Image Registration Method, Apparatus, Electronic Device, and Storage Medium

Abstract

The present invention discloses an image registration method, apparatus, electronic device, and storage medium. The method includes: respectively obtaining a first vascular centerline image corresponding to a first vascular image to be registered and a second vascular centerline image corresponding to a second vascular image to be registered; determining key points corresponding to the first vascular centerline image and key points corresponding to the second vascular centerline image; performing prediction of the starting and ending points of the blood vessels based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image to obtain predicted starting and ending points of the blood vessels; determining a first set of vascular segments based on the key points corresponding to the first vascular centerline image and the predicted starting and ending points of the blood vessels, and determining a second set of vascular segments based on the key points corresponding to the second vascular centerline image; and determining a vascular registration image based on the first set of vascular segments and the second set of vascular segments. The above technical solution improves the registration accuracy by predicting the starting and ending points of the blood vessels and re-determining the correct starting and ending points of the blood vessels.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an image registration method, device, electronic equipment and storage medium. Background Art

[0002] In conventional vascular interventional surgery, doctors usually use intraoperative 2D angiography images with clear structure, dynamic real-time, such as (Digital subtraction angiography, DSA) as guiding images. However, due to the nature of intraoperative 2D angiography image projection, it is difficult for doctors to make clear and intuitive decisions on intervention methods and treatment plans during surgery. Preoperative 3D images, such as (Computed Tomography Angiography, CTA), just make up for this defect. Through three-dimensional reconstruction, information such as the shape of blood vessels and the location of lesions can be presented intuitively and stereoscopically. Therefore, performing vascular interventional surgery by fusing real-time 2D angiography images and 3D images during surgery can greatly improve the success rate of doctors' operations and prevent the occurrence of complications during patients' operations.

[0003] Most of the current 3D / 2D vascular image registration technologies assume that the beginning and end of the vessels to be fused are the same. However, since the 2D angiography images change in real time during surgery, the beginning and end of the vessels also deform accordingly. Therefore, the beginning and end of the vessels in the 2D angiography images may not correspond to the beginning and end of the corresponding vessels in the 3D image, which leads to a decrease in the registration accuracy of the vascular images. Summary of the invention

[0004] The present invention provides an image registration method, device, electronic equipment and storage medium to improve the image registration accuracy.

[0005] According to one aspect of the present invention, there is provided an image registration method, comprising:

[0006] Respectively acquiring a first blood vessel centerline image corresponding to the first blood vessel image to be registered and a second blood vessel centerline image corresponding to the second blood vessel image to be registered;

[0007] Determining key points corresponding to the first blood vessel centerline image and key points corresponding to the second blood vessel centerline image;

[0008] Predicting the start and end points of a blood vessel based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image to obtain predicted blood vessel start and end points;

[0009] Determine a first blood vessel segment set based on key points corresponding to the first blood vessel centerline image and the predicted blood vessel start and end points, and determine a second blood vessel segment set based on key points corresponding to the second blood vessel centerline image;

[0010] Determine a vascular registration image based on the first set of vascular segments and the second set of vascular segments.

[0011] According to another aspect of the present invention, there is provided an image registration device, comprising:

[0012] A vascular centerline image acquisition module, configured to respectively acquire a first vascular centerline image corresponding to a first vascular image to be registered and a second vascular centerline image corresponding to a second vascular image to be registered;

[0013] A key point determination module, configured to determine key points corresponding to the first vascular centerline image and key points corresponding to the second vascular centerline image;

[0014] A vascular start and end point prediction module, configured to perform vascular start and end point prediction based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image, to obtain predicted vascular start and end points;

[0015] A vascular segment set determination module, configured to determine a first set of vascular segments based on the key points corresponding to the first vascular centerline image and the predicted vascular start and end points, and determine a second set of vascular segments based on the key points corresponding to the second vascular centerline image;

[0016] A vascular registration image determination module, configured to determine a vascular registration image based on the first set of vascular segments and the second set of vascular segments.

[0017] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0018] At least one processor; and

[0019] A memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the image registration method according to any embodiment of the present invention.

[0021] According to another aspect of the present invention, there is provided a computer-readable storage medium, the computer-readable storage medium storing computer instructions, and the computer instructions are used to implement the image registration method according to any embodiment of the present invention when executed by a processor.

[0022] In the technical solution of the embodiment of the present invention, the extraction of the blood vessel centerline is realized by respectively obtaining the first blood vessel centerline image corresponding to the first blood vessel image to be registered and the second blood vessel centerline image corresponding to the second blood vessel image to be registered; further, the starting and ending points of the blood vessels are predicted based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image, and the correct predicted starting and ending points of the blood vessels are re-determined; furthermore, the first blood vessel segment set is determined according to the key points corresponding to the first blood vessel centerline image and the predicted starting and ending points of the blood vessels, and the second blood vessel segment set is determined based on the key points corresponding to the second blood vessel centerline image, thereby improving the accuracy of the blood vessel segment set; furthermore, the blood vessel registration image is determined according to the first blood vessel segment set and the second blood vessel segment set, thereby improving the blood vessel registration accuracy.

[0023] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0025] Figure 1 is a flowchart of an image registration method provided in Embodiment 1 of the present invention;

[0026] Figure 2 is a schematic diagram of blood vessel key points provided in Embodiment 1 of the present invention;

[0027] Figure 3 is a flowchart of an image registration method provided in Embodiment 2 of the present invention;

[0028] Figure 4 is a schematic diagram of blood vessel segmentation provided in Embodiment 2 of the present invention;

[0029] Figure 5 is a schematic diagram of blood vessel radius provided in Embodiment 2 of the present invention;

[0030] Figure 6 is a schematic diagram of the estimated result of blood vessel starting and ending point mapping provided in Embodiment 2 of the present invention;

[0031] Figure 7 is a flowchart of an image registration method provided in Embodiment 3 of the present invention;

[0032] Figure 8 It is a schematic structural diagram of an image registration device provided in Embodiment 4 of the present invention;

[0033] Figure 9 It is a schematic structural diagram of an electronic device for implementing the image registration method of the embodiment of the present invention. Specific Embodiments

[0034] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0036] Embodiment 1

[0037] Figure 1 It is a flowchart of an image registration method provided in Embodiment 1 of the present invention. This embodiment is applicable to the case of registering blood vessel images in different dimensions. This method can be executed by an image registration device, which can be implemented in the form of hardware and / or software, and the image registration device can be configured in a computer terminal and / or a server. As Figure 1 shown, the method includes:

[0038] S110. Obtain a first blood vessel centerline image corresponding to a first blood vessel image to be registered and a second blood vessel centerline image corresponding to a second blood vessel image to be registered respectively.

[0039] In this embodiment, the first blood vessel image to be registered refers to the blood vessel image to be registered. Similarly, the second blood vessel image to be registered refers to the image for blood vessel registration with the first blood vessel image to be registered. The first blood vessel image to be registered or the second blood vessel image to be registered can be one of real-time two-dimensional, three-dimensional or multi-dimensional images, which is not limited herein. It should be noted that the first blood vessel image to be registered and the second blood vessel image to be registered can be images of different dimensions. For example, the first blood vessel image to be registered is a three-dimensional blood vessel image, and the second blood vessel image to be registered can be a two-dimensional blood vessel image. The number of the first blood vessel image to be registered and the second blood vessel image to be registered can be one or more, which is not limited herein.

[0040] Exemplarily, the first blood vessel image to be registered or the second blood vessel image to be registered can be medical images containing blood vessels such as computed tomography angiography (CTA) and digital subtraction angiography (DSA).

[0041] In this embodiment, the first blood vessel centerline image refers to the blood vessel centerline image obtained by morphological processing and other operations on the first blood vessel image to be registered. Similarly, the second blood vessel centerline image refers to the blood vessel centerline image obtained by morphological processing and other operations on the second blood vessel image to be registered.

[0042] In some alternative embodiments, respectively obtaining the first blood vessel centerline image corresponding to the first blood vessel image to be registered and the second blood vessel centerline image corresponding to the second blood vessel image to be registered includes: obtaining the blood vessel images to be registered, where the blood vessel images to be registered include the first blood vessel image to be registered and the second blood vessel image to be registered; segmenting the first blood vessel image to be registered to obtain a first blood vessel segmentation image, and extracting the centerline from the first blood vessel segmentation image to obtain the first blood vessel centerline image corresponding to the first blood vessel image to be registered; segmenting the second blood vessel image to be registered to obtain a second blood vessel segmentation image, and extracting the centerline from the second blood vessel segmentation image to obtain the second blood vessel centerline image corresponding to the second blood vessel image to be registered.

[0043] Wherein, the first blood vessel segmentation image refers to the initial blood vessel segmentation image of the first blood vessel image to be registered. Similarly, the second blood vessel segmentation image refers to the initial blood vessel segmentation image of the second blood vessel image to be registered.

[0044] Exemplarily, the first vascular image to be registered can be a three-dimensional preoperative CTA image, and the second vascular image to be registered can be a two-dimensional real-time intraoperative DSA image. The formats of the first vascular image to be registered and the second vascular image to be registered are not limited herein. For example, they can be in the DICOM format, etc. The above process can be achieved through image preprocessing. The image preprocessing can include, but is not limited to, a CTA image data segmentation module, a CTA image data centerline extraction module, a DSA image data segmentation module, and a DSA image data centerline extraction module. Among them, the CTA image data segmentation module includes: in the training stage, a large amount of CTA data is labeled for blood vessels. Based on the labeling results, a three-dimensional (3D) segmentation model for blood vessel segmentation is trained. The network architecture of the three-dimensional segmentation model can be 3DU-net, V-net, etc.; in the inference stage, the trained three-dimensional segmentation model is used to predict new CTA image data, and a three-dimensional initial blood vessel segmentation image is obtained through post-processing operations. The CTA image data centerline extraction module: performs a series of morphological operations on the three-dimensional initial blood vessel segmentation image, and uses a smoothing algorithm to smooth the image to obtain a three-dimensional blood vessel centerline image. The DSA image data segmentation module: in the training stage, a large amount of DSA data is labeled for blood vessels. Based on the labeling results, a two-dimensional (2D) segmentation model for blood vessel segmentation is trained. The network architecture of the two-dimensional segmentation model can be 2DU-net, V-net, etc.; in the inference stage, the trained two-dimensional segmentation model is used to predict new DSA image data, and a two-dimensional initial blood vessel segmentation image is obtained through post-processing operations. The DSA image data centerline extraction module: performs a series of morphological operations on the two-dimensional initial blood vessel segmentation image, and uses a smoothing algorithm to smooth the image to obtain a two-dimensional blood vessel centerline image.

[0045] S120. Determine the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image.

[0046] In this embodiment, the key points refer to the main constituent points in the blood vessel centerline, which can include, but are not limited to, the blood vessel starting point, the blood vessel ending point, the blood vessel branching point, etc. In other words, the number of key points can be one or more, which is not limited herein. Exemplarily, Figure 2 is a schematic diagram of a blood vessel key point provided in this embodiment. Figure 2 The left side is the preoperative CTA image. Figure 2 The right side is the intraoperative DSA image, and the key points are the dots in the figure.

[0047] Specifically, a trained key point tracking model can be used for key point detection, or key point detection can be performed according to the adjacency relationship of the centerline points, which is not limited herein.

[0048] S130. Predict the starting and ending points of the blood vessels based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image, and obtain the predicted starting and ending points of the blood vessels.

[0049] In this embodiment, the predicted starting and ending points of the blood vessels refer to the re-determined starting and ending points of the blood vessels, which may include the predicted starting point of the blood vessel and the predicted ending point of the blood vessel.

[0050] It should be noted that in this embodiment, by predicting the starting and ending points of the blood vessels based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image, the starting point and the ending point of the blood vessel can be re-determined, so that the starting and ending points of the first blood vessel image to be registered and the second blood vessel image to be registered are corresponding, thereby improving the registration accuracy of the blood vessel images.

[0051] Specifically, a blood vessel starting and ending point prediction model can be established according to the matching relationship between the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image, and the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image are input into the blood vessel starting and ending point prediction model to obtain the predicted starting and ending points of the blood vessels.

[0052] In some alternative embodiments, predicting the starting and ending points of the blood vessels includes predicting the starting point of the blood vessel and the ending point of the blood vessel; determining the first blood vessel segment set based on the key points corresponding to the first blood vessel centerline image and the predicted starting and ending points of the blood vessels, including: based on the predicted starting point and the predicted ending point of the blood vessel, replacing the starting point and the ending point of the blood vessel corresponding to the key points corresponding to the first blood vessel centerline image to obtain the target key points corresponding to the first blood vessel centerline image; determining the first blood vessel segment set based on the target key points corresponding to the first blood vessel centerline image.

[0053] It should be emphasized that the starting point and the ending point of the blood vessel in the key points corresponding to the first blood vessel centerline image can be replaced with the predicted starting point and the predicted ending point of the blood vessel, so that the starting and ending points of the blood vessel are more accurate, and thus the blood vessel is segmented according to the re-determined target key points, improving the segmentation accuracy of the first blood vessel segment set. Among them, the target key points refer to the set of key points after the replacement of the starting and ending points of the blood vessel.

[0054] S140. Determine the first blood vessel segment set based on the key points corresponding to the first blood vessel centerline image and the predicted starting and ending points of the blood vessels, and determine the second blood vessel segment set based on the key points corresponding to the second blood vessel centerline image.

[0055] In this embodiment, the first blood vessel segment set refers to a set of blood vessel segments obtained by re - dividing blood vessels based on the key points corresponding to the first blood vessel centerline image and the predicted start and end points of the blood vessels, which can be used to represent the overall profile of the blood vessels in the first blood vessel image to be registered. The second blood vessel segment set refers to a set of blood vessel segments obtained by dividing blood vessels based on the key points corresponding to the second blood vessel centerline image, which can be used to represent the overall profile of the blood vessels in the second blood vessel image to be registered.

[0056] S150. Determine the blood vessel registration image based on the first blood vessel segment set and the second blood vessel segment set.

[0057] Specifically, the first blood vessel segment set and the second blood vessel segment set can be densely matched to obtain the corresponding relationships of the blood vessels in the first blood vessel segment set and the second blood vessel segment set. Then, based on the corresponding relationships of the blood vessels in the first blood vessel segment set and the second blood vessel segment set, the first blood vessel image to be registered and the second blood vessel image to be registered can be registered.

[0058] The technical solution of the embodiment of the present invention realizes the extraction of the blood vessel centerline by respectively obtaining the first blood vessel centerline image corresponding to the first blood vessel image to be registered and the second blood vessel centerline image corresponding to the second blood vessel image to be registered. Further, based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image, the start and end points of the blood vessels are predicted, and the correct predicted start and end points of the blood vessels are re - determined. Then, based on the key points corresponding to the first blood vessel centerline image and the predicted start and end points of the blood vessels, the first blood vessel segment set is determined, and based on the key points corresponding to the second blood vessel centerline image, the second blood vessel segment set is determined, which improves the accuracy of the blood vessel segment set. Furthermore, the blood vessel registration image is determined based on the first blood vessel segment set and the second blood vessel segment set, which improves the blood vessel registration accuracy.

[0059] Embodiment Two

[0060] Figure 3 As shown in the flowchart of an image registration method provided in Embodiment Two of the present invention, the method of this embodiment can be combined with each optional solution in the image registration method provided in the above - mentioned embodiment. The image registration method provided in this embodiment is further optimized. Optionally, the predicting the start and end points of the blood vessels based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image to obtain the predicted start and end points of the blood vessels includes: determining the blood vessel segment length set or the key point radius set based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image; and determining the predicted start and end points of the blood vessels based on the blood vessel segment length set or the key point radius set.

[0061] As Figure 3 shown, the method includes:

[0062] S210. Obtain the first blood vessel centerline image corresponding to the first blood vessel image to be registered and the second blood vessel centerline image corresponding to the second blood vessel image to be registered respectively.

[0063] S220. Determine the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image.

[0064] S230. Determine a blood vessel segment length set or a key point radius set based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image.

[0065] S240. Determine the predicted start and end points of the blood vessel based on the blood vessel segment length set or the key point radius set.

[0066] S250. Determine a first blood vessel segment set based on the key points corresponding to the first blood vessel centerline image and the predicted start and end points of the blood vessel, and determine a second blood vessel segment set based on the key points corresponding to the second blood vessel centerline image.

[0067] S260. Determine a blood vessel registration image based on the first blood vessel segment set and the second blood vessel segment set.

[0068] In this embodiment, the blood vessel segment length set refers to the set of lengths of each blood vessel segment divided according to the key points. The key point radius set refers to the set of blood vessel radii corresponding to each key point.

[0069] Specifically, taking each key point as a blood vessel segmentation point, segment the blood vessels in the first blood vessel centerline image and the second blood vessel centerline image to obtain a blood vessel segment length set. Taking each key point as a starting point, calculate the distance from each key point to the blood vessel wall, and generate a blood vessel radius set based on the distances from each key point to the blood vessel wall.

[0070] After obtaining the blood vessel segment length set or the key point radius set, the predicted start and end points of the blood vessel can be determined according to the blood vessel segment length set or the key point radius set. Specifically, the difference information between the first blood vessel centerline image and the first blood vessel centerline image can be determined according to the blood vessel segment length set, and the predicted start and end points of the blood vessel can be determined based on this difference information. Or, the difference information between the first blood vessel centerline image and the first blood vessel centerline image can be determined according to the blood vessel radius set, and the predicted start and end points of the blood vessel can be determined based on this difference information. Among them, the difference information can be the distance difference between the start and end points of the blood vessel.

[0071] In some alternative embodiments, the set of vascular segment lengths includes a first set of vascular segment lengths and a second set of vascular segment lengths; determining the set of vascular segment lengths based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image includes: determining the first set of vascular segment lengths based on the key points corresponding to the first vascular centerline image, and determining the second set of vascular segment lengths based on the key points corresponding to the second vascular centerline image; correspondingly, determining the predicted start and end points of the blood vessel based on the set of vascular segment lengths or the set of key point radii includes: determining the vascular length ratio coefficient based on the first set of vascular segment lengths and the second set of vascular segment lengths; determining the extended lengths of the start and end segments based on the vascular length ratio coefficient and the lengths of the start and end segments in the second set of vascular segment lengths; determining the predicted start and end points of the blood vessel based on the start and end points in the first set of vascular segment lengths and the extended lengths of the start and end segments.

[0072] Among them, the extended lengths of the start and end segments include the extended length of the start segment and the extended length of the end segment.

[0073] Exemplarily, Figure 4 is a schematic diagram of blood vessel segmentation provided in this embodiment. Figure 4 The left part is a schematic diagram of blood vessel segmentation corresponding to the preoperative three-dimensional CTA image. The three-dimensional CTA blood vessel segments can be represented by , and p1...p5 represent the key points corresponding to the first vascular centerline image; Figure 4 The right part is a schematic diagram of blood vessel segmentation corresponding to the intraoperative two-dimensional DSA image. The two-dimensional DSA blood vessel segments can be represented by , and q1...q5 represent the key points corresponding to the second vascular centerline image. Among them, and represent the starting blood vessel segment and the ending blood vessel segment of the three-dimensional CTA respectively, and represent the starting blood vessel segment and the ending blood vessel segment of the two-dimensional DSA respectively. The first set of vascular segment lengths can be the set of vascular segment lengths corresponding to the preoperative three-dimensional CTA image, which can be represented by , where C represents the preoperative CTA image, represents the length of the i-th blood vessel segment of the three-dimensional CTA preoperative image blood vessel; the second set of vascular segment lengths can be the set of vascular segment lengths corresponding to the intraoperative two-dimensional DSA image, which can be represented by , where D represents the intraoperative DSA image, represents the length of the i-th blood vessel segment of the two-dimensional DSA intraoperative image blood vessel. Further, the vascular length ratio coefficient can be obtained by the following formula:

[0074]

[0075] Further, obtain the lengths of the starting and ending blood vessel segments of the two-dimensional DSA mapped to the three-dimensional CTA respectively:

[0076]

[0077]

[0078] According to the determined starting segment extension length A and the position of the first key point of the three-dimensional CTA blood vessel, extend forward by A to obtain the corresponding 3D blood vessel starting end. Similarly, according to the determined ending segment extension length B and the position of the last key point of the three-dimensional CTA blood vessel, extend backward by B to obtain the corresponding 3D blood vessel ending end. The determination method of the blood vessel length may include but is not limited to: determining the blood vessel length based on the number of pixels or voxels between blood vessel segments, or uniformly sampling between each blood vessel segment and representing the blood vessel length with the number of intermediate points.

[0079] In some alternative embodiments, the key point radius set includes a first key point radius set and a second key point radius set; determining the key point radius set based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image includes: determining the first key point radius set based on the key points corresponding to the first blood vessel centerline image, and determining the second key point radius set based on the key points corresponding to the second blood vessel centerline image; correspondingly, determining the predicted blood vessel start and end points based on the blood vessel segment length set or the key point radius set includes: determining the blood vessel radius ratio coefficient based on the first key point radius set and the second key point radius set; determining the start and end point predicted radii based on the blood vessel radius ratio coefficient and the start and end point radii in the second key point radius set; determining the predicted blood vessel start and end points based on the first key point radius set and the start and end point predicted radii.

[0080] Among them, the start and end point predicted radii include the predicted blood vessel radius corresponding to the starting key point and the predicted blood vessel radius corresponding to the ending key point.

[0081] Exemplarily, Figure 5 is a schematic diagram of a blood vessel radius provided for this embodiment. Figure 5 The left part is a schematic diagram of the blood vessel radius corresponding to the three-dimensional CTA preoperative image. The key points in the three-dimensional CTA blood vessel can be represented by P = {p1, p2,..., p n-1}; Figure 5 The right part is a schematic diagram of the blood vessel radius corresponding to the two-dimensional DSA intraoperative image. The key points in the two-dimensional DSA blood vessel can be represented by Q = {q1, q2,..., q n-1}; p0 and p n respectively represent the starting key point and the ending key point of the three-dimensional CTA blood vessel, and q0 and q n respectively represent the starting key point and the ending key point of the two-dimensional DSA blood vessel. represents the blood vessel radius of the jth key point of the three-dimensional CTA blood vessel, represents the vessel radius of the j-th key point of the two-dimensional DSA vessel;

[0082] Furthermore, the vessel radius proportionality coefficient μ is obtained by the following formula:

[0083]

[0084] Furthermore, the radii of the starting and ending vessel ends of the two-dimensional DSA mapped to the starting and ending vessel ends of the three-dimensional CTA are obtained respectively:

[0085] p 01 = q0 * μ

[0086] p n1 = q n * μ

[0087] where p 01 represents the predicted vessel radius corresponding to the starting key point, and p n1 represents the predicted vessel radius corresponding to the ending key point. Furthermore, the center of the circle where the radius with the smallest distance error from p 01 or p n1 can be determined as the predicted starting and ending points of the vessel based on distance formulas such as the mean square error. The calculation method of the vessel radius can include but is not limited to the distance from the vessel centerline to the image background, etc.

[0088] Exemplarily, Figure 6 is a schematic diagram of the estimation result of the mapping of the starting and ending points of the vessel provided in this embodiment. The mapping of the starting and ending points of the vessel is the above-mentioned method for predicting the starting and ending points of the vessel, which can include the method for estimating the mapping of the starting and ending points of the vessel based on the vessel length and the method for estimating the mapping of the starting and ending points of the vessel based on the vessel radius. Figure 6 The left part is the mapping estimation of the starting point of the vessel, Figure 6 and the right part is the mapping prediction of the ending point of the vessel. The estimated starting and ending points are indicated by "+".

[0089] The technical solution of the embodiment of the present invention realizes the re-determination of the starting point and the ending point of the vessel by determining the vessel segment length set or the key point radius set according to the key points corresponding to the first vessel centerline image and the key points corresponding to the second vessel centerline image, and then determining the predicted starting and ending points of the vessel according to the vessel segment length set or the key point radius set, so that the starting and ending points of the vessels in the first vessel image to be registered and the second vessel image to be registered are corresponding, thereby improving the registration accuracy of the vessel images.

[0090] Embodiment Three

[0091] Figure 7The flowchart of an image registration method provided in Embodiment 3 of the present invention. The method in this embodiment can be combined with each optional solution in the image registration method provided in the above embodiments. The image registration method provided in this embodiment is further optimized. Optionally, determining the vascular registration image based on the first vascular segment set and the second vascular segment set includes: sampling each vascular segment in the first vascular segment set to obtain a set of intermediate points of the first vascular segments; sampling each vascular segment in the second vascular segment set to obtain a set of intermediate points of the second vascular segments; determining the vascular registration relationship based on the set of intermediate points of the first vascular segments and the set of intermediate points of the second vascular segments; and registering the first vascular image to be registered and the second vascular image to be registered based on the vascular registration relationship to obtain the vascular registration image.

[0092] As Figure 7 shown, the method includes:

[0093] S310. Respectively obtain the first vascular centerline image corresponding to the first vascular image to be registered and the second vascular centerline image corresponding to the second vascular image to be registered.

[0094] S320. Determine the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image.

[0095] S330. Perform vascular start and end point prediction based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image to obtain predicted vascular start and end points.

[0096] S340. Determine the first vascular segment set based on the key points corresponding to the first vascular centerline image and the predicted vascular start and end points, and determine the second vascular segment set based on the key points corresponding to the second vascular centerline image.

[0097] S350. Sample each vascular segment in the first vascular segment set to obtain a set of intermediate points of the first vascular segments.

[0098] In this embodiment, the set of intermediate points of the first vascular segments refers to the set of sampling intermediate points of each vascular segment in the first vascular segment set.

[0099] Exemplarily, the sampling method can be uniform sampling. The first vascular segment set can be the vascular segment set corresponding to the three-dimensional CTA pre-operative image. Specifically, each vascular segment in the vascular segment set corresponding to the three-dimensional CTA pre-operative image can be uniformly sampled to obtain the set of intermediate points of the i-th vascular segment of the vascular centerline corresponding to the three-dimensional CTA pre-operative image (i = 1,..., n), where represents the set of intermediate points of the first vascular segments, C represents the CTA pre-operative image, and mix Represents the x-th midpoint of the i-th blood vessel segment.

[0100] S360. Sample each blood vessel segment in the second blood vessel segment set to obtain a set of midpoints of the second blood vessel segments.

[0101] In this embodiment, the set of midpoints of the second blood vessel segments refers to the set of sampled midpoints of each blood vessel segment set in the second blood vessel segment set.

[0102] Exemplarily, the sampling method can be uniform sampling, and the second blood vessel segment set can be the set of blood vessel segments corresponding to the intraoperative two-dimensional DSA images. Specifically, uniform sampling can be performed on each blood vessel segment in the set of blood vessel segments corresponding to the intraoperative two-dimensional DSA images to obtain the set of midpoints of the i-th blood vessel segment of the blood vessel centerline corresponding to the intraoperative two-dimensional DSA images (i = 1, …, n), where represents the set of midpoints of the second blood vessel segments, D represents the intraoperative two-dimensional DSA image, and n iy Represents the y-th midpoint of the i-th blood vessel segment.

[0103] S370. Determine the blood vessel registration relationship based on the set of midpoints of the first blood vessel segments and the set of midpoints of the second blood vessel segments.

[0104] In this embodiment, the blood vessel registration relationship refers to the corresponding relationship between the midpoints in the set of midpoints of the first blood vessel segments and the midpoints in the set of midpoints of the second blood vessel segments.

[0105] Exemplarily, the blood vessel registration relationship can be γ = {(m i1 , n i1 ), (m i2 , n i2 ), …, (m ix , n iy )}, where m ix represents the x-th midpoint of the i-th blood vessel segment of the preoperative three-dimensional CTA image, n iy represents the y-th midpoint of the i-th blood vessel segment of the intraoperative two-dimensional DSA image, and γ represents the blood vessel registration relationship. Specifically, the minimum distance cost matching between the midpoints in the set of midpoints of the first blood vessel segments and the set of midpoints of the second blood vessel segments can be determined according to the dynamic time warping algorithm, so as to obtain the blood vessel registration relationship. It can be understood that the corresponding relationship between the midpoints in the blood vessel registration relationship can be a one-to-one, one-to-many, or many-to-one relationship, which is not limited herein.

[0106] It should be noted that the blood vessel registration relationship establishment method in this embodiment belongs to a dense matching method, and the blood vessel registration relationship determined by this dense matching method can be applied to blood vessel images that undergo elastic deformation in real time, improving the adaptability of image registration.

[0107] S380. Based on the vascular registration relationship, register the first vascular image to be registered and the second vascular image to be registered to obtain a vascular registration image.

[0108] Exemplarily, the first vascular image to be registered can be a three-dimensional CTA preoperative image, and the second vascular image to be registered can be a two-dimensional DSA intraoperative image. Specifically, according to the vascular registration relationship, the spatial position of the blood vessels in the three-dimensional CTA preoperative image can be transformed so that the spatial position of the blood vessels in the transformed three-dimensional CTA preoperative image is consistent with the spatial position of the blood vessels in the two-dimensional DSA intraoperative image, and a registered vascular registration image is generated.

[0109] In some alternative embodiments, determining the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image includes: inputting the first vascular centerline image into a pre-trained first key point tracking model to obtain the key points corresponding to the first vascular centerline image, or determining the centerline points in the first vascular centerline image that meet the adjacency relationship condition as the key points corresponding to the first vascular centerline image; inputting the second vascular centerline image into a pre-trained second key point tracking model to obtain the key points corresponding to the second vascular centerline image, or determining the centerline points in the second vascular centerline image that meet the adjacency relationship condition as the key points corresponding to the second vascular centerline image.

[0110] Among them, the key point tracking model refers to a network model for predicting segmented key points. The key point tracking model can be obtained by initially training a network based on multiple vascular centerline sample images and the labels corresponding to the vascular centerline sample images.

[0111] Exemplarily, the adjacency relationship condition may include: if the current centerline point has only one adjacent point, indicating that the current centerline point is the starting point or the ending point of the blood vessel, then the current centerline point is determined as a key point; if the current centerline point has three or more adjacent points, indicating that the current centerline point is a blood vessel branch point, then the current centerline point is determined as a key point.

[0112] The technical solution of the embodiment of the present invention samples each blood vessel segment in the first blood vessel segment set to obtain a set of intermediate points of the first blood vessel segment; samples each blood vessel segment in the second blood vessel segment set to obtain a set of intermediate points of the second blood vessel segment; determines a vascular registration relationship based on the set of intermediate points of the first blood vessel segment and the set of intermediate points of the second blood vessel segment, and this vascular registration relationship can be applied to blood vessel images that undergo elastic deformation in real time, improving the adaptability of image registration.

[0113] Embodiment 4

[0114] Figure 8The figure is a schematic structural diagram of an image registration device provided in the fourth embodiment of the present invention. As Figure 8 shown, the device includes:

[0115] A blood vessel centerline image acquisition module 410, configured to respectively acquire a first blood vessel centerline image corresponding to a first to-be-registered blood vessel image and a second blood vessel centerline image corresponding to a second to-be-registered blood vessel image;

[0116] A key point determination module 420, configured to determine key points corresponding to the first blood vessel centerline image and key points corresponding to the second blood vessel centerline image;

[0117] A blood vessel start and end prediction module 430, configured to perform blood vessel start and end prediction based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image to obtain predicted blood vessel start and end points;

[0118] A blood vessel segment set determination module 440, configured to determine a first blood vessel segment set based on the key points corresponding to the first blood vessel centerline image and the predicted blood vessel start and end points, and determine a second blood vessel segment set based on the key points corresponding to the second blood vessel centerline image;

[0119] A blood vessel registration image determination module 450, configured to determine a blood vessel registration image based on the first blood vessel segment set and the second blood vessel segment set.

[0120] The technical solution of the embodiment of the present invention realizes the extraction of the blood vessel centerline by respectively acquiring a first blood vessel centerline image corresponding to a first to-be-registered blood vessel image and a second blood vessel centerline image corresponding to a second to-be-registered blood vessel image; further, performs blood vessel start and end prediction based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image, and re-determines the correct predicted blood vessel start and end points; then determines a first blood vessel segment set according to the key points corresponding to the first blood vessel centerline image and the predicted blood vessel start and end points, and determines a second blood vessel segment set based on the key points corresponding to the second blood vessel centerline image, improving the accuracy of the blood vessel segment set; and then determines a blood vessel registration image according to the first blood vessel segment set and the second blood vessel segment set, improving the blood vessel registration accuracy.

[0121] In some alternative embodiments, the blood vessel centerline image acquisition module 410 is specifically configured to:

[0122] Acquire a to-be-registered blood vessel image, where the to-be-registered blood vessel image includes a first to-be-registered blood vessel image and a second to-be-registered blood vessel image;

[0123] Segment the first blood vessel image to be registered to obtain a first blood vessel segmentation image, and extract the centerline of the first blood vessel segmentation image to obtain a first blood vessel centerline image corresponding to the first blood vessel image to be registered;

[0124] Segment the second blood vessel image to be registered to obtain a second blood vessel segmentation image, and extract the centerline of the second blood vessel segmentation image to obtain a second blood vessel centerline image corresponding to the second blood vessel image to be registered.

[0125] In some alternative embodiments, the key point determination module 420 is specifically configured to:

[0126] Input the first blood vessel centerline image into a pre-trained first key point tracking model to obtain the key points corresponding to the first blood vessel centerline image, or determine the centerline points in the first blood vessel centerline image that satisfy the adjacency relationship condition as the key points corresponding to the first blood vessel centerline image;

[0127] Input the second blood vessel centerline image into a pre-trained second key point tracking model to obtain the key points corresponding to the second blood vessel centerline image, or determine the centerline points in the second blood vessel centerline image that satisfy the adjacency relationship condition as the key points corresponding to the second blood vessel centerline image.

[0128] In some alternative embodiments, the blood vessel start and end point prediction module 430 includes:

[0129] A set determination unit, configured to determine a blood vessel segment length set or a key point radius set based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image;

[0130] A blood vessel start and end point prediction unit, configured to determine predicted blood vessel start and end points based on the blood vessel segment length set or the key point radius set.

[0131] In some alternative embodiments, the blood vessel segment length set includes a first blood vessel segment length set and a second blood vessel segment length set;

[0132] The set determination unit is specifically configured to:

[0133] Determine a first blood vessel segment length set based on the key points corresponding to the first blood vessel centerline image, and determine a second blood vessel segment length set based on the key points corresponding to the second blood vessel centerline image;

[0134] Correspondingly, the blood vessel start and end point prediction unit is specifically configured to:

[0135] Determine a blood vessel length ratio coefficient based on the first blood vessel segment length set and the second blood vessel segment length set;

[0136] Determine the start and end segment extension lengths based on the blood vessel length ratio coefficient and the start and end segment lengths in the second blood vessel segment length set;

[0137] Determine the predicted blood vessel start and end points based on the start and end points in the first blood vessel segment length set and the start and end segment extension lengths.

[0138] In some alternative embodiments, the key point radius set includes a first key point radius set and a second key point radius set;

[0139] The set determination unit is specifically configured to:

[0140] Determine the first key point radius set based on the key points corresponding to the first blood vessel centerline image, and determine the second key point radius set based on the key points corresponding to the second blood vessel centerline image;

[0141] Correspondingly, the blood vessel start and end point prediction unit is further specifically configured to:

[0142] Determine the blood vessel radius ratio coefficient based on the first key point radius set and the second key point radius set;

[0143] Determine the start and end point predicted radii based on the blood vessel radius ratio coefficient and the start and end point radii in the second key point radius set;

[0144] Determine the predicted blood vessel start and end points based on the first key point radius set and the start and end point predicted radii.

[0145] In some alternative embodiments, the predicted blood vessel start and end points include a predicted blood vessel starting point and a predicted blood vessel ending point; the blood vessel segment set determination module 440 is specifically configured to:

[0146] Based on the predicted blood vessel starting point and the predicted blood vessel ending point, replace the blood vessel starting point and the blood vessel ending point corresponding to the key points of the first blood vessel centerline image to obtain the target key points corresponding to the first blood vessel centerline image;

[0147] Determine the first blood vessel segment set based on the target key points corresponding to the first blood vessel centerline image.

[0148] In some alternative embodiments, the blood vessel registration image determination module 450 is specifically configured to:

[0149] Sample each blood vessel segment in the first blood vessel segment set to obtain a set of first blood vessel segment midpoints;

[0150] Sample each blood vessel segment in the second blood vessel segment set to obtain a set of second blood vessel segment midpoints;

[0151] Determine a vascular registration relationship based on the set of midpoints of the first vascular segment and the set of midpoints of the second vascular segment;

[0152] Based on the vascular registration relationship, register the first vascular image to be registered and the second vascular image to be registered to obtain a vascular registered image.

[0153] The image registration device provided by the embodiments of the present invention can execute the image registration method provided by any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution of the method.

[0154] Embodiment Five

[0155] Figure 9 FIG. shows a schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0156] As Figure 9 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0157] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0158] The processor 11 may be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the image registration method, which includes:

[0159] respectively obtaining a first blood vessel centerline image corresponding to the first blood vessel image to be registered and a second blood vessel centerline image corresponding to the second blood vessel image to be registered;

[0160] determining key points corresponding to the first blood vessel centerline image and key points corresponding to the second blood vessel centerline image;

[0161] performing blood vessel start and end point prediction based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image to obtain predicted blood vessel start and end points;

[0162] determining a first blood vessel segment set based on the key points corresponding to the first blood vessel centerline image and the predicted blood vessel start and end points, and determining a second blood vessel segment set based on the key points corresponding to the second blood vessel centerline image;

[0163] determining a blood vessel registration image based on the first blood vessel segment set and the second blood vessel segment set.

[0164] In some embodiments, the image registration method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the image registration method described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the image registration method by any other suitable means (e.g., by means of firmware).

[0165] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), systems-on-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0166] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0167] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0168] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0169] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0170] A computing system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0171] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and this is not limited herein.

[0172] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An image registration method, characterized in that Including: Respectively obtain a first vascular centerline image corresponding to a first vascular image to be registered, and a second vascular centerline image corresponding to a second vascular image to be registered; Determine key points corresponding to the first vascular centerline image, and key points corresponding to the second vascular centerline image; Based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image, perform prediction of the beginning and end points of the blood vessel to obtain predicted beginning and end points of the blood vessel; Based on the key points corresponding to the first vascular centerline image and the predicted beginning and end points of the blood vessel, determine a first set of vascular segments, and based on the key points corresponding to the second vascular centerline image, determine a second set of vascular segments; Based on the first set of vascular segments and the second set of vascular segments, determine a vascular registration image.

2. The method according to claim 1, characterized in that The step of respectively obtaining a first vascular centerline image corresponding to a first vascular image to be registered, and a second vascular centerline image corresponding to a second vascular image to be registered includes: Obtain a vascular image to be registered, where the vascular image to be registered includes a first vascular image to be registered and a second vascular image to be registered; Segment the first vascular image to be registered to obtain a first vascular segmentation image, and extract the centerline of the first vascular segmentation image to obtain a first vascular centerline image corresponding to the first vascular image to be registered; Segment the second vascular image to be registered to obtain a second vascular segmentation image, and extract the centerline of the second vascular segmentation image to obtain a second vascular centerline image corresponding to the second vascular image to be registered.

3. The method according to claim 1, wherein The step of determining key points corresponding to the first vascular centerline image, and key points corresponding to the second vascular centerline image includes: Input the first vascular centerline image into a pre-trained first key point tracking model to obtain key points corresponding to the first vascular centerline image, or determine the centerline points in the first vascular centerline image that meet the adjacency relationship condition as the key points corresponding to the first vascular centerline image; Input the second vascular centerline image into a pre-trained second key point tracking model to obtain key points corresponding to the second vascular centerline image, or determine the centerline points in the second vascular centerline image that meet the adjacency relationship condition as the key points corresponding to the second vascular centerline image.

4. The method according to claim 1, characterized in that The step of performing prediction of the beginning and end points of the blood vessel based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image to obtain predicted beginning and end points of the blood vessel includes: Based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image, determine a set of vascular segment lengths or a set of key point radii; Based on the set of vascular segment lengths or the set of key point radii, determine the predicted beginning and end points of the blood vessel.

5. The method according to claim 4, wherein The set of vascular segment lengths includes a first set of vascular segment lengths and a second set of vascular segment lengths; The step of determining a set of vascular segment lengths based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image includes: Determine a first set of blood vessel segment lengths based on the key points corresponding to the first blood vessel centerline image, and determine a second set of blood vessel segment lengths based on the key points corresponding to the second blood vessel centerline image; Correspondingly, the determining of the predicted blood vessel start and end points based on the set of blood vessel segment lengths or the set of key point radii includes: Determine a blood vessel length ratio coefficient based on the first set of blood vessel segment lengths and the second set of blood vessel segment lengths; Determine the start and end segment extension lengths based on the blood vessel length ratio coefficient and the start and end segment lengths in the second set of blood vessel segment lengths; Determine the predicted blood vessel start and end points based on the start and end points in the first set of blood vessel segment lengths and the start and end segment extension lengths.

6. The method according to claim 4, wherein The set of key point radii includes a first set of key point radii and a second set of key point radii; The determining of the set of key point radii based on the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image includes: Determine a first set of key point radii based on the key points corresponding to the first blood vessel centerline image, and determine a second set of key point radii based on the key points corresponding to the second blood vessel centerline image; Correspondingly, the determining of the predicted blood vessel start and end points based on the set of blood vessel segment lengths or the set of key point radii includes: Determine a blood vessel radius ratio coefficient based on the first set of key point radii and the second set of key point radii; Determine the predicted start and end point radii based on the blood vessel radius ratio coefficient and the start and end point radii in the second set of key point radii; Determine the predicted blood vessel start and end points based on the first set of key point radii and the predicted start and end point radii.

7. The method according to claim 1, wherein The predicted blood vessel start and end points include a predicted blood vessel start point and a predicted blood vessel end point; The determining of the first set of blood vessel segments based on the key points corresponding to the first blood vessel centerline image and the predicted blood vessel start and end points includes: Based on the predicted blood vessel start point and the predicted blood vessel end point, replace the blood vessel start point and the blood vessel end point corresponding to the key points corresponding to the first blood vessel centerline image to obtain the target key points corresponding to the first blood vessel centerline image; Determine a first set of blood vessel segments based on the target key points corresponding to the first blood vessel centerline image.

8. The method according to claim 1, characterized in that The determining of the blood vessel registration image based on the first set of blood vessel segments and the second set of blood vessel segments includes: Sample each blood vessel segment in the first set of blood vessel segments to obtain a set of intermediate points of the first blood vessel segments; Sample each blood vessel segment in the second set of blood vessel segments to obtain a set of intermediate points of the second blood vessel segments; Determine the blood vessel registration relationship based on the set of intermediate points of the first blood vessel segments and the set of intermediate points of the second blood vessel segments; Based on the blood vessel registration relationship, register the first blood vessel image to be registered and the second blood vessel image to be registered to obtain a blood vessel registration image.

9. An image registration device, characterized in that, Includes: A blood vessel centerline image acquisition module for respectively acquiring a first blood vessel centerline image corresponding to a first blood vessel image to be registered and a second blood vessel centerline image corresponding to a second blood vessel image to be registered; A key point determination module for determining the key points corresponding to the first blood vessel centerline image and the key points corresponding to the second blood vessel centerline image; A vascular start and end prediction module, configured to perform vascular start and end prediction based on the key points corresponding to the first vascular centerline image and the key points corresponding to the second vascular centerline image, so as to obtain predicted vascular start and end points; A vascular segment set determination module, configured to determine a first vascular segment set based on the key points corresponding to the first vascular centerline image and the predicted vascular start and end points, and determine a second vascular segment set based on the key points corresponding to the second vascular centerline image; A vascular registration image determination module, configured to determine a vascular registration image based on the first vascular segment set and the second vascular segment set.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute the image registration method according to any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the image registration method according to any one of claims 1-8 when executed by a processor.

Images