Medical image registration method, system, device, medium, and program product

Abstract

This application relates to a medical image registration method, system, computer device, storage medium, and computer program product. The method includes: acquiring a first vascular feature corresponding to a preoperative medical image; acquiring an intraoperative medical image captured by a vascular imaging device and identifying a second vascular feature in the intraoperative medical image; calculating a first registration relationship between the preoperative medical image and the intraoperative medical image based on the first and second vascular features; acquiring a second registration relationship between the image coordinate system of the vascular imaging device and the world coordinate system; and generating operation navigation information based on the first and second registration relationships. This method improves accuracy through vascular registration, and the use of a vascular imaging device avoids incisions and minimizes harm to the patient.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to a medical image registration method, system, computer device, storage medium, and computer program product. Background Technology

[0002] Neurosurgery is unique compared to other types of surgery because of its intricate network of nerves and blood vessels, the delicate structure and vital functions of brain tissue, and the fact that even slight mishaps can damage nerves and blood vessels, leading to severe disability or even death. Furthermore, the nervous system has limited mobility and cannot be easily manipulated during surgery. Therefore, traditional neurosurgery often sacrifices significant portions of skull structure to expose lesions. Neuronavigation systems are among the most important auxiliary devices in minimally invasive neurosurgery, serving as invaluable tools for neurosurgeons. Similar to car navigation, they provide real-time information about the surgeon's current location, enabling more accurate predictions and decisions.

[0003] Traditional techniques for registering preoperative and intraoperative medical images include the following methods:

[0004] (1) Rigid body registration based on bone screws: This method requires inserting multiple bone screws or other devices into the patient's skull before taking images. It is an invasive registration method that causes additional harm to the patient.

[0005] (2) Face-based registration: This method uses point cloud registration, with two point clouds: one extracted from the face using DICOM images preoperatively, and the other acquired during surgery. However, since the human face is mainly soft tissue, it is prone to deformation, which reduces the similarity between the two point clouds and lowers the registration accuracy. Other products use certain facial feature points (such as the eye socket and forehead) for registration, but because the registration area is far from the surgical area, the registration error is amplified, resulting in a decrease in surgical accuracy.

[0006] Therefore, the current method of registering preoperative and intraoperative medical images has relatively low registration accuracy. Summary of the Invention

[0007] Therefore, it is necessary to provide a medical image registration method, system, computer equipment, storage medium, and computer program product that can improve registration accuracy in response to the above-mentioned technical problems.

[0008] In a first aspect, this application provides a medical image registration method, the method comprising:

[0009] Obtain the first vascular features corresponding to the preoperative medical images;

[0010] Acquire intraoperative medical images captured by a vascular imaging device and identify second vascular features in the intraoperative medical images;

[0011] Based on the first vascular feature and the second vascular feature, the first registration relationship between the preoperative medical image and the intraoperative medical image is calculated;

[0012] Obtain the second registration relationship between the image coordinate system and the world coordinate system of the vascular imaging instrument;

[0013] Operation navigation information is generated based on the first registration relationship and the second registration relationship.

[0014] In one embodiment, prior to acquiring the first vascular feature corresponding to the preoperative medical image, the process includes:

[0015] Acquire the corresponding preoperative medical images and segment the preoperative medical images to obtain a vascular mask;

[0016] The blood vessel surface model is obtained by reconstructing and classifying the blood vessel mask;

[0017] The first vascular feature corresponding to the preoperative medical image is obtained based on the vascular surface model; the first vascular feature includes at least one of vascular topology, vascular point cloud data, vascular point cloud features, and vascular image features.

[0018] In one embodiment, obtaining the first vascular feature corresponding to the preoperative medical image based on the vascular surface model includes at least one of the following:

[0019] Extract the first skeleton of the blood vessel surface model, and perform a traversal analysis on the first skeleton to obtain the blood vessel topology; or

[0020] The isosurface of the blood vessel surface model is used to obtain blood vessel point cloud data; or

[0021] Based on the blood vessel surface model and the blood vessel point cloud data, the space containing the points in the blood vessel point cloud data is divided, and point cloud features are extracted from the blood vessel point cloud data in each divided space to obtain blood vessel point cloud features; or

[0022] Based on the blood vessel surface model and the blood vessel point cloud data, the pose of the surface blood vessels is calculated; a blood vessel image is generated based on the pose of the surface blood vessels, and image feature extraction is performed on the blood vessel image to obtain blood vessel image features.

[0023] In one embodiment, acquiring intraoperative medical images captured by a vascular imaging device and identifying second vascular features in the intraoperative medical images includes at least one of the following:

[0024] The initial point cloud data of different parts of the target object collected by the vascular imaging instrument is obtained, the initial power data is fitted to obtain the second skeleton, and the second skeleton is traversed and analyzed to obtain the second vascular feature, which is the vascular topology.

[0025] Acquire initial point cloud data of different parts of the target object collected by a vascular imaging instrument; stitch the initial point cloud data together to obtain a second vascular feature, which is vascular point cloud data; or

[0026] Acquire target point cloud data of the target object collected by a vascular imaging device, and extract point cloud features from the target point cloud data to obtain a second vascular feature, wherein the second vascular feature is a vascular point cloud feature; or

[0027] A two-dimensional image of the target object is acquired by a vascular imaging instrument, and image features are extracted from the two-dimensional image to obtain a second vascular feature, which is a vascular image feature.

[0028] In one embodiment, calculating the first registration relationship between the preoperative medical image and the intraoperative medical image based on the first vascular feature and the second vascular feature includes at least one of the following:

[0029] The first registration relationship is obtained by registering the vascular topology corresponding to the preoperative medical image with the vascular topology corresponding to the intraoperative medical image.

[0030] The first registration relationship is obtained by performing point cloud registration between the vascular point cloud data corresponding to the preoperative medical images and the vascular point cloud data corresponding to the intraoperative medical images; or

[0031] Feature matching is performed between the point cloud features corresponding to the preoperative medical images and the point cloud features corresponding to the intraoperative medical images to determine a first initial registration region from the preoperative medical images. The vascular point cloud data in the first initial registration region is then registered with the vascular point cloud data corresponding to the intraoperative medical images to obtain a first registration relationship; or

[0032] By matching the image features corresponding to the preoperative medical image with the image features corresponding to the intraoperative medical image, a second initial registration region is determined from the preoperative medical image. The vascular point cloud data in the second initial registration region is then matched with the vascular point cloud data corresponding to the intraoperative medical image to obtain a first registration relationship.

[0033] In one embodiment, obtaining the second registration relationship between the image coordinate system and the world coordinate system of the vascular imaging instrument includes:

[0034] Acquire the calibration plate image captured by the vascular imaging instrument;

[0035] The calibration board image is registered with the real calibration board to obtain the third registration relationship;

[0036] The pose relationship between the real calibration plate and the target corresponding to the real calibration plate, the first transformation relationship between the target corresponding to the real calibration plate and the coordinate system of the positioning device, and the second transformation relationship between the coordinate system of the positioning device and the target corresponding to the vascular imaging instrument are obtained.

[0037] Based on the third registration relationship, the pose relationship, the first transformation relationship, and the second transformation relationship, the second registration relationship between the image coordinate system and the world coordinate system of the vascular imaging instrument is determined.

[0038] In one embodiment, generating operation navigation information based on the first registration relationship and the second registration relationship includes:

[0039] Obtain the position of medical instruments in the world coordinate system during surgery;

[0040] Based on the first registration relationship and the second registration relationship, the target object in the preoperative medical image is mapped to the world coordinate system;

[0041] Operational navigation information is generated based on the position of the target object in the world coordinate system and the position of the intraoperative medical device in the world coordinate system.

[0042] Secondly, this application also provides a medical image registration system, the system comprising:

[0043] A vascular imaging device is used to acquire intraoperative medical images;

[0044] The processor is used to execute the medical image registration method described above to generate intraoperative navigation information.

[0045] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method in any of the above embodiments.

[0046] In another aspect, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0047] The aforementioned medical image registration method, system, computer equipment, storage medium, and computer program product acquire first vascular features from preoperative medical images and second vascular features from intraoperative medical images using a vascular imaging device. Based on these first and second vascular features, a first registration relationship between the preoperative and intraoperative medical images is calculated. Then, a second registration relationship between the image coordinate system of the vascular imaging device and the world coordinate system is obtained. Operational navigation information is generated based on these first and second registration relationships. This vascular registration method improves accuracy, and the use of a vascular imaging device avoids incisions and minimizes harm to the patient. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of a medical image registration system in one embodiment;

[0049] Figure 2 This is a diagram illustrating an application scenario of the surgical robot in one embodiment;

[0050] Figure 3 This is a schematic diagram of a processor module in one embodiment;

[0051] Figure 4 This is a flowchart illustrating a medical image registration method in one embodiment;

[0052] Figure 5 This is a schematic diagram of a blood vessel imaging device in one embodiment;

[0053] Figure 6 This is a schematic diagram of the first blood vessel feature extraction process in one embodiment;

[0054] Figure 7 This is a flowchart of the blood vessel topology extraction step in one embodiment;

[0055] Figure 8 This is a flowchart of the blood vessel point cloud data extraction step in one embodiment;

[0056] Figure 9 This is a flowchart of the blood vessel point cloud feature extraction step in one embodiment;

[0057] Figure 10 This is a schematic diagram of an octree in one embodiment;

[0058] Figure 11 This is a flowchart of the vascular image feature extraction step in one embodiment;

[0059] Figure 12 This is a schematic diagram of the extraction of second blood vessel features in one embodiment;

[0060] Figure 13A flowchart illustrating the registration of blood vessel point cloud data in one embodiment;

[0061] Figure 14 A flowchart for vascular point cloud data registration in another embodiment;

[0062] Figure 15 This is a flowchart for vascular point cloud data registration in another embodiment;

[0063] Figure 16 This is a schematic diagram of the calibration process of a blood vessel imaging device in one embodiment;

[0064] Figure 17 This is a schematic diagram of a surgical scenario in one embodiment;

[0065] Figure 18 This is a schematic diagram of a surgical scenario in another embodiment;

[0066] Figure 19 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0068] Specifically, in combination Figure 1 As shown, a medical image registration system is provided, which includes a vascular imaging device 100 and a processor 200. The vascular imaging device 100 is used to acquire intraoperative medical images, such as vascular images of a target area during surgery, such as the surgical area. The processor 200 can acquire the intraoperative medical images acquired by the vascular imaging device 100, identify second vascular features in the intraoperative medical images, and configure the second vascular features with the corresponding first vascular features in the preoperative medical images to obtain a first registration relationship between the preoperative and intraoperative medical images. In addition, the processor 200 can also acquire a second registration relationship between the image coordinate system of the vascular imaging device 100 and the world coordinate system. Thus, operation navigation information is generated based on the first and second registration relationships.

[0069] For ease of explanation, an application scenario diagram of a surgical robot is provided. This application scenario includes a robotic surgical system 300, an operating table 400, and medical imaging equipment, such as a visual tracker 304. Combined with... Figure 2 As shown, Figure 2This is a schematic diagram of a robotic surgical system 300 in one embodiment. The robotic surgical system 300 includes a visual navigation cart 301, a surgical control cart 302, and a surgical robotic arm 303, etc. The surgical robotic arm 303 can be a seven-axis surgical robotic arm 303 or other types, without specific limitations. A processor 200 is installed inside the surgical control cart 302, and a visual tracker 304 is installed on the visual navigation cart 301 and connected to the processor 200. The angiography device 100 can be controlled by the surgical robotic arm or fixed to the operating table 400.

[0070] For ease of explanation, the following is combined Figure 3 As shown, Figure 3 This is a schematic diagram of the processor modules in one embodiment. In this embodiment, the processor 200 includes a preoperative planning module 500 and an intraoperative planning module 600. The preoperative planning module 500 includes an image import unit 501, a vascular feature extraction unit 502, and a storage unit 503. The intraoperative planning module 600 includes a calibration unit 601, an image acquisition unit 602, a feature extraction unit 603, a registration unit 604, and a display unit 605. The image import unit 501 is used to parse imported preoperative medical images, such as DICOM images. The vascular feature extraction unit 502 is used to extract blood vessels and features from the imported preoperative medical images. The storage unit 503 stores the extracted vascular features. The calibration unit 601 is used to calibrate the image acquisition unit 602 and establish a mapping relationship between the local coordinate system of the vascular imaging instrument 100 and the positioning device located on it. The image acquisition unit 602 is used to acquire the target site during surgery, such as three-dimensional point cloud data and two-dimensional image data mixed with superficial blood vessels of the head. The feature extraction unit 603 is used to extract blood vessel features from the three-dimensional point cloud data and two-dimensional image data obtained by the image acquisition unit 602. The registration unit 604 is used to establish the mapping relationship between the preoperative and intraoperative point clouds. The display unit 605 is used to display the surgical instruments in real time through the mapping relationship obtained by the registration unit 604.

[0071] In one embodiment, such as Figure 4 As shown, a medical image registration method is provided, which is applied to... Figure 1 Taking the processor in the example, the following steps are included:

[0072] S402: Obtain the first vascular features corresponding to the preoperative medical images.

[0073] Specifically, preoperative medical images are medical images of the surgical site acquired using medical imaging equipment before the surgical procedure. The first vascular feature is obtained by performing vascular feature recognition on the imported preoperative medical images. This first vascular feature may include at least one of vascular topology, vascular point cloud data, vascular point cloud features, and vascular image features.

[0074] In this embodiment, preoperative medical images are imported, and then the processor 200 extracts blood vessels and features from the preoperative medical images to obtain the first blood vessel features.

[0075] S404: Acquire intraoperative medical images from the angiography system and identify secondary vascular features in the intraoperative medical images.

[0076] Specifically, specifically can be combined with Figure 5 As shown, the vascular imaging device 100 projects near-infrared light of a specific frequency onto the skin surface. Because hemoglobin in the blood absorbs infrared light more strongly than other tissues, when the light returns to the device, photosensitive components collect the light signal and process it to generate a vascular distribution contour map. This image 700 is then clearly projected onto the skin surface using micro-projection technology. The intraoperative medical image mentioned here is obtained by processing the light signal collected by the photosensitive components; it can be point cloud data or a two-dimensional image, and the two-dimensional image is projected onto the skin surface.

[0077] The second vascular feature is obtained by identifying vascular features in intraoperative medical images. The second vascular feature may include at least one of vascular topology, vascular point cloud data, vascular point cloud features, and vascular image features.

[0078] One point to note is that during the procedure, a 100-degree angiography system is used to monitor blood vessels in real time and verify the registration results. If any deviation of the blood vessels is detected, it can be corrected in real time.

[0079] S406: Based on the first and second vascular features, the first registration relationship between the preoperative medical images and the intraoperative medical images is calculated.

[0080] Specifically, the first registration relationship is obtained by registering preoperative medical images and intraoperative medical images. The registration method is obtained by registering the same type of first and second vascular features, such as point cloud registration and feature registration, etc., without specific limitations here.

[0081] S408: Obtain the second registration relationship between the image coordinate system and the world coordinate system of the angiography instrument.

[0082] S410: Generate operation navigation information based on the first registration relationship and the second registration relationship.

[0083] Specifically, the second registration relationship is the process of calibrating the angiography system. Its main purpose is to obtain the second registration relationship between the image coordinate system of the angiography system and the world coordinate system. The world coordinate system can refer to the coordinate system corresponding to the optical positioning instrument 800.

[0084] The subsequent processor 200 acquires the position information of the medical device in the world coordinate system through the optical locator 800, transforms it to the image coordinate system of the angiography system through a second registration relationship, and then transforms it to the preoperative medical image through a first registration relationship. This allows the position of the medical device to be displayed in real time in the preoperative medical image, or for operation navigation based on the position of the medical device in the preoperative medical image and preoperative planning, etc., without specific limitations. In other embodiments, the preoperative planning can be mapped to the intraoperative medical image through the first registration relationship, and the position of the medical device can be mapped to the intraoperative medical image according to the second registration relationship. This allows the position of the medical device and the preoperative planning to be displayed in the intraoperative medical image, and operation navigation to be performed based on the position of the medical device and the preoperative planning.

[0085] In the above embodiments, by acquiring the first vascular features of preoperative medical images and the second vascular features of intraoperative medical images acquired by a vascular imaging device, the first registration relationship between the preoperative and intraoperative medical images can be calculated based on the first and second vascular features. Then, the second registration relationship between the image coordinate system of the vascular imaging device and the world coordinate system is obtained. Operation navigation information is generated based on the first and second registration relationships. The accuracy is improved by vascular registration, and the use of a vascular imaging device can avoid incisions and avoid harm to the patient.

[0086] Furthermore, vascular registration eliminates the need for preoperative bone screws or head frames, avoiding the harm caused by such instruments and making the surgery safer. Vascular registration allows for registration within the surgical area, improving surgical precision. It also enables real-time monitoring of the patient's posture using a vascular imaging device, allowing for real-time correction of the registration results and enhancing surgical accuracy and safety. Using vascular registration reduces contraindications, such as the requirement for relatively loose facial tissues for facial point cloud registration, which is not a limitation in this embodiment, improving the system's versatility. The vascular imaging device described in the vascular registration method projects incision locations that avoid blood vessels onto the patient's epidermis, helping the surgeon avoid blood vessels during incision.

[0087] In one embodiment, before obtaining the first vascular feature corresponding to the preoperative medical image, the method includes: obtaining the corresponding preoperative medical image and segmenting the preoperative medical image to obtain a vascular mask; reconstructing and classifying the vascular mask to obtain a vascular surface model; obtaining the first vascular feature corresponding to the preoperative medical image based on the vascular surface model; the first vascular feature includes at least one of vascular topology, vascular point cloud data, vascular point cloud features, and vascular image features.

[0088] Specific combination Figure 6 As shown, Figure 6 This is a schematic diagram of the first blood vessel feature extraction process in one embodiment. In this embodiment, the processor 200 first acquires imported preoperative medical images, such as DICOM images, then segments the preoperative medical images to obtain a blood vessel mask, and reconstructs a blood vessel surface model based on the mask. The first blood vessel feature is obtained by extracting point cloud features or image features from the blood vessel surface model. The skeleton of the blood vessel surface model is extracted, and the extracted skeleton is pruned, traversed, numbered, and processed to establish a topological structure. Finally, the extracted first blood vessel feature is stored. Optionally, the processor 200 can store the blood vessel mask, the blood vessel surface model, and the first blood vessel feature.

[0089] In the above embodiments, vascular features are extracted by segmenting and reconstructing preoperative medical images.

[0090] In one embodiment, obtaining the first vascular feature corresponding to the preoperative medical image based on the vascular surface model includes at least one of the following: extracting the first skeleton of the vascular surface model and performing traversal analysis on the first skeleton to obtain the vascular topology; or extracting isosurfaces from the vascular surface model to obtain vascular point cloud data; or dividing the space where the points in the vascular point cloud data are located based on the vascular surface model and the vascular point cloud data, and extracting point cloud features from the vascular point cloud data in each divided space to obtain vascular point cloud features; or calculating the pose of the surface blood vessels based on the vascular surface model and the vascular point cloud data; generating a vascular image based on the pose of the surface blood vessels, and extracting image features from the vascular image to obtain vascular image features.

[0091] Specifically, this embodiment mainly presents the extraction process for various first blood vessel features.

[0092] See details Figures 7 to 11 As shown, where Figure 7 This is a flowchart of the blood vessel topology extraction step in one embodiment. Figure 8 This is a flowchart of the blood vessel point cloud data extraction step in one embodiment. Figure 9 This is a flowchart of the blood vessel point cloud feature extraction step in one embodiment. Figure 10This is a schematic diagram of an octree in one embodiment. Figure 11 This is a flowchart of the vascular image feature extraction step in one embodiment.

[0093] See Figure 7 As shown, preoperative medical images are imported, and blood vessels are segmented and extracted using a segmentation algorithm to obtain a vascular mask. Morphological processing, such as hole filling, is applied to the vascular mask to make its 3D model a single connected domain. The vascular mask is then classified, and superficial blood vessels are extracted. One classification method uses the SVN classifier; other methods can also be used in other embodiments. The bronchial mask after hole filling is refined, and its skeleton is extracted.

[0094] Traverse the skeleton, analyze it using the 26-neighborhood method or other methods, mark the vessel segment number, number of bifurcations, and other information of each point, and establish the vessel topology.

[0095] See Figure 8 As shown, preoperative medical images are imported, and a segmentation algorithm is used to segment and extract blood vessels to obtain a vascular mask; morphological processing such as hole filling is performed on the vascular mask to make its three-dimensional model a single connected domain; target algorithms, such as Marching Cubes algorithm, are used to extract isosurfaces from the mask to generate point cloud data.

[0096] See Figure 9 and Figure 10 As shown, preoperative medical images are imported, and a segmentation algorithm is used to segment and extract blood vessels to obtain vascular mask and superficial vascular point cloud data; based on the size of the point cloud bounding box, the space where the point cloud is located is divided into octrees (e.g., ...). Figure 10 As shown), the smallest unit of spatial division refers to the imaging parameters of the vascular imaging instrument 100; feature vectors are extracted from the point cloud data in each space, such as SIFT features. In other embodiments, PFH, FPFH, SHOT, C-SHOT, RSD, 3D shape descriptors, etc. can also be used. The extracted features are the point cloud feature data.

[0097] See Figure 11 As shown, preoperative medical images are imported, and a segmentation algorithm is used to segment and extract blood vessels to obtain vascular masks and superficial vascular point cloud data; different poses around the superficial blood vessels are calculated to generate two-dimensional images; various parameters of the vascular imaging instrument 100 are used to enable the rendering engine to generate two-dimensional images based on the current pose; finally, vascular image features, such as ORB features, are extracted. In other embodiments, other image features can also be used.

[0098] In the above embodiments, various methods are used to extract vascular features.

[0099] In one embodiment, acquiring intraoperative medical images collected by the vascular imaging device 100 and identifying second vascular features in the intraoperative medical images includes at least one of the following: acquiring initial point cloud data of different parts of the target object collected by the vascular imaging device 100, fitting the initial point cloud data to obtain a second skeleton, and performing traversal analysis on the second skeleton to obtain second vascular features, wherein the second vascular features are vascular topology structures; acquiring initial point cloud data of different parts of the target object collected by the vascular imaging device 100, performing point cloud stitching on the initial point cloud data to obtain second vascular features, wherein the second vascular features are vascular point cloud data; or acquiring target point cloud data of the target object collected by the vascular imaging device 100, performing point cloud feature extraction on the target point cloud data to obtain second vascular features, wherein the second vascular features are vascular point cloud features; or acquiring a two-dimensional image of the target object collected by the vascular imaging device 100, performing image feature extraction on the two-dimensional image to obtain second vascular features, wherein the second vascular features are vascular image features.

[0100] Specifically, in combination Figure 12 As shown, for the vascular imaging instrument 100, the pose of the local coordinate system of the vascular imaging instrument 100 in the intraoperative physical space is first acquired to complete the calibration of the vascular imaging instrument 100. Then, the acquisition unit is used to acquire superficial vascular data of the patient's head in real time. As described in the preoperative feature extraction process, feature extraction is performed on the point cloud data or the two-dimensional image. The acquired vascular point cloud data is fitted to generate a skeleton, and then pruned, traversed, and numbered to establish a topological structure. Finally, the features extracted preoperatively and during the operation are used to register the blood vessels to obtain the patient's pose in the intraoperative physical space. Based on the patient's pose in the intraoperative physical space, the positional relationship between the surgical instruments and the patient is given in real time.

[0101] In one embodiment, a first registration relationship between the preoperative medical image and the intraoperative medical image is calculated based on the first vascular feature and the second vascular feature, including at least one of the following: registering the vascular topology corresponding to the preoperative medical image with the vascular topology corresponding to the intraoperative medical image to obtain the first registration relationship; registering the vascular point cloud data corresponding to the preoperative medical image with the vascular point cloud data corresponding to the intraoperative medical image to obtain the first registration relationship; or performing feature matching by matching the point cloud features corresponding to the preoperative medical image with the point cloud features corresponding to the intraoperative medical image to determine a first initial registration region from the preoperative medical image, and registering the vascular point cloud data in the first initial registration region with the vascular point cloud data corresponding to the intraoperative medical image to obtain the first registration relationship; or performing feature matching by matching the image features corresponding to the preoperative medical image with the image features corresponding to the intraoperative medical image to determine a second initial registration region from the preoperative medical image, and registering the vascular point cloud data in the second initial registration region with the vascular point cloud data corresponding to the intraoperative medical image to obtain the first registration relationship.

[0102] Specifically, see Figures 13 to 15 As shown, Figure 13 This is a flowchart illustrating the registration of blood vessel point cloud data in one embodiment. Figure 14 This is a flowchart of blood vessel point cloud data registration in another embodiment. Figure 15 This is a flowchart for vascular point cloud data registration in another embodiment.

[0103] Among them, see Figure 13 As shown, during the operation, a vascular imaging device 100 is used to acquire point cloud data of the patient's superficial blood vessels, i.e., intraoperative medical images. During the operation, data of a specific blood vessel can be acquired, and the vascular imaging device 100 can be moved to acquire data from different parts of the patient's head. Based on the pose of the vascular imaging device 100 under the optical positioning device or a target, the point cloud data acquired by the vascular imaging device 100 is stitched together. The stitched point cloud data is then subjected to noise reduction processing, and ICP registration is performed with the point cloud data corresponding to the preoperative medical images to obtain the first registration relationship mentioned above, thereby establishing a mapping relationship between the patient and the optical positioning device or a target.

[0104] Combination Figure 14As shown, this is an optimized algorithm for ICP registration. Due to the limited number of point clouds acquired during surgery, ICP registration with all point clouds corresponding to preoperative medical images is slow. Coarse registration can be performed using features of preoperatively generated 2D images from different locations to determine a coarse region. Then, precise registration is performed based on the point cloud data of this coarse region. Specifically, during surgery, a vascular imaging device 100 acquires a frame of superficial vascular point cloud data and a 2D image of the patient. Using the same feature extraction unit 603 as before surgery, ORB features (Oriented Fast and Rotated Brief) are extracted from this 2D image. These features are compared and scored with the preoperative ORB features, and the ORB feature closest to the current frame is selected. The pose space corresponding to this ORB feature is used as the initial registration result. Point cloud registration is then used for re-registration to correct the result.

[0105] Combination Figure 15 As shown, this is an optimized algorithm for ICP registration. Due to the limited number of point clouds acquired during surgery, ICP registration with all point clouds corresponding to preoperative medical images is slow. Coarse registration can be performed using feature vectors extracted preoperatively from different spaces to determine a coarse region. Then, precise registration is performed based on the point cloud data of this coarse region. Specifically, during surgery, a frame of superficial blood vessel point cloud data is acquired using a vascular imaging device 100. The same feature extraction unit 603 as preoperatively is used to extract SIFT (Scale-Invariant Feature Transform) features from this point cloud data. These features are compared and scored with the preoperative SIFT features, and the SIFT feature closest to the current frame is selected. The partitioned space corresponding to this SIFT feature is used as the initial registration result. Point cloud registration is then used for re-registration to correct the result.

[0106] The above embodiments provide a variety of registration methods to achieve rapid registration of blood vessels.

[0107] In one embodiment, obtaining the second registration relationship between the image coordinate system of the vascular imaging device 100 and the world coordinate system includes: acquiring a calibration plate image captured by the vascular imaging device 100; registering the calibration plate image with a real calibration plate 900 to obtain a third registration relationship; acquiring the pose relationship between the real calibration plate 900 and the target corresponding to the real calibration plate 900, the first transformation relationship between the target corresponding to the real calibration plate 900 and the coordinate system of the positioning device, and the second transformation relationship between the coordinate system of the positioning device and the target corresponding to the vascular imaging device 100; and determining the second registration relationship between the image coordinate system of the vascular imaging device 100 and the world coordinate system based on the third registration relationship, the pose relationship, the first transformation relationship, and the second transformation relationship.

[0108] Specifically, in combination Figure 16 As shown, Figure 16This is a schematic diagram of the calibration process of a blood vessel imaging device in one embodiment. In this embodiment, the optical locator 800 can identify the poses of the first target 801 and the second target 802, that is, provide the poses of the first target 801 and the second target 802 in the coordinate system of the optical locator 800. The pose relationship between the second target 802 and the actual calibration plate 900 can be calibrated using engineering data, a coordinate measuring machine, or other methods. The blood vessel imaging device 100 can acquire the calibration plate image of the calibration table, thereby obtaining the pose of the target 801 and the second target 802 in the calibration table. The image can be registered with the real calibration plate 900 to obtain the third registration relationship, which is the third registration relationship between the image coordinate system and the real calibration plate 900. Then, based on the pose relationship between the second target 802 and the real calibration plate 900, the pose relationship between the second target 802 and the optical locator 800, and the pose relationship between the first target 801 and the optical locator 800, the second registration relationship between the image coordinate system of the vascular imaging instrument 100 and the first target 801, that is, the world coordinate system, can be obtained.

[0109] In one embodiment, generating operation navigation information based on a first registration relationship and a second registration relationship includes: obtaining the position of the intraoperative medical device in the world coordinate system; mapping the target object in the preoperative medical image to the world coordinate system based on the first registration relationship and the second registration relationship; and generating operation navigation information based on the position of the target object in the world coordinate system and the position of the intraoperative medical device in the world coordinate system.

[0110] Specifically, in this embodiment, the target object in the preoperative medical image is mapped to the world coordinate system according to the first registration relationship and the second registration relationship. In this way, operation navigation information is generated based on the position of the target object in the world coordinate system and the position of the intraoperative medical device in the world coordinate system, so as to facilitate intraoperative navigation.

[0111] To make it easier to understand, you can combine... Figure 17 and Figure 18 As shown, Figure 17 This is a schematic diagram of a surgical scenario in one embodiment. Figure 18 This is a schematic diagram of a surgical scenario in another embodiment.

[0112] exist Figure 17In this process, the third target 803 is used to calibrate the vascular imaging instrument 100, the fourth target 804 is used to calibrate the position of the patient's head, and the fifth target 805 is used to calibrate the position of the medical device 806. The vascular imaging instrument 100 scans the patient's head to obtain intraoperative medical images, which are then registered with preoperative medical images. Since the vascular imaging instrument 100 is pre-calibrated, a mapping relationship between the image coordinate system of the fourth target 804 and the vascular imaging instrument 100 is established. Subsequently, since the medical device 806 carries the fifth target 805, the positional relationship between the medical device 806 and the patient can be obtained in real time and visualized.

[0113] exist Figure 18 In this process, the third target 803 is used to calibrate the angiography system 100, and the fifth target 805 is used to calibrate the position of the medical device 806. The patient's head is fixed to the operating table or other device using a head fixation frame to prevent relative displacement between the patient and the optical positioning system 800. Intraoperative medical images are acquired by scanning the patient's head with the angiography system 100 and registered with preoperative medical images. Since the angiography system 100 is pre-calibrated, a mapping relationship between the image coordinate systems of the optical positioning system 800 and the angiography system 100 is established. Subsequently, because the medical device 806 carries the fifth target 805, the positional relationship between the medical device 806 and the patient can be obtained in real time and visualized.

[0114] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0115] Based on the same inventive concept, this application also provides a medical image registration apparatus for implementing the medical image registration method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more medical image registration apparatus embodiments provided below can be found in the limitations of the medical image registration method described above, and will not be repeated here.

[0116] In one embodiment, a medical image registration device is provided, comprising:

[0117] The vascular feature extraction unit 502 is used to acquire the first vascular feature corresponding to the preoperative medical image;

[0118] The feature extraction unit 603 is used to acquire intraoperative medical images collected by the vascular imaging instrument 100 and identify second vascular features in the intraoperative medical images.

[0119] The registration unit 604 is used to calculate the first registration relationship between the preoperative medical image and the intraoperative medical image based on the first vascular feature and the second vascular feature.

[0120] The calibration unit 601 is used to obtain the second registration relationship between the image coordinate system and the world coordinate system of the vascular imaging instrument 100;

[0121] Display unit 605 is used to generate operation navigation information based on the first registration relationship and the second registration relationship.

[0122] In one embodiment, the vascular feature extraction unit 502 is used to acquire the corresponding preoperative medical image and segment the preoperative medical image to obtain a vascular mask; reconstruct and classify the vascular mask to obtain a vascular surface model; and obtain a first vascular feature corresponding to the preoperative medical image based on the vascular surface model; the first vascular feature includes at least one of vascular topology, vascular point cloud data, vascular point cloud features, and vascular image features.

[0123] In one embodiment, the vascular feature extraction unit 502 obtains the first vascular feature corresponding to the preoperative medical image based on the vascular surface model, including at least one of the following: extracting the first skeleton of the vascular surface model and performing traversal analysis on the first skeleton to obtain the vascular topology; or performing isosurface extraction on the vascular surface model to obtain vascular point cloud data; or dividing the space where the points in the vascular point cloud data are located according to the vascular surface model and the vascular point cloud data, and performing point cloud feature extraction on the vascular point cloud data in each divided space to obtain vascular point cloud features; or calculating the pose of the surface blood vessels according to the vascular surface model and the vascular point cloud data; generating a vascular image according to the pose of the surface blood vessels, and performing image feature extraction on the vascular image to obtain vascular image features.

[0124] In one embodiment, the feature extraction unit 603, in acquiring intraoperative medical images acquired by the vascular imaging instrument 100 and identifying second vascular features in the intraoperative medical images, includes at least one of the following: acquiring initial point cloud data of different parts of the target object acquired by the vascular imaging instrument 100, fitting the initial point cloud data to obtain a second skeleton, and performing traversal analysis on the second skeleton to obtain a second vascular feature, wherein the second vascular feature is a vascular topology; acquiring initial point cloud data of different parts of the target object acquired by the vascular imaging instrument 100, performing point cloud stitching on the initial point cloud data to obtain a second vascular feature, wherein the second vascular feature is vascular point cloud data; or acquiring target point cloud data of the target object acquired by the vascular imaging instrument 100, performing point cloud feature extraction on the target point cloud data to obtain a second vascular feature, wherein the second vascular feature is a vascular point cloud feature; or acquiring a two-dimensional image of the target object acquired by the vascular imaging instrument 100, performing image feature extraction on the two-dimensional image to obtain a second vascular feature, wherein the second vascular feature is a vascular image feature.

[0125] In one embodiment, the registration unit 604 calculates a first registration relationship between the preoperative medical image and the intraoperative medical image based on the first vascular feature and the second vascular feature, including at least one of the following: registering the vascular topology corresponding to the preoperative medical image with the vascular topology corresponding to the intraoperative medical image to obtain a first registration relationship; registering the vascular point cloud data corresponding to the preoperative medical image with the vascular point cloud data corresponding to the intraoperative medical image to obtain a first registration relationship; or performing feature matching by matching the point cloud features corresponding to the preoperative medical image with the point cloud features corresponding to the intraoperative medical image to determine a first initial registration region from the preoperative medical image, and performing point cloud registration by matching the vascular point cloud data in the first initial registration region with the vascular point cloud data corresponding to the intraoperative medical image to obtain a first registration relationship; or performing feature matching by matching the image features corresponding to the preoperative medical image with the image features corresponding to the intraoperative medical image to determine a second initial registration region from the preoperative medical image, and performing point cloud registration by matching the vascular point cloud data in the second initial registration region with the vascular point cloud data corresponding to the intraoperative medical image to obtain a first registration relationship.

[0126] In one embodiment, the calibration unit 601 is used to acquire the calibration plate image collected by the vascular imaging instrument 100; register the calibration plate image with the real calibration plate 900 to obtain a third registration relationship; acquire the pose relationship between the real calibration plate 900 and the target corresponding to the real calibration plate 900, the first transformation relationship between the target corresponding to the real calibration plate 900 and the coordinate system of the positioning device, and the second transformation relationship between the coordinate system of the positioning device and the target corresponding to the vascular imaging instrument 100; and determine the second registration relationship between the image coordinate system of the vascular imaging instrument 100 and the world coordinate system based on the third registration relationship, the pose relationship, the first transformation relationship, and the second transformation relationship.

[0127] In one embodiment, the display unit 605 is used to acquire the position of the intraoperative medical device in the world coordinate system; map the target object in the preoperative medical image to the world coordinate system according to the first registration relationship and the second registration relationship; and generate operation navigation information according to the position of the target object in the world coordinate system and the position of the intraoperative medical device in the world coordinate system.

[0128] Each module in the aforementioned medical image registration device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor 200 in the computer device, or stored in the memory of the computer device as software, so that the processor 200 can call and execute the corresponding operations of each module.

[0129] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 19 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a medical image registration method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0130] Those skilled in the art will understand that Figure 19 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0131] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0132] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0133] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0134] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A medical image registration method, characterized by, The method comprises: obtaining first blood vessel features corresponding to a preoperative medical image; obtaining an intraoperative medical image collected by a blood vessel imaging device and identifying second blood vessel features in the intraoperative medical image; calculating a first registration relationship between the preoperative medical image and the intraoperative medical image according to the first blood vessel features and the second blood vessel features; obtaining a second registration relationship between an image coordinate system of the blood vessel imaging device and a world coordinate system; generating operation navigation information according to the first registration relationship and the second registration relationship; The method comprises: obtaining a calibration board image collected by the blood vessel imaging device; registering the calibration board image with a real calibration board to obtain a third registration relationship; obtaining a pose relationship between the real calibration board and a target corresponding to the real calibration board, a first conversion relationship between the target corresponding to the real calibration board and a coordinate system of a positioning device, and a second conversion relationship between the coordinate system of the positioning device and a target corresponding to the blood vessel imaging device; determining the second registration relationship between the image coordinate system of the blood vessel imaging device and the world coordinate system according to the third registration relationship, the pose relationship, the first conversion relationship, and the second conversion relationship.

2. The method of claim 1, wherein, Before the method comprises: obtaining a corresponding preoperative medical image and segmenting the preoperative medical image to obtain a blood vessel mask; reconstructing and classifying the blood vessel mask to obtain a blood vessel surface model; obtaining first blood vessel features corresponding to the preoperative medical image according to the blood vessel surface model; the first blood vessel features include at least one of blood vessel topology, blood vessel point cloud data, blood vessel point cloud features, and blood vessel image features.

3. The method of claim 2, wherein, The method comprises: extracting a first skeleton of the blood vessel surface model and performing traversal analysis on the first skeleton to obtain blood vessel topology; or extracting an isosurface from the blood vessel surface model to obtain blood vessel point cloud data; or dividing a space in which a point in the blood vessel point cloud data is located according to the blood vessel surface model and the blood vessel point cloud data, and extracting point cloud features of blood vessel point cloud data in each divided space to obtain blood vessel point cloud features; or calculating a pose of a surface blood vessel according to the blood vessel surface model and the blood vessel point cloud data, generating a blood vessel image according to the pose of the surface blood vessel, and extracting image features of the blood vessel image to obtain blood vessel image features.

4. The method of claim 1, wherein, The method comprises: obtaining initial point cloud data of different parts of a target object collected by a blood vessel imaging device, fitting the initial point cloud data to obtain a second skeleton, and performing traversal analysis on the second skeleton to obtain second blood vessel features, the second blood vessel features being blood vessel topology; obtaining initial point cloud data of different parts of a target object collected by a blood vessel imaging device, performing point cloud splicing on the initial point cloud data to obtain second blood vessel features, the second blood vessel features being blood vessel point cloud data; or obtaining target point cloud data of a target object collected by a blood vessel imaging device, performing point cloud feature extraction on the target point cloud data to obtain second blood vessel features, the second blood vessel features being blood vessel point cloud features; or obtaining a two-dimensional image of a target object collected by a blood vessel imaging device, performing image feature extraction on the two-dimensional image to obtain second blood vessel features, the second blood vessel features being blood vessel image features.

5. The method of claim 1, wherein, The calculating, according to the first blood vessel features and the second blood vessel features, of the first registration relationship between the preoperative medical image and the intraoperative medical image comprises at least one of the following: performing registration on blood vessel topological structures corresponding to the preoperative medical image and blood vessel topological structures corresponding to the intraoperative medical image to obtain the first registration relationship; performing point cloud registration on blood vessel point cloud data corresponding to the preoperative medical image and blood vessel point cloud data corresponding to the intraoperative medical image to obtain the first registration relationship; or performing feature matching on point cloud features corresponding to the preoperative medical image and point cloud features corresponding to the intraoperative medical image to determine a first initial registration region from the preoperative medical image, and performing point cloud registration on blood vessel point cloud data in the first initial registration region and blood vessel point cloud data corresponding to the intraoperative medical image to obtain the first registration relationship; or performing feature matching on image features corresponding to the preoperative medical image and image features corresponding to the intraoperative medical image to determine a second initial registration region from the preoperative medical image, and performing point cloud registration on blood vessel point cloud data in the second initial registration region and blood vessel point cloud data corresponding to the intraoperative medical image to obtain the first registration relationship.

6. The method according to any one of claims 1 to 5, characterized in that, The generating, according to the first registration relationship and the second registration relationship, of operation navigation information comprises: obtaining a position of an intraoperative medical instrument in a world coordinate system; mapping a target object in the preoperative medical image to the world coordinate system according to the first registration relationship and the second registration relationship; generating operation navigation information according to the position of the target object in the world coordinate system and the position of the intraoperative medical instrument in the world coordinate system.

7. A medical image registration system, characterized by The system comprises: a blood vessel imaging device configured to collect an intraoperative medical image; a processor configured to perform the following steps: obtaining first blood vessel features corresponding to a preoperative medical image; obtaining an intraoperative medical image collected by the blood vessel imaging device and identifying second blood vessel features in the intraoperative medical image; calculating a first registration relationship between the preoperative medical image and the intraoperative medical image according to the first blood vessel features and the second blood vessel features; obtaining a second registration relationship between an image coordinate system of the blood vessel imaging device and a world coordinate system; generating operation navigation information according to the first registration relationship and the second registration relationship; The obtaining of the second registration relationship between the image coordinate system of the blood vessel imaging device and the world coordinate system comprises: obtaining a calibration board image collected by the blood vessel imaging device; The third registration relationship is obtained by registering the calibration board image with the real calibration board; The pose relationship between the real calibration board and the target corresponding to the real calibration board, the first conversion relationship between the target corresponding to the real calibration board and the coordinate system of the positioning device, and the second conversion relationship between the coordinate system of the positioning device and the target corresponding to the blood vessel imaging instrument are obtained; The second registration relationship between the image coordinate system of the blood vessel imaging instrument and the world coordinate system is determined according to the third registration relationship, the pose relationship, the first conversion relationship, and the second conversion relationship.

8. The system of claim 7, wherein, Before the processor performs the obtaining of the first blood vessel feature corresponding to the preoperative medical image, the following steps are included: The corresponding preoperative medical image is obtained, and the preoperative medical image is segmented to obtain a blood vessel mask; The blood vessel mask is reconstructed and classified to obtain a blood vessel surface model; The first blood vessel feature corresponding to the preoperative medical image is obtained according to the blood vessel surface model; the first blood vessel feature includes at least one of a blood vessel topological structure, blood vessel point cloud data, blood vessel point cloud features, and blood vessel image features.

9. The system of claim 8, wherein, The processor performs the obtaining of the first blood vessel feature corresponding to the preoperative medical image according to the blood vessel surface model, including at least one of the following: The first skeleton of the blood vessel surface model is extracted, and the blood vessel topological structure is obtained by traversing and analyzing the first skeleton; or The blood vessel point cloud data is obtained by performing isosurface extraction on the blood vessel surface model; or According to the blood vessel surface model and the blood vessel point cloud data, the space in which the points in the blood vessel point cloud data are located is divided, and the blood vessel point cloud features are obtained by performing point cloud feature extraction on the blood vessel point cloud data in each divided space; or According to the blood vessel surface model and the blood vessel point cloud data, the pose of the surface blood vessel is calculated; the blood vessel image is generated according to the pose of the surface blood vessel, and the blood vessel image features are obtained by performing image feature extraction on the blood vessel image.

10. The system of claim 7, wherein, The processor performs the obtaining of the intraoperative medical image collected by the blood vessel imaging instrument and the identification of the second blood vessel feature in the intraoperative medical image, including at least one of the following: The initial point cloud data of different parts of the target object collected by the blood vessel imaging instrument is obtained, the second skeleton is obtained by fitting the initial point cloud data, and the second blood vessel feature is obtained by traversing and analyzing the second skeleton; the second blood vessel feature is a blood vessel topological structure; The initial point cloud data of different parts of the target object collected by the blood vessel imaging instrument is obtained, and the second blood vessel feature is obtained by performing point cloud splicing on the initial point cloud data; the second blood vessel feature is blood vessel point cloud data; or The target point cloud data of the target object collected by the blood vessel imaging instrument is obtained, and the second blood vessel feature is obtained by performing point cloud feature extraction on the target point cloud data; the second blood vessel feature is blood vessel point cloud features; or The two-dimensional image of the target object collected by the blood vessel imaging instrument is obtained, and the second blood vessel feature is obtained by performing image feature extraction on the two-dimensional image; the second blood vessel feature is blood vessel image features.

11. The system of claim 7, wherein, The processor performs the calculation of the first registration relationship of the preoperative medical image and the intraoperative medical image according to the first blood vessel feature and the second blood vessel feature, including at least one of the following: The first registration relationship is obtained by registration of the blood vessel topology structure corresponding to the preoperative medical image and the blood vessel topology structure corresponding to the intraoperative medical image; The first registration relationship is obtained by point cloud registration of the blood vessel point cloud data corresponding to the preoperative medical image and the blood vessel point cloud data corresponding to the intraoperative medical image; or The first registration relationship is obtained by feature matching of the point cloud feature corresponding to the preoperative medical image and the point cloud feature corresponding to the intraoperative medical image, to determine a first initial registration area from the preoperative medical image, and by point cloud registration of the blood vessel point cloud data in the first initial registration area and the blood vessel point cloud data corresponding to the intraoperative medical image; or The first registration relationship is obtained by feature matching of the image feature corresponding to the preoperative medical image and the image feature corresponding to the intraoperative medical image, to determine a second initial registration area from the preoperative medical image, and by point cloud registration of the blood vessel point cloud data in the second initial registration area and the blood vessel point cloud data corresponding to the intraoperative medical image.

12. The system of any one of claims 7 to 11, wherein, The processor performs the generation of the operation navigation information according to the first registration relationship and the second registration relationship, including: Obtaining the position of the intraoperative medical instrument in the world coordinate system; Mapping the target object in the preoperative medical image to the world coordinate system according to the first registration relationship and the second registration relationship; Generating operation navigation information according to the position of the target object in the world coordinate system and the position of the intraoperative medical instrument in the world coordinate system.

13. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor performs the computer program to realize the steps of the method in any one of claims 1 to 6.

14. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 6.

Images