producing misaligned background features in the registered sequence, which are thus blurred and attenuated when temporal integration is applied to the sequence.

[0060] Various techniques can be used to perform temporal integration of image intensity. In one example, temporal integration involves calculating a weighted average of pixel intensities in mutually registered images within a time series. For instance, temporal integration could include applying weights to pixel intensities when selecting the last N images in the time series, and calculating the pixel intensities in image representation 160 as a weighted average of the corresponding pixel intensities. The weights applied to an image can be the same for all pixels and for all N images. Alternatively, the weights can be different for different images and / or they can differ within an image. For example, different weights can be applied to an image based on the estimation accuracy of one or more motion vectors of the image or the image quality. Alternatively, different weights can be applied inside and outside the region of interest (ROI) of the image to emphasize the importance of features within the ROI relative to features outside the ROI.

[0061] In one example, the region of interest includes calcified vascular segments. Calcified vascular segments are commonly present in occlusions because vascular calcification is a stage in the development of some types of occlusions. Calcification has relatively higher X-ray attenuation than vascular tissue, and therefore calcified vascular segments can be used to indicate the path of an occluded vessel in the absence of contrast agents. Therefore, if the region of interest includes calcified vascular segments, processing operation S150 (e.g., temporal integration of image intensity in the region of interest across multiple mutually registered images) improves the enhancement of image intensity within the vessel, and thus provides a clearer representation of the occluded vessel in image representation 160.

[0062] Instead of performing time integration of image intensity in operation S150, other techniques can be alternatively used to provide an image representation 160 of the occluded blood vessel. For example, a trained neural network can be used to predict the image representation 160 of the occluded blood vessel based on mutually registered angiographic images provided by operation S140. The neural network can be trained using a sequence of mutually registered images of the occluded blood vessels, and for each sequence, the neural network is trained using a corresponding ground truth image of the blood vessel in which the occlusion is visible. The ground truth image can be obtained using an imaging modality different from the angiographic image 120, and wherein the occlusion is visible. For example, the ground truth image may include an external ultrasound image or a CT image. In this example, the image representation 160 predicted by the neural network is used to provide an image representation of the occluded blood vessel, wherein the visibility of the occlusion is enhanced compared to its visibility in the sequence of angiographic images 120.

[0063] Now refer to the above reference. Figure 1 and Figure 2 The mentioned operation S160; in this operation, an image representation of the blocked blood vessel is output 160. The image representation can look similar to Figure 3The angiographic image 120 shown further includes image features representing a blockage 130. This image representation can be referred to as an enhanced image representation of the blocked vessel because the visibility of the blocked vessel is enhanced compared to a sequence of angiographic images.

[0064] Image representation 160 can be output in various ways, including output to display devices, such as output to... Figure 1 The monitor 220 shown may output to a virtual / augmented reality display device, a printer, or a computer-readable storage medium, etc. Typically, the image representation 160 may be provided as a single image, or it may be provided as video.

[0065] An image representation 160 in the form of a single image can be generated from a sequence 120 of two or more angiographic images by performing detection S120, extraction S130, mutual registration S140, and processing S150 on the images in the sequence and then outputting a single image representation 160 of the images in the sequence S160.

[0066] Alternatively, the image representation 160 in video form can be generated by outputting an image representation 160 for a specified number (e.g., k≥2) of angiographic images 120 and then updating the image representation 160 when taking into account additional images from the sequence 120 of angiographic images. For example, the image representation 160 for a specified number (e.g., k≥2) of angiographic images 120 can be generated by performing detection S120, extraction S130, mutual registration S140, and processing S150 on all k images. The image representation 160 for the k images can then be output. The image representation 160 can then be updated whenever additional images from the sequence are taken into account by performing detection S120, extraction S130, mutual registration S140, and processing using the additional images. In this example, the visibility of the occlusion in the video image representation 160 is improved when each additional image is taken into account. Furthermore, due to its improved visibility, the portion 140 of the adjacent occlusion 130 of the vessel is also improved when each subsequent image is taken into account. p 140 d Extending toward blockage 130.

[0067] In one example, one or more processors 110 are configured to predict portions 140 of adjacent occlusions 130 of blood vessels. p 140 dThe extension toward blockage 130. In this example, image representation 160 includes the predicted extension. Various techniques can be used to predict the extension in this example. These include the local minimum path technique described in the document “Fast objectsegmentation by growing minimal paths from a single point on 2D or 3D images” by Benmansour, F. et al. (J Math Imaging Vis 33, 209–221, 2009, https: / / doi.org / 10.1007 / s10851-008-0131-0). As another example, a minimum path technique such as fast-moving minima can be used to predict the extension. Such minimum path techniques attempt to find the path that minimizes the value of the cost function. Alternatively, the extension in this example can be predicted using a neural network, such as nnU-Net described in the document by Isensee, F. et al. cited above.

[0068] exist Figures 6-8 The illustration shows various examples of image representations of blocked blood vessels, including such extensions of portions adjacent to the blocked blood vessel. Figures 6 to 8 The illustrations in the diagrams are represented by the extensions indicated by the dashed lines in each image.

[0069] In one example, a single extension is predicted from multiple angiographic images. Figure 6 The image below illustrates this example. Figure 6 This is a first example of an image representation 160 of a blocked blood vessel according to some aspects of this disclosure, the image representation 160 including an adjacent portion of the blocked blood vessel extending toward the blockage. In this example, multiple angiographic images (such as...) Figure 5 Operations S120, S130, S140, and S150 are performed on the two angiographic images shown in 5a) and 5b). This produces an image representation 160 of the occluded vessel. In operation S150, the image representation 160 of the occluded vessel can be generated by temporal integration of the image intensity of the region of interest on the two mutually registered images. Figure 5 a) and Figure 5 Compared to the individual angiography images shown in b), time integration provides a clearer representation of the occluded vessel. Then, the aforementioned minimum path technique is used to predict from the image representation of the occluded vessel 160. Figure 6 The extension is shown. This extension is then output in the image representation of the blocked blood vessel, for example, as a dashed line, such as... Figure 6 As shown.

[0070] Refer to the above Figure 6In the described technique, the result of a time integration operation can be used to predict a single extension, where any number of angiographic images from the sequence are considered. Typically, the sharpness of an image representation 160 of a clogging vessel increases with the number of angiographic images used in the time integration operation. However, clogging vessels can cause distortions throughout the sequence of angiographic images. This distortion can limit the reliability of the time integration operation because it tends to blur clogging vessels in the image representation 160. In turn, vessel distortion can limit the reliability of the extension predicted based on the result of time integration. In one example, the effect of such vessel distortion is mitigated by iteratively predicting the extension. In this example, one or more processors 110 are configured to iteratively perform extraction S130, mutual registration S140, processing S150, and prediction, wherein each iteration uses portions 140 of the vessel from the previous iteration. p 140 d The predicted extension is used as the detected portion of the blood vessels extracted from S130.

[0071] The effect of iteratively performing the above operations is to provide a more reliable location of the portion of the blood vessel used as the starting point for integration at each time step. This, in turn, improves the reliability of the predicted extension, and so on.

[0072] The result of a single iteration in the example above corresponds to the single extension described above. Figure 7 The diagram illustrates an example of the image representation provided after two iterations of the above operations. Figure 7 This is a second example of an image representation 160 of a blocked blood vessel according to some aspects of this disclosure, which includes an adjacent portion of the blocked blood vessel extending toward the blockage. Figure 8 The diagram illustrates an example of the image representation provided after three iterations of the above operations. Figure 8 This is a third example of an image representation 160 of a blocked blood vessel according to some aspects of this disclosure, which includes an adjacent portion of the blocked blood vessel extending toward the blockage. Figure 8 The extended diagram shown illustrates the full length of blockage 130, i.e. Figure 3 The complete length of the blockage between points marked A and A' in vessel 130 shown. From Figures 6 to 8 As can be understood from the illustration, in this example, both the predicted length of the extension and the clarity of the blockage increase with each iteration.

[0073] In operation S160, Figures 6-8 The image shown represents 160, which can be output as a video. The video illustrates the incremental increase in the predicted extension of the blood vessel in each iteration. It is output as a video showing the growth of the extension in each iteration. Figures 6-8 The images shown help users observe the extended development, instead of outputting in video format. Figures 6-8 The image shown represents 160, and a single image representation 160 can be output in operation S160, for example, representing Figure 8 The image shown represents the final iteration or an intermediate iteration in the iterative process, as shown in Figure 160.

[0074] As described above, typically, in operation S120, a single portion of the adjacent blockage 130 of the blood vessel 140 can be detected, namely the proximal portion or the distal portion, or alternatively both, namely the proximal and distal portions. Referring to the latter case, in one example, the above reference... Figure 1 The described system is used to detect the proximal portion of adjacent occlusion 130 140 of vessel S120. p And extract S130 to represent the proximal portion 140 of the blood vessel. p The motion vector of the motion is 150. In this example, one or more processors 110 are also configured to:

[0075] Detecting the distal portion of the adjacent occlusion 130 of vessel 140 in sequence 120 of angiographic images 140 d ;

[0076] Based on the detected distal portion of the blood vessel, the distal portion 140 representing the blood vessel is extracted from sequence 120 of angiographic images. d The motion vector of the motion;

[0077] Furthermore, one or more processors 110 are configured to also be based on the distal portion 140 representing a blood vessel. d The extracted motion vectors are used to register the sequence of angiographic images 120 to each other 140.

[0078] In this example, the distal portion 140 of the blood vessel d The detection and representation of the distal portion of the blood vessel 140 d The extraction of the motion vector of the motion can be done in the same way as above for the proximal part 140. p The same method described is used. According to this example, using both the proximal and distal portions of the blood vessel can improve the reliability of the image representation of the blood vessel, because the path of the blood vessel is constrained at both ends by the detected portions of the blood vessel.

[0079] In one example, an image representation 160 of a blocked blood vessel is registered to a live sequence of fluorescence fluoroscopic images. In this example, one or more processors 110 are configured to:

[0080] Receive fluorescence fluoroscopic image data, which represents a live sequence of fluorescence fluoroscopic images of the blockage 130 included in the blood vessel 140;

[0081] The image representation of the blocked blood vessel was 160-degree registered to a live sequence of fluorescence fluoroscopy images; and

[0082] One or more processors 110 are configured to output a live sequence of fluorescence transillumination images, and wherein one or more of the following are performed:

[0083] The live sequence of the image representation 160° and fluorescence perspective image is displayed as an overlay image;

[0084] The images represent a live sequence of 160° and fluorescence perspective images, which are displayed adjacent to each other;

[0085] Extract the path of the blocked blood vessel 140 from image representation 160 and display it in a live sequence of fluorescence fluoroscopic images;

[0086] One or more features of the blocked blood vessel 140 are extracted from the image representation 160 and displayed in a live sequence of fluorescence fluoroscopic images.

[0087] Live sequences of fluoroscopic images are often acquired during the procedure to navigate the interventional device to the occlusion. Fluoroscopic images indicate radiopaque materials (such as interventional devices) and also indicate dense portions of anatomical structures, such as bone. However, the vascular system and soft tissues are poorly represented in fluoroscopic images. To visualize the vascular system, contrast agents can be intermittently injected into the vascular system during the fluoroscopic imaging procedure. However, the use of contrast agents poses health risks, and therefore, it is desirable to reduce the use of contrast agents during fluoroscopic imaging. Furthermore, in cases of occlusion, the use of intermittently injected contrast agents is completely ineffective.

[0088] In this example, the registration of the image representation 160 of the occluded vessel with a live sequence of fluoroscopic images provides a spatial correspondence between the occluded vessel and the fluoroscopic images. Registration can be performed using the technique used in the Dynamic Coronary Roadmapping product sold by Philips Healthcare, Best, The Netherlands, and described in document WO 2008 / 104921 A2. Improved guidance is provided to physicians by each of the following: displaying the image representation 160 and the live sequence of fluoroscopic images as overlay images; displaying the image representation 160 and the live sequence of fluoroscopic images adjacent to each other; displaying the path of the occluded vessel 140 in the live sequence of fluoroscopic images; and displaying one or more features of the occluded vessel 140 in the live sequence of fluoroscopic images. Furthermore, this additional guidance is provided without requiring additional contrast agent injections. Therefore, it also reduces the health risks associated with contrast agent use.

[0089] In this example, in the overlay image, the image representation of the blocked blood vessel 160 can be distinguished from the live sequence of the fluorescence fluoroscopic image by using various colors or by using various markers to highlight the blocked blood vessel.

[0090] In this example, for instance, one or more features of the blocked blood vessel 140 extracted from image representation 160 may include features such as calcification. Calcifications are visible in the image representation due to their relatively higher X-ray attenuation values ​​compared to blood vessels. Calcification features can be extracted from image representation 160 using segmentation techniques such as those described above.

[0091] In a relevant example, the live sequence of fluorescence fluoroscopic images includes an interventional device in blood vessel 140. One or more processors 110 are configured to:

[0092] Detecting portions of the interventional device in a live sequence of fluorescence fluoroscopic images;

[0093] Motion vectors representing the motion of the interventional device are extracted from the live sequence of fluorescence fluoroscopy images based on the portion detected by the interventional device; and

[0094] In this process, one or more processors 110 are configured to also register sequences 120 of angiographic images to each other based on extracted motion vectors representing the motion of that portion of the interventional device, S140.

[0095] In this example, the movement of a portion of the interventional device, as determined by fluorescence fluoroscopy images, and the adjacent blockage of a portion 140 of a blood vessel, as determined by angiography images, 130, are considered. p 140 dThe movement of the device is used in conjunction with the radiopaque material to register angiographic images. In this example, the detected portion of the interventional device can be, for example, the radiopaque tip of the device, or a portion of a guidewire attached to the device, made of radiopaque material, etc. The radiopaque nature of such interventional device features facilitates their detection in fluoroscopic images in the absence of contrast agents. Therefore, this portion of the interventional device serves as an additional feature for registering angiographic images. This portion of the interventional device serves as a reliable feature for registering angiographic images because the device is within the same blocked vessel in both the angiographic and fluoroscopic images, and therefore the movement of the device determined from the fluoroscopic image is similar to the movement of the vessel determined from the detected portion of the vessel in the angiographic image. Another advantage of using this portion of the interventional device to register angiographic images is that it does not require a contrast agent to make it visible in the fluoroscopic image. A further advantage of using the interventional device to perform this registration is that it follows the path of the blockage as the device advances into the blockage in the fluoroscopic image. Therefore, the movement of the interventional device, as determined by fluorescence fluoroscopy images, provides reliable information about the movement of the occluded blood vessel and its shape within the occlusion itself. This, in turn, improves the reliability of the image representation of the occluded blood vessel 160.

[0096] In another example, a volumetric angiography image including blood vessel 140 is used to estimate the shape of blood vessel 140 in a sequence 120 of angiography images. In this example, the image data received in operation S110 includes projected image data, and the sequence of angiography images includes a sequence of projected angiography images, and one or more processors 110 are further configured to:

[0097] Receive volumetric image data representing a volumetric angiography image including blood vessel 140; and

[0098] The shape of the blood vessel 140 in the sequence 120 of the angiography images is estimated based on the volumetric angiography images; and

[0099] One or more processors 110 are configured to detect portions 140 of adjacent occlusions 130 of the blood vessel 140 in a sequence 120 of angiographic images based on the estimated shape of the blood vessel 140. p 140 d .

[0100] In this example, volumetric angiography images are used to estimate the shape of vessel 140 in angiography image 120, and the estimated shape of vessel 140 in angiography image 120 is used to detect portions 140 of adjacent occlusions 130 of vessel S120 in sequence 120 of angiography images. p 140d This can improve the effect on the vascular portion by 140. p 140 d The reliability of the detection, or its ability to reduce the portion of the blood vessel being detected by 140. p 140 d The time spent is reduced because, compared to the projected angiography image received in operation S110, the volumetric angiography image includes additional information about the shape and movement of portions of the vessel 140. Therefore, it improves the reliability of the image representation 160 of the occluded vessel and also reduces the time spent generating the image representation 160.

[0101] In this example, the shape of the blood vessel 140 in the sequence of angiography images can be estimated by identifying the blood vessel 140 in the volumetric angiography image and then registering the volumetric angiography image including the identified blood vessel to the sequence of projected angiography images acquired in operation S110. The portion 140 of the adjacent occlusion 130 of the blood vessel in the sequence 120 of angiography images can be determined by finding the (minimum) path constrained by the estimated shape of the projected blood vessel in the angiography image path. p 140 d Blood vessels 140 can be identified in volumetric angiography images using segmentation techniques such as model-based segmentation and semantic segmentation. Registration can be performed by projecting a volumetric angiography image including the identified blood vessel to provide a synthetic projection image from the same viewpoint as the sequence of the projected angiography images from the projection X-ray imaging system, and then (flexibly) registering the synthetic projection image to the projected angiography image. This provides a synthetic projection image that includes the identified blood vessel 140 and corresponds to the projected angiography image, and therefore it facilitates the identification of the identified blood vessel in the projected angiography image.

[0102] In this example, the volumetric angiography image can be, for example, a (preoperative) CT image or a (preoperative) MRI image. In the former case, if the vessel is in the heart, the volumetric angiography image can be referred to as a coronary computed tomography angiography “CCTA” image.

[0103] In another example, a volumetric angiography image including blood vessel 140 is registered to an angiography image. In this example, the image data received in operation S110 includes projected image data, and the sequence of angiography images includes a sequence of projected angiography images, and one or more processors 110 are further configured to:

[0104] Receive volumetric image data representing a volumetric angiography image including blood vessel 140;

[0105] Based on the extracted motion vectors, the volumetric angiography image is elastically registered to the projected angiography image; and

[0106] Among them, the image representation 160 of the occluded blood vessel includes a volumetric angiographic image that is elastically registered to the occluded blood vessel.

[0107] In this example, motion vectors can be extracted in the manner described above.

[0108] In another example, one or more processors 110 are configured to evaluate an accuracy parameter for the image representation 160 and output a value for the accuracy parameter. The value of the accuracy parameter can be calculated based on factors such as the sharpness of the image representation 160 within the region of interest, or based on a confidence metric for the operation used to detect (S120) adjacent occlusions of a blood vessel. Alternatively or additionally, motion vectors can be used to calculate the value of the accuracy parameter. For example, a relatively low accuracy parameter can be provided if the motion field has excessive amplitude, or if the motion field indicates a high degree of curvature. Providing a value for the accuracy parameter allows a physician to assess the extent to which the image representation 160 can be relied upon.

[0109] In the relevant example, portion 140 of the blood vessel is provided. p 140 d The estimation extension. In this example, in response to the accuracy parameter value falling below a predetermined threshold, one or more processors 110 are configured to:

[0110] Estimated adjacent blockage of blood vessel 130 / 140 p 140 d The extension, which points toward the blockage; and

[0111] Output the portion 140 of the blood vessel in the image representation 160 of the blocked blood vessel. p 140 d The estimated extension.

[0112] In this example, the estimated extension can be determined by techniques such as interpolation (e.g., using spline curves) or weighted least squares regression of parametric shapes, where the weights are determined based on the values ​​of the accuracy parameter.

[0113] In another example, the value of the accuracy parameter can be indicated in the image representation. For example, a color corresponding to the value of the accuracy parameter can be provided for the region of interest adjacent to that part of the blood vessel, or for the predicted extension of the blood vessel, or for the estimated extension of the blood vessel. This indicates to the physician the degree to which the image representation can be relied upon.

[0114] In another example, sequence 120 of angiographic images includes an interventional device in vessel 140, and one or more processors 110 are configured to:

[0115] Detecting portions of the interventional device in sequence 120 of angiographic images;

[0116] Motion vectors representing the motion of the interventional device are extracted from sequence 120 of angiographic images based on the detected portion; and

[0117] In this process, one or more processors 110 are configured to also register sequences 120 of angiographic images to each other based on extracted motion vectors representing the motion of that portion of the interventional device, S140.

[0118] In this example, the detected portion of the interventional device can be, for example, the radiopaque tip of the interventional device, or a radiopaque marker attached to the interventional device, a portion of a guidewire formed of radiopaque material, etc. This radiopaque characteristic of interventional devices facilitates their detection in angiographic images. The detected portion of the interventional device provides a reliable landmark, in addition to the detected portion 140 of the adjacent occlusion 130 of the blood vessel. p 140 d In addition, this reliable marker is also used to register the sequences 120 of angiographic images to each other S140. Using this additional marker improves the reliability of the registration, and therefore improves the reliability of the image representation 160 of the occluded vessel. If a portion of the interventional device comes into contact with the occlusion, an additional benefit is provided, because in this arrangement, the detected portion of the interventional device provides an accurate indication of the movement of the vessel near the occlusion, i.e., at a point in the vessel where the portion 140 of the vessel adjacent to the occlusion 130 is... p 140 d Visibility begins to decrease due to congestion.

[0119] It should be noted that in the above example, system 100 may also include one or more of the following: an angiography imaging system for providing image data received in operation S110, or for providing fluorescence fluoroscopy image data, such as, for example Figure 1 The projection X-ray imaging system 210 shown; display device, such as Figure 1 The monitor 220 shown is a display device used to display an image representation 160 of the blocked blood vessel, a fluorescence fluoroscopic image, and other data output by one or more processors 110 of the system 100; a patient bed 230; and a syringe ( Figure 1 (Not shown in the image), the syringe is used to inject contrast agent into the object; and the user input device ( Figure 1(Not shown in the figure), it is used to receive user input related to system 100, such as a keyboard, mouse, touch screen, etc.

[0120] In another example, an angiography imaging system including system 100 is provided.

[0121] In another example, a computer-implemented method is provided for providing an image representation of a blocked blood vessel. The method includes:

[0122] Receive image data sequence 120 representing an angiographic image of a blockage 130 included in blood vessel 140, S110;

[0123] Detecting a portion of the adjacent occlusion 130 of vessel S120 in sequence 120 of the angiographic image 140. p 140 d ;

[0124] Based on the detected blood vessels, S130, representing the blood vessel portion 140, is extracted from sequence 120 of the angiography image. p 140 d The motion vector of the motion is 150;

[0125] Based on the extracted motion vector 150, the sequences 120 of the angiography images are mutually registered S140;

[0126] Processing S150 in the portion 140 with blood vessels p 140 d The image intensity of adjacent regions of interest in a sequence of mutually registered angiographic images is used to provide an image representation of the occluded vessel 160; and

[0127] The output image of the blocked blood vessel is represented as S160.

[0128] In another example, a computer program product is provided. The computer program product includes instructions that, when executed by one or more processors 110, cause the one or more processors to perform a method of providing an image representation of a blocked blood vessel. The method includes:

[0129] Receive image data sequence 120 representing an angiographic image of a blockage 130 included in blood vessel 140, S110;

[0130] Detecting a portion of the adjacent occlusion 130 of vessel S120 in sequence 120 of the angiographic image 140. p 140 d ;

[0131] Based on the detected blood vessels, S130, representing the blood vessel portion 140, is extracted from sequence 120 of the angiography image.p 140 d The motion vector of the motion is 150;

[0132] Based on the extracted motion vector 150, the sequences 120 of the angiography images are mutually registered S140;

[0133] Processing S150 in the portion 140 with blood vessels p 140 d The image intensity of adjacent regions of interest in a sequence of mutually registered angiographic images is used to provide an image representation of the occluded vessel 160; and

[0134] The output image of the blocked blood vessel is represented as S160.

[0135] The examples described above should be understood as illustrative of this disclosure, not restrictive. Further examples are also contemplated. For instance, the examples described with respect to the system may also be provided in a corresponding manner by a computer-implemented method or by a computer program product or by a computer-readable storage medium. It should be understood that features described with respect to any one example may be used alone or in combination with other described features, and may be used in combination with one or more features of another example or a combination of other examples. Furthermore, equivalents and modifications not described above may be employed without departing from the scope of the invention as defined in the appended claims. In the claims, the word "comprising" does not exclude other elements or operations, and the words "a" or "an" do not exclude a plurality. Although specific features are recited in dissimilar dependent claims, this does not indicate that combinations of these features cannot be advantageously used. Any reference numerals in the claims should not be construed as limiting their scope.

Claims

1. A system (100) for providing an image representation of a blocked blood vessel, the system comprising one or more processors (110) configured to: Receive (S110) image data, the image data representing a sequence (120) of angiographic images of a blockage (130) in a blood vessel (140). In the sequence (120) of the angiographic images, the portion (140) of the blood vessel adjacent to the blockage (130) is detected (S120). p 140 d ); Based on the detected portion of the blood vessel, motion vectors (S130) are extracted (150) from the sequence (120) of the angiography images, the motion vectors representing the portion (140) of the blood vessel. p 140 d ) movement; The sequences (120) of the angiography images are mutually registered based on the extracted motion vectors (150) (S140). Processing (S150) at the portion (140) of the blood vessel p 140 d The image intensity of adjacent regions of interest in a sequence of mutually registered angiographic images is used to provide an image representation of the occluded vessel (160); and Output (S160) the image representation of the blocked blood vessel (160).

2. The system according to claim 1, wherein, The one or more processors (110) are configured to process (S150) the image intensity in the sequence of mutually registered angiographic images by performing time integration of the image intensity in the region of interest on a plurality of mutually registered images.

3. The system according to claim 2, wherein, The one or more processors (110) are also configured to predict the portion (140) of the blood vessel adjacent to the blockage (130). p 140 d The image representation (160) extends toward the blockage (130), and wherein the image representation (160) includes the predicted extension.

4. The system according to claim 3, wherein, The one or more processors (110) are configured to iteratively perform the extraction (S130), the mutual registration (S140), the processing (S150), and the prediction, wherein each iteration uses the portion (140) of the blood vessel from the previous iteration. p 140 d The predicted extension is taken as the detected portion of the blood vessel in the extraction (S130).

5. The system according to any of the preceding claims, wherein, The portion of the blood vessel (140) includes the proximal portion (140). p ), and wherein the one or more processors (110) are further configured to: In the sequence (120) of the angiographic images, the distal portion (140) of the blood vessel (140) adjacent to the blockage (130) is detected. d ); Based on the detected distal portion of the blood vessel, the distal portion (140) representing the blood vessel is extracted from the sequence (120) of the angiography images. d The motion vector of the motion; Furthermore, the one or more processors (110) are configured to also base their work on the extracted distal portion (140) representing the blood vessel. d The motion vectors of the motion will register the sequence of angiographic images (120) with each other (S140).

6. The system according to any of the preceding claims, wherein, The one or more processors (110) are also configured to: Receive fluorescence fluoroscopic image data, which represents a real-time sequence of fluorescence fluoroscopic images of the blockage (130) included in the blood vessel (140); The image representation of the blocked blood vessel (160) is registered to the live sequence of the fluorescence fluoroscopic images; and The one or more processors (110) are further configured to output a live sequence of the fluorescence transillumination images, and wherein one or more of the following operations are performed: The image representation (160) and the real-world sequence of the fluorescent perspective image are displayed as an overlay image; The image representation (160) and the real-time sequence of the fluorescent perspective image are displayed adjacent to each other; The path of the blocked blood vessel (140) is extracted from the image representation (160), and the path is displayed in the live sequence of the fluorescence fluoroscopic images; One or more features of the blocked blood vessel (140) are extracted from the image representation (160), and the one or more features are displayed in the live sequence of the fluorescent fluoroscopic images.

7. The system according to claim 6, wherein, The live sequence of the fluorescence fluoroscopic images includes an interventional device in the blood vessel (140), and wherein the one or more processors (110) are further configured to: Detecting portions of the interventional device in a live sequence of the fluorescence imaging; Based on the detected portion of the interventional device, motion vectors representing the motion of that portion of the interventional device are extracted from the live sequence of the fluorescence imaging; and The one or more processors (110) are configured to also register the sequence (120) of the angiography images with each other based on the extracted motion vectors representing the motion of the portion of the interventional device (S140).

8. The system according to any of the preceding claims, wherein, The received image data includes projected image data, and the sequence of angiographic images includes a sequence of projected angiographic images, wherein the one or more processors (110) are further configured to: Receive volumetric image data, the volumetric image data representing a volumetric angiography image including the blood vessel (140); and The shape of the blood vessel (140) in the sequence (120) of the angiography images is estimated based on the volumetric angiography images; and One or more processors (110) are configured to detect (S120) the portion (140) of the blood vessel adjacent to the blockage (130) in the sequence (120) of the angiographic images based on the estimated shape of the blood vessel (140). p 140 d ).

9. The system according to any of the preceding claims, wherein, The one or more processors (110) are also configured to: Evaluate the accuracy parameters for the image representation (160); and Output the value of the accuracy parameter.

10. The system according to claim 9, wherein, In response to the accuracy parameter value being lower than a predetermined threshold, the one or more processors (110) are further configured to: Estimate the portion (140) adjacent to the blockage (130) of the blood vessel. p 140 d The extension of the blockage is directed toward the obstruction; and Output the portion (140) of the blocked blood vessel in the image representation (160). p 140 d The estimated extension of ).

11. The system according to any of the preceding claims, wherein, The sequence (120) of angiographic images further includes an interventional device in the blood vessel (140), and wherein the one or more processors (110) are further configured to: A portion of the interventional device is detected in the sequence (120) of the angiographic images; Based on the detected portion of the interventional device, motion vectors representing the motion of that portion of the angiography image sequence (120) are extracted; and The one or more processors (110) are configured to also register the sequence (120) of the angiography images with each other based on the extracted motion vectors representing the motion of the portion of the interventional device (S140).

12. The system according to any of the preceding claims, wherein, The region of interest includes calcified vascular segments.

13. The system according to any of the preceding claims, wherein, The image representation (160) of the blocked blood vessel includes the portion (140) of the blood vessel adjacent to the blockage (130). p 140 d ) towards the extension of the blockage (130).

14. A computer-implemented method for providing an image representation of a blocked blood vessel, the method comprising: Receive (S110) image data, the image data representing a sequence (120) of angiographic images of a blockage (130) in a blood vessel (140). In the sequence (120) of the angiographic images, the portion (140) of the blood vessel adjacent to the blockage (130) is detected (S120). p 140 d ); Based on the detected portion of the blood vessel, motion vectors (S130) are extracted (150) from the sequence (120) of the angiography images, the motion vectors representing the portion (140) of the blood vessel. p 140 d ) movement; The sequences (120) of the angiography images are mutually registered based on the extracted motion vectors (150) (S140). Processing (S150) at the portion (140) of the blood vessel p 140 d The image intensity of adjacent regions of interest in a sequence of mutually registered angiographic images is used to provide an image representation of the occluded vessel (160); and Output (S160) the image representation of the blocked blood vessel (160).

15. A computer program product comprising instructions that, when executed by one or more processors (110), cause the one or more processors to perform the method according to claim 14.

Images