Catheter registration apparatus and method

Abstract

The application relates to a catheter registration device and method. The catheter comprises a positioning assembly and an optical coherence tomography (OCT) assembly. The catheter registration device comprises a collection module, a processing module and a registration module. The collection module is used for collecting first pose information of the catheter in a mirror insertion process through the positioning assembly of the catheter, and collecting local bronchial OCT images of the catheter in the mirror insertion process through the OCT assembly. The processing module is used for performing point cloud extraction and three-dimensional splicing on the local bronchial OCT images according to the relative position relationship between the positioning assembly and the OCT assembly and the first pose information, so as to obtain first bronchial surface point cloud data of the catheter in the mirror insertion process. The registration module is used for registering the first bronchial surface point cloud data and second bronchial surface point cloud data, and determining the mapping relationship between the first bronchial surface point cloud data and the second bronchial surface point cloud data. The device can improve the registration accuracy of the catheter.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot control, in particular to a catheter registration device and method. BACKGROUND

[0002] With the development of computer technology and medical imaging technology, in order to improve the accuracy and convenience of surgery, bronchoscope navigation system has been more and more widely used. Bronchial registration during the process of bronchoscope in surgery is a key technology. Bronchial registration is to establish the mapping relationship between the bronchus in the preoperative image and the bronchus of the patient in the real operating room, and to associate the preoperative planned bronchus with the real patient based on the mapping relationship, thereby assisting real-time navigation in surgery.

[0003] In related technologies, in order to ensure the timeliness of registration, three-dimensional data of the bronchus can be collected before surgery, and the inner cavity or inner surface point cloud data of the bronchus can be extracted. In the surgery, a magnetic sensor or an optical fiber can be arranged on the bronchoscope, and the bronchoscope can traverse in the bronchus, so that the magnetic sensor or the optical fiber collects a large amount of point cloud data in the bronchus cavity. Finally, the preoperative point cloud data and the intraoperative point cloud data are matched, so as to realize the registration of the bronchoscope.

[0004] However, due to the limitation of the catheter traversal operation of the bronchoscope in the bronchus, the point cloud data obtained in the surgery cannot accurately reflect the morphological structure of the bronchus cavity or surface, thereby resulting in low registration accuracy. SUMMARY

[0005] Therefore, it is necessary to provide a catheter registration device and method capable of improving the registration accuracy of the catheter in view of the above technical problems.

[0006] In a first aspect, the present application provides a catheter registration device. The catheter comprises a positioning assembly and an optical coherence tomography (OCT) assembly, and the device comprises:

[0007] An acquisition module is configured to acquire first pose information of the catheter in the process of inserting the catheter into the bronchus by the positioning assembly of the catheter, and acquire local bronchus OCT images of the catheter in the process of inserting the catheter into the bronchus by the OCT assembly. The local bronchus OCT images are tomographic images of the local bronchus generated by the OCT assembly through optical interference.

[0008] A processing module is configured to perform point cloud extraction and three-dimensional splicing on the local bronchus OCT images according to the relative position relationship between the positioning assembly and the OCT assembly and the first pose information, so as to obtain first bronchus surface point cloud data of the catheter in the process of inserting the catheter into the bronchus.

[0009] The registration module is used to register the first bronchial surface point cloud data and the second bronchial surface point cloud data to determine the mapping relationship between the first bronchial surface point cloud data and the second bronchial surface point cloud data. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion into the endoscope.

[0010] In one embodiment, the processing module is specifically configured to: determine the second pose information of the local bronchial image in the positioning space based on the relative positional relationship between the positioning component and the OCT component and the first pose information; fill the local bronchial OCT image into the global bronchial image based on the second pose information until the reconstruction cutoff condition of the global bronchial image is reached; determine the filling value of all cavities in the cavity filling area of ​​the global bronchial image that has reached the reconstruction cutoff condition; and fill the cavities based on the filling values ​​of all cavities to obtain a complete bronchial image.

[0011] In one embodiment, the processing module is specifically used to fill the voxels in the local bronchial OCT image into the global bronchial image according to the second pose information.

[0012] In one embodiment, the processing module is specifically configured to: determine the pose information of a first voxel in the space of the local bronchial OCT image, wherein the first voxel is any voxel in the local bronchial OCT image; determine the pose information of the first voxel in the positioning space based on the second pose information and the pose information of the first voxel in the space of the local bronchial OCT image; determine the target index of the first voxel in the space of the global bronchial image based on the pose information of the first voxel in the positioning space; determine the voxel value of the second voxel mapped in the global bronchial image based on the voxel value of the first voxel and the target index; and fill the global bronchial image with the second voxel based on the voxel value of the second voxel.

[0013] In one embodiment, the processing module is further configured to identify the global bronchial image that meets the reconstruction cutoff condition as a search cube; perform multiple cavity searches on the search cube until the number of cavity searches reaches a threshold; and merge the result cubes identified by the last cavity search to obtain the cavity filling area.

[0014] In one embodiment, the hole search process includes:

[0015] The search cube is divided into multiple sub-cubes;

[0016] The result cube in the search cube is determined based on the number of non-hollow voxels in each sub-cube;

[0017] Update the search cube based on the result cube.

[0018] In one embodiment, the processing module is specifically configured to determine the fill value of all the cavities based on the values ​​of the non-cavities adjacent to each cavity.

[0019] In one embodiment, the reconstruction cutoff condition includes the cavity ratio of the global bronchial image or the number of times the global bronchial image is filled.

[0020] In one embodiment, the processing module is specifically configured to traverse the complete bronchial image in cells of a preset size, determine the vertex index of the isosurface based on the tomographic value of the vertex of each cell and a preset tomographic threshold of the isosurface; obtain the coordinates of the vertex of the isosurface in the positioning space based on the vertex index of the isosurface of each cell, the second pose information, and the position of the image origin in the positioning space; and determine the first bronchial surface point cloud data based on the coordinates of the vertex of the isosurface in the positioning space.

[0021] In one embodiment, the processing module is further configured to acquire third pose data of the calibration object in the positioning space and fourth pose data of the positioning component in the positioning space, wherein the calibration object and the conduit are fixed in the positioning space; determine the relative positional relationship between the calibration object and the positioning component based on the third pose data and the fourth pose data; extract fifth pose data of the calibration object in the image coordinate system from the three-dimensional image; and determine the relative positional relationship between the positioning component and the OCT component based on the relative positional relationship between the calibration object and the positioning component and the fifth pose data.

[0022] In one embodiment, the positioning component and the OCT component are disposed within the catheter; or, the catheter is provided with a working channel, the OCT component is disposed within the positioning component, and the OCT component extends into the catheter through the working channel and moves synchronously with the insertion of the catheter.

[0023] Secondly, this application provides a catheter registration method, wherein the catheter includes a positioning component and an optical coherence tomography (OCT) component, the method comprising:

[0024] The positioning component acquires the first position information of the catheter during the insertion process, and the OCT component acquires local bronchial OCT images of the catheter during the insertion process. The local bronchial OCT images are tomographic images of the local bronchus generated by the OCT component through optical interference.

[0025] Based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud extraction and three-dimensional stitching are performed on the local bronchial OCT image to obtain the first bronchial surface point cloud data of the catheter during the insertion process.

[0026] The first bronchial surface point cloud data and the second bronchial surface point cloud data are registered to determine the mapping relationship between the first bronchial surface point cloud data and the second bronchial surface point cloud data. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion of the endoscope.

[0027] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the catheter registration method of the first aspect described above.

[0028] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the catheter registration method of the first aspect described above.

[0029] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the catheter registration method of the first aspect described above.

[0030] The aforementioned catheter registration device and method, wherein the catheter includes a positioning component and an optical coherence tomography (OCT) component, the catheter registration device includes an acquisition module, a processing module, and a registration module. The acquisition module is used to acquire the first pose information of the catheter during endoscope insertion via the positioning component, and to acquire local bronchial OCT images of the catheter during endoscope insertion via the OCT component. The local bronchial OCT images are tomographic images of the local bronchus generated by the OCT component through optical interference. The processing module is used to extract point clouds and perform 3D stitching on the local bronchial OCT images based on the relative positional relationship between the positioning component and the OCT component and the first pose information, to obtain the first bronchial surface point cloud data during endoscope insertion. The registration module is used to register the first bronchial surface point cloud data and the second bronchial surface point cloud data, determining the mapping relationship between them. The second bronchial surface point cloud data is point cloud data generated by the catheter based on the 3D image of the target object before endoscope insertion. By extracting point clouds and stitching three dimensions from local bronchial OCT images based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud data of the first bronchial surface during catheter insertion is obtained. This allows the first bronchial surface point cloud data to fully reflect the actual morphological structure of the bronchus. Catheter matching based on the first bronchial surface point cloud data can improve bronchial registration accuracy without significantly reducing registration time performance. Attached Figure Description

[0031] Figure 1 An application environment diagram of a catheter registration method provided in this application embodiment;

[0032] Figure 2 A schematic flowchart illustrating a catheter registration method provided in an embodiment of this application;

[0033] Figure 3 This is a schematic diagram of the structural composition of a catheter provided in an embodiment of this application;

[0034] Figure 4 This is a schematic diagram of the structure of a calibration object provided in an embodiment of this application;

[0035] Figure 5 This is a schematic diagram illustrating the extraction of point cloud data from the surface of a bronchus, provided in an embodiment of this application.

[0036] Figure 6 A schematic diagram illustrating the filling of a local bronchial OCT image into a global bronchial image, as provided in an embodiment of this application;

[0037] Figure 7 A schematic diagram of a three-line interpolation method provided in an embodiment of this application;

[0038] Figure 8 A schematic diagram of a search cavity filling area provided in an embodiment of this application;

[0039] Figure 9 A schematic diagram illustrating a cavity filling method provided in an embodiment of this application;

[0040] Figure 10 A registration diagram provided for an embodiment of this application;

[0041] Figure 11 A schematic flowchart of another catheter registration method provided in an embodiment of this application;

[0042] Figure 12 A schematic flowchart illustrating another catheter registration method provided in an embodiment of this application;

[0043] Figure 13 A schematic flowchart illustrating another catheter registration method provided in this application embodiment;

[0044] Figure 14 A structural block diagram of a catheter registration device provided in an embodiment of this application;

[0045] Figure 15 This is an internal structural diagram of a computer device provided in an embodiment of this application. Detailed Implementation

[0046] 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.

[0047] The catheter registration method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown.

[0048] like Figure 1As shown, the bronchoscope 101 and the terminal device 102 are connected. The bronchoscope 101 includes a positioning component and an optical coherence tomography (OCT) component. During bronchoscope insertion, the terminal device 102 can acquire the first pose information of the bronchoscope 101 during insertion via the positioning component, and acquire local bronchial OCT images via the OCT component. The local bronchial OCT images are tomographic images of the local bronchus generated by the OCT component through optical interference. Subsequently, based on the relative positional relationship between the positioning component and the OCT component and the first pose information, the terminal device 101 performs point cloud extraction and 3D stitching on the local bronchial OCT images to obtain the first bronchial surface point cloud data during catheter insertion. Finally, the terminal device 101 registers the first bronchial surface point cloud data and the second bronchial surface point cloud data to determine the mapping relationship between the first bronchial surface point cloud data and the second bronchial surface point cloud data. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion into the endoscope.

[0049] In some embodiments, before surgery, a three-dimensional image of the human chest can be obtained by scanning with optical coherence tomography (OCT), computed tomography (CT), or nuclear magnetic resonance (MR) equipment, and then point cloud data can be acquired from the three-dimensional image to obtain point cloud data of the second bronchus surface.

[0050] In one embodiment, such as Figure 2 As shown, a catheter registration method is provided, which can be applied to... Figure 1 The following explanation will be based on the terminal equipment in the example, including S201-S203:

[0051] S201. Acquire the first position information of the catheter during the insertion process through the positioning component, and acquire local bronchial OCT images of the catheter during the insertion process through the OCT component. The local bronchial OCT images are tomographic images of the local bronchus generated by the OCT component through optical interference.

[0052] In this application, during bronchoscopy, the bronchoscope catheter moves within the bronchi of the human body. At this time, the terminal device can acquire the first position information of the catheter during the insertion process through the positioning component on the catheter, and acquire local bronchial OCT images of the catheter during the insertion process through the OCT component.

[0053] It should be understood that the embodiments of this application do not limit the positioning components. In some embodiments, the positioning components include magnetic positioning components or fiber optic positioning components.

[0054] In some embodiments, the positioning component and the OCT component are disposed within the conduit. For example, Figure 3 This is a schematic diagram of the structural composition of a catheter provided in an embodiment of this application, as shown below. Figure 3 As shown, an OCT component and a positioning component can be set inside the positioning component, so that the catheter, the OCT component and the positioning component can be combined to form a bronchial catheter with a positioning OCT.

[0055] In other embodiments, a working channel is provided inside the catheter, and a positioning component is provided inside the OCT assembly. The OCT assembly extends into the catheter through the working channel and moves synchronously with the catheter during endoscope insertion. That is, the positioning component is fixedly disposed inside the OCT assembly, and the OCT assembly can extend into the catheter through the working channel and be fixed to the catheter, so that the positioning component and the OCT assembly can move synchronously with the catheter during endoscope insertion.

[0056] It should be understood that the aforementioned first-position information and local bronchial OCT images can be acquired in real time according to the settings, and the first-position information and local bronchial OCT images acquired at the same time correspond one-to-one. For example, they can be acquired once at preset time intervals, such as once every 1 second; or they can be acquired once after advancing a certain distance, such as once every 0.1 cm advancement of the catheter.

[0057] S202. Based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud extraction and three-dimensional stitching are performed on the local bronchial OCT image to obtain the point cloud data of the first bronchial surface during the catheter insertion process.

[0058] The relative positional relationship between the positioning component and the OCT component can be predetermined before the catheter is inserted into the endoscope. In some embodiments, the terminal device can first acquire the third pose data of the calibration object in the positioning space and the fourth pose data of the positioning component in the positioning space, with the calibration object and the catheter fixed in the positioning space. Subsequently, the terminal device determines the relative positional relationship between the calibration object and the positioning component based on the third and fourth pose data. Finally, the terminal device extracts the fifth pose data of the calibration object in the image coordinate system from the 3D image, and determines the relative positional relationship between the positioning component and the OCT component based on the relative positional relationship between the calibration object and the positioning component and the fifth pose data.

[0059] The aforementioned positioning space can be the coordinate system in which the positioning component is located. Alternatively, it can be a spatial coordinate system derived from the coordinate system in which the positioning component is located.

[0060] The calibration object can be a calibration phantom containing measurable feature structures. These feature structures can be detected by the OCT component and measured in local bronchial OCT images. Simultaneously, these feature structures can be measured within the positioning space of the positioning component. When determining the relative positional relationship between the positioning component and the OCT component, the calibration object can be fixed within the positioning space of the positioning component.

[0061] For example, after the calibration object is fixed in the positioning space of the positioning component, the terminal device can collect the third pose data M of the feature structure in the calibration object in the positioning space of the positioning component. ref And collect the fourth pose data M of the positioning component in the positioning space. s Subsequently, the terminal device can calculate the relative positional relationship M between the calibration object and the positioning component using formula (1). ref→s .

[0062]

[0063] For example, after calculating the relative positional relationship M between the calibration object and the positioning component... ref→s Afterwards, three-dimensional images can be acquired, and the fifth pose M of the feature structure in the calibration object in the coordinate system of the three-dimensional image can be extracted and calculated. ref→img Subsequently, the relative pose relationship M between the three-dimensional image coordinate system and the pose of the positioning sensor was calculated using formula (2). img→s The relative positional relationship between the positioning component and the OCT component.

[0064]

[0065] In some embodiments, to eliminate errors, the relative pose relationship M between multiple three-dimensional image coordinate systems and the pose of the positioning sensor can be calculated. img→s Then, using the error elimination method, the final optimal M is calculated. img→s This error elimination method can include the averaging method.

[0066] For example, Figure 4 This is a schematic diagram of the structure of a calibration object provided in an embodiment of this application, such as... Figure 4As shown, the characteristic structure of the calibration object consists of two mutually perpendicular cylinders. The central axis of each cylinder serves as a coordinate axis of the characteristic structure, and the intersection of the central axes of the two cylinders represents the position of the characteristic structure. During the calibration process of the relative positional relationship between the positioning component and the OCT component, the coordinate system of the positioning space, the bronchial tube, and the calibration object maintain a fixed relationship. After each calibration of the relative positional relationship between the positioning component and the OCT component is completed, the fixed relationship between the coordinate system of the positioning space, the bronchial tube, and the calibration object can be changed before the next calibration.

[0067] In some embodiments, the terminal device can first determine the second pose information of the local bronchus OCT image in the positioning space based on the relative positional relationship between the positioning component and the OCT component and the first pose information. Subsequently, based on the second pose information, point cloud extraction and 3D stitching are performed on the local bronchus OCT image to obtain the first bronchus surface point cloud data.

[0068] For example, the second pose information of local bronchial OCT images in the positioning space It can be calculated using formula (3).

[0069]

[0070] in, Let r be the pose of the positioning component in the positioning space, and r be used to characterize the positioning space.

[0071] It should be understood that there are no restrictions on how the embodiments of this application perform point cloud extraction and three-dimensional stitching on local bronchial OCT images to obtain the first bronchial surface point cloud data. Point cloud extraction can be performed first and then three-dimensional stitching can be performed, or three-dimensional stitching can be performed first and then point cloud extraction can be performed.

[0072] In some embodiments, the terminal device may first extract point clouds from the local bronchial OCT image to obtain local bronchial surface point cloud data. Subsequently, the terminal device may perform three-dimensional stitching of the local bronchial surface point cloud data according to the second pose information to obtain first bronchial surface point cloud data.

[0073] In other embodiments, the terminal device can first perform three-dimensional stitching on the local bronchial OCT image based on the second pose information to obtain a complete bronchial image. Subsequently, point cloud extraction is performed on the complete bronchial image to obtain the first bronchial surface point cloud data.

[0074] For example, Figure 5 This is a schematic diagram illustrating the extraction of point cloud data from the bronchial surface, provided as an embodiment of this application. Figure 5As shown, the local bronchial 3D OCT images can be stitched together and reconstructed into a complete bronchial OCT image. Then, intraoperative bronchial data can be extracted from the complete bronchial OCT image to obtain the aforementioned first bronchial surface point cloud data.

[0075] It should be understood that the embodiments of this application do not limit how to perform three-dimensional stitching of local bronchial OCT images based on the second pose information. In some embodiments, the terminal device may first fill the local bronchial OCT image into the global bronchial image based on the second pose information until the reconstruction cutoff condition of the global bronchial image is reached. Subsequently, the filling values ​​of all cavities in the cavity filling area of ​​the global bronchial image that has reached the reconstruction cutoff condition are determined. Finally, cavity filling is performed based on the filling values ​​of all cavities to obtain a complete bronchial image.

[0076] The aforementioned reconstruction cutoff conditions include the cavity ratio of the global bronchial image or the number of times the global bronchial image is filled. For example, the number of fillings can be 10 times, 20 times, etc.; the cavity ratio of the global bronchial image can be 90%, 95%, etc. A cavity can be understood as all pixels in the bronchial image without initial values, which are no longer cavities after being filled.

[0077] In some embodiments, the terminal device can fill local bronchial OCT images based on the pose change amplitude of the second pose information. For example, when the pose change amplitude of the second pose information exceeds a threshold, local bronchial OCT images can be acquired and filled. The threshold may include position, direction, angle, etc.

[0078] In some embodiments, the local bronchial OCT images may be preprocessed before filling them. For example, the preprocessing of the local bronchial OCT images includes noise reduction.

[0079] It should be understood that the embodiments of this application do not limit how to fill local bronchial OCT images. In some embodiments, the filling method includes nearest neighbor value, linear interpolation, spline interpolation, and hybrid methods.

[0080] In some embodiments, when the filling threshold is reached, the cavity-filling region in the global bronchial image can be calculated. The calculation method includes octree calculation, etc. The cavity-filling region can be the entire target image.

[0081] In some embodiments, within the cavity filling area, all cavities in the global bronchial image can be traversed, and the cavity filling value can be calculated and filled based on the values ​​of all its neighboring non-cavities. The calculation method may include the nearest neighbor method, interpolation method, etc.

[0082] In some embodiments, before filling the global bronchial image with the local bronchial OCT image, the volume attributes of the global bronchial image can be set or calculated to facilitate the filling of the bronchial image. These volume attributes include position, orientation, size, resolution, and dimensions, and are defined in the positioning space.

[0083] It should be understood that the embodiments of this application do not limit how the volume properties of the global bronchial image are set or calculated. In some embodiments, the volume properties of the global bronchial image can be calculated based on the three-dimensional image before endoscopic examination.

[0084] For example, the bronchial attributes of the 3D image before intubation can be extracted first, and the maximum bounding box of the bronchial point cloud coverage area can be calculated based on the bronchial attributes of the 3D image before intubation to determine the size of the global bronchial image, that is, the length, width and height of the maximum bounding box. Subsequently, the resolution of the global bronchial image can be calculated. The dimensionality of the global bronchial image was determined by the ratio of its size to its resolution. Finally, the location of the global bronchial image is calculated in the positioning space using the positioning component. and posture This allows us to obtain the location of the global bronchial image in the localization space. As shown in formula (4):

[0085]

[0086] In other embodiments, the terminal device may set the pose of the initial global bronchial image to be the same as the pose of the first local bronchial OCT image. During the acquisition of the local bronchial OCT image, the position of the global bronchial image is calculated and updated in real time. The calculation method may include Simultaneous Localization and Mapping (SLAM).

[0087] In some embodiments, the terminal device may fill voxels from local bronchial OCT images into global bronchial images based on the second pose information.

[0088] A voxel, also known as a volume pixel, is a unit of measurement in 3D space. A solid containing voxels can be represented through stereo rendering or by extracting polygonal isosurfaces from a given threshold contour. A voxel is the smallest unit of 3D spatial segmentation and is commonly used in fields such as 3D imaging, scientific data, and medical imaging.

[0089] For example, Figure 6This is a schematic diagram illustrating the process of filling a local bronchial OCT image into a global bronchial image, as provided in an embodiment of this application. Figure 6 As shown, after acquiring local bronchial OCT images, the second pose information of the local bronchial OCT images in the positioning space can be calculated, and the local bronchial OCT images are preprocessed. Subsequently, the voxels of the local bronchial OCT images are mapped onto the global bronchial images, and the corresponding areas of the global bronchial images are filled. This process of acquiring and filling local bronchial OCT images continues until the cutoff condition is reached.

[0090] It should be understood that the embodiments of this application do not limit how voxels from local bronchial OCT images are filled into global bronchial images. In some embodiments, the terminal device may first determine the pose information of a first voxel in the space of the local bronchial OCT image, where the first voxel is any voxel in the local bronchial OCT image. Secondly, the terminal device determines the pose information of the first voxel in the positioning space based on the second pose information and the pose information of the first voxel in the space of the local bronchial OCT image. Thirdly, the terminal device determines the target index of the first voxel in the space of the global bronchial image based on the pose information of the first voxel in the positioning space. Finally, the terminal device determines the voxel value of the second voxel mapped by the first voxel in the global bronchial image based on the voxel value of the first voxel and the target index, and fills the global bronchial image with the second voxel based on the voxel value of the second voxel.

[0091] For example, the terminal device can traverse each voxel in the local bronchial OCT image. For the first voxel traversed, the pose information of the first voxel in the space of the local bronchial OCT image can be calculated by formula (5), and the pose information of the first voxel in the positioning space can be calculated by formula (6).

[0092]

[0093]

[0094] Where i, j, and k are the indices of the first voxel. This represents the resolution of a local bronchial OCT image. This provides the pose information of local bronchial OCT images in the positioning space.

[0095] For example, the terminal device can map the first voxel to the global bronchial image using formula (7) to obtain the pose information of the first voxel in the global bronchial image, and then obtain the target index i of the first voxel in the space of the global bronchial image using formulas (8)-(10). t j t k tSubsequently, the voxel value of the first voxel and the global bronchial image were combined at target index i. t j t k t The new voxel value is calculated at the voxel value at the target index i, and the global bronchial image is updated. t j t k t The voxel at the location.

[0096]

[0097]

[0098]

[0099]

[0100] in, This refers to the pose information of the global bronchial image in the positioning space.

[0101] The following explains how to calculate the voxel value at the target index for global bronchial images.

[0102] It should be understood that global bronchial imaging in target index i t j t k t The location of the voxel at a given location in the spatial dimension of a global bronchial image is usually not equivalent to the mapping location of the local bronchial OCT image in the spatial dimension of the global bronchial image. t y t z t .

[0103] In some embodiments, candidate voxel values ​​can be calculated first, then the voxel value at the target index can be obtained, and finally the candidate voxel values ​​and the voxel value at the target index can be mixed to obtain the final voxel value.

[0104] For example, candidate voxel value V c The nearest neighbor method can be used, that is, the value can be determined by the corresponding voxel value of the local bronchial OCT image.

[0105] For example, candidate voxel value V c Alternatively, interpolation methods (such as three-line interpolation) can be used. Figure 7 This is a schematic diagram of a three-line interpolation method provided in an embodiment of this application. Figure 7 As shown, the global bronchial image at target index i can be calculated first. t j t k t The voxel position P(P) x P yP z ). Then according to x t y t z t and P x P y P z The positional relationship is determined by identifying the eight nearest voxels in local bronchial OCT images, including C... 000 C 001 C 011 C 010 C 100 C 101 C 111 C 110 Subsequently, based on the voxel position P and its eight nearest voxels, the weights of the eight nearest voxels relative to the voxel value of P are calculated. Finally, the voxel value V of P is calculated based on the values ​​of the eight nearest voxels and their weights relative to P. c , which is the aforementioned candidate voxel value.

[0106] It should be understood that the embodiments of this application do not impose any restrictions on how the candidate voxel values ​​and the voxel values ​​at the target index are mixed.

[0107] For example, a mixing coefficient A for candidate voxel values ​​can be set. c Mixing coefficient A with the current voxel value src And A c +A src =1. If A src If the value is 0, then the candidate voxel value is the final voxel value. Obtain the target index i. t j t k t If the voxel value at a given location is a hole, then the candidate voxel value is the final voxel value. If it is not a hole, then the final voxel value V is calculated using formula (11). d .

[0108] V d =V c ×A c +V src ×A src (11)

[0109] The following section explains how to search for and fill voids.

[0110] It should be understood that using local bronchial OCT images to fill global bronchial images may result in residual cavities, requiring further filling. Cavity filling can be performed on the entire global bronchial image. Due to the anatomical characteristics of the bronchi, only a small portion of the global bronchial image is an effective area containing the bronchi; therefore, obtaining the minimum area surrounding the bronchi can greatly improve the efficiency of cavity filling.

[0111] It should be understood that the terminal device can calculate the cavity-filling region of the global bronchial image using an octree. In some embodiments, the terminal device can first identify the global bronchial image that meets the reconstruction cutoff criteria as the search cube. Subsequently, the terminal device can perform multiple cavity searches on the search cube until the number of cavity searches reaches a threshold. Finally, the terminal device can merge the result cubes determined by the last cavity search to obtain the cavity-filling region.

[0112] The hole search process includes: dividing the search cube into multiple sub-cubes, then determining the result cube within the search cube based on the number of non-hole voxels in each sub-cube. Finally, updating the search cube based on the result cube.

[0113] The number of attempts threshold can be set according to the specific circumstances, such as 3 or 4. The number of sub-cubes into which the search cube is divided can also be set according to the specific circumstances, such as 8.

[0114] Figure 8 This is a schematic diagram illustrating a method for searching and filling voids, as provided in an embodiment of this application. Figure 8 As shown, a maximum number of searches, N, can be set, where N is a non-zero integer, and the global bronchial image is set as the initial search cube. Subsequently, multiple cavity searches are performed until the maximum number of searches, N, is reached. During each search, each search cube can be divided into 8 equal sub-cubes. For each sub-cube, the number of non-cavitary voxels is calculated. If the number of non-cavitary voxels is not zero, it is marked as a result cube; otherwise, it is discarded. Finally, all result cubes from the Nth search are merged, and the result is the cavity-filling region.

[0115] Figure 9 This is a schematic diagram illustrating a cavity filling method provided in an embodiment of this application. Figure 9 As shown, all cavities within the cavity filling area can be traversed. Then, based on the values ​​of all its neighboring non-cavity values, the cavity filling value is calculated and used to fill the cavity. Finally, all cavities within the cavity filling area are filled, resulting in a complete bronchial image.

[0116] In some embodiments, the fill value for all cavities can be determined based on the value of the non-cavities adjacent to each cavity.

[0117] For example, for each void, 13 directed rods and a maximum search length are defined. For each directed rod, non-voids are searched along both directions of the rod. If a non-void is found in both directions, a valid rod is obtained, and the voxel values ​​of the two non-voids are V0 and V1 respectively. i1 and V i2The distances from the two non-voids to the current void are L and L respectively. i1 and L i2 The length L of the rod i =L i1 +L i2 Density of the rod If the maximum search length is reached and no two non-hole bars are found, the bar is discarded. For all valid bars, the hole-filling value can be calculated. Where N is the number of effective bars.

[0118] It should be understood that the embodiments of this application do not limit how to extract point clouds from complete bronchial images to obtain first bronchial surface point cloud data. In some embodiments, a marching cubes algorithm can be used for extraction. For example, the complete bronchial image can first be traversed using cells of a preset size. Based on the tomographic value of each cell's vertex and a preset tomographic threshold for the isosurface, the vertex index of the isosurface is determined. Then, based on the vertex index of the isosurface in each cell, the second pose information, and the position of the image origin in the positioning space, the coordinates of the isosurface vertices in the positioning space are obtained. Finally, based on the coordinates of the isosurface vertices in the positioning space, the first bronchial surface point cloud data is determined.

[0119] The size of the cell can be set according to the actual situation; for example, it can be set to 2*2*2.

[0120] 203. Register the first bronchial surface point cloud data and the second bronchial surface point cloud data to determine the mapping relationship between the first bronchial surface point cloud data and the second bronchial surface point cloud data. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion into the endoscope.

[0121] In this step, after the terminal device extracts point clouds and stitches three dimensions of the local bronchial OCT image to obtain the point cloud data of the first bronchial surface during the insertion of the catheter, the point cloud data of the first bronchial surface and the point cloud data of the second bronchial surface can be registered.

[0122] The target object is the bronchus to be examined by the endoscope.

[0123] The aforementioned second bronchial surface point cloud data can be extracted from the target object's 3D image before endoscopy or surgery. Extraction methods can include marching cubes algorithms, image-based meshing algorithms, etc. For example, the marching cubes algorithm can be used, similar to the process of extracting point clouds from a complete bronchial image.

[0124] For example, an isosurface CT threshold can be set, which lies between the air CT value and the bronchial wall tissue CT value. Then, the 3D image of the target object is traversed in 2x2x2 cells. For each traversal, isosurfaces can be calculated based on the relationship between the CT values ​​of the cell's eight vertices and the isosurface CT threshold, resulting in N isosurfaces and M isosurface vertex indices i, j, and k. For isosurface points, the coordinates of the isosurface vertices in the positioning space can be calculated, and then the first bronchial surface point cloud data is acquired.

[0125] For example, if the resolution of the 3D image of the target object is E x R y R z The position of the target object's 3D image at the origin in human body space is The pose of the 3D image of the target object is (D x D y D z If the coordinates of the vertex of the isosurface in the positioning space can be determined by formula (12).

[0126]

[0127] Figure 10 This is a registration diagram provided for an embodiment of this application. Figure 10 As shown, a registration algorithm can be used to register the point cloud data of the first bronchus surface and the second bronchus surface to obtain the mapping relationship between them. This mapping relationship can be represented by a mapping matrix M.

[0128] It should be understood that the registration algorithm in this application is not limited and may include Iterative Closest Point (ICP) algorithm, Robust Point Matching (RPM) algorithm, and Coherent Point Drift (CPD) algorithm.

[0129] The catheter registration method provided in this application combines an OCT component with a bronchial catheter. Based on the OCT component, high-quality bronchial tissue information is acquired, such as the complete bronchial lumen and inner wall, thus accurately and completely reflecting the actual morphology and structure of the bronchus and forming dense, accurate, and complete first bronchial surface point cloud data. Registration is then performed between this first bronchial surface point cloud data and the corresponding second bronchial surface point cloud data from preoperative images, thereby improving bronchial registration accuracy without significantly reducing registration time.

[0130] Meanwhile, the aforementioned OCT component can be an OCT component based on the imaging characteristics of bronchial OCT. It eliminates the need for the bronchial tube to repeatedly traverse the bronchus for extended periods. A large amount of data can be collected through simple unidirectional advancement, greatly reducing the difficulty of operation, shortening the intraoperative time, reducing the patient's anesthesia risk, and also reducing the risk of bronchial injury.

[0131] The catheter registration method provided in this application first acquires the first pose information of the catheter during endoscope insertion using a positioning component, and then acquires local bronchial OCT images of the catheter during insertion using an OCT component. Subsequently, based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud extraction and 3D stitching are performed on the local bronchial OCT images to obtain the first bronchial surface point cloud data during catheter insertion. Finally, the first bronchial surface point cloud data and the second bronchial surface point cloud data are registered to determine the mapping relationship between them. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the 3D image of the target object before insertion. By extracting point clouds and stitching three dimensions from local bronchial OCT images based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud data of the first bronchial surface during catheter insertion is obtained. This allows the first bronchial surface point cloud data to fully reflect the actual morphological structure of the bronchus. Catheter matching based on the first bronchial surface point cloud data can improve bronchial registration accuracy without significantly reducing registration time performance.

[0132] The calibration process for the relative positional relationship between the positioning component and the OCT component is explained below. Figure 11 A schematic flowchart of another catheter registration method provided in this application embodiment is shown below. Figure 11 As shown, this catheter registration method includes S301-S307:

[0133] S301. Obtain the third pose data of the calibration object in the positioning space and the fourth pose data of the positioning component in the positioning space.

[0134] The calibration object and the catheter are fixed in the positioning space.

[0135] The aforementioned positioning space can be the coordinate system in which the positioning component is located, or it can be obtained by transforming the coordinate system in which the positioning component is located.

[0136] In some embodiments, the calibration object can be a calibration phantom containing measurable feature structures. These feature structures can be detected by the OCT component and measured in local bronchial OCT images. Simultaneously, these feature structures can be measured within the positioning space of the positioning component. When determining the relative positional relationship between the positioning component and the OCT component, the calibration object can be fixed within the positioning space of the positioning component.

[0137] S302. Based on the third pose data and the fourth pose data, determine the relative positional relationship between the calibration object and the positioning component.

[0138] S303. Extract the fifth pose data of the calibration object in the image coordinate system from the 3D image.

[0139] S304. Determine the relative positional relationship between the positioning component and the OCT component based on the relative positional relationship between the calibration object and the positioning component and the fifth pose data.

[0140] In some embodiments, to eliminate errors, the relative pose relationships between multiple 3D image coordinate systems and the pose of the positioning sensor can be calculated, and then an error elimination method can be used to calculate the final optimal value. This error elimination method may include the averaging method.

[0141] S305. Acquire the first position information of the catheter during the insertion process using the positioning component, and acquire local bronchial OCT images of the catheter during the insertion process using the OCT component.

[0142] S306. Based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud extraction and three-dimensional stitching are performed on the local bronchial OCT image to obtain the point cloud data of the first bronchial surface during the catheter insertion process.

[0143] S307. Register the point cloud data of the first bronchus surface and the point cloud data of the second bronchus surface to determine the mapping relationship between the point cloud data of the first bronchus surface and the point cloud data of the second bronchus surface.

[0144] Among them, the point cloud data of the second bronchus surface is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion of the endoscope.

[0145] In this embodiment, by using the relative positional relationship between the positioning component and the OCT component, point cloud extraction and three-dimensional stitching of local bronchial OCT images can be performed based on the first pose information, thereby simplifying the extraction process of point cloud data of the first bronchial surface during catheter insertion and accelerating the registration speed.

[0146] The following explains how to extract point clouds and stitch 3D images of local bronchial OCT images. Figure 12 A schematic flowchart of another catheter registration method provided in the embodiments of this application is shown below. Figure 12 As shown, this catheter registration method involves 3D stitching followed by point cloud extraction, including S401-S405:

[0147] S401. Acquire the first position information of the catheter during the insertion process using the positioning component, and acquire local bronchial OCT images of the catheter during the insertion process using the OCT component.

[0148] S402. Based on the relative positional relationship between the positioning component and the OCT component and the first pose information, determine the second pose information of the local bronchial OCT image in the positioning space.

[0149] S403. Based on the second pose information, perform three-dimensional stitching on the local bronchial OCT images to obtain a complete bronchial image.

[0150] S404. Extract point cloud data from the complete bronchial image to obtain the first bronchial surface point cloud data.

[0151] S405. Register the point cloud data of the first bronchus surface and the point cloud data of the second bronchus surface to determine the mapping relationship between the point cloud data of the first bronchus surface and the point cloud data of the second bronchus surface.

[0152] Among them, the point cloud data of the second bronchus surface is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion of the endoscope.

[0153] Figure 13 A schematic flowchart of another catheter registration method provided in the embodiments of this application is shown below. Figure 13 As shown, this catheter registration method involves point cloud extraction followed by 3D stitching, including S501-S505:

[0154] S501. Acquire the first position information of the catheter during the insertion process using the positioning component, and acquire local bronchial OCT images of the catheter during the insertion process using the OCT component.

[0155] S502. Based on the relative positional relationship between the positioning component and the OCT component and the first pose information, determine the second pose information of the local bronchial OCT image in the positioning space.

[0156] S503. Point cloud extraction is performed on the local bronchial OCT image to obtain local bronchial surface point cloud data.

[0157] S504. Based on the second pose information, perform three-dimensional stitching on the local bronchial surface point cloud data to obtain the first bronchial surface point cloud data.

[0158] S505. Register the point cloud data of the first bronchus surface and the point cloud data of the second bronchus surface to determine the mapping relationship between the point cloud data of the first bronchus surface and the point cloud data of the second bronchus surface.

[0159] Among them, the point cloud data of the second bronchus surface is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion of the endoscope.

[0160] The catheter registration method provided in this application first acquires the first pose information of the catheter during endoscope insertion using a positioning component, and then acquires local bronchial OCT images of the catheter during insertion using an OCT component. Subsequently, based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud extraction and 3D stitching are performed on the local bronchial OCT images to obtain the first bronchial surface point cloud data during catheter insertion. Finally, the first bronchial surface point cloud data and the second bronchial surface point cloud data are registered to determine the mapping relationship between them. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the 3D image of the target object before insertion. By extracting point clouds and stitching three dimensions from local bronchial OCT images based on the relative positional relationship between the positioning component and the OCT component and the first pose information, point cloud data of the first bronchial surface during catheter insertion is obtained. This allows the first bronchial surface point cloud data to fully reflect the actual morphological structure of the bronchus. Catheter matching based on the first bronchial surface point cloud data can improve bronchial registration accuracy without significantly reducing registration time performance.

[0161] 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.

[0162] Based on the same inventive concept, this application also provides a catheter registration device for implementing the catheter registration method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more catheter registration device embodiments provided below can be found in the limitations of the catheter registration method described above, and will not be repeated here.

[0163] In one embodiment, such as Figure 14 As shown, a catheter registration device 600 is provided, including: a data acquisition module 601, a processing module 602, and a registration module 603, wherein:

[0164] The acquisition module 601 is used to acquire the first position information of the catheter during the insertion process through the catheter positioning component, and to acquire local bronchial OCT images of the catheter during the insertion process through the OCT component. The local bronchial OCT images are tomographic images of the local bronchus generated by the OCT component through optical interference.

[0165] The processing module 602 is used to extract point clouds and stitch three-dimensionally on the local bronchial OCT image based on the relative positional relationship between the positioning component and the OCT component and the first pose information, so as to obtain the point cloud data of the first bronchial surface during the insertion of the catheter.

[0166] The registration module 603 is used to register the first bronchial surface point cloud data and the second bronchial surface point cloud data to determine the mapping relationship between the first bronchial surface point cloud data and the second bronchial surface point cloud data. The second bronchial surface point cloud data is the point cloud data generated by the catheter based on the three-dimensional image of the target object before insertion into the endoscope.

[0167] In one embodiment, the processing module 602 is specifically configured to: determine the second pose information of the local bronchial image in the positioning space based on the relative positional relationship between the positioning component and the OCT component and the first pose information; fill the local bronchial OCT image into the global bronchial image based on the second pose information until the reconstruction cutoff condition of the global bronchial image is reached; determine the filling value of all cavities in the cavity filling area of ​​the global bronchial image that has reached the reconstruction cutoff condition; and fill the cavities based on the filling values ​​of all cavities to obtain a complete bronchial image.

[0168] In one embodiment, the processing module 602 is specifically used to fill the voxels in the local bronchial OCT image into the global bronchial image according to the second pose information.

[0169] In one embodiment, the processing module 602 is specifically configured to: determine the pose information of a first voxel in the space of a local bronchial OCT image, wherein the first voxel is any voxel in the local bronchial OCT image; determine the pose information of the first voxel in a positioning space based on the second pose information and the pose information of the first voxel in the space of the local bronchial OCT image; determine the target index of the first voxel in the space of a global bronchial image based on the pose information of the first voxel in the positioning space; determine the voxel value of the second voxel mapped by the first voxel in the global bronchial image based on the voxel value of the first voxel and the target index; and fill the global bronchial image with the second voxel based on the voxel value of the second voxel.

[0170] In one embodiment, the processing module 602 is further configured to identify the global bronchial image that meets the reconstruction cutoff condition as a search cube; perform multiple cavity searches on the search cube until the number of cavity searches reaches a threshold; and merge the result cubes identified by the last cavity search to obtain the cavity filling area.

[0171] In one embodiment, the hole search process includes:

[0172] Divide the search cube into multiple sub-cubes;

[0173] The result cube in the search cube is determined based on the number of non-hollow voxels in each sub-cube;

[0174] Update the search cube based on the result cube.

[0175] In one embodiment, the processing module 602 is specifically configured to determine the fill value of all cavities based on the values ​​of the non-cavities adjacent to each cavity.

[0176] In one embodiment, the reconstruction cutoff condition includes the cavity ratio of the global bronchial image or the number of fills in the global bronchial image.

[0177] In one embodiment, the processing module 602 is specifically configured to traverse the complete bronchial image in cells of a preset size, determine the vertex index of the isosurface based on the tomographic value of the vertex of each cell and the preset tomographic threshold of the isosurface; obtain the coordinates of the vertex of the isosurface in the positioning space based on the vertex index of the isosurface of each cell, the second pose information and the position of the image origin in the positioning space; and determine the first bronchial surface point cloud data based on the coordinates of the vertex of the isosurface in the positioning space.

[0178] In one embodiment, the processing module 602 is further configured to acquire third pose data of the calibration object in the positioning space and fourth pose data of the positioning component in the positioning space, wherein the calibration object and the conduit are fixed in the positioning space; determine the relative positional relationship between the calibration object and the positioning component based on the third pose data and the fourth pose data; extract fifth pose data of the calibration object in the image coordinate system from the three-dimensional image; and determine the relative positional relationship between the positioning component and the OCT component based on the relative positional relationship between the calibration object and the positioning component and the fifth pose data.

[0179] In one embodiment, the positioning component and the OCT component are disposed inside the catheter; or, the catheter is provided with a working channel, the OCT component is provided with a positioning component, and the OCT component extends into the catheter through the working channel and moves synchronously with the insertion of the catheter.

[0180] Each module in the aforementioned catheter 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 in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0181] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 15As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium 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 medium. The input / output interface is used for exchanging information between the processor and external devices. 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 conduit registration method.

[0182] Those skilled in the art will understand that Figure 15 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.

[0183] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described catheter registration method.

[0184] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described catheter registration method.

[0185] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the catheter registration method described above.

[0186] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0187] 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. When executed, the computer program 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.

[0188] 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.

[0189] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this 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 catheter registration device, characterized by, The catheter comprises a positioning assembly and an optical coherence tomography (OCT) assembly, and the device comprises: a collection module, configured to collect first pose information of the catheter in a process of mirror insertion by the positioning assembly of the catheter, and collect a local bronchus OCT image of the catheter in the process of mirror insertion by the OCT assembly, the local bronchus OCT image being a tomographic image of a local bronchus generated by the OCT assembly through optical interference; a processing module, configured to perform point cloud extraction and three-dimensional splicing on the local bronchus OCT image according to a relative positional relationship between the positioning assembly and the OCT assembly and the first pose information, to obtain first bronchus surface point cloud data of the catheter in the process of mirror insertion; a registration module, configured to register the first bronchus surface point cloud data and second bronchus surface point cloud data, and determine a mapping relationship between the first bronchus surface point cloud data and the second bronchus surface point cloud data, the second bronchus surface point cloud data being point cloud data generated based on a three-dimensional image of a target object before the catheter is inserted; the processing module is specifically configured to determine second pose information of the local bronchus OCT image in a positioning space according to the relative positional relationship between the positioning assembly and the OCT assembly and the first pose information, fill the local bronchus OCT image into a global bronchus image according to the second pose information, until a reconstruction cutoff condition of the global bronchus image is reached, determine filling values of all cavities in a cavity filling area of the global bronchus image that reaches the reconstruction cutoff condition, and perform cavity filling according to the filling values of the all cavities, to obtain a complete bronchus image; the processing module is specifically configured to determine a target index of a first voxel in a space of the global bronchus image according to pose information of the first voxel in the positioning space, the first voxel being any voxel in the local bronchus OCT image, determine a voxel value of a second voxel mapped in the global bronchus image from the first voxel according to a voxel value of the first voxel and the target index, and fill the second voxel in the global bronchus image according to the voxel value of the second voxel.

2. The apparatus of claim 1, wherein, the processing module is specifically configured to determine pose information of a first voxel in a space of the local bronchus OCT image, the first voxel being any voxel in the local bronchus OCT image, and determine pose information of the first voxel in the positioning space according to the second pose information and the pose information of the first voxel in the space of the local bronchus OCT image.

3. The apparatus of claim 1, wherein, the processing module is further configured to determine a search cube as the global bronchus image that reaches the reconstruction cutoff condition, perform multiple cavity searches on the search cube until a number of times of the cavity searches reaches a number threshold, and merge a result cube determined by the last cavity search, to obtain the cavity filling area.

4. The apparatus of claim 3, wherein, the process of the cavity search comprises: dividing the search cube into a plurality of sub-cubes; determine a result cube in the search cube according to a number of non-hole voxels in each sub-cube; update the search cube according to the result cube.

5. The apparatus of claim 1, wherein, The processing module is specifically configured to determine the filling value of the all holes according to the value of the non-hole adjacent to each hole.

6. The device of any one of claims 1-5, wherein, The reconstruction stop condition comprises a hole ratio of the global bronchus image or a filling number of the global bronchus image.

7. The device of any one of claims 1-5, wherein, The processing module is specifically configured to traverse the complete bronchus image in a unit cell of a preset size, determine a vertex index of an isosurface according to a tomography value of a vertex of each unit cell and a tomography threshold value of the isosurface, obtain coordinates of the vertex of the isosurface in the positioning space according to the vertex index of the isosurface of each unit cell, the second pose information and a position of an image origin in the positioning space, and determine the first bronchus surface point cloud data according to the coordinates of the vertex of the isosurface in the positioning space.

8. The apparatus of claim 1, wherein, The processing module is further configured to obtain third pose data of a calibration object and fourth pose data of the positioning assembly in the positioning space, the calibration object and the catheter being fixed in the positioning space, determine a relative position relationship between the calibration object and the positioning assembly according to the third pose data and the fourth pose data, extract fifth pose data of the calibration object in an image coordinate system in the three-dimensional image, and determine the relative position relationship between the positioning assembly and the OCT assembly according to the relative position relationship between the calibration object and the positioning assembly and the fifth pose data.

9. The device of any one of claims 1-5, wherein, The positioning assembly and the OCT assembly are arranged in the catheter, or the catheter is provided with a working channel, the positioning assembly is arranged in the OCT assembly, and the OCT assembly extends into the catheter through the working channel and moves synchronously with the catheter.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The steps implemented by the computer program when executed by the processor include: acquiring first pose information of the catheter in a process of inserting the catheter into the body through a positioning assembly in the catheter, and acquiring a local bronchus OCT image of the catheter in the process of inserting the catheter into the body through an OCT assembly in the catheter, the local bronchus OCT image being a tomographic image of the local bronchus generated by the OCT assembly through optical interference; performing point cloud extraction and three-dimensional splicing on the local bronchus OCT image according to the relative position relationship between the positioning assembly and the OCT assembly and the first pose information, to obtain first bronchus surface point cloud data of the catheter in the process of inserting the catheter into the body; registering the first bronchus surface point cloud data and second bronchus surface point cloud data, determining a mapping relationship between the first bronchus surface point cloud data and the second bronchus surface point cloud data, the second bronchus surface point cloud data being point cloud data generated based on a three-dimensional image of a target object before the catheter is inserted into the body; and According to the relative positional relationship between the positioning assembly and the OCT assembly and the first pose information, second pose information of the local bronchial OCT image in a positioning space is determined; according to the second pose information, the local bronchial OCT image is filled into a global bronchial image until a reconstruction cutoff condition of the global bronchial image is reached; a filling value of all cavities in a cavity filling area of the global bronchial image reaching the reconstruction cutoff condition is determined; cavity filling is performed according to the filling value of all cavities, and a complete bronchial image is obtained; The filling of the local bronchial OCT image into the global bronchial image according to the second pose information comprises: according to the pose information of a first voxel in the positioning space, a target index of the first voxel in the space of the global bronchial image is determined, the first voxel being any voxel in the local bronchial OCT image; according to the voxel value of the first voxel and the target index, a voxel value of a second voxel mapped by the first voxel in the global bronchial image is determined; according to the voxel value of the second voxel, the second voxel is filled in the global bronchial image.

Images