A method, system, and apparatus for surface reconstruction angle selection

Abstract

The embodiment of the present specification provides a curved surface reconstruction angle selection method, system and device. The system can determine the included angle between the distance vector between the sampling points on the center line of the reconstruction object and the candidate interest vector representing the candidate observation angle, and then determine whether the candidate observation angle corresponding to the candidate interest vector is selected according to the included angle. In this way, the candidate observation angle leading to the self-crossing phenomenon can be automatically identified, and the self-crossing part in the curved surface reconstruction image can be avoided to mislead the doctor.

Description

Technical Field

[0001] This specification relates to the field of medical imaging, and in particular to a method, system, and apparatus for selecting the angle of surface reconstruction. Background Technology

[0002] In the 3D visualization of medical imaging data, surface reconstruction methods can extract specific curved surfaces from 3D medical images, which is of great significance for displaying curved structures such as blood vessels. Taking blood vessels as an example, in order to examine the lesions in blood vessels (such as plaque size and composition), doctors usually need to manually adjust the viewing angle of the blood vessel and examine the surface reconstruction image of the blood vessel from each angle (between 0 and 360 degrees). [Reference] Figure 1 From certain viewing angles, self-crossing portions may appear in the surface reconstruction images of blood vessels (see the location indicated by the arrow), which can be visually misleading to doctors, for example, causing them to mistakenly believe that the blood vessel branches at that location. Therefore, doctors need to carefully examine the surface reconstruction images from different viewing angles for self-crossing portions and delete the surface reconstruction images containing self-crossing portions, which is a very tedious and labor-intensive operation. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide an intelligent method for selecting the angle of surface reconstruction.

[0004] One embodiment of this specification provides a method for selecting the angle of surface reconstruction, which includes: obtaining one or more candidate interest vectors, each candidate interest vector representing a candidate observation angle for surface reconstruction; for each candidate interest vector, performing the following operations: determining multiple pairs of sampling points on the centerline of the reconstructed object; for each of the multiple pairs of sampling points, determining the angle between the distance vector between the pair of sampling points and the candidate interest vector; and determining whether the candidate observation angle corresponding to the candidate interest vector is selected based on the angle corresponding to the multiple pairs of sampling points.

[0005] In some embodiments, determining whether a candidate observation angle corresponding to a candidate interest vector is selected based on the included angles corresponding to the multiple pairs of sampling points includes: determining the minimum included angle from the included angles corresponding to the multiple pairs of sampling points; and determining whether a candidate observation angle corresponding to a candidate interest vector is selected based on the relationship between the minimum included angle and a preset included angle threshold.

[0006] In some embodiments, determining multiple pairs of sampling points on the centerline of the reconstructed object includes: determining at least one feature point from multiple sampling points on the centerline; and determining the multiple pairs of sampling points based on the at least one feature point.

[0007] In some embodiments, each feature point can be determined by the following method: determining three consecutive sampling points on the center line, wherein adjacent sampling points among the three sampling points constitute a first reference sampling point pair and a second reference sampling point pair; determining a first angle between a first distance vector corresponding to the first reference sampling point pair and a preset vector; determining a second angle between a second distance vector corresponding to the second reference sampling point pair and the preset vector; determining whether the first angle and the second angle satisfy a preset condition; and in response to determining that the first angle and the second angle satisfy the preset condition, determining the middle sampling point among the three sampling points as the feature point.

[0008] In some embodiments, the at least one feature point includes only one feature point, and determining the plurality of pairs of sampling points based on the at least one feature point includes: dividing the centerline into a first segment and a second segment based on the feature point; and determining the plurality of pairs of sampling points based on sampling points in the first segment and sampling points in the second segment, wherein each pair of sampling points includes one sampling point in the first segment and another sampling point in the second segment.

[0009] In some embodiments, the at least one feature point includes a plurality of feature points, and determining the plurality of pairs of sampling points based on the at least one feature point includes: determining at least one feature segment on the center line based on the plurality of feature points, each feature segment having a pair of adjacent feature points among the plurality of feature points as endpoints; for each of the at least one feature segment, determining one or more pairs of sampling points based on the sampling points in the feature segment and the sampling points in other segments on the center line other than the feature segment; wherein each pair of sampling points includes one sampling point in the feature segment and another sampling point in the other feature segments.

[0010] In some embodiments, the method further includes: for each candidate vector of interest, determining whether an output image obtained by surface reconstruction based on the centerline is likely to contain a self-intersecting portion based on the shape and rotation axis of the centerline; performing the operation in response to determining that the output image may contain a self-intersecting portion; or, in response to determining that the output image is unlikely to contain a self-intersecting portion, skipping the operation and determining that a candidate viewing angle corresponding to the candidate vector of interest is selected.

[0011] In some embodiments, determining multiple pairs of sampling points on the centerline of the reconstructed object includes: determining a sampling interval based on a preset surface width; determining multiple sampling points on the centerline according to the sampling interval; and combining the multiple sampling points in pairs to obtain multiple pairs of sampling points.

[0012] In some embodiments, determining multiple pairs of sampling points on the centerline of the reconstructed object includes: dividing the centerline into multiple segments; for each of the multiple segments, determining the sampling interval of the segment based on the smoothness of the segment; and determining sampling points on the segment based on the sampling interval corresponding to the segment.

[0013] One embodiment of this specification provides a surface reconstruction angle selection system, which includes an acquisition module, a sampling point pair determination module, an angle determination module, and a selection module. The acquisition module is used to acquire one or more candidate interest vectors, each candidate interest vector representing a candidate observation angle for surface reconstruction. For each candidate interest vector: the sampling point pair determination module is used to determine multiple pairs of sampling points on the centerline of the reconstructed object; the angle determination module is used to: for each of the multiple pairs of sampling points, determine the angle between the distance vector between the pair of sampling points and the candidate interest vector; and the selection module is used to: based on the angles corresponding to the multiple pairs of sampling points, determine whether the candidate observation angle corresponding to the candidate interest vector is selected.

[0014] One embodiment of this specification provides a surface reconstruction angle selection device, which includes a processor and a storage device. The storage device is used to store instructions, and when the processor executes the instructions, it implements the surface reconstruction angle selection method as described in any embodiment of this specification.

[0015] This specification provides a method, system, and apparatus for selecting angles in surface reconstruction. The system can determine the angle between the distance vector between sampling points on the centerline of the reconstructed object (such as a blood vessel) and a candidate vector of interest representing a candidate observation angle, and then determine whether the candidate observation angle corresponding to the candidate vector of interest is selected based on this angle. In this way, candidate observation angles that cause self-crossing can be automatically identified, avoiding misleading doctors by self-crossing portions appearing in the surface reconstruction image. Attached Figure Description

[0016] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0017] Figure 1 This is a schematic diagram of a surface reconstruction image containing self-intersecting portions;

[0018] Figure 2 These are schematic diagrams illustrating application scenarios of medical imaging systems based on some embodiments of this specification;

[0019] Figure 3This is an exemplary block diagram of a surface reconstruction angle selection system according to some embodiments of this specification;

[0020] Figure 4 This is an exemplary flowchart of a surface reconstruction angle selection method according to some embodiments of this specification;

[0021] Figure 5A and Figure 5B This is a schematic diagram of the centerline and the vector of interest in the surface reconstruction;

[0022] Figure 6A and Figure 6B This is a schematic diagram used to analyze the reasons for the appearance of self-intersecting parts in curved surface images;

[0023] Figure 7 It is a schematic diagram of the center line containing multiple feature points. Detailed Implementation

[0024] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0025] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0026] As indicated in this specification, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0027] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0028] Figure 2These are schematic diagrams illustrating application scenarios of medical imaging systems based on some embodiments of this specification. For example... Figure 2 As shown, system 100 may include scanning device 110, processing device 120, user terminal 130, storage device 140 and network 150.

[0029] The scanning device 110 can be used to acquire scan data of the object to be reconstructed. The scan data can be used for image reconstruction, for example, for the reconstruction of 2D or 3D images. In some embodiments, the object to be reconstructed can refer to an organism (such as a patient) or a part of an organism (such as the upper body). In some embodiments, the object to be reconstructed can include tubular tissues, such as blood vessels, bile ducts, trachea, small and large intestines, etc. In some embodiments, the scanning device 110 can send the scan data to other devices (such as processing device 120) for image reconstruction. In some embodiments, the scanning device 110 can perform image reconstruction independently.

[0030] In some embodiments, the scanning device 110 may include a CT (Computed Tomography) device, a PET (Positron Emission Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, a CAT (Computer-Aided Tomography) device, a DSA (Digital Subtraction Angiography) device, or any combination thereof.

[0031] The processing device 120 can be used to process information and / or data. In some embodiments, the processing device 120 can reconstruct a 3D image based on scanned data. In some embodiments, the processing device 120 can perform surface reconstruction based on the 3D image to obtain a surface-reconstructed image. In some embodiments, the processing device 120 can intelligently select the surface reconstruction angle, where the surface reconstruction angle refers to the viewing angle of the surface reconstruction. For more details on surface reconstruction and surface reconstruction angle selection, please refer to [reference needed]. Figure 4 And its related descriptions.

[0032] In some embodiments, the processing device 120 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, the processing device 120 may be local or remote. In some embodiments, the processing device 120 may be implemented on a cloud platform. By way of example only, the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-tiered cloud, etc., or any combination thereof.

[0033] User terminal 130 can be used to enable interaction with a user (such as a doctor). In some embodiments, the user can manipulate one or more components in system 100 through user terminal 130. For example, user terminal 130 can send instructions to processing device 120 to generate a surface reconstruction image at a specific reconstruction angle. Alternatively, user terminal 130 can convey image reconstruction instructions to processing device 120. In some embodiments, user terminal 130 can display medical images generated in system 100 to the user. For example, user terminal 130 can display reconstructed images (such as surface reconstruction images) from scanning device 110 or processing device 120 to the user.

[0034] In some embodiments, the user terminal 130 may include a smartphone 130-1, a tablet computer 130-2, a laptop computer 130-3, a desktop computer 130-4, or any combination thereof. In some embodiments, the user terminal 130 may be integrated into the processing device 120.

[0035] Storage device 140 can be used to store data and / or instructions. For example, storage device 140 can be used to store reconstructed images (such as surface reconstruction images) from scanning device 110 or processing device 120. As another example, storage device 140 can be used to store surface reconstruction angle selection instructions. When the processor of processing device 120 executes these instructions, the surface reconstruction angle selection method provided in the embodiments of this specification can be implemented. Furthermore, storage device 140 can be used to store surface reconstruction algorithms.

[0036] In some embodiments, storage device 140 may include mass storage, removable storage, volatile read-write storage, read-only storage, or any combination thereof. In some embodiments, storage device 140 may be integrated into processing device 120.

[0037] Network 150 can be used to facilitate the exchange of information and / or data within system 100. For example, scanning device 110 can send scan data to processing device 120 via network 150, and processing device 120 can also send reconstructed images based on the scan data to user terminal 130 and / or storage device 140 via network 150. As another example, processing device 120 can read surface reconstruction angle selection instructions from storage device 140 via network 150.

[0038] In some embodiments, network 150 may include public networks (such as the Internet), private networks (such as local area networks, wide area networks, etc.), wired networks (such as Ethernet), wireless networks (such as Wi-Fi networks, Bluetooth networks, etc.), and any combination thereof.

[0039] Figure 3This is an exemplary block diagram of a surface reconstruction angle selection system according to some embodiments of this specification. In some embodiments, system 200 can... Figure 1 It is implemented on the processing device 120 shown.

[0040] like Figure 3 As shown, system 200 may include acquisition module 210, sampling point pair determination module 220, included angle determination module 230 and selection module 240.

[0041] The acquisition module 210 can be used to acquire one or more candidate interest vectors, each of which can represent a candidate viewing angle for surface reconstruction.

[0042] For each candidate vector of interest: the sampling point pair determination module 220 can be used to determine multiple pairs of sampling points on the center line of the reconstructed object; the angle determination module 230 can be used to determine the angle between the distance vector between the pair of sampling points and the candidate vector of interest for each of the multiple pairs of sampling points; the selection module 240 can be used to determine whether the candidate observation angle corresponding to the candidate vector of interest is selected based on the angles corresponding to the multiple pairs of sampling points.

[0043] For more details about System 200 and its modules, please refer to [link / reference]. Figure 4 And its related descriptions.

[0044] It should be understood that Figure 2 , Figure 3 The systems and modules shown can be implemented in various ways. For example, in some embodiments, the systems and modules can be implemented by hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the methods and systems described above can be implemented using computer-executable instructions and / or included in processor control code, for example, on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and modules of this specification can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., but also by software, for example, executed by various types of processors, or by a combination of the aforementioned hardware circuits and software (e.g., firmware).

[0045] It should be noted that the above description of the system and its modules is for convenience only and should not be construed as limiting this specification to the embodiments described. It is understood that those skilled in the art, after understanding the system's principles, may arbitrarily combine the modules or construct subsystems connected to other modules without departing from these principles. For example, in some embodiments, the angle determination module 230 and the selection module 240 may be two modules or combined into one module. Such modifications are all within the scope of this specification.

[0046] Figure 4 This is an exemplary flowchart illustrating the selection of surface reconstruction angles according to some embodiments of this specification. For example... Figure 4 As shown, process 400 may include the following steps.

[0047] Step 410: Obtain one or more candidate vectors of interest.

[0048] Each candidate Vector of Interest (VOI) can represent a candidate viewing angle for surface reconstruction. It is described as "candidate" because subsequent steps (step 430) will filter the candidate VOIs / candidate viewing angles based on relevant angles. Furthermore, since VOIs and viewing angles are mutually corresponding, filtering viewing angles is equivalent to filtering VOIs; therefore, before making a selection, both the VOI and its corresponding viewing angle are considered candidates.

[0049] The vector of interest is one of the important concepts in Curved Planar Reformation (CPR). The following section combines... Figure 5A and Figure 5B Introduce the vector of interest.

[0050] CPR (Continuous Plane Reconstruction) is an extension of MPR (Multiple Plane Reconstruction). The goal of CPR visualization is to present the complete structure of curved structures within a living organism (e.g., blood vessels or other tubular structures) in a single image. To achieve this, prior information about the curved structure is needed, particularly its central axis (or centerline). Typically, the spatial location and shape of the centerline determine which parts of the curved structure will be visualized. Figure 5AAn exemplary centerline is illustrated in a three-dimensional coordinate system (e.g., the patient coordinate system defined in the DICOM standard). This three-dimensional coordinate system is an orthogonal coordinate system containing x-axis, y-axis, and z-axis, where one axis will be determined as the rotation axis. Specifically, the vector connecting the first and last points on the centerline can be projected onto the x, y, and z axes respectively to obtain three projected vectors, and the coordinate axis containing the longest of the three projected vectors is used as the rotation axis. For example... Figure 5A As shown, the length of the projection vector on the z-axis is greater than that of the other two axes, so the z-axis will be used as the rotation axis.

[0051] Since a surface cannot be well defined by a curve in three-dimensional space, surface reconstruction also introduces a vector of interest (ROI) to represent the viewing angle of the reconstructed object. When reconstruction is performed based on a specific ROI, the structure seen when looking at the reconstructed object from the viewing angle (or sampling angle) corresponding to that ROI vector can be obtained. By setting ROI vectors in different directions, different viewing angles of the reconstructed object can be achieved. In some embodiments, the ROI vector can be rotated around a rotation axis to obtain multiple candidate ROI vectors in different directions to represent multiple candidate viewing angles in the range of 0° to 360°. For example, refer to... Figure 5B Here, vi represents the vector of interest. Multiple candidate vectors of interest can be obtained by rotating vi around the z-axis at fixed angular intervals (e.g., 1°, 3°, 5°). Candidate vectors of interest parallel to the y-axis correspond to a candidate observation angle of 0°, while other vectors of interest correspond to candidate observation angles ranging from 0° to 360°. In some embodiments, the vector of interest vi and a point on the center line together define a straight line, which can be called the line of interest (LOI), for example... Figure 5B A line of interest (LI) is shown. During surface reconstruction, all voxels traversed by the line of interest can be acquired for resampling along the line of interest.

[0052] After introducing the vector of interest, the surface to be reconstructed can be defined using the centerline and the vector of interest. This surface can pass through multiple sampling points on the centerline and extend along the direction corresponding to the vector of interest. For example... Figure 5B As shown, the gray surface is the surface to be reconstructed. It is worth noting that in some embodiments, the surface defined by the centerline and the vector of interest can have a certain thickness and width. For example, Figure 5B The surface in the system can have a certain thickness in the x-axis direction and a certain width in the y-axis direction. The surface thickness and width can be set by the system default value or by the user (e.g., a doctor). In some embodiments, the system will sample according to the preset surface width.

[0053] In some embodiments, considering symmetry, candidate observation angles within the range of 0° to 180° can be filtered according to process 400. After completing the angle filtering within the range of 0° to 180°, 180° can be added to each selected candidate observation angle to obtain the angle filtering results within the range of 180° to 360°. For example, if candidate observation angle 20° is selected, it means that candidate observation angle 200° can also be selected.

[0054] For each candidate interest vector (denoted as v) i The processing device 120 can execute steps 420-440 to determine the candidate interest vector v. i The corresponding candidate observation angle (denoted as) Whether it was selected.

[0055] Step 420: Determine multiple pairs of sampling points on the center line of the reconstructed object.

[0056] In some embodiments, the reconstructed object may include tubular tissues, such as blood vessels, ribs, spine, ureters, bronchi, colon, etc.

[0057] This specification does not impose any restrictions on the method of obtaining the centerline of the reconstructed object. As an example only, the processing device 120 can acquire an original 3D image containing the reconstructed object, perform image segmentation on the original 3D image, and obtain a segmented image of the reconstructed object (such as a blood vessel segmentation image). Furthermore, the processing device 120 can determine the centerline of the reconstructed object based on the segmented image. Optionally, the centerline can be stored as a spatial curve equation, and the start and end points of the centerline can be stored as spatial coordinates. Optionally, the centerline can be stored as the spatial coordinates of multiple points on it.

[0058] In some embodiments, the sampling point determination module 220 may perform uniform (equal-interval) or non-uniform sampling along the centerline of the reconstructed object to obtain multiple sampling points. It is worth noting that uniform sampling can refer to sampling at equal intervals along the path of the centerline, or it can refer to sampling along a fixed direction (such as...). Figure 5A The sampling points are sampled at equal intervals along the z-axis. For example, the sampling point pair determination module 220 can divide the path length of the center line (denoted as s) into n equal parts, and correspondingly, the interval between adjacent sampling points on the path of the center line is equal to s / n. As another example, refer to... Figure 5A The sampling point pair determination module 220 can determine the length of the z-axis projection of the center line (denoted as s). z The sample is divided into n equal parts, and correspondingly, the interval between adjacent sampling points along the z-axis is equal to s. z / n.

[0059] The sampling interval can be the system's default setting, manually set, or determined by the sampling point pair determination module 220 based on actual conditions. In some embodiments, the sampling point pair determination module 220 can determine the sampling interval based on a preset surface width and / or voxel size, and determine multiple sampling points along the center line according to the sampling interval. Typically, the sampling interval does not exceed the preset surface width to exclude sampling point pairs with excessively large intervals. The surface width is the width of the reconstructed two-dimensional surface image (e.g., along...). Figure 5B The width of the surface (as shown in the Y-axis direction) can be the system default or user-defined. Under this premise, the sampling interval can be a preset multiple of the voxel size; for example, the sampling interval can be equal to 3 times the voxel size.

[0060] Once multiple sampling points are determined, the sampling point pair determination module 220 can combine the multiple sampling points in pairs to obtain multiple pairs of sampling points. In some embodiments, for each sampling point, it is necessary to combine the sampling point with each other sampling point to generate a corresponding sampling point pair. That is, multiple pairs of sampling points include all possible combinations of sampling points. Assuming there are n sampling points, n*(n-1) / 2 pairs of sampling points can be generated. In some embodiments, the sampling point pair determination module 220 can determine at least one feature point from multiple sampling points on the centerline, and then determine multiple pairs of sampling points based on the at least one feature point. The determination of feature points helps to reduce the number of sampling point pairs that need to participate in the angle determination step (step 430), thereby improving the screening efficiency of the surface reconstruction angle.

[0061] The following is combined with Figure 6A An exemplary method for determining feature points is provided.

[0062] refer to Figure 6AThe sampling point pair determination module 220 can determine three consecutive sampling points on the center line, such as Pi, Pk, and Pj. It should be noted that these three sampling points may or may not be consecutive. Adjacent sampling points among the three sampling points constitute a first reference sampling point pair and a second reference sampling point pair, for example, the first reference sampling point pair Pi-Pk and the second reference sampling point pair Pk-Pj. It should be noted that the reference sampling point pair refers to the sampling point pair used to determine feature points, which is different from the sampling point pair finally determined in step 420. The sampling point pair determination module 220 can determine the first angle between the first distance vector corresponding to the first reference sampling point pair (e.g., Pi-Pk) and a preset vector, and determine the second angle between the second distance vector corresponding to the second reference sampling point pair (e.g., Pk-Pj) and the preset vector. The distance vector can be a vector formed by connecting two sampling points, and the preset vector can be parallel to the rotation axis of the vector of interest. The sampling point pair determination module 220 can determine whether the first angle and the second angle satisfy a preset condition. In response to determining that the first included angle and the second included angle satisfy a preset condition, the sampling point pair determination module 220 can determine the middle sampling point (e.g., Pk) among the three sampling points as a feature point. The preset condition may include that the cosine values ​​of the first included angle and the second included angle have opposite signs. (Reference) Figure 6A Assuming the preset vector is the rotation axis in the diagram and points downwards, the first angle between the first reference sampling point and Pi-Pk is acute, while the angle between the second reference sampling point and Pk-Pj is obtuse. According to the characteristics of the cosine function, the cosine values ​​of the first and second angles have opposite signs. Figure 6A As can be seen, the angle between the distance vector between sampling points (such as Pi and Pj) on both sides of Pk and the interest vector perpendicular to the rotation axis tends to approach 0°, which will lead to self-crossing in the surface reconstruction image. Therefore, Pk can be used as a feature point to determine sampling point pairs. More information on the causes of self-crossing can be found later.

[0063] In some embodiments, the sampling point determination module 220 can analyze all three consecutive sampling points on the center line to determine whether a sampling point can be used as a feature point.

[0064] In some embodiments, the determined at least one feature point may include only one feature point, and the sampling point pair determination module 220 may divide the centerline into a first segment and a second segment based on this feature point. Furthermore, the sampling point pair determination module 220 may determine the plurality of pairs of sampling points based on sampling points in the first segment and sampling points in the second segment, wherein each pair of sampling points includes one sampling point in the first segment and another sampling point in the second segment. For example, referring to... Figure 6AThe sampling point pair determination module 220 can divide the centerline into a first segment P0-Pk and a second segment Pk-Pn based on the feature point Pk. Furthermore, each pair of sampling points determined by the sampling point pair determination module 220 may include one sampling point from the first segment P0-Pk and another sampling point from the second segment Pk-Pn.

[0065] In some embodiments, the at least one feature point may include multiple feature points, and the sampling point pair determination module 220 may determine at least one feature segment on the centerline based on the multiple feature points, each feature segment having a pair of adjacent feature points among the multiple feature points as endpoints. Furthermore, for each of the at least one feature segment, the sampling point pair determination module 220 may determine one or more pairs of sampling points based on sampling points in the feature segment and sampling points in other segments on the centerline besides the feature segment. Each pair of sampling points includes one sampling point in the feature segment and another sampling point in the other feature segments. For example, refer to... Figure 7 The sampling point pair determination module 220 can determine the feature segment Ps-Pt based on the feature points Ps and Pt. Furthermore, the sampling point pair determination module 220 can determine one or more pairs of sampling points based on the sampling points in the feature segment Ps-Pt and the sampling points in other segments (including P0-Pr and Py-Pn) on the center line other than the feature segment Ps-Pt. Each pair of sampling points may include one sampling point in the feature segment Ps-Pt and another sampling point in the other segment.

[0066] In some embodiments, the sampling point pair determination module 220 can flexibly set the sampling interval according to the smoothness of different segments of the centerline, so as to reasonably control the number of determined sampling point pairs and thus ensure the screening efficiency of the surface reconstruction angle. Specifically, the sampling point pair determination module 220 can divide the centerline (e.g., randomly or equally) into multiple segments. For each of the multiple segments (denoted as Li), the sampling point pair determination module 220 can determine the sampling interval of segment Li based on the smoothness of segment Li. It can be understood that the smoother the segment Li, the larger the sampling interval can be set for it.

[0067] Step 430: For each of the multiple pairs of sampling points, determine the distance vector and the candidate interest vector (e.g., v) between the pair of sampling points. i The angle between them.

[0068] As mentioned above, some reconstructed objects may produce surface reconstruction images with self-crossing phenomena under certain candidate viewing angles. Self-crossing refers to the appearance of branches in the surface reconstruction image of reconstructed objects that originally did not have branches (such as blood vessels), which may lead to diagnostic errors.

[0069] The following is combined with Figure 6A and Figure 6B Analyze the reasons for the appearance of self-intersecting parts in the reconstructed surface image. For example... Figure 6A As shown, during surface reconstruction, it is necessary to traverse the sampling points along the centerline. When accessing Pi from P0, the distance vector between the sampling points along the centerline and the candidate interest vector v are... i The included angles are relatively large, and the reconstructed image from the curved surface will not have self-intersecting parts. When point Pi is visited, the vector between Pi and Pj and the candidate interest vector v i Parallel. When generating a surface reconstruction image by traversing the centerline of the Pi-Pk segment of the centerline, such as... Figure 6B As shown, in the reconstructed surface image, Pi and Pj are at the same horizontal position. When traversing the Pk-Pj segment of the centerline, as... Figure 6B As shown, the pixels (voxels) from Pj to Pi will be repeatedly sampled, resulting in self-intersections in the surface reconstruction image.

[0070] Generally, when the distance vector between sampling points is similar to the candidate interest vector (e.g., v), i If the angle between the sampling points is too small, self-intersecting parts will appear in the surface reconstruction image. Therefore, the angle determination module 230 can determine the distance vector between the sampling points and the candidate interest vector (such as v). i The angle between the sampling points is used to intelligently filter the surface reconstruction angle based on the corresponding angle of each sampling point.

[0071] It is worth mentioning that, in practical applications, the angle determination module 230 can determine the distance vector and the vector of interest (such as v). i The angle between the vectors is limited to the range of 0° to 90° for subsequent use (step 440). Specifically, the angle determination module 230 can ignore the distance vector and the vector of interest v. i The direction is chosen from the four included angles formed after the vectors intersect, within the range of 0° to 90°. For example, refer to... Figure 6A If we consider the vector direction, the angle between the distance vector from Pi to Pj and the vector of interest should be 180°, but the angle determination module 230 can take 0° as the angle.

[0072] In some embodiments, any angle mentioned herein can be quantified not only by degrees but also by trigonometric function values ​​(such as cosine values). For example, the angle 0° can also be recorded as 1 (cos0° = 1). Accordingly, the magnitude relationship between angles can be determined based on the magnitude relationship between trigonometric function values. For example, assuming we want to determine the magnitude relationship between two angles α and β within the range of 0° to 90°, we can determine the magnitude relationship between cosα and cosβ. According to the monotonicity of the cosine function, when cosα is greater than cosβ, α is less than β.

[0073] Step 440: Based on the included angles corresponding to the multiple pairs of sampling points, determine the candidate interest vector (e.g., v). i The corresponding candidate observation angle (e.g.) Whether it was selected.

[0074] With candidate interest vector v i For example, when the included angle between any pair of sampling points (at least one pair of sampling points) is too small (e.g., less than a preset included angle threshold), based on the candidate interest vector v i The obtained surface reconstruction image (i.e., the viewing angle is) The reconstructed surface image may exhibit self-intersecting regions. Based on this, when the included angle corresponding to any pair of sampling points (at least one pair of sampling points) is too small (e.g., less than a preset included angle threshold), the selection module 240 can determine the candidate interest vector v. i Corresponding candidate observation angles Not selected. However, when the included angles of all sampling points are sufficiently large (e.g., not less than a preset included angle threshold), the selection module 240 can determine the candidate interest vector v. i Corresponding candidate observation angles Selected.

[0075] The preset angle threshold can be determined through sample statistics. Specifically, the processing device 120 can acquire a certain number of samples (e.g., 100, 500, or 1000). Each sample can include a surface reconstruction image that is manually determined to have no self-intersecting parts, along with its related parameters (e.g., centerline, sampling interval, and vector of interest). The threshold can be obtained by analyzing these samples. For example, the minimum angle between the distance vector between sampling points and the vector of interest can be calculated based on the related parameters, and the minimum angle obtained can be used as the threshold. For example, the preset angle threshold can be 10°, 5°, 3°, 2°, 1°, etc.

[0076] In some embodiments, the selection module 240 can determine candidate interest vectors (such as v) based on the relationship between the included angle corresponding to each pair of sampling points and a preset included angle threshold. i The corresponding candidate observation angle (e.g.) Whether a candidate vector of interest (V) is selected. In some embodiments, the selection module 240 can determine the minimum angle from the angles corresponding to the plurality of pairs of sampling points. Furthermore, the selection module 240 can determine the candidate vector of interest (e.g., v) based on the relationship between the minimum angle and a preset angle threshold. i The corresponding candidate observation angle (e.g.) Whether a sample is selected. It can be understood that when the minimum included angle is not less than a preset included angle threshold, it means that the included angles of all sampling points are not less than the preset included angle threshold. In this case, the selection module 240 can determine the candidate vector of interest (e.g., v). i The corresponding candidate observation angle (e.g.) () was selected.

[0077] In some embodiments, for an unselected viewing angle, if a surface reconstruction image of that viewing angle has been obtained, the processing device 120 can delete the surface reconstruction image of that viewing angle. Thus, the doctor does not need to manually filter and delete surface reconstruction images containing intersecting portions.

[0078] In some embodiments, for unselected viewing angles (such as...) The processing device 120 can abandon processing based on the corresponding interest vector (such as v). i Perform surface reconstruction. This effectively avoids self-intersecting parts in the generated surface reconstruction image.

[0079] In some embodiments, the processing device 120 can predict whether the surface reconstruction image may contain self-intersecting portions, and then decide whether to perform surface reconstruction angle selection based on the prediction result. Specifically, for each candidate interest vector (denoted as v... i The processing device 120 can determine the candidate interest vector v based on the shape of the centerline and the rotation axis. i The output image obtained by surface reconstruction (i.e., based on the candidate interest vector v) i Is it possible for the obtained reconstructed surface image to contain self-intersecting portions? (See also: For example only) Figure 6A The processing device 120 can treat the rotation axis as an elevation line. When the center line (such as the Pi-Pj segment) exhibits an up-and-down oscillating pattern, the processing device 120 can determine that the output image may contain a self-crossing portion; otherwise, the processing device 120 can determine that the output image cannot contain a self-crossing portion. For example, the processing device 120 can input the center line shape parameters, rotation axis parameters, and candidate interest vector parameters into a pre-trained machine learning model. This model can determine whether the output image may contain a self-crossing portion based on the input. In response to determining that the output image may contain a self-crossing portion, the processing device 120 can execute steps 420-440. Alternatively, in response to determining that the output image cannot contain a self-crossing portion, the processing device 120 can skip steps 420-440 and determine the candidate interest vector v. i The corresponding candidate viewing angles are selected. This improves the efficiency of angle selection for surface reconstruction.

[0080] It should be noted that the above description of the process is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to the process under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0081] The beneficial effects that the embodiments of this specification may bring include, but are not limited to: (1) providing an intelligent screening method for surface reconstruction angles, which can automatically identify surface reconstruction images containing self-intersecting parts instead of doctors, or directly discard some observation angles to avoid the appearance of self-intersecting parts in surface reconstruction images; (2) by determining feature points, the number of sampling point pairs can be reduced, thereby improving the screening efficiency of surface reconstruction angles; (3) controlling the sampling interval to not exceed the preset surface width to exclude sampling point pairs with excessively large intervals; (4) by predicting whether self-intersecting is possible, the efficiency of surface reconstruction angle selection can be improved; (5) by setting the sampling interval according to the smoothness of different segments of the center line, the number of sampling point pairs can be reasonably controlled, thereby ensuring the screening efficiency of surface reconstruction angles. It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced can be any one or a combination of the above, or any other possible beneficial effects.

[0082] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation on the embodiments of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to the embodiments of this specification. Such modifications, improvements, and corrections are suggested in the embodiments of this specification, and therefore, such modifications, improvements, and corrections still fall within the spirit and scope of the exemplary embodiments of this specification.

[0083] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0084] Furthermore, those skilled in the art will understand that various aspects of the embodiments of this specification can be described and illustrated through several patentable types or situations, including any new and useful combinations of processes, machines, products, or substances, or any new and useful improvements thereto. Accordingly, various aspects of the embodiments of this specification can be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. All of the above hardware or software may be referred to as a "data block," "module," "engine," "unit," "component," or "system." Furthermore, various aspects of the embodiments of this specification may be embodied as a computer product located on one or more computer-readable media, the product including computer-readable program code.

[0085] Computer storage media may contain a propagated data signal containing computer program code, for example, on baseband or as part of a carrier wave. This propagated signal may take various forms, including electromagnetic, optical, and suitable combinations thereof. Computer storage media can be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, apparatus, or device to enable communication, propagation, or transmission of a program for use. The program code located on the computer storage medium can be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0086] The computer program code required for the operation of each part of the embodiments in this specification can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C, Visual Basic, Fortran2003, Perl, COBOL2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby, and Groovy, or other programming languages. This program code can run entirely on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or processing device. In the latter case, the remote computer can be connected to the user's computer through any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as Software as a Service (SaaS).

[0087] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in the embodiments of this specification are not intended to limit the order of the processes and methods of the embodiments of this specification. Although some inventive embodiments that are currently considered useful have been discussed by way of various examples in the foregoing disclosure, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments. Rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented solely by software solutions, such as installing the described system on existing processing devices or mobile devices.

[0088] Similarly, it should be noted that, in order to simplify the description of the embodiments disclosed in this specification and thereby aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the objects of the embodiments in this specification require more features than those mentioned in the claims. In fact, the embodiments have fewer features than all the features of the single embodiments disclosed above.

[0089] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with this specification, as well as documents that limit the broadest scope of the claims of this application (currently or subsequently appended to this application). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0090] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of the embodiments described herein. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for selecting the angle in surface reconstruction, characterized in that, include: Obtain one or more candidate interest vectors, each of which represents a candidate viewing angle for surface reconstruction; For each of the candidate interest vectors, Based on the shape and rotation axis of the centerline of the reconstructed object, determine whether the output image obtained by surface reconstruction based on the candidate interest vector may contain self-intersecting parts; In response to determining that the output image may contain self-intersecting portions, the following operations are performed: Multiple pairs of sampling points are determined along the center line; For each of the multiple pairs of sampling points, determine the angle between the distance vector between the pair of sampling points and the candidate interest vector; as well as Based on the included angles corresponding to the multiple pairs of sampling points, it is determined whether the candidate observation angle corresponding to the candidate interest vector is selected.

2. The method of claim 1, wherein, The step of determining whether a candidate observation angle corresponding to a candidate interest vector is selected based on the included angles corresponding to the multiple pairs of sampling points includes: Determine the minimum included angle from the included angles corresponding to the multiple pairs of sampling points; and Based on the relationship between the minimum included angle and the preset included angle threshold, it is determined whether the candidate observation angle corresponding to the candidate interest vector is selected.

3. The method of claim 1, wherein, The determination of multiple pairs of sampling points on the center line includes: Determine at least one feature point from a plurality of sampling points along the center line; and The plurality of sampling points are determined based on the at least one feature point; Each of the aforementioned feature points is determined using the following method: Three consecutive sampling points are determined on the center line, and adjacent sampling points among the three sampling points constitute a first reference sampling point pair and a second reference sampling point pair; Determine the first angle between the first distance vector corresponding to the first reference sampling point pair and the preset vector; Determine the second angle between the second distance vector corresponding to the second reference sampling point pair and the preset vector; Determine whether the first included angle and the second included angle satisfy a preset condition; and In response to determining that the first included angle and the second included angle satisfy the preset condition, the middle sampling point among the three sampling points is determined as the feature point.

4. The method of claim 3, wherein, The at least one feature point includes only one feature point, and determining the multiple pairs of sampling points based on the at least one feature point includes: Based on this feature point, the centerline is divided into a first segment and a second segment; and The plurality of pairs of sampling points are determined based on the sampling points in the first segment and the sampling points in the second segment, wherein each pair of sampling points includes one sampling point in the first segment and another sampling point in the second segment.

5. The method of claim 3, wherein, The at least one feature point includes multiple feature points, and determining the multiple pairs of sampling points based on the at least one feature point includes: At least one feature segment is determined on the center line based on the plurality of feature points, and each feature segment takes a pair of adjacent feature points among the plurality of feature points as its endpoints; For each of the at least one feature segment, one or more pairs of sampling points are determined based on the sampling points in the feature segment and the sampling points in other segments on the center line other than the feature segment; wherein each pair of sampling points includes one sampling point in the feature segment and another sampling point in the other feature segment.

6. The method as described in claim 1, characterized in that, The determination of multiple pairs of sampling points on the center line includes: The sampling interval is determined based on the preset surface width; Multiple sampling points are determined along the center line according to the sampling interval; and The multiple sampling points are combined in pairs to obtain multiple pairs of sampling points.

7. The method as described in claim 1, characterized in that, The determination of multiple pairs of sampling points on the center line includes: The centerline is divided into multiple segments; For each of the plurality of segments, Based on the smoothness of the segmentation, the sampling interval of the segment is determined; and Sampling points are determined on the segments based on the sampling intervals corresponding to the segments.

8. A surface reconstruction angle selection system, characterized in that, It includes an acquisition module, a sampling point pair determination module, an angle determination module, and a selection module; The acquisition module is used to acquire one or more candidate interest vectors, each of which represents a candidate viewing angle for surface reconstruction; For each candidate interest vector, the acquisition module is further configured to determine, based on the shape of the centerline of the reconstructed object and the axis of rotation, whether the output image obtained by surface reconstruction based on the candidate interest vector may contain a self-intersecting portion; In response to determining that the output image may contain self-intersecting portions: The sampling point pair determination module is used to determine multiple pairs of sampling points on the center line; The angle determination module is used to: for each of the plurality of pairs of sampling points, determine the angle between the distance vector between the pair of sampling points and the candidate interest vector; and The selection module is used to: determine whether the candidate observation angle corresponding to the candidate interest vector is selected based on the included angles corresponding to the multiple pairs of sampling points.

9. A surface reconstruction angle selection device, characterized in that, It includes a processor and a storage device, the storage device being used to store instructions that, when executed by the processor, implement the method as described in any one of claims 1 to 7.

Images